arbete och hälsa | vetenskaplig skriftserie isbn 91-7045-559-7 issn 0346-7821 http://www.niwl.se/ah/ nr 2000:8 Ergonomics of Protective Clothing Proceedings of nokobetef 6 and 1st European Conference on Protective Clothing held in Stockholm, Sweden, May 7–10, 2000 Kalev Kuklane and Ingvar Holmér (eds.) National Institute for Working Life ARBETE OCH HÄLSA Editor-in-chief: Staffan Marklund Co-editors: Mikael Bergenheim, Anders Kjellberg, Birgitta Meding, Gunnar Rosén och Ewa Wigaeus Hjelm © National Institut for Working Life & authors 2000 National Institute for Working Life S-112 79 Stockholm Sweden ISBN 91–7045–559–7 ISSN 0346–7821 http://www.niwl.se/ah/ Printed at CM Gruppen The National Institute for Working Life is Sweden’s national centre for work life research, development and training. The labour market, occupational safety and health, and work organi- sation are our main fields of activity. The creation and use of knowledge through learning, information and documentation are important to the Institute, as is international co-operation. The Institute is collaborating with interested parties in various develop- ment projects. The areas in which the Institute is active include: • labour market and labour law, • work organisation, • musculoskeletal disorders, • chemical substances and allergens, noise and electromagnetic fields, • the psychosocial problems and strain-related disorders in modern working life. IForeword The European Directives on personal protective devices have increased the interest in protective and functional properties of work clothing and intensified standardisation work as well as stimulated research in areas with limited knowledge. There is a long tradition of research and information exchange in the Nordic countries on the subject. The NOrdisk KOrdineringsgruppe om BEskyttelseskläder som TEknisk Forebyggelsesmiddel (Nordic Coordination Group on Protective Clothing as a Technical Preventive Measure) was founded in 1984. NOKOBETEF is an independent society of professionals from the Nordic as well as other countries. NOKOBETEF has since its foundation organised symposia in Copenhagen (1984), Stockholm (1986), Gausdal, Norway (1989), Kittilä, Finland (1992), and Elsinor, Denmark (1997). The conferences have long had a good attendance from European countries and from overseas. The 6th Nokobetef conference was organised as the 1st European Conference on Protective Clothing to emphasize the European dimension. During the conference the European Society for Protective Clothing was founded. One of its first tasks will be to prepare for the 2nd conference to be held in Switzerland in 2003. The proceedings of this conference cover a broad spectrum of the subject protective clothing. Emphasis was given to the ergonomics aspects, which is in line with the present interest and priorities of the European standardisation bodies (CEN). A functional and comfortable use of protective clothing is a key element for a succesful implementation of this kind of preventive and protective measures in the workplaces. A total of 77 papers are presented in this book. They represent a qualified source of new, valuable and useful information for the advancement of the knowledge and the application of protective clothing. Solna in May 2000 Ingvar Holmér II Program Committee Organising Committee Helena Mäkinen, Finland Ingvar Holmér, Sweden Ingvar Holmér, Sweden Helena Mäkinen, Finland Ruth Nielsen, Denmark Birgitta Carlsson, Sweden Randi Eidsmo Reinertsen, Norway Ann-Mari Lindqvist, Sweden Ken Parsons, UK Håkan Nilsson, Sweden Traugott Zimmerli, Switzerland Désirée Gavhed, Sweden Peter Heffels, Germany Kalev Kuklane, Sweden Noëlle Valentin, France The organisers are indebted to the following sponsors of the conference: The Swedish Council for Work Life Research www.ralf.se Taiga AB www.taiga.se Arbesko AB www.arbesko.se Hestra-Handsken AB www.hestragloves.se Ullfrotté AB www.ullfrotte.se Kansas A/S www.kansas.dk III Table of contents Past, present and future trends in protective clothing 1 Traugott Zimmerli Integrated CAD for functional textiles and apparel 8 Yi Li, Edward Newton, Xiaonan Luo, Zhongxuan Luo Influence of air permeability on thermal and moisture transport through clothing 12 René Rossi, Markus Weder, René Gross, Friedrich Kausch New algorithms for prediction of wind effects on cold protective clothing 17 Håkan O. Nilsson, Hannu Anttonen, Ingvar Holmér Limitations of using a single-exponential equation for modelling clothing ventilation 21 Mark Bentley, Lisa M. Bouskill, George Havenith, Reginald W. Withey Effects of skin pressure by clothing on thermoregulation and digestive activity 25 Hiromi Tokura Ergonomics of protective clothing 26 George Havenith, Ronald Heus Application of the product planning chart in quality function deployment to improve the design of a fireman’s safety harness 30 Neil Parkin, Dave J. Stewardson, Michael Peel, Mike Dowson, Joe F. L. Chan An adaptive approach to the assessment of risk for workers wearing protective clothing in hot environments 34 Ken Parsons Radiation protective clothing in a hot environment and heat strain in men of different ages 38 Anna Marszalek, Maria Konarska, Juhani Smolander, Krzysztof Soltynski, Andrzej Sobolewski Management of Safety and Health Protection on building sites – under special consideration of use of personal protective equipment 41 Bernd Ziegenfuß, Nicola Klein Clothing trials as a part of worker training 44 Tanja Risikko, Juhani Hassi, Tiina M. Mäkinen, Liisa Toivonen Properties of foul weather clothing for construction workers after use 48 René Rossi, Markus Weder, Friedrich Kausch Physiological optimisation of protective clothing for users of hand held chain saws 53 Volkmar T. Bartels, Karl-Heinz Umbach The need for a rational choice of cold protective equipment in a refrigerated working environment 57 Shin-ichi Sawada Diversified design needs of personal protective devices and clothing in cold climate: An example in the design needs of protective outdoors winter shoes 62 John Abeysekera Footwear for cold work: a limited questionnaire survey 67 Kalev Kuklane, Désirée Gavhed, Eva Karlsson, Ingvar Holmér, John Abeysekera Footwear for cold work: a field study at a harbour 71 Kalev Kuklane, Désirée Gavhed, Eva Karlsson, Ingvar Holmér IV Footwear for cold work: a field study about work on high masts 75 Kalev Kuklane, Désirée Gavhed, Ingvar Holmér Innovations in fibres and textiles for protective clothing 79 Roshan Shishoo High visibility warning clothing 88 Doina Toma, Eftalea Carpus, Iuliana Cohea The effectiveness of phase change materials in outdoor clothing 90 Huensup Shim, Elizabeth A. McCullough Protective equipment against heat and/or fire produced from performant fibres 94 Doina Toma, Eftalea Carpus, Emilia Visileanu Dynamics of sweat vapour sorption as the function of physical parameters of textile packets under protective barrier 98 *UD*\QD %DUWNRZLDN Psycho-physiological mechanisms of thermal and moisture perceptions to the touch of knitted fabrics 102 Junyan Hu, Yi Li Combined effects of fabric moisture absorbancy and air permeability on thermophysiological responses in the warm environments 107 Hiromi Tokura Fibres, textiles and materials for future military protective clothing 108 Richard A. Scott Woven technical textiles for ballistic protection 114 Carmen Mihai, Eftalea Carpus, Emilia Visileanu, Doina Toma, Nicolae Scarlat, Mircea Milici Thermal protective textiles: Correlation between FR properties and static propensity 119 Jose A. Gonzalez, Martin W. King, Amit Dhir Testing and evaluation of electrostatic behaviour of electric inhomogeneous textiles with core- conductive fibers 123 Jürgen Haase, Christian Vogel Features of electric arc accidents in Finland 1996-1999 127 Sanna Mustonen, Helena Mäkinen Electric arc testing with heat flux measurement for FR clothing materials 131 Sanna Mustonen, Helena Mäkinen, Kalevi Nieminen Needs for research for protective clothing standards 135 Eero Korhonen A new structure of Ergonomic Standards for PPE – Proposal from Kommission Arbeitsschutz und Normung – KAN (Commission for Occupational Health and Safety and Standardization) 137 Dorit Zimmermann Main non-conformities of protective clothing detected in the Spanish market 141 Ignacio Cáceres, José Bahima, Eva Cohen Evaluating the cutting resistance of protective clothing materials 145 Jaime Lara, Serge Massé VTesting materials against small hot metal drops - Development of a new test method 150 Helena Mäkinen, Sanna Raivo, Sanna Karkkula, Erkki Rajamäki Revision of test methods: Better screening of PPE materials against liquid pesticides 154 Anugrah Shaw, Eva Cohen and Torsten Hinz A new British Standard: The assessment of heat strain for workers wearing personal protective equipment 159 Margaret Hanson Assessment of the scientific validity of ISO 7933/EN 12515 163 Robin Howie The influence of the number of thermal layers on the clothing insulation of a cold-protective ensemble 167 Désirée Gavhed, Kalev Kuklane, Ingvar Holmér Thermal insulation of multi-layer clothing ensembles measured on a thermal manikin and estimated by six individuals using the summation method in ISO 9920 171 Désirée Gavhed, Kalev Kuklane, Ingvar Holmér Effect of the number, thickness and washing of socks on the thermal insulation of feet 175 Kalev Kuklane, Désirée Gavhed, Ingvar Holmér Use of manikins in protective clothing evaluation Methods for cold protective clothing evaluation 179 Håkan O. Nilsson, Hannu Anttonen, Ingvar Holmér Research on typical medical work clothing on humans and on a thermal manikin 183 Krzysztof Soltynski, Maria Konarska, Jerzy Pyryt, Andrzej Sobolewski Comparative evaluation of the methods for determining thermal insulation of clothing ensemble on a manikin and person 188 Ralemma F. Afanasieva, Nina A. Bessonova, Olga V. Burmistrova, Vyacheslav M. Burmistrov, Ingvar Holmér, Kalev Kuklane Evaporative resistance of various clothing ensembles measured on standing and walking manikin 192 Krzysztof Blazejczyk, Ingvar Holmér Rain tightness of protective clothing – Prenormative interlaboratory tests using a manikin 196 Peter Heffels Development of the research and technology group flammability manikin systems 200 James D. Squire Hand protection Thermal properties of protective gloves measured with a sweating hand 204 Harriet Meinander Manual performance after gripping cold surfaces with and without gloves 208 Qiuqing Geng, Eva Karlsson, Ingvar Holmér Cold protective gloves in meat processing industry - product development and selection 212 Hannu Anttonen, Piritta Pietikäinen, Hannu Rintamäki and Sirkka Rissanen VI Protective gloves against mechanical and thermal risks 216 Doina Toma, Eftalea Carpus A case study on the selection and development of cut resistant protective gloves for household appliance assembly industries 218 Jaime Lara, Chantal Tellier Issues and challenges in chemical protective clothing 222 Jeffrey O. Stull Sweat effects on adsorptive capacity of carbon-containing flannel 226 Hubin Li, Jiangge Liu, Lei Li, Zhiqiang Luan Dynamic elongation test to evaluate the chemical resistance of protective clothing materials 230 Jaime Lara, Gérald Perron, Jacques E. Desnoyers Physiological strain and wear comfort while wearing a chemical protective suit with breathing apparatus inside and outside the suit in summer and in winter 235 Raija Ilmarinen, Harri Lindholm, Kari Koivistoinen, Petteri Helistén Performance criteria for PPE in agri- and horticulture 239 Torsten Hinz, Eberhardt Hoernicke Limits of recycling in protective apparel 243 Serhiy Zavadsky Protective clothing and survival at sea 245 Hilde Færevik Current and future standards of survival suits and diving suits 252 Arvid Påsche Heat preservation behavior of diving suit 255 Zhongxuan Luo, Edward Newton, Yi Li, Xiaonan Luo The effect of the distribution of insulation in immersion suits on thermal responses 259 Randi Eidsmo Reinertsen Lifevests - what is the value of the certification? 262 Arvid Påsche Pass/fail criteria to evaluate the strength of buoyancy aids (50 N) and lifejackets (100 N) in accordance to EN 393:1993, EN 395:1993 and the A1:1998 263 Hanna Koskinen, Raija Ilmarinen The effect of protective clothing on thermoneutral zone (TNZ) in man 267 Drude Markussen, Gro Ellen Øglænd, Hilde Færevik, Randi E. Reinertsen Passenger survival suits - a new emergency equipment 268 Arvid Påsche Protective clothing for firefighters Aspects of firefighter protective clothing selection 269 Mandy Stirling Investigating new developments in materials and design via statistically designed experiments 273 Dave J. Stewardson, Shirley Y. Coleman, John Douglass Design of UK firefighter clothing 277 Richard Graveling, Margaret Hanson VII Effects of clothing design on ventilation and evaporation of sweat 281 Emiel A. den Hartog Physiological load during tunnel rescue 285 Ulf Danielsson, Henri Leray Effectiveness of a light-weight ice-vest for body cooling in fire fighter’s work 289 Juhani Smolander, Kalev Kuklane, Désirée Gavhed, Håkan Nilsson, Eva Karlsson, Ingvar Holmér Fire fighter garment with non textile insulation 293 Michael Hocke, Lutz Strauss, Wolfgang Nocker Assessing fire protection afforded by a variety of fire-fighters hoods 296 James R. House, James D. Squire, Ron Staples Fire fighters’ views on ergonomic properties of their footwear 300 Helena Mäkinen, Susanna Mäki, José S. Solaz, Dave J. Stewardson Participant list 304 Author index 312 1Past, present and future trends in protective clothing Traugott Zimmerli EMPA Swiss Federal Laboratories for Materials Testing and Research, CH-9014 St.Gallen, Switzerland Introduction Protective clothing has a long history. If not already the fig-leafs of Adam and Eve, at least the armour of ancient warriors and the medieval knights may be designated the first real protective clothing. For the purpose of this paper, however, we will not look so far to the past but we will concentrate on the last few decades, the time period during which the most of the development of modern protective clothing has happened. Anyhow, due to space limitations, this review will be far from complete. An extensive compilation of the past development of protective clothing has been presented several years ago in an issue of Textile Progress (Bajaj, 1992). Protective clothing is used to achieve safety for people in professional and other sur- roundings. Safety is defined as ”Freedom from unacceptable risk of harm” (ISO, 1986). The measures to achieve safety can be divided into three levels: 1. First of all, processes, equipment and products have to be made safe, which means that they have to be conceived in such a way that any risk of harm is excluded or spatially separated from the people involved. 2. If for any reason persons nevertheless have to come near to the source of risk, they have to be protected by appropriate protective equipment. 3. A last mean to avoid people being exposed to a risk is to put a warning sign in front of the source of risk. From this concept we can see that the use of protective clothing is clearly not the first choice among the safety measures. However, it is nevertheless a very important measure and protective clothing of all kinds will in the future be of growing importance in the oc- cupational sector as well as in the field of leisure and sport. After some general remarks, the technical development and trends will be shown at the example of two different types of protective clothing. The thermophysiological comfort of protective clothing, which is a very important aspect, will also be highlighted. A look at the test methods, standards and market development will conclude this review. General remarks Looking at the general tendencies in the development of protective clothing, one can see that in the beginning people were looking at the protective properties of normal clothing, which they then tried to improve in one way or other. If for example they realised that a clothing material had good thermal insulation properties and therefore offered certain protection against heat, they used this material for the whole garment and, if necessary, in a greater thickness. At a later stage, specialised materials with optimised protective properties were developed and used for the manufacture of protective clothing. These were the so-called technical textiles. Later it was realised how strong the influence of the 2manufacturing on the protective properties of the ready-made garments is. Therefore, protective clothing is now developed more and more as a complete protective system, using modern materials, sometimes also the so-called intelligent materials. This trend will continue and even become stronger in the future. Technical development and trends Thermal protective clothing Protection against convective (flames), radiative and contact heat, against sparks and drops of molten metal, against severe cold and frost is a prime requirement of protective clothing in occupational, leisure and sports application. Key properties of materials used in this domain are thermal conductivity, flammability and heat resistance. To achieve a high insulation of textile materials, it was soon realised that it is necessary to develop bulky materials with much air enclosed and with low compressibility. For protection against heat radiation highly reflective outer materials were used. The necessary insula- tion is determined by the limiting heat flow that does just not create harm to the wearer of the clothing. In the case of heat protective clothing it is the tolerance of human tissue against burn injuries (Stoll & Chianta, 1969). Most of the performance requirements of heat protective clothing are based on these values. In order to reduce the flammability and increase the heat resistance – these two properties are strongly coupled – two different ways have been chosen. The first is to treat natural fibres, mainly wool and cellulosic fibres, but also not inherently flame- resistant man-made fibres chemically, in order to make them flame retardant. A lot of chemicals have been developed which fulfil this purpose. The other way is to develop inherently flame-resistant man-made fibres. In this field a lot of work has been done during the last decades: Aramid, PBI, Chlorofibres, carbon and mineral fibres to mention just a few. In the construction of protective clothing against heat and flame these chemi- cally treated or inherently flame-resistant fibres were used for the complete clothing or at least for the outer shell, depending of the severity if risk the user is exposed to. In the last years also thermophysiological aspects have more and more been taken into account (see next chapter). This means that ways have been sought to get rid of the sweat and exces- sive body heat without neglecting the protective demands against external heat. This can be done either by including special openings or by using new, specially designed mate- rials like phase change material (PCM) or materials with variable insulation or humidity absorbing capacity, so-called intelligent materials. Similar techniques apply for cold protective clothing with the exception that the use of heat resistant and flame-retardant materials is not necessary. The trend in the future will go into the direction of complete multifunctional protection systems using optimised manufacturing techniques and new, intelligent materials. High attention will be given to thermophysiological aspects and other use properties. Chemical protective clothing Modern technological developments have brought with them a multifold increase in the kinds of chemical hazards to which a worker is exposed. These hazards range from liquids (spray), gases to dust occurring in sectors like chemical, pharmaceutical, petro- chemical, electroplating industry and agriculture (fertilisers). They necessitate the wear- ing of clothing that is impermeable and resistant to chemicals, provides a tight seal against toxic gases, or filters dangerous dust. The big variety of hazardous chemicals and 3the manifold of the application of these chemicals seem to create the necessity to develop an unlimited set of chemical protective clothing. One very essential step in the past de- velopments was therefore to create a systematic classification of the necessary protection systems. The different ways of influence of the chemicals led to a series of about six types, some of them even subdivided, of protective clothing, ranging from totally encap- sulated, gas tight suits to clothing protecting only parts of the body. The selection of the materials used for the manufacture of the clothing depends on the chemicals against which it has to protect. Here too a systematic of chemicals was created which divided them into a series of classes of substances having similar penetration, permeation and degradation effects on the materials. For all these effects also relevant test methods with appropriate levels of performance were developed and standardised. As it was soon recognised that it was not possible to find one material, which protects against all chemi- cals, new, multilayered material combinations were created which could offer a broad range of protection. In the manufacturing of the clothing new joining and sealing tech- niques were developed. For totally encapsulated suits ventilation and cooling systems were worked out which were stationary or portable, depending on the use of the suit. A big problem that had and will still have to be solved is the decontamination and/or the disposal of used. The fact that in many cases the disposal is easier than the decontamina- tion led to the increasing use of single use disposable garments mostly made from non- wovens. The trend in chemical protective clothing in the future will certainly go in two direc- tions. One is the development of sophisticated, broad range protective systems using new high-tech materials and considering all aspects of use, wearing properties and decontami- nation. These products will be in the high-price segment. The other direction is the in- creasing use of cheap, single use, disposable garments for which the protective per- formance is optimised for one specific hazard situation. Thermophysiological aspects It has been highlighted already in different papers (Stull, 2000; Zimmerli, 1996 and 1998) that the thermophysiological properties of protective clothing are not, as the com- monly used term ”clothing comfort” would suggest, some kind of luxury, but that they have a very strong safety aspect. The higher the risk is, against which the clothing (e.g. fire fighters’ clothing, totally encapsulated chemical protective suit) has to protect, the less permeable for body heat and evaporated sweat is the clothing. By this fact and the amount of work the wearer has to perform, the risk of heat stress becomes very great. At the end, it doesn’t help the user that he wears a garment with a protection suitable for the external risks he is exposed, when the same garment on the other hand creates itself an internal risk of overheating the body by the excess of metabolic heat. The statistics of NFPA (Washburn et al, 1999) show that about 50 % of the fatal accidents of fire fighters in the USA are due to heat stress. So the modern design philosophy for protective cloth- ing is concentrated on optimising protection and simultaneously making the wearer more comfortable and more productive. The trend for the future in this respect will also go in the direction of developing complete protective systems, using new, “intelligent” mate- rials and considering the physiological aspects as much as the demands of protection. 4Test methods and standards There are different reasons why standardised test methods and performance requirements for protective clothing are necessary. The users of protective clothing need to be certain that they are sufficiently protected. The manufacturers want to show to the users that their product fulfils their needs of protection. And the test laboratories want to have approved and standardised test methods in order to get reproducible results and standardised per- formance requirements as a guideline for the certification of products. The International Organisation for Standardisation (ISO) started to develop standards in the field of protective clothing in 1964 in the Technical Committee (TC) 94 "Personal safety - Protective clothing and equipment". In 1966 the Subcommittee (SC) 11 "Protec- tive clothing against chemical products" and in 1968 SC 9 "Protective clothing against heat and fire" held their first meetings. In 1981 it was decided that SC 9 and SC 11 should be amalgamated to form the new SC 13 "Protective clothing”. At present SC 13 consists of 6 working groups (WG). When the European Community (EC) decided to establish the European common mar- ket by the end of 1992, the 'New approach' was formulated. The philosophy of this 'New approach' is that the EC does not establish detailed legislation on the rules for the com- mon market but it restricts itself on the edition of the so-called 'New approach' directives. All the details are then regulated in harmonised European standards, which ensure that the essential requirements of the directive are fulfilled. In the field of personal protective equipment (PPE) there exist two 'New approach' directives, one (EEC, 1989/2) for the manufacturing and another (EEC, 1989/1) for the use of PPE. As a consequence of the 'New approach', a mandate was given to the European Committee for Standardisation (CEN) by EG to establish harmonised European standards in the field of PPE. In 1989, among others, the CEN/TC 162 "Protective clothing in- cluding hand and arm protection and lifejackets" started to work. At present TC 162 has 12 WGs. In the last years efforts have been made to have identical standards in CEN and ISO. The "Vienna Agreement" between ISO and CEN, signed in 1991, is a tool to de- velop standards only once, to have parallel votes on identical documents in ISO and CEN and finally to have also identical standards. In the United States, the committee F23 of the American Society for Standardisation and Materials, ASTM develops standards on protective clothing since 1977. The work is done in 9 different SCs. The National Fire Protection Association (NFPA) writes per- formance standards for fire fighters’ clothing, based on test methods standardised by ASTM. A co-operation in the standardisation work between ISO, ASTM and CEN started several years ago. Since 1994, the chairmen of the relevant committees in the three organisations have regular meetings. However, it is much more difficult to have a similar co-operation as the one between CEN and ISO fixed in the Vienna agreement. This is due to different reasons. The main reason is that ISO and ASTM started their work independently and, as a consequence, two different sets of standards existed already when the co-operation started. A lot of laboratories had bought or built the test equipment based on their respective standards and it was unacceptable to either side to abandon its test methods and equipment. One way to overcome this problem has been tried in the ISO standard on protective clothing for fire fighters (ISO, 1999). Therein the CEN and NFPA standards are amalgamated, leaving to the user the choice, which test methods and cor- responding performance requirements he wants to select. Another reason to render the co-operation more difficult lies in the different working procedures and standard format of ASTM in comparison with CEN and ISO. But certainly these differences may be over- 5come and hopefully in the near future we will reach the goal to have only one set of stan- dards for protective clothing all over the world. The trend in the test methods followed the development of the protective clothing it- self. At the beginning mostly material tests were used and in some cases these methods were widened to make the assessment of the properties of seams, joints and closures pos- sible. The problem of standardised tests is always that they have, for the sake of repro- ducibility and repeatability, to be conceived so that the test conditions are far away from the conditions in real use (Zimmerli, 1996). In the last years more and more the under- standing came up, that the complete protective clothing has to be tested, either in a prac- tice test with test persons or with an instrumented manikin (Zimmerli, 2000). In addition, it will be necessary to assess the protective and the comfort properties simultaneously, because in most cases there is a strong interaction between both. Market development The following information on the market situation for protective clothing fabrics is based on a study which David Rigby Associates (DRA), Manchester, UK, made for the 1997 TechTextil Messe in Frankfurt. It shows that even on a relatively conservative definition, the European protective clothing market is substantial and continues to grow at an attrac- tive rate. DRA estimate that over 200 million m2 of fabric are consumed in Western Europe in the production of protective clothing, about 60 % thereof being nonwovens. Included in this figure are only the conventional types of protective clothing (against fire, heat, gas, chemicals, dust, particulate, NBC agents and extreme cold as well as high visi- bility clothing). Not considered in this compilation are the products for cut and abrasion protection (mostly gloves), for ballistic protection, foul weather clothing and protective garments for purely sporting applications. How these 200 million m2 are distributed among the major product functions and end-use segments is shown in Table 1. The medical sector represents by far the largest individual end-use sector in terms of fabric consumption. The trend in this domain goes in the direction of disposable non- woven products with high level of barrier performance. In the industrial sector there is in Europe a steady decline of people working in traditional manufacturing and heavy in- dustry and in addition a reduction of the exposure to risks at the workplace. On the other side there is an overall increase in the level of protection (more protective layers and/or higher performing products). Due to the more stringent regulations and the higher in- surance liability of the employers, protective clothing is more generally used. As a con- sequence the conventional protective clothing against fire, heat, stab and abrasion forms still a considerable part of the overall consumption and will be growing in the future. In order to have a general idea of the future development of the protective clothing market, DRA has made a forecast, the result of which is shown in Table 2. For the purpose of this forecast protective clothing has been more broadly defined, in- cluding foul weather clothing and protective garments for non-occupational use. Two main trends can be seen from this forecast. The first is that the nonwoven consumption is growing at a much higher rate than that of “conventional” fabrics (knits and wovens). The second is that, although the European market is still growing at an attractive rate, some product/market segments will reach saturation, whereas the market in developing countries will show a faster expansion. 6Table 1. Estimated West European Consumption of Fabric in Protective Clothing, 1996, (million m 2) (Davies, 1998). Product function End use Public utilities Military Medical Industry, construction, agriculture Total Flame retardant, High temperature Woven/knit Nonwoven 5 - 2 - - - 15 - 22 - Total 5 2 - 15 22 Dust and parti- culate (Barrier) Woven/knit Nonwoven - - - - 12 62 22 10 34 72 Total - - 74 32 106 Gas and chemical Woven/knit Nonwoven 1 3 1 - - - 4 47 6 50 Total 4 1 - 51 56 Nuclear, biological, chemical (NBC) Woven/knit Nonwoven - - 2 2 - - - - 2 2 Total - 4 - - 4 Extreme cold Woven/knit Nonwoven - - 1 - - - 2 - 3 - Total - 1 - 2 3 High Visibility Woven/knit Nonwoven 11 - 1 - - - 3 - 15 - Total 11 1 - 3 15 Totals Woven/knit Nonwoven 17 3 7 2 12 62 46 57 82 124 Total 20 9 74 103 206 Table 2. Forecast annual growth rates for ”Protective Clothing” consumption by product type, by region, 1995-2005, (% per annum, weight terms) (Davies, 1998). Knits/Wovens Nonwovens Total Western Europe 2.5 % 7.7 % 5.7 % North America 2.7 % 3.2 % 3.0 % Rest of World 5.1 % 14.4 % 10.0 % World Total 3.6 % 8.0 % 6.3 % Conclusions The development of protective clothing, as highlighted in this paper, showed a consider- able improvement over the last decades. From the use of normal clothing with some pro- tective properties until the conception of complex, multifunctional protection systems using sophisticated modern materials and manufacturing techniques was a long way to go. The variety of end use sectors has widened too in the course of time. Whereas in ear- lier times the occupational sector dominated, nowadays the use of protective clothing in the leisure and sports sectors has gained great importance. This is due to the fact that the modern society sets a high value on leisure activities and extreme sports. On the technical side, the requirements protective clothing has to fulfil have become much more complex. When in the beginning only the protective function was dominat- ing, today the combination of different, partly contradictory requirements like protection, comfort, fashion and other functional properties makes the development of a modern protective clothing a complex task. But, seeing which improvement of safety, health and quality of life it can produce, it is also a highly satisfying task. 7References Bajaj P & Sengupta AK (1992) Protective Clothing. Textile Progress, 22 (2/3/4). EEC (1989/1) COUNCIL DIRECTIVE of 30 November 1989 on the minimum requirements for safety and healthcare at the use of personal protective equipment by the employees at work (89/656/EEC). Official Journal of the European Communities, No L 393/18, 30.12.89. EEC (1989/2) COUNCIL DIRECTIVE of 21 December 1989 on the approximation of the laws of the Member States relating to personal protective equipment (89/686/EEC). Official Journal of the European Communities, No L 399/18, 30.12.89. Davies B (1998) Growth prospects in the protective clothing market. ITB Nonwovens – Industrial Textiles, (3), pp. 10-12. ISO (1986) ISO/IEC Guide 2:1986, Definition 2.5. International Organisation of Standardisation, Geneva, Switzerland. ISO (1999) ISO 11613:1999, Protective clothing for fire fighters – Laboratory test methods and performance requirements. International Organisation of Standardisation, Geneva, Switzerland. Stoll AM & Chianta MA (1969) Aerospace Medicine, 41, pp. 1232-1238. Stull J (2000) Cooler fabrics for Protective Apparel. Industrial Fabric Products Review, March, pp 62-68. Washburn AE, LeBlanc PR & Fahy RF (1999) Fire fighters’ Fatalities. NFPA Journal, July/August, pp. 55-70. Zimmerli, T (1996) Standardised and practice oriented tests; comfort and protection of clothing: Two contradictions in one? Environmental Ergonomics, Recent Progress and New Frontiers, pp. 369- 372, Y. Shapiro, D.S. Moran and Y. Epstein, Eds., Freud Publishing House, Ltd., London and Tel Aviv. Zimmerli T (1998), Schutz und Komfort von Feuerwehrbekleidung (Protection and Comfort of Fire- fighters’ clothing). Textilveredlung 33 (3/4), pp. 52-56. Zimmerli, T (2000) Manikin Testing of Protective Clothing - A Survey. Performance of Protective Clothing: Issues and Priorities for the 21st Century: Seventh Volume, ASTM STP 1386, C. N. Nelson and N. W. Henry, Eds., American Society for Testing and Materials, West Conshohocken, PA. 8Integrated CAD for functional textiles and apparel Yi Li, Edward Newton, Xiaonan Luo, Zhongxuan Luo Institute of Textiles and Clothing, The Hong Kong Polytechnic University, Hung Hom, Kowloon, Hong Kong Clothing requirements of modern consumers Extensive consumer research has shown that modern consumers require clothing to not only look good, but also feel good in dynamic wear situations. The comfort and superior functional performance of clothing have been identified as the most important attributes demanded by modern consumers, especially under dynamic wear situations (Figure 1). It has been noted that sports buffs are focusing on functional products and classic style as fashion is now of secondary importance. A recent survey in the US showed that 81 % of US consumers signalled comfort as their top choice (Hong Kong TDC, 1999). In China, consumers ranked comfort in the top three most important attributes of apparel product. Therefore, comfort and functional performance have become a major focal point for manufacturers to gain competitive advantages in global apparel markets. Over years of research, it has been found that clothing comfort consists of three major sensory factors: thermal-moisture comfort, tactile comfort and pressure comfort, as shown in Figure 2. The three sensory factors contribute up to 90 % of overall comfort perceptions, and the relative importance of individual factors varies with different wear conditions. For active sportswear, thermal-moisture comfort is the most important factor, followed by tactile comfort and pressure comfort. Thermal-moisture comfort is deter- mined by the heat and moisture transfer behaviour of clothing during dynamic interac- tions with human body and external environment. Tactile and pressure comfort is related to the mechanical behaviour of clothing during wear. Therefore, heat and moisture trans- fer and the mechanical behaviour of clothing materials are the two major dimensions in determining the comfort and functional performance of apparel products. Price 0 10 20 30 40 50 60 70 80 Australian Asian European Comfort fit Style Quality Color Easy care Fabric Brand Consumer Requirements Sensory Perceptions Tactile Thermal- wet Pressure PrickleItch RoughScratch Cla mm y Damp Hot Co ld Sticky Stiff SoftSnug SmoothLoose Figure 1. Clothing consumption requirements of modern consumers. Figure 2. Sensory comfort of apparel products. Computer technology has successfully been used in the textile and apparel industries, and CAD techniques are widely used for fashion and textile design. The main purpose behind the utilization of CAD is to increase productivity and flexibility during clothing 9fashion design process. As modern consumers demand personal comfort, CAD for fashion design alone cannot satisfy the needs of manufacturers to develop functional and comfortable products that can meet the requirements of consumers. However, CAD for clothing functional design has not been developed and applied in fashion industry. One of the major reasons is that the heat and moisture transfer and the mechanical behavior of textiles and clothing are extremely complex. Sound scientific understanding and mathe- matical simulation of the coupled heat and moisture and fabric mechanical behavior are essential requirements for developing CAD technologies for the functional design of ap- parel and textiles. CAD for fashion design Obviously, fashionable outlook of clothing is a major attribute that influences the psy- chological comfort and satisfaction, as well as the purchase decision of consumers. There are a number of dimensions in fashion design such as colour, texture, pattern, drape (or appearance) style and fit. Colour, texture and pattern are important components of artistic creativity during design processes, which have been enhanced successfully by CAD technology for textile design and directly linked to printing and dyeing processes. Com- mercial technological packages including software and hardware have been developed and applied successfully in fashion industry. Fabric drape is more difficult to be simu- lated and visualized by computing technology alone, as it is determined largely by the mechanical behaviour of clothing materials and its dynamic interaction with the body and external mechanical forces such as air movement. There is some CAD packages provid- ing artificial simulations by computing image manipulations without considering the me- chanical behaviour of clothing materials. Extensive research activities have been carried out around the world to develop numerical simulation of the drape effect on basis of fab- ric mechanics. Fashion Design Thermal Analysis and Functional Design Thermal Functional Analysis Thermal Functional design Figure 3. 3D CAD technology for fashion design. Figure 4. CAD technology for thermal functional design. The effect of style and fit is related to body size and shape, 2D fabric cuttings and 3D wrapping to human body, as well as the mechanical behaviour of clothing materials. To simulate and visualize the 3D effect, we need measuring body size and generate 3D geo- metric body shape (i.e. numerical geometric human model), on which 2D fabric cuttings determined by style and fit can be wrapped. By adding on the effects of colour, pattern, texture and drape, designers and/or consumers are able to view the artistic and fashion- able effect. Extensive R&D activities have been in both academic institutions and com- mercial organizations to develop such technology. 10 CAD for thermal functional design On the basis of the numerical geometric human model, a model simulating the thermo- regulation of human body (i.e. numerical thermal human model) needs to be developed. The numerical thermal human model will be integrated with the model of heat and moisture transfer in clothing materials and in the external environment to simulate the heat and moisture generation and transfer processes of the body-clothing-environment system as the basis of thermal functional design. Using such a numerical simulation system, we are able to investigate the influence of fibers, fabrics, clothing, the physical activities of the body and external environment on the thermal comfort and functional performance, as shown in Table 1. The mathematical models developed and improved by various researchers such as Henry (1939) and Farnworth (1986) to describe the complex coupled heat and moisture transfer in textiles have laid a sound scientific basis to achieve this goal. For instance, Li and Holcombe (1998) interfaced a fabric heat and moisture transfer model with Gagge's two-node thermo-regulatory model of the body to investigate the impact of fiber hygroscopicity on the dynamic thermoregulatory responses of the body during exercise and on protection of the body against rain. Table 1. Input and Output variables in thermal functional design Input variables Output variables x Fiber structural and properties, such as fiber diameter, fiber density, moisture sorption isotherm, heat of sorp- tion, and water diffusion coefficient, specific heat; x Fabric structural and thermal properties, such as thick- ness, porosity, tortuosity, thermal conductivity and volumetric thermal capacity; x Skin thermal properties: thickness, thermal conduc- tivity, water diffusion coefficient, volumetric thermal capacity; x Ambient boundary conditions: temperature, relative humidity and air velocity; x Style and fit of apparel products. x profile of temperature in the fabric; x profile of moisture content of fibers; x profile of moisture in the air of the fabric void space; x profile of temperature at the skin surface; x the neurophysiological responses of thermal receptors in the skin; x Intensity of subjective perception of thermal and moisture sensations. CAD for mechanical functional design On the basis of the numerical geometric human model, a model simulating the biome- chanical behaviour of human body (i.e. numerical mechanical human model) needs to be developed. The numerical mechanical human model will be integrated with the model of fabric mechanics to simulate the dynamic mechanical interactions between the body and clothing. Using such clothed numerical mechanical human model, similar to thermal function design, we are able to study the effect of structural and mechanical properties of fibers, yarns and fabrics, and clothing style and fit on the mechanical comfort and func- tional performance of apparel products (Figure 5). The extensive research on modeling fabric mechanics in the last century has laid down a sound scientific knowledge founda- tion to achieve this aim. For example, Zhang and Li et al (1999) studied the physical mechanisms of woven fabric bagging and developed mathematical simulation of fabric bagging behavior. During bagging, fabrics are exposed to sophisticated multi-dimen- sional deformation inserted by the contact force from human body parts such as the knees. The understanding of physical mechanisms and modeling methodology of fabric bagging can be applied to simulate the mechanical behavior of garments and mechanical comfort of the wearer by modifying the boundary conditions and specifying different fi- ber mechanical properties and fabric structural characteristics. 11 Mechanical Analysis and Functional Design Mechanical Functional Analysis Mechanical Functional design Body Model Mechanical Model Body Measurement Fashion Thermal DesignThermal Model Mechanical design Integrated Design Figure 5. CAD technology for mechanical functional design. Figure 6. Integrated CAD technology for design of functional apparel products. Integrated CAD technology The fundamental research in modelling and simulating the heat and moisture in textiles and fabric mechanics has establish a good foundation to develop integrated CAD, which is able to introduce science into the apparel design process. By integrating the CAD tech- nologies for fashion design, thermal functional design and mechanical functional design, we are able to reveal the outlook, the comfort and functional of clothing before it is actu- ally made, as shown in Figure 6. Using the mathematical models with advanced compu- tational techniques, we are able to simulate the dynamic heat and moisture transfer processes from the human body and clothing to the environment, and the dynamic me- chanical interaction between the body and clothing. The simulation results can be visualized and characterized to show the dynamic temperature and moisture distribution profiles in human body, clothing and environment and stress distributions in clothing and on the body. Thus, we are able to demonstrate how changes in physical activities, envi- ronmental conditions and/or different design of clothing will influence the thermal and mechanical comfort of the wearer. Therefore, on the basis of the scientific mathematical models we can develop integrated CAD technologies that are workable as advanced en- gineering design tool for textile and clothing industry. Acknowledgement We would like to thank The Hong Kong Polytechnic University for the funding of this research through the Area of Strategic Development in Apparel Product Development and Marketing. References Hong Kong T.D.C. (1999) German Sporting Goods Market. International Marketing News, 15(10): p. 3. Henry PSH (1939) The Diffusion in Absorbing Media, Proc. Roy. Soc. 171A, 215-241. Farnworth B (1986) A Numerical Model of the Combined Diffusion of Heat and Water Vapor Through Clothing, Textile Res. J., 56, 653-665. Li Y & Holcombe BV (1998) Mathematical Simulation of Heat and Mass Transfer in a Human-Clothing -Environment, Text. Res. J., Vol.67 (5), pp389-397 Zhang X, Li Y, Yeung KW, Yao M (1999) Mathematical Simulation of Fabric Bagging. Textile Res. J., (accepted for publication), 1999. 12 Influence of air permeability on thermal and moisture transport through clothing René Rossi, Markus Weder, René Gross, Friedrich Kausch EMPA St.Gallen, Lerchenfeldstrasse 5, CH-9014 St.Gallen, Switzerland Introduction The physiologic properties of clothing are usually assessed under well-defined condi- tions. In practice however, the climatic conditions can change quite rapidly and influence the insulation of the clothing. Depending on the air permeability of the clothing, wind will more or less go through the textile layers and favour the release of heat and the evaporation of moisture. This effect of wind will change the range of use of clothing and can therefore cause a hypothermia of the body. The goal of this study was to analyse the effect of wind on the thermal insulation and the water vapour permeability of ready-made, single-layered garments. For the measure- ments, a “sweating arm” was used. Identical sleeves were made from fabrics with different air permeability, put on the sweating arm and exposed to different climatic conditions (variable temperatures, relative humidity and wind speeds). The heat loss was assessed either in dry conditions or with release of moisture and correlated to the air permeability of the fabrics. Methods Test apparatus For this study, the sweating arm (Weder et al., 1996) was used, that corresponds from its dimensions and geometry to a man’s arm. It is usually heated up to 35 °C, corresponding to skin temperature and releases moisture in vaporous form. As the measurements are made under non-isothermal conditions, the effects of condensation in the textile layers can be assessed. The apparatus is divided into five parts, which can be heated up to de- fined temperatures (usually 35 °C). Additionally, the forearm and the upper arm can "sweat", that is they are equipped with different nozzles, which can release as much water as a human being would. The humidity can be released in liquid or vaporous form. The water supply is regulated through two pumps for the two parts of the arm. The supplied water is absorbed by a cotton fabric and distributed homogeneously over the whole surface. A cellophane foil is placed as outermost layer to avoid that liquid water comes into contact with the fabric sample and to obtain a moisture release only in va- porous form. In order to avoid the loss of heat through conduction, the sweating arm is protected by two guards placed at both ends of the arm. The guards can be heated up to the same tem- perature as the arm to avoid heat exchanges by conduction. The effect of wind is simu- lated by a fan. 4 different wind velocities (1, 2, 4.4 and 13.3 m/s) and 3 different outside temperatures (5, 15 and 20 °C) were chosen to analyse the effect of wind on thermal and mass transfer, and the influence of condensation in the sleeves. The sleeves were closed at both ends to avoid any transfer through the openings. 13 Samples Identical sleeves were made of seven different polyester fleece fabrics. The air perme- ability of the fabrics was determined according to EN ISO 9237 and is shown in Table 1. Table 1. Air permeability of the samples. Sample IA IB IIA IIB IIIA IIIB IIIC Air perm. in l/m2s 225 > 950 252 647 149 0 250 Results and discussion Thermal resistance Depending on the air permeability of a fabric, wind will change the convective heat flow around the sample, but also penetrate the fabric and disturb the still air layers in the microclimate between the arm and the sample. The influence of wind on a cylindrical sample is not the same as on a flat one as wind can pass round the same in the first case but must pass through the fabric in the other. Furthermore, the heat loss is not symmetri- cal: while it is very dependent on the wind speed on the windward side, heat loss on the lee (opposite) side is relatively low and insensitive to the wind (Kind et al., 2000). Rct*1000 in m2K/W 0 100 200 300 400 500 600 700 0 5 10 15 Wind speed in m/s IA IB IIA IIB IIIA IIIB IIIC Water vapour permeability in g/m2h 0 100 200 300 400 500 600 700 800 900 0 1 2 3 4 5 Wind speed in m/s Figure 1. Thermal resistance at different Figure 2. Water vapour permeability at wind speeds. different wind speeds. The reduction of insulation of the different samples is dependent on their air per- meability (Figure 1). Sample IIIB with an air permeability of 0 l/m2s has the lowest re- duction in thermal resistance whereas samples IB and IIB have the highest. Sample IIB has only about 10 % insulation left at 13.3 m/s compared with 1 m/s. The reduction of the thermal resistance with the wind speed is nearly linear for the samples with air per- meabilities between 150 and 250 l/m2s (IA, IIA, IIIA and IIIC). 14 Water vapour permeability The moisture transport behaviour of the different sleeves was similar to the one of ther- mal transfer (Figure 2). The samples with the higher air permeability had the highest in- crease of water vapour permeability with increasing wind speed. Stuart et al. (1983) de- veloped a model that describes the heat and moisture loss from a cylinder as depending linearly on the air permeability and on the second power of the wind velocity. Lamb et al. (1990, 1992) used a more complex model that depended, among others, on material pa- rameters as thickness and thermal conductivity as well as physical constants of the air as the viscosity or the density. The results of the present study (Figure 2) showed that the relationship between water vapour permeability and wind speed can be approximated quite well with a polynom of second order for low to moderate wind speeds (<10 m/s). For higher wind speeds, the increase slowed down for the samples with high air permeability (IB and IIB) as the sweat release of the sweating arm could not rise indefinitely. The relationship between the moisture transfer (as well as the heat transfer) and the air permeability can very roughly be approximated by a linear curve (Figure 3) as suggested by Stuart et al. (1983), although an equation of the following type gives more precise results: n)AP(2k1kWVP  with n between 0 and 1 (k1, k2: constants) (Equation 1) (WVP: Water vapour permeability, AP: Air permeability) 0 200 400 600 800 1000 1200 0 200 400 600 800 1000 Air permeability in l/m2s W at er v ap ou r pe rm ea bi lit y in g /m 2h 1 m/s 2 m/s 4.4 m/s 13.3 m/sWind speed: y = c1 + c2*(AP)0.49 R =0.92 y =d1 + d2*(AP)0.003 R =0.66 y = a1 + a2*(AP)0.26 R =0.99 y = b1 + b2*(AP)0.73 R =0.98 Figure 3. Water vapour permeability in dependence of the air permeability (measurements at 20 °C, 65 % RH). The results of water vapour permeability were nearly identical for the three outside conditions (20, 15 and 5 °C), although the amount of condensation was logically much higher for the tests at 5 °C (over 100 g/m2h) for some samples) than for the other two at- mospheric conditions. The permeability of the samples was therefore not disturbed by the moisture uptake of the materials for the whole duration of the test (40 minutes). It can 15 nevertheless be foreseen that with a longer test duration, the condensed moisture would create a barrier to humidity. Influence of air layers within the clothing system The correlation between a standardised test to determine the water vapour resistance (sweating guarded hot plate according to EN 31092) and the sweating arm were assessed for different fabrics for fire fighter’s clothing (Rossi, 1999). The measurements on the upper arm alone (without air layers) were made under the same climatic conditions as required by EN 31092 (35 °C, 40 % RH) but with a wind speed of 2 m/s instead of 1 m/s with the sweating guarded hot plate. 0.00 10.00 20.00 30.00 40.00 50.00 60.00 70.00 0 10 20 30 40 50 Ret (in m2Pa/W) [EN 31092] W at er v ap ou r re sis ta nc e (in m 2P a/W ) [ sw ea tin g a rm ] . WVR upper arm WVR whole arm Figure 4. Water vapour resistance (WVR) on the sweating arm compared to the standardised test according to EN 31092. There is a good correlation between the upper arm measurement and the standardised test (Figure 4, correlation coefficient R = 0.97). The values on the upper arm are lower than the standardised test, confirming previous findings (Meinander, 1985). This is due to a possible chimney effect and a larger outer surface of the sample onto the cylinder. The results are less differentiated with the upper arm (between 7 and 21 m2Pa/W) than with the plate (between 4 and 36 m2Pa/W). This could be due to the fact that the still air layer is completely removed on the plate whereas the wind flow passing round the upper arm will not ideally remove the air layer as the wind was only generated from one direction. The relationship between the water vapour resistance on the plate and on the whole arm seems not to be linear any more and shows an exponential tendency. A possible explanation of this outcome is that the measurements on the whole arm were made under non-isothermal conditions (outside climate: 20 °C and 65 % RH). When the resistance of the material increases, there will be more and more condensation during the non-isother- mal tests and the moisture amount released to the environment may decrease. 16 Conclusions In this study, a relationship between the air permeability and thermal and water vapour resistances could be established. Nevertheless, the dependency of the water vapour per- meability with the air permeability is not generally valid. Especially if several windtight materials were used, the results would have been totally different. It has also been shown that the relationship between the water vapour resistances measured with air layers (sweating arm) and without air layers (only upper arm or sweating guarded hot plate) is not linear, even if the making of the sleeves is identical. Under identical atmospheric conditions, however, the results on a cylindrical apparatus (upper arm of the sweating arm) correlate very well with the results on a flat plate (sweating guarded hot plate). References EN 31092 (1993) Textiles – Physiological effects – Measurement of thermal and water vapour resistance under steady-state conditions (sweating guarded-hotplate test) [European Standard]. Brussells: Comité Européen de Normalisation EN ISO 9237 (1995) Textiles – Determination of permeability of fabrics to air [European Standard]. Brussells: Comité Européen de Normalisation Kind R. J. and Broughton C. A. (2000), Reducing Wind-Induced Heat Loss Through Multilayer Clothing Systems by Means of a Bypass Layer. Textile Res. J. 70(2), 171-176 Lamb G. E: R. and Yoneda M. (1990), Heat Loss from a Ventilated Clothed Body, Textile Res. J. 60, 378- 383 Lamb G. E. R: (1992), Heat and Water Vapor Transport in Fabrics Under Ventilated Conditions, Textile Res. J. 62(7), 387-392 Meinander H. (1985), Introduction of a new test method for measuring heat and moisture transmission through clothing materials and its application on winter work wear, Technical Research Centre of Finland VTT Stuart I. M. and Denby E. F. (1983), Wind Induced Transfer of Water Vapor and Heat Through Clothing, Textile Res. J. 53, 655-660 Rossi R. (1999), FOKUS – Research Project Comfort and Protective Clothing, EMPA Report No. 243 Weder, M. S, Zimmerli, T., and Rossi, R. M (1996), A Sweating and Moving Arm for the Measurement of Thermal Insulation and Water Vapour Resistance of Clothing, Performance of Protective Clothing: Fifth Volume, ASTM STP 1237, James S. Johnson and S. Z. Mansdorf, Eds., American Society for Testing and Materials 17 New algorithms for prediction of wind effects on cold protective clothing Håkan O. Nilsson1, Hannu Anttonen2, Ingvar Holmér1 1 Programme for Respiratory Health and Climate, National Institute for Working Life, S - 171 84 Solna, Sweden 2 Oulu Regional Institute of Occupational Health, Aapistie 1, F - 902 20 Oulu, Finland Introduction The European Pre-Standard for protective clothing against cold ENV 342:1998 suggests that the testing of thermal insulation should be made on a moving thermal manikin. In this way the value can be used to match requirements specified by the IREQ-method (ISO/TR-11079, 1993) or as realistic input for prediction of thermal stress in other stan- dards. It is known from prior work (Olesen et al, 1982; McCullough & Hong 1992; Nilsson et al, 1997) that the clothing insulation can be reduced by wind and body move- ments by up to 70 % from the value measured on a standing thermal manikin. This paper will focus on the principals of reduction made on total insulation calculated according to methods presented in ENV 342:1998. Materials and Methods The thermal manikin used is one in the TORE-series that has been described earlier (Hänel, 1983; Nilsson et al, 1992). The power transmission, in the walking apparatus, has been made with pneumatic cylinders, which gives a simple and durable construction with a minimum of mechanical components. TORE was positioned in the controlled environment of the climatic chamber until steady state was reached. Then the insulation was calculated from the measured heat loss. In this investigation 10 types of cold protective working clothes with total insulation values from 2.24 to 4.61 clo with an air permeability between 1 and 1000 l/m2s. Figure 1. The moving thermal manikin TORE, measuring according to ENV 342:1998. 18 The walking speed of the manikin was set to 0 to 1.2 m/s and the wind speed was set in six levels from 0.4 to 18 m/s. The repeatability for the method used for determination of insulation values was high, the difference between double determinations was less than 5 % of the mean value of the two measurements based on 228 independent measurements. Results The results are given as percentage of the total insulation (It) measured with a standing manikin during wind still conditions (0.3 to 0.5 m/s, ENV 342:1998). Calculations where also made of the intrinsic value (Icl) where the insulation of the air layer was subtracted as recommended in the standard without correction for the increased surface area. 0 20 40 60 80 100 0 3 6 9 12 15 18 Insulation reduction standing Undressed 0.55 clo, O layers Suite 1.49 clo, 2 layers, <1000 l/m2s Overall w. underw. 1.53 clo, 2 layers, <1000 l/m2s Standard (prEN 342) 1.98 clo, 2 layers, <1000 l/m2s Wool permeable 2.84 clo, 3 layers, <1000 l/m2s Synthetic 2.43 clo, 3 layers, <50 l/m2s Wool 2.58 clo, 3 layers, <50 l/m2s Cotton 2.64 clo, 3 layers, <50 l/m2s Basic 2.69 clo, 3 layers, <50 l/m2s Wool thick 3.46 clo, 3 layers, <50 l/m2s Gore Tex 2.90 clo, 3 layers, <1 l/m2s Low, 1 l/m2s Medium, 50 l/m2s High, 1000 l/m2s Wind speed (m/s) Figure 2. The combined effect of air permeability, wind on TORE standing still. The relationship with wind and walk influence as well as different clothing air perme- ability have been examined. The insulation reduction (It,r /It) as a function of air perme- ability (p, l/m2s), wind speed (v, m/s) and walking speed (w, m/s) is now calculated with: I t, r I t  0.54  e  0.15 ˜ v  0.22 ˜ w  p0.075  0.06  ln(p)  0.5 (1) The equation is derived from three dependent regressions, one for wind and walk (R = 0.885) and one for the inclination of the permeability (R = 0.965) and one for the inter- cept of the permeability (R = 0.998). The standard deviation of the difference between measured and calculated data (It,r /It) is 4 % (Max/Mean/Min 15/5/0) based on all 228 independent data sets. The validity interval for the equation is 0.4 - 18 m/s wind speed, 0 - 1.2 m/s walking speed and an air permeability of 1 to 1000 l/m2s. The equation is plot- ted in Figure 3 for low air permeability - 1 l/m2s (Figure 3a) as well as high air perme- ability - 1000 l/m2s (Figure 3b). Discussion The clothing insulation is strongly affected by wind and movements. A reduction of the clothing insulation measured with a static thermal manikin is consequently needed. The 19 insulation is reduced exponentially with increased step frequency (walking speed) and increased wind speed (Nilsson & Holmér, 1997). 0 2 4 6 8 1012141618 0.00 0.30 0.60 0.90 1.20 0 20 40 60 80 100 % of total standard insulation wind speed (m/s) walking speed (m/s) Low permeability 80-100 60-80 40-60 20-40 0-20 0 2 4 6 8 1012141618 0.00 0.30 0.60 0.90 1.20 0 20 40 60 80 100 % of total standard insulation wind speed (m/s) walking speed (m/s) High permeability 80-100 60-80 40-60 20-40 0-20 Figure 3a - b. The combined effect of wind and walking speed for TORE while walking at 0 to 1.2 m/s with wind speed from 0.4 to 18 m/s. For clothing combinations with 2-3 layers with a total insulation of 1.49 - 3.46 clo and an air permeability of 1 l/m2s (left) and 1000 l/m2s (right). In Figure 2 are three lines shown for high, medium and low air permeability respec- tively. In our study 8 out of 10 ensembles fitted into these categories, as most winter clothing do. It is also clearly seen in Figure 2 that that the air permeability has little influence on the insulation for wind speeds below 2 m/s. For calculations below such a limit the formula then could be reduced, by insertion of i.e. p = 1 in equation 1, to only take wind and walk into consideration. I t, r I t  0.54  e 0.15˜ v0.22˜w  0.5 (v  2m / s) (2) Conclusions A general equation that takes into account the insulation reduction effects from wind, permeability and walk has been developed. The equation makes it possible to calculate more realistic values for the actual insulation during different activity and weather condi- tions for most winter work clothing, if the static clothing insulation is known from meas- urements or tables. The air permeability has little influence on the insulation for wind speed below 2 m/s. For calculations below such a limit the air permeability could be omitted. In the future only measurements on standing manikin should be needed. The wind, permeability and walk reductions will be calculated from this value. To validate the relationships more measurements on subjects exposed to wind and ac- tivity in working life are needed. Acknowledgements The Swedish Council for Work Life Research and Finnish Work Environment Fund has supported this work. 20 References ENV 342:1998 (1998) Protective clothing against cold. Comité Européen de Normalisation, Brussells. ISO/TR-11079 (1993) Evaluation of cold environments - Determination of required clothing insulation (IREQ). International Standards Organisation, Geneva. Hänel S-E (1983) A joint Nordic project to develop an improved thermal manikin for modelling and measuring human heat exchange. In: Aspect médicaux et biphysiques des vêtements de protection. Lyon: Centre de Recherche du Service de Santé des Armées, 280-282. McCullough EA & Hong S (1992) A data base for determining the effect of walking on clothing insulation. Proceedings of the Fifth International Conference on Environmental Ergonomics, Maastricht, the Netherlands, 68-69. Nilsson HO, Gavhed DCE, Holmér I (1992) Effect of step rate on clothing insulation. – Measurement with a moveable thermal manikin. In: Lotens W, Havenith G (ed.), Environmental Ergonomics V, Masstricht, Netherlands. Nilsson H & Holmér I (1997) Development of Clothing Measurement Methods with the Thermal Manikin TORE. Proceedings of 5th Scandinavian Symposium on Protective Clothing (NOKOBETEF V). Helsingør, Denmark. 30-35. Nilsson H, Holmér I, Ohlsson G & Anttonen H (1997) Clothing insulation at high wind speeds. Proceedings of Problems with Cold Work, Holmér I & Kuklane K (ed.), Arbete och Hälsa 1998:18, 114-117. Olesen BW, Sliwinska E, Madsen TL & Fanger PO (1982) Effect of body posture and activity on the thermal insulation of clothing. Measurement by a movable thermal manikin. ASHRAE Transcations, 88:2, 791-805. 21 Limitations of using a single-exponential equation for modelling clothing ventilation Mark Bentley1, Lisa M. Bouskill2, George Havenith2, Reginald W. Withey1 1 Centre for Human Sciences, DERA, Farnborough GU14 0LX, UK 2 Department of Human Sciences, Loughborough University, LE11 3TU, UK Introduction The thermal load that a clothing ensemble might impose can be calculated from its ther- mal and evaporative resistances, traditionally measured using a static thermal manikin. However, physical activity of the wearer and increased environmental airspeed greatly reduce these resistances - by disrupting the boundary air layer and by increasing ex- change between air trapped within the clothing and the surrounding air (clothing ventila- tion). Clearly, if the effect of clothing on human heat balance is to be understood fully, it is essential that these influences be quantified. Many equations have been developed to describe the disruption of the boundary air layer, all are of the same form (see, for example, Winslow et al., 1939) but the heat ex- change by clothing ventilation is more complex. Nilsson et al. (1992) used an articulated, thermal manikin to derive empirical corrections to static manikin thermal resistance to take account of the clothing ventilation. Havenith et al. (1990) extended these corrections to include evaporative resistance. However, corrections based on manikin measurements clearly have a practical limit – manikins cannot replicate the complex movement of hu- mans in the work-place. A more general approach would be to measure clothing ventila- tion when the clothing ensemble is worn in the work-place by human test subjects, and use the clothing ventilation data to calculate its effect on heat balance. Clothing ventilation is the product of the rate of air exchange (ie the number of times per unit time that a volume of air in the clothing equal to the clothing microenvironment volume is exchanged with ambient air) and the clothing microenvironment volume. A method for measuring the rate of exchange between air trapped within the clothing and surrounding air was developed by Crockford et al. (1972). The method has two stages: First, record the dilution curves obtained by flushing the clothing with a tracer gas (nitro- gen), then monitoring the concentration of oxygen as it returns to the clothing. Second, calculate the rate constant of the dilution curve by fitting a single exponential. The rate constant can be shown to be equal to the rate of air exchange. Birnbaum & Crockford (1978) developed a method to measure the clothing micro- environment volume. They multiplied this value (VT) by the measured rate of air ex- change (r) to calculate what they called the ‘ventilation index’. However, this quantity is not an index, so in this paper we refer to it as clothing ventilation. 22 The Problem Crockford et al. (1972) used a single exponential equation (Equation 1) to calculate the clothing ventilation from the oxygen dilution curves:   rt air epptp   1 where: r is the rate of air exchange (min-1) p(t) is the concentration of oxygen in the clothing microenvironment (%) p1 is such that pair-p1 is the initial concentration of oxygen in the clothing microenvironment at t=0 (%) p air is the concentration of oxygen in the surrounding air (%) Equation 1. Single exponential model of clothing ventilation. The single exponential equation of clothing ventilation is based on a single compart- ment model of a clothing ensemble (Figure 1), which makes several assumptions, that: 1. there is complete mixing within the clothing ensemble so that the concentration of oxygen in the clothing microenvironment is homogeneous; 2. the rate of air exchange is constant; 3. the volume of the clothing microenvironment is constant; 4. the air (and oxygen) in the clothing microenvironment do not permeate the skin. where: CLV x is the clothing ventilation (l.min-1) VT is the volume of the clothing microenvironment in clothing layer 1 (l) Figure 1. 1-layer model of clothing ventilation. It is not certain that these assumptions are valid. For example, Lotens & Havenith (1988) showed that higher-order exponentials could be a better fit, because the con- centration of oxygen is not homogeneous - there are pockets of trapped nitrogen, regions of slow air exchange, and clothing layers that restrict gas diffusion. The purpose of this investigation was to determine the magnitude of any errors introduced into the calcula- tion of clothing ventilation when assumption 1 above was violated. Methods Numerical exploration of implications of the assumptions To explore the effectiveness of modelling 2-layer ensembles with the single exponential equation, a simple numerical 2-compartment model, representing 2 layers of clothing, was produced (Figure 2). In the 2-compartment model an additional assumption was made - that the rate of air exchange between the compartments ( diffVx ) is constant. 23 where: CLiV x is the clothing ventilation for the layer i (l.min-1) ( CL2V x includes ventilation through fabric layer 2) VLi is the microenvironment volume of clothing layer i (l) pLi(t) is the concentration of oxygen in the clothing layer i (%) p air is the concentration of oxygen in the surrounding air (%) diffV x is the rate of air exchange between the clothing layers (l.min-1) Figure 2. 2-compartment model of clothing ventilation. Application of the numerical model The numerical model of a 2-layer clothing ensemble was used to produce oxygen dilution curves for known clothing ventilation and rates of air exchange between clothing layers ( diffVx ). The single exponential equation was fitted in 2 ways: 1. to the average oxygen concentration in the 2-layers together; 2. to the oxygen concentration in each layer sepa- rately. In each case, the predicted rate of air exchange (r) was derived. The predicted clothing ventilation was then calculated as the product of r and the microenvironment volumes used in the 2-layer model. This value was compared to the ventilation used in the model. The error of the single exponential approach was defined as the percentage difference (error) between the model’s known ventilation and the predicted values. This process was repeated for different rates of air exchange between the 2 clothing layers. Results Inputs: CL1V x = 5 l.s-1, VL1 = 20 l, pL1(0) = 10 %, CL2V x = 10 l.s-1, VL2 = 20 l, pL2(0) = 15 %, pair = 20.6 % 0 5 10 15 20 25 0.0 2.0 4.0 6.0 8.0 10.0 Air exchange between layers (l.s-1) 0 20 40 60 80 100 0.0 2.0 4.0 6.0 8.0 10.0 Air exchange between layers (l.s-1) Layer 1 Layer 2 Figure 3. Error when using a single expo- nential to calculate ventilation of a 2-layer clothing ensemble. Figure 4. Error when using a single expo- nential to calculate ventilation in each layer of a 2-layer clothing ensemble. 24 The calculated errors when applying the single exponential to both clothing layers, and to each layer separately, are shown in Figures 3 and 4 respectively. These graphs show only the magnitude of the errors, not their direction. In Figure 3 the single exponential under- estimates clothing ventilation. In Figure 4 the ventilation in layer 1 is over-estimated; ventilation in layer 2 is underestimated. Fitting a single exponential to the 2 layers together resulted in errors greater than 10 % for low rates of air exchange between layers. Errors were less than 10 % for the higher exchange rates. Fitting a single exponential to each of the layers individually (Figure 4) gave errors that increased with increasing rates of air exchange between layers. Only at very low exchange rates was the error less than 10 %. Discussion The microenvironment next to the skin is often the most important determinant of the ensemble’s ability to transfer heat away from the body. Therefore, measurements of ven- tilation in this layer (including the exchange with adjacent layers) must be accurate. Figures 3 and 4 show that using a single exponential equation may not give the required accuracy. They also imply that using a single exponential equation for multi-layer cloth- ing with low water vapour or air permeable layers leads to inaccurate calculation of clothing ventilation. Some protective and military clothing ensembles exhibit these properties. Air exchange between layers cannot be calculated using a single exponential. In order to calculate the clothing ventilation of multi-layer clothing with internal layers with low water vapour and air permeabilities, or to calculate the ventilation of layer 1 (next to the skin), current experimental procedures need to be extended. The micro- environment volume of each clothing layer must be measured; the concentration of oxy- gen in each layer must be recorded; a model more complex than the single exponential (single compartment) one is needed. Acknowledgements This study was funded by Technology Group 5 (Human Sciences and Synthetic Environments) of the UK Ministry of Defence Corporate Research Programme. References Birnbaum RR & Crockford GW (1978). Measurement of clothing ventilation index. Applied Ergonomics, 9(4), 194-200. Crockford GW, Crowder M & Prestidge SP (1972). A trace gas technique for measuring clothing microclimate air exchange rates. British Journal of Industrial Medicine, 29, 378-386. Havenith G, Heus R & Lotens WA (1990). Clothing ventilation , vapour resistance and permeability index: changes due to posture, movement and wind. Ergonomics, 33(8), 989-1006. Lotens WA & Havenith G (1988). Ventilation of rainwear determined by a trace gas method. In: Mekjavic IB, Banister EW & Morrison JB, Eds. Environmental Ergonomics: Sustaining human performance in harsh environments. London: Taylor & Francis, 162-176. Nilsson HO, Gavhed DCE & Holmer I (1992). Effect of step rate on clothing insulation – measurement with a moveable thermal manikin. In: Lotens WA and Havenith G, Eds. Environmental Ergonomics V. Proceedings of the Fifth International Conference on Environmental Ergonomics, Maastricht, Netherlands, 02-06 November 1992. Winslow C-EA, Gagge AP & Herrington LP (1939). The influence of air movement upon heat losses from the clothed human body. American Journal of Physiology, 127(3), 505-518. 25 Effects of skin pressure by clothing on thermoregulation and digestive activity Hiromi Tokura Department of Environmental Health, Nara Women’s University, Nara 630-8506, Japan Abstract I talk on physiological parameters, i.e., core temperature, skin temperatures, salivary melatonin, urinary noradrenaline and amounts of feces under the influences of skin pressure (foundation). Main findings are summarized as follows: 1) Rectal temperature was kept significantly higher throughout day and night under the influence of skin pressure due to foundation garments. 2) Diurnal decrease of leg skin temperature was significantly greater and its nocturnal increase was significantly suppressed under the skin poressure, which was responsible for upward shift of core temperature. 3) Nocturnal increase of salivary melatonin was significantly suppressed by skin pressure due to foundation. 4) Urinary noradrenaline was significantly suppressed by skin pressure due to foundation. 5) The amount of feces was significantly smaller while wearing the body compensatory brassiere than while not wearing it. These results suggest how deeply human body is influenced by skin pressure due to clothing. It seems problematic in terms of health maintenance. 26 Ergonomics of protective clothing George Havenith1, Ronald Heus2 1 Department of Human Sciences, Human Thermal Environments Laboratory, Loughborough University, LE11 3TU, UK; 2 TNO Human Factors, Soesterberg, 3769ZG, NL. Introduction Testing of workwear and specialised protective clothing has in the past mainly con- centrated on testing the properties of the materials used. A typical example of this can be found in the testing and evaluation of fire-fighting garments. Presently, these fire-fighter garments are usually tested to standards, which concentrate mostly on requirements for the materials (e.g. prEN469). Additional standards are under development, e.g. for as- sessment of the risk of burn injury using an instrumented manikin, but mostly tests refer to material heat and vapour resistance, flammability, retro-reflection properties etc. Though all these tests evaluate important properties of the clothing, this testing often neglects the effect the manufacturing process of the garment has on the material proper- ties (stitching, seems, treatments), the effects of clothing design, sizing and fit, as well as the interaction of the clothing with other components of the standard gear for the profes- sion. Such effects can only be tested by looking at the protective clothing ensemble as a whole, and such test should be incorporated in more detail into future PPE standards (e.g. prEN469). This paper deals with methods of testing protective clothing ensembles, which go be- yond the material standards. These tests, often referred to as tests on the ergonomics of clothing, consider the complete outfit, and involve manikin (insulation) and human sub- ject testing. The tests described here, which will be the human subject tests only, are based on extensive developments of ergonomic testing methods at TNO Human Factors over two decades (Heus et al, 1995, 1999), and were applied to various types of PPE (fire-fighting, motorcycle clothing, military clothing, chemical protective clothing). Hu- man subjects tests for comfort, physiological load, cold or heat protection, ergonomical design, fit, loss of performance, rain/moisture protection and conspicuity/visibility of the clothing will be described and proposed for further development into standards for ergo- nomic clothing evaluation. In the description of these methods, the example of testing fire-fighting gear will be followed, but the principles of the testing can be used in most types of protective clothing. Selection of tests and conditions An important advantage of many material test methods is that they are designed to pro- duce accurate and repeatable results, both within and between laboratories, at minimal cost. In order to achieve this, test conditions are chosen to provide a high sensitivity for differences in the parameter under study. In practice this also implies however that the test circumstances do not always reflect the actual use conditions. Real use conditions can be highly variable however, and it is unfeasible to test clothing ensembles in all pos- sible conditions. Hence, the first step in any clothing evaluation should be a proper task analyses to identify the main functions of the clothing, during what tasks it would be 27 used, in what environments that would be, and how the different requirements should be rated in terms of priorities (risk assessment, see e.g. prEN469). Doing this should identify the main requirements for the clothing to which subsequently testing procedures can be adapted. To follow the example of testing of fire-fighting gear, ‘typical’ scenarios may involve different ‘typical’ climatic circumstances:  Indoor fire attack High air temperature, medium to high radiation, smoke, water exposure, high ac- tivity levels  Outdoor extinguishing activities Low air temperature, low to high radiation, water exposure, low to high activity levels  Roadside rescue Ambient temperature, low radiation, exposure to rain, low to high activity levels, risk from traffic Within these scenarios, typical aspects of the clothing that may be considered as de- termining clothing performance are:  Insulation (protection from cold, heat (convective and radiant)  Vapour permeability  Air permeability  Visibility  Water protection (repellency, waterproofness, absorption)  Mechanical protection In addition to measuring/determining these parameters in the materials tests, they should now be determined in relevant scenarios. The main protective and other functional aspects of the clothing present in the fire fighter scenario’s can be determined in the fol- lowing tests: Test for physiological load Relevant practical conditions: Exposure where time is limited by SCBA or by heat strain: high air temperature (above skin temperature) with medium radiation and medium work load allows for working times up to 30 minutes. The heat strain will be mainly de- termined by the ability of the wearer to lose his body heat through the clothing to the en- vironment. Considering the high air temperatures, dry heat transfer will be towards the body, and heat loss will have to be through evaporation. Experiment: Different ensembles are compared to each other and to the clothing currently in use, using a within subject design. Each subject wears all clothing types in a balanced fashion to avoid any order effects on the results. The clothing, including SCBA, is worn while performing moderate work (walking at ± 3.5 km per hour, stepping over low benches [±250 W·m-2) in a room with air temperature of ± 60 °C and a black globe tem- perature of about 95 °C. Body core temperatures are monitored, as is sweat loss and sweat absorption/desorption in the clothing. The clothing’s performance is measured in terms of tolerance time (time for the wearer to reach an 8 J·g-1 body heat storage), dry heat gain, evaporative heat loss and evaporative heat loss efficiency (sweating effi- ciency). Subjective ratings of subjects are taken as well. 28 Test for heat protection Relevant practical conditions: Exposure where time is limited by heat protection against high radiant loads. The high radiant load will increase suit temperature and the heat can penetrate the suit fast leading to skin burns. Experiment: Using the same experimental design as in the physiological load test, sub- jects in the clothing to be tested are now exposed to very high radiant loads (7 kW·m-2) in a hot chamber (>100 °C). In front of well-controlled propane burners delivering this load they perform a light bench stepping exercise. The clothing will increase in temperature very quickly, as will skin temperature. The clothing’s performance is measured as the tolerance time, which is either defined by the subjects withdrawal, or by any skin tem- perature of the subject coming within 1 °C of the pain threshold limit of 45 °C. The maximal exposure time for this test is 120 seconds. The function of the exercise is mainly to keep the clothing moving to and from the skin. A high risk of burns would be present if clothing was left static. In that case it’s temperature could increase to very high values, and then when the person wanted to withdraw the hot clothing would touch the skin and cause a burn. Experience has shown that subjects voluntary withdraw at skin temperatures around 44 °C. Thus, as long as precautions are taken that the clothing isn’t static, with the asso- ciated risk of high clothing temperatures, no risk of skin burns in this type of testing is present. Water protection Relevant practical conditions: Exposure to neutral ambient conditions where ‘comfort’ and health risk is determined by rain or spray-water protection. Experiment: In the same within-subject design as above, the clothing is worn in a well- defined artificial rain environment. Subjects need to walk in line with the others while doing a small obstacle course at low speed. This consists of climbing over objects, as well as crawling under objects and e.g. moving crates from low to high positions and vice versa. These movements test the design of the clothing (no gaps when bending over), the waterproofness of materials, seems etc., in normal conditions and under pressure (knees, elbows). The clothing’s performance is measured in terms of water absorption, water penetration to the underclothing (both by weighing the respective garments), the leakage locations and the subjects subjective judgement. Freedom of movement Relevant practical conditions: Task performance where freedom of movement is deter- mining movement speed, flexibility and energy expenditure. Any hampering of the wearer’s movements increases risk. Experiment. In the same within-subject design as above, subjects in the gear (without SCBA) are asked to complete an obstacle course at maximum speed. This involves rele- vant activities for the profession (climbing ladders, through windows, over, under and through obstacles). The clothing’s performance is measured in terms of time to complete the obstacle course, in relation to the time to do it in a training suit. In addition various small tests (time needed for donning, sprint, running in 8 shaped pattern underneath a lowered bar, sit and reach and stand and reach test, sargeant’s jump) can be added to evaluate the way in which the clothing hampers the movement. 29 Ergonomical design assessment Relevant practical conditions: When all the clothing and equipment worn during a cer- tain task is combined the clothing still should be functioning properly and provide the basic functions (protection, storage, etc.) to the wearer. Assessment: For this aspect an expert panel (ergonomics and topic specialist) assesses the clothing. While the clothing is worn by subjects of different statures and builds, as well as by panel members, the clothing is evaluated for freedom of movement, proper design (overlap between jacket and trousers, arm length), compatibility with other equipment (e.g. accessibility of pockets while SCBA is worn). The clothing performance is measured using consumer evaluation type tables (scores much better, better, equal to, lower/worse etc. than average or than old suit or a reference suit) Visibility/Conspicuity Relevant practical conditions: Exposure during tasks where the wearer’s visibility is im- portant. Any activity in traffic, or in obscure areas will have a high risk if visibility of the wearer is insufficient. Experiment: As not only the materials used and their total surface area, but also the pat- terns used are relevant in this case, tests are done in a set-up modified from the CEN tests on retro-reflective clothing materials (EN471). In a set-up simulating lighting condition of a car approaching in full darkness, using the car’s driver viewpoint when the car is 100 meters away from the subject, pictures with defined exposure time are taken of the suits in frontal, lateral and rear view. Later, these pictures are evaluated by the expert panel. The clothing’s performance is measured in terms of visibility of the subject (total visible surface), as well as in terms of recognition as a human (pattern of visibility material). Human subjects as evaluation tools As mentioned before, material test standards are geared towards accurate and repro- ducible results, In the tests proposed here, measurements are performed using human subjects as ‘evaluation tools’. This implies that the experimenter will have to expect hu- man variability to influence the results. While inter-subject variability is desirable, in or- der to evaluate the PPE for use by different populations, intra-subject variation (dif- ferences between days etc.) is undesirable and efforts should be made to minimise this. Pre-test conditioning of clothing and subjects are thus important, as is the subjects moti- vation level which needs to remain high. Though this variability will increase noise levels in the data, experience shows that the tests described here can discriminate well between different clothing ensembles for various PPE types, and that they are very sensitive in exposing typical weak spots in the clothing design or manufacturing process. We are therefore convinced that adaptation of these or similar test procedures for use in PPE standards would advance the quality of PPE testing substantially. References EN 469 (1995) Protective clothing for firefighters. European standard, Brussels: CEN EN 471 (1994) High visibility warning clothing. European standard, Brussels: CEN Heus R, Wammes LJA, Kistemaker JA, Havenith G (1995) Improvement of the Amsterdam Firefighters garments (in Dutch). Report TNO-TM 1995 C 06. TNO Human Factors Research Institute, Soesterberg. Heus, R. & Heukelom, J. (1999). Fire fighting and human factors; what does EN469 mean for fire fighters. Proceedings of the international conference on thermal protection of man under hot and hazardous conditions, March 1999, Paris, France, 51-58. 30 Application of the product planning chart in quality function deployment to improve the design of a fireman’s safety harness Neil Parkin1, Dave J. Stewardson1, Michael Peel2, Mike Dowson2, Joe F. L. Chan1 1 Industrial Statistics Research Unit (ISRU) University of Newcastle, Mechanical, Materials, and Manufacturing Engineering, University of Newcastle, UK. 2 Draeger Ltd, Northumbria, UK. Introduction This paper discusses the use of a systematic planning tool called Quality Function Deployment (QFD), which has been used to improve the design of a belt/buckle as- sembly used on a Fireman’s safety harness. QFD is a tool, which provides the develop- ment team with a structured planning approach to new product development that is driven by the ‘voice’ of the customer. This is achieved by linking customers requirements to ways of achieving them, while taking into account both customer and competitors’ product performance. As a tool QFD has been growing in influence both in Japan and the USA, as well as in Europe. The planning process used in Quality Function Deployment is usually split into Four Phases, and charts are used to help facilitate planning during each phase. The first phase is used to define the overall requirements that are deemed to satisfy the customer, the re- maining three are used to define the parts, processes and production requirements in de- velopment (Bossert, 1991; Day, 1993). For the purpose of this paper, the first phase in the QFD planning process is discussed. As a consequence, the format of this paper is based around the activities that are required to build the first phase QFD chart formally known as the Product Planning Chart. From the analysis of the Product Planning Chart, and knowledge of the company’s existing budgetary constraints, the development team decided to begin development that looked at alternate buckle designs that would make the replacement and refitting of the Belt/Buckle easier. QFD project planning Before the project could commence, senior management with the aid of a Product Manager formed a multi-functional development team, with members from the Product Design, Manufacturing, Purchasing, Accounts, Technical Documentation, Marketing, and Quality business functions. Once the team was formed, based on the Project Goal to ‘Improve the existing design of the Belt/Buckle Assembly on a given range of Safety Harnesses’, the objective to Investigate the implications on Belt/Buckle Material, Testing and Physical Size were agreed. From these objectives a schedule of work was formulated highlighting the tasks to be completed, making sure that this project tied in with the company’s existing busi- 31 ness plan. The remainder of this paper is based on the schedule of work relating to build the first QFD Chart for Product Planning, which can be found in Table 1. Customer ‘voice’ data collection and analysis Using an interview guide, a representative sample of customers were interviewed to identify their requirements. A questionnaire was then developed based on the findings of the interviews and used to establish the importance for each customer requirement. From the questionnaire it was found that the most important customer requirement for the Belt/Buckle Assembly was that “the unit should be easily replaced and refitted”. The Customer Requirements were then grouped using Affinity Diagrams so that they could be put into the QFD Chart in a structured tree format, along with the importance ratings. In order to generate Customer Competitive Assessment Data, the development team obtained belt buckles used by a leading competitor and asked company personnel outside the development activities to rate the belt/buckle assembly against the competitor’s buckle on the issues that came to light in the questionnaire. The Customer Competitive Assessment Data was then put in the QFD Product Planning Chart, and the development team met to establish specific objectives for the Belt/Buckle Assembly project, and brain-storm the Technical Requirements from the Customer Requirements. Once all of the low-level Technical Requirements had been identified, the development team then grouped them into categories using Affinity Diagrams. As the Belt/Buckle used by the company in question was very similar in design to the competitor’s buckle, the Customer Competitive data revealed that the customer and com- petitor ratings were similar, however the customer rated the competitor’s buckle much higher for “ease of replacement/refit”. Conduct technical competitive benchmark studies At the end of the first meeting, it was agreed that the team members from the Product Design and Manufacturing Functions should carry out relevant tests on their own, and on their competitors’, Belt/Buckle assemblies. By carrying out such tests Technical Competitive Data were then generated, preliminary target values were computed, and the orientation for each Technical Requirement formulated. QFD product planning chart building With the Customer and Technical Competitive Data incorporated into the Product Planning Chart, the development team had a second meeting to complete the Relationship Matrix, establish the Absolute and Relative Importance Weightings for each Technical Requirement, and review the whole of the QFD Chart. In this meeting, the QFD chart was built-up using software that was projected onto a screen for the develop- ment team to observe. In order to complete the Relationship Matrix, the development team established the level of relationship (strong, medium, or weak) between the Technical and Customer Requirements by assigning a correlation value 9, 3, 1 respectively (by brainstorming). The Absolute and Relative Importance Values for each Technical Requirement were then calculated by the QFD software using the following equations: 32 Absolute Importance I jabs   W Ri ij i i n * 1 (After Santow and Clauing, 1993) Relative Importance I j rel   I I j j j j n 1 (After Santow and Clausing, 1993) where: I Technical Requirement Importance value W Customer Requirement Rating (eg: any value 1 to 10) R Correlation factor (eg: 1, 3, or 9) between each Customer Requirement and Technical Requirement Subscripts i Customer Requirement Row Reference Number j Quality Characteristic, Design Requirement (or HOW) Column Reference Number n Any number of Customer Requirements and Technical Requirements >1 Consequently, Ij abs is the absolute importance value of the jthTechnical Requirement, Wi is the weight of the ith Customer Requirement, and Rij is the correlation value between the two. Ij rel is the relative (technical) importance value for the jth Technical Requirement, di- vided by the arithmetic summation of all Technical Requirement relative importance values from j=1, to j=n expressed as a percentage. Analysis of the chart data quite clearly showed that one weakness in the existing de- sign of the buckle was that it could be improved in relation to ease of replacement and refitting. Also, the customer rated both companies as average for “operation when hot or cold”, consequently this could be seen as a marketing opportunity. From the analysis of the Product Planning Chart, and knowledge of the company’s existing budgetary constraints, the development team decided to begin a development programme that looked at alternate buckle designs that would enable easier replacement and refitting of the Belt/Buckle. References Bossert JL (1991) Quality Function Deployment: A Practitioner’s Approach. ASQC Quality Press, MI, WI. Day RG (1993) Quality Function Deployment: Linking a Company with It’s Customers. ASQC Quality Press, Milwaukee, WI. Santow K & Clausing DP (1993) Integration of Quality Function Deployment with Further Methods of Quality Planning. Report No. LMP-93-005, Working Paper of the Laboratory for Manufacturing and Productivity, MIT, Cambridge, MA. Ta bl e 1. Q FD Pr od uc t P la nn in g M at rix fo r B el t/B u ck le A ss em bl y. 34 An adaptive approach to the assessment of risk for workers wearing protective clothing in hot environments Ken Parsons Human Thermal Environments Laboratory, Loughborough University, Loughborough, LEICS LE11 3TU, United Kingdom Introduction The human body responds to the microclimate between the skin and the clothing and any risk of heat strain will be as a response to that climate. The properties of the microclimate will be determined by the external environment and the clothing heat and vapour transfer characteristics as well as the metabolic heat produced by the body due to activity. Tradi- tional heat stress assessment methods will attempt to take the above factors into account and predict the thermal strain caused to the clothed person. The risk of unacceptable heat strain is therefore directly related to the environment, clothing and activity. In practical situations the risk of heat strain is not dependant upon such a passive system. If people are subject to unacceptable environments then they will attempt reduce the effects of the environment. There are a number of ways in which this can be done including reducing work rate, altering clothing or posture, moving out of the hot environment and more. The risk of heat strain can then be viewed as the lack of ability, potential or opportunity for a person to act or adapt to reduce the heat strain. This provides an adaptive approach to as- sessment where by contrast with traditional methods, for the same environment, clothing, activity and person, a context with restricted opportunity to avoid heat strain would pro- vide a much greater risk than a context where it is possible to act to reduce the heat strain. Current methods for defining the thermal properties of clothing take no account of hu- man behaviour. How people wear clothing and its dynamic interaction with tasks, work- places and environments, however is of major importance in human thermoregulation with often dominant significance for human comfort, performance and survival. The degree of freedom of an individual to behave appropriately in any context therefore should be a major component in any assessment of risk of thermal strain in hot and cold environments. This paper presents the concept of the adaptive approach to assessing risk and demonstrates how it can be used in practical application with particular reference to the effects of protective clothing. Specification of clothing properties Specification for the thermal properties of clothing has evolved from an index of the thermal properties of materials, involving the measurement of heat and vapour resistance using materials fixed to a ‘flat plate’ under standardised conditions, to indices which rep- resent those properties in clothing on the human body. Indices include the Tog value for materials (Pierce & Rees, 1946) where 1 Tog = 0.1 m2·K·W-1, and vapour transfer proper- ties measured in m2·kPa·W-1 . For these indices the term ‘m2 ’represent one square metre of material. This can be contrasted with indices of the thermal properties of clothing 35 where the Clo value is used (Gagge et al., 1941), where 1 Clo = 0.155 m2·K·W-1 and is the thermal resistance provided to the human body by clothing as if the clothing had been spread across the whole body. A similar concept for the vapour transfer properties of clothing provides units of m2·kPa·W-1. For clothing indices however, the ‘m2’ term refers to a square metre of the human body surface (often taken as 1.8m2 for an average man). The thermal resistance of a pair of socks would therefore have a material resistance similar to that of trousers or a shirt (e.g. 0.1 m2·K·W-1) whereas the socks would have a Clo value of around 0.1 Clo (0.155 m2·K·W-1), as the surface area of the socks is small compared to that of the human body when compared to a shirt or trousers where surface areas are greater and Clo values will be around 0.3 to 0.4 Clo. The properties of clothing are measured on thermal manikins which represent the shape and temperature distribution of the human body. They can also be calculated from the physiological responses of human subjects when wearing the clothing. Intrinsic re- sistance values represent the ‘intrinsic’ properties of clothing and are independent of en- vironmental conditions. Total clothing insulation values include not only the intrinsic properties of clothing but also the resistance of the environment surrounding the body. This will depend upon environmental conditions and human body movement. Develop- ments in quantifying the thermal properties of clothing include the ventilation index (Birnbaum & Crockford, 1978) which quantifies the air exchange between the microcli- mate within clothing and the external environment. All of the above indices can be used in the assessment of the thermal strain of clothed workers in hot environments. A practical example Consider a worker in protective clothing (estimated insulation value of 1.0 Clo, normal vapour permeation properties) carrying out heavy manual work (estimated 200 W·m-2) for 1 hour in an air temperature of 35 °C with no significant thermal radiation, 50 % relative humidity and low air flow (0.2 ms-1). Using a simple 2-Node model of human thermo- regulation, based upon Gagge et al (1971), the predicted responses of the workers are that skin temperature would rise rapidly and the skin would be completely wet with sweat within 15 minutes of work. Core temperature would rise above 38.0 °C within 40 minutes and would be at around 38.5 °C after one hour of work. An assessment of the risk of unacceptable heat strain in those conditions would reasonably conclude that there is some risk to the worker after one hour of exposure. In practice, whether there is unacceptable risk will depend upon the opportunity for the worker to adapt to the work. For example, when the worker becomes ‘too hot’ or ‘too sweaty’ then it may be possible to take off a jacket or slow down. A reduction in Clo value to 0.7 Clo and metabolic heat production to 150 W·m-2 after 15 minutes of expo- sure, according to the 2-Node model, slows the rate of increase of skin temperature and sweating and maintains a body core temperature below 38 °C even after one hour of ex- posure. If an opportunity to adapt is available then there is a low risk and if the opportu- nity to adapt is not available then there is high risk. The 2-Node model of human thermoregulation was used to represent a ‘traditional’ static model of assessment and other methods would provide a similar result (e.g. ISO 7933, 1989). When assessing the risk of unacceptable heat strain for workers wearing protective clothing, therefore, it is important to determine the ‘degrees of freedom’ a worker has to adapt to the environment. 36 Risk assessment – an adaptive approach An assessment of risk of unacceptable heat strain for work in hot environments must con- sider factors that influence human response, hence air temperature, radiant temperature, humidity and air movement making up the environment and metabolic rate and clothing of the worker. These provide insight into how to use an adaptive approach in risk assess- ment. That it, a consideration should be given to the opportunity to adapt to or adjust the environment, work rate or activity level, and clothing. Environment A method of reducing heat strain would be to reduce the strain caused by the environ- ment. Examples of how this could be achieved include moving away from the hot area, shielding from radiation or increasing air flow across the body. An ability to recover in a cool area is also important. Activity High levels of activity with protective clothing should always be regarded as high risk. The ability to vary the pace of work will provide a major method for reducing thermal strain and avoidance of heat stroke. Many jobs will allow this and varying activity level and hence metabolic heat production, probably explains why workers can perform jobs when traditional assessment methods predict that heat stroke will occur. Some jobs how- ever will not allow variation in activity level. For example, in situations where exposure time may be limited, the environment is hot and protective clothing must be worn and the task must be completed. This should be regarded as high risk. The social context of the work is also relevant. Highly motivated people may increase their workload and maintain it even to heat stroke. Military personnel and sportsmen and women, may be particularly susceptible and any adaptive opportunity should be viewed in a social, cultural and moti- vational context. Clothing Clothing restricts evaporative heat loss which is the main physiological mechanism for the body to lose heat in a hot environment. The ability to remove clothing or adjust it to allow heat loss will provide a major adaptive opportunity in a hot environment and clothing design should take this into account. However, protective clothing is often im- permeable and closed to protect against hazards such as airborne toxins. It may not be possible therefore to adjust clothing. The ability to remove clothing if a worker collapses or to place gloved hands into cold water for example will be important in reducing ther- mal strain and the avoidance of injury. Clothing designed to allow air exchange through pumping will provide an adaptive opportunity. Other adaptive methods Other factors which provide adaptive opportunity include technological ‘solutions’ such as opening windows and turning on fans. These would alter the environment as described above. Changing posture can also be regarded as an adaptive response. If a person must consistently work in a crouched posture, the surface area for heat loss from the body will be greatly reduced. 37 Practical questions for risk assessment The above discussion has demonstrated that for an effective assessment of risk in hot en- vironments the ability for workers to adapt should be considered. In practice this would add to a checklist or procedure for assessment. The questions that should be asked could be as follows.  Is there an opportunity for the worker to change the environmental conditions in terms of air temperature, radiant temperature, humidity or air velocity?  Could a worker change his or her level of activity if he or she became too hot?  Could a worker remove or adjust clothing if he or she became too hot?  Could the worker change posture to reduce heat strain if he or she became too hot?  Does the social, motivational and cultural climate allow adaptive opportunities to be taken?  Can you identify other adaptive opportunities that will reduce the risk of unacceptable heat strain?  Are the workers sufficiently trained to take advantage of adaptive opportunities?  Have the workers been advised of adaptive opportunities at this workplace? Conclusion This paper has demonstrated that the opportunities, or lack of opportunities, for workers to adapt to hot environments by, for example, varying work rate or reducing clothing levels, has significance in assessing the risk of heat strain. It provides the basis for a more comprehensive study into the inclusion of the adaptive approach for consideration in risk assessment. References Birnbaum RR & Crockford GW (1978) Measurement of the clothing ventilation index. Applied Ergonomics, 9(4), 194-200. Gagge AP, Burton AC & Bazett HC (1941) A practical system of units for the description of the heat exchange of man with his thermal environment. Science, N.Y. 94, 428-30. Gagge AP, Stolwijk JA J & Nishi Y (1971) An effective temperature scale based upon a single model of human physiological temperature response. ASHRAE Transactions 77, 247-262. ISO 7933 (1989) Hot environments. Analytical determination and interpretation of thermal stress using calculation or required sweat rate. ISO Geneva. Pierce FT & Rees WH (1946) The transmission of heat through textile fabrics. Part II. Journal of Textile Institute, 37, 181-204. 38 Radiation protective clothing in a hot environment and heat strain in men of different ages Anna Marszalek1, Maria Konarska1, Juhani Smolander2, Krzysztof Soltynski1, Andrzej Sobolewski1 1 Central Institute for Labour Protection, Warsaw, Poland 2 University of Helsinki, Helsinki, Finland Introduction The aim of the study was to compare thermal strain of older men of different ages at rest and during exercise in a hot environment without and with heat radiation and then with- out and with radiation protective clothing and to assess if older men are at a greater risk for hyperthermia in these conditions. In a hot working environment it is often necessary to use reflective protective clothing against radiant sources. This kind of clothing is aluminized and greatly disturbs heat dis- sipation from the human body and may result in excessive heat storage. Older working population is constantly growing in industrialized countries. More older people are also exposed to occupational thermal stress. Some studies related to thermoregulation indicated that middle-aged (45-64 years) men and women tolerate work in the heat less and sustain a greater physiological strain during heat acclimation than young ones (Hellon et al. 1956). However other studies showed only low attenuation of thermoregulatory responses from the fifth and sixth decades of life for aerobically trained and relatively lean men and women (Armstrong & Kenney 1993). Thermal strain of men and women of different ages during acute heat exposure or ac- climation at rest and during exercise is in general the same or even lower in middle-aged men and women if they are matched for level of aerobic fitness and selected morphologi- cal factors such as: surface area, weight, percentage body fat (Kenney 1988, Tankersley et al. 1991). Methods Studies were conducted at rest (24 subjects) and during exercise (20 subjects) in a hot environment at maximal thermal load determined by hygienic standards of around 29 ºC WBGT and around 26 ºC WBGT at rest and during exercise, respectively (Table 1). Table 1. Thermal environment conditions in the study at rest and during exercise (Cx r SD). Rest study Exercise studyThermal environment conditions No protective clothing Protective clothing No protective clothing Protective clothing WBGT (ºC) 29.1 r 0.6 28.3 r 1.2 26.5 r 0.6 25.7 r 0.6 tg (ºC) 37.6 r 0.8 41.4 r 1.5 36.5 r 0.6 40.5 r 1.0 t nwb (ºC) 25.4 r 0.6 22.2 r 2.0 22.3 r 1.0 19.4 r 0.8 Note: WBGT – wet bulb globe temperature, tg – black globe temperature, tnwb – natural wet globe temperature. 39 In experiments without protective clothing subjects wore only shorts, socks and tennis shoes (0.09 clo) whereas in experiments with protective clothing garments included cot- ton long-sleeve and long-leg underwear, a three-part aluminized protecting set (jacket, trousers, gloves), socks and leather shoes (1.17 clo). During the experiments, rectal (t re ) and skin temperatures, heart rate, heat storage, body mass loss, and the subjective ratings of thirst, thermal sensation and perceived effort (in the exercise study) were collected. The subjects were divided into three age groups: young (Y, 20-29 years), middle-aged (M-A, 41-55 years) and older (O, 58-64 years). The rest study lasted one hour and the age groups were matched on the basis of the subjects’ activity level in leisure time. In the exercise study, groups Y and M-A repre- sented average while the subjects in the O group had high maximal oxygen uptake (VO2max) for the respective age groups. In the study, subjects exercised for 30 minutes on a cycloergometer at 40 % VO2max. Results and discussion At rest in experiments where protective clothing was used in all groups a tendency to higher body mass loss was observed: 189.0 r 79.2 g, 191,9 r 89.0 g and 167.5 r 69.4 g in the “no-protective clothing” experiment 279.9  106.0 g, 245.3  64.4 g and 302. 8  124.9 g in protective clothing in groups Y, M-A and O respectively. There were signifi- cantly worse subjective ratings of sweating in all groups compared to experiments with- out protective clothing (Table 2). There were no significant differences in thermoregulatory responses during passive heat exposure of men of different ages. The thermal load applied in the study was probably not high enough because Lind et al. (1970) indicated that older men tolerated heat significantly less than young ones when thermal conditions were more acute com- pared to moderate ones. Table 2. Subjective ratings at the end of experiments in a hot environment at rest and during exercise ( C x r SD). Rest study Exercise study Subjective ratings Groups No protective clothing Protective clothing No protective clothing Protective clothing Thirst (scale 0-4) Y M-A O 2.7 r 0.9 2.4 r 0.7 1.5 r 0.5*1 2.8 r 1.0 2.6 r 0.7 2.0 r 0.9*1 2.50 r 0.50 2.33 r 0.75 1.67 r 0.75*1 2.75 r 0.83 2.67 r 0.75 1.83 r 0.69*1 Sweating (scale 1-7) Y M-A O 5.7 r 0.7 5.4 r 1.0 5.3 r 0.8 6.9 r 0.3*2 6.3 r 0.9*2 6.7 r 0.5*2 6.7 r 0.3 6.2 r 0.4 6.3 r 0.0 6.7 r 0.0 5.8 r 0.0 6.5 r 0.2 * 1 p<0.05 compared to Y, *2 p<0.05 compared to “no protective clothing” experiment During 30-min exercise in a hot environment older men with higher physical capacity for their age had significantly lower increase in t re and lower heat storage than in young men with average physical capacity (Table 3). Simultaneously body mass loss in older men was similar to that in young men. 40 Table 3. Heat strain indices during exercise in a hot environment (Cx r SD). Heat strain indices Groups No protective clothing Protective clothing ' t re (ºC) Y M-A O 0.53 r 0.04 0.58 r 0.05 0.38 r 0.07*1 0.52 r 0.05 0.55 r 0.06 0.42 r 0.08*1 Heat storage (Wm-2) Y M-A O 51.7 r 10.4 62.3 r 11.4 41.1 r 11.1*2 68.2 r 9.9 73.2 r 7.8 52.1 r 17.8*2 Body mass loss (g) Y M-A O 354 r 112.6 350.0 r 70.2 351.6 r 80.8 437 r 97.2 412 r 112.1 370.0 r 146.9 * 1 p< 0.05 compared to Y, *2 p< 0.05 compared to M-A. In the study significantly lower thirst of older men compared to young ones was ob- served in all the experiments – at rest and during exercise (Table 3). Philips et al. (1991) obtained similar results but at different stimulations. They indicated reduced thirst of older men after infusions of hypertonic and isotonic saline and lower drinking during 30- min rehydration period compared to young men. These authors concluded that reduced thirst in older men was primarily related to a lower thirst sensitivity to hypertonicity and these older men showed trends towards an increased thirst threshold compared with younger men. Conclusions It was shown that the older group did not tolerate protective clothing worse than young one. In the study, the tolerance of passive heat exposure was similar in the young and older groups matched on the basis of their physical activity. It was shown that the tolerance of 30-minute exercise at temperate intensity and at similar relative physical load was even better in the older group with higher physical capacity for their age than in the young group with average physical capacity. However, lower thirst in older men observed in all the experiments indicates that prolonged exposure of older men to a hot environment may carry a potential risk factor of dehydration, especially at a similar sweating rate as in young men. References Armstrong CG & Kenney WL (1993) Effects of age and acclimation on responses to passive heat exposure. J. Appl. Physiol. 75(5): 2162-2167. Hellon RF, Lind AR & Weiner JS (1956) The physiological reaction of men of two age groups to a hot environment. J. Physiol. (Lond) 133: 118-131. Kenney WL (1988) Control of heat-induced cutaneous vasodilatation in relation to age. Eur. J. Appl. Physiol. 57: 120-125. Lind AR, Humphreys PW, Collins KJ, Foster K & Sweetland KF (1970) Influence of age and daily duration of exposure on responses of men to work in heat. J. Appl. Physiol. 28: 50-56. Tankersley CG., Smolander J, Kenney WL & Fortney SM (1991) Sweating and skin blood flow during exercise: effects of age and maximal oxygen uptake. J. Appl. Physiol. 71(1): 236-242. Philips PA, Bretherton M, Johnston CI & Gray L (1991) Reduced osmotic thirst in healthy elderly men. Am. J. Physiol. 261: R166-R171. 41 Management of Safety and Health Protection on building sites – under special consideration of use of personal protective equipment Bernd Ziegenfuß, Nicola Klein Mplus – Managementgesellschaft zur Optimierung von Arbeitsbedingungen mbH, Germany Introduction In many working environments the "management of safety and health protection" empha- sises the use of "personal protective equipment". The EC - guideline 89/656/EEC estab- lished the ordinance on the use of personal protective equipment, which contains the ob- ligation of employers to select and provide personal protective equipment for the employees, if other measures of the maintenance of industrial health and safety standards do not offer any sufficient protection for the employees. In addition to that it is necessary to instruct the employees on how to use their personal protective equipment the right way. The mentioned topic shall be clarified in the following with the example "management of safety and health protection on building sites". Compared with other fields of work the employees on building sites are exposed to particularly high accident and health risks. Changing working conditions and weather influences as well as appointment pressure and the simultaneous activities of employees of different employers permanently increase the endangering risks on building sites. Methodology Years of experience in coordinating safety and health protection on over 50 construction sites in Germany and Austria with a building volume between 250.000 and 7.000.000 EUR delivered the basic data for the discussion of the above mentioned topic. The empirical value of these data are emphasizing the need for an effective manage- ment of safety and health protection on building sites under special consideration of the use of personal protective equipment. Safety engineers who have been trained to coordinate safety and health protection on construction sites are witnessing dangerous situations involving people who are working on construction sites everyday. A lot of these situations could be less dangerous if employees would be sufficiently instructed and job sequences organized and optimized. Safety engineers on construction sites are confronted with these – often difficult – tasks in their daily work and their experiences helped to develop training measures. Results and Discussion In Germany the accident rate (accidents per 1000 full workers) lies more than twice as high in the building sector than in the average of the commercial economy - both at the 42 reported and the particularly serious work accidents. Almost every second fatal accident still happens in the building and construction industry - according to the statistical evaluations of fatal work accidents. Table 1. Number of work accidents - Building and construction sector compared to the average annual accident rate of the Commercial economy (HVBG, 1997). 1970 1980 1985 1990 1995 1996 1997 Building and construction sector 420.485 306.101 214.607 251.742 364.773 321.958 312.975 Commercial economy 143.600 110.087 83.319 95.100 101.099 90.461 87.252 Most of the fatal accidents are due to falling of scaffoldings, roofs etc.. About 60 per cent of the fatal accidents happen in connection with transport and subsidies building machines, processing machines as well as scaffoldings, in which every fifth piece of equipment shows safety technical defects. It can be assumed that every eighth accident wouldn't have gone out fatally, if personal protective equipment had got specified, posed and used. Table 2. Ranking of severe accidents on building sites. Cause of accident Percentage of accidents Falls from great height and other falls 38 % Transport events: construction site vehicles and construction site machines 19 % Collapses - also collapses with masses in movement 14 % Objects and materials falling down 10 % Electric blow 8 % Suffocating, drowning 4 % Fires, explosions 3 % Other 4 % Conclusions Because of the reasons mentioned above the management of safety and health protection under special consideration of personal protective equipment is so important on building sites. Especially on building sites it proves difficultly to check the retention of protection regulations, such as carrying personal protective equipment. To improve the safety and health protection of the employees on building sites the EC - building site guideline 92/57/EEC was passed in June 1989. In the meantime this ordinance was converted into national law in the member countries of the EC and shall improve the working conditions at the building sites as well as reduce work accidents, occupational diseases and work conditional health dangers. Especially on building sites it is crucial for the employees that executives can be taken as examples in wearing personal protective equipment. That is the only way to guarentee that the employees sense of duty to wear personal protective equipmet can be increased. Architects - for example - who enter a building site without helmet and safety shoes are just one unsuitable negative example for the employees. Building sites offer a special challenge, because a lot of executives do not act as models for their employees. The employees observe certain behaviors and take these over as their own behavior. This kind of ‚learning by observing the behavior of other people‘ may lead to negative behavioral aspects which are very hard to reverse once they are established. To reverese these be- 43 haviors intensive training programmes are indispensable. The goal must be to prevent these behaviors from developing by improving the executive duties to manage safety and health protection under special consideration of personal protective equipment. It is therefore inevitable to train those who are responsible for empoyees on construc- tion sites to enable them to manage safety and health protection – paying special attention to the use of personal protective equipment. Special training classes have been developed during the last two years and have already proved to be successful. Evaluating data will have to be elevated. References Bundesministerium für Arbeit und Sozialordnung (1999) Occupational Safety and Health Protection on construction sites, Bonn. Hustadt H & Ziegenfuß B (1999) Legal aspects of the coordination of Safety and Health. Protection on construction sites. Spektrum der Arbeit 4/99, 7-9. Tepasse R (1999) Handbook – Coordination of Occupational Safety and Health Protection on construction sites. Berlin: Erich Schmidt Verlag. Ziegenfuß B (1990) The meaning of personal protective equipment for employees on construction sites in Europe. Jahresbericht der Bau – Berufsgenossenschaft Wuppertal. Ziegenfuß B (1999). Everything in the plumbline – Coordination on construction sites – management of safety and health protection. Spektrum der Arbeit 3 /99, 4-7. 44 Clothing trials as a part of worker training Tanja Risikko, Juhani Hassi, Tiina M. Mäkinen, Liisa Toivonen Finnish Institute of Occupational Health, Cold Work Action Program, Aapistie 1, FIN-90220 Oulu, Finland Introduction According to an extensive survey in Finland the construction workers are most occupa- tionally exposed to cold in their work (Hassi et al, 1998). There are a large number of risk assessment methods as well as safety norms and recommendations for the construction industry (Lappalainen et al, 1991). These recommendations do not pay attention to the effects of the cold on human beings. An ISO standard for working practises in the cold is under development. It will provide a comprehensive strategy for risk assessment and management practices of cold risks (ISONP 15743). A development project was conducted in order to improve the management of cold re- lated health and safety risks in the construction industry. The cold risks in house building work were assessed at a target site by means of a questionnaire and ISO standardised as- sessment methods. According to the questionnaire the workers had experienced cold re- lated disease symptoms, cold injuries, discomfort, decline in working capacity, and re- duced working motivation. The most problematic ambient factors were wind, cold mate- rials and wet conditions. The work place assessments showed a risk of extremity cooling in many work phases and a risk of whole body cooling during the cold season in tasks involving low heat production. From the basis of risk assessment, immediate development measures were carried out. As a learning organization, the company participated actively in developing its organiza- tion culture (Ruohotie 1993). At first foremen and key persons of the construction site were trained to recognise and manage the cold risks in construction work. The worker training was then conducted in information campaign at the target site. Further written learning and guidance material was provided. Information was given about the effects of cold on physical performance and health as well as cold protection. Field trials of protec- tive clothing and equipment were an important part of the training. Methods At first, the workers were informed about the aim of the project and the forthcoming ac- tions at the target site. The cold induced health and safety risks were assessed at the target site by using a primary questionnaire and ISO standardized methods. The analysis of the present state showed needs for immediate development measures. The workers received the cold guide booklet (Figure 1) as a basic information package about the cold. Information about the effects of cold and protection against cold was de- livered at the target site in 30 minutes information sessions. Written material was included in the dissemination of information (Figure 2). In the development actions the concept of "learning by doing" was applied. The cloth- ing trials were connected with the information sessions. The tested garments and equip- 45 ment were chosen from the needs that had risen in the risk assessment stage. The aim of the trial and the function of the test garments were explained to the workers. THE COOLING EFFECT OF WIND & WIND PROTECTION Cooling of the bare skin The wind increases the adverse effects of cold. The skin cools asymmetrically, the extremities cool most. Windchill index describes the cooling effect of wind on bare skin areas, such as face and hands. The numbers in the table equal the combined effect of ambient temperature and wind. Protection against wind: •Thin gloves are recommended for working in manual dexterity requiring jobs insted of working barehands. Face mask forms an isolating air layer on the skin. Solely a thin wool, cotton or silk balaclava prvents frostbite effectively in cold and wind. (Oily creams are not recommended for cold protection of the face). Windchill index describes the cooling effect of wind on bare skin air temperature, °C wind, m/s cold freezing cold frostbite risk on bare skin bare skin freezes in max 30 seconds 10 5 0 -5 -10 -15 -20 -25 -30 -35 -40 -45 -50 calm 10 5 0 -5 -10 -15 -20 -25 -30 -35 -40 -45 -50 2 9 4 -1 -6 -11 -16 -21 -26 -31 -37 -42 -47 -52 4 5 -1 -7 -13 -18 -24 -30 -37 -43 -49 -55 -61 -67 6 3 -4 -10 -17 -24 -30 -37 -43 -50 -56 -63 -69 -76 8 1 -6 -13 -20 -27 -34 -41 -48 -55 -62 -69 -76 -83 10 0 -8 -15 -22 -30 -37 -44 -52 -59 -66 -73 -81 -88 12 -2 -9 -17 -24 -32 -39 -47 -54 -62 -69 -77 -84 -92 14 -2 -10 -18 -26 -33 -41 -49 -56 -64 -72 -79 -87 -95 16 -3 -11 -19 -27 -34 -42 -50 -58 -65 -73 -81 -89 -97 18 -3 -11 -19 -27 -35 -43 -51 -59 -67 -75 -83 -90 -98 20 -4 -12 -20 -28 -36 -44 -52 -60 -68 -76 -84 -91 -99 Figure 1. The Finnish Cold Guide. Figure 2. Example of given information. Results and discussion After the winter period the workers evaluated the information they had received about the different issues by using a scale of "a lot of new information - some new information - no new information". This feedback was given anonymously. The clothing trials were evaluated by open questions. The workers indicated that they had received new knowledge about the effects of cold and cold protection. The feedback about the given information is shown in Table 1. The workers' subjective evaluations about the clothing trials are shown in Table 2. The con- cept of multilayer clothing helped in maintaining the thermal comfort in changing work situations and reduced excess bulkiness of the clothing. Especially protection of the hands against the cold felt to be improved. Table 1. The workers´ feedback about the information blitzs given at their workplace (% of the answerers). The workers were asked to evaluate the new information they had received. A lot of new information (%) Some new information (%) No new information (%) 1. Underwear as a part of multilayer clothing 36 64 0 2. Mid-layer as a part of multilayer clothing 45 45 9 3. The water-repellency of outer garnments 55 45 0 4. The cooling effect of wind and protection against it 36 64 0 5. The effects of cold on physical performance 27 64 9 6. The effects of cold on health 45 55 0 7. Decreasing the contact cooling 45 45 9 8. Cold protection of the feet 27 45 27 9. Clothing as personal protective equipment 18 73 9 10. Rati-riti-ralla, Cold quide for the Finns 27 73 0 46 Table 2. The workers´ feedback about the clothing trials and immediate development actions conducted at the target site. The trial Products Subjective evaluations Underwear as a part of multilayer clothing underwear, 30 % PP / 70 % WO undershirt, mesh 100 % PP x very good: breathing, comfortable on the skin x mesh shirt in spring, mesh + polo shirt in cold x high collar protects well / collar too tight Mid-layer as a part of multilayer clothing pile mid-layer, 100 % PES x really warm x never had better x really warm: it makes you sweat when working The water-repellency of outer garments water-repellency treatment of outer garment x it works for a while x the garments are treated, but not yet in use The cooling effect of wind and protection against it construction worker's winter work clothing x protects angainst the wind x best I ever had, I would surely wear this throughout the winter season x a PP mesh shirt and a CO sweatshirt are sufficient underwear and mid-layer The cooling effect of wind and protection against it wind proof insulated vest x it was warm and allowed more room in sleeves, the clothing does not press the arms, it was easier to work than with a sweater Protection of the hands against cold and wetness, decreasing contact cooling knitted gloves, 100 % PP x good under rubber gloves (no cold sensations, no sweating) x good under rubber gloves when T a < -10 q C x keep the fingers warmer and dryer Decreasing contact cooling insulating tool handles with rubber tape x I could better hold the tools in my hands x non-slippery, insulated Cold protection of the feet sock with terry sole, 40 % WO / 40 % PAC / 20 % PA x the sole insulates well against coldnes x I already had similar ones, really good x warm, good Cold protection of the feet loose inner soles x keep the feet warm x reduced cold pain in my heels Abbreviations: CO cotton, PA polyamid, PAC polyacrylic, PES polyester, PP polypropylene, WO wool. Conclusions The training associated with clothing trials had an impact on the attitudes of the workers. The workers felt that they had received new information about the effects of cold and means for protection against it. The training encouraged the workers to take responsi- bility for their own well-being while working in the cold. In the follow-up survey the workers also showed willingness to participate financially in purchase of additional clothing and equipment for cold work. The project will proceed by developing an action model for assessing and managing the cold related occupational health and safety risks at construction sites. The model is integrated into the occupational health and safety manangemant system of the company. The development work standard proposal ISO NP 15743 "Working practices in cold en- vironments" is used as a framework for the action model. 47 References Hassi J, Juopperi K, Remes J et al. (1998) FINRISKI’97 Kylmäaltistusalaotos: Tutkimus suoma-laisten kylmäaltistuksesta, -haitoista ja kylmältä suojautumisen tavoista. Tutkimuksen toteutus ja perustaulukot. Oulun aluetyöterveyslaitos, raportti 4, Oulu 1998. ISONP 15743 (new project) Ergonomics of the thermal environment - Working practices in cold environments. Genève: International Organization for Standardization. Lappalainen J, Oksa P, Mattila M, Kivi P (toim) (1991) Miten selvitetään rakentamisen terveysriskit. Työpaikkaselvitysmenetelmät ja niiden käyttö. Helsinki: Työterveyslaitos 1991. 53s. Ruohotie P (1993) Ammatillinen kasvu työelämässä. Ammattikasvatussarja 8. Hämeenlinna 1993. 48 Properties of foul weather clothing for construction workers after use René Rossi, Markus Weder, Friedrich Kausch EMPA, Lerchenfeldstrasse 5, CH – 9014 St.Gallen, Switzerland Introduction Construction workers often have to use foul weather clothing to be able to work effi- ciently at their workplace and to prevent them from conditions injurious to health. One of the most important criteria of this sort of clothing is the protection against bad weather in all circumstances and a sufficient mechanical resistance. The aim of this project that was carried out jointly by the Centre for Safety Engineer- ing (Erkrath, Germany) and EMPA (Switzerland) was to analyse the properties of typical foul weather clothing used by construction workers in a new state as well as after use (Heffels et al., 1999). The main focus was put on watertightness of the jackets, which was assessed according to different methods. Standardised tests (hydrostatic pressure (EN 20811) and Bundesmann rain shower test (EN 29865)) and own developments like the rain tower test with a manikin were used. Methods Six jacket types were analysed in this study (Table 1), corresponding to clothing com- monly used by construction workers in Germany. For this study, two jackets of types 1, 2, and 3 were available and one of all the other types. Table 1. Description of the samples. Jacket Watertight layer Closure system Hood Damages in the used jackets 1N (new) 1U (used) PVC Push buttons with vertical gutter yes Stains on the outside, no visible damage 2N (new) 2U (used) Compact PU Buttons no Mildew stains, small damages in the coating 3N (new) 3U (used) Compact PU Buttons no Only light stains, no visible damage 4N (new) 4U (used) Compact PU Zipper yes Stains, one hole in the back region 5N (new) 5U (used) PES membrane Push buttons with zipper underneath yes Stains, Zipper torn out, no visible damage 6N (new) 6U (used) PTFE membrane Push buttons with vertical gutter yes Stains, no visible damage Several methods were used to assess the watertightness, the mechanical properties and the wearing comfort of the jackets:  Hydrostatic pressure (EN 20811) on the fabrics and on the seams  Bundesmann rain shower test (EN 29865)  Raintightness of the whole garment 49  Tear strength (ISO 4674 A2)  Seam strength (prEN ISO 13935-2) Raintightness of the whole garment This method (Rossi et al., 1998) consists of exposing the jacket to be tested, worn by a manikin, to a defined rain. The manikin is equipped with 22 humidity sensors and wears a cotton shirt underneath the jacket to get a good water uptake and to be able to make a visual assessment of the wet surface after the exposure to the rain. The tests were per- formed with three kinds of rain: cloudburst, persistent rain and drizzle. For the cloudburst (about 450 l/m2h), the rain comes down onto the manikin from a height of 10 m to get similar dynamics as in practice. The raindrops (diameter about 5 mm) correspond to those described in the Bundesmann rain shower test standard (EN 29865). For the two other types of rain, nozzles are placed at about 3 m height, which release drops of about 0.5 mm (40 l//m2h) for the drizzle and 1 mm (100 l//m2h) for the persistent rain. Results and discussion Watertightness according to the standardised test methods Table 2. Watertightness according to the hydrostatic pressure test and the Bundesmann rain shower test. Jacket Hydrost. Pr. (cm) [StD] Bundesmann (ml/h) [StD] surface mat. seam 1N - *) 379 [242] - *) 1U 123 [196] 33 [11] 3 [6] 2U 11 [4] 4 124.8 [135.1] 3U 500 [0] 0 0 [0] 4N - *) 500 [0] - *) 4U 18 [11] 134 [244] 21.1 [22.6] 5N - *) 408 [184] - *) 5U 151 [235] 33 [11] 0 [0] 6N - *) 500 [0] - *) 6U 15 [2] 15 [3] 1.1 [1.4] *) not tested The watertightness according to EN 20811 was assessed up to a hydrostatic pressure of 500 cm. In order to be able to calculate the mean value of the different samples, the value of 500 cm was used even if the sample had reached higher pressures. This test was only performed on seams for the new samples as it was admitted that the surface material was tight in any case (Table 2). Jackets 2N and 3N were not assessed, as their seams were not sealed. The Bundesmann rain shower test (EN 29865) was also only performed on the used jackets. For both test methods, the samples were taken from the shoulder region. The seam seals of the jackets 1N, 4N, 5N and 6N can be qualified as good, as they reach the best classification (class 3, without pre-treatment) according to ENV 343 (hydrostatic pressure  130 cm). Nevertheless, the standard deviation of jackets 1N and 5N are quite large as for each jacket, one sample got a lower result (16 cm for jacket 1N and 132 cm for jacket 5N). In the case of jacket 1, one seam was not properly sealed. The results of the used jackets were totally different: only jacket 3U was completely watertight. Jackets 1U, 4U (on seam) and 5U had each one sample with high water- tightness (respectively 472, 500 and 500 cm), explaining the high standard deviation. All 50 other results of the jackets were lower than 80 cm, which is the limit to reach the lowest class of water penetration (class 1) according to ENV 343. The consistently low results show that most of the shoulder region of all the used jackets was damaged. On the Bundesmann rain shower test, the used jackets 3U and 5U were tight. The result of jacket 5U shows that the hydrostatic pressure test is more stringent for the membrane than the Bundesmann test. It allows the detection of very small holes in the structure, not visible from the naked eye. For jacket 2U, a connection could be found between the watertightness and the surface covered by mildew, showing that mildew attacked the coating. The distribution of mildew was obviously not uniform, which explains the large standard deviation of this jacket. An informal test of the repellency rate on only one sample per new jacket showed that all the jackets reached low values: jacket 6N obtained a rate of 2-3; jackets 1N, 2N and 3N a rate of 2 and jackets 4N and 5N a rate of 1. This low rate could favour the water ingress through wicking effect if it gathered near the end of the sleeves or at the hem, or near a non-sealed seam. Raintightness of the whole garment The test with the cloudburst is very stringent and thus, a jacket that remains tight for one hour in this type of rain will withstand most of the weather conditions. Jackets 1N, 4N, 5N and 6N resisted the rain for over one hour (Figure 1). However, jacket 4N stayed only tight if the hood was covered by a plastic bag to avoid water ingress through the collar region. The hood seam of this jacket was not sealed properly; this is a quite common mistake of the making. Jacket 5N resisted 59 minutes in the cloudburst but the results were very different for the three repetitions of the test. This was due to the fact that water mostly penetrated through the zipper and the collar region, where the water penetration is very much dependent on how the jacket is placed on the manikin. The relatively bad results of the jackets 2N and 3N are not very astonishing, as the seams of these jackets were not sealed. However, the water ingress mostly occurred at the end of the sleeves, showing that water gathered at the lower ends and was wicked inside through the seams. 1 10 100 1000 drizzle pe rs is te nt rain cloudburs t T im e fo r w a te r pe n e tr a tio n in m in . 1 2 3 4 5 6 7 8 9 10 11 12 360 240 120 60 Figure 1. Time for water penetration with the 3 types of rain. 51 From all the used jackets, Nr. 1U and 3U showed the best results. All the others had water ingress after no longer than 5 minutes. Water mostly penetrated the shoulder region, showing a weakening of the material or a detachment of the seam seals. The ranking of the jackets remained approximately the same with the persistent rain and the drizzle, obviously with higher values. Only jacket 6U obtained a very good result with the drizzle, whereas it leaked very quickly in the cloudburst. This means that this jacket did not resist drops with higher kinetic energy. The drizzle had not a sufficient im- pact to press the drop through the jacket. A survey among the users of this type of clothing (Heffels et al., 1999) showed that they complained mostly about the tightness of seams, shoulder and collar regions. The tests with the manikin confirm that these areas (especially the shoulder region) are par- ticularly delicate for used jackets. Tear strength and seam strength All the jackets (apart from 2N, resp. 2U) showed a reduction of the tear strength for the used jackets. Jacket 2N obtained a significantly worse result in warp direction (i.e. the weft threads are torn) than jacket 2U, which raise the question whether the construction of both jackets was identical or if the two jackets had not the same slippage resistance of the warp-weft compound. The standard ENV 343 requires at least 25 N and even though there is a reduction during use, the used jackets all reach this value. There were no significant changes in the seam strength between the new and used jackets. In two cases (jackets 3U and 5U), the value of the used jacket was even higher than the new one. EN 343 requires seam strength of minimum 225 N, which was only reached by the two breathable jackets 5N and 6N (resp. 5U and 6U). Table 3. Mechanical properties of the jackets. Jacket 1N 1U 2N 2U 3N 3U 4N 4U 5N 5U 6N 6U Tear str., warp *) 60.5 47.8 56.7 58.7 60.9 46.9 - **) 44.0 64.4 55.2 74.7 61.8 Tear str., weft 40.8 43.6 39.9 65.2 44.0 28.1 42.7 27.7 35.6 27.9 52.9 52.3 Seam str. 105.0 111.9 199.4 187.5 224.1 361.8 184.3 154.0 374.1 586.0 308.6 294.6 *) weft threads are torn **) not tested Conclusions In this study, the tightness and the mechanical properties of foul weather clothing for construction workers were analysed in a new as well as in a used state. The results show that the used jackets, that are usually replaced after about two years are used for a too long period of time as they all offered insufficient rain protection, mostly in the shoulder region. As this area is the most exposed to rain, it would be sensible to reinforce this re- gion with a watertight material that offers a good resistance to ageing. The repellency of the jackets should in any case be improved to avoid any wicking of water to the inside of the jacket. The survey among the users revealed that they wished a better mechanical resistance of the material. As all the samples complied with the requirements of ENV 343 for the tear strength, the wishes of the end users should be considered during a revision of this stan- dard. It is nevertheless astonishing that all the compact coated jackets did not reach the minimum requirement of the seam strength. 52 References Heffels P., Hospach F., Rossi R. & Weder M. (1999) Wetterschutzkleidung für den Bau – Untersuchung zur Wasserdichtheit unter Berücksichtigung des Tragekomforts, Zentrum für Sicherheitstechnik, Erkrath, Germany Rossi R, Weder M & Rüedi M (1998) Praxisnahe Prüfungen der Wasserdichtheit von konfektionierter Bekleidung, 37. Internationale Chemiefasertagung, Dornbirn, Austria ENV 343 (1998) Protective clothing – Protection against foul weather [European Standard]. Brussels. Comité Européen de Normalisation. EN 20811 / ISO 811 (1981) Textile fabrics – Determination of resistance to water penetration- Hydrostatic pressure test [International Standard]. Geneva. International Organization for Standardization EN 29865 (1993) / ISO 9865 (1991) Textiles – Determination of water repellency of fabrics by the Bundesmann rain-shower test [International Standard]. Geneva. International Organization for Standardization ISO 4674 (1977) Fabrics coated with rubber or plastics – Determination of tear resistance [International Standard]. Geneva. International Organization for Standardization prEN ISO 13935-2 (1994) Textiles – Seam tensile properties of fabrics and made-up textile articles – Part 2: Determination of seam maximum force, Grab method [International Standard]. Geneva. International Organization for Standardization 53 Physiological optimisation of protective clothing for users of hand held chain saws Volkmar T. Bartels, Karl-Heinz Umbach Dept. of Clothing Physiology, Hohenstein Institutes, D - 74357 Boennigheim, Germany Introduction Users of hand held chain saws like forest workers, members of technical welfare organi- sations, or fire-fighters have to wear cut resistant protective clothing permanently or oc- casionally. However, up to now the available protective clothing is criticised or even re- fused by the wearers. This is mainly because of a too small water vapour permeability ("breathability") of the clothing system. It could hardly been used over a longer period of time without leading to a critical overheating of the body, especially during the warm season and with heavier physical activity. Therefore, in a research project the textile's construction and pattern of the single components (trousers with cut resistant inlays, un- derwear, foul weather suit) of a protective clothing system for users of hand held chain saws were investigated. The aim of the research project was to derive constructional guidelines for a cut resistant protective clothing system, which offers sufficient protection and noticeably better physiological wear properties as the products presently available. Within the research project, different aspects of wear comfort have been investigated: - The thermophysiological comfort describes, how the thermoregulation of man is sup- ported by clothing. It contains parameters like thermal insulation, water vapour per- meability, and buffering capacity against sweat. - The skin sensorial comfort deals with the direct mechanical contact between skin and textile. It is related to perceptions like softness or smoothness, but also to stiffness, scratching, or itching. - The ergonomic comfort is a matter of garment design. The pattern and fit have also an influence on the thermophysiological properties, because they influence convection and ventilation in the clothing system. More detailed descriptions can be found in (Umbach, 1987a; Umbach, 1987b; Umbach, 1993). Materials and methods In order to survey the thermophysiological properties of textiles, constructions from the actual production of textile manufacturers as well as possibly improved samples have been tested with the Skin Model (Sweating Guarded-Hotplate Test) according to ISO 11092 and internal standard-test specifications (Umbach, 1987a; Umbach, 1987b; Umbach, 1993; Bartels & Umbach, 1999a). These measurements simulate different wear situations from "normal" (insensible perspiration) up to "heavy sweating" (liquid sweat on the skin) including stationary as well as instationary experiments. 54 The skin sensorial properties of textiles have been tested by 5 specific apparatus, measuring the clinging to sweat wetted skin, the surface roughness, the number of contact points between skin and textile, the wettability, and the stiffness. Physiological tests of garment systems have been performed with the manikin "Charlie" (see Figure 1) measuring the thermal insulation. Combined with Skin Model experiments, these data also allowed to determine the water vapour resistance of the clothing system and to perform predictive calculations of the physiological im- pact on the wearer (Umbach, 1987a; Umbach, 1987b; Umbach, 1993; Bartels & Umbach, 1999b). Results and discussion The research project led to numerous results, how to optimise the physiological function of protective clothing for users of hand held chain saws. These were converted into constructional guidelines, some of which are compiled in Table 1. There, the innovative constructional parame- ters are given together with their physiological advantages. These results have been imple- mented in a prototype which is shown in Figure 1. In the following, some of these constructional guidelines are discussed. Cut resistant inlays Protective clothing for users of hand held chain saws contains a package of 9 to 10 layers of a synthetic woven, in order to be protected against injuries caused by the saw. Because of its thickness, this package has a high thermal insulation and a high water vapour re- sistance, which can cause a significant heat stress for the wearer. Hence, the thickness of the cut resistant inlay has to be reduced. From a pure physiological point of view, a reduction of the number of cut resistant layers would be most effective. However, in this case the level of protection becomes un- satisfactory, therefore a decreased number of layers cannot be recommended. A better solution to reduce thickness is a wide stitching, keeping the level of protection unaffected. In this case, the single layers cannot rub against each other and the inlay is not swelling up by the movements of the wearer or by washing. A typical example of a 10-layer inlay is given in Figure 2: Here stitching reduces the thermal insulation by 32 % and the water vapour resistance by 25 %, and improves the buffering capacity against va- porous sweat (moisture regulation index) by 10 %. Figure 1. Prototype of a physiologically optimised protective clothing system for users of hand held chain saws during the tests with the thermal manikin "Charlie". 55 Lining of the cut resistant inlay The cut resistant inlay is separated from skin by a lining. It has been shown during the project that currently available lining constructions are not optimal: Cotton linings are soaked up with sweat during high levels of activity and dry too slowly. On the other hand, synthetic linings are usually too smooth and have too many contact points with the skin, thus leading to clinging on sweat wetted skin and a poor skin sensorial comfort. In the research project, a new material has been proposed which is based on a polyester knitwear. The recommended textile is brushed on its in- side and hydrophilic. It has a very good transport of va- porous and liquid sweat, dries quickly, and offers a superior skin sensorial wear comfort. Outer fabric Similar problems as de- scribed with linings were also found for so far available outer fabrics of protective clothing for users of hand held chain saws: Cotton products were showing thermophysiologi- w ithout w ith 0 100 200 300 in s u m m e r therm al insulation w o rs e b e tt e r - 32 % R c t / 10 - 3 m 2 K W - 1 s titch ing w ithout w ith 0 10 20 30 - 25 % b e tt e r w o rs e w ater vapour resistance stitch ing R e t / m 2 P a W - 1 w ithout w ith 0.0 0.1 0.2 + 10 % b e tt e r w o rs e m o is ture regulation index F d s titch ing Figure 2. Physiological advantages of stitching a cut resistant inlay with 10-layers. Table 1. Overview on main results obtained in the described research project for the physiological optimisation of protective clothing for users of hand held chain saws. Constructional guidelines have been used for the prototype given in Figure 1. clothing component innovative construction physiological advantages cut protective clothes cut resistant inlay stitching low thermal insulation good breathability outer fabric PES/CO 65/35 215 g/m² 2/2 twill-weave brushed inside not hydrophobic good skin sensorial properties short drying time lining PES-knitwear brushed inside very good skin sensorial properties very high liquid sweat transport short drying time pattern with waistband (no dungarees) vents adjustable rubber band at waist pair of braces (no belt) more width low thermal insulation high ventilation rate wider range of utility high breathability underwear double face PES inside, CO outside high sweat transport short drying time foul weather suit 3-layer-laminate PTFE-membrane good breathability high buffering capacity against vaporous sweat 56 cal disadvantages, whereas synthetic materials offered an only poor skin sensorial com- fort, being especially too stiff. Most materials were hydrophobic, thus leading to a poor transport of liquid sweat. Therefore, in the project a 2/2 twill-weave PES/CO 65/35 mate- rial with a weight of 215 g/m² is proposed, which is again brushed on the inside and hydrophilic. It leads to a fast transport of sweat, dries quickly, and offers a good skin sen- sorial wear comfort. Pattern The pattern of the protective clothing system has been investigated as well. It has been shown, that a high ventilation rate can be obtained by a wider cut, using no belt but a pair of braces, and including vents. In the prototype (see Figure 1), the high ventilation rate can reduce the thermal insulation and the water vapour resistance by up to 25 %. Conclusions In the research project presented a systematic physiological optimisation of protective clothing for users of hand held chain saws was carried out. As a result, thermophysiologi- cal as well as skin sensorial improvements have been proposed and realised in a proto- type. The research project has provided the knowledge, how to construct the different components of this type of protective clothing, in order to achieve optimal physiological wear properties and additionally fulfil the necessary safety demands. Acknowledgement We gratefully acknowledge funding by the Federal Institute for Occupational Safety and Health, Dortmund, Germany, via Grant no. F1279. References Bartels VT & Umbach KH (1999a, March 24–26) Assessing the physiological function of protective clothing. Thermal Protection of Man Under Hot and Hazardous Conditions, Paris, France. Bartels VT & Umbach KH (1999b, October 12–13) Assessment of the physiological wear comfort of garments via a thermal manikin. 3rd International Meeting on Thermal Manikin Testing, Stockholm, Sweden. ISO 11092 (1993) Measurement of thermal and water-vapour resistance under steady-state conditions [International Standard]. International Standard Organisation. Umbach KH (1987a) Physiological tests and evaluation models for the optimization of the performance of protective clothing. In Mekjavic IB ed. Evironmental Ergonomics. London: Taylor and Francis Ltd. Umbach KH (1987b) Methods of measurement for testing physiological requirements of civilian, work and protective clothing and uniforms. Melliand English 68, E379–E383. Umbach KH (1993) Evaluation of the physiological impact of protective apparel on the wearer. 2nd International Symposium on Consumer Environmental Issues, St. Petersburg FL, USA. 57 The need for a rational choice of cold protective equipment in a refrigerated working environment Shin-ichi Sawada Division of Work Stress Control, National Institute of Industrial Health, Kawasaki, Kanagawa 214-8585, Japan Introduction In Japan, the frozen food chain stores have developed and the capacity of refrigerated warehouses has increased remarkably. Even after the Japanese bubble economy collapsed in 1992, the still high growth is shown (Figure 1). The setting temperatures in the re- frigerated warehouses have become lower and the cold environments below –20 °C have increased. Several studies (Tanaka et al., 1993; Tochihara et al., 1990) have been carried out until now to investigate the work environments in refrigerated warehouse industry and the health problems and physiological strains of the workers, but the details have not been fully clarified. The objective of the present study was to investigate the present state and problems in the refrigerated warehouse industry from the viewpoint of industrial health. To achieve this, we carried out a field survey for three refrigerated warehouses in Japan. Special attention was paid to how much and how long the cold workers are cooled during work and how they use cold protective equipment in the workplaces. 1000 1500 2000 2500 3000 3500 19 82 19 83 19 84 19 85 19 86 19 87 19 88 19 89 19 90 19 91 19 92 19 93 19 94 19 95 19 96 19 97 19 98 19 99 Year Ca pa cit y (T en th ou sa nd cu bic m ete rs) 3500 3700 3900 4100 4300 4500 4700 4900 N um be r o f U ni ts Capacity (Ten thousand cubic meters) Number of Units Figure 1. Recent change in the capacity and number of units of refrigerated warehouses in Japan (Jpn Assoc Refrig Storage, 1999) Methods A field survey for three refrigerated warehouses in Japan was carried out during March in 1999. The refrigerated warehouse “SA”, “YO”, and “KA” is located in Sapporo (the an- nual mean temperature is 8.2 °C; region near sub-arctic in Japan), Yokohama (15.2 °C; temperate zone region), and Kagoshima (17.6 °C; region near sub-tropical), respectively 58 (Table 1). To investigate the present state and industrial hygienic problems in the re- frigerated warehouse work, we interviewed the foremen of the refrigerated warehouses about the scale of the workplace, the working conditions, the working hours, the work type, the type and brand of work clothes and cold protective equipment which are used during work, the work/rest schedule, what is in trouble during work and to be improved, the complaints or health problems which are conspicuous for the cold workers etc. To estimate how long the workers are exposed to cold stress and how much they are cooled, skin temperature of peripheral parts of the body (finger and hand), body core temperature (oral temperature) and the ambient temperature of the vicinity of the workers were also measured for some of the workers from the start to the end of a working day. Results Characteristics of the refrigerated warehouse work As shown in Table 1, air temperatures in the refrigerated warehouses were set at below 20 °C and above 0 °C. The average age of the workers was 46.1 years in SA, 42.5 years in YO, 36.0 years in KA, respectively and a small proportion of workers (10 %) was made up of young persons below 30 years old. Table 1. The refrigerated warehouses investigated in the present study. Refrigerated Warehouse Region Number of cold workers Average and range in age of workers Setting temperature in the warehouses (°C) SA Sapporo (Sub-arctic) 10 46.1 (32-55) 0, -20, -21, -24.5 YO Yokohama (Temperate) 6 42.5 (28-57) 0, 5, -23 KA Kagoshima (Sub-tropical) 10 36.0 (22-48) 13, 1, -23, -25, -50 Main job of the workers was to carry frozen objects in or out of the refrigerated ware- houses by operating a forklift (forklift work) or by pulling a loaded hand cart (manual work). Cold exposure time per one entrance was shorter for forklift work than for manual work, but frequency of entrance into refrigerated warehouse was higher. The workers tended to work for hours without taking few regular rests except for lunch break (12:00- 13:00). There was some individual variation in methods of getting back body warmth during rest periods. Use of cold protective equipment The companies supplied both cold protective equipment and work clothes for the workers. The cold protective equipment consisted of insulated jacket, insulated trousers, helmet, hat etc. These were chosen by the foremen of the companies on the basis of their experiences without any information available for their thermal insulation level. Ther- mally low-insulated gloves such as cotton gloves were commonly used as a result of giving priority to the workability. Thermal loads and health problems The forklift workers were frequently exposed to large temperature differences by coming in and out of refrigerated warehouses many times in a working day (Figure 2). Especially 59 the facial cooling seemed to occur frequently in forklift-operating workers because their faces were not covered with any cold protective equipment during work. The manual workers tended to stay longer in the inside of the refrigerated warehouses than the forklift workers. The finger temperature in some manual workers decreased to below 10 °C during work (Figure 3). This appeared to be due to the fact that the manual workers usually loaded and unloaded the frozen objects directly by their hands during work. Despite this excessive finger cooling, the workers tended to have little complaints about pain and cold sensation. -30 -20 -10 0 10 20 30 40 08 :3 0 09 :0 0 09 :3 0 10 :0 0 10 :3 0 11 :0 0 11 :3 0 12 :0 0 12 :3 0 13 :0 0 13 :3 0 14 :0 0 14 :3 0 15 :0 0 15 :3 0 16 :0 0 16 :3 0 Time of Day Ex po su re te m pe ra tu re (° C) Lunch Figure 2. A typical continuous recording in an ambient temperature of the vicinity of the worker who is engaged in the forklift work in the refrigerated warehouse in a working day. -30 -20 -10 0 10 20 30 40 50 08 :3 0 09 :0 0 09 :3 0 10 :0 0 10 :3 0 11 :0 0 11 :3 0 12 :0 0 12 :3 0 13 :0 0 13 :3 0 14 :0 0 14 :3 0 15 :0 0 15 :3 0 16 :0 0 16 :3 0 17 :0 0 17 :3 0 18 :0 0 Time of Day Te m pe ra tu re (° C) Ambient temperature Finger temperature LUNCHREST REST Figure 3. A typical continuous recording in skin temperature of a finger and ambient tempera- tures of the vicinity of the worker who is engaged in the manual handling work in the re- frigerated warehouse in a working day. 60 Figure 4 shows the oral temperature change of the workers in the refrigerated ware- house “KA”. Four out of seven workers (worker 1,4,6,7) had excessive body core cooling as indicated by oral temperature below 36 °C. Despite this excessive body core cooling, some of the workers tended to have little complaints about subjective discomfort and cold sensations. As the main health problems, backache, hemorrhoid, hypertension, frostbite, and mu- cus during work were reported by the foremen and workers. Greater increase in subjective thermal loads during work in summer was complained by the workers even in the refrigerated warehouse “SA” of sub-arctic region as well as “YO” of temperate and “KA” of sub-tropical regions. This seemed to be due to the greater temperature difference between inside and outside of the refrigerated warehouse in summer. 34.0 34.5 35.0 35.5 36.0 36.5 37.0 37.5 08:30 10:00 11:00 12:00 13:00 15:00 17:00 Time of Day O ra l t em pe ra tu re (° C) Worker 1 Worker 2 Worker 3 Worker 4 Worker 5 Worker 6 Worker 7 Figure 4. Oral temperature changes of the seven workers in the refrigerated warehouse “KA”. Four out of seven workers (Worker 1,4,6,7) had excessive body core cooling as indicated by oral temperature below 36 °C. Discussion The recent problems in refrigerated warehouse work indicated by the present survey are (1) few young workers in the workplace, (2) longer working time and shorter resting time, (3) frequent and intermittent exposures to large temperature differences by frequent coming in and out of refrigerated warehouses, (4) common use of thermally low- insulated gloves as a result of giving priority to the workability, (5) increase in subjective thermal loads during summer, (6) excessive cooling of peripheral body parts during manual work, (7) intermittent facial cooling of forklift-operating workers with low meta- bolic rate, (8) individual variation in methods of getting back body warmth during rest periods, and (9) greater decline in oral and finger temperatures without subjective dis- comfort or cold sensation in some workers. Problem (9) is consistent with the results of our previous laboratory studies (Sawada et al.,2000; Sawada et al.,1998) which examined the effect of intermittent cooling of fingers and the whole body. The findings of the pre- sent study strongly suggest the need for a rational choice of cold protective equipment, which are not dependent on the workers’ subjective judgement in refrigerated warehouse work. 61 References Jpn Assoc Refrig Storage (1999) Overview of the Public Refrigerated Warehouse Industry in Japan. Sawada S, Araki S & Yokoyama K (2000) Changes in Cold-induced Vasodilatation, Pain and Cold Sensation in Fingers Caused by Repeated Finger Cooling in a Cool Environment. Industrial Health, 38, 79-86. Sawada S & Kobayashi T (1998) Thermal Comfort, Thermal Sensation, and Associated Behavioural and Physiological Responses during Alternately Repeated Exposure to Cold and Warm Environments. Proceedings of Second International Conference on Human-Environment System, 279-282, Yokohama National University, Japan. Tanaka M, Nakamura K, Shimai S & Takahashi H (1993) Work at Cold Storage. Journal of Thermal Biology, 18(5/6), 647-650. Tochihara Y, Ohkubo C, Uchiyama I, Komine H, Honda Y, Aoki N, Ishibashi K, Ohno R, Shimozono N & Nakajima H (1990) A Survey on Work Loads of Forklift-truck Workers in Cold Storages. Bulletin of Institute of Public Health, 39(2/3), 29-36. 62 Diversified design needs of personal protective devices and clothing in cold climate: An example in the design needs of protective outdoors winter shoes John Abeysekera Division of Industrial Ergonomics, Luleå University of Technology, S-971 87 Luleå, Sweden Introduction The use of personal protective devices (ppd) and clothing have become essential in many hazardous work situations. But unless the workers wear the ppds during the maximum period of exposure, protection from the hazards is not guaranteed. Many surveys in the use of ppd have revealed that they have not been adequately worn by the workers due to mainly the discomfort caused by the lack of wearability needs in their design (Abeysekera and Shahnavaz, 1988). During the past 10 years there has been a con- tinuously increasing interest to improve the human factors or user needs of ppds. This is clearly evident from the Conference Proceedings of NOKOBETEF during the past 16 years. From a very negligible number of papers presented on ergonomics or wearability aspects in NOKOBETEF 1, 2, and 3, over the years human factors of ppds have gained importance. The current NOKOBETEF 6 and the First European Conference on Protective Clothing has as its theme 'Ergonomics of Protective Clothing'. The breakdown of the papers presented in NOKOBETEF is shown in Table 1. Table 1. From papers presented in NOKOBETEF Symposia over the years. Year and Number On Protective Aspects On Ergonomic Aspects 1984 NOKOBETEF 1 21 3 1986 NOKOBETEF 2 11 3 1989 NOKOBETEF 3 26 3 1992 NOKOBETEF 4 36 6 1997 NOKOBETEF 5 31 7 2000 NOKOBETEF 6 46 16 In safety helmets used to protect from falling objects in indoor climates the protection needed is for one type of hazard. The wearability needs e.g. fitness, low weight, ventila- tion, etc., for such helmets usually are not difficult to achieve (Abeysekera et al., 1997). However when ppds are used in the cold climate, apart from the protection from a work hazard the ppd must also protect the worker from the cold exposure. While trying to achieve both protections the wearability needs of such ppds become very difficult to in- corporate (Bergquist and Abeysekera, 1994a). A safety helmet worn outdoors in cold climate must have the possibility to protect the head from impact injuries as well as pro- vide insulation of the head from the cold. In trying to achieve both demands the correct fitting of the helmet is affected. Ppds used in cold climate can become clumsy and un- comfortable for the wearer in trying to achieve the double protection demand. In safety gloves used in cold environments the wearer must be protected from impact hazard or chemical hazard as well as from the cold exposure of the hands. Achieving both protection demands can result in significant loss of hand dexterity (Geng et al., 1997). 63 In very rare situations one type of ppd has to meet the 3rd protection demand. A good example is safety footwear worn outdoors on snow and ice. The three protection demands are: protection from foot impact injuries, protection of feet from exposure to cold climate and protection from slips and fall accidents. To satisfy the 3 demands, a steel toecap, in- sulated shoe material and high friction sole material respectively are the basic design needs. Providing wearability needs for such ppds is a challenge to the shoe designers. These needs are so diversified and appropriate priorities obtained from research have to be adopted in such designs. An example using protective footwear in cold climates Working under cold conditions present more problems than working in indoor climates. The two protection needs viz. protection from impact injuries and protection from the cold lack co-ordination in their demands. This will cause the wearing of such shoes clumsy and bad fitting. A questionnaire survey was conducted by Bergquist and Abeysekera (1994b) among 100 outdoor workers in the north of Sweden, 100 manufac- turing companies from 12 countries and 15 experts. The respondents were asked to rank the importance of 11 ergonomic demands when designing safety shoes for the cold cli- mate. 90% of users, 20% of manufacturers and 75% of the experts responded to the questionnaire. The ergonomic demands were ranked from the most demanded to the least demanded in the following order: Fitness: Thermal Comfort: Protection from work hazards: Low weight: Anti slip: Mobility: Ease to don / doff: Durability: Adjustability: Good appearance and Moderate price. The extremities viz. hands and feet are affected by cold exposure as they are in fre- quent contact with cold surfaces. Therefore both design needs e.g. protection from impact as well as protection from cold injuries are important. In consideration of thermal properties of winter shoes, factors that are important are heat conductivity of material, avoiding dampness of the fabric and allowing air trapped inside the fabric for insulation. A steel toecap is incorporated in most safety shoes for protection of the feet from impact injuries. Experiments and surveys carried out in the past on thermal effects of steel toecaps have revealed contradictory results. Påsche et al (1990) reported no cooling effect from steel toecap. A questionnaire survey conducted by Bergquist and Abeysekera (1994b) on pro- tective shoes revealed that a major problem regarding thermal comfort was the alleged cooling effect of the steel toecap. However Kuklane and others (1999) in their investiga- tions have observed small differences of insulation of some types of boots with and with- out steel toecaps. They have also shown that steel toecaps may show an "after-effect" which may be caused by higher mass and thermal inertia of the toecaps, which affects cooling of the feet. The most demanded ergonomic need of the users was fitness of the shoe (Bergquist and Abeysekera, 1994b). The fitness has to be adjusted to the insulation needed, which in turn depends on the air trapped inside the fabric and between foot and shoe. Therefore shoes must be big enough to accommodate thick socks. Shoes must also allow the pumping effect for the air-layer inside the shoe to escape out which is also influenced by the shoe fitness. It has been reported that increased shoe weight can increase the work load of the wearer (Smolander, 1989; Jonas et al., 1984). Walking on snow layers can also be physi- cally demanding (Pandolf et al., 1976). 64 Slips and falls on ice and snow The third and by no means the least protection need in safety outdoor winter shoes is the prevention of slips and falls. The major focus on current research on protective footwear for cold environments has been on thermal insulation (Kuklane, 2000). Prevention of slips and falls however has not caught much attention. The available research studies on slips and falls have concentrated more on improving the friction of shoes on oily and wet shop floor surfaces. Considerable interest on the design of anti-slip winter shoes on snow is fast developing as in Nordic region, slips and falls on icy roads are a major problem during winter. Recently several studies have been carried out on shoes used on ice and snow and researchers have come up with information on the aetiology of slips and falls, which have to be considered by the designers of such shoes. Poor grip and low friction between footwear and the underfoot surface have been considered as major factors for slips which calls for increasing the friction of the shoes as well as the walking surfaces (Grönqvist, 1995; Gao and Abeysekera, 1999). Many other factors have been revealed in studies that contribute to prevention of slips: viz. postural control and gait (Grönqvist, 1995), perception of risk, hardness (low) of sole (Leclereq et al., 1994), roughness to in- crease the friction (Manning et al., 1990). Studies on tread pattern and contact area have not yet given conclusive results (Stevenson, 1997; Grönqvist, 1995). Recent studies car- ried out by Gao (1999) have shown no significant effect from the Center of Gravity (COG), hardness, roughness and sole contact area (static and dynamic) on the coefficient of friction (COF) of the shoes. Experiments by Gao and Abeysekera (1999) have clearly indicated that safety shoes, which are to be used outdoors in winter, are not adequately provided with anti-slip sole material. Results of observations of 25 subjects walking on five types of icy roads wearing 4 types of shoes (4x5x25 trials) on ice at 0 0C are shown in Table 2. COF values measured on 4 different pairs of miscellaneous shoes and 3 types of safety shoes showed that COF values of safety shoes (Group B) were significantly lower (p<0.0006) than COF of miscellaneous shoes (Group A, Table 3). It can be seen from the number of occasions of tendency to slip (during 500 trials) and low values of COF in safety shoes that the designer of safety shoes has incorporated no special anti-slip precautions. It is observed from recent research on slips and falls on ice and snow that unlike the design demands for protection from impact injuries and insulation from cold climates, the design demands for prevention of slips and falls in safety shoes are seldom provided. Table 2. Results of tendency to slip. Shoe Type Times tendency to slip/fall Winter shoes 156 PU occupational footwear 175 Safety boot with steel toe-cap 150 Farmers protection boot 136 PU: Æ Polyurethane Table 3. Comparison of COF values. Group A COF Winter with metal stud 0.14 Shoes with anti-slip fibre component 0.18 Work shoes 0.18 Normal footwear 0.13 Group B Safety shoes with nitrile soling 0.09 Safety boots 0.09 Safety shoes with PU soling 0.09 Group A: Æ Winter, work and normal shoes Group B: Æ Safety shoes with steel toecap 65 Discussion and Conclusions Users frequently demand the comfort and wearability of ppds if not provided will make them very unpopular. Protection and comfort are contradictory demands. When the same ppd has more than one protection to provide it becomes harder for the designer to provide human factor needs. In a questionnaire conducted on users of ppd in the cold climate the following complaints have been made (Bergquist and Abeysekera, 1994a): Cold sensa- tion from steel toecap; lack of dexterity from gloves; difficult to work with large gloves; helmets are heavy and not fitting; walking difficulties with heavy safety shoes and trauma from the shoes. From the example of the use of safety shoes used outdoors in the cold the diversified design demands in protection and their influence on wearability are shown in Figure 1. Protection Demands Design Demands Influence on Wearability Impact Injuries Steel Toe Cap Thermal Discomfort Falling Objects Thick Shoes Heaviness Thick heels & soles Walking Difficulty Cold Climate Insulation Cold Injuries Heat conductivity Fitness Permeability Thermal Comfort Leak proofness Pumping effect Slips and Falls Friction Weight Roughness, Hardness Perception of risk Tread pattern Gait Contact area Postural balance Durability Figure 1. Protection and design demands and their influence on wearability of winter safety footwear Regarding the design and use of ppds Crockford (1979) quoted as follows: "From the designers point of view he designs a shell or a piece of environmental armour which keeps all the nasty aspects of the environment away from the wearer. From the wearer's point of view, the designer places a barrier around him which makes work more diffi- cult." From the above discussion it will be necessary to complement Crockford's state- ment with the following: "From the ppd Standard Committee's point of view the main aim is to achieve a distinct level of protection performance to protect the user from injury or disease. From the researcher's point of view research in ppds has to be developed in a specific area e.g. protection, comfort etc, to achieve the maximum possible level. Lastly in the human factor specialists point of view a ppd must achieve optimum protection at optimum comfort or wearability to the wearer. Some of the demands are contradictory with each other. Therefore future research in ppds should be directed towards an integrated or systems approach to achieve the opti- mum protection as well as optimum wearability. Researchers, users and designers should adopt an iterative process in designing ppds requiring more than one type of protection. Standard setting committees and designers must take cognisance of contradictory protec- tion demands and their influence to wearability needs of ppds 66 References Abeysekera J & Shahnavaz H (1988) Ergonomic evaluation of modified industrial helmets for use in tropical environments. Ergonomics, 31(9), 1317-1329. Abeysekera J, Holmér I, Liu X, Gao C, and Wu Z (1996) Some design recommendations to improve comfort in helmets: A case study from China. Journal of Human Ergology, 25, 145-154. Bergquist K & Abeysekera J (1994a) Research needs to improve wearability of personal protective devices and clothing (ppds) in the cold climate. Proceedings of Polar Tech '94, Luleå, Sweden, 369-375. Bergquist K & Abeysekera J (1994b) Ergonomic aspects of safety shoes worn in the cold climate. Proceedings of the 3rd Pan Pacific Conference on Occupational Ergonomics, Seoul, Korea, 590- 594. Crockford GW (1979) Physiological problems of protective clothing. In: Clothing and Working Man - The ergonomics of protective clothing in relation to the thermal environment. The Ergonomics Society, 75 Swingbridge Road, Loughborough, Leicestershire, UK, 1730. Gao C (1999) The anti-slip effect and wearability assessment of footwear on icy surfaces. Masters thesis, 107 CIV, Division of Industrial Ergonomics, Luleå University of Technology, Sweden Gao C & Abeysekera J (1999) Slip resistant properties of footwear on ice. Proceedings, 10th Year Anniversary Ergonomics Conference, Luleå University, Sweden, 117-123. Geng Q, Chen F. and Holmér I (1997) The effect of protective glove on manual dexterity in the cold environments. International Journal of Occupational Safety and Ergonomics, Vol. 3, No. 1-2, 15- 29. Grönqvist R (1995) A dynamic method for assessing pedestrian slip resistance. Research Report, Finnish Institute of Occupational Health, Helsinki. Jones BH, Toner MM, Daniels WL, Knapik JJ (1984) The energy cost and heart rate response of trained and untrained subjects walking and running in shoes and boots. Ergonomics, 27(8), 895-902. Kuklane K, Geng Q & Holmér I (1999) Thermal effects of steel toe caps in footgear. International Journal of Industrial Ergonomics, 23(5-6), 431-438. Kuklane, K (2000) Footwear for cold environments - Thermal properties, performance and testing. Doctoral Thesis, 1999:23, ISBN 91-7045-540-6, Luleå Univ. of Technology & National Institute for Working Life, Sweden. Leclereq S, Tisserant M, and Saulnier H (1994) Slip resistant footwear: A means for the prevention of slipping. In F. Aghazadeh (eds), Advances in Industrial Ergonomics and Safety VI, 329-337. Manning DP, Jones C & Bruce M (1990) Proof of shoe slip-resistance by a walking traction test. Journal of Occupational Accidents, 12, 255-270. Pandolf KB, Haisman MF, Goldman RF (1976) Metabolic energy expenditure and terrain coefficients for walking on snow. Ergonomics, 19, 683-690. Påsche A, Holand B, Myrseth E (1990) Cold climate work clothing. Proceedings Polartech '90 International Conference on development and commercial utilisation of technologies in polar regions, 325-334. Smolander J, Louhevaara V, Hakola T (1989) Cardiorespiratory strain during walking in snow with boots of different weights. Ergonomics, 32(1), 3-13. Stevenson M (1997) Evaluation of slip resistance of six types of women's safety shoe using a newly developed testing machine. Journal of Occupational Health Safety - Australia and New Zealand, 13(2), 175-182. 67 Footwear for cold work: a limited questionnaire survey Kalev Kuklane1, 2, Désirée Gavhed1, Eva Karlsson1, Ingvar Holmér1, John Abeysekera2 1 National Institute for Working Life, Respiratory health and climate, Solna, Sweden 2 Luleå University of Technology, Dept. of Human Work Sciences, Luleå, Sweden Introduction This questionnaire survey was initiated to acquire information for preparation of a series of field studies (Gavhed, Kuklane & Holmér, 1999a; Gavhed, Kuklane, Karlsson & Holmér, 1999b). The paper deals with questions that were related to the thermal responses of the feet and footwear performance. Other data is shown for reference and background. Materials and methods A brief questionnaire was sent out to various companies about their workers’ cold expo- sure and if they were interested to take part in a more thorough study about work in cold. Three companies answered positively and to them was sent a thorough questionnaire (Gavhed et al., 1999b). Thirty questions were directly related to feet and/or footwear. Results and discussion Work organisation and background data In total 43 questionnaires out of 66 were returned: 18 from a harbour company (HC), 16 from a telecommunications company (TC), 8 from a customs office (CO) and 1 from a construction worker working at the institute by contract. The persons represented a wide variety of jobs that have to be carried out in cold working conditions. The age of the workers was between 21 and 60 years. In average they had worked in the company for 18.510.4 years. The work load for an ordinary day was rated as very light by 18.7 % of the respon- dents, light 23.2 %, moderate 21.5 %, heavy 29.3 % and very heavy 7.3 %. 57 % of the work time was spent in cold environment. The total amount of cold exposure during a full working day depended on the environmental temperature and was at average more than 4 hours at 0 °C, about 4 hours at -10 °C and 2-4 hours at -20 °C. The length of cold exposure was dictated by the job needs. Temperature of -20 °C was considered enough severe to limit the exposure time. The length of one work period in cold was usually 1-2 hours. At -20 °C it was under 1 hour, except for the CO where work periods were always around 1-2 hours. It was possible to take a warm break when a person felt cold, but there could be as well occasions where there was not possibility to take a break because of feeling cold. Sometimes the workers themselves did not choose to take a break for warming up even when there was a possibility. Generally, there were breaks in warm lo- cations about 3 times per day. The length of an average break was about 30 minutes. If there were more breaks, then these were usually shorter. There were seldom (1-2 times) 68 tasks at warm locations. At the CO some tasks were done in warm locations up to about 5 times per day. The work periods in warm places usually lasted 15-60 minutes. Most of workers (72 %) considered the wintertime to be the most troublesome season regarding the footwear. Cold was the main reason for that. The other reasons were wet- ness, snow and ice, bulkiness (weight) of the winter boots that makes it difficult to carry out some tasks (reduced moving ability), darkness. Two workers (<5 %) considered the spring and fall periods as the worst. One of them considered in his answer the difficulty to choose the right footwear for work due to the quick weather changes. Nine workers did not answer the question and 1 had not had any trouble with the footwear. The footwear Only one of 43 workers said that he did not use the safety shoes (footwear with steel toe cap). However, 8 (3 from TC and 5 from CO) said that the safety shoe was not compul- sory (including one who didn’t wear them) and one did not answer on the question. Generally the same pair of the footwear was worn longer than 6 months (44 % 6-12 months, 28 % 1-2 years and 26 % more than 2 years). 67 % used the same safety foot- wear every day and 30 % used the same footwear both during summer and winter. At CO all wore the same footwear every day (at HC 2/3 and at TC more than half) and only one used different boots during summer and winter. At the TC only two persons reported use of the same footwear during summer and winter (HC 5). At the HC most of the workers (12) used footwear with warm lining during winter. Half of them used also footwear without lining occasionally (5) or regularly (4). The TC workers used warm boots and one of them used footwear without lining occasionally while at the CO all the workers regularly wore footwear without lining. They had high leather boots with a zipper and/or a cord. At other companies the leather was the most common footwear material, too. However, some footwear of synthetic material (3 at both) or rubber boots (2 at HC and 1 at TC) were used as well. At TC the used footwear was usually of boot type, while at HC on 50 % of cases the shoes were used. The most common high footwear was with the zipper, while low shoes commonly had cords. General problems with cold Most of the persons reported trouble with cold. Often the cold became most unpleasant when it was connected with wind (23 % always, 40 % at least once per week and 23 % 1- 2 times per month). The most common cold body parts were feet and toes (84 %) fol- lowed by hands and fingers (79 %) and face with its parts (ears, chin, cheek etc., 79 %). The cold sensation was the severest in hands and fingers, followed by feet and toes and then face and its parts. The legs were the next to feel cold. The three most highly ranked factors for cold sensation of feet were boot material (42 %), contact with cold surfaces (37 %) and work at varying activity levels (14 %). In addition, the jobs that require long periods of standing still were mentioned as a factor for cold feet. Some workers (14 %) had a cold injury at work and some (14 %) during free time. One person reported a cold injury both at work and free time at the same location. Some had cold injury at two or more locations. The listed sites were toes (5 cases), fingers (3), ears (3), nose (1), under chin (1), eyelid (1), cheek (1) and an unspecified location on face (1). Some workers had stumbled because their feet were cold (16 %). 50 % of workers were satisfied with their clothing. However, this picture was different in various companies. At HC 40 % answered that their clothing does not protect well enough against cold. At TC the percentage of dissatisfied was 30, while none from CO 69 was satisfied with cold protection of their clothing. The footwear was considered to be the worst (30 %) followed by outer garments and gloves (both 23 %). At the companies the dissatisfaction was different: at HC (18 workers) the three biggest problems were footwear (33 %), gloves (28 %) and outer garments (22 %). At TC (16) there was most few complains on clothing: footwear 19 %, gloves 13 % and on socks, cap and gloves all 6 %. The most dissatisfied were CO workers (8): 75 % complained on outer garments and 50 % on footwear and gloves. At all companies 44 % used thick warm socks. Extra insoles were used by 19 %. Some of the complaints were cold steel toe cap in footwear and “footwear too small to allow additional insoles or socks”. Problems with footwear in cold 47 % of the workers (HC 50 %, TC 31 % and CO 75 %) had experienced problems with footwear in the cold climate. This was well related to the type of the used footwear. Twenty one workers gave their comments on what could be the problem. Nine of them (43 %) directly point out steel toe cap or other steel objects in footwear as a cause of cold feet. Other problems were: cold soles (6), slippery soles (3), shoes were experienced colder than boots (however, some jobs, e.g. driving a forklift truck, were seen more comfortable with them), jobs that require long periods of standing still, fresh snow after heavy snowfall (more than 20-30 cm), missing a good support at ankle, too low footwear, not insulated footwear (especially CO) or just cold itself. Finally, there was a suggestion to use boots with cords that fit better around the ankle instead of ordinary leather boots. The most common problems connected with footwear were feeling cold in feet (30 of 43) followed by risk to slip (16) and wet (15) or sweaty (12) feet. The main reasons for cold feet was seen to be footwear without warm lining or bad insulation (3), steel toe cap in safety footwear (3), work that requires long periods of standing still (2) or low sole insulation (1) and severe cold in itself (1). The supposed reasons for slipping were wrong choice of the sole, incorrect soles and that the rubber becomes stiff with cold. Stiffness and missing good support at the ankle joint were the problems connected with boot mate- rial and construction (2). Slippery working/walking surfaces were mentioned by 2 workers. Wet or sweaty feet were related to water from outside (snow, water and rain, 3) or inside (sweating, 1), shifting between cold and warm locations during day (1), no ven- tilation in footwear (1), wrong choice of footwear (1) and no insulation (1). Bad fit (9) was related to not enough tight around shin, bad support for ankle, stiffness, wrong choice of shoes, can’t find the shoe fitting particularly one’s foot (all 1). Stiffness was brought out separately on 8 cases. Difficulties to walk (6) and restricted mobility (6) were related to stiffness (3), clumsy shoes, cold and slipperiness (all 1). Five (5) workers thought that their shoes affect the working and one (1) related it to the need to be more careful. Four (4) workers had had chafed feet. For one of them the reason was a hard heel-cup. One (1) mentioned the difficulty to put on/take off the shoes. Because of miss- ing good support at the ankle and slipperiness the footwear was considered not to give enough protection against injuries (1). One (1) recommended laced boots against sprain- ing and 1 mentioned the footwear tightness to be the reason for wet feet. Slipping Slipping was considered as the next biggest problem after cold. 30 (70 %) workers had slipped and 22 (51 %) had fallen during last two winter seasons. Only one of 43 workers had used some kind of anti-slip device and even that was used seldom and when it was a part of the safety footwear. Another worker mentioned that he used the boots with ribbed 70 soles. Both of them thought that such footwear functions quite well. Of these 30 workers who had slipped, 6 had slipped often (at least once per week), 12 sometimes (1-3 times per month) and the rest seldom. Of the 22 workers who fell half had fallen sometimes and half seldom. In five cases it had ended with an injury: ligament injury, caudal vertebrae contusion, muscle strain and tooth injury, foot sprain and lumbago. The biggest slipping risks differed somewhat at various workplaces. At HC the highest risk was connected with slippery deck (2), stepping on and off the machines (2), climbing up to crape off ice from machine windows, new machines, on top of the containers, under the crane and at small height differences (all 1). At TC the highest risks were connected with icy masts (4) and climbing up the mast, walking at parking lots near stations or just on ground (all 1). At CO were the problems connected with climbing on and off the trucks (4), oily ter- minals, slippery deck and brisk walk (all 1). The workers recommended to use salt (2), sand, chip the ice and shovel it away and better soles (all 1). Some proposals on how to improve the situation were given: to use laced boots (simi- lar to hiking boots) that are more tight around the ankle and shin area (2); there are avail- able good boots on market, but the company (employer) provides just one type (2); pur- chase boots with better quality (2); boots with warm lining needed (2); use another mate- rial for toe cap instead of steel; cover the seam at sole with sole material (rubber) to re- strict water entering the footwear; use materials with high friction (tested for oily and icy surfaces) for soles; use materials of GoreTex type on sides to improve ventilation (all 1). Conclusions According to the respondents the problem with footwear was low insulation that resulted in cold feet. This supports the need for further research and development of footwear. Often the companies do provide footwear to their workers. Better communication be- tween workers and employers could be helpful for purchasing footwear according to the workers’ needs. Simultaneously, it requires more knowledge and information on products and insulation requirements in various conditions. Slipping was ranked second after cold by the workers. Half of the respondents had fallen due to slippery surfaces or soles. Some of the falls were accompanied by serious injuries. Further research in this area is needed to reduce the risks and injuries. Acknowledgements Thanks to the Port of Göteborg AB, TERACOM AB and the Swedish Board of Customs, and their workers for co-operation. Special thanks to Taiga AB for help to establish contacts. References Gavhed D, Kuklane K & Holmér I (1999a) Work on masts in cold (Mastarbete i kyla). Arbetslivsrapport 1999:22, Solna: National Institute for Working Life. Gavhed D, Kuklane K, Karlsson E & Holmér I (1999b) A pilot study on work in the cold - questionnaire survey and field study (En pilotstudie om arbete i kyla - frågeundersökning och fältstudie). Arbetslivsrapport 1999:4, Solna: National Institute for Working Life. 71 Footwear for cold work: a field study at a harbour Kalev Kuklane, Désirée Gavhed, Eva Karlsson, Ingvar Holmér National Institute for Working Life, Respiratory health and climate, Solna, Sweden Introduction Quite a few laboratory studies dealing with feet and footwear have been reported (Kuklane, Geng & Holmér, 1998; Rintamäki & Hassi, 1989; Tochihara, Ohnaka, Tuzuki & Nagai, 1995). Field studies of cold work are more rare and are often related to jobs in cold stores (Enander, Ljungberg & Holmér, 1979). Needs to investigate the thermal con- ditions with special regards to the feet at various cold workplaces initiated a series of field studies (Gavhed, Kuklane & Holmér, 1999a; Gavhed, Kuklane, Karlsson & Holmér, 1999b). The studies were expected to give information on:  how the results from various laboratory experiments could be used in practice,  how to design future tests that will be directly related to actual working conditions,  how to validate and improve existing prediction models and develop new requirements. This paper deals with a study at a harbour and the main attention is paid to the thermal responses of the feet and footwear performance. Other data are shown as reference and background data. Materials and methods The study was carried out in February (Gavhed et al., 1999b). February is usually one of the coldest months in Sweden. However, on the measuring week the weather was rela- tively warm and calm. Eight male workers were the subjects: 31-62 years, 89 kg (70-104), 180 cm (170-187). They were selected among volunteers who carried out four different jobs in pairs. Two persons (construction workers) were cutting asphalt (oxygen consumption (VO2) 0.91 and 0.94 l/min), 2 were securing steel to platforms for loading to ships (VO2 1.23 and 1.35 l/min), 2 were securing paper to platforms for loading to ships (VO2 0.95 and 1.30 l/min) and 2 were controlling the loading of the trailers to the ships (VO2 0.85 and 1.29 l/min). The mean air temperature (Table 1), globe temperature, wind velocity and relative hu- midity for each job and day were following: cutting asphalt 4.0 °C, 7.6 °C, 2.5 m/s (max 5.2 m/s), 56 %, securing steel to platforms 7.3 °C, 8.6 °C, 3.7 m/s (max 11.6 m/s), 70 %, securing paper to platforms 6.7 °C, 7.3 °C, 0.14 m/s (max 0.3 m/s), 68 % and signalling during loading the trailers 8.0 °C, 0.6 m/s (max 2.5 m/s), 79 %. The workers could choose clothing of their personal choice from the company’s store. All the clothes were weighed separately in the beginning and at the end of the workday. The clothing was manufactured by Fristads AB and Taiga AB Sweden. The warm winter boots were provided by the experimenters and they were model 520 (Woodman) manu- factured by Stålex, Arbesko Gruppen AB. The insulation of a boot (size 41) was meas- ured on a thermal foot model (Kuklane & Holmér, 1998). The insulation when standing dry with a load of 35 kg and a sock was 0.34 m2°C/W. As the sizes 45 and bigger were not present, one subject used his own footwear. His shoes were Stålex model 732. The 72 shoes were without warm lining. The insulation of the shoe was estimated to around 0.23 m 2 °C/W. Two types of socks were in use: A. 67 % wool, 26 % nylon and 7 % lycra (Arbesko, model 106), B. 80 % cotton, 20 % stretch. Subjects 1-5 used sock A. The others used sock B. Temperature sensors (StowAway temperature logger, Onset Computer Corporation) were taped to the second toe and dorsal foot. The workers were observed during the whole workday. Their subjective responses on thermal sensation were recorded and at the end of the day they filled in a questionnaire on that particular day (Gavhed et al., 1999b). Four questions were directly related to feet and/or footwear. Results and discussion The data are shown in Table 1. As the weather was relatively mild throughout the whole study period, the toe and foot temperatures stayed at high levels in the insulated winter boots. Only one subject with his own shoes experienced cold feet. Table 1. The footwear and sock weight, foot and toe skin temperature (mean (min-max)), air temperature (mean (min)), moisture accumulation, oxygen consumption and thermal sensation in feet (mean over work period (lowest during work period)) data of the subjects. Data was ranked by foot skin temperature. Footwear (subject) Footwear (sock) weight (g/pair) Foot temp. (°C) Toe temp. (°C) Air temp. (°C) Total moisture . accumulation (g) Thermal sensation Shoes (7) 1447 (80) 29.2 (27.8-30.8) 21.5 (17.9-25.3) 8.0 (7.3) 16.6 -0.7 (-1) WS (2) 1784 (79) 29.9 (28.3-30.9) 19.7 (17.3-22.9) 4.0 (0.3) 32.5 2.2 (2) WS (3) 2021 (74) 32.7 (31.5-34.2) 22.1 (19.8-25.1) 7.3 (6.7) 23.6 0.2 (0) WS (4) 1764 (64) 32.7 (31.3-34.3) 27.0 (20.1-33.5) 7.3 (6.7) 60 2.0 (2) WS (1) 1756 (81) 33.1 (32.3-34.6) 28.6 (21.9-33.9) 4.0 (0.3) 44.3 1.8 (1) WS (5) 1863 (82) 33.2 (30.4-34.6) 29.1 (19.8-33.6) 6.7 (5.5) 27.8 1.7 (1) WS (6) 2021 (70) 33.2 (30.9-33.9) 30.3 (23.6-32.0) 6.7 (5.5) 59.7 1.0 (0) WS (8) 1736 (82) 33.4 (29.8-35.5) 31.7 (25.4-35.0) 8.0 (7.3) 51.2 1.5 (0) Questionnaire at the end of the day During the day the workloads varied and for each type of job the workload was some- what different. In average the test persons rated their exertion very light 23 %, light 20 %, moderate 47 % and heavy 10 % of total work time. In average they were working out- doors 4-6 hours and the length of an average work period outdoors was 1-2 hours. In average the test persons had 2-4 breaks during the day and the length of an average break was 30-45 minutes. The duration of warm work was less than 15 minutes per day. The workers were not disturbed by cold or were disturbed just occasionally. The workers re- lated the cold sensation in feet to the footwear material (1 subject) and sweating (1) and considered these days similar or warmer compared to usual cold season workdays. The footwear was considered comfortable except by one worker - he had chosen a too big size. Remaining problems with the footwear were: difficulty to walk, limited mobility and poor fit (1 subject), limiting effect on work ability (2 subjects), cold feet (the subject with shoes) and sweaty feet (1 subject). A person, who often had to climb up and down platforms, considered the boots to be too heavy. Generally the boots were considered good and some workers were eager to get similar shoes. Two workers mentioned the boots to be “very good safety boots”. No subject had problem with slipping. 73 3 2 2 3 2 2 2 2 1 1 1 1 1 11 2 1 2 3 3 2 2 2 2 2 2 1 2 2 2 10 15 20 25 30 35 06 :3 0 07 :0 0 07 :3 0 08 :0 0 08 :3 0 09 :0 0 09 :3 0 10 :0 0 10 :3 0 11 :0 0 11 :3 0 12 :0 0 12 :3 0 13 :0 0 13 :3 0 14 :0 0 14 :3 0 15 :0 0 15 :3 0Time (hh:mm) Te m pe ra tu re (° C) -4 -3 -2 -1 0 1 2 3 4 Th er m al s en sa tio n Thermal sensation 1 Thermal sensation 2 Toe 1 Toe 2 Foot 1 Foot 2 Figure 1. Foot and toe temperatures of subjects 1 and 2 (asphalt cutting). Thick lines at time axis show work periods in the cold. Columns and numbers show the thermal sensation in feet (1 - subject 1; 1 - subject 2) at given time point. Foot skin temperatures The foot skin temperature over a day is shown in Figure 1. The foot skin temperatures stayed relatively high, usually over 30 °C. In subject 2 the foot skin temperature was somewhat lower already in the beginning. However, during the workday it raised higher than 30 °C. Enander et al. (Enander et al., 1979) showed similar values for food process- ing workers in cold halls doing various jobs at an air temperature of 1-13 °C. The warm boots kept the foot skin temperature relatively constant for the whole workday. The thermal sensations of the feet (scale from +4 to -4) were often warm or very warm. The exception was subject 7 who had his own shoes and during most of the day felt slightly cold in the feet. His foot skin temperatures stayed below 30 °C for most of the workday. However, the toe temperatures varied more than the foot temperatures. This could be related to the lower insulation of the toe zone (Kuklane & Holmér, 1998) and to the fact that the vasomotor activity is high in peripheral body parts. The lowest toe temperatures were occasionally between 17-18 °C. Commonly they stayed higher than 25 °C for most of the workday. In the beginning of the day most of the subjects had toe skin tempera- tures between 20-25 °C. It could depend on the preparation of the study, i.e. taping the sensors to the skin, while subjects were minimally dressed. Subjects 5 and 6 had the most stable toe temperatures if to exclude the start of the day. These subjects took off the foot- wear during the breaks and wore slippers instead. It is difficult, however, to say if the high and relatively stable toe temperature depended on that. The toe temperatures were in average 6 °C lower than foot skin temperatures. Toe temperature changes did not considerably affect the thermal sensation of feet as has been shown before (Kuklane et al., 1998; Tochihara et al., 1995). The reason could be that the skin temperatures stayed relatively high. 74 Moisture accumulation in footwear The moisture accumulation in the footwear was at average 20 g per day and foot (range 10 - 30). If the moisture accumulation rate is assumed equal to sweat rate then it corre- sponds to about 1.3-3.8 g/h. Around 2 g of that stayed in the socks at the end of the day. The amount of moisture, that was accumulated in the footwear of subjects 5 and 6 (took off the footwear during breaks) did not differ from the others. Sweating seems to affect footwear insulation depending on sweat rate, sweat amount, and probably also absorption capacity and evaporation resistance of the footwear. During sweating tests on a thermal foot model (Kuklane & Holmér, 1998) a considerable reduction of footwear insulation was shown. During those tests a sweat rate of 10 g/h was simulated. At sweat rates of this study the insulation reduces less, resulting in drier socks and thus more comfortable feet. Conclusions The provided boots had too high insulation (0.34 m2°C/W) for the particular weather conditions (+4 to +8 °C). At the same time the insulation of another type of shoes (0.23 m 2 °C/W) seemed to be insufficient (one subject). Under the studied climatic and work conditions the recommended insulation would be between 0.25 and 0.30 m2°C/W. The foot temperatures of the subjects who took off footwear during warm breaks stayed at a constant and high level. This supports the recommendation to doff the foot- wear during breaks for rewarming and ventilation. The wetness of footwear, mainly due to sweating, was a source of some complaints. Sweating intensity can be related to high insulation of footwear, bursts of high activity and warm weather, but also to individual differences. Acknowledgements Thanks to the Port of Göteborg AB and the workers for co-operation. Special thanks to Taiga AB for help to establish contacts and Arbesko AB for providing footwear. References Enander A, Ljungberg A-S & Holmér I (1979) Effects of work in cold stores on man. Scandinavian Journal of Work, Environment & Health, 5, 195-204. Gavhed D, Kuklane K & Holmér I (1999a) Work on masts in cold (Mastarbete i kyla), Arbetslivsrapport 1999:22, Solna: National Institute for Working Life. Gavhed D, Kuklane K, Karlsson E & Holmér I (1999b) A pilot study on work in the cold - questionnaire survey and field study (En pilotstudie om arbete i kyla - frågeundersökning och fältstudie), Arbetslivsrapport 1999:4, Solna: National Institute for Working Life. Kuklane K, Geng Q & Holmér I (1998) Effect of footwear insulation on thermal responses in the cold. International Journal of Occupational Safety and Ergonomics, 4, (2) 137-152. Kuklane K & Holmér I (1998) Effect of sweating on insulation of footwear. International Journal of Occupational Safety and Ergonomics, 4, (2) 123-136. Rintamäki H & Hassi J (1989) Foot temperature and thermal sensations in the foot in the naked and clothed man. In Mercer J B (Ed), Thermal physiology 1989 (pp 173-176). Amsterdam: Elsevier. Tochihara Y, Ohnaka T, Tuzuki K & Nagai Y (1995) Effects of repeated exposures to severely cold environments on thermal responses of humans. Ergonomics, 38, (5) 987-995. 75 Footwear for cold work: a field study about work on high masts Kalev Kuklane, Désirée Gavhed, Ingvar Holmér National Institute for Working Life, Respiratory health and climate, Solna, Sweden Introduction A series of field studies at cold workplaces were initiated (Gavhed, Kuklane & Holmér, 1999a; Gavhed, Kuklane, Karlsson & Holmér, 1999b). The studies were expected to give information on:  how the results from various laboratory experiments could be used in practice,  how to design future tests that will be directly related to actual working conditions,  how to validate and improve existing prediction models and develop new requirements. This paper deals with the study at a telecommunication company and the main atten- tion is paid to the thermal responses of the feet and footwear performance. Other data are shown as a reference and background data. Materials and methods The study was carried out in February (Gavhed et al., 1999a). Eight male workers were the subjects with a mean age of 40 (22-50) years, mean weight of 87 kg (72-105) and mean length of 177 cm (173-184). They were selected among volunteers who carried out various jobs on high masts on four different days. The work was done in pairs on four different masts in Sweden. A short period of time (10-20 minutes) was used to measure oxygen consumption (VO2) during climbing the masts. The maximum VO2 during climbing was 2.45 l/min (1.18-3.33 l/min). The average VO2 during the work periods over whole workday was estimated to lay around 0.74 l/min (0.64-0.86 l/min). The length of the cold exposure depended on the task. The shortest total work time in cold on the mast was on the third day (Day 3, changing a warning light): 2 hours 40 minutes. At the same time it was one of the longest cold exposures. The longest total work time in cold was on the second day (Day 2, fastening new cables to the mast, half time operating in a mast elevator): 4 hours and 55 minutes (length of a whole work day 6.5 hours). It was divided into two work periods. The longest work day was 7 hours and 30 minutes (Day 4 mounting up two new parabolas). 4 hours and 30 minutes of that was work outdoors near or on the mast. This day was divided into 4 work periods in cold. The first day (Day 1, fastening new cables and taking down old) was 5 hours 40 minutes long and 4 hours of that was spent on or near the mast. The work outdoors was divided into two periods. The time to drive to the mast, preparation for the measurements and ending the measurements were not included above. In addition, communication and waiting for the details took some time, too. The mean outside air temperature (measured near body) and wind velocity are shown in Table 1. The workers used their ordinary clothing. Standard (company recommended) clothing were Ullfrotté underwear and outer layer by Taiga AB. All the clothes were weighted 76 separately in the beginning and at the end of the workday. During the breaks the workers usually changed footwear and clothing. Subjects 1-4, 7 and 8 had boots of model 520 (Woodman) that were manufactured by Stålex, Arbesko Gruppen AB. Subject 5 used Graninge boots and subject 6 model 529 (Varm-varm, Stålex, Arbesko Gruppen AB). The weight of the footwear is shown in Table 1. The insulation of model 520 (size 41) was measured on a thermal foot model (Kuklane & Holmér, 1998). The insulation when standing dry with a load of 35 kg and a sock (34 g) was 0.34 m2°C/W. The insulation of the other boots was estimated to be at the same insulation level, i.e. between 0.33 and 0.35 m2°C/W. The workers used two pairs of socks (thin and thick), except subject 6 who used 1 pair of thin cotton socks. Usually various woollen socks were worn, but also cotton, fibre-pile and synthetic socks, usually together with a woollen sock. By the weight socks could be divided into three groups: thin (54-82 g), thick (100-107 g) and extra thick (133-165 g, Table 1). The socks could add about 0.02-0.05 m2°C/W to the insulation. Temperature sensors (StowAway temperature logger, Onset Computer Corporation) were taped to the second toe and dorsal foot. The workers were observed during the whole workday. Their subjective responses on thermal sensation were recorded over the day and at the end of the day they filled in a questionnaire about the workday. Four ques- tions were directly related to feet and/or footwear. Results and discussion The data are shown in Table 1. Some data were lost, probably due to the strong electro- magnetic field near the antennas. Table 1. The footwear and sock weight, foot and toe skin temperature (mean (min-max)), air temperature near body (mean (min)), air velocity (mean (max)), moisture accumulation and thermal sensation in feet (mean over work period (lowest during work period)) data of the sub- jects. Data was ranked by foot skin temperature. Footwear (subject) Footwear (sock) weight (g/pair) Foot temp. (°C) Toe temp. (°C) Air temp. (°C) Air velocity (m/s) Total moisture . accumulation (g) Thermal sensation Varm-varm (6) 1828 (63) 26.7 (25.9-27.1) Missing 1.9 (-2.4) 9.7 (13.5) 6 0.1 (0) WS (3) 1988 (69+104) 27.7 (20.8-31.2) 23.2 (11.0-30.5) 1.5 (-11.5) 3.6 (6.1) 35 0.8 (-2) Graninge (5) 1808 (69+143) 29.3 (27.9-31.2) 17.2 (14.9-19.5) 2.1 (-2.1) 9.7 (13.5) 12 -0.1 (-1) WS (4) 1982 (54+106) 30.4 (25.9-34.3) 23.4 (9.8-33.7) 3.9 (-0.6) 3.6 (6.1) 56 0.0 (-2) WS (2) 1681 (80+100) 30.8 (27.1-34.3) 26.9 (12.7-34.7) 2.9 (-0.6) 3.4 (9.1) 54 0.1 (-2) WS (1) 1754 (65+107) 31.4 (28.2-33.8) 17.1 (13.9-20.9) -4.1 (-10.9) 3.4 (9.1) 63 0.7 (0) WS (8) 1910 (67+133) 31.8 (26.7-34.8) 29.6 (18.0-35.2) 3.8 (-5.6) 1.7 (4.6) 94 0.8 (-1) WS (7) 1783 (82+165) 32.0 (29.3-34.2) Missing 8.8 (-1.2) 1.7 (4.6) 39 0.7 (-1) Questionnaire at the end of the day The observed work times fit relatively well with work times from the questionnaire. The test persons estimated their work exertion and the time worked at certain exertion level the in following way: very light in average 1.6 (1-2.5) hours (4 persons), light 2.1 (1-4) hours (7 persons), moderate 1.4 (1.5-6) hours (4 persons), heavy 0.5 (0.5-1) hours (5 per- sons) and very heavy 0.5 hours (4 persons). Five workers worked outdoors 4-6 hours, one 2-4 hours and two 0.5-2 hours (observed 2.3-5 hours). The length of an average work pe- riod outdoors was more than 2 hours for 4 persons, 1-2 hours for 2 and less than 1 hour for one person (observed average 1.8 hours). One did not answer the question. Three per- sons had 3-5 breaks per day, three had 1-2 and 2 did not have any breaks. The length of a break was estimated to be 0.5-1 hour (3 persons), 15- 30 minutes (1) and less than 15 minutes (2). Two worked once or twice in warm locations, but duration of warm work 77 was less than 15 minutes. The workers were not disturbed by cold (2 persons) or were disturbed just occasionally (6). The workers related the cold sensation in feet to the wind (2 subjects), contact with cold surfaces (2), footwear material, the way of using footwear and standing still (each 1). The subjects felt to be similar (6) or warmer (2) compared to usual cold season workdays. The footwear was considered comfortable except by 1 worker and 1 had no opinion. Problems with the footwear were reported to be stiff in the cold (1 person), cold feet (1), sweaty feet (2) and wet feet (1). One person slipped during the day, but did not fall. 2 1 1 1 0 -2 1 1 0 0 0 4 1 0 -1 -1 -2 1 00 0 1 10 15 20 25 30 35 08 :3 0 09 :0 0 09 :3 0 10 :0 0 10 :3 0 11 :0 0 11 :3 0 12 :0 0 12 :3 0 13 :0 0 13 :3 0 14 :0 0 14 :3 0 15 :0 0 15 :3 0 16 :0 0 16 :3 0Time (hh:mm) Te m pe ra tu re (° C) -4 -3 -2 -1 0 1 2 3 4 Th er m al s en sa tio n Thermal sensation 3 Thermal sensation 4 Toe 3 Toe 4 Foot 3 Foot 4 Figure 1. Foot and toe temperatures of subjects 3 and 4. Thick lines at time axis show work pe- riods. Columns and numbers show the thermal sensation in feet (1 - subject 3; 1 - subject 4) at given time point. Foot skin temperatures The foot skin temperature over a day is shown in Figure 1. During colder weather the foot skin temperatures were lower than during warmer weather. There was also a ten- dency that before lunch the foot, and especially the toe temperatures were lower. This could be related to lower air temperatures before noon. The subjective responses fol- lowed the skin temperatures. The lowest average foot skin temperature was 26.7 °C and this was measured in subject 5 (one pair of thin socks). However, his subjective ratings were usually at comfort (0). This subject had also the smallest amount of sweat accumu- lated in the footwear and sock. Probably due to the influence from the electromagnetic field his foot data recording was stopped relatively early and toe temperature data were lost. The lowest measured foot skin temperature was 20.7 °C. This subject (3) had also the next lowest average foot skin temperature (27.73.1 °C). His job in the morning con- sisted of standing on the elevator roof and fastening of new cables to the mast. The highest mean foot skin temperatures were in subject 7 (32.01.4 °C, toe skin temperature missing). 78 The footwear (0.34 m2°C/W, use of extra socks not considered) used during work on high masts at lower mean environmental temperatures (-2 to -8 °C) gave in average 2.2 °C lower foot and 3.4 °C lower toe temperatures than during work in harbour (same footwear, ambient temperature +4 to +8, Gavhed et al., 1999b). The subjective responses were in average colder by about 1-2 points (scale 4 to -4) during work on masts, however being still somewhat warmer than neutral. The mean difference between the toe and foot skin temperatures was 7 °C. All the subjects had toe temperatures under 18 °C at least once (data for 2 subjects are missing). Subject 1 had the lowest mean toe skin temperature (17.12.2 °C). Subject 4 had the lowest measured toe temperature (9.8 °C). This subject was also the one who in the questionnaire pointed out cold feet. The highest mean toe skin temperature was measured on subject 8 (29.65.7 °C). He had also the second highest foot skin temperature, and the highest sweat gain in footwear and socks (94 g). The high foot and toe temperatures of subjects 7 and 8 can be related to the warmer weather and lower air velocity than during other days. These subjects had also relatively high activity throughout the whole day. Moisture accumulation in footwear The moisture accumulation in the footwear was on average 37 g (6-94 g) per day. Around 4 g of that stayed in the socks at the end of the day. However, the length of the workday varied. If the moisture accumulation rate is assumed equal to sweat rate then it corre- sponds to 7.53.7 g/h (2.2-12.9 g/h). Conclusions The footwear in combination with the chosen socks worked well in the particular weather conditions (-2 to -8 °C). However, more attention should be paid to the cold protection of toes. During low activity the toe temperatures dropped relatively quick at air tempera- tures of about -10 °C and/or high wind speeds (about 10 m/s). The workers reported relatively few complaints on their footwear. However, in some cases the foot and especially toe skin temperatures dropped to low levels that are con- nected with cold and discomfort sensation. Acknowledgements Thanks to TERACOM AB and the workers for co-operation. Special thanks to Taiga AB for help to establish contacts. References Gavhed D, Kuklane K & Holmér I (1999a) Work on masts in cold (Mastarbete i kyla). Arbetslivsrapport 1999:22, Solna: National Institute for Working Life. Gavhed D, Kuklane K, Karlsson E & Holmér I (1999b) A pilot study on work in the cold - questionnaire survey and field study (En pilotstudie om arbete i kyla - frågeundersökning och fältstudie). Arbetslivsrapport 1999:4, Solna: National Institute for Working Life. Kuklane K & Holmér I (1998) Effect of sweating on insulation of footwear. International Journal of Occupational Safety and Ergonomics, 4, (2) 123-136. 79 Innovations in fibres and textiles for protective clothing Roshan Shishoo IFP Research AB, Sweden Millions of people world-wide have working environment which exposes them to spe- cific risks from which their bodies need protection. In many industrial sectors, military and energy services, hospital environments, human beings are subjected to various types of risks and each sector has its own requirements for protective clothing. The end-use applications for protective clothing include  Chemical splash and vapour protection  Clean-room apparel  Cut resistant gloves  Dirt and dust  Fire fighting  Ballistic protection  Paint spray  Puncture resistant clothing  Pharma cantical manufacturing  Dry chemical handling The performance requirement of all types of protective clothing often demand the balance of widely different properties of drape, thermal resistance, liquid barrier, water vapour permeability, anti-static, stretch etc. The seemingly contradictory requirement of creating a barrier e.g., towards heat, cold, chemicals, bacteria, and breathability in high functional clothing has placed challenging demands on new technologies for producing fibres, fabrics and clothing design. Among the contributing factors responsible for suc- cessful marketing of such products have been advances in polymer technology and pro- duction techniques for obtaining sophisticated structures of fibres, yarns and fabrics (Shishoo, 1988). Innovations in fibres The evolution of fibre developments have gone through the phases of conventional fibres, high-functional fibres and high-performance fibres (Mukhopadhyay, 1993). Improved fibre spinning techniques in melt spinning, wet spinning, dry spinning and new techniques such as gel spinning, bicomponent spinning, microfiber spinning, have made it possible to produce fibres with characteristics more suitable for use in protective clothing. Today a wide range of high performance fibers is commercially available for technical and industrial applications. Among the specialty fibers already established can be men- tioned the following types.  Aramid fibres - p-aramid fibre to provide high strength and ballistics. - m-aramid fibre to provide flame and heat resistance. 80  Ultra-high tenacity polyethylene fibers (UHMWPE) Gel spun, ultra-high molecular weight polyethylene fibers with extremely high spe- cific strength and modulus, high chemical resistance and high abrasion resistance.  Polyphenylene sulphide fibers (PPS) Crystalline thermoplastic fiber with mechanical properties similar to regular polyester fiber. Excellent heat and chemical resistance.  Polyetheretherketone fibers (PEEK) Crystalline thermoplastic fiber with high resistance to heat and to a wide range of chemical.  Novoloid (Cured phenol-aldehyde) fibers High flame resistance, non melting with high resistance to acid, solvents, steam, chemical and fuels. Good moisture regain and soft hand.  Polybenzimidazole fibers (PBI) Moisture regain 15 %, high resistance to chemicals especially at elevated temperatures. Aramid fibres A number of different types of p-aramid fibres are commercially available as the basis material for protective clothing. The brands include Kevlar by DuPont, Twaron by Acordis and Technora by Teijin. Para-aramid fibre’s combination of high strength, non- flameability and high temperature resistance makes this type of fibre suitable in applica- tions such as  ballistic protection in armour vests and helmets,  cut through protection in safety gloves, aprons, work wear and shoes for high risk jobs,  high temperature protection e.g., spatter-resistant clothing. These fibres are also well-known for their dynamic energy absorbing properties (Van Zijl et al., 1999). For thermal stability purposes m-aramid fibres has the right attributes. Meta-aramids have high temperature resistance, moderate tenacity and low modulus but excellent re- sistance to heat (Gorashi & Stocks, 1995). These fibres are very useful when outstanding thermal protection and electrical insula- tion properties are required. The brands include Nomex and Corex. The m-aramid fibre Kermel is currently used in protective clothing. It has excellent thermal properties against high temperature. “flash” exposures and is easily blendable with other fibres (Cassat & Hoessi, 1995). DuPont-Toray in Japan have also produced new spun-dyed Kevlar p-aramid fibres. Twaron made by Acordis is also being produced as a microfibre with a dtex of 0.93. High-performance polyethylene fibres UHMPE with high modulus and high strength together with exceptional strength-to- weight ratios are available with trade names Dyneema from DSM, Spectra from Allied Signals, Tekmilon from Mitsui. This type of fibres has  very high specific strength, specific modulus and high energy to break  low specific gravity 81  very good abrasion resistance  excellent chemical and electrical resistance  good UV resistance  low moisture absorption In protective clothing applications, many impact helmets used by runners and moun- tain climbers, for example, are made of UHMPE composites. High performance PE fibres are also used for protection against cutting, sawing, puncturing, and ballistics. A very high growth in the use of fibres such as Dyneema and Spectra is predicted for use as body armour. Knitted fabrics made of UHMPE are used for personal protection viz., gloves, chain saw protection and fencing suits (Jacobs & Mencke, 1995; Van Gorp & Van der Loo, 1995). Novoloid fibres Kynol woven and nonwoven fabrics display the following advantages: In flame:  high flame resistance; non-melting at any temperature  retention of textile integrity; no embrittlement or breakage  minimal evolution of smoke, nearly no shrinkage  virtually no toxic off-gassing (no HCN, halogens, etc.) And in general:  outstanding resistance to acids, solvents, bleaches, fuels and other chemicals, steam  excellent thermal and electrical insulation  good moisture regain and soft hand; comfortable to wear and use  light weight (specific gravity 1.27)  available as 100 % novoloid and in various blends. Melamine-based fibres Basofil (BASF), a high temperature and fire resistance fibre, is a melamine-based staple fibre which has a LOI of 31-33 with no melt dripping and a continuous service tempera- ture of about 200 ºC. (Berber, 1995; Ott, 1995). PEN & PBO fibres Recently some new exciting fiber types such as PEN and PBO have been introduced in the market. Compared to the standard polyester, PEN, polyethylene-2, 6-naphtalat, fiber yarns have a significantly higher modulus, high dimensional stability, higher glass- transition temperature and better resistance to hydrolysis and LOI of 31. Toyobos’ high performance PBO fiber, Zylon, p-phenylene-2, 6-benzobisoxazole, has strength and modulus far exceeding than any of the known fibers. PBO fiber has the de- composition temperature of 650 ºC, tenacity 5-8 Gpa and modulus of 180-250 Gpa (Shishoo, 2000; Toyobo, 1998; KoSa Co, 1999). Other types The market for FR-treated celllulosic fibres for use in apparel. Flame-retardant arments made from cellulosic fibres continue to find application even in the most demanding of 82 working environments such as those encountered by fire-fighters. Polyester Trevira is another example of a flame retardant fibre. Flame-resistant heavy-duty PES multifilament yarns are also available from e.g. KoSa. Electro-conductive fibres Many synthetic fibres are excellent electrical insulators and therefore very susceptible to static charges. One approach for solving this problem is to increase fibre conductivity by applying conductive finishes to the surface. Most of these finishes are easily removed by washing however, hence do not provide durable protection against the hazards of static electricity. Metal fibres have been successively replaced by electro-conductive organic fibres in the past decade. One commercial approach to make a permanently antistatic fibre is to blend a conducting polymer into the non-conducting fibre. Many polyblend fibres of this type now are commercially available (Shishoo, 1988). Novel yarn spinning technology Novel yarn spinning technologies are commercially available today for producing hybrid yarns for various applications including protective clothing. Two technologies are used to manufacture such yarns  the conventional spinning by intimately blending two different yarns  the core yarn spinning making a core of a certain fibre type covered with a sheath of a different fibre type. One can thus combine the functions of two different fibre qualities to produce fabric of varied functions. For example the core can be made of p-aramid, the sheath of m-aramid, cotton or polyester. Sensitive core-materials can be protected by he sheath fibres (Bontemps, 1995). High functional and high performance fabrics Microfibre based light woven constructions The very first example within this group is the ventile fabrics made of 100 % cotton, where a very large reduction in the interfibre pore sizes is achieved as a result of swelling in fibres in wet state. New techniques for spinning extremely fine polyester filaments of 0.1-0.3 denier have resulted in many interesting developments of woven and nonwoven structures of high functional characteristics. For example the superfine yarns are woven into various constructions such as Oxford and taffeta, in which individual filaments are packed so tightly that the resultant fabric develops high hydrostatic resistance in combination with good water-vapour permeability and drape. These properties can be further improved by treatment with liquid repellent finishes. Some of the most interesting developments lie in the production of multi-layer knitted and woven constructions. A multi-knit fabric of 2 or 3layered structure using 100 % polyester and polypropylene yarn has characteristics of quick water absorption, ability to evaporate water and a dry touch, being capable of transporting perspiration from the skin to the outer surface and then quickly dispersing it (Shishoo, 1988). 83 Breathable fabrics Laminated fabrics (microfibrous membrane) One of the most significant developments in breathable waterproof was the introduction of the GORE-TEX rainwear fabric in 1976. GORE-TEX is a microporous polymeric film made of polytetrafluoroethylene (PTFE). The GORE-TEX film is supposed to contain micropores of size 0.2 micron at the rate of more than 1.3 billion/cm², and it provides a barrier to water, airborne particles and bacteria. Being very hydrophobic, PTFE will resist wetting of the surface. At the same time because of the pore distribution water vapour can readily diffuse through the film. GORE-TEX film has been bonded to a variety of substrates and constructions. Both 2- layer construction, where a single layer of fabric is bonded to one side of the film, and 3- layer, where fabrics are bonded to both sides of the film, are available as high functional fabrics. A similar principle is used in MICRO-TEX film from Japan. A fabric for outwear is laminated with porous film of PTFE resin containing micropores of size 0.6 micron at the rate of approx 1 billion/cm² (Shishoo, 1988). Coated fabrics (microporous caoting) Polyurethane is one of the most widely used polymers for coating of apparel fabrics. One reason for this is the availability of numerous combinations of polyols, isocynates and amines for its synthesis. Poromeric polyurethane coatings are produced by either a wet coagulation process or a direct coating system. The general procedure in a wet coagula- tion system is to impregnate the fabric with a special polyurethane formulation dissolved in an appropriate solvent. Before evaporation of the solvent the coated fabric is immersed in a water bath which precipitates a coherent but highly porous polyurethane layer. In a direct coating system, the microporous structure develops during subsequent drying and curing process. Numerous developments in microporous polymer structures for use in direct coating, and as a film for laminating into 2 or 3 layer structures, have taken place world-wide. These microporous structures function by allowing the passage of water vapour mole- cules (approx 0.0004 micron in dia) whereas large diameter (>100micron) water drops get blocked by these structures. For a given porosity and thickness of the coating the wa- ter vapour permeability increases if one reduces the pore size (Shishoo, 1988). Laminated fabrics (hydrophilic membrane) High functional fabrics are engineered by combining with hydrophilic membranes. Water-vapour transmission through these membranes is achieved by the physical pro- cesses of adsorption, diffusion and desorption. Because the membrane is a non-porous material one can expect no clogging of pores. AKZO Nobel has introduced SYMPATEX, which is a nonporous high breathable polyester membrane. This product is washable and dry cleanable and is claimed as wa- tertight, windtight and having high wear resistance. Its development has included work in polymeric composition, membrane process, laminating process and the seam-sealing technique. AKZO Nobel has developed a special seam-sealing tape for this purpose (Spijkers, 1995). 84 Thermal insulation materials Waterproof and breathable fabrics are often used in combination with synthetic heat in- sulating materials. Because of fashion and function requirements and demands, develop- ments have taken place for producing thin but warm fibrefills. Since air has a very low conductivity in the newly developed materials, heat insulating property is imparted and improved by confining as much air as possible in the microspaces between and/or within the fibres and so checking heat losses by convection. These materials are made by pro- ducing Coweb-like structures from superfine fibres, or by making the fibres hollow in order to contain air and prevent air movements. These new materials have 2-3 times as much thermal insulation as that of conventional polyester wadding or down of the same thickness (Shishoo, 1988). Nonwoven protective clothing Protective coveralls, suits, gowns, lab coats and accessories are used in industry and institutions to protect workers from exposure to hazardous materials and to protect sensi- tive products from human contamination. Nonwoven are used for limited use protective clothing and as components in reusable clothing. Spunbonded olefins are the leading materials used and are often used in composites with barrier films. Medical/surgical disposal represent about 11% of the European nonwoven market whereas these disposal are the second largest market for nonwoven in the US. This product group includes surgical packs, drapes and gowns, surgical face masks, caps and head covering, patient and staff clothing. Spunlaced pulp/polyester, spunbonded, spun- bonded/meltblown components and wet laid pulp/polyester nonwovens are the leading fabric types used for medical disposables. Heatbonded and calendered polyethylene spun-bonded fabric Tyvek from DuPont is a fine example of a very successful nonwoven material for protective clothing. Tyvek 1431N, nonwoven structure made of 100% HDPE offers excellent particle hold and properties down to particle size of 3 microns. For protection from particles smaller than 3 microns, DuPont recommends the coated fabric Tyvek C and laminated fabric Tyvek F. For dust protection Corvin Company’s spunbonded nonwoven “Multidenier” using dif- ferent spunbonded technologies demonstrate a new group of innovative products (Bernstein & Thiele, 1995). Photoluminescent material Inorganic luminescent pigments which are composed of zinc-sulphide crystals. The crystal lattice of the pigment is able to absorb visible high energy and to store that energy through various levels of the electrons. The energy will be set free under the emission of light when the activating light source disappears. This process can be repeated as often as wanted. The cause of photoluminescence is a transport of electrons. These materials are free of radioactive substances and are non-toxic. The size, formation and the quantity of the zinc-intensity of the light emission. The base material carrying the photoluminescent is a soft-coating blend of polyester and cotton (Simpson, 1995). 85 Interactive textile materials There have been some interesting developments taking place in recent yarns regarding smart textile materials. These materials readily interact with human/environmental con- ditions to produce change in material properties. Shape Memory Polymers The shape memory effect is observed in metal alloys and polymers and results in an ob- ject reverting to a previously held shape when heated. Early shape memory polymers were blends of glassy thermoplastics and elastomers. More recently, Mitsubishi has made shape memory polyurethanes available. These are thermoplastic polymers that can have reversion temperatures in the range –30 ºC to +100 ºC. Shape memory polyurethanes are block copolymers having hard and soft segments. The hard segments, which contain the urethane linkage and chain extenders, are present in sufficient numbers to form a continuous crystalline phase. Constituted from polyether or polyester diols, the soft segments make up the glassy phase (Russel et al., 1999; Technical Textiles International, 1999b). In effect, the shape memory polyurethanes are thermoplastic elastomers that can have glass transition temperatures within an unusually interesting range. For instance, this in- cludes ambient temperatures (about 25 ºC), the temperature of the human body (37 ºC) and the temperature of boiling water (100 ºC). The prototype design is a laminated film consisting of:  a film of shape memory polymer having a glass transition temperature of 25 ºC;  a layer of a compatible elastomeric, a thermoplastic polyurethane having a much lower glass transition point (such as a conventional shoesoling grade). The films can be made on an extrusion/calendering line and the laminates can be com- pression moulded in conventional equipment. To promote the circulation of air and moisture vapour within the interstitial space, as well as reducing the weight of the film, holes and cut-outs can be made in the laminate. On cooling below 25 ºC, the shape memory layer should shrink linearly by some 3 % and become rigid while the conventional elastomer remains largely unaltered. As a result, an out-of-plane deformation of the laminate is expected to occur. It is anticipated that deformed films of these laminates will provide a reversible response to cold conditions. Phase Change Materials (PCM) Protective textiles with microencapsulated Phase Change Material are now commercially available, e.g. Outlast Technologies. The PCMs used in this technology consists of car- bohydrates with different chain lengths, whose phase change take place in a temperature range close to that of the human skin (Pause, 1995; Pause, 1999; Technical Textiles International, 1999a; Leitch & Tassinari, 2000). The PCM, are encapsulated in small spheres in order that they are contained when in a liquid state. The microcapsules possess approximate diameter of 1-10 m and are resis- tant to abrasion, pressure, heat and chemicals. At present this-technology is only appli- cable to acrylic fibres using the wet-spinning process. For coating applications, the PCM microcapsules are dispersed in e.g. a polyurethane coating, which is then applied to a fabric. 86 For foam applications, the PCM microcapsules are dispersed in a polyurethane foam matrix. These foams are often laminated to a fabric. Textile structures with PCM microcapsules for protective clothing have following in- teractive functions:  absorption of surplus body heat;  an insulation effect – caused by heat emission of the PCM into the textile structure;  a thermo-regulating effect – which keeps the microclimate temperature nearly constant. Cooling fibres Other relatively new technology of interest is cooling fibres by the introduction of water- retaining fibres into the fabric structure. These fibres are sandwiched between a breath- able out fabric e.g. cotton and Nomex and an inner layer conducting heat and moisture from the body (Leitch & Tassinari, 2000; Stull, 2000). Concluding remarks The driving technology force in material development for protective clothing has been spear-headed by advances in fibres, polymers, chemical technology and fabric/web forming technologies. The technological trends and challenges ahead will be determined by market pull demands, increasing environmental awareness, personal safety and com- fort, and performance requirements. The advances in material and technologies should lead to products with sought-after characteristics e.g., laminates able to meet multiple functional requirement, coating which can be tailored for specific end-uses, fibres and fibre-blends for demanding applications and sophisticated fibrous structures. References Berber H (1995) Textile Processing of Fibres based on Melamine Resin. Techtextil Symposium 1995, lecture 2.14, Messe Frankfurt. Bernstein U & Thiele R (1995) New Dustproof Textiles from Composite Nonwovens. Techtextil Symposium 1995, lecture 2.56, Messe Frankfurt. Bontemps G (1995) Schappe Core Yarns for Protection against Heat. Techtextil Symposium 1995, lecture 2.41, Messe Frankfurt. Cassat R & Hoessi J (1995) Kermel Tech – A New Fibre for Industrial Applications. Techtextil Symposium 1995, lecture 2.56, Messe Frankfurt. Gorashi HM & Stocks AI (1995) Nomex Omega – a new fibre for active fire protection. Techtextil Symposium 1995, lecture 2.16, Messe Frankfurt. Jacobs MJN & Mencke JJ (1995) New Technologies in Gel-Spinning the World’s Strongest Fibres. Techtextil Symposium 1995, lecture 2.13, Messe Frankfurt. KoSa GmbH & Co (1999) PEN-fibres. Technical Information, April 1999 Leitch P & Tassinari TH (2000) Interactive Textiles; New Materials in the New Millennium. Part 1. J. Industrial Textiles Vol. 29, No 3, January 2000, 173-190. Mukhopadhyay SK (1993) High Performance Fibres Textile Progress Vol. 25, No 3/4, The Textile Institute. Ott K (1995) Development of Heavy Fire and Heat Protection Suits from Basofil for the BASF Fire Brigade. Techtextil Symposium 1995, lecture 2.43, Messe Frankfurt. 87 Pause B (1995) Development of Heat and Cold Insulating Membrane Structures with Phase Change Material. J. of Coated Fabric, 25 (7), 59-68. Pause B (1999) Development of Self-conditioning Sport Shoe with PCM. Techtextil Symposium 1999, lecture 305, 22-26, Messe Frankfurt. Russel A, Hayashi S & Yamada T (1999) The Potential Use of Shape Memory Film in Clothing. Techtextil Symposium 1999, lecture 302, Messe Frankfurt. Shishoo R (1988) Technology for Comfort. Textile Asia, June 1988. Shishoo R (2000) Global Advances in Technical Fibres and Fibrous Products for Speciality Applications. Techtextil North American Symposium 2000, March 22-24, 2000, Atlanta. Simpson U (1995) New Possibilities through the Making of Protective Clothing for Drakness with Luminescent Fabric Film. Techtextil Symposium 1995, lecture 2.44, Messe Frankfurt. Spijkers JCW (1995) New Application of Sympatex Polyester Membranes in Personal Protection. Techtextil Symposium 1995, lecture 2.52, Messe Frankfurt. Stull JO (2000) Cooler Fabrics for Protective Apparel. Industrial Fabric Products Review, Vol. 76, March 2000, No 11, 62-68. Technical Textiles International (1999a) Phase change materials show potential for medical applications. September, 23-26. Technical Textiles International (1999b) Potential uses of shape memory film in clothing. October, 17-19. Toyobo Co (1998) Super High Performance Fiber ZYLON. Technical Information. Van Gorp EHM & Van der Loo LLH (1995) Dyneema Nonwovens for Ballistic Protection. Techtextil Symposium 1995, lecture 2.53, Messe Frankfurt. Van Zijl N, Ren I & Chiou M (1999) High Comfort Systems with KEVLAR for Protection against Sharp and Pointed Weapons. Techtextil Symposium 1999, New Protective Textiles, lecture 301, Messe Frankfurt. 88 High visibility warning clothing Doina Toma, Eftalea Carpus, Iuliana Cohea The National R-D Institute for Textile and Leather, Bucharest Introduction High visibility warning clothing is necessary to signal out the presence of the worker to detect and quickly percept of the worker that work in dangerous conditions, in any lumi- nosity conditions, on day time (including on bad weather conditions: mist, rain etc.), on night time (including headlights) or working places where adequate lighting condition can not be ensured (for example mines, work in pipes, tunnels). In order to produce high visibility warning clothing there are used textile fabrics which should correspond the main performing parameters, concerning:  chromacity characteristics  colour fastness to rubbing, perspiration, washing, bleaching and washing agent  mechanical characteristics: tensile strength, tearing strength In this respect the poster will present the accomplished results of the researches concerning:  designing and accomplishing the woven structures fluorescent Yellow and fluorescent Red-Orange  accomplishing of 2nd and 3rd class of high visibility warning clothing High visibility materials The higher visibility materials can be classified as retro-reflective, photo-luminescent and fluorescent. They should provide:  signs of the worker presence  quick detection and perception of the workers who are working in:  dangerous conditions  any conditions of luminosity  day time (including the presence of rain and haze)  night time (including the light of the headlights)  working places where it is not provided an adequate lighting (mine, tunnel, pipes) The photometric characteristics of the background material are shown in Table 1. The physico-mechanical characteristics of the background material are following:  tensile strength (minimal) warp 1000 N weft 850 N  tear strength (minimal) warp 50 N weft 30 N  abrasion strength (minimal) 500 cycle  water permeability (minimal) 50 mg/sqcm 24 h 89 Table 1. Photometric characteristics of the background material. Fluorescent Yellow Orange - Red Chromaticity co-ordinates X = 0.4235 Y = 0.5296 X = 0.5980 Y = 0.3641 Reduced luminance factor ( E ) 0.9400 0.4008 Dyeing resistance on  abrasion  perspiration  washing  chemical cleaning  bleaching  ironing 4 – 5 4 – 5 4 – 5 4 – 5 4 5 4 4 4 – 5 4 – 5 4 5 High-visibility warning clothing: 2nd class Structure Background material: 85 % PES / 15 % Co fabric 50 mm width retro- reflective stripes, 100 % Co fabric. 2nd visibility class (Figure 1):  background material: 0.70 sqm  retro-reflective material: 0.20 sqm Destination Protective clothing for preventing hurting the person, by coming up against objects, that are moving with max. 50 km/h velocity, because this type of clothing provide the signs of the person in any humidity conditions against superficial mechanical aggression of medium intensity (abrasion, tearing). End-uses  public roads, with velocity restriction, for moto / auto vehicles  railway traffic  wood industry High-visibility warning clothing: 3rd class Structure Background material: 85 % PES / 15 % Co fabric 50 mm width retro-reflective stripes. 3rd visibility class (Figure 2):  background material: 2.50 sqm  retro-reflective material: 0.37 sqm Destination Protective clothing: for preventing hurting the person, by coming up against objects, that are moving with high velocity, because this type of clothing provide the signs of the person in any humidity conditions against superficial mechanical aggression of medium intensity (abrasion, tearing). End-uses  public roads, for moto / auto vehicles  railway traffic  wood industry Figure 1. 2nd visibility class. Figure 2. 3rd visibility class. 90 The effectiveness of phase change materials in outdoor clothing Huensup Shim, Elizabeth A. McCullough Institute for Environmental Research, Kansas State University, Manhattan, KS 66506 USA Introduction Phase change materials (PCMs) used in textiles are combinations of different types of paraffins - each with different melting and crystallization (i.e., freezing) points. The PCMs are enclosed in a microcapsule to prevent leakage of the material during its liquid phase (Bryant & Colvin, 1992). PCM microcapsules can be incorporated into the spin- ning dope of manufactured fibers (e.g., acrylic), incorporated into the structure of foams, and coated onto fabrics. When the encapsulated PCM is heated to the melting point, it absorbs heat energy as it goes from a solid state to a liquid state. This phase change produces a temporary cooling effect in the clothing layers. The heat energy may come from the body or from a warm environment. Once the PCM has completely melted, the storage of heat stops. If the PCM garment is worn in a cold environment where the temperature is below the PCM’s freez- ing point, and the fabric temperature drops below the freezing point, the PCM liquid will change back to a solid state, generating heat energy and a temporary warming effect. The developers and producers of phase change materials in textiles claim that garments made with PCMs will keep a person warm longer than conventional insulations when worn in cold environments and decrease the thickness and weight of the clothing required (Pause, 1998). In order for a phase change material to improve the thermal comfort characteristics of a clothing ensemble in the cold, it must produce enough heat in the garment layers to reduce the heat loss from the body to the environment. Since there is a temperature gradient from the warm skin surface to the clothing layers to the cold envi- ronment, some of the PCMs may not change phase. In addition, small increases and decreases in skin temperature resulting from changes in body heat production (activity) may not be enough to keep the phase change process going in the cold. The magnitude and duration of the phase change heating and cooling effects on a clothed body have not been documented in the scientific literature. Most of the published research that is available was conducted on small pieces of fabric (Pause,1995; 1998). Therefore, there is a need for independent research on the use of PCMs in clothing systems. The purpose of this project was to quantify the effect of PCMs in fabric-backed foams on heat loss from a thermal manikin’s surface to the environment during environmental temperature transients and on human subjects’ physiological responses and comfort per- ceptions during environmental temperature transients and changes in activity. Identical fabrics and garments - with and without the PCMs - were compared. 91 Methodology An open-cell, hydrophilic polyurethane foam was produced directly on a fabric substrate of polyester knitted fleece. The experimental foam contained 60 % PCM microcapsules and 40 % foam by weight. Approximately 40 % of the PCM microcapsules changed phase at 28.3 °C (83 °F), and 60 % of them changed phase at 18.3 °C (65 °F). Suits for the manikin tests consisted of a fitted long-sleeved top and long pants. Ski jackets and pants were made for the human tests. Environmental step change tests with a manikin A life-size, computerized, thermal manikin was used to simulate the heat loss from a per- son to a cooler environment and to measure the insulation (clo) value of the clothing systems. The manikin has 18 body segments which provide independent temperature control and measurement. His skin temperature distribution simulates that of a com- fortable human. The environmental temperature transient was achieved by using two adjacent chambers under still air conditions: one simulated a warm indoor environment: 25 °C (77 °F) air temperature, 13.9 °C (57 °F) dew point temperature, and one was a cold outdoor envi- ronment: 10 °C (50 °F) air temperature, 5.8 °C (42.4 °F) dew point temperature. First, the insulation value of the clothing and the average level of body heat loss was measured in the warm chamber according to ASTM F 1291. Then the manikin (hanging from a track by a hook in his head) was moved quickly to the adjacent cold chamber, and the power to the manikin was recorded every minute during the transient. After the tran- sient was over, the process was repeated as the manikin was moved back to the warm chamber. Exercise/rest tests with human subjects Sixteen male subjects wore ski ensembles with and without PCMs during a 90 minute exercise/rest protocol that simulated skiing. The experiment began with the subjects sit- ting (1 MET) in a warm chamber with an air temperature of 25 °C (77 °F) and a dew point temperature of 13.9 °C (57 °F). Then the subjects sat in an adjacent cold chamber with an air temperature of -4 °C (24.8 °F) and a dew point temperature of -7.1 °C (19.3 °F). Every 15 minutes, they alternated walking on a treadmill at 4.0 mph and a 3 % incline (5.5 MET) and sitting in the cold. The subjects’ four skin temperatures were measured every minute during the exercise/ rest protocol using a computerized data acquisition system. Subjects’ perceptions were measured at the end of each 15 minute period using ballots. Thermal sensation was rated from #1 very cold to #5 neutral to #9 very hot. Clothing comfort was rated from #1 uncomfortable to #5 comfortable and from #1 clammy to #5 dry using a semantic differential scale. In addition, the amount of unevaporated sweat in the ski ensemble was determined by weighing the garments before and after the 90 minute experiment. Results of manikin tests during environmental transients Insulation values for the PCM suits were slightly higher than those measured for the suits without PCM. The one layer suit with PCM was 1.57 clo and without PCM was 1.48 clo. 92 The two layer suit with PCM was 2.07 clo and without PCM was 1.95 clo. The insulation values for the ski ensembles with and without PCM were the same (2.19 clo). Heat loss data from two replications of the environmental transient tests with the manikin were averaged for each minute of the data collection period. Unfortunately, the experimental suits – with and without the PCM – were not exactly equivalent with re- spect to their heat loss and insulation values under steady-state conditions. To correct this problem, the average heat loss measured for each suit at steady-state was used to correct the transient data. Then the difference between the PCM curve and the no-PCM curve was plotted for each suit type. The difference curves indicated that the effect of PCMs on body heat loss lasted approximately 15 minutes. Therefore, the magnitude of this buffer- ing effect was calculated and reported as the integrated total difference in heat loss for the 15 minute period (J) and as the average difference in the rate of heat loss during the 15 minute period (W). (See Table 1.) This is not the amount of heat actually produced or stored by the PCM in the material. Some of this heat is lost to or gained from the envi- ronment. Table 1. Heating effects of clothing for first 15 minutes of warm to cold temperature transients.a Garments with PCM compared to garments without PCM Average heat- effect (W) Total energy gain (J) One layer suit with PCM vs. without PCM 6.5 5 850 Two layer suit with PCM vs. without PCM 13.2 11 880 Two layer suit (inner layer with PCM, outer layer without PCM) vs. two layer suit without PCM 4.5 4 050 Two layer suit (outer layer with PCM, inner layer without PCM) vs. two layer suit without PCM 9.9 8 910 Two layer suit with PCM on top 39 % of body no PCM on bottom 42 % of body vs. two layer suit without PCM 6.3 5 670 PCM ski ensemble vs. ski ensemble without PCM 5.9 5 310 a Cooling effects were similar to heating effects, so they were not given here to conserve space. The manikin lost an average 6.5 W less heat during the first 15 minutes of wearing the PCM suit as compared to the control suit in a warm to cold transient. The heating effect of the two layer suit was approximately double that for the one layer suit. The heating effect of the two layer suits containing only one layer of PCM were less than the effects produced by the two layer PCM suit. There are less PCM microcapsules in one layer of clothing (as compared to two layers) to go through the phase change. In addition, there was a larger heating effect when the PCM layer was on the outside than when it was on the inside, next to the body. These layering differences may not be as great in clothing systems that fit more loosely on the body and contain thicker air layers. The heating ef- fect of a two layer suit containing PCM on the top half of the body only was less than the effect produced by the two layer PCM suit with full coverage. In conclusion, the magni- tude of the heating and cooling effects associated with phase change materials in clothing increases as the number of PCM garment layers increases and as the amount of body coverage with PCM fabrics increases. Results of exercise/rest tests with human subjects An analysis of variance was conducted on the skin temperature data and subjective data collected near the end of each 15 minute period. When the subjects moved from sitting in 93 a warm chamber to sitting in a cold one, their average leg skin temperature was signifi- cantly higher when wearing the PCM ski ensemble (31.5 °C, 88.7 °F) than the ensemble without PCMs (30.3 °C, 86.5 °F). The chest, arm, and finger temperatures and subjective responses were not statistically different, probably because subject variability was high and the average heating effect of the PCM ski ensemble was low (about 6 W). There were no significant differences in the skin temperatures and thermal sensations of males wearing the PCM ski ensemble compared to the control during the four remaining exercise/rest periods in the cold. When the subjects were walking at 5.5 MET in the cold environment, they were producing more than five times the amount of heat than they did while sitting (1 MET). This extra body heat did not raise their skin temperatures to pro- portionately higher levels and heat the PCMs so that they would change back to liquid. Instead, the body dissipated the extra heat energy by sweating. When the experiment was over, there was significantly more unevaporated sweat in the PCM ensemble, and the subjects felt significantly more clammy when wearing it during the last exercise period and rest period of the test. Conclusions Phase change materials produced a small, temporary heating or cooling effect when gar- ments made of the material went through a step change in temperature. The PCM heating and cooling effects changed body heat loss by an average 2-13 W for the first 15 minutes of the environmental transient, and then the effects were over. The magnitude of the ef- fect increased as the number of garment layers and the amount of body coverage with PCM fabrics increased. When 16 male subjects moved from a warm to a cold environment, their average leg skin temperature was significantly higher when wearing the PCM ski ensemble as com- pared to the control. However, their other skin temperatures and their subjective re- sponses were not significantly different during the environmental step change at low ac- tivity. There were no significant differences in the skin temperatures and thermal sensa- tions of males wearing the PCM ski ensemble during the four remaining exercise/rest pe- riods in the cold. The effect of PCMs in clothing would probably be maximized if the wearer were repeatedly going through temperature transients or intermittently touching hot or cold objects with PCM gloves. References Bryant YG & Colvin DP (1992) Fibers with enhanced, reversible thermal energy storage properties. Techtextil-Symposium, 1-8. Pause BH (1995) Development of heat and cold insulating membrane structures with phase change material. Journal of Coated Fabrics, 25 (7), 59-68. Pause BH (1998) Development of new cold protective clothing with phase change material. International Conference on Safety & Protective Fabrics ‘98, 78-84. 94 Protective equipment against heat and/or fire produced from performant fibres Doina Toma, Eftalea Carpus, Emilia Visileanu The National R-D Institute for Textile and Leather, Bucharest, Romania Introduction The effects on health and physical integrity of the persons exposed to risks caused by heat and/or fire are the most serious. This led to paying special attention to this individual protective equipment (IPE) category. In order to assure the protective level suitable for different exposure and intensity levels, there have been created special types of IPE. This requirement was imposed by the different ways the risks of heat and fire appear, for instance for the heat (various types with specific effects: contact, convective or ra- diant) and for the fire (short time occasional contact with flame, projection of incandes- cent or molten metal particles, etc.). Being used against heat rays and/or against fire, the IPE has fulfilled the following requirements:  the heat quantity transmitted to the user has to be small enough so as the whole quan- tity accumulated during the entire period of operation in that part of the body which needs to be protected not to reach the pain, s break point or other boundaries that could damage one person, s health;  IPE has to obstruct, if necessary, the liquid or the vapor to penetrate and also has to avoid burns at the contact between the protective layer and the user. The risk of inflaming the protective clothing can be caused by:  the occasional contact with the flame;  high intensity heat rays;  direct contact with glowing materials. In order to establish the potential range of use was necessary to emphasize the concrete usage conditions of the IPE against heat and/or fire. We started from the assumption that only a fair evaluation of the accident or rendering sick risks results in choosing and utilizing in high efficiency conditions the IPE. The analysis of the risks for the working places emphasizes the presence of one type of risks very rare, in most of the cases we are faced with the cumulative action of many types of risk. The most frequent case in this respect is a cumulating of risks determined by the superficial mechanical aggressions as a result of working with pieces endowed with rough surfaces and sharp ribs. In order to be able to assure an adequate protection, in designing and accomplishing some new IPE-s against heat and fire, it is necessary that we consider these risks. Experimental part In order to assure a suitable protection against fire and heat, heat-resisting fibers are used. Heat-resistance of a polymer is commonly characterized by three aspects: 95  the level of basic characteristics (physical and chemical);  to preserve these characteristics in a given temperature area;  to maintain these characteristics in a certain time period at a given temperature. In the particular case of heat-resisting fibers destined for thermal IPE, the most impor- tant factors are:  the level of the basic characteristics imposed by the specific domain of usage (me- chanical resistance, resistance to chemical agents);  to preserve the integrity of the surface of the protection textile structure in the case of: - short time exposure to an open radiant flux (heat resistance: many more seconds at 1000 °C); - flame closeness (heat resistance: many more minutes at 350 °C). Accomplishment of IPE involves:  estimation of the dangers and of the risk levels;  delimitation of the basic characteristics, specific and additional of the textile support;  designing the textile support taking into account the performance requirements im- posed by the domain of usage;  testing according to the European standards;  designing and accomplishing the IPE-s in order to obtain the protection, ergonomic and comfort characteristics;  certification of the IPE. For the thermal IPE, the textile support is submitted to the most severe requests of fire resistance and temperature, the primary and the most important condition being not to maintain the burning. That’s one of the reasons for which in accomplishing the respective woven fabric, some fibers with special characteristics are used, namely inherent inflam- mable fibers, resistant to different chemical agents, and in the same time with small com- bustion gas emission, even at carbonizing temperatures. The Twaron para-aramidic fibers produced by Akzo combine the good processing properties of the high temperatures resistant textile fibers with the high level resistance of the organic fibers (Table 1). Table 1. Characteristics of Twaron fibers. Length Density, dtex 1.78 Individual Length, mm 38.5 Tensile Breaking Strength, cN 27.6 Breaking elongation, % 5.3 Tenacity, cN/tex 15.5 Breaking modulus, cN/dtex 290.5 The fibers used for experimentation have been processed on the conventional spinning systems, such achieving 37 tex yarns (Table 2). Table 2. Performance characteristics of the obtained 100 % Twaron textile support. Length density (warp and weft), tex 37x2 Weight, g/m2 294 Breaking resistance, N, warp/weft 5500/4700 Breaking Elongation, %, warp/weft 12/7 Tearing Resistance, N, warp/weft 750/680 Wearing Resistance (00 abrasive) no. of cycles more than 2000 Water vapor permeability, mg/cm2 24 h 665 Air permeability, l/m2 - s 200 96 Testing the protection performances against heat and/or fire In order to finalize the usage domain of the IPE-s accomplished from 100 % Twaron woven fabrics, some specific tests have been made (Table 3). Table 3. The tests that were carried out on 100 % Twaron woven fabrics Characteristic Obtained performances Testing method Limited flame propagation Corresponds to 6/2 SR EN 531 and EN 533:  doesn’t propagate the flame: post burning time after removing the ignition flame  1 s;  during the exposure the flame doesn’t touch the superior or the vertical margin of the test bar;  doesn’t release any molten, burning or burnt rests;  test bars don’t get perforated;  medium time of after glowing- 1 s. 6.2.2./SR EN 531 after a perpendicular flame exposure in accordance to SR EN 532 Fire behavior In accordance to 5.1/SR EN 407, in the initial state and after washing:  medium time of flame’s persistence- 0 s;  medium time of residual after glowing- 0 s. 6.3 SR EN 407 + 5.1 SR EN 407 Convection heat resistance In accordance to 5.3/SR EN 407 Performance level 4 HTI ! 18  time to increase the internal force temperature with 40 °C at a direct flame exposure in a 50 mm distance, caloric flux of about 80 kw/m2: more than 25 s Romanian Standard adjusted after 6.5/SR EN 407 + EN 367 Contact heat resistance In accordance to 5.2/SR EN 407 Performance level 3  breaking point time: 20 s for a contact temperature (of the plate) of 350 °C 6.4/SR EN 407 + Romanian Standard adjusted after pr. EN 702 Small fluid metal projections In accordance to 5.5/SR EN 407 Performance level 2  the number of particles that may lead to a growth of the internal force temperature with 40 °C: 15 min. 6.7/SR EN 407 + EN 348 Large fluid metal projections In accordance to 5.6/SR EN 407 Performance level 1  the quantity of melted cast iron that doesn’t bring about any flow signs or any changes in the exterior surface of PVC foil 6.8/SR EN 407 + Romanian Standard adjusted after EN 373 Taking into account the very good performances assured by the accomplished textile supports in both protection against heat and/ or fire (guaranteed for the entire period of service because of the maintenance of parameters by using performance Twaron para- aramidic fibers) and in mechanical characteristics with reference to breaking and tearing resistance, the following prototypes were designed and accomplished:  anticaloric clothing used as fire-proof clothing against convection heat; this clothing is specially designed for works effected near by some important heat sources (low level risk, located heat and / or fire exposure) where short time contact with the open flame, or with incandescent molten metal particles (sparks) is possible, or in the presence of risks caused by superficial mechanical aggressions (medium level risk – abrasion, punch, hanging);  anticaloric gloves with one finger, accomplished from three layers of textile materials:  they are designed for works effected near by important heat sources, medium level, which assure the thermal isolation against contact heat (when handling hot objects with temperature higher than 350 °C); 97  they are used against convection heat, against projection of particles or medium quantities of molten cast iron, as well as against short time chance contact with the open flame;  their main characteristics are: abrasion, punch, tearing and cutting resistance. The accomplished products, tested and certified in accordance to the law in force satisfy the main security and health requirements, suitable for different usage domains. Conclusions  To assure the security and the health is a main concern of the working process, and we have to refer to it when creating the working life, if we want to obtain its maxi- mum efficiency.  The protection effect of an IPE depends mainly on its structure. In order to achieve the best structure it is required to use performance fibers, which assure quality, multi- functionality and comfort.  In order to accomplish the woven and/or knitted structures for IPE against thermal and /or mechanical aggressions Twaron para-aramidic fibers can be used. Acknowledgments We wish to express our thanks to AKZO NOBEL Comp. for their outstanding support. Literature Adanur S (1995) Safety and Protective Textiles, Wellington Sears Handbook of Industrial Textiles. Bajaj P & Sengupta AK (1992) Protective Clothing, Textile Progress, 22 (2/3/4). 98 Dynamics of sweat vapour sorption as the function of physical parameters of textile packets under protective barrier Grayna Bartkowiak Central Institute for Labour Protection, 90-133 ód, Wierzbowa 48, Poland Introduction Protective clothing made from barrier fabrics very often puts a thermal strain on its wearer. This takes mostly place during intensive manual labour (Beckett et al., 1986). Under these conditions, the body begins to intensely perspire, while, during wear of her- metic barrier clothing, only a small quantity of sweat can evaporate from the skin. The effect is a rapid growth in the humidity of the microclimate and the accumulation of moist sweat on the skin. As proven by many research studies, there is a significant influence of textiles worn near the body on the intensity of heat and sweat removal - a determinant of under- garment microclimatic parameters as well as of a level of discomfort (Umbach, 1980, Umbach Sonderdruk). The phenomenon particularly applies to fabrics worn under pro- tective barrier clothing. When moist sweat appears on the skin, a fabric of under-barrier clothing ought to enable its removal and then its effective absorption. In order to elimi- nate the sensation of wetness, resulting from moisture sorption, the layer of an under- barrier fabric which adjoins the skin should be made from a raw material characterized by minimal moisture sorption and a structure providing good capillary transport. It is thus advisable to assume that the under-barrier fabric should have a two-layer structure. The first, inner, conductive-diffusive layer should enable the diffusion of water vapour and moisture capillary transport, whereas the second, sorptive one should arrest humidity within its structure. The purpose of the research work was to characterize an effect of diversely structured textile packets used under the protective barrier on shaping the under-garment microcli- mate and, consequently, to select the packet optimal for use during intensive physical effort. Materials and methods The basic test materials were one-component knitted fabrics made from five different types of textile raw materials, i.e., polyester (PES), polypropylene (PP), cotton (CO), viscose (CV), and wool (WO), manufactured in two stitches: interlock and float special pique. The investigation also encompassed unwoven fabrics containing superabsorbents in the form of fibres and powder. Characteristics of the fabrics produced for the investi- gation were presented in Table 1. The materials were also tested in two-layer systems: the near-body hydrophobic layer covered with the hydrophilic one. The dynamics of humidity of the microclimate was examined at the test stand shown in Figure 1 (Zieliski, 1997). The lower part of the chamber contains a porous glass agglomerate, supplied with water and kept at the constant temperature of 30±1 °C, which 99 simulates the perspiring human skin. Initially, the humidity in the microclimatic zones was equal to approximately 15±2 %. The velocity of water vaporization in the experi- ment amounted to 0.85 ml/m2·min. The investigation was carried out in the chamber at the constant temperature of 30±1 °C. The packet of the tested materials was placed above the skin surface: - an under-barrier clothing fabric or system of fabrics, - the protective barrier (polyamide-polyester fabric coated with PVC). Table 1. Characteristics of under-barrier fabrics. Variant Kind of material Raw materials Mass per square meter [g/m2] Thickness [mm] pressure 0.5 kPa Water vapour resistance*) [m2˜Pa/W] a1 knitted fabric - interlock PES textured 218 1.01 3.9 a2 knitted fabric - interlock PP textured 228 1.09 3.8 a3 knitted fabric - interlock CO 299 1.42 4.5 a4 knitted fabric - interlock CV 277 1.05 3.3 a5 knitted fabric - interlock WO 309 1.70 5.8 b1` knitted fabric – RL float special pique PES textured 127 1.48 4.6 b2 knitted fabric – RL float special pique PP textured 158 1.66 4.7 b3 knitted fabric – RL float special pique CO 212 1.41 4.5 b4 knitted fabric – RL float special pique CV 182 1.22 3.2 b5 knitted fabric – RL float special pique WO 272 2.25 6.9 p unwoven fabric PES, PP powder Favor SAB-954 366 1.58 6.3 w unwoven fabric PP, CV, fibres Fibersorb 900 94 0.36 2.9 * ) according to ISO 11092 THERMOSTAT Water Measuring system Sweating surface Textile packet Barrier (o) Figure 1. Diagram of the apparatus measuring humidity dynamics of the microclimate. 100 The layers of the packet divided the under-barrier space into the two zones (I, II) where the humidity of the microclimate was investigated. Besides, the humidity above the bar- rier (zone III) and temperatures on the surface of each layer of the packet and the ‘perspiring skin’ were recorded. The hermetic water-steam-tight barrier used in the ex- periment caused that the humidity in zone III of the microclimate remained constant. Re- sults of the measurements were expressed as graphs illustrating the dynamics of humidity changes in zones I, II and III of the microclimate and the temperature on the surface of the textile packet (Figure 2). 0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 time (min) R e la tiv e hu m id ity (% ) . 30 31 32 33 34 35 36 37 38 39 40 Te m pe ra tu re (°C ) . M Q M , M , , M , ,, M S temp II III IIII - humidity in zone I, II and III respectively for the packet: knitted fabric a4 + barrier Ip In - humidity in zone I and II respectively for the barrier itself (o) Figure 2. Results of measurements of dynamics of the microclimate humidity and temperature on the surface of the under-barrier fabric for the packet: viscose knitted fabric (design a4) + hermetic barrier. In order to make a comparative analysis of the results of measurements, indices characterizing dynamics of the microclimate were assumed. The dynamics of humidity growth above the ‘perspiring skin’ was calculated by means of the formula (Figure 2): where: II - humidity in zone I of the microclimate, Ip - humidity in zone I of the microclimate without the under-barrier fabric, t - time. The dynamics of humidity growth above the under-barrier fabric was calculated by means of the following formula (Figure 2):   120 0 120 0 )( )( dtt dtt Dn n II I I where: III - humidity in zone II of the microclimate, I n - humidity in zone II of the microclimate without the under-barrier fabric, t - time.   120 0 120 0 )( )( dtt dtt Dp p I I I 101 Indices Dp10 and Dn10, characterizing the dynamics of humidity growth above the ‘perspiring skin’ and the under-barrier fabric, were calculated similarly 10 minutes after the beginning of the experiment. Results The fabrics were assessed in comparison to the proposed two-layer model designed for use under the protective barrier. The first layer - made from a hydrophobic fabric - should be characterized by high permeability and low absorptivity of water vapour. Therefore, indices Dp and Dn as well as Dp10 and Dn10 ought to be high. In this respect, the best prop- erties were shown by the float special pique fabrics manufactured from synthetic fibre textured yarn (b1, b2), especially the polyester knitted fabric - design b1 (Dp and Dn near 1). The second layer of the under-barrier fabric should show high sweat vapour sorption and, thanks to this quality, be able to reduce humidity of the microclimate near the skin. Therefore, indices Dp and Dn ought to be adequately low. Comparing the raw textiles used for the sorptive layer, a high rating was given to viscose and wool yarn knitted fabrics. The best properties were shown by the interlock wool knitted fabric (Dp =0.62, Dn =0.50), yet it is necessary to add that this design was at the same time characterized by high su- perficial mass. Bearing in mind the low superficial mass of the unwoven fabric with high-sorptive fibres, one should take note of high water vapour sorption efficiency (Dp =0.76, Dn =0.66). On the other hand, no direct correlation between the of water vapour resistance and the effect of the fabrics on shaping the under-garment microclimate was observed. It follows that in order to analyse an influence of fabrics worn near the body upon forming the under-barrier microclimate it is necessary to research into the dynamics of humidity for the whole packet of clothing (under-barrier fabric + barrier). Conclusion The research work done made it possible to describe the impact of the structure of the fabric packets placed underneath the hermetic protective barrier upon shaping the dy- namics of the under-clothing microclimate. Taking into consideration the proposed two- layer fabric model designed for use under the hermetic barrier, it was affirmed that the most suitable materials to be used for the near-body layer are the open-structured pique knitted fabrics made of synthetic fibres. The materials best fitted for the second, sorptive layer are those characterized by the tight structure and made from the raw materials of high water vapour sorption, particularly in the first phase of the experiment - i.e., viscose or wool knitted fabrics, or high-sorptive fibre unwoven fabrics. Reference literature Beckett WS, Davis JE, Vroman N, Nadig R & Fortney.S (1986) Heat stress associated with the use of vapour-barrier garments. Journal of Occupational Medicine, 28 (6), 1986, 411-414. Umbach KH (1980) Hautnahe Synthetics mit gutem Tragekomfort. Chemiefasern/Textilindustrie, august, pp 628-636 Umbach KH Tragekomfort von Kleidungskombinationen, Sonderdruck. Bekleidungsphysiologisches Institut Hohenstein e.V. Zieli  ski J & Majewski P (1997) A Measuring System for the Assessment of Moisture Transport Through Textiles, Fibres and Textiles in Eastern Europe, July/September, 57-59. 102 Psycho-physiological mechanisms of thermal and moisture perceptions to the touch of knitted fabrics Junyan Hu, Yi Li Institute of Textiles and Clothing, The Hong Kong Polytechnic University, Hung Hom, Hong Kong, China Introduction Being defined as freedom from pain and discomfort, comfort is a fundamental and uni- versal need of consumers. Meanwhile, closely integrated with human life, comfort of clothing plays an important role in human sensory experience (Hatch, 1993) and for the clothing industry, the final goal is improve the quality of life and the survival of human beings. Many articles in the literature investigated the factors involved in clothing com- fort (Lake & Hughes, 1980; Nielsen & Endrusick, 1990; Schneider & Holcombe, 1991; Li et al., 1996,). Thermal and moisture sensations have been widely recognised as the most important factors contributing discomfort sensations. In this paper, we study the psycho-physiological mechanisms of the thermal and moisture sensations by conducting a series of perception trials using knitted fabrics. Neurophysiological basis of the thermal and moisture sensations Based on Hensel's research (Hensel, 1981), Li et al developed a mathematical equation as following to describe the neurophysiological response of this process: C t tyTktyTKtyQ skdsks  w w  ),(),(),( (1) Where Q (y, t) is the pulse output response of the thermoreceptors as a function of time t and at the depth y of the thermoreceptors below the skin surface in impulses. K s = -0.72, Kd = -50 and C=28.1. The integrals of the thermoreceptor frequency output curves, named as psychosensory intensity (PSI value), were calculated to describe the intensity of the thermal sensations and the stated as equation (2):   t dttyQPSI 0 ),( (2) To study the psychophysical relationship, the maximal skin temperature change rate (MTC) also defined as: max|t TMTC    (3) Experimental All the tests were carried out in a conditioned laboratory, controlled at 20 °C ±0.5 °C and relative humidity of 63 %±1 %. Fifteen subjects (seven males and eight females, ranging in age from 20 to 41) participated in the trial. Each session consisted of testing two sets of specimens, one for dampness perception and another for coolness perception. A series of fabrics were used in this study and detailed in Table 1. Fabrics were considered at 103 “equilibrium regain", when the moisture content of fabrics was in equilibrium with the ambient condition. Additional water added to fabrics was termed as “excess moisture”, which was expressed as a percentage of fabric weight at equilibrium regain. Three levels of moisture were used: the 0 % (i.e. the “dry” fabric in equilibrium with ambient condi- tion), 4 % and 16 % of excess moisture. Table 1. Fabrics used in experimental program. Sample Code Fiber Content Fabric Construction Mass/Unit area g/m2 Thickness at 98 pa pressure mm K1 Nylon knitted 90.6 0.27 K2 Darcon knitted 178.5 0.95 K4 Orlon knitted 252.0 1.23 K8 30 % Cotton / 70 % Polyester knitted 200.4 0.62 K9 Cotton knitted 356.6 0.94 K15 20 % polyester / 70 % wool / 10 % nylon knitted 454.6 3.06 K16 25 % Cotton / 75 % Polyester knitted 319.4 1.52 K17 Nylon knitted 231.4 2.64 The weights of test fabrics were determined prior to the testing after conditioning in the ambient condition for 24 hours. Before test, the desired moisture contents for an indi- vidual fabric were introduced by adding water to the fabric to required level of excess moisture. And then stored in individual seal storage bags overnight to make sure they have reached equilibrium at the certain excess moisture. In the trial, non-comparative un- balanced and forced rating scales are used for tester. Subjective scores on coolness and dampness were recorded on two scales, which range from “dry” to “Extremely Damp” and “Slight Warm” to “Cold” respectively. The scales marked with text labels faced to the subjects and the numeral scales faced to the operator on the rear of the ruler. A pointer, which was indicated by a subject on the text scale, provided a corresponding numerical value that was recorded by the operator. Prior to the trial, we fixed two type-T thermocouples on the inner surface of subject’s forearm and connected to a portable PC. During the trial, we began to record the temperatures before the fabric was placed on the forearm and then continued to record during the fabric-skin contact. In the experiment, the arm of subjects was kept static and the operator placed the fabric samples across the subject’s forearm. The apparatus for temperature measuring can be shown in Figure 1. Typical skin temperature-changing curve during the fabric contact is shown in Figure 2 and the calculation of measured temperature-changing rate and Q(t) at the skin surface are shown in Figure 3. 29 29.5 30 1 3 5 7 Time (sec) Te m pe ra tu re a t s kin su rfa ce (de g C ) Figure 1. Apparatus for temperature measurement. Figure 2. Measured temperature changes at the skin surface. 104 -0.6 -0.2 0.2 1 3 5 7 Time (sec) dT /dt Figure 3. Calculation of temperature change rate and impulse on the skin surface. Results and discussion To reduce the influence of individual bias on subjective coolness and dampness judge- ment, all the subjective perception data had been standardised before the mean calculated and the results are shown in Figure 4. The means of MTC and PSI are shown in Figure 5 too. Both subjective perceptions and psycho-physiological indexes on knitted fabric are highly correlated with the water content in the fabric and increased with it. 0 20 40 60 80 100 Dry 4% 16% Water Content Sl igh t W ar m - - Co ld K! K2 K4 K8 k9 K15 K16 K17 0 25 50 75 100 Dry 4% 16% Water Content Dr y - - Ex tre m ely D am p K! K2 K4 K8 k9 K15 K16 K17 Figure 4. The means of subjective perception on coolness and dampness. -1.2 -0.8 -0.4 0.0 Dry 4% 16% Water Content Dt /dt m ax K! K2 K4 K8 k9 K15 K16 K17 0 50 100 150 200 250 300 Dry 4% 16% Water Content PS I K! K2 K4 K8 k9 K15 K16 K17 Figure 5. Means of MTC and PSI on the skin surface. Using the GLM General Factorial and GLM Multivariate methods, we obtained the significance of influence of water content, fibre content on the perception of coolness and dampness, which are shown in Tables 2~4 respectively. As shown in the tables, the level of water content in the fabrics and fibre content have significant influence on the subjec- tive sensations with P<0.001, suggesting that coolness and dampness sensations are not only determined by the level of water content in the fabric, but also the types of fibres in the fabric. Similarly, the objectively measured skin temperature changes and PSI are also significantly influenced by the excess moisture with p<0.001 and fibre content with P<0.002. These results show that both subjective perceptions and the neurophysiological indexes (MTC and PSI) depend on the excess moisture and fibre content, indicating that 105 the subjective sensations may be determined by the two factors through the neuro- physiological responses of the thermoreceptors. Table 2. Analysis of variance on subjective coolness ratings. Source Type III Sum of Squares df Mean Square F Sig. Corrected Model 164536.0 20 8226.8 22.8 .000 Intercept 740681.7 1 740681.7 2049.1 .000 Water content 107214.7 2 53607.3 148.3 .000 Fiber content 49025.5 6 8170.9 22.6 .000 Table 3. Analysis of variance on subjective dampness ratings. Source Type III Sum of Squares df Mean Square F Sig. Corrected Model 153786.0 20 7689.3 16.2 .000 Intercept 609691.6 1 609691.6 1286.0 .000 Water content 102457.6 2 51228.8 108.1 .000 Fiber content 37560.0 6 6260.0 13.2 .000 Table 4. Analysis of variance on objectively measured MTC and PSI. Source Dependent Variable Type III Sum of Squares df Mean Square F Sig. Corrected Model MTC 13.0 20 .7 5.2 .000 PSI 589910.7 20 29495.5 6.1 .000 Intercept MTC 78.2 1 78.2 628.8 .000 PSI 4569584.1 1 4569584.1 943.5 .000 Water content MTC 8.6 2 4.3 34.6 .000 PSI 449219.5 2 224609.8 46.4 .000 Fiber content MTC 3.0 6 .5 4.0 .001 PSI 130212.8 6 21702.1 4.5 .000 Furthermore, the dampness sensation is highly correlated with the coolness sensation, as shown in Figure 6, suggesting that the perception of temperature and moisture may be determined by the same psycho-physiological mechanisms: the neurophysiological responses of skin thermorecptors to skin temperature changes during the fabric-skin contact. The relationships between the sub- jective perceptions of coolness and dampness the neurophysiological in- dexes (MTC and PSI) are shown in Figure 7~10. The coolness perception is highly correlated with the PSI and MTC with R2 of 0.79 to 0.82 respectively, which are significant at level of P<0.001. The linear relationship between the two suggests that the coolness perception be determined by the psycho-physiological mechanism of thermoreceptors' response to the skin temperature changes. Similarly, the dampness perception is also linearly related to MTC and PSI with the R2 of 0.81 and 0.84 respectively, showing that the same psycho- physiological mechanism determine the subjective perception of moisture in the fabrics. Comparing the results for coolness and dampness sensations, good agreement between the two confirms our assumption that perception of moisture is determined by the same psycho-physiological mechanism as the perception of temperature in the body. y = 0.9089x + 0.5659 R2 = 0.9389 0 50 100 0 20 40 60 80 100 Slight Warm -- Cold D ry -- Ex tr em el y Da m p Figure 6. The relationship between coolness and dampness. 106 y = 0.3584x R2 = 0.7883 0 25 50 75 100 0 50 100 150 200 250 PSI Sl ig ht W ar m - - C o ld y = 79.987x R2 = 0.8153 0 25 50 75 100 0.0 0.5 1.0 1.5 abs(MTC) Sl ig ht W ar m - - Co ld Figure 7. Coolness and PSI. Figure 8. Coolness and MTC. y = 0.3318x R2 = 0.8064 0 25 50 75 100 0 50 100 150 200 250 PSI D ry -- Ex tr em el y Da m p y = 74.158x R2 = 0.8445 0 25 50 75 100 0.0 0.5 1.0 1.5 abs(MTC) D ry -- Ex tr em el y Da m p Figure 9. Dampness and PSI. Figure 10. Dampness and MTC. Conclusions By conducting a series of psycho-physiological perception trials and measuring the skin temperature changes during the fabric-skin contact process, we found that subjective per- ception of coolness and dampness is highly correlated with each other and significantly influenced by fabric excess moisture and fibre content. The perception of moisture and temperature is shown determined by the same psycho-physiological mechanism: the neu- rophysiological responses of thermoreceptors to the skin temperature change during the fabric-skin contact. Acknowledgement We would like to thank The Hong Kong Polytechnic University for the funding of this research. References Hatch KL (1993) Textile Science. West Publishing Company. 459. Hensel H (1981) Thermoreception and temperature regulation. New York: Academic Press. Lake B & Hughes JL (1980) Moisture studies in the domestic environment. 1. Dampness perception in laundered articles. Journal of Consumer Studies and Home Economics, 4(1): p. 87-95. Li Y, Holcombe BV & de Dear R (1996) Enhancement of Coolness to the Touch by Hygroscopic Fibers. Textile Research Journal, September, 66(9): p. 587-594 Nielsen R & Endrusick TL (1990) Sensations of temperature and humidity during alternative work/rest and the influence of underwear knit structure. Ergonomics, 33(2): p. 221-234. Schneider AM & Holcome BV (1991) Properties influencing coolness to the touch of fabrics. Textile Res. J., 61(8): P488-494. 107 Combined effects of fabric moisture absorbancy and air permeability on thermophysiological responses in the warm environments Hiromi Tokura Department of Environmental Health, Nara Women's University, Nara 630-8506, Japan Abstract I talk on combined influences of fabric moisture absorbancy and air permeability in the warm environments. Two kinds of sportswear, i.e., 100 % cotton with moisture regain of 6.1 % and air permeability of 143 ml/square centimeter/second (Type A) and 100 % cot- ton with moisture regain of 7.0 % and air permeability of 26.8 ml/square centimeter/sec- ond (Type B) were prepared. Thickness and mass were controlled as equally as possible. Participants wearing either Type A or Type B performed cycle exercise, which intensity corresponded with 60 % of their maximal oxygen uptake, for 60 min under wind of 1.5 m/s or no wind at an ambient temperature of 30 degree C. Rectal temperature during 60 min exercise was significantly inhibited to increase in Type A compared with Type B under windy condition, while there did not exist any differences in rectal temperature in- crease under no windy condition between Type A and Type B. This fact suggests that the fabric with high moisture regain plus high air permeability can inhibit an increase of core temperature during exercise and effective evaporative cooling can be accelerated by combined function of fabric high moisture absorbancy and air permeability under light windy condition. 108 Fibres, textiles and materials for future military protective clothing Richard A. Scott Research & Technology Group, Defence Clothing & Textiles Agency, Flagstaff Rd., Colchester, Essex, CO2 7SS, United Kingdom. Introduction The Defence Clothing & Textiles Agency of the Defence Logistics Organisation is re- sponsible for providing all Clothing and Textile items for the U.K Armed Forces. It has tri-service responsibility, in that it covers all requirements for the Royal Navy, the Army and the Royal Airforce (excluding aircrew clothing, and diving clothing for the Navy) (Scott, 2000). The Research and Technology Group carries out research, development and Quality Assurance for this organisation. A large proportion of the scientific effort involves re- search on novel materials and clothing systems for the future. One of the main research programmes is entitled “CRUSADER 21” – future integrated combat clothing system (Beyts & Scott, 1999). Another (Core) programme provides underpinning research in the form of background information, new techniques, and computer modelling of high-risk human protection scenarios. A third programme, (Inventive Corporate) involves high- risk, long term, innovative research topics. Improvements in future military combat clothing Many of the requirements for future combat and protective clothing are aimed at solving a set of perennial problems. These are: a. Improved Protection - against the natural environment and battlefield threats. b. Maintenance of Thermo-physiological comfort - or survival in extreme conditions. c. Improved compatibility - between and within different components in the clothing assembly. d. Reduction in weight and bulk - especially load carriage systems and ballistic protec- tive clothing. e. Integration of Clothing items - in which the clothing items are considered to be parts of a multi-role system. f. Reduction in life cycle costs - future systems may be more expensive, but may be more effective, durable, may consist of fewer components, and could be recyclable. Research objectives The research programme is split into a range of threat-based objectives. The aim is to provide protection of the individual combatant, whilst maintaining full operational effec- tiveness. The following list approximates to the threat priority order: 109 1. The Natural and Operational Environment: mainly world climatic extremes, but in- cludes sea water, dust, mud, altitude, terrain and vegetation. The operational envi- ronment includes protection from low speed impacts, foreign object damage, petrol, oils and lubricants, noise, abrasion, wear, and static electricity. 2. Ballistic Protection: against fragmenting munitions, such as bombs, mines and shells, bullets, and other projectiles such as flechettes. 3. Surface Heat and Blast: this includes open flames, heat, weapon flash and blast. 4. Detection, Identification and Recognition: involves camouflage, concealment and deception against a wide spectrum of surveillance threats. 5. Biological and Chemical Protection: against toxic chemicals such as mustard gas, nerve agents, bacteria, viruses, and bio-regulators. 6. Nuclear and Radiological Protection: against the effects of nuclear weapons. 7. Directed Energy Weapons: mainly battlefield range-finding lasers. 8. Mobility and Load Carriage: involves the provision of appropriate footwear, and means of carrying personal loads. 9. Integration, Interfaces and Design: deals with all aspects of integrated clothing sys- tems. 10. Human Factors: considers such factors as Thermo-Physiological comfort, military performance factors, anthropometry and fit of clothing. Materials research It is predicted that concentrating on innovative materials can make significant improve- ments in military protective clothing in the future. This has led DCTA to consider the use of “smart/reactive“ materials for protection. Smart / reactive materials These are designed to operate so that the wearer is comfortable and effective during nor- mal duties. When the threat changes, such materials respond and change state in order to oppose the threat, and provide increased protection just when it is required. Examples are included in the following sections of the paper. Materials for environmental protection Thermal Insulation - here the aim is to provide “smart” adjustable insulation in one gar- ment, rather than adding or removing multiple layers. The approaches include Variable Pile Fabrics ( Patent GB.2,234,705B) or Inflatable Mesh Systems ( Patent GB.2,242,609B).(Forshaw, 1999) These materials can be erected or inflated to increase thermal insulation when ambient temperatures and activity levels are low. Conversely, they can be collapsed to offer minimal thermal insulation when ambient temperatures are high, or activity is high. A variety of materials are being used in the pile fabrics, includ- ing monofilament polyester, core yarns and microfibres to give optimum properties. The inflatable meshes have utilised rubber, PVC and polyurethane films. New polyester filling fibres combining the properties of high insulation, low bulk, and high recovery from compression are now available in many forms and are being evaluated. 110 High Performance Wicking Materials - are being investigated for next-to-skin clothing. A range of purpose designed polyester fibres such as Coolmax®, Thermostat®, and ny- lon microfibres are being evaluated for their ability to transfer liquid moisture from the skin and dry quickly. They are being examined in a range of knitted fabric constructions. The incorporation of controlled release biocidal treatments to improve hygiene during longer wear periods is under investigation. Waterproof / Water Vapour Permeable (WVP) Materials - are in wide usage in U.K military foul weather clothing. Development continues to produce the optimum balance of properties for particular uses. Some require high WVP and waterproofness for in- fantry, Marines and special forces, whereas Navy personnel require high waterproofness with lower WVP for their particular tasks. Cost is a particular current driver for rainwear. Research is also examining the concept of fabrics with “smart pores”, which are designed to open when the internal humidity is high (i.e. sweating conditions), and close when humidity is low. Multi-Role Textiles - are designed to minimise the number of separate clothing items to be carried and maintained. We can currently demonstrate a laminate which combines 8 levels of functionality comprising water proofness, vapour permeability, wind proofness, flame-retardance, non-melting, and camouflaged in the visible, near infra-red and acous- tic parts of the spectrum. Phase Change Materials - are designed to buffer temperature changes in clothing sys- tems by absorbing or releasing latent heat. These are based upon the use of micro- encapsulated waxes in fibres or fabrics. The buffering effects are proving to be rather limited at this stage of development, but promise future improvements (Lennox-Kerr, 1998). Improved Repellent Finishes - involving new chemistry to covalently bind fluorocarbon chains to textile fibres have achieved some limited success in providing permanent high levels of oil and water repellency for the life of the material. Materials for ballistic protection Light weight flexible protection for the individual requires the use of specialist high modulus fibres for body armour. They are also used for light weight composites for hel- mets. Such fibres combine very high tensile strength with very low elongation at break. They are effective against bomb fragments and low velocity bullets. Fibres available include ballistic nylon, para aramids (aromatic polyamides) such as Kevlar® (Du Pont), and Twaron® (Enka). Kevlar in particular is available in a variety of continuous filament decitexes and finishes. More recently these have been joined by a new fibre Zylon® (Toyobo). This is a PBO fibre (paraphenylene polybenzobis oxazole). Early evaluations show some promise, although the price is currently prohibitive (Payne, 2000). A range of Ultra High Modulus Poly Ethylene (UHMPE) fibres have been developed in recent years. They are typically Gel Spun Poly-Ethylene (GSPE) fibres with trade- names such as Dyneema® (DSM), and Spectra ® (Allied Signal). A needle punched felt fabric Fraglight® having chopped, randomly laid GSPE fibres, offers a very low density (0.97 g/ml) ballistic pack, albeit with a degree of bulk. The choice of ballistic fibre and fabric depends on ballistic performance, weight, flexibility and cost. 111 Protection against high velocity bullets and projectiles is provided additionally by ce- ramic plates, often based upon alumina, or boron carbide for more severe ballistic threats. The plates are limited in size - because of their weight - and are designed to protect vital organs such as the heart. Fibres for flame and heat protection Flames, heat and flash threats in operational conditions are deliberate threats to military personnel. Significant R&D efforts are expended in order to provide adequate protection at reasonable cost, and to maintain thermo-physiological comfort. A limited range of Flame Retardant (FR) fibres and finishes are actually used in UK military clothing, although the number of options available is increasing. The most widely used FR fibre is Proban® treated cotton – hydroxymethyl phospho- nium oxide or chloride. Fabrics currently contain up to 30 % of polyester in a blend to improve durability, dimensional stability, and smart appearance. The major user is the Royal Navy, whose working and action clothing on board ship is made from this blend. Certain coveralls used in workshops and dockyards are also supplied in FR treated cotton. When Naval crews are required to fight fires they currently don a heavy insulative coverall in Zirpro® treated wool. There are plans to replace this in the future. The UK Nuclear, Biological and Chemical (NBC) protective over suit currently has an outer fabric made from a (FR) modacrylic warp and nylon weft. This gives a degree of flash and flame resistance to the wearer. A wide range of meta aramid fibres now exist. Examples such as Nomex® (Du Pont), Conex® (Teijin), and Kermel® (Rhone-Poulenc) are becoming more widely used, and is largely replacing other current products, such as wool, especially in the fire- fighting sector of the business. Nomex® is available in a range of blends with Kevlar® para-aramid, some of which contain small percentages of a P140 antistatic fibre for specific applications (Mountford, 2000). MOD uses some of these variants in Tank Crew clothing, Explosive Ordnance Disposal (EOD) clothing, Aircrew and Special Forces clothing, and is currently worn by submarine crews. The newer variants are being evaluated for uses in future multi-role textiles (para.4.b.4), aircrew clothing, FR underwear, Firefighters ,and specialist protec- tive clothing systems. MOD civilian fire brigades have also recently trialled turn- out suits in PBI “Gold” fibre (Poly Benz-Imidazole fibre) Reactive materials for flame and heat protection One of the main problems with protection is providing sufficient insulation to prevent skin burns whilst maintaining wearer thermal comfort. Multiple layers of clothing and air gaps will provide the required protection. Reactive materials are designed to provide minimal insulation during normal wear, but maximum insulation when the heat threat impinges. The DCTA has examined several novel approaches: Intumescent treatments - are formulations containing a carbonific source, a catalyst, and a spumificant, or gas blowing agent. These would normally be in the form of thin, low in- sulation coatings on a lining fabric. When activated by excessive heat or flames the for- mulation swells instantly to form an inert insulative char, protecting the wearer. Such in- 112 tumescent treatments are currently being optimised for use in clothing systems (Scott, 1999). Shaped Memory Alloys - are metals which can assume different shapes when the tem- perature is changed. In this case we wish to establish a large air gap during the threat phase. This can be achieved by using a coil spring between two fabric layers which adopts a flat conformation during normal operations, but which rapidly forms an ex- tended helical shape when the heat threat arrives. These can increase protection times against 2nd degree skin burns by a factor of between 50 and 85 %, depending on the heat flux and clothing system (Congalton, 1999). Photochromic Dyes - are colorants which change colour with light intensity. To provide flash protection for the eyes we would ideally envisage an optical eye protector which is perfectly clear under normal conditions, but which instantaneously attains an optical den- sity of at least 3 when nuclear weapon flash occurs. Such systems have shown some promise for this application. Thermochromic Dyes - are textile colorants which change colour with temperature. In order to provide radiant flash protection, we ideally want to use dyes, which normally exhibit the correct visual camouflage print, but which bleach white instantaneously when the flash arrives. This reflects a proportion of the incident radiant energy. Such dyes will react in nanoseconds, although it is currently difficult to provide the effects in all 4 colours of the UK camouflage print. Camouflage, concealment and deception Full spectrum camouflage for combat clothing encompasses the ultra violet, visible, (0.4 to 0.8 micron) near infra red (0.8 to 1.2 micron), and thermal far infra-red (2.0 to 5.6 and 8 to 14 micron) We also desire to defeat surveillance systems operating in the radar and acoustic bands. Adaptive Camouflage - Camouflage patterns on UK clothing are designed to mimic woodland and desert backgrounds. The natural background changes markedly with the seasons – there being a preponderance of green hues in summer, many of which disap- pear in winter. To cope with these changes research is examining the use of thermochro- mic textile dyes that will change colour when triggered by a significant seasonal tem- perature change. The desired effect must mimic the background vegetation in both the visual and near infra-red parts of the spectrum, as surveillance devices such as Image Intensifiers and low light television can detect contrasts in the near IR. Thermal Camouflage - Reduction of the human thermal signature is a complex problem. It can be achieved by providing thermally insulating clothing, but this can cause prob- lems with heat stress and discomfort, since the face and hands must also be screened. An additional approach is to lower the emissivity of the outer materials by applying metallised coatings which also provide the correct visual and near IR properties. Re- search work is producing interesting solutions to these multi spectral problems. Selectively permeable materials for chemical protection Nuclear, Biological and Chemical protective hand and foot wear is currently provided by impermeable rubber gloves and overboots. These can cause skin maceration within a few 113 hours wear, due to entrapped sweat. A research programme is examining the possibility of providing rubber with selective permeability, which allows perspiration vapour to es- cape whilst preventing the ingress of toxic chemical warfare agents. Bibliography Beyts NGB, Scott RA (1999) Crusader 21 Research Programme. International Soldier Systems Conference, Orlando, USA, Sept. 1999. Congalton D (1999) Thermally active Clothing systems using Shape Memory Alloys and Polymers. Intelligent Textiles Conference. Heriot-Watt University, Sept 1999 Forshaw PL (1999) Variable Insulation in Military Clothing. Ibid., Sept 1999. Lennox-Kerr P (1998) Comfort in Clothing through Thermal Control. Textile Month, pp 8-10, November 1998. Mountford R (2000) Individual Fighting Systems Working with Du Pont Nomex®. Various presentations, Jan. 2000. Payne M (2000) Toyobo puts its faith in Speciality and Functional Textiles. Technical Textiles International, Pp 11-13, April 2000. Scott RA (1999) Intumescent Treatments for Flame & Heat Protection. Intelligent Textiles Conference. Heriot-Watt University, Galashiels, Sept 1999. Scott RA (2000) Textiles in Defence. Chapter 16 in Technical Textiles, Eds. Horrocks R and Anand. SC, Woodhead Press. 114 Woven technical textiles for ballistic protection Carmen Mihai1, Eftalea Carpus1, Emilia Visileanu1, Doina Toma1, Nicolae Scarlat2, Mircea Milici2 1 The National Research-Development Institute for Textile and Leather, Bucharest, Romania 2 Scientific Research Center for N.B.C. Protection Technics, Bucharest, Romania Introduction All scientific discoveries led the military research results to high modernization limits. Following the path of all scientific achievements a major issue is soldiers total protection, improving survival conditions. In this respect, one of the textile industry performances is the military protective equipment meant to assure humans normal life conditions during extreme fighting situations. At research boundaries there are some specific directions such as: strength to ballistic bullets impact and protection in nuclear biological or chemical contaminated field. Ac- cording to this background the textile research industry takes into account some major requirements:  physical requirements: durability to prolonged exposure to inclement weather and heavy wear, good tensile and tear strength and abrasion resistance;  environmental requirements: water repellency, air permeability;  physiological requirements: low weight, easy to wear, minimum heat stress, air moisture and vapor permeability and comfort;  battlefield requirements: ballistic protection, flame resistance, resistance to chemical and biological agents, resistance to long range thermal effects of nuclear weapons, good camouflage properties and low noise generation. Considerations regarding the used yarns for ballistic protection During the projectile impact with the textile support, specific and complex phenomena take place, phenomena which were not completely explained until now. The ballistic and shell fragment protection is assured by using the woven fabrics with content of yarns characterized through high tenacity, low breaking elongation and a very good capacity of impact energy absorption. Nowadays, on the market of the ballistic protection equipment are know two types of fibers:  para–aramide fibers : Kevlar (Du Pont) Twaron (AKZO) Technora (Teijin)  high tenacity polyethylene fibers : Dyneema (DSM) Spectra (Allied Signal) In Table 1 are shown the comparative values of the characteristics determined for dif- ferent fibers used in manufacturing the ballistic protection equipment. 115 Table 1. Comparative characteristics for the fibers used in ballistic protection equipment. Fiber type Density g/cm3 Tenacity N/Tex Specific modulus N/Tex Elongation at break % Sonic velocity m/s Polyamide HT 1.14 0.80 5 20.0 2 200 Aramid - ballistic 1.44 2.35 52 3.6 8 200 HP polyethylene 0.97 3.30 101 3.7 10 000 In ballistic protection, the degree and the mechanism of impact energy absorption are very important. For example, the specific modulus and the density determine the velocity of the shock wave propagation into fiber. Moreover, the specific modulus and the sonic velocity in fiber determine the ballistic potential of the textile complex, this becoming more performant as the velocity of sound propagation into fiber is higher. The capacity of impact energy absorption is determined by the tenacity and elongation of the yarn. It is obvious the fact that this is direct proportional with the two size multi- plied. It should be emphasized the fact that woven fabrics with content of para–aramidic or polyethylene yarns are preferred those of polyamide HT, although the capacity of energy absorption is almost similar, because a high breaking elongation determines the increasing of the backface deformation and implicit of the traumatic shock (Figure 1). 4 3 2 1 0 0 5 10 15 20 [%] Elongation Tenacity [N/ Tex] Tenacity [N/ den] 40 30 20 10 Aramid HPPE Polyester Nylon Figure 1. Tenacity vs. elongation for fibers used in ballistic protection. The studies and researches made until nowadays have shown that approx. 80 % of the dead and the wounded in battle are due to the shell fragments and not to the bullets. For this reason, the military ballistic vests are designed and are tested in order to reduce these looses. For establishing the military ballistic vest performances it is necessary to deter- mine the parameter ballistic limit velocity- V50. “The ballistic limit velocity for a material or armour is defined as that velocity for which the probability of penetration of the cho- sen projectiles is exactly 0.5“ (STANAG 2920). In Figure 2 there are presented the V50 values for Kevlar complexes tested according to STANAG 2920 with 9 mm FMJ ammunition. 116 3.0 3.5 4.0 4.5 5.0 5.5 6.0 [kg / sqm] KEVLAR pack areal density 550 500 450 400 350 300 V50 [m/s], 9 mm FMJ KEVLAR Comfort KEVLAR Ht KEVLAR 29 Figure 2. V50 vs. pack areal density. Kevlar Comfort has a 10 % higher ballistic resistance V50 compared with Kevlar HT and also, has 35 % higher ballistic resistance V50 than Kevlar 29. The Figure 3 shows V50 values for different fabric panels and contrasts the Twaron standard type and Twaron CT (high tenacity) with Twaron CT Microfilament. In figure 4 there are shown, in comparison, the V50 value for HPPE and para–aramidic tested according to STANAG with 1.1 g FSP ammunition. 3.0 3.5 4.0 4.5 5.0 5.5 6.0 Weight [kg / sqm] 550 500 450 400 350 300 V50 [m/s], 9 mm FMJ Kinetic energy [Nm] TWARON CT Microfilament TWARON CT TWARON standard 1000 810 640 490 360 Figure 3. V50 and kinetic energy vs. weight. The test has been done according to STANAG 2920 with 9 mm FMJ. 117 3.0 3.5 4.0 4.5 5.0 5.5 6.0 [kg / sqm] Areal density 550 500 450 400 350 300 V50 [m/s], 9 mm FMJ HPPE para - aramid Figure 4. V50 values for HPPE and para–aramid. Own technological experiments for producing ballistic woven fabrics Taking into account the conditions imposed to ballistic structure, two woven fabrics have been made at The National Research – Development Institute of Textile and Leather, woven fabrics composed of 100 % Kevlar yarn 930 dtex and 220 dtex respectively, whose main characteristics are presented in Table 2. Table 2. Characteristics of Kevlar yarns used for ballistic woven fabrics. SymbolCharacteristics VB – 60 – K129 K - 160 - PG Filament Type 964 C 964 C Yarn count, warp / weft dtex 930 / 930 220 / 220 Weight, g/sqm 216 88 Weave plain plain Tensile strength, warp / weft daN /5 cm 900 / 900 310 / 310 I have to mention that the woven structure K – 160 PG has been designed according to the requirements imposed in using at the brake parachutes for supersonic airplanes type MIG 29. The reason of using woven fabric made of 220 dtex yarns, in order to realise armours, had as result the reducing of the weight with about 14 %, assuring at the same time special comfort in normal using conditions. The two woven fabrics have been tested at Scientific Research Center for N.B.C. Protection Technics from Ministery of National Defence. The ballistic Test Method con- sisted in sending a projectile to a target (textile support placed in a plastiline block) at 5 meters distance. In these circumstances has been evaluated the diameter of the trauma produced in the plastiline block by deforming the textile suport during the impact. Ballistic protection for the two supports have been tested according to NIJ 0101.03/87, for the following types of ammunition: -  357 Magnum JSP (Jacketed Soft Point); - 9 X 19 mm FMJ (Full Metal Jacketed). The results are presented in the Tables 3 and 4. 118 Table 3. Ballistic performances for VB – 60 – K129 woven fabric. Ammunition No. of layers Weight kg/sqm Velocity m/s Strength Backface deformation mm 9X19 FMJ Yes 35 9X19 FMJ Yes 18 9X19 FMJ Yes 27 9X19 FMJ 360 ± 15 Yes 24  357 Magnum JSP Yes 25  357 Magnum JSP 18 3.88 400 -425 Yes 30  357 Magnum JSP Yes 30  357 Magnum JSP 400 - 425 Yes 30 9X19 FMJ Yes 27 9X19 FMJ Yes 29 9X19 FMJ Yes 30 9X19 FMJ Yes 30 9X19 FMJ 17 3.67 Yes 31 9X19 FMJ 16 3.46 360 ± 15 No - Table 4. Ballistic performances for K – 160 – PG woven fabric. Ammunition No. of layers Weight kg/sqm Velocity m/s Strength Backface deformation mm 9X19 FMJ Yes 30 9X19 FMJ 360 ± 15 Yes 44  357 Magnum JSP Yes 50  357 Magnum JSP 34 2.99 Yes 47  357 Magnum JSP 32 2.82 400 -425 Yes 50 9X19 FMJ 28 2.46 360 ± 15 No - It can be seen that VB – 60 – K129 woven fabric with 17 layers structure can hold the II Level of ballistic protection, and K – 160 – PG woven fabric with 34 layers can hold the II Level of ballistic protection with backface deformation bigger than 44 mm. For the VB – 60 – K129 woven fabric has been determined the V50 ballistic limit ve- locity tested according to STANAG with 1.1 g FSP ammunition which is of 570 m/s. The determined V50 indicates also a good ballistic resistance to shell fragments. Conclusions The construction of the military protective equipment meant to assure humans normal life conditions during extreme fighting situations is one of the textile industry performances. Taking into account the conditions imposed to ballistic structure, two woven fabrics have been made at The National Research – Development Institute of Textile and Leather, woven fabrics made of 100 % Kevlar yarn 930 dtex and 220 dtex. The two woven fabrics have been tested at Scientific Research Center for N.B.C. Protection Technics from Ministery of National Defence:  The 930 dtex woven fabric symbolised VB – 60 – K129 with 17 layers structure can hold the II Level of ballistic protection.  The 220 dtex woven fabric symbolised K – 160 – PG with 34 layers can hold the II Level of ballistic protection with backface deformation bigger than 44 mm.  The determined V50 for VB – 60 – K129 woven fabric indicates also a good ballistic resistance to shell fragments. 119 Thermal protective textiles: Correlation between FR properties and static propensity Jose A. Gonzalez, Martin W. King, Amit Dhir Department of Clothing and Textiles, University of Manitoba, Winnipeg, Manitoba, R3T 2N2, Canada Introduction All clothing is protective to some extent. It is the degree of protection that is important, particularly for those working in hazardous occupations, such as fire fighters, foundry and utility workers. Concern over the health and safety of workers in various industries has stimulated research and development in the area of personal protective clothing. Some of this research was focused on thermal protective clothing, which is designed to extend people’s physical and physiological limitations in response to fire and heat expo- sure. The performance of thermal protective clothing depends on its ability to insulate and to maintain structural integrity when exposed to high heat or flash fires (Raheel et al., 1994). However in certain circumstances, such protective clothing may add to, rather than reduce, the hazard. For example, such clothing may lead to electrostatic discharges (sparks) at low humidities, which can cause explosions in the presence of flammable gases, and lead to the potential for injury and the loss of human life. In Canada and in some other parts of the world, many industrial workers are required to carry out their duties in very cold and therefore very dry conditions. It has been shown that when worn under such low humidity conditions, clothing made from thermal protec- tive materials, such as aramid or flame-retardant cotton, may generate enough energy during a discharge to ignite a fuel vapour-air mixture (Rizvi et al., 1998). Ignition of flammable materials due to a human spark is a significant concern in the oil, military, chemical, electronic, and other industries, leading to a considerable amount of research on the charge generation characteristics of various types of textiles used in clothing. Many cases have been documented (Scott, 1981; Crow, 1991) where charges generated on an object reach the level at which the resistance of the air-gap between the object and a conductor at a lower potential breaks down, producing a spark. The most serious effect of an electrostatic discharge (ESD) is its ability to ignite flammable gases, vapours, or powders, resulting in fires and explosions and the possible loss of human life. For example, the average individual walking across a non-conductive floor or sliding off a car seat can generate discharge potentials up to 15 kV at low humidity (Matisoff, 1986; Rizvi et al., 1995). Wilson (1977/78) showed that the minimum ignition energy (MIE) of coal gas and air is 0.03 mJ. The MIE required to ignite methane and air in a closed cham- ber by a spark between a finger and an earthed electrode has been measured at 1.1 mJ (Tolson, 1980), and as low as 0.5 mJ (Crugnola & Robinson, 1959). The purpose of this research was to study both the flame resistance (FR) properties and the electrostatic propensity of different thermal protective textile materials, and determine if the electrostatic propensity is influenced by the degree of FR protection. Of a major concern was to determine if an increase in thermal protection could cause an increase in the electrostatic propensity of thermal protective textile materials. 120 Methods FR properties were evaluated in terms of char length (mm) and thermal protective per- formance (kW · s/m2) following CAN/CGSB 4.2 Method No 27.1, Flame Resistance – Vertical Burning Test, and ASTM Method D 4108 –Test Method for Thermal Protective Performance of Materials on Clothing by Open-Flame Method, respectively. The electrostatic propensity of the materials was evaluated in terms of surface peak potential (V) and charge decay rate to 10 % of the initial charge (V/s) following a modi- fied method of the UA ESD Test System, which has been described by Gonzalez et al. (In press). The modification to the UA ESD Test System involves the use of a movable specimen holder. At the end of the charging process, the specimen holder is moved under a sensor head connected to a static voltmeter1, which is then connected to a flatbed re- corder2 and a digital oscilloscope3. The oscilloscope is set in roll mode and used in com- bination with WaveStar version 2.0 software for data acquisition. Both the voltmeter and recorder are activated immediately after placing specimen under the sensor head to measure and record peak potential and charge decay from the fabric surface. Thermal stable and FR finished fabrics were tested, including FR cotton, meta-aramid, meta-aramid/carbon, meta-aramid/FR viscose, para-aramid, and para-aramid/PBI (Table 1). Specimens for flammability tests were conditioned according to their respective stan- dard method; those for electrostatic tests were prepared following CAN/CGSB-4.2 Method 15, non-fibrous materials on textiles, and conditioned at 0 % relative humidity (RH) by drying specimens out at 105 °C for an hour, cooling them off inside a dessicator, and then storing the specimens in sealed polyethylene bags until testing. Table 1. Fabrics used in the experiment. Code Fibre Content Weave Fabric Count Fabric Mass (Yarns/cm) (g/m2) 01 FR cotton Twill 35 x 19 320 02 Meta-aramid Plain 32 x 30 203 03 Meta-aramid/carbon Twill 30 x 24 251 04 Meta-aramid/FR viscose Twill 32 x 21 264 05 Para-aramid Plain 11 x 11 208 06 Para-aramid/PBI Plain 32 x 18 182 Using commercially available, SPSS version 10.0 software, the following statistical analyses were performed, with the level of significance for testing hypothesis set at p<0.05. 1) One-way ANOVA (analysis of variance), to test the null hypothesis (H01) that there was no significant difference in char length, thermal protective performance (TPP), surface peak potential and decay rate among different thermal protective fabrics. 2) Pearson’s correlation coefficient, to test the null hypothesis (H02) that there was no significant correlation among char length, TPP, surface peak potential, and charge decay rate. 1 Electro-Tech System, Inc. model 105 2 Cole-Parmer flatbed recorder model 08376 with a millivolt/volt module model 8376-65 3 Tektronix model TDS 340A 121 Results and discussion The null hypothesis H01 was rejected; One-way ANOVA found a significant main effect of fabric on all measured variables, indicating that there were significant differences. Duncan’s multiple range test results for each dependent variable (Table 2) imply that these parameters measured from different fabrics differ significantly (p <.05) from each other, although there are some homogeneous subsets. Table 2. Analysis of variance: Char length, TPP, peak potential and charge decay rate. Fabric TPP Char Length Peak Potential Decay Rate Mean S. D. Mean S. D. Mean S. D. Mean S. D. FR cotton 8.43a 0.15 8.20e 1.23 2304c,d 291.1 0.42a 0.11 Para-aramid 10.90b 0.18 0.90a 0.21 2942e 268.9 0.64a 0.20 Para-aramid/PBI 11.21c 0.29 3.35b 0.88 985b 186.7 5.21b 3.67 Meta-aramid 11.46c 0.37 5.44c 1.84 2606d,e 161.9 0.21a 0.11 Aramid/FR viscose 11.96d 0.29 6.26c,d 1.05 1921c 140.6 18.65c 5.18 Aramid/carbon 14.34e 0.53 6.70d 0.79 141a 11.5 0.09a 0.02 a,b,etc In each column, means with the same letter superscript indicate homogeneous subsets (highest and lowest means are no significantly different) when subjected to Duncan’s multiple range test (p <.05). To test the null hypothesis H02, Pearson’s correlation analysis was performed between FR and electrostatic parameters (Table 3). Correlations between peak potential and either char length or TPP were significant; thus, the null hypothesis was rejected for these three parameters. Figure 1 graphically shows the relationship between thermal protective per- formance and the surface peak potential. Table 3. Correlations (R) between char length and TPP, with peak potential and decay rate. Parameter Peak Potential Decay Rate Char length -.300* .107 TPP -.613* .115 *Correlation is significant at the .05 level (2-tailed) Conclusions The main purpose of this study was to determine whether the electrostatic propensity was affected by the degree of FR protection of thermal protective fabrics. A vertical burning test was used to assess the resistance of a series of fabrics to burning, and the TPP test was used to rate the same textile materials for thermal resistance and insulation. In addi- tion, the amount of surface static charge generated by tribo-electrification and its subse- quent dissipation was evaluated by ESD testing. Different properties and characteristics of the fibres such as polymer chemical composition and polymer structure are at work during flammability tests. Surface charge generation depends to a large extent on the na- ture of the material rubbed and other external factors; while charge dissipation depends on the geometric, dielectric and resistivity conditions of the material and surrounding environment. Results from this investigation suggest that surface peak potential increases as the thermal protection performance decreases, although these changes occur at different rates. Since both parameters are highly dependent on the polymer structure, it implies that changes in polymer structure for improving FR protection may cause variations in the material’s propensity to be tribo-charged. 122 0.00 2.00 4.00 6.00 8.00 10.00 12.00 14.00 16.00 aramid/carbon aramid/PBI aramid/FR viscose FR cotton meta-aramid para-aramid Fabric kW·s/m² 0.00 500.00 1000.00 1500.00 2000.00 2500.00 3000.00 3500.00 Potential (V) TPP Peak Potential Figure 1. Correlation between thermal protective performance and surface peak potential. Further work is therefore needed to determine how much influence the polymer’s structure has on both thermal protective performance and the generation of static charge. Acknowledgements The authors gratefully acknowledge for the invaluable assistance given by Dr. Elizabeth Crown, Ms. Helena Perkins and Ms. Elaine Bitner of the Department of Human Ecology, University of Alberta, Edmonton, Alberta. Thanks are also due to Ms. Jennifer Duncan for her effective work during long hours of testing at the laboratory. References Crow RM (1991). Static electricity. A literature review (DREO technical note No. 91-28). Canada: Defense Research Establishment Ottawa. Crugnola AM & Robinson HM (1955). Measuring and predicting the generation of static electricity in military clothing (Report No. 110). USA: Quartermaster Research and Engineering Centre. Gonzalez JA, Rizvi SAH, Crown EM & Smy PR (in press). A laboratory protocol to assess the electrostatic propensity of protective clothing systems. Journal of The Textile Institute. Accepted for publication. Matisoff BS (1986). Handbook of electrostatic discharge controls. New York: Van Norstrand Reinhold. Raheel M, Perenich TA & Kim CJ (1994). Heat- and fire-resistant fibres for protective clothing. In Raheel M (Eds), Protective Clothing Systems and Materials. Pp 197-223, USA: Marcel Dekker. Rizvi SAH, Crown EM, Osei-Ntiri K, Smy PR & Gonzalez JA (1995). Electrostatic characteristics of thermal-protective garments at low humidity. Journal of The Textile Institute, 86(4), 549-558. Rizvi SAH, Crown EM, Gonzalez JA & Smy PR (1998). Electrostatic characteristics of thermal-protective garment systems at various low humidities. Journal of The Textile Institute, 89(4), 703-710. Scott RA (1981). Static electricity in clothing and textiles. UK: Commonwealth Defense Conference. Tolson P (1980). The stored energy needed to ignite methane by discharges from a charged person. Journal of Electrostatics, 8, 289-293. Wilson N (1977/78). The risk of fire or explosion due to static charges on textile clothing. Journal of Electrostatics, 4, 67-84. 123 Testing and evaluation of electrostatic behaviour of electric inhomogeneous textiles with core-conductive fibers Jürgen Haase, Christian Vogel Sächsisches Textilforschungsinstitut e.V. (STFI), 09125 Chemnitz, Germany Introduction The energy of electrostatic discharges from charged textile materials can exceed the minimum ignition energy of flammable gas- air atmospheres. To avoid the risk of explo- sion for persons working in such environments, besides other measurements the textile materials for this aims incorporate conductive components (Holme et al., 1998; IEC 61340-1, 2000). The test method (EN 1149-1, 1995) for the evaluation of the surface resistance is often not suitable, especially in the case of modern textiles with highly conductive fibers in core –matrix –structure. Materials of this kind are able to dissipate static electricity by several physical mechanisms (Löbel, 1995, 1987; Holdstock, 1996):  Dissipation by homogeneous conductivity of material components: time constant of the decay is dependent on the surface resistivity  Dissipation owing to polarisation effects in polymer material: initial decay effect with time constant in the range 10 - 100 ms;  Induction field and shielding effect of this caused by the presence of highly conduc- tive fibres (metal or carbon) in the fabric: time constant for generation of shielding ef- fects after induction charging is practically zero (range s … ns) because of the high conductivity of the metal or carbon components;  Neutralisation of charges by corona discharging from the fibres within the fabric. All of these mechanisms require a specific test method. Methods The STFI developed so called induction decay-method (Löbel, 1995, Löbel & Haase, 1995). This method allows all of the above processes to be observed if they are present in a fabric specimen. The principle of the test method is shown in Figure 1 (see also Holme, 1998; EU-Project No. SMT4-CT96-2079, 2000). It is based on placing a circular field electrode (diameter 70 mm) directly below an earthed specimen but not in contact with it (distance 4 mm). This electrode delivers a momentary(within 30 s) 1,2 kV-step from a piezo-electric generator. The resulting electric field induces a quickly initial charge of opposite polarity on the conductive specimen. The test specimen is clamped in a circular, earthed holder (diameter 300 mm). The field screening and decay-effects of the fabric is monitored by circular field- measuring probe (diameter 30 mm ) above the sample. The field strength is measured via a two-channel charge amplifier connected to a two-channel high speed recorder (ASTRO-Med Inc. Type DASH IV). 124 Figure 1. Schematic outline of induction method. Key: A Charge amplifier B Recording device C Field-measuring probe D Guard ring E Specimen clamping ring F Test specimen G Support ring H Field-electrode I Voltage generator Results and discussion The typical charging and charge decay profiles of the different types of material are summarised in Figure 2. Insulator hom ogeneous m aterial core -like m aterial m etallic -like m aterial E charging potential: 1 ,2 kV step duration: 30 Ps t EP ER ER E100 Figure 2. Charge decay profiles of different materials. Key: E field signal on the field probe as a function of time E100 initial field signal on the field probe E50 field signal on the field probe after 50 % decay t time (it be used two time scales: 0,00005s/mm recorder paper; 0,04 s / mm) t50 half decay time, i.e. time to reach E50 R resistance C capacitance of measuring device (14 pF) EP field signal on the field probe at maximum of “core peak” ER field signal on the field probe at minimum of core peak, i.e. where the residual field of the high conductivity component transitions to the second, long decay time component Ref C D EE HRef A B F GG I 125 From the measured decay curves E = E(t) several parameters - besides of the charge decay time -can be evaluated depending of type of the specimen conductivity: Resistance R for homogeneous materials (long time decay): 50 100 50 ln E EC t R  Equivalent resistance Ri for inhomogeneous materials with core conductivity (short time decay):       RP R i EE EEC tR 100ln Electrostatic shielding value S: S E E R  1 100 S depends substantially on the distribution and grid density of the conductive fibres within a fabric. When the grid density is high, i.e. the spacing between conductive fibres is small, then ER/ E0 0and the value S1. Table 1 shows typical test results for characteristic specimens (details see EU-Project (2000)). Table 1. Typical test results for characteristic specimens. t 50 Ssample typ fibre material high con- ductive material EN1149-1 spec. surface resistance (Ohm/sq) Ri short time decay (Ohm) R long time decay (Ohm/sq) short time decay (s) long time decay (s) PPC17 hom. CO FR - 2.1E+12 2.2E+12 - 0.6 - PPC18 metal Aramid 1 % steel f. 1.1E+05 <<1.0E+06 >1.0E+13 <0.01 >50 0.61 PPC19 hom. Aramid/ Visk FR - 5.6E+12 3.2E+12 - 0.84 - PPC20 hom. Aramid - >1.0E+13 >1.0E+13 - >50 - PPC23 core Aramid 2 % C-core >1.0E+13 1.4E+06 >1.0E+13 <0.01 >50 0.86 PPC25 core Aramid/ Visc FR 3 % C-core 1.5E+12 1.1E+06 1.0E+12 <0.01 0.5 0.91 PPC41 core PES 5x5 grid trilobal core fiber >1.0E+13 1.2E+06 >1.0E+13 <0.01 >10 0.76 PPC42 core PES 20 mm stripe trilobal core fiber >1.0E+13 1.0E+06 >1.0E+13 (<0.01) >10 0.21 PPC43 core PES 10x10grid trilobal core fiber >1.0E+13 1.0E+06 >1.0E+13 <0.01 >10 0.68 PPC44 core PES 10 mm stripe trilobal core fiber >1.0E+13 1.4E+06 >1.0E+13 <0.01 >10 0.6 The specimens with core conductive fibers have a equivalent resistance in the range of 106 Ohm and decay -time <0.01 s. The antistatic quality is good. 126 Conclusion For testing antistatic behaviour of electric inhomogeneous textiles with core – high con- ductive fibers, resistance measurements are unhelpful. Testing must be based on charge/ discharge principles. Frictional charging is the most realistic but is difficult to reproduce and does not allow the short-term effects to evaluate. Corona charging likewise does not allow the short-term effects to evaluate. The inductive method generally very good dis- criminates between samples of all typs of conductivity. Short time effects can be evalu- ated. The method is involved in preparation of the European Standard EN 1149-3. References EN 1149-1:1995 (1995) Electrostatic properties, part 1 surface resistance, test methods and requirements [European Standard]. EU-Project SMT4-CT96-2079 (2000) The Evaluation of the Electrostatic Safety of Personal Protective Clothing for Use in Flammable Atmospheres. Final Report (Summary) 2000-01-04 Holdstock P (1996) Evaluating Personal Protective Clothing constructed from electrically inhomogeneous Textiles. Conference “Static and Textiles“, Institute of Physics London, Proceedings Volume 1 Holme I, McIntyre JE & Shen ZJ (1998) Electrostatic charging of textiles. Textile Progress, 28(1), 1-85 IEC 61340-1/ TC 101 (2000) Guide to the principles of electrostatic phenomena and their measurement. Project- No. 101/61340-1/TR/Ed. 1.0 2000-03-03 Löbel W (1987) Antistatic mechanism of internally modified synthetics and quality requirements for clothing textiles. Inst. Phys. Conf. Ser. No. 85, Section 2, Electrostatics ´87 Löbel W.(1995) Zur Wirkung permanenter antistatischer Applikkationen bei Textilien und deren komplexe Prüfung. Melliand Textilberichte 10, 916-919 Löbel W.& Haase J. (1995) Elektrostatische Sicherheit von Schutzkleidung und ihre Prüfung. Internat. Text. Bull. Vliesstoffe-Techn. Textilien 3, 6-12 127 Features of electric arc accidents in Finland 1996-1999 Sanna Mustonen, Helena Mäkinen Finnish Institute of Occupational Health, Vantaa, Finland Introduction An electric arc is an electric current, which passes from one electrode to another through hot ionised gas. The temperature of an electric arc burning in the air can be 6000  – 20000 C. About half of the energy of an electric arc is consumed by heating the air, the equipment, walls and other structures, as well as ionising the air. Depending on the mate- rial of the electrode, 5-20 % of the energy of an electric arc is used for melting and va- porising the electrodes. The rest of the energy mainly evaporates into the environment as radiant heat (Martikainen, 1978). The amount of energy that is released in an electric arc affects essentially the extent of the damages. Electric arc energy less than 0.5 MJ can be considered harmless, while energy greater than 0.5 MJ is dangerous, and energy greater than 4 MJ causes a great risk of serious bodily injury. The magnitude of electric arc energy is characteristically 0.4-5 MJ in the low voltage range, and 3-120 MJ on high voltage. Burn injuries from an electric arc are mainly caused by the radiant heat and splashes of molten metal. Some of the gases that are formed as combustion products of the electric arc are toxic. The fumes of the vaporised electrode materials are the most dangerous ones of these. The phenome- non includes a loud noise caused by rapid expansion of the air, and a brilliant flash of light, which can cause a temporary eye injury. (Askaner, 1972; Martikainen, 1978; Meyer, 1960; Pälli, 1988) Nominal voltages below 1000 V are generally called low voltage. Nominal voltages between 1 and 100 kV are referred to as high voltage. In high voltage switchgear, the voltage in an electric arc short-circuit is relatively small compared to the supply voltage. In the low voltage range, electric arc voltage is of the same magnitude as the nominal voltage. (Martikainen, 1978) In Finland, since 1986, clothing used in work involving a hazard of electric arc, has had to be made of a flame-retardant material. After the European Union directive of per- sonal protective equipment came into in force 1995, all protective clothing used in electrical work must be CE labelled. CENELEC CLC/TC78 is preparing a standard test method for materials and garments used in work entailing risk from exposure to an electric arc. (Centre of electrical inspection, 1986; CLC/TC 78, 2000) The aim of this study is to investigate the characteristics of electric arc accidents that have happened to electricians, in order to obtain information on the present risk level. We examined the type of electrical equipment, electric arc current, as well as the clothing worn by the workers, damage of the clothing, and the seriousness of the workers’ inju- ries. Methods The study material consisted of electric arc accidents that occurred to electricians in Finland from 1996 to June 1999. Two official registers were used to collect the informa- 128 tion on the accidents. Only occupational accidents were included in the study. In 1996- 1999, 40 electric arc accidents had been recorded. Altogether 171 electrical accidents have been reported to the Safety Technology Authority of Finland during the same period. We contacted the employers of electric arc accident victims and asked whether they would like to participate in the study. We interviewed of 25 electric arc accident victims or employers’ safety delegates using a questionnaire. The purpose of the interviews was to find out features of the electrical installations involved in the accidents, as well as the type of the clothing and damage to the clothing that was worn by the electricians in the accident situation. We also recorded the seriousness of their injuries. The damage to the clothing was evaluated by dividing them into three categories. 1. Minor damage: There was no visible damage at all; or fabric became sooty; it was dotted by splashes of molten metal; or it become brittle after laundering 2. Moderate damage : The fabric hardened but did not break open, got small holes from splashes of molten metal 3. Significant damage: The fabric melted, burned or hardened and broke open in the accident Results Electrical equipment In several cases, the electric arc was ignited when a tool touched live parts. Some acci- dents occurred when a peeled or a broken cable touched live parts. Other reasons for the ignition of an electric arc were faulty equipment, installation or operation. Over two- thirds of the electric arc accidents took place in low voltage installations. On the average, the duration of the electric arc was estimated to be 0.5 s. The short-circuit current varied from 0.1 to 50 kA. In 11 accidents, the branch of activity was the production or distribution of energy. Eight of the interviewed electricians worked in electrical installation enterprises. Acci- dents occurred also in the forest industry, the metal industry, shipbuilding, and several other branches. Every third accident happened to a subcontractor’s employee. Burn injuries Third degree burn injuries were at least 2.5 times as common in high voltage accidents as in low voltage accidents. Third degree burn injuries occurred in at least 43 % of the high voltage electric arc accidents. In the low voltage accidents, 11 % of the accident victims got third degree burns. About three-quarters of the victims got burn injuries on their hands. Nearly every second accident caused facial damage. Half of the interviewed electricians had worn flame-retardant protective clothing when the accident took place. Unexpectedly, the seriousness of the burn injuries was not di- rectly connected to flame retardancy of the clothing material. Instead, the seriousness of the injuries correlated to the damage to the clothing. The clothing was significantly damaged in 75 % of the cases in which the victims got third degree burn. When the inju- ries were first or second degree burns, the clothing was only slightly damaged in 50 % of the cases (Figure 1). 129 SERIOUSMODERATE MINORUNKNOWN 3RD DEGREE BURNS 1ST AND 2ND DEGREE BURNS 25 % 75 % 10 % 50 % 20 % 20 % 0 % 10 % 20 % 30 % 40 % 50 % 60 % 70 % 80 % DAMAGE TO CLOTHING Figure 1. Damage to the clothing and burn injuries in electric arc accidents. In most low voltage accidents, the damage to the clothing was minor, while in high voltage accidents the clothing got severe damage in 50 % of the cases. Both flame- retardant and untreated garments had damages in all three categories. Discussion The material of this study might consist of accidents that are more serious than the average. The Safety Technology Authority of Finland estimates that a large percentage of electric arc accidents, as well as many other accidents, remains unreported. However, se- rious accidents are always investigated and reported. The injuries caused by an electric arc were more serious when they took place in high voltage installations, than they were in low voltage installations. The circumstances varied greatly in the studied accidents. When the energy of an electric arc is high enough, the clothing is destroyed regardless of its material, and serious burn injuries are caused. In the CENELEC draft, a 400 V supply voltage, 0.5 s duration time for an electric arc is and 4 and 7 kA currents are given as test parameters for electric arc testing of flame- retardant clothing. These parameters are in line with our observations, and represent average real life electric arc accidents reasonably well. Half of the interviewed electricians used flame-retardant protective clothing. One third used other work clothing and the rest wore regular clothing. The work was often carried out in the immediate proximity of live electric parts without following live working pro- cedures. A tool falling on or touching live parts was a typical cause of igniting of an electric arc. Using appropriate tools could diminish the accident hazard. Many of the ac- cidents happened to very experienced workers. 130 Conclusions Most of the investigated electric arc accidents took place in low voltage installations, while accidents in high voltage installations were more serious. Third degree burns were at least 2.5 times as common in high voltage accidents as in low voltage accidents. The damage to the clothing was directly proportional to the seriousness of the injuries. Be- cause the circumstances of the accidents varied greatly, both flame-retardant and un- treated clothing were damaged to diverse degrees. Acknowledgements The Finnish Work Environment Fund supported this study. References Askaner, S. (1972) Electric arc faults in at most 24 kV indoor switchgear. Presentation in annual meeting of Swedish fire-fighters electricity council. (In Swedish) Centre of electrical inspection (1986) Announcement T73-86. (In Finnish) CLC/TC 78 (2000) Tools for live working. Draft standard. IEC/TC 78 (1999) Live working – clothing for thermal protection of workers. Martikainen, S. (1978): Examining electric arc and short-circuit endurance of a 660 V switch bay centre. MSc thesis. (In Finnish) Meyer, F. (1960) Control of short-circuit electric arcs in open and encapsulated medium high voltage switchgear. Siemens Zeitschrift 34, Heft 1, 53-59. (In German) Pälli, I. (1988) Protective clothing in electrical work, in which there is a hazard of electric arc. MSc thesis. (In Finnish) 131 Electric arc testing with heat flux measurement for FR clothing materials Sanna Mustonen, Helena Mäkinen, Kalevi Nieminen Finnish Institute of Occupational Health, Vantaa, Finland Introduction Protective clothing used in work involving a hazard of electric arc should be CE-labelled in accordance with the European Union directive on personal protective equipment. As of yet, there is no harmonised standard, which would explain the requirements of the direc- tive. CENELEC CLC/TC 78 is preparing a standard test method for materials and gar- ments used in electrical work. Various European countries have had their own methods for electric arc testing, and research work on the workers’ exposure to electric arc, and testing of clothing materials have been carried out for example at Manitoba University, Canada, and by AMYS, Spain. (King et al., 1988 and 1992; AMYS 1998) The aim of this study was to perform electric arc testing of flame-retardant fabrics, utilising two different test methods. In addition, the rise in temperature underneath the fabric sample was recorded. We evaluated the thermal performance of the fabrics by cal- culating the heat flux through the samples, and compared it to the Stoll curve (Stoll, Chianta, 1969). We also assessed the relation of the changes in the materials to the heat flux peak value. Material and method The test methods used were based on the Swedish EBR test and the CENELEC draft test. The test equipment was constructed according to these test method instructions, respec- tively (CLC/TC 78, 2000; EBR, 1997). With the EBR method, we used 11 and 13-kA test currents. We performed the tests with the CENELEC method using mostly 4-kA cur- rent. In the CENELEC method, the use of a plaster testbox intensifies the effect of the electric arc, and thus a lower test current can be used. For each test, we combined a tem- perature measurement modified from the IEC standard proposal test method A, arrange- ment of two-sensor panels (IEC/ TC78, 1999). There were two calorimeters on a test panel, each consisting of four K-type thermoelements connected to a thin copper plate. In addition to this, there were two thermistors on each test panel. An IR detector was used to obtain a measurement of the fabric’s surface temperature. Total heat flux from the electric arc was measured by exposing the temperature- measuring equipment to an electric arc without any fabric between the sensor panel and the ignition source. The tested samples consisted of single and double layer fabrics. The outer layer was always flame-retardant (FR). The single layer test samples include cotton (CO), cotton and polyester blends (CO/PES) and aramid (AR). We tested two different weight classes of each material type. The outer layer fabrics of two layer samples were FR cotton, FR polyester (PES), and a cotton and polyester blend (Tables 1 and 2). 132 Table 1. Tested single layer materials. Material no. Composition Construction Weight (g/m2) Thickness (mm) Test 1 FR 100% CO Satin 320 0.68 CENELEC, EBR 2 FR 100% CO Twill 175 0.43 CENELEC 3 FR 75% CO, 25 % PES Twill 340 0.72 CENELEC 4 FR 70% CO, 30 % PES Twill 240 0.44 CENELEC 5 100% AR Twill 260 0.51 CENELEC 6 100% AR Twill 165 0.39 EBR Table 2. Tested double layer materials. Material no. Composition Construction Weight (g/m2) Thickness (mm) Test 7 Outer: FR 100% CO Under: FR 100% CO Twill Plain 175 180 0.43 0.42 CENELEC 8 Outer: FR 100% PES Under: FR 100% CO Twill Plain 230 190 0.51 0.41 CENELEC 9 Outer: FR 100% CO Under: 100% CO Satin Plain 320 180 0.68 0.42 CENELEC 10 Outer: FR 75% CO, 25 % PES Under: 100% CO Twill Plain 340 180 0.72 0.42 CENELEC When tested with the CENELEC method, the fabrics were conditioned for at last 24 h in an atmosphere with a temperature of 20  2 C and relative humidity 65  5 %. The samples used in the EBR test were not conditioned. No prewashing was done. All tests were conducted outdoors under a shelter. The ambient temperature was about 10 C when the tests with the EBR method were performed. When testing with the CENELEC method, the temperature was around 0 C. At the beginning of each test, the calorimeters were approximately at ambient temperature. Results The total heat flux from the electric arc was measured without a fabric sample. The total heat flux was 265 kW/m2 in the EBR test and 261 kW/m2 in the CENELEC test. All the tested single-layer fabrics would pass class 1 in the CENELEC test, which, according to the test criteria, does not allow holes greater than 5 mm in any direction. The maximum heat flux value of the Stoll curve is 50 kW/m2; exceeding this value indicates the onset of a second-degree burn injury. The maximum heat flux value for material 1 in CENELEC test was 71 kW/m2. For material 2, the maximum heat flux was 97 kW/m2. With the EBR method, the maximum heat flux for material 1 was 49 kW/m2. Comparing these results indicates a second-degree burn injury would occur in the two former cases but not in the last. However, charring of the fabric was more severe in the test sample tested by the EBR method. Maximum value of heat flux for material 3 was 55 kW/m2 and for material 4, it was 77 kW/m2. The maximum heat flux value in the EBR test for material 6 was 73 kW/m2. The maximum heat flux value in CENELEC test for material 5 was 53 kW/m2. The difference in the results is mainly due to the different weight of the fabrics. Surprisingly, the typical colour change of aramid fabrics was significantly slighter in the samples tested with the CENELEC method. The heat flux for double layer fabrics 7, 8, 9 and 10 was below the Stoll curve. The peak value of the heat flux greatly depended on the weight of the fabric. The heavier the 133 fabric, the lower was the heat flux peak value measured. Even sample 8 with a 100 % polyester outer layer, which melted completely from the exposed area, conformed to this pattern. -10 0 10 20 30 40 50 60 TIME (s) HEAT FLUX (kW/m2) FR CO 175 g/m2 FR C0/PES 340 g/m2 FR CO 320 g/m2 FR PES 230g/m2 STOLL Figure 1. Heat flux curves for two-layer fabric combinations. Discussion The formation of holes in the outer material seems to have a relatively minor effect on the actual heat flux through a test sample consisting of two FR fabric layers. However, the heat flux through a fabric could in some cases be significant even when the fabric is unbroken. This is the case particularly with lightweight FR fabrics. This suggests that the integrity of the fabric alone would not be a sufficient criterion for deciding which mate- rials should be recommended for the protective clothing of electricians. Because the total heat flux values of the two test methods are in a very limited range compared to the inherent variation in an electric arc, we concluded that it is relevant to compare the results of the tests performed with these methods and test currents. In the CENELEC test, the colour of test samples turned dull and muted compared to samples tested with the EBR test, and compared to garments that have been worn in an actual electric arc accident. It is possible that some plasterdust gets comes from the testbox. To- gether with the vapours from electrode materials which condense on the surface of the fabric, the dust forms a layer that may reduce the material changes in this test. All testing was performed in ambient temperature and humidity conditions, which makes it difficult to compare the results to tests performed at some other temperature. However, the results are valid for material comparisons. Burning was observed in the test-box in virtually every test performed. Presumably the molten electrode material had ignited the plaster, because holes were visible in the test- boxes. The burning times in the box varied from less than second to a few seconds. The energy, which is released in the burning process in the testbox, is probably negligible compared to the energy of an electric arc. While its influence can be observed in the heat flux curve, its effect on the test result remains negligible. 134 Conclusions We performed electric arc testing on flame-retardant fabrics. In the CENELEC method, the use of a plaster testbox intensifies the effect of the electric arc. We found that the 4- kA CENELEC test and 11-kA EBR test are comparable in terms of total heat flux. How- ever, the fabric changes in the CENELEC test were lesser, while the heat flux values were higher for samples of the same material. All single-layer samples of the presented selection of FR fabrics pass class 1 (4-kA) according to the CENELEC method test crite- ria. The heat flux for each combination sample was below the Stoll curve. The thicker and heavier the fabric combination, the lower was the heat flux peak value observed. Acknowledgements Tekes, the National Technology Agency supported this study. References Asociacion de medicina y seguridad en el trabajo de unesa para la industria electrica (AMYS) (1998) Thermographic measuring of damage on fabric due to electric arc Asociacion de medicina y seguridad en el trabajo de unesa para la industria electrica (AMYS) (1998) Results of fabric testing (in Spanish) CLC/TC 78 (2000) Tools for Live Working. Draft Standard Doughty et al. (1997) Testing Update On Protective Clothing & Equipment for Electric Arc Exposure. Copyright material IEEE Paper No. PCIC –97-35 EBR, Electricity distributors of Sweden (1997) Branch Requirements for FR clothe. (In Swedish) IEC/TC 78 (1999) Live Working – Clothing for Thermal Protection of Workers. King et al. (1988) Thermal Protective Performance Of Single-Layer And Multiple-Layer Fabrics Exposed To Electrical Flashovers. Performance of Protective Clothing: Second Symposium, ASTM STP 989. King et al. (1992) Evaluating The TPP Of Clothing Exposed To Electric Arcs: A Standard Laboratory Test Apparatus. Performance of Protective Clothing: Fourth Volume, ASTM STP 1133. Stoll A, Chianta M (1969) Method and Rating System for Evaluation of Thermal Protection. Aerospace Med. 40 (11): 1232-1237. 135 Needs for research for protective clothing standards Eero Korhonen Sarvimäentie 26, 04200 Kerava, Finland Introduction After the adoption of the New Approach and Directive 89/686/EEC for Personal Protective Devices (PPE) CEN committees prepare about 300 standards or draft stan- dards. This work has been done in a very short time period and the quality of the stan- dards has not been always good. There has been no time enough to conduct necessary pre- or co-normative research on the working conditions where the use of special protec- tive clothing is needed. The validation of the standardized test methods has not been pos- sible and numerous interlaboratory tests should be done and new better methods developed. Research needs for protective clothing standards Research on the foreseeable use conditions The basis for the standardization of special work clothing is the knowledge about the foreseeable use conditions, about the risks against which the clothing shall give protec- tion as well as the physiological and environmental factors. Based on this clarification, it is evaluated, which essential safety and health requirements of the directive are relevant and should be taken into account in the drafting of the standard. Unfortunately in many cases the requirements for the clothing are given on the basis of the existing products on the market, not on the basis of a carefully conducted risk analysis. The requirements for the clothing should be based on reliable data produced by re- search. There are many good research reports, but unfortunately these are not always taken into account. The standards specify normally several protection levels. The relationship of these protection levels and the existing risk levels are difficult or even impossible to under- stand. The number of performance levels should relate to the different risk levels in fore- seeable conditions of use. In many standards the existing products are classified, not the risk levels. Due to the uncertainty of the measurements, the classes may overlap. A systematic procedure in the drafting of a standard shall be followed. The first step shall be reliable data, which can only be produced by the research. Ergonomic requirements The first part of the essential safety and health requirements contains mainly require- ments related to the ergonomic properties of the devices. There are only few test methods, which can be used when the conformity of the product with the directive is assessed. The human tolerance values against different kind of mechanical, physiological and physical and chemical threats are not available. For instance protective clothing used in 136 various sport activities are mostly protecting against mechanical impacts or abrasion to- wards different part of the body. A general database for these limit values is missing and such data shall be produced. The test methods try normally to simulate the accident situation. All different possible cases can not be simulated. A reliable test method giving clear pass/fail limit should be given for each requirement. A global horizontal harmonization of the methods is needed. The prerequisite for this harmonization is the research on suitable and reliable test methods and validation of those by interlaboratory testing and research co-operation The overall uncertainty of the measurements is found to be big. During the standardi- zation process, all major sources of the uncertainty should be identified and pre- and co- normative research is needed during the whole drafting period. Optimum level of protection The Directive 89/686/EEC requires that PPE be designed to provide the highest possible level of protection. The "highest possible level" is explained in other clauses of Annex II of the directive. It is required that the protection level be only so high that the wearer can use the equipment in foreseeable conditions of use without excessive discomfort or stress. The aim is to get the maximum benefit from the use of PPE taking into account its protection efficiency and wearer acceptance. The standards are prepared considering this basic principle and present technological possibilities. The problem is that this optimum level is very difficult to specify when only few research results are available from the real use situations. The realistic practical per- formance test shall be created on the basis of the field studies. For the time being the practical performance tests are only existing possibility to assess the different properties connected to the ergonomics or usability of the clothing. Chemical safety The horizontal question, how to specify the requirements for the chemical safety of the materials used in the design of the clothing, is still open. The risk determination of various chemicals, which may come into contact with the skin of the user, is expensive and time consuming. Possibilities for a better practical solution should be clarified. Conclusions Research to support the standardization work is needed. When the existing standards are revised, the work shall be based on more reliable data on work conditions, physiological and other relevant factors. The main problem is that the same persons who are respon- sible for the standardization work are also busy with certification or testing of the protec- tive clothing. The compulsory interlaboratory tests shall also be done. The time for the basic and applied research needed is normally not existing. The situation can be im- proved through increased co-operation of the research institutes. There should be also more financial support. 137 A new structure of Ergonomic Standards for PPE – Proposal from Kommission Arbeitsschutz und Normung – KAN (Commission for Occupational Health and Safety and Standardization) Dorit Zimmermann Commission for Occupational Health and Safety and Standardization (KAN), Alte Heerstr. 111, D-53757 Sankt Augustin, Germany Introduction Ergonomic aspects are important for good product design for personal protective equip- ment (PPE), because PPE - which is designed to protect people – is in close contact with the body. It is therefore important that PPE does not affect the comfort of the user or his physical abilities more then is absolutely necessary. PPE should be for example not to heavy and should reduce as little as possible the mobility and dexterity of the user. The European Directive 89/686/EEC demands, in its safety and health requirements, the ergonomic design of PPE. But at the moment, the harmonised European Standards for PPE do not reflect ergonomic aspects sufficiently. It is important to note that it is generally difficult to specify ergonomic parameters and suitable test methods in PPE standardization because it is not easy to establish objective ergonomic requirements. One important step towards solving this problem was the establishment of a joint working group within the European Technical Committee for Ergonomics (CEN/TC 122 JWG 9). The joint proposal from KAN (Commission for Occupational Health and Safety and Standardization in Germany) and the German standards committee for Ergonomics within DIN for a new structure of ergonomics standardization might be a next step. The legal situation in Europe: product requirements versus requirements concerning the health and safety at the workplace Directives based on Article 95 are mainly concerned with the free movement of products in the European Union or the European Economic Area. The target group includes manu- facturers, distributors, public authorities and notified bodies. Such directives normally contain basic health and safety requirements with respect to the products, regulations on conformity assessment, testing and certification and regulations on market surveillance. The aim of the directives is total harmonisation. Therefor the directives must be trans- ferred into national law by the member states without any changes. Because the "essential health and safety requirements" in the Directives are kept quite general, the European Commission intends to supplement these requirements with har- monised European Standards. Although the use of harmonised European Standards is not obligatory for the manufacturer it bears great advantages. When the manufacturer applies these standards, it will be presumed that his products are in conformity with the essential health and safety requirements of the Directives. Therefore, harmonised European Standards play an important role in the harmonisation concept of the European 138 Commission. Product standards developed under these product related directives may contain fixed requirements and limit values if those are considered necessary. Directives based on Article 137 are aimed at the improvement of the safety and health of workers at work. The target group of these Directives includes employers, employees and supervising authorities. The Directives contain company regulations and regulations with respect to the workplace and the work environment. They contain minimum require- ments, and the Member States are obliged to transfer at least these minimum require- ments into national law. But the member states are free to define completing and more stringent provisions in their national law. Harmonised standards are not intended to further specify the requirements of these Directives. But they might be helpful to define terminology or measuring procedures. In no case should these standards contain requirements or limit values, because these might conflict with national legal requirements. The new concept of ergonomic standards The concept takes the legal situation into account and will lead to a hierarchical structure of ergonomic standards and thus to more clarity. This is positive for the users of the stan- dards as well as for the experts developing ergonomic standards. In addition, the concept will help to avoid the development of standards conflicting with existing national legal regulations on health and safety at the work place. The new concept certainly wants to influence the structure of future standards. This does not mean however, that existing standards now have to be replaced by new stan- dards in conformity with the new concept. If the concept is generally accepted, it is ad- visable to adapt the existing standards during their periodical revisions. Many of the existing CEN and ISO standards in ergonomics are already now fitting into the new concept just as if they were designed accordingly. In many cases, this adoption could merely consist in adding a few new headlines and rearranging the existing contents under these new headlines. The concept envisages three levels of standards (Table 1). There is the first level of ba- sic standards the second level of generic standards and the third level of product stan- dards. This concept is to a certain degree comparable with the CEN standardization con- cept under the Machinery Directive. Table 1. Proposed concept for ergonomics standards Type of standard Content Product requirements (Art. 95) Design of the working environment (Art. 137) Basic (B) Principles and methods to describe human factors Generic (G) Product groups (GP) Work systems (GW) Product (P) Integration of data from B and GP in specific product standards Basic standards Basic standards (B) should contain principles and methods to describe human factors such as - terminology 139 - human characteristics (e.g. anthropometric data, human forces, stress/strain) - measuring and evaluation - general design principles for standards on ergonomics. This means that basic standards should describe principles and methods independent of the design of products or work. An example for a basic standard is prEN 979 “Basic list of definitions of human body dimensions for technical design”. The document provides a basic list of anthropometric measurements for use in the establishment of common com- parative definitions of population groups. Generic standards The generic standards (G) must be divided into two groups: - Those standards dealing with product groups (P) and - those standards dealing with work systems (W). In generic standards for product groups those parts of the basic standards which are rele- vant for the product group under consideration are used. These standards may contain quantitative and qualitative ergonomic requirements, they can deal with measuring and test methods and they could contain guides for the application of generic product stan- dards. In many cases it would be useful if basic data described in basic standards was re- peated in these generic standards for product groups. Typical examples for generic product standards are the standards developed in CEN/TC 122 JWG 9 for the field of personal protective equipment (PPE):  WI “Personal Protective Equipment - Ergonomic principles - Part 3: Biomechanical characteristics”  WI “Personal Protective Equipment - Ergonomic principles - Part 4: Thermal characteristics”,  WI “Personal Protective Equipment - Ergonomic principles - Part 6: Sensory factors”. Generic Product standards on special ergonomic aspects, for example PPE-related data on problems such as “sweating” or “skin contact” do not yet exist. Therefore, TC 159 “Hearing protection” itself specified the ergonomic requirements in product standards. On the basis of the new concept, we would suggest to standardize these data in a generic standard. Generic standards for work systems should contain those elements from the basic stan- dards, which are relevant for the work system under consideration. The standards should contain qualitative design information for work systems without setting limits. This is very important since work systems usually describe the situation at the workplace, which is regulated by national law. Generic standards for work systems may also include a de- scription of measuring methods, which can be used in the different cases. For the special field of PPE, there are no standards. Product standards Product standards of the third level should not be developed in Technical Committees dealing with ergonomics. Usually, these product standards are developed in TCs, which are specialised in the technical field the product belongs to. Ergonomic data from Basic standards and generic product standards should be integrated in specific product stan- dards. But at that point it is extremely important that the experts in ergonomics contribute 140 to the development of product standards by offering generic product group standards to other Technical Committees. It does not make much sense if for example an expert group standardizing earth moving machines develops at the same time detailed ergonomic re- quirements for this type of machinery. This should be done by experts of ergonomics on the level of generic product standards. The standards organisations play an important role by informing the different Technical Committees about the existence of generic product standards containing details on ergonomics. Examples for product standards are:  EN 340 "Protective Clothing – General requirements". EN 340 specifies general re- quirements for protective clothing such as ergonomics, the size system and marking. But at the moment, there are some problems concerning the quantification of ergo- nomic parameters. Ergonomic requirements are specified in the form of recommen- dation. It has shown in practice that specifying ergonomic parameters objectively is generally extremely difficult.  EN 420 "General requirements for gloves" specifies design principles and require- ments concerning the possible effects of protective materials on the skin (ph value, chromium(VI) content) and comfort (size, water vapour permeability). The require- ments concerning the possible effects of protective materials on the skin should be standardized in a generic product standard, too. In the field of article 137 of the EC Treaty, no product standards are developed. Conclusion The proposed concept would not mean that existing PPE standards have to be pressed into a bureaucratic pattern. It is much rather intended to consider the structure during the revision of standards. The existing PPE standards are to a large extent already in confor- mity with the proposed structure. KAN sees a need for the documents of CEN/TC 122 JWG 9 to be completed. These mostly belong to the group of generic product standards. What is also needed is the de- velopment of guidelines for PPE TCs on how to use generic product standards when drafting product standards. When ergonomic aspects of a general nature are already specified at the level of PPE product standards, these aspects should be included in GP standards if not partly in basic standards. This would make these aspects available to standardizers when drafting other PPE product standards. What should furthermore be done is check, during the revision of a standard, whether an adaptation of the product standard to the requirements of the relevant GP standard is necessary. For this kind of work, close cooperation between the PPE TCs and CEN/TC 122 JWG 9 is necessary. 141 Main non-conformities of protective clothing detected in the Spanish market Ignacio Cáceres, José Bahima, Eva Cohen Centro Nacional de Medios de Protección, Ins. Nac. de Seguridad e Higiene en el Trabajo Sevilla, Spain Introduction The European Directive 89/686/EEC lays down the obligation for each Member State to survey that the PPE placed in its market comply with the provisions of the Directive, spe- cially that they do not endanger the health and safety of people when used for its intended purpose. In Spain, the Ministry of Industry, in order to effectively fulfil this obligation, and in the exercise of the competence conferred on him by the Spanish legislation, has been carrying out since 1994 annual campaigns for the surveillance of PPE in the Spanish Market, with the collaboration of the Centro Nacional de Medios de Protección, centre of the Spanish National Institute for Occupational Safety and Health specialised on PPE. These control campaigns have reached through all these years a wide variety of PPE, including protective clothing (comprising clothing and gloves) against mechanical, ther- mal, chemical and electrical hazards. Methodology In the course of each annual campaign a number of commercial premises (in most cases specialised shops and superstores) distributed all around Spain were visited. Among the models of PPE available in each store a certain number of them were randomly selected and purchased. For each selected model, the number of purchased units depended on the checks to be carried out to that type of PPE. These checks consisted basically of: ‰ Existence and correctness of the CE marking. ‰ Existence and completeness, in accordance with Annex II, art. 1.4 of the PPE Directive, of the instructions for use provided with the equipment. A great impor- tance was attached to this point since the safety of a PPE can only be assessed in relation to its intended use according to the manufacturer’s instructions. ‰ Compliance with the most representative health and safety requirements, verified by testing in accordance with the applicable European harmonised standards. Table 1. Distribution of the inspected models depending on the type of clothing. Mec. and therm. gloves Thermal clothing Chemical gloves Chemical clothing Electrical gloves 55 % 13 % 18 % 4 % 10 % Table 2. Distribution of the inspected models depending on the origin. Spain Other EU countries Other countries Unknown origin 42 % 33 % 7 % 18 % In the 6 campaigns already concluded 136 different models of protective clothing have been inspected, implying the purchase and testing of around 600 units. Tables 1 and 2 142 show the distribution of these 136 models depending on the type of protective clothing and on the origin of the manufacturer (not necessarily coincident with the country where the equipment is manufactured). The great majority of the models coming from non-EU countries and those of unknown origin (most of them later known to come from third countries) were purchased in superstores, where they are usually commercialised. Results In the first campaign, in 1994, the protective clothing presented a very high level of non- compliance with the Directive. This situation improved continuously and at a good rate till 1997, and then the level of compliance became stabilised. Two clearly different pe- riods may therefore be considered, 1994-96 and 1997-99, with the following results: 47% 26% 30% 78% 84% 64% 37% 81% 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% Models with CE marking Models with information Complete information Models passing all the tests 1994-1996 1997-1999 Figure 1. Protective clothing: levels of compliance with the requirements of the PPE Directive. The rate of complete information is referred to the number of models with instructions. For the detailed analysis of the results that follows, and in order to simplify and make it easier to draw conclusions, we will start from the following considerations: ‰ Three large groups of protective clothing will be considered: gloves and clothing against mechanical and thermal hazards, gloves and clothing against chemical hazards, and electrical insulating gloves. The reason for this is the homogeneity ob- served in each of these groups as regards the type and rate of non-conformities. ‰ Only the results obtained in the period 1997-1999 are presented and analysed, since these are the results that reflect the present situation of the protective clothing. ‰ For the analysis of the results according to the origin of the models, two large groups are considered: the EU models and the rest of the models. The reason is the similarity of the results offered by the models coming from Spain and from the rest of the EU, and the difference with the results offered by the other ones. Protective clothing against mechanical and thermal hazards As shown in Figure 2, the main detected problem relates to the manufacturer’s instruction for use: a considerable percentage of the models did not bring any instructions at all or these were incomplete. The most frequently missing information are the levels of protec- tion and limits of use (in fact the most important information for a safe use of the product), followed by the information about the manufacturer and the certifying body, and the significance of the markings affixed on the equipment. This lack of information about the levels of protection implied that in many cases only the minimal levels of per- formance laid down in the standards (usually easy to pass) were verified. This fact may 143 have helped for the very high rate of compliance with the tests obtained, mainly in the case of the non-EU models where the said information was very often missing. 84% 84% 66% 68% 92% 68% 27% 45% 10% 100% 0% 20% 40% 60% 80% 100% 120% Models w ith CE marking Correct CE markings Models w ith information Complete information Models pass ing the tests EU m o dels N on-EU m o dels Figure 2. Results for the models of protective clothing against mechanical and thermal hazards. As regards the CE marking, the most common defect is the affixing of the two digits of the year and the number of the certifying body, and this is partly due to the amendments of the Directive in relation to the marking. Finally, it is necessary to highlight the dif- ference between the results of the EU and non-EU models, much worse for the latter. Protective clothing against chemical hazards 90% 37% 73% 9% 73% 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Models with CE marking Correct CE markings Models with information Complete information Models passing all the tests Figure 3. Results for the models of protective clothing against chemical hazards. Practically all the inspected models come from the EU. As shown in Figure 3, important levels of non-compliance have been detected for the 3 verified aspects of the Directive: CE marking: It is rather a question of classification of the chemical gloves as PPE of category 2 or 3, what implies different CE marking. While some EU countries, including Spain, interpret the Directive so that chemical gloves (except those of category 1 for protection against minimal risks) belong all to category 3, other EU countries consider that they may be included either in category 2 or 3. Information: We have on the one hand equipment that only bring an address or a tele- phone number where to obtain information and on the other equipment with incomplete instructions for use. The most often missing information are again the levels of protection and limits of use (mainly the resistance to permeation of chemicals), followed by the in- formation about the manufacturer and certifying body, the obsolescence, and the cleaning and servicing instructions. Tests: The tests presenting the highest rate of failure are the permeation test and the jet test (only for type 3 clothing). In the permeation test according to EN 374-3 (which not always could be carried out due to the lack of information) differences of up to 5 levels were detected between the results obtained and the values indicated in the manufacturer’s instructions. As regards the jet test according to EN 463, the different interpretation about the incidence angle of the jet may lead the equipment to pass or fail the test. 144 Electrical insulating gloves 100% 85% 85% 45% 46% 0% 20% 40% 60% 80% 100% 120% Models with CE marking Correct CE markings Models with information Complete information Models passing all the tests Figure 4. Results for the models of electrical insulating gloves. All the inspected models come from the EU. It is worrying the high percentage of models failing some of the tests, in all cases electrical tests (proof and withstanding voltage tests according to EN 60903). Another general problem, not shown in the graph, is the absence of the additional marking requested for this PPE in the Directive (annex II, art. 3.8), and the bad quality of the marking (60 % of the models failed the durability test of EN 60903). Conclusions ‰ The evolution of the results along the successive campaigns proves the effectiveness of this type of market surveillance, especially when we take into account that the in- spections insisted on the equipment and tests which offered the worst results in pre- vious campaigns ‰ The present levels of non-compliance make it advisable the maintenance of the sur- veillance intensity with the aim of protecting the users and preventing unfair compe- tition. It should be increased the control of the equipment coming from third countries usually commercialised in superstores. ‰ It is necessary a better control of the manufacturing process by the manufacturers, so as to ensure the homogeneity of production and the conformity with the certified type. Furthermore, it is surprising that the level of non-compliance with the tests for the whole of the PPE is similar in the case of PPE of category 2 and 3, despite the fact that the latter are required to undergo a periodic quality control by a Notified Body. This result questions to what extent this requirement is being fulfilled and, when that is the case, if that external control is being effective. ‰ The different European forums devoted to PPE, such as the Standing Committee 89/392, the ADCO Group, the Co-ordination of Notified Bodies, etc., must act quickly to solve the problems of interpretation of the Directive and the harmonised standards detected in the market surveillance, which create confusion among the dif- ferent parties involved: manufacturers, distributors, users, etc. References Directive 89/686/CEE on the approximation of the laws of the Member States relating to personal protective equipment (PPE), and its amendments. 145 Evaluating the cutting resistance of protective clothing materials Jaime Lara, Serge Massé IRSST, 505 West de Maisonneuve, Montreal, Quebec, Canada H3A 3C2 Basic principles of a cut test method To cut through a material, it is necessary to slide a cutting component or cutting probe (blade, knife), applying a given force to cross the material. The higher the force necessary to cut through the material, the higher the cut resistance will be. To determine the cut re- sistance level, the first criterion is that the cutting probe should have a controlled quality so that the results obtained with different cutting probes are reproducible. The second criterion is that the test apparatus be reliable and produce a constant normal force through- out the test at the point of contact of the cutting probe with the sample material. It is also necessary to have a valid test procedure that allows cutting probe sharpness to be con- trolled. Forces involved in a cutting test Most standardized cutting test methods use a blade (circular or straight blade) that slides on the sample materials to be cut through, when a given force is applied. The following forces are present in a cutting process: a normal force FN, which is applied at the blade- material point of contact; a frictional force FF, which develops when the blade slides and penetrates the material; the resultant cutting force FR, which is the resultant vector of FN and FF. FN FF FR FN FF FR a) b) Figure 1. Example of forces involved in a cutting process. Figure 1a represents the case of a material where the coefficient of friction FF is much higher than FN, which is typical of some rubber materials. Figure 1b represents the case of a material for which FN is higher than FF, which is typical for knitted fiber materials such as Spectra® and Kevlar®. The magnitude of these forces, which differ from material to material, are important when selecting the right test method for evaluating the cut resis- tance of protective materials. Some of the test methods included in the standards (ASTM, 1997; ISO13997, 1998; EN 388, 1994) will be presented here. 146 Cutting test method EN 388 The test method described in EN 388 uses a round blade that has a diameter of 45 mm. The test apparatus described in EN 388 standard consists of a pivoting arm fixed at one extremity through a pivot to the motor. At the other extremity of the arm, the blade is at- tached at the bottom, with the weight installed directly on the blade. A weight of 5 N is used for the cutting test. At the bottom is the straight sample holder, 90 mm in length with 5 slots, each 5 mm wide. The blade slides into the slots on the material, driven by a motor describing simultaneously an alternative horizontal movement and rotating in the direction opposite its movement. The sample material is fixed on a conductive material. When the blade comes in contact with the conductive material, the test stops and a coun- ter indicates the number of cycles the blade has made to achieve cut-through. A cotton reference material is used to control blade sharpness. A test is performed on the control material before and after a test on the sample material. Thus, the test sequence is repeated five times. The material's cut resistance is determined by the following equation: n C n T n C n i )(   where i n is the index for a given test sequence; C n is the average value for the number of blade cycles for the cotton reference material measured before and after the test on the sample mate- rial; and T n represents the number of blade cycles for the sample material. This test method works well for materials having a low cut resistance level. However, for high cut resistant materials, significant blade sharpness degradation is observed. The results presented in Table 1 are a typical example of this. Table 1. Cutting test results for Spectra glove material. Blade C n T n C n 2 0.8 50 (without cut) 25.5 " 25.5 50-200 (without cut) 24.4 " 27.3 200-400 (without cut) 27.2 " 27.2 400-558 28.1 These results represent a test performed with a Spectra glove where cut-through was obtained after 558 cycles. After 50 cycles without cut-through, the test was stopped and the blade sharpness was verified with the cotton control material. Initially, the number of cycles on the control material was 0.8 cycles, whereas after the test on the sample mate- rial, the test on the control material was 25.5 cycles. This result demonstrated that blade sharpness degrades before material cut-through. Standard EN 388 specifies that when the number of cycles on the control material tested after testing the sample material is higher than 3, the blade should be replaced with a new one. The test was continued in spite of the fact that blade sharpness had already degraded. Blade sharpness was verified with the control material after 200 and 400 cycles without cut-through and the results show no significant changes after the blade degradation detected after 50 cycles. Advantages: Easy to use, Compact Disadvantages: Limited to materials of low cut resistance and without abrasive surfaces 147 Load vs distance concept At the end of the 1980's, a new concept to evaluate the cut resistance of fabrics was de- veloped. It considers the relationship between the load applied to the material-blade point of contact and the blade displacement needed to cut through the material. The basic prin- ciple for a test apparatus complying with this concept is that the normal force applied should remain constant throughout the test and that the blade displacement to cut through the material should be measured precisely (Lara et al., 1996). In the study, blade degra- dation was characterized. Figure 2 shows the effect on blade edge sharpness of repetitive tests on a Neoprene sample material. For example, a new blade cuts through the material at approximately 2 mm when a normal force of 3.92 N is applied. Under the same ex- perimental conditions, its takes approximately 15 mm after reusing the blade for 40 tests and 23 mm after 70 tests. Figure 2 shows how blade edge degradation increases regularly with the number of tests. For this reason, it is recommended that a new blade be used for each cutting test. Furthermore, a procedure to control blade sharpness using Neoprene 1.57 mm thick from Fairprene was proposed. BLADE CUTTING EDGE DEGRADATION Consecutive tests with the same blade 0 5 10 15 20 25 30 0 10 20 30 40 50 60 70 80 NUMBER OF TESTS Blade A Blade B Blade C Test parameters Blade speed: 150 mm/min Load: 3.92 N Blade: Razor type Material: Neoprene (0.79 mm) B LA D E D IS PL A CE M EN T (m m) Figure 2. Consecutive tests performed with blades A (80 tests), B (46 tests) and C (18 tests). In the same study, a procedure to perform the cutting tests was developed. It considers using a straight blade 70 mm long4, and testing a material with at least three different loads, with five tests at each load. Typical curve results are presented in Figure 3. The higher the load applied, the lower the blade displacement will be to cut through the mate- rial. A conductive material is placed directly under the tested material. When cut-through is achieved, the blade comes in contact with the conductive material and the test auto- matically stops. The blade displacement is precisely measured with an LVDT to 0.1 mm. BLADE DISPLACEMENT (mm) 0 4 8 12 16 20 LO AD AP PL IE D (N ) 0 10 20 30 40 50 Kevlar Neoprene A Figure 3. Typical results obtained with cutting test prototype using the load vs distance concept with Neoprene and Kevlar materials. 4 American Safety Razor Company, Stauton, VA 24402-0500, model number 88-0121 148 The curves presented in Figure 3 show that a decreasing exponential relationship exists between the load applied and the blade distance to cut through the material. This type of curve is typical for each material and is used to determine the cut resistance level. ASTM F 1790 cutting test method Figure 4a is a schematic representation of the ASTM cutting test method. It consists of a motor-driven balance arm holding the blade, which moves vertically with a sinusoidal speed. The force is applied normally to the blade at the blade-material contact point. Weights installed on the platen on the horizontal arm result in twice the normal force applied on the blade before a test (static mode). This is due to the 2:1 ratio of the arms, as illustrated in the figure. However, when the blade moves down (dynamic mode), the fric- tional force developed when the blade slides and penetrates the material pushes the mechanism down, which results in an increased cutting force equal to F  N + F  F. For the same reason, when the blade goes up, the cutting resultant force is F  N - F  F , which re- sults in a lower cut resistance value. The higher the frictional coefficient, the higher the error will be in the cutting results obtained with this test apparatus. The use of this test apparatus is limited to materials with very low frictional coeffi- cients and no thicker than 3 mm. Before performing a test, the equilibrium of the test ap- paratus should be verified. This equilibrium is quite difficult to verify. Motor Blade Weight Sample Holder FN 1X 2X Blade direction a) b) Figure 4. Schematic representation of ASTM F 1790 (a) and ISO 13997 Cut Test Methods (b) ISO 13997 cutting test method Figure 4b is a schematic representation of the cut test apparatus in standard ISO 13997. In this test method, the blade slides horizontally at a controlled rate. The force at the material-blade point of contact is applied normally to the blade through a double Watt's mechanism. This mechanism ensures that the sample holder has a precise vertical move- ment. As shown in Figure 4b, the arm lengths have a ratio of 2:1 so that the weight in- stalled on the platen results in twice the normal force applied directly on the sample holder. This test apparatus is sensitive to 0.02 N, is easy to operate, and the equilibrium of its mechanism is easy to establish before a cutting test. The particular Watt's mecha- nism used on this apparatus ensures that equilibrium is reached even if the level arm is not in a perfect horizontal position. It also ensures that materials up to 20 mm thick can be tested without introducing errors into the test results. Furthermore, the mechanism used eliminates the problems caused by frictional forces. Details on the design of the test apparatus were recently published (Massé et al., 1997). Blade Weight Watt's system Sample holder 149 Table 3. Main differences between ISO 13997 and ASTM F 1790 cutting test methods. ISO 13997 ASTM F 1790 Normal force constant Normal force variable and a function of the material’s coefficient of friction Blade speed constant Blade speed sinusoidal Cut resistance calculated at 20 mm blade displacement Cut resistance calculated at 25 mm blade displacement Sample installation: the sample is placed on the double phase tape in direct contact with the conductive material Sample installation: the sample is placed on the double phase tape. The blade should cut through the sample material and the double phase tape to contact the conductive material Correction for blade sharpness as follows: C = 20/l (l is the cutting stroke on neoprene at 5 N) Correction for blade sharpness as follows: C = 25/l (l is cutting stroke on neoprene at 400 g) Because of the differences in the test apparatus and procedures, the results obtained with the ISO 13997 and ASTM F 1790 cutting test methods cannot be compared. Dif- ferences between the results obtained with both test methods are smaller for materials with negligible coefficients of friction, and significant differences are expected for mate- rials with high coefficients of friction. Conclusions The load vs distance concept is emerging as the more valuable approach for measuring cut resistance. It is important that the normal force applied between the blade and the material to cut-through remain constant throughout the test. The use of instruments where this basic principle is not respected will result in erroneous results and incorrect informa- tion about material cut resistance. Furthermore, this incorrect information could affect the development of new and better-performing materials. Progress in the development of a test method to evaluate material cut resistance has been slow because of the complexity of the cutting phenomenon. The development of test methods based on the load vs distance concept have open possibilities for developing knowledge on the breaking mechanism by using instrumented apparatus to measure the forces developed during cut-through. A better understanding of the contribution of these forces will be helpful in the development of improved materials with higher cut resistance. References American Society for Testing and Materials (1997) F 1790-97 Standard Test Method for Measuring Cut Resistance of Materials Used in Protective Clothing. Philadelphia, PA. European Committee for Standardisation (1994) Standard EN 388, Protective Gloves Against Mechanical Risks. CEN Secretariat, rue de Stassart 36, B-1050, Brussels. ISO 13997 (1998) Protective Clothing - Mechanical Properties - Determination of Resistance to Cutting by Sharp Objects. International Standard. Lara J, Turcot D, Daigle R & Boutin J (1996) A New Test Method to Evaluate the Cut Resistance of Glove Materials. Performance of Protective Clothing: Fifth Volume. ASTM STP 1237, Johnson JS, Mansdorf SZ, Eds., American Society for Testing and Materials. Massé S, Lara J, Sirard C & Daigle R (1997) Basic Principles Used in the Development of a New Cut-Test Machine for Standardization. Performance of Protective Clothing: Sixth Volume. ASTM STP 1273, Stull JO, Schwope AD, Eds., American Society for Testing and Materials. 150 Testing materials against small hot metal drops - Development of a new test method Helena Mäkinen, Sanna Raivo, Sanna Karkkula, Erkki Rajamäki Finnish Institute of Occupational Health Introduction There is a diversity of conditions in which splashes of molten metal may come into con- tact with garments used by welders. The current test method for determining the behaviour of materials on impact of small splashes of molten metal is presented in EN 348:1992. This method measures the insula- tion of the material when drops weighing 0.50 g fall on the same point at a velocity of 20 drops per minute. A steel rod is melted in the flame of an oxyacetylene welding torch, and the molten drops are guided through a funnel made of polytetrafluoroethylene onto the fabric attached to a sensor support block. The number of drops required to raise the temperature of the sensor by 40 K is measured (EN 348:1992). The method is quite com- plicated because of many variables which are difficult to adjust and calibrate. Factors re- lated to the formation of a molten metal drop are difficult to control, as are the heat trans- fer phenomena occurring during drop travel, and its subsequent collision with the test specimen. The large spread in the results obtained from three separate interlaboratory trials indicated that there is a need to develop this test method further. Also individual laboratories experience difficulties in reproducing their results. The aim is this paper is to describe the development process of a new method with better accuracy. The method should give satisfactory repeatability between different labo- ratories, and give still a realistic view of the protection performance of the material used for welders’ protective clothing or gloves. The method should be based on the measure- ment of parameters which can be calibrated and are traceable. Method Different types of alternative test methods were sought and the problem was simplified to trying to find methods with clear measurable parameters. Already during the interlabo- ratory trials according to EN 348, some laboratories were looking for an alternative test method. The following test methods were considered: 1. Contact heat test method in accordance with EN 702 (EN 702:1994) 2. Repeated short contact. In this method the number of hot contacts, contact-heat element (900 °), needed to raise the temperature by 20 Kelvin was counted under the sample. 3. Hot ball test method 3.1 One hot ball. In this method a hot steel ball is projected at a point on a horizontally placed test specimen, and the time needed for a 40 C temperature rise is measured. Changes, e.g. holes, adhering of the drop to the fabric, ignition, are recorded. 151 3.2 Several balls. In this method several hot balls are heated by induction and dropped protected on the sample as in EN 348. The number of balls which raise the temperature under the sample to 40 Kelvin is counted. 4. Field trials Parallel with the material tests described above, garments, made of the same fabrics as tested in the laboratories, were given for field trial in a shipyard. The welders evaluated the properties of the garments by filling out a questionnaire on insulation against small splashes, the adhering of the drops in the folds of the fabric or on the fabric, ignition of the fabric. The results of field trials were compared with the inter- laboratory test results according to EN 348 and to alternative test methods. Seven different materials were used for testing. Table 1 summarizes the fabrics used for different interlaboratory trials. The garments for the field trials were made of fabrics 1, 2 and 4. Repeatability and reproducibility was calculated in accordance with ISO 5725-2 (ISO 5725-2:1994). Table 1. Materials for interlaboratory trials. Sample Material Thickness (mm) Weight (g/m²) 1 Pyrovatex 100 % CO, satin 1.17 470 2 CO/PES, 2/2 twill 0.70 380 3a Proban 100 % CO, twill 0.78 360 3b 100% Aramid 0.70 265 4 Proban 100 % CO, satin 0.71 330 5 100 % CO 0.74 310 6 FR 100 % CO 1.06 440 Results The results of the hot ball test method correlated well with the results of the EN 348 method, and also with the welders’ experiences of the fabrics in normal use (Table 2). Table 2. Protection against welding drops. Method Fabric 1 Fabric 2 Fabric 4 Impact of molten drops (EN 348) 18-36 (27.4) 11-21 (15.3) 11-22 (16.8) Repeated hot contacts 13.3 16.9 15.7 Hot ball test (time measured) 13.1 s 5.7 s 6.5 s Welders’ subjective experiences: 1) Insulation 2) Nonadherence 2.33 new 2.43 used 3 new 2.43 used 1 new 2 used 1 new 1.83 used 1.67 new 2.3 used 2.67 new 2.43 used 3=good, 2=moderate, 1= poor Prototype 1 In June 1998 after three interlaboratory trials with EN 348, and after considering alterna- tive test methods, the decision was made in CEN TC 162/WG 2 to build the test appara- tus in one laboratory and to circulate it to other laboratories. The equipment was based on suitable parts of the current EN 348. It consisted of: 152  A funnel, made of heat-insulating material which directs a steel ball onto the speci- men; the drop guide according to EN 348 was used  An oven as a heat source, capable of maintaining the steel ball at a temperature of 900±20 °C. The ball was heated in an oven in a cone. When the temperature of the steel ball was 750±20 °C it was guided onto the sample from the cone.  A steel ball weighing 0.5 g with diameter of 4.2 mm  A timer, capable of measuring the elapsed time Results were obtained from seven laboratories. The variation between the laboratories was considerable. There was also variation within one laboratory. This was anticipated. The testing needs many manual operations and practising is necessary in order to perform the tests successfully. On the other hand, the results from some of the laboratories were very similar. Figures 1 and 2 show the repeatability (r) and reproducibility (R) for mate- rials 1, 2 and 5. For material 3b r was 3.56, and R 18.2 of the results of all laboratories. For the results of selected laboratories, r was 2.28, and R 1.49. For material no 4, r was 6.41 and 1.62, and R 18.05 and 3.05, respectively. Prototype 2 The reasons for the great variation in the results were analysed, and the improvements tested at the FIOH. Some parts of the equipment were modified by FIOH and distributed to voluntary laboratories in 1999 with new samples and work instructions, including a video presentation. The modified parts were the cone, the sensor, the sensor support block, the solenoid with an electrical switch, and the lowest part of the drop guide. The tested materials were nos. 1, 2 and 5 from the previous trial, and also tighter FR cotton material (no. 6, in Table 1). The repeatability and reproducibility with the second prototype apparatus are shown in Figures 1 and 2 for materials 1, 2 and 5. For material 6, r was 1.6 and R was 2.97. 0 2 4 6 8 10 1995 1997 1998/ All 1998/ Sel 1999 Trial R ep ea ta bi lity Material 1 Material 2 Material 5 Figure 1. The development of repeatability. 153 0 5 10 15 20 1995 1997 1998/ All 1998/ Sel 1999 Trial Re pr od uc ib ilit y Material 1 Material 2 Material 5 Figure 2. The development of reproducibility. Figures 1 and 2 show that the repeatability and reproducibility of the test results from the first trial with EN 348 in 1995 to the trial with the second prototype hot ball test method in 1999 have improved remarkably. Conclusion The proposed method is based on normal parameters which are easy to trace and calibrate, i.e. temperature and time. Until now, only very traditional fabrics used for welders' garments are tested. Tests with additional fabrics used for welders' clothing and gloves (leather, Kevlar knit and new CO and viscose blend fabrics with aramid and other new fibres) are going on to see how this method screens out different types of fabrics. Totally 12 European laboratories have participated in the trials in some phases (four labs to all trials) The work has been conducted in laboratories as additional work without any supporting money. The laboratories have only paid for the basic material and the construction work cost of the new equipment for the co-ordinating laboratory. This shows that testing houses really want to find ways to carry out the tests in a reliable and reproducible manner. This also shows that the development is a long process. References EN-348. (1992) Protective Clothing - Test Method: Determination of behaviour of materials on impact of small splashes of molten metal [European Standard]. Brussells: Comité Européen de Normalisation. EN 702 (1994) Protective clothing - Protection against heat and flame - Test method: Determination of the contact heat transmission through protective clothing or its materials [European Standard]. Brussells: Comité Européen de Normalisation. ISO 5725-2 (1994) Accuracy (trueness and precision) of measurement methods and results-Part 2: Basic method for the determination of repeatability and reproducibility of a standard measurement method. [International standard] 154 Revision of test methods: Better screening of PPE materials against liquid pesticides Anugrah Shaw1, Eva Cohen2 and Torsten Hinz3 1 University of Maryland Eastern Shore, Princess Anne, MD, U.S.A. 2 Centro Nacional de Medios de Protection, I.N.S.H.T., Sevilla, Spain 3 Federal Agricultural Research Centre, Braunschweig, Germany Introduction Use of appropriate Personal Protective Equipment (PPE) is recommended while handling pesticides. Currently three test methods are proposed or used as national/international standards to measure repellency, penetration and/or retention properties of PPE materials against liquid pesticide formulations. A previous study indicated that the three test methods used to measure performance of textile materials against liquid pesticides pro- duced significantly different results (Shaw et al., 1999). These variations are primarily due to the differences in approach in the development of the test methods. The gutter test (EN 368) was developed to measure fabric performance against splashes or accidental liquid spillage. The atomizer method (BBA –3-3/2 German guideline) was designed to measure the performance of textile materials against atomized liquid pesticide typically used for field spraying. Whereas, the pipette method (proposed ASTM method) was de- signed to measure the performance of textile materials against known volumes of liquid pesticides (concentrate and field strength). The objective of this study was to revise the three test methods such that the results reported were a true indication of the performance of textile materials against liquid pesticides applied under various exposure conditions in agriculture. Methods Atrazine (46 % a.i., flowable liquid) diluted to 0.23 % a.i. was used for all three test methods. Physical characteristics of the six fabrics are given in Table 1. Table 1. Physical characteristics of the fabrics Code Fiber Content Fabric weight (g/m2) Fabric construction Yarn count (yarns/cm) Water repellency (range 0-6) Comments A polyethylene 72 nonwoven* NA 3 nonporous material B cotton 277 twill weave 44 x 23 6 fluorochemical finish C polypropylene 61 nonwoven NA 3 Proshield 1® D cotton/polyester 245 twill weave 40 x 25 0.5 E cotton/polyester 248 twill weave 41 x 26 0 Fabric D washed F cotton 127 plain weave 56 x 28 0 --- Note: *nonwoven with polyethylene film laminated to the face of the fabric (Tyvek®QC) 155 Gutter test The gutter test was modified to allow additional contact time between the pesticide and test specimen. This was achieved by changing the angle of the incline and/or the rate and pressure at which the pesticide was applied. Based on the preliminary test results, three modifications (45º, 10 mL in 10 sec. was used for the previous study) were selected: 1. 45º incline, 10 mL applied in 2 min 50 s as drip (high incline, low rate and thus no pressure) 2. 10º incline, 10 mL applied in 10 s (low incline, high rate and thus with pressure) 3. 10º incline, 10 mL applied in 2 min 50 s as drip (low incline, low rate and thus no pressure) The following procedure was used for the gutter test modifications:  The fabric specimen (top layer) and Benchkote paper (collector layer) were weighed separately and then clipped to the gutter.  A syringe was used to apply 10 mL of the pesticide formulation to the surface of the fabric.  Pesticide runoff was collected in a pre-weighed beaker placed at the bottom of the gutter.  A cover was then held against the fabric surface to ensure contact with the collector layer.  Both layers were separated and weighed. Amount of liquid in the fabric layer was used to measure pesticide retention and that in the collector layer to measure penetra- tion. The liquid collected in the beaker was used to measure repellency. Total amount applied was calculated by adding the repellency, pesticide retention, and penetration values for each fabric specimen. Results were reported as a percentage of amount applied. Pipette test The pipette test was modified to measure fabric performance against two levels of con- tamination. For lower level 0.1 mL and for higher level 0.2 mL of liquid pesticide was used. The revisions were based on tests conducted using different volumes and con- centrations. The following procedure was used for both levels of contamination:  Test assembly consisted of 8x8 cm fabric specimen (top layer) and 8x8 cm Benchkote Plus paper (collector layer) held together between two 10x10 cm Plexiglas plates – a base plate and a cover plate with a 6x6 cm opening in the center.  A pipettor with a disposable tip was used to apply the pesticide as a single drop to the center of the fabric specimen.  After 10 min., a second Benchkote Plus paper was placed between the fabric and the cover plate to measure repellency. After another two minutes, the three layers were separated and extracted. Each layer was extracted twice in 50 mL of acetone in an or- bital shaker for 30 minutes at 200 rotations per minute. The two aliquots were com- bined and analyzed using a gas chromatograph with a N/P detector.  Pesticide extracted from the paper placed on top of the fabric was used to measure repellency. The amount extracted from the fabric was used to measure pesticide re- tention and that in the collector layer to measure penetration. For this study the 156 amount applied was calculated by adding the amount extracted from each layer. Re- sults were reported as percentages of amount applied. Atomizer test Preliminary tests were conducted to determine factors that would enable the atomizer to apply higher volumes and/or concentrations of the liquid pesticide. Syringe needles with different inner diameter were used. Based on the preliminary test results, 2 mL of the pesticide was atomized. Of the 2 mL applied, approximately 0.5 mL to 0.6 mL was de- posited on the surface of the fabric specimen. The following procedure was used for the atomizer test.  Test assembly consisted of 8x8 cm fabric specimen (top layer) and 8x8 cm Benchkote Plus paper (collector layer) held together between two circular metal plates – a base plate and a cover plate with 7 cm diameter opening in the center.  The test assembly was enclosed in a cylindrical test chamber with a suction device and an atomizer.  A syringe was used to dispense 2 mL of the pesticide under 300-hPa air pressure into the test chamber. The fabric specimen was exposed to the atomized spray in the chamber for 3 min. The test assembly was then removed from the chamber and placed under the fume hood for 30 min.  The layers were separated for extraction. The liquid on the surface of the fabric specimen was extracted with the specimen.  An orbital shaker was used to extract the fabric specimen and the collector layer in acetonitrile for 30 minutes at 200 rotations per minute. They were extracted once in 20 mL and then twice in 15 mL. The three aliquots were combined and analyzed using HPLC.  The amount extracted from the collector layer was used to measure penetration. Re- pellency and pesticide retention were not measured for this method. Amount applied was calculated by adding the amount extracted from the fabric and collector layer. Penetration was calculated as a percentage of the amount applied. Results and discussion Percent repellency Percent repellency was measured for gutter and pipette tests. The mean values for Fabric A ranged from 92.7 % to 99.6 % and for Fabric B from 85 % to 99.5 %. Fabric A had a polyethylene film on the surface and Fabric B had a fluorochemical finish that resulted in higher repellency values for both fabrics. For the gutter tests, the low angle of incline and low rate at which the pesticide was applied affected the repellency values for Fabrics C, D and F. Percentage of repellency was reduced drastically due to increased contact time between the liquid and the fabric. These results indicate that the lower angle and rate may be better suited to measure the repellency. There was minimal or no measurable percent repellency for Fabrics C, D, E, and F using the pipette method. For Fabric C, a previous test conducted using the same conditions had measurable percent repellency. The dif- ferences were due to the variation in the nonwoven fabric. 157 Percent pesticide retained The amount extracted from the fabric specimen is reported as pesticide retained. This may be pesticide on the surface, or within the fabric structure. Percent pesticide retained was measured for the pipette and gutter tests. The results ranged from 0.5 % to 6.7 % for fabric A and 0.8 % to 13.7 % for fabric B. For Fabrics C, D and E, most of the pesticide was retained by the fabric when 0.1 mL was applied using the pipette method. The per- cent pesticide retained by the material was lower for 0.2 mL as the liquid holding ca- pacity of the fabric had been reached and so the pesticide penetrated through the fabric. Gutter Test - 10 ml in 2 min 51s, 45o 0 20 40 60 80 100 A B C D E F Gutter Test - 10 ml in 2 min 51s, 10o 0 20 40 60 80 100 A B C D E F Pipette Test - 0.1 mL 0 20 40 60 80 100 A B C D E F Gutter Test - 10 ml in 10s, 10o 0 20 40 60 80 100 A B C D E F Pipette Test - 0.2 mL 0 20 40 60 80 100 A B C D E F Atomizer Test - 0.5 to 0.6 mL 0 20 40 60 80 100 A B C D E F Repellency Pesticide Retention Repellency + Pesticide Retention Penetration Figure 1. Means for repellency, pesticide retention and penetration for fabrics using different test methods (repellency and pesticide retention was measured as one value for atomizer test). Percent penetration Pesticide penetration was measured for all tests. Fabric F had the maximum penetration using all test methods. There was minimal or no penetration through Fabrics A and B with the exception for fabric B (3.8 %) when the pesticide was applied at a higher rate at 10o incline. For the pipette method, the percent penetration increased with the increase in volume applied to Fabrics C, D, E, and F. This, as mentioned previously, was because the holding capacity of these materials had been reached. For atomizer test, 0.24 % pene- trated through Fabric C, 14.5 % through Fabric F, and no measurable amount through the 158 other four fabrics. As in the case of 0.1 mL pipette method, the holding capacity of Fabrics C, D and F had not been reached. As repellency is not measured for the atomizer test, reduced penetration due to repellency could not be differentiated from amount re- tained by the fabric. The differences in the penetration data for the atomizer and pipette tests were due to the size and distribution of the droplets on the material. Conclusion Test method selection should be based on the pesticide exposure conditions to which the PPE materials will be subjected. Gutter test with lower angle and rate and pipette method with higher contamination level are more suitable to evaluate PPE materials against high exposure with dilute formulations e.g., conditions where there may be equipment leakage and/or rubbing against contaminated foliage in addition contamination due to spray drift. The atomizer method is suitable for measuring the penetration through materials when there is low contamination only due to fine droplets. References Shaw A, Cohen E and Hinz T (1999) Protection from Pesticides: A Comparison of Three Methods of Measuring Contamination of PPE. The XIVth International Plant Protection Congress. Jerusalem, Israel 159 A new British standard: The assessment of heat strain for workers wearing personal protective equipment Margaret Hanson Human Sciences Section, Institute of Occupational Medicine, Edinburgh, EH8 9SU, UK Introduction In many industrial situations workers are required to wear personal protective equipment (PPE) to protect against hazards. Although offering protection, it is recognised that some forms of PPE may increase the risk of heat strain. Methods for estimating potential heat stress include the WBGT index (BS EN 27243) and Required Sweat Rate index (BS EN 12515) (see Parsons, 1999). However, these methods assume that the clothing worn is water vapour permeable; the WBGT index also assumes it is relatively light. Since most forms of PPE either have a higher insulative value than that assumed or are not water va- pour permeable, these standards cannot be accurately applied when PPE is worn. A British standard (BS) was developed to allow interpretation of these standards for workers wearing PPE. This paper outlines the development and contents of the BS (to be published as BS 7963: 2000) and its application. Development of the British standard A detailed work programme was undertaken to obtain information relevant to the BS. The research comprised of a literature review and discussion with experts to provide relevant information; two questionnaire surveys of potential users of the BS (PPE manu- facturers and health and safety experts) to identify their needs and current use of thermal standards; and limited validation of the BS using physiological data from work situations and experiments (see Hanson, 1999). The research was overseen by a steering group ap- pointed by BSI subcommittee PH/9/1. The draft BS was circulated for public comment in May 1998, discussed at the UK Clothing Science Group meeting in June 1998 and amended based on comments received during this consultative process. The remainder of this paper outlines the contents of the BS. It covers the effect of PPE on heat balance, and the application of methods to interpret existing thermal standards to take account of PPE. Content of the British standard: the effect of PPE on heat balance The heat balance equation can be expressed as: M - W = Cres + Eres + C + E + R + K + S The effect of PPE on heat balance can be summarised in these terms:  Metabolic rate (M) can be increased due to the weight of the PPE or the restrictions it imposes on the wearer’s movements.  Convection and evaporation in the respiratory tract (Cres + Eres) can be affected by the temperature and humidity of the air breathed, e.g. through BA.  Convection (C) and conduction from the skin (K) can be affected by the amount of the body covered by PPE and by its thermal insulation properties. 160  Evaporation from the skin (E) can be affected by the amount of the body covered by PPE, by its thermal insulation properties, and by its evaporative resistance.  Radiation from the skin (R) can be affected by the amount of the body covered by PPE. In environments where there is a high radiant load, heat may be gained by the body.  There can be a heat load through chemical activity of PPE (e.g. closed circuit BA). W is the mechanical work done by the body during the physical activity and S is heat storage in the body. If heat storage exceeds heat loss, heat strain will occur. The specific guidance in the BS concerning how PPE may affect these factors, is summarised briefly below. Full guidance is contained within the BS. In assessing the risk of heat stress when wearing PPE, the environment (e.g. temperature, humidity, air ve- locity), metabolic rate and individual factors (e.g. acclimatisation) obviously need to be considered in relation to the PPE worn. The effect of PPE on metabolic rate PPE can increase the wearer’s metabolic rate due to its weight or the restrictions it im- poses on movement. This impact will depend on activity, and will be greater at higher metabolic rates. Typical increases in metabolic rate when PPE is worn, based on experi- mental data, are shown in Table 1 below. These increases should be added to the meta- bolic rate estimated for the activity being undertaken (see BS EN 28996: 1994). Where a combination of items is worn, the increase in metabolic rate due to each form of PPE should be added to the metabolic rate for the activity being undertaken. Table 1. Estimated increases in metabolic rate (M) due to wearing PPE. Increase in M due to wearing PPE (W·m-2) PPE item Resting Low M Mod. M High M V. high M Safety shoes / short boots 0 5 10 15 20 Safety boots (long) 0 10 20 30 40 Respirator (low/moderate performance, e.g. P1, P2) 5 10 20 30 40 Respirator (high performance, e.g. P3) 5 20 40 60 80 Self-contained breathing apparatus 10 30 60 95 125 Light, water vapour permeable chemical coverall (e.g. disposable) 5 10 20 30 40 Chemical protective water vapour impermeable ensemble (e.g. PVC) with hood, gloves and boots 10 25 50 80 100 Highly insulating, water vapour semi-permeable ensemble (e.g. firefighters’ gear) 15 35 75 115 155 The thermal insulation of the material covering the body Thermal insulation values (I cl) for clothing ensembles are given in BS ISO 9920: 1995. Few values are given for different forms of PPE, but where these are available they have been included in the BS. The evaporative resistance of the material covering the body The evaporative resistance (RT) of material is a significant factor in heat stress, but is of- ten not quantified. Materials with low evaporative resistance will allow evaporation of sweat through them, and thus reduce heat stress. However, where evaporative resistance 161 is high, a condensation-evaporation cycle can develop within the garment, allowing some cooling. Closure of garments to the body ‘The pumping effect’ (ventilation of the garment caused by the wearer’s movement) will provide some cooling, dependent on activity. At metabolic rates below 100 W.m-2 ther- mal insulation of clothing should be reduced by 10 %; at metabolic rates above this it should be reduced by 20 %. These reductions in thermal insulation should not be applied to garments that are sealed to the body, since they provide limited potential for this effect. PPE which compresses the ensemble (e.g. harnesses, belts) also reduces the potential for this effect; the reduction in thermal insulation should not be applied if this type of PPE is worn. The proportion of the body covered by PPE Heat stress will be affected by the amount of the body that is covered by PPE. Based on data from BS ISO 9920: 1995, the proportion of the body covered by PPE was estimated. Where several garments, that overlap are worn, the extent of overlap is considered. The effect of respiratory protective equipment A supply of compressed air (open circuit BA, airlines) may provide some cooling to the body, primarily through convective heat loss and evaporation of water from the lungs. With an air line, some air can be vented into the suit and provide additional (evaporative and convective) cooling. In contrast, closed circuit BA usually delivers warmer air with high humidities, and this may result in some heat gain. However, wearing BA will in- crease the metabolic rate and reduce the movement of air within the garment (the pump- ing effect) and thus the potential for cooling. On balance, the net effect of air delivered from a cylinder worn on the body can be to increase heat stress, while the net effect of air provided by an airline can be to decrease heat stress. Table 2. Correction factors to be applied to BS EN 27243:1994 WBGT reference values when PPE is worn (from ACGIH, 1990) Clothing type Thermal insulation (clo) Correction factor for WBGT (qC) Summer work uniform Cotton coveralls Winter work uniform Water barrier, water vapour permeable ensemble Water vapour impermeable ensemble (fully encapsulating suit, hood, gloves, boots) 0.6 1.0 1.4 1.2 1.2 0 -2 -4 -6 a -10 a, b a Because these WBGT correction factors were derived from experimental data, it is not necessary to increase metabolic rate due to these items of clothing. If PPE in addition to that listed is worn (e.g. BA) the metabolic rate should be increased to take account of only these additional items. b Use the WBGT reference value for ‘no sensible air movement’ with this correction factor. Also, when a water vapour impermeable ensemble is worn, the ambient dry bulb temperature is more important in determining heat stress than the wet bulb temperature as sweat cannot be evaporated into the environment. In effect, the dry bulb temperature replaces the WBGT. 162 Content of the British standard: Application to existing thermal standards BS EN 27243: 1994, WBGT index BE EN 27243: 1994 provides limit values for the WBGT above which action should be taken to reduce the risk of heat stress. To interpret this for workers wearing PPE, the normal data required for this standard should be collected. Two additional steps should be taken to account for the PPE: firstly, apply a correction to metabolic rate due to PPE (see Table 1), if required; secondly, if the worker is not wearing a ‘standard working garment’ (see BS EN 27243:1994), apply a correction to the WBGT limit values, see Table 2. The correction factor for PPE should be applied to the WBGT limit value. Any measured WBGT value above this will require action to reduce the risk of heat stress. BS EN 12515: 1997, Calculation of required sweat rate BS EN 12515 provides a more detailed method for assessing thermal stress, taking ac- count of the thermal environment, clothing worn and the work undertaken. It includes a computer program which calculates the sweat evaporation rate required for thermal balance. This requires information about: duration of exposure; dry bulb air temperature; globe temperature; psychometric wet bulb temperature; air velocity; metabolic rate; pro- portion of the body exposed to environment; and thermal insulation of clothing. PPE can affect the metabolic rate, proportion of the body exposed to the environment and clothing thermal insulation (including the pumping effect). Based on the guidance in this BS these input variables can be adjusted directly to take account of the PPE. The effects of water vapour impermeable PPE also need to be accounted for. When such clothing is worn, the potential to evaporate sweat into the environment will be neg- ligible. Therefore, the wet bulb temperature (water vapour pressure) is not relevant to potential heat loss; it should be set to equal the measured dry bulb temperature. Also, air velocity will have only a limited effect on the amount of heat lost from the body. The air velocity input should be set to 0.1 ms-1 to reflect this limited potential for heat loss. Respiratory protective equipment will affect inspired air temperature. The variable for inspired air temperature in the BASIC program should be altered to 45 [C] if closed cir- cuit BA is worn; 25 [C], if open circuit BA is worn; and to the dry bulb temperature if a filter which does not chemically react with the inspired or expired air is used. The BS provides guidance on the impact of PPE on heat strain, and methods for inter- preting existing heat stress standards for workers wearing these garment. Three worked examples using the methods described above for BS EN 27243 and BS EN 12515 are given in the Annex of the BS. References ACGIH (1990) 1990-1991 Threshold limit values for chemical substances and physical agents and biological exposure indices. American Conference of Governmental Industrial Hygienists, 1330 Kemper Meadow Drive, Cincinnati, OH 45240-1634. Hanson MA (1999) Development of a draft British Standard: The assessment of heat strain for workers wearing personal protective equipment. Annals of Occupational Hygiene, Vol 43 No 5, 309-319. Hanson MA & Graveling RA (1999) Development of a draft British Standard: The assessment of heat strain for workers wearing personal protective equipment. IOM Research Report TM/99/03 Parsons KC (1999) International Standards for the assessment of the risk of thermal strain on clothed workers in hot environments. Annals of Occupational Hygiene, Vol 43 No 5, 297-308 163 Assessment of the scientific validity of ISO 7933/EN 12515 Robin Howie Robin Howie Associates, 12, Morningside Road, Edinburgh, Scotland Introduction Ergonomic standards such as ISO 7933 and ISO 9920 are based on a number of assump- tions that do not adhere to conventional thermodynamic principles, e.g. the concept of a thick layer of stationary air on the outermost surface of the garment and such air layer being opaque to low frequency radiation, or which are contrary to common experience, e.g. the change in clo value due to wearing gloves is effectively nil yet personal expe- rience in cold weather is that gloves do make a substantial difference. European Standards for high performance protective clothing will shortly be published. To meet the performance requirements of these new standards clothing may be manu- factured from fabrics which have poorer air and water vapour permeability and which are better sealed to the wearer’s body than many current garments. These new garments may therefore have greater potential than current garments for creating or exacerbating heat strain. ISO 7933 has been adopted as EN 12515, and may therefore be used to assess the potential for protective clothing to generate heat strain, e.g. EN 12515 has been incorpo- rated into a draft UK standard for such assessment, BS 7963:2000. It is therefore essential that the scientific validity of ISO 7933/EN 12515 be assessed. Methodology In assessing the scientific validity of ISO 7933/EN 12515 it is useful to compare the rele- vant heat transfer equations described in basic thermodynamics with those used in the ISO/EN standards. The text referred to hereinafter is that from EN 12515:1997. The simplest situation for quantifying heat transfer between the body and the environ- ment is where the inner surface of the clothing is in contact with the skin. If such contact does not occur heat is transferred between the skin and the inner surface of the clothing by natural convection, forced convection caused by pumping of the in-garment microcli- mate generated by movement of the body and/or radiation. Sensible heat transfer between the body and the environment through impermeable clothing sealed to the wearer’s body and not in contact with external surfaces occurs by: heat transfer between the skin and the inner surface of the clothing by physical contact, radiation or natural or forced convection; heat transfer through the clothing by conduc- tion; heat transfer between the outer surface of the clothing and the environment by con- vection and radiation, i.e. algebraically, conduction = convection + radiation. 164 That is, for clothing in contact with the skin, the steady-state net sensible heat transfer by convection and radiation in still air is given by: k c (t sk - tso) = hc(tso - ta) +  p[(tso + 273)4 - (tr + 273)4] 5.7 x 10 -8 where: k c thermal conductivity of clothing t sk skin temperature t so clothing outer surface temperature h c convection coefficient t a dry bulb air temperature H radiant absorbency of outermost surface p posture correction t r mean radiant temperature Note: if t r >> t a then t so can exceed t a because the surface gains more energy by radiation than it loses by convection. h c for natural convection can be calculated in a stepwise fashion. 1) The Grashof number, Gr, = g (t so - t a )L3/ 2 where: g gravitational acceleration E 1/[273 + 0.5(t so + t a )] L Characteristic (vertical) length Q kinematic viscosity at 0.5(t so + t a ) 2) h c = k a C(Gr.Pr)a/L where: k a thermal conductivity of air C = 0.59 and a = 0.25 for Gr.Pr in range 104-109, fluid flow in the boundary layer is laminar C = 0.13 and a = 0.33 for Gr.Pr > 109, fluid flow in the boundary layer is turbulent Pr Prandtl number at 0.5(t so + t a ) The above equations can be easily solved by successive approximation. Assessment of validity of ISO 7933/EN 12515 The validity of the above equations for heat transfer between the body and the environ- ment can be assessed for a thermal manikin as specified in ISO 9920 (1995). Assuming the manikin is 1.7 m tall, has a surface area of 1.8 m2, a mean skin temperature of 33 ºC,  = 0.95 and a posture factor of 0.77 and is in a thermal environment where t a =t r = 25 ºC, the mean heat loss is 25 W/m2 convective and 37 W/m2 radiative, midway between the overall specified limits of 40-80 W/m2. Note that flow in the boundary layer is turbulent, i.e. h c varies as (t so - t a )0.33. 165 The above equations differ critically from those adopted in EN 12515. 1) Equations (4), convection, and (5), radiation, in EN 12515 are based on the dif- ference between t sk and ta and tr rather than between tso and ta and tr respectively. As fluid flow in the boundary is generally turbulent so that the temperature dif- ference is raised to the power 1.33 for convection and to the power 4 of the abso- lute temperature for radiation, the error of assuming t sk rather than tso as the basis for temperature differences cannot be corrected for by altering the values of the linear coefficients in the equations unless within a limited range of applicability. The range of such applicability is currently unspecified. 2) Equation A.3 for convective heat transfer assumes laminar flow rather than tur- bulent flow. However, for other than minimal temperature differences between t so and t a flow in the boundary layer is turbulent. Consequently h c should vary as (t so - t a )0.33 rather than as (t so -t a )0.25 as in EN 12515. As above, the difference between (t so -t a )1.33 rather than (t so -t a )1.25, for other than small temperature differences cannot be corrected for by altering the value of the linear coefficient unless within a limited range of applicability. The range of such applicability is currently unspecified. 3) Equation A.7 for radiant heat transfer is based on the absorbency/emissivity of skin. For radiative heat transfer between an effectively fully clothed body and the environment the only radiative transfer between the skin and the environment will occur from uncovered skin, e.g. the hands or face. For a body which is effectively completely clothed, the relevant radiant absorbency/emissivity is therefore that of the clothing outer surface(s). Equation A.7 is therefore invalid for fully clothed bodies. However, radiation between the skin and the environment can be impor- tant for uncovered skin and between the skin and the inner surfaces of the cloth- ing can be important in some situations. 4) From the form of Equations 4, 5 and A.8 heat transfer by convection and radiation must both be either positive or negative, i.e., EN 12515 cannot accommodate situations where t r >> t a such that t so > t a . However, when wearing a garment with a conductivity of 5 W/m2/K and an  of 0.95 in a t a of 50 ºC, convective heat transfer becomes negative when t r > 63 ºC. Such conditions are commonly en- countered by workers such as restaurant chefs, asbestos strippers, welders and fire-fighters. 5) EN 12515 uses clo values derived on the basis of the assumption that there is a thick layer of stationary air attached to the outermost surface of the garment. Authors such as Havenith (1999) describe a clothing material as having “a still air layer attached to its outer surface” up to 6 mm thick. From the above application of basic thermodynamic principles to a thermal manikin there is no need to pos- tulate a thick stationary air layer adhering to the surface of the manikin as as- sumed in ISO 9920 and ISO 7933/EN 12515. The assumption of a thick stationary air layer may be based on a failure to understand that although the boundary may be a number of mm thick, there is significant air move- ment within the boundary layer and that within the boundary layer there is a very thin layer of stationary fluid adjacent to the surface through which energy eventually trans- ferred by convection must initially be conducted. In EN 12515 it is further assumed that the stagnant air layer affects radiant transfer. If a thick stagnant air layer was present, such an air layer would be effectively transparent to radiant energy of the frequencies of interest and therefore would not affect radiative heat 166 transfer. The radiative transfer calculated from ISO 7933/EN 12515 is therefore likely to be underestimated. This could be critical in high radiant energy situations. Conclusion By comparison with the principles of basic thermodynamics, the scientific basis of ISO 7933/EN 12515 is faulty. However, the standard may be applicable through a limited, presently undefined, range of conditions. A thorough scientific review of the standard is therefore required. Until completion of the review and replacement by a scientifically valid standard, the limits of applicability of ISO 7933/EN 12515 should be clearly stated. Such scientific review should be extended to all other standards in which clo values de- rived from tests based on the assumptions of a thick stationary boundary layer and/or heat transfer being driven by the differences between skin temperature and ambient dry bulb or radiant temperatures, e.g. ISO 9920:1995. Such review should ensure that all assumptions on which the standards are based and all individual steps within the standards are scientifically valid and not simply balanced to match observations over a limited, unspecified, range of conditions. References BS 7963 (2000) Ergonomics of the thermal environment - Guide to the assessment of heat strain of workers wearing personal protective equipment. London: British Standards Institution. Havenith G (1999) Heat balance when wearing protective clothing. Annals of Occupational Hygiene, 43(5), 289-296. ISO 9920 (1995) Ergonomics of the thermal environment - Estimation of the thermal insulation and evaporative resistance of a clothing ensemble. Geneva: International Standards Organisation. Pitts DR & Sissom LE (1977) Schaum’s outline of theory and problems of heat transfer. USA: McGraw- Hill Book Company. 167 The influence of the number of thermal layers on the clothing insulation of a cold-protective ensemble Désirée Gavhed, Kalev Kuklane, Ingvar Holmér Programme for Respiratory Health and Climate, National Institute for Working Life, S - 112 79 Stockholm, Sweden Introduction Clothing is essential for wellbeing and performance in cold climate. The thermal proper- ties of the clothing are especially important. The required thermal insulation in a specific environmental condition can be predicted by use of the IREQ (Insulation REQuired) method, described in ISO document ISO/TR-11079, Evaluation of cold environments - Determination of required clothing insulation (IREQ) (ISO/TR-11079, 1993). The method is used for assessment of cold work environments and to provide recommenda- tions of clothing insulation for specific cold exposures. If possible, work clothing with an insulation value that is equivalent to the IREQ value is of course preferred, both from the employers and the employee's view. However, sometimes the predicted insulation value, IREQ, is not achieved by the actual work clothing. Then, a time limit for the cold expo- sure (DLE, Duration Limited Exposure) may be calculated to prevent the worker to cool unacceptably. The thermal insulation of a clothing ensemble can be measured with a thermal manikin (ENV-342, 1998) or estimated according to the international ergonomics standard ISO 9920, "Ergonomics of the thermal environment - Estimation of the thermal insulation and evaporative resistance of clothing ensemble" (ISO-9920, 1993). Thermal manikin measurements of clothing insulation are relatively costly and are limited to certain test laboratories. To compose a clothing ensemble with appropriate insulation for certain conditions an alternative method is to use tables of garments in the standard ISO 9920, annex 2. The estimation method is discussed in another paper presented in this conference (Gavhed et al., 2000). The standard includes a number of tables, which provide the thermal insulation of clothing ensembles with specified garments, as well as the insulation of single garments. Additionally, the type and weight of the ensembles and garments are described in the tables. The basic thermal insulation of single garments (I clu) can be summated to form the basic insulation (I cl) of a whole clothing ensemble. The values of Iclu and Icl in the tables ISO 9920 are obtained from measurements on thermal manikins. In cold environments more than one layer of thermal underwear is commonly used. Thermal underwear is usually made out of knitted fabrics. Numerous pockets of air (air is a very good insulator) are trapped between the fibres in the fabric. In addition to reduc- tion of the air layers between clothing layers, the fibres in knitted garments may be com- pressed when worn in tight fitting multiple layers. If compressed, the air volume inside the knit would be reduced and, concurrently, the insulation. Addition of thermal layers to an ensemble is therefore not likely to have a simple additive effect on the thermal insula- tion of the whole ensemble, i.e. the sum of I clu of single garments may differ from the measured basic insulation, I cl. The influence of the number of clothing layers on the in- sulation of clothing ensembles has not been systematically investigated. 168 The purpose of the study was to investigate the resulting thermal insulation of clothing ensembles with different numbers of thermal underwear and the relative contribution of the underwear to the insulation of a whole cold-protection clothing ensemble. Methods The total insulation (clothing + boundary air layer insulation) of different combinations of garments was measured on a standing thermal manikin. The basic insulation (I cl, the insulation of the clothing and the intrinsic air layers without the outer surface air layer insulation) was calculated from the total insulation according to ENV 342 (ENV 342, 1998). Seven pieces of clothing, sweaters, pants and coveralls were combined in different numbers of layers (1-6 layers). The sweaters and pants were manufactured from Ullfrotte Original™ (Ullfrotte AB, Sweden) "U", a terry material which is knitted from wool and synthetic fibres. Three qualities of Ullfrotte™ were tested (Table 1). The garments were thermal sweaters with turtle neck and long back (fabric qualities U2, U4 and U6, Table 1), thermal pants (U2 and U4, Table 1), a coverall with thin polyester lining (CLI) and a heavy insulated coverall with polyester filling (CHI). The outer layer of the coveralls had low air permeability. One to five thermal sweaters/jackets and one to two thermal pants combinations were measured alone, and combined with CHI and CLI, respectively. All garments were measured with gloves and double pairs of thick socks. The air velocity was < 0.3 m/s during the measurements. Table 1. Description of the fabric materials. Code Mass/area (g/m2) Fibre materials U2 200 60 % wool, 25 % polyester, 15 % polyamide U4 400 70 % wool, 18 % polyamide, 12 % aramide U6 600 70 % wool, 18 % polyamide, 12 % aramide Results and discussion The measured basic insulation values (I cl) of the ensembles are given in Figure 1. The re- sulting I cl of two layers of knitted underwear was not purely the sum of Icl of the two knitted clothing layers measured and worn as single layers. The sum of I cl for U2 sweater/ U2 pants and U4 sweater /U4 pants was 0.29 clo higher than I cl of the combination of U2+U4 sweaters/U2+U4 pants, which corresponded to 22 % of the insulation of the single layers added together (Figure 1). 1 clo = 0.155 m2·°C/W. Similarly, the sum of I cl for U6 sweaters /U4 pants and U2+U4 sweaters/U2+U4 pants was 0.48 clo higher than I cl of the combination of U2+U4+U6 sweaters /U2+U4 pants, which corresponded to 31 % of the insulation of the single layers added together (Figure 1). Further, combination U2+U4+U6 sweaters /U2+U4 pants insulated even less than the sum of I cl for U2 sweater/U2 pants and U4 sweater/U4 pants, although one more sweater was worn in the combination. Havenith et al reported 15 % reduction of the insulation of two layers when worn under a coverall in comparison to worn without coverall (Havenith et al., 1990) I cl of the ensemble with U4 sweater+pants and the ensemble with U6 sweater/U4 pants as single layer were similar (Figure 1). The weight of U6 was higher than U4, but the thickness did not differ much (about 0.5 mm). Thus, probably about the same amount of air was trapped in both fabric qualities. The thickness is correlated with insulation, but is certainly not the only determinant for insulation. 169 2.76 2.47 2.42 2.28 2.15 2.10 2.26 1.94 1.77 1.63 1.40 1.67 1.29 1.06 0.81 0.82 0.53 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 2+4+4+6+6/2+4+4 2+4+6/2+4 2+4/2+4 6/4 4/4 2/2 Co m bi n a tio n Insulation (clo) No coverall CLI CHI Figure 1. Insulation in different combinations of thermal underwear and coveralls together with standard gloves and socks. For explanation of the codes, see Methods. The code before the slash in the y-axis labels denotes the tested quality of the sweater and the code after the slash denotes the tested quality of the pants. A third layer worn under CHI on the upper body barely contributed to I cl, only by 0.05 clo, compared with two layers. A corresponding relevant table value would be about 0.30 clo. Under CLI, the contribution to I cl was slightly larger, 0.14 clo (Figure 1). The small contribution to I cl of the ensemble was probably because the clothing and air layers under the shell were probably slightly compressed especially when three layers were worn. The compression was probably larger in the thick coverall. The coveralls had normal fit for the manikin body and the knitted garments were of the same size, fitting the manikin body. However, as the number of layers increased, the garments, which were put on top of other layers stretched more. In addition to compression, the stretching may have re- duced the insulation of the garment in comparison to being the innermost layer. During standing the convection along the skin and clothing surfaces inside the outer shell was rather limited. However, when the body moves, the clothing is ventilated and air flows through the openings of the clothing, known as the "pumping effect". Air may also be exchanged through air permeable fabrics. Pumping during walking and wind re- duces the insulation of a multi-layer clothing ensemble by up to 70 % (Havenith et al., 1990, Nilsson et al., 1998). The influence by the number of clothing layers on insulation is most likely otherwise during walking (Havenith et al., 1990) and wind than during standing. Therefore studies of the influence of air velocity and body movements on the insulation of the above combinations are planned. The effect on the insulation of clothing ensembles with multiple layers of compression and stretching of fabrics has obviously not been taken into consideration in the standard method, probably partly because most ensembles in the tables are indoor clothing en- sembles. Thus, the insulation of multi-layer clothing ensembles risks to be overestimated by summation for standing (Gavhed et al., 2000). The estimated value of I cl is for ex- ample used for calculations of Duration Limited Exposure, DLE (ISO/TR-11079, 1993). The assessment of DLE, and other climate indices may thus be inaccurate if summated table values are used without correction. For example, according to IREQ, the clothing 170 ensemble with U2+U4+U6 sweaters /U2+U4 pants and CHI (I cl 2.47 clo), would not al- low thermal comfort at standing for eight hours below +8 °C. At -10 °C, the DLE for maximal allowed cooling in this clothing is 85 min. If the I cl of the single layers were summated, I cl of the combined ensemble would be overestimated by at least 0.48 clo ac- cording to this study. The corresponding DLE would then be one hour longer, 145 min. To improve the insulation of an ensemble by adding a layer, the garments must be large enough. Otherwise, the garments may be compressed and stretched. Then, the in- sulation will not increase proportionally to the insulation of added garments. Further, dif- ferent fibres and knitting structures may collapse under pressure easier than others. The insulation would probably reduce more at compression of thermal underwear made from soft fibres and knits with high loft. Conclusions The basic insulation of knitted wear combined in layers were 22-31 % lower than I cl of the sum of the garments in single layer. The insulation values of multi-layered clothing ensembles of this kind would be over- estimated by summation of the I clu of individual garments. Overestimation leads to wrong recommendations of cold-protective clothing or working time limits, at least during standing, but probably also during physical activity. Acknowledgements The authors thank Ullfrotte AB, Östersund, Sweden and TAIGA, Varberg, Sweden for providing garments for the study. References ENV-342 (1998) Protective clothing - Ensembles for protection against cold. Brussells: CEN. Gavhed D, Kuklane K, Holmér I (2000) Thermal insulation of multi-layer clothing ensembles measured on a thermal manikin and estimated by six individuals using the summation method in ISO 9920. In: Proceedings of the NOKOBETEF 6 and 1st European Conference on Protective clothing, Stockholm, May 7-10, 2000. Havenith G, Heus R, Lotens WA (1990) Resultant clothing insulation: a function of body movement, posture, wind, clothing fit and ensemble thickness. Ergonomics, 33(1), 67-84. ISO-9920 (1993) Ergonomics - Estimation of the thermal characteristics of a clothing ensemble. Geneva: International Standards Organisation. ISO/TR-11079 (1993) Evaluation of cold environments - Determination of required clothing insulation (IREQ). Geneva: International Standards Organisation. Nilsson H, Holmér I, Ohlsson G, Anttonen H (1998) Clothing Insulation at High Wind Speeds. In: Holmér I & Kuklane K eds. Problems with cold work. Arbete och Hälsa 1998:18. Pp 114-117, Stockholm: National Institute for Working Life. 171 Thermal insulation of multi-layer clothing ensembles measured on a thermal manikin and estimated by six individuals using the summation method in ISO 9920 Désirée Gavhed, Kalev Kuklane, Ingvar Holmér Programme for Respiratory Health and Climate, National Institute for Working Life, S-112 79 Stockholm, Sweden Introduction The thermal insulation of a clothing ensemble can be measured with a thermal manikin (ENV-342, 1998) or estimated according to the international ergonomics standard ISO 9920, "Ergonomics of the thermal environment - Estimation of the thermal insulation and evaporative resistance of clothing ensemble" (ISO-9920, 1993). Estimation is commonly used, since thermal manikin measurements are more costly and time consuming. In the standard there are tables of thermal insulation both for a number of specific clothing en- sembles (annex 1) and separate tables for single garments (annex 2). The tables in annex 2 contain information about the insulation of the garments, fabrics of the garment, type of garment and weight. The insulation values of single garments are summated to provide an estimate of the insulation of a whole clothing ensemble. The thermal insulation of clothing is used in assessment of the thermal environment of workers. The measurement of clothing insulation is restricted to certain test laboratories, why estimation is the most used method at thermal environment assessments. Estimates of the insulation of a clothing ensemble by summation of garments may naturally differ between individuals. Deviations from the actual value are probably higher when more garments are included in the ensemble. The tables are based on measurements with thermal manikins mainly before 1985 (McCullough et al., 1985, Olesen and Nielsen, 1983). Most garments in ISO 9920 are sleepwear, daily wear or industrial work wear. Some types of clothing are not well repre- sented, such as heavy insulated outdoor wear, boots, gloves and headgear. A large part of the measurements on garments were originally done as a complement for assessment of the thermal indoor environment (ISO-EN-7730, 1995) and thus rather few garments in the tables are for cold protection. Further, new materials have been developed since the tables were completed. For the practitioner it can be difficult to estimate the insulation of clothing ensembles comprising pieces of clothing or types of garments that are not found in the tables. The purpose of this study was to investigate i) the error of the estimate of the insulation of cold-protective clothing ensembles comprising two or more clothing layers and ii) the interrater reliability of the estimate. Methods 21 pieces of clothing (two coveralls, sweaters, thermal pants, t-shirt, socks, gloves, hood and helmet) were combined into seven clothing ensembles (Table 1). The ensembles were used at mast work in Sweden in winter. The thermal insulation of the ensembles 172 was measured with a standing thermal manikin at < 0.2 m/s. The basic insulation was calculated from the total insulation (clothing + boundary air layer insulation) according to ENV 342 (ENV 342, 1998). Six individuals ("raters") were instructed how to use the tables in ISO 9920, annex 2. They were asked to estimate the basic insulation of the gar- ments (I clu), which constituted the measured clothing ensembles. The basic insulation of the whole ensemble was estimated by summation of the I clu of the single garments ac- cording to the ISO standard. Table 1. Description of the garments and the seven clothing ensembles measured. Besides the garments in the table, all clothing ensembles comprised cotton briefs, gloves, thick boot socks, hood and safety helmet. UllfrotteTM terry knit (Ullfrotte AB, Sweden) "thin": 60 % wool/ 40 % synthetic fibres, "medium" and "thick": 70 % wool/ 30 % synthetic fibres. Garment, type Material Clothing ensemble A B C D E F G Long-legged underpants Ullfrotte, thin x x x x x Long-legged underpants Ullfrotte, medium thickness x x Thermal underpants Fibre pile x Thermal undershirt Ullfrotte, thin x x x x Thermal undershirt Ullfrotte, medium thickness x x T-shirt Cotton x x Sweater Cotton x x x x Jacket Fibre pile x Coverall, outdoor Multi-component x x x Coverall, heavy insulated Multi-component with filling x x x x Socks Ullfrotte, thin x x x x x Socks Ullfrotte, medium thickness x x x x x Socks Ullfrotte, thick x Socks Synthetic x x Socks Fibre pile x Results and discussion Five out of six raters underestimated the basic insulation (I cl) of most clothing ensembles. The maximal deviation of the estimated basic thermal insulation (I clEST) from the measured basic thermal insulation (I clMEAS) was 67 % (underestimation). IclMEAS of four ensembles was underestimated by on average 15 %, one ensemble was overesti- mated by on average 10 % and the average I clEST of two ensembles were similar to I clMEAS (Table 2). One rater systematically overestimated IclMEAS. In contrast to the other raters, this rater used the fabric of the garment as a determinator for the insulation estimate. Body surface area covered and fabric thickness are the most important determi- nators for thermal insulation of garments (Afanasieva, 1977), but not fabric material (McCullough et al., 1985). I clEST by the two raters without experience deviated more from I clMEAS (average difference 0.5 clo) than estimates done by the four more expe- rienced raters (< 0.3 clo). McCullough et al (1985) concluded that the estimates of the insulation of clothing ensembles made by a group of educated raters were unacceptably inaccurate. 173 Table 2. Measured, I clMEAS, and estimated basic insulation, IclEST, by six raters (1 clo=0.155 m2°C/W). U="UllfrotteTM terry knit, C=cotton, F=fibre pile, HIC=heavy insulated coverall, NFC=coverall without filling. SD=standard deviation of the mean. Ensemble code Layers on the torso, Material/ type Layers on the lower body, Material/ type I MEAS (clo) I clEST average (clo) SD Minimum Maximum A 3, U,U,HIC 2 U,HIC 2.35 2.08 0.53 1.55 3.07 B 3, U,C,HIC 2 U,HIC 2.35 2.00 0.45 1.55 2.86 C 2, U,HIC 2 U,HIC 2.34 1.79 0.48 1.25 2.69 D 4, C,U,U,HIC 2 U,HIC 2.35 2.19 0.48 1.75 3.13 E 3, U,U,NFC 3 U,U,NFC 1.74 1.72 0.18 1.41 1.94 F 3, C,U,NFC 2 U,NFC 1.70 1.73 0.23 1.37 2.07 G 3, C,F,NFC 2 F,NFC 1.87 2.06 0.24 1.68 2.28 The measured basic insulation may have been somewhat underestimated since the area surface enlargement factor (f cl) was not taken into account. According to equations pre- sented by McCullough et al (McCullough et al., 1985) f cl may be 1.7 for multilayer en- sembles. The estimated value for ensembles A-F would then be even more distant from I clMEAS, while IclEST for ensemble G would be closer to IclMEAS. The insulation of the ensemble with only one layer on the upper and lower body be- neath HIC (ensemble C) seemed to be most difficult to estimate correctly. This may partly be explained by the properties of the clothing ensemble. HIC was heavy insulated, the outer material was made of tight weave and the air permeability was low. About the same amount of air was probably trapped inside the outer shell regardless of the clothing beneath this shell. The air contributed to the total intrinsic (basic) insulation. The insu- lating air layers between the clothing layers beneath the outer shell are not accounted for in the estimation method. When there were more layers than one beneath the shell, the insulation of these garments was added to the insulation of the whole ensemble. The in- sulation of the air, which was not taken into account in the estimate, was then "replaced" by the insulation of the garments. Similarly, McCullough, et al (1985) observed that the number of garments, but also the type of clothing influenced the accuracy of the estimate. An additional explanation of the discrepancy between I clMEAS and IclEST is that a simi- lar coverall as the one in the study could not be found among the tables. The smaller garments (helmet, gloves, socks and hood) gave the highest variability among raters, coefficient of variation 23-73 %. This was probably due to the lower ab- solute insulation value, which made a small over- or underestimation of the absolute value relatively large, and to the fact that I clu of helmets and hood were not found in the tables, while the gloves were only represented by one type and value. Further, the I clu of the insulated coverall showed an average deviation of 59 % of the mean. The large varia- tion was mainly due to the same reason as for the smaller garments, that a similar cover- all to the one used in the study could not be found in the tables of ISO 9920. The raters commented that the tables were not well organised and that many garments were lacking in the tables, e g heavy insulated coverall and boots. I clMEAS was measured at standing at low air velocity. The estimate of insulation is adequate for this condition. However, with body movements, air will move beneath the clothing and through openings (pumping effect), which causes the insulation to decrease. The reduction of thermal insulation depends on factors such as the air permeability of the garment and openings of the clothing. The thermal insulation of a multi-layer clothing ensemble in dry conditions can be reduced by about 30 % during walking (Nilsson et al., 1992). The relationship between the number of layers, type of clothing material and in- 174 sulation needs further study to provide better estimations of insulation of multi-layer clothing ensembles. The estimated value of I cl derived from the tables in the standard ISO 9920 is, for ex- ample, used for calculations of Duration Limited Exposure, DLE (ISO/TR-11079, 1993) to limit work time if the cold protection is not sufficient for the conditions at work. The estimates with the highest error in this study (+0.72 clo and -1.1 clo, respectively), would render a calculated DLE for standing at -10 °C in calm air, which is about 40 minutes longer and about 30 min shorter than the actual DLE, respectively. These values cor- respond to 50-90 % of the DLE based on I clMEAS. Further, the underestimation by 1.1 clo would limit work to less than an hour at the same conditions according to DLE, while the working time with I clMEAS would be thermally comfortable for more than eight hours. Conclusions The estimations of insulation were more accurate for the ensembles with lower insulation with only one layer beneath the outer shell than with higher insulation and multiple layers. The raters considered the tables to be complicated to use. Experience to do insu- lation estimations tended to improve the estimate. Acknowledgements The authors thank Ullfrotte AB, Östersund, Sweden and TAIGA, Varberg, Sweden for providing garments for the study. References Afanasieva RF (1977) Hygienic basis for designing cold protective clothing. Moscow: Legkaja Industria. ENV-342 (1998) Protective clothing - Ensembles for protection against cold. Brussells: CEN. ISO-9920 (1993) Ergonomics - Estimation of the thermal characteristics of a clothing ensemble. Geneva: International Standards Organisation. ISO-EN-7730 (1995) Moderate thermal environments - Determination of the PMV and PPD indices and specification of the conditions for thermal comfort (pp. 19). Geneva: International Standards Organisation. ISO/TR-11079 (1993) Evaluation of cold environments - Determination of required clothing insulation (IREQ) (pp. 31). Geneva: International Standards Organisation. McCullough EA, Jones BW, Huck J (1985) A comprehensive data base for estimating clothing insulation. ASHRAE Trans, 91 part 2A, 29-47. Nilsson HO, Gavhed DCE, Holmér I (1992) Effect of step rate on clothing insulation-measurement with a moveable thermal manikin. The Fifth International Conference on Environmental Ergonomics, Maastricht. TNO Institute for Perception, Soesterberg, The Netherlands, 1992: 174-175. Olesen BW, Nielsen R (1983) Thermal insulation of clothing measured on a movable thermal manikin and on human subjects. ECSC Programme Research No 7206/00/914, Copenhagen: Technical University of Denmark. 175 Effect of the number, thickness and washing of socks on the thermal insulation of feet Kalev Kuklane, Désirée Gavhed, Ingvar Holmér National Institute for Working Life, Respiratory health and climate, Solna, Sweden Introduction An effective way to increase the insulation on feet is to add socks. However, the effect of adding new layers on feet has not been studied enough. There is limited data on insula- tion of socks and their combination with footwear. In this study 4 types of socks with dif- ferent fabric thickness and 3 types of footwear were tested on a thermal foot model to study the effect of the number, thickness and washing on the thermal insulation of feet. Methods Four types of socks, manufactured by Ullfrotté (Table 1) were tested separately and in combination with the others. Some socks were tested in combination with 3 types of footwear (Table 2). The cotton sock, and boots AS and WS were used in thermal foot measurements earlier (Kuklane et al., 1997; Kuklane et al., 1998; Kuklane et al., 1999b; Kuklane et al., 1999c). Table 1. Sock types, measured on thermal foot model. Sock Manufacturer Material Material weight (g/m2) Weight of a sock (g) Cotton 70 % cotton, 30 % polyester 20 2 Ullfrotté 60 % wool, 40 % polyamide 200 21 4 Ullfrotté 65 % wool, 35 % polyamide 400 39 6 Ullfrotté 60 % wool, 40 % polyamide 600 53 8 Ullfrotté 85 % wool, 15 % polyamide 800 77 Table 2. Boot types, measured on thermal foot model. Boot Manufacturer Material Weight (g) AS Arbesko, Sweden Leather 801 SS Steitz Secura, Germany GoreTex 827 WS Arbesko, Sweden Impregnated leather, Thinsulate, nylon fur 815 The thermal foot model has 8 zones (toes, sole, heel, mid-foot, ankle, lower calf, mid- calf and a guard zone) that are controlled separately by a computer program. At a con- stant ambient temperature (in this study 6.00.5 C, air velocity <0.2 m/s) a certain power is needed to keep the foot surface temperature constant (in this study 34.00.1 C). The supplied power is directly related to the heat losses from the foot. From the tem- perature gradient and the heat losses it is possible to calculate the insulation for each zone or whole foot. The insulation was defined as the total insulation of the foot zones (toes, sole, heel, mid-foot) and ankle zone. All insulation values, given in the paper, include the air layer 176 insulation. The air layer insulation (0.108 m2C/W) was measured earlier (Kuklane et al., 1999a). Results and discussion There is a clear difference between socks with various material thickness/weight (Figure 1). By combining different socks to the same material weight the effect of layers becomes obvious (Figure 2). The effect of layers is even clearer if one compares Figures 1 and 3. Adding one layer of thin sock to each sock type increases more noticeably the insulation of thinner socks (Figure 4) and the difference between the insulation provided by diffe- rent sock types decreases (Figures 1 and 4). 0.000 0.050 0.100 0.150 0.200 0.250 0.300 Cotton 2 4 6 8 Socks In su la tio n (m 2 ° C/ W ) Figure 1. Insulation of single socks. 0.000 0.050 0.100 0.150 0.200 0.250 0.300 8 6+2 4+4 2+2+2+2 Socks In su la tio n (m 2 ° C/ W ) Figure 2. Insulation of sock combinations giving the total material weight 800 g/m2. 0.000 0.050 0.100 0.150 0.200 0.250 0.300 2+2 2+4 2+6 2+8 8+2 Socks In su la tio n (m 2 ° C/ W ) Figure 4. Effect of adding one thin sock to every type and the effect of having it inside or outside. 0.000 0.050 0.100 0.150 0.200 0.250 0.300 2 2+2 2+2+2 2+2+2+2 Socks In su la tio n (m 2 ° C/ W ) Figure 3. Effect of sock layers on the insula- tion on the example of the thinnest sock. 0.000 0.050 0.100 0.150 0.200 0.250 0.300 2 (new, 21 g) 4 (new, 39 g) 4 (used, 37 g) 4 (used, 28 g) 4 (used, 18 g) Socks In su la tio n (m 2 ° C/ W ) Figure 6. The effect of wear. 0.000 0.050 0.100 0.150 0.200 0.250 0.300 4 new 4 used, not washed 4 used, washed 8 new 8 used, not washed 8 used, washed Socks In su la tio n (m 2 ° C/ W ) Figure 5. Effect of the use and washing. 177 Figure 5 shows the effect of use and washing. The slightly higher insulation of washed socks could be related to the properties of wool fibres that become more loose with washing thus increasing material thickness and trapping more air. With use, dirt and compression diminish this effect. Such behaviour of wool fibres could also explain, why a cotton sock of the same weight has slightly lower insulation than a woollen one (Figure 1). Figure 6 shows, how the insulation reduces with the loss of material on the example of a sock material of 400 g/m2 and in comparison with a sock material of 200 g/m2. A used sock had higher insulation than a new one after washing. This could be related to the loosening of the wool fibres. When the material had become more worn out, the insula- tion decreased. However, when the sock had lost the loose wool fibres and the material had reduced to the weight of a thinner sock, then the insulation had decreased accordingly. Similar effect of layers can also be observed in the case of footwear in Figure 7. It shows clearly that higher insulation of the footwear is related to lower gain of insulation corresponding to the added socks. It means that the well insulated footwear traps air rela- tively well and adding socks could even press out some air. However, the sock layers would gain importance during sweating as moisture transport and absorption media, thus contributing to better comfort. This question should be studied further. 0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 No sock Cotton 2 2+cotton cotton+2 8+2 Socks In su la tio n (m 2 ° C/ W ) AS SS WS Figure 7. Insulation of the footwear without and in the combination with socks. Conclusions  It is more effective to use several layers for higher insulation.  Thin layers give relatively higher insulation gain than thick socks with the same num- ber of layers.  Washing increases the insulation of woollen socks to a certain extent. However, each use and washing reduces the amount of material in the sock and in long run the insu- lation will decrease.  Footwear insulation increases with added socks. The effect of insulation gain is bigger in footwear with lower insulation. Acknowledgements Thanks to Ullfrotté AB for providing the socks. 178 References Kuklane K, Afanasieva R, Burmistrova O, Bessonova N & Holmér I (1999a) Determination of heat loss from the feet and insulation of the footwear. International Journal of Occupational Safety and Ergonomics, 5(4), 465-476. Kuklane K & Holmér I (1997) Reduction of footwear insulation due to walking and sweating: a preliminary study. In: Holmér I & Kuklane K eds. Problems with cold work. Vol. Arbete och Hälsa 1998:18. Pp 96-98, Stockholm: National Institute for Working Life. Kuklane K & Holmér I (1998) Effect of sweating on insulation of footwear. International Journal of Occupational Safety and Ergonomics, 4(2), 123-136. Kuklane K, Holmér I & Giesbrecht G (1999b) Change of footwear insulation at various sweating rates. Applied Human Science, 18(5), 161-168. Kuklane K, Holmér I & Giesbrecht G (1999c) One week sweating simulation test with a thermal foot model. In: Nilsson HO & Holmér I eds. Proceedings of the Third International Meeting on Thermal Manikin Testing, 3IMM. Vol. Arbete och Hälsa 2000:4. Pp 106-113, National Institute for Working Life, October 12 - 13, 1999, Stockholm, Sweden. 179 Methods for cold protective clothing evaluation Håkan O. Nilsson1, Hannu Anttonen2, Ingvar Holmér1 1 Programme for Respiratory Health and Climate, National Institute for Working Life, S - 171 84 Solna, Sweden 2 Oulu Regional Institute of Occupational Health, Aapistie 1, F - 902 20 Oulu, Finland Introduction For evaluation of the protection requirements in cold environments it is suggested that thermal insulation values from a moving thermal manikin (ENV 342) can be used to match requirements specified by the IREQ-method (ISO/TR-11079, 1993). It is known from prior work (Olesen et al, 1982; McCullough & Hong 1992; Nilsson et al, 1997) that the clothing insulation can be reduced by wind and body movements by up to 70 % from the value measured on a standing thermal manikin. This paper deals with the assessment of the effects of wind and walking movements on insulation reductions of clothing and how it can be used for evaluation of work clothing with a modified IREQ-method. Figure 1. After ten years' construction work, the Øresund Bridge will open to traffic on Saturday, July 1 2000. Materials and methods The thermal manikin used is one in the TORE-series that has been described earlier (Hänel, 1983; Nilsson et al, 1992). The power transmission, in the walking apparatus, has been made with pneumatic cylinders, which gives a simple and durable construction with a minimum of mechanical components. The manikin surface sensors and the air tem- perature sensors were calibrated prior to the tests (ASL F25, Uncertainty: ± 0.015 °C, Pentronic, Sweden). Three different types of cold protective clothing were investigated. They will be used by the Swedish police force on the Øresund bridge between Sweden and Denmark. The ensembles have a total insulation of up to 5.0 clo (ENV 342:1998) and an air 180 permeability of 8 l/m2s. The ensembles comprised (A) an outer layer consisting of an insulative jacket and trousers, (B) an insulative overall and (C) jacket and trousers on top of the overall. The walking speed of the manikin was set to 0 (standing) and 0.8 m/s at an average wind speed, over a cross section of the wind tunnel in front of the manikin, of 0.38 ± 0.07 m/s (Low Velocity Transducer 54R10, Uncertainty: ± 0.02 m/s, Dantec, Denmark). Figure 2. Pictures of TORE dressed in the different clothing combinations used in the investiga- tion (A to the left and C to the right). A. Winter overall Icl = 0.344 m2K/W B. Jacket and trousers Icl = 0.378 m2K/W C. Overall as well as jacket and trousers on top. Icl = 0.458 m2K/W Insulation reduction (It,r /It) as a function of air permeability (p, l/m2s), wind speed (v, m/s) and walking speed (w, m/s) is calculated with (Nilsson et. al, 2000): I t, r I t  0.54 e 0.15˜ v0.22˜w p0.075  0.06 ln(p)  0.5 (1) The equation for air layer reduction is produced from (1) by inserting p = 10000 l/m2s and I a = 0.085 m2K/W. The equation then becomes: Ia, r  0.092 e  0.15 ˜ v  0.22 ˜ w  0.0045 (2) Using the relationship below from ENV 342, equation 4 is derived. In all the following calculations f cl = 1 is used in consequence of ENV 342, although this can be modified in the future: I cl, r  I t, r  Ia, r (A.3) ENV342 (3) This way of calculating the basic clothing insulation by not taking into account the clothing area factor is questioned. The increase in clothing area is in the region of 30 to 70 % for heavy winter clothing. The value of I cl.r will be underestimated. I cl  I cl ,r  (0.092 e 0.15˜v 0.22˜w  0.0045) (0.54 e 0.15˜v 0.22 ˜w p0.075  0.06 ln(p)  0.5)  0.085 (4) h c  1 I a  h r (5) The reduction of the air layer is calculated with equation 2, the Icl,r /Icl correction is now calculated with equation 4. The convective heat transfer coefficient is produced with 181 equation 5. The equations 2, 4 and 5 replace corresponding equations in ISO TR 11079 for the calculation of IREQ. Results The results of the analysis of 8 different conditions on the bridge are shown in Table 1. Weather information, provided by The Øresund Link Consortium, shows that the tem- perature at Drogden very seldom is below -5 °C during the coldest month. Wind speed data from the same location show values that very seldom exceeds 15 m/s. Relative hu- midity most of the time is around 80 to 90 %. Two metabolic rates are used for the cal- culations (EN 28 996:1993) - 100 W/m2 (Low metabolic rate) and 165 W/m2 (Moderate metabolic rate). Table 1. IREQ calculations with wind, permeability and walk reductions made with the new equations. Low metabolic rate Moderate metabolic rate M (W/m2) 100 100 100 100 165 165 165 165 t a (°C) 5 5 -5 -5 5 5 -5 -5 t r (°C) 5 5 -5 -5 5 5 -5 -5 p (l/m2s) 8 8 8 8 8 8 8 8 w (m/s) 0 0 0 0 0.8 0.8 0.8 0.8 v (m/s) 5.0 15.0 5.0 15.0 5.0 15.0 5.0 15.0 rh % 85 85 85 85 85 85 85 85 I cl- available 0.458 0.458 0.458 0.458 0.458 0.458 0.458 0.458 IREQ (m2K/W) 0.298 0.332 0.416 0.451 0.168 0.197 0.241 0.269 I cl- needed 0.415 0.679 0.591 0.946 0.235 0.380 0.351 0.548 DLE (hours) >8.0 1.1 1.8 0.5 >8.0 >8.0 >8.0 1.7 The low metabolic rate is presumed to correspond to a standing activity and the moderate metabolic rate to conditions with light movements. The corrections in IREQ (neutral, I cl,r) are now depending on the prevailing wind, permeability and walking conditions. Discussion The clothing insulation is strongly affected by wind and to some extent by body move- ments. The combined effect of body movements and wind increases the heat loss from the human body. A reduction of the clothing is consequently needed. At the most severe situation in table 1 (-5 °C and 15 m/s) the evaluation suggests a work duration (exposure time) for half an hour - if standing still. The time limit prevents body cooling to unacceptably low levels. With the higher activity this time can be in- creased to approximately one and a half hour. Another possibility is, of course, to in- crease the insulation at these extreme cases. On the other hand is it recommendable to decrease the insulation, e.g. by taking of the jacket, if the temperature goes up to 5 °C or if the wind decreases to 5 m/s. More investigations have to be made with the clothing area reduction factor f cl for mul- tilayer winter clothing at high wind speeds and during movements. Special consideration has to be made concerning the fact that the wind sometimes is compressing the clothing material and consequently reducing area increase. 182 Conclusions The new set of equations makes it possible to calculate the insulation reduction due to activity and wind for most winter work clothing, if the static clothing insulation is known from the manufacturer or manikin measurements. The analysis provides an overall assessment of the protective value of the three en- sembles, that can be used as a guideline for organisation and performance of work during extreme weather conditions. It is necessary to emphasise that the final adjustment of clothing must be made by the user according to his or her situation, requirements and preferences. To validate the relationships measurements on subjects exposed to real conditions on the bridge would be valuable. Acknowledgements The Swedish Council for Work Life Research and The Swedish National Police Board have supported this work. References EN 28 996:1993 (1993) Ergonomics - Determination of metabolic heat production. Comité Européen de Normalisation, Brussells. ENV 342:1998 (1998) Protective clothing against cold. Comité Européen de Normalisation, Brussells. ISO/TR-11079 (1993) Evaluation of cold environments - Determination of required clothing insulation (IREQ). International Standards Organisation, Geneva. Hänel S-E (1983) A joint Nordic project to develop an improved thermal manikin for modelling and measuring human heat exchange. In: Aspect médicaux et biphysiques des vêtements de protection. Lyon: Centre de Recherche du Service de Santé des Armées, 280-282. McCullough EA & Hong S (1992) A data base for determining the effect of walking on clothing insulation. Proceedings of the Fifth International Conference on Environmental Ergonomics, Maastricht, the Netherlands, 68-69. Nilsson H, Anttonen H, Holmér I (2000) New algorithms for prediction of wind effects on cold protective clothing. Kuklane K & Holmér I (ed.), Proceedings of 1th European Conference on Protective Clothing (NOKOBETEF6). Stockholm, Sweden. Nilsson HO, Gavhed DCE, Holmér I (1992) Effect of step rate on clothing insulation. – Measurement with a moveable thermal manikin. In: Lotens W, Havenith G (ed.), Environmental Ergonomics V, Masstricht, Netherlands. Nilsson H, Holmér I, Ohlsson G & Anttonen H (1997) Clothing insulation at high wind speeds. Proceedings of Problems with Cold Work, Holmér I & Kuklane K (ed.), Arbete och Hälsa 1998:18, 114-117. Olesen BW, Sliwinska E, Madsen TL & Fanger PO (1982) Effect of body posture and activity on the thermal insulation of clothing. Measurement by a movable thermal manikin. ASHRAE Transcations, 88:2, 791-805. 183 Research on typical medical work clothing on humans and on a thermal manikin Krzysztof Soltynski, Maria Konarska, Jerzy Pyryt, Andrzej Sobolewski Central Institute for Labour Protection, ul. Czerniakowska 16, 00-701 Warsaw, Poland Introduction Many models and indices for predicting comfort or thermal stress make use of the value of the insulation of the clothing worn in a specific situation. Normally, this insulation value is unknown and has to be determined usually on humans, but more often on ther- mal manikins. For the same clothing ensemble the insulation value determined on hu- mans was generally found to be lower than the corresponding value determined on a manikin, with no satisfactory explanation of the observed differences. Our ultimate ob- jective is to confirm and test this effect. The present work – thermal insulation measure- ments of medical work clothing using a thermal manikin compared to measurements on humans – is only the initial phase in the realisation of this task. We would also like to look at those results from the point of view of ergonomics and show how they can be used to determine the parameters of the working environment in which medical personnel in that clothing will be at thermal comfort. Methods Garments Three clothing ensembles were selected for this study: Ensemble A - shoes, socks; underwear, briefs and a shirt (cotton); surgical ensemble (cotton-like nonwoven with hydrophic viscose fibres, material square metre weight [sq]=65 g/m27, good permeability of air and water vapour). Ensemble B - Ensemble A plus a surgical apron (hygienic foil-covered two-layer non- woven made of fibres and polypropylene foil, material square metre weight [sq]=45 g/m22), not permeable to any liquids). Ensemble C - Ensemble A plus a surgical apron (thermoplastic two-layer hygienic nonwoven made of propylene fibres and hydrophilic viscose fibres, material square metre weight [sq]=35 g/m22, good permeability of air). The ensembles were chosen as typical medical work clothing, made according to WHO recommendations and the requirements of ISO 9001 and EN 49000. Thermal manikin Clothing insulation of the tested ensembles was determined using a thermal manikin type TM3 (made in cooperation with the Thermal Insulation Laboratory, Technical University of Denmark). The measurements were performed according to ISO 9920. 184 Posture and movements One posture and one movement type were tested: rest in standing posture. Wind One wind condition was tested: no wind (air speed  0.1 m/s). That is why all the ex- periments were performed in a climatic chamber. Experimental procedure Manikin - I a , IT test: The experiments were carried out in a climatic chamber in which air (T a ) and mean radiant temperature (T r ) were maintained to within 0.1 °C. Relative hu- midity was kept constant at 405 % and air velocity at 0.10.03 ms-1. Two series of ex- periments were performed. In series I (nude manikin), I a dependence was determined for various values of (T ms – T o ) and for various values of H. There was a total of 12 experimental sessions. Each session lasted about 4 hours. In series II (clothed manikin), IT insulation was determined for the clothing ensembles (A, B, C) selected for this study. During the experiments the operating temperature in the climatic chamber was kept constant at 23.50.1 °C. There was a total of 15 experimental sessions. Each session lasted about 4 hours. During this period all measurements were performed every second and recorded as minute averages. The measurements were also continuously displayed graphically, enabling an easy check of the steady situation. Humans – I a , IT test: Seven healthy male subjects (Table 1) volunteered for these experi- ments. The studies on humans are conducted for the same clothing ensembles (A, B, C), in exactly the same conditions as for the studies with the manikin (rest standing posture, practically identical thermal load conditions). Subjects began the test at a mean skin tem- perature of 33  0.5 C (comfort). Table 1. Physical characteristics of subjects Subjects Body weight(kg) Height (m) AD* (m2) Age (years) AS JT TT PR KK AK ML 70 65 71 67 77 81 80 1.72 1.79 1.72 1.73 1.74 1.79 1.77 1.82 1.82 1.83 1.79 1.91 1.99 1.97 23 27 29 23 23 26 22 Mean 73.0 1.75 1.87 25 *AD Hardy and DuBois Body area. A measurement session in the climatic chamber lasted for 120 minutes. There was a total of 28 experimental sessions. Maximal oxygen uptake in the subjects was calculated by the indirect method of Astrad-Ryhming. Metabolic heat production was calculated from oxygen consumption. Oxygen content and the respiratory quotient were measured with a Beckman analyser. Rectal temperature and mean skin temperature were measured with the accuracy of 0.15 C using standard thermocouples, and recorded on a 15-channel Ellab point recorder. 185 Mean skin temperature was calculated in accordance with the 8-point Hardy and DuBois formula; heart rate – by counting QRS of the EKG. The body and clothes were weighed before and after the experimental period (accuracy  5 g). Physical basic and derived parameters were measured according to ISO 7726. All these measurements and calculations were carried out every 5 min. Calculations Insulation (manikin): Total insulation, IT (in m2 °C/W), is the insulation from the surface of the manikin to the environment, including the effect of the increased surface area, f cl, and the resistance on the surface of the manikin, I a . where: T cms =mean manikin surface temperature, in °C; T o =operating temperature, in °C; H=heat loss from clothed thermal manikin, in W/m2. Surface air insulation (I a ) and effective clothing insulation (I cle) can be similarly calculated: Insulation (humans): During the relative steady states of body temperatures observed at the end of our experiments, (C+R) sk is determined from the heat balance equation: (C+R) sk = M – E – Eres – Cres – S, where M is metabolic rate, E is evaporative heat loss rate, E res , C res are evaporative and convective respiratory heat rates respectively, S is heat storage rate, all in W m-2. Total clothing insulation IT and effective insulation Icle (all in m2 °C/W) can be similarly calculated: where I a – resistance at the surface of the clothed body I a = 1/(h c + h r ) determined from experi- ments in “nude” conditions (briefs and shoes). The measurements and calculations were performed according to ISO 9920. Results from the point of view of ergonomics: Assuming typical activities, typical envi- ronmental parameters and knowing thermal clothing insulation (ensemble A and C) for surgeons in the operating theatre, operating temperatures at which medical personnel will be at thermal comfort were estimated on the basis of heat balance and comfort defined with the PMV index. Statistics: A statistical package STATGRAPHIC and analysis of variance were used to determine the effects of the parameters under investigation on insulation values. Two- Sample Analysis and ANOVA were used to test differences between separate levels of significant parameters. A significance level of =0.05 was accepted. Results and discussion The results of the manikin and humans measurements are presented in Table 2. The in- sulation of the surface air layer of the manikin/humans is 0.78 clo. H TTI ocmsT   H TTI onmsa   IH TTI aocmscle    SCEEM TTI resres osk T    ISCEEM TTI a resres osk cle     186 Table 2. Mean values and standard deviations of total (IT) and effective (Icle) thermal clothing insulation calculated for each experimental session (three ensembles measured on the standing manikin and standing subjects at air velocity d 0.1 m/s). The differences between the average values (subjects, manikin) for ensembles A and B are statistically significant (Two-Sample Analysis, s.l.=0.02 y 0.03). IT IcleEnsemble (m2K/W) (clo) (m2K/W) (clo) A 0.206 r 0.0010.25 r 0.03 1.33 r 0.01 1.6 r 0.2 0.084 r 0.001 0.13 r 0.03 0.54 r 0.01 0.8 r 0.2 B 0.218 r 0.0020.25 r 0.03 1.40 r 0.01 1.6 r 0.2 0.096 r 0.002 0.13 r 0.03 0.62 r 0.01 0.8 r 0.2 C 0.269 r 0.0010.28 r 0.05 1.73 r 0.01 1.8 r 0.3 0.147 r 0.001 0.16 r 0.05 0.95 r 0.01 1.0 r 0.3 Manikin The relative measurement error of clothing thermal insulation (IT) treated as an error of the complex value is approximately 3 %. The partial input of particular measured values is as follows: operating temperature measurement error (T o ) – 56 %, manikin surface temperature (T ms ) – 35 % and heat loss from manikin surface (H) – 9 %. The relative error assessed on the basis of the measurements does not exceed 2 %. Thus, the role of the cli- matic chamber is important for these measurements. The number of necessary measure- ments assessed on the basis of the average variance of results for estimate precision of 0.01 clo is n11, and for 0.02 clo it is n3. Therefore, for testing clothing in practice it is sufficient to repeat measurements three times for one tested ensemble. The differences between the average values for ensembles are statistically significant (ANOVA, F=1000). Humans The relative measurement error of clothing thermal insulation (IT) treated as an error of the complex value is approximately 6 %. The partial input of particular measured values is as follows: metabolic heat production error (M) – 52 %, skin temperature (T sk) – 24 %, operating temperature (T o ) –16 % and evaporative respiratory heat rate (E res ) – 6 %. The relative error assessed on the basis of the measurements changes from 12 % to18 %. The number of necessary measurements assessed on the basis of the average variance of re- sults for estimate precision of 0.02 clo is n600, and for 0.2 clo it is n8. That is why, in order to increase precision of the measurements of thermal insulation on humans, the precision of the measurements of metabolic rate and skin temperature should be in- creased, with the same number of measurements. The differences between the average values for ensembles are not statistically significant. Humans/manikin – comparisons To make a correct comparison possible, the conditions of the experiment were chosen such that a manikin, which cannot sweat or move and which is made to stand, was com- pared to a standing human at rest in thermoneutral conditions (a limited amount of ex- creted sweat). For this experimental condition for the same clothing ensemble, the insu- lation value determined on humans was generally found to be slightly higher than the corresponding value determined on the manikin (it was 13 %). This difference can addi- tionally increase after diffusion evaporation through the skin has been considered in cal- culating thermal insulation of the clothing. 187 Clothing insulation-point of view of ergonomics Ergonomic conditions (thermal comfort) for medical personnel is determined by limit values for operating temperature: Ensemble A (20 CT o 24 C); Ensemble C (16 CT o 20 C). Conclusions To summarise, it is risky to draw general conclusions from measurements on the three garments tested in one set of thermal load conditions of our experiment. However, the measurements show clearly that: 1. for the same clothing ensemble, the insulation value determined on humans was generally found to be slightly higher than the corresponding value determined on a manikin, 2. taking diffusion evaporation into consideration can additionally increase this difference, 3. a thermal manikin makes it possible to determine thermal clothing insulations a few times more precisely compared to measurements performed on humans, 4. greater precision in determining thermal clothing insulations on humans can be achieved by making measurements of oxygen consumption more precise and – to a lesser degree – by making measurements of skin temperature more precise. References Nishi Y Gonzalez RR & Gagge AP (1975) Direct measurement of clothing heat transfer properties during sensible and unsensible heat exchange with thermal environment. ASHRAE Transactions, 81, 183-199. Olesen B, Sliwinska EE, Madsen T, and Fanger PO (1982) Effect of body posture and activity on the insulation of clothing. Measurement by a movable thermal manikin. ASHRAE Transactions, 32, 791-805. Vogt JJ, Meyer JP, Candas V, Libert JP & Sagot JC (1983) Pumping effects on thermal insulation of clothing worn by human subjects. ERGONOMICS, vol 26 No 10, 963-974. Havenith G, Heus R, & Lotens WA (1990) Resultant clothing insulation: a function of body movement, posture, wind, clothing fit and ensemble thickness. ERGONOMICS, vol 33 No 1 67-84. Parsons KC (1988) Protective clothing: heat exchange and physiological objectives. ERGONOMICS, vol 31 No 7, 991-1007. 188 Comparative evaluation of the methods for determining thermal insulation of clothing ensemble on a manikin and person Ralemma F. Afanasieva1, Nina A. Bessonova1, Olga V. Burmistrova1, Vyacheslav M. Burmistrov1, Ingvar Holmér2, Kalev Kuklane2 1RAMS Institute of Occupational Health, Moscow, Russia 2National Institute for Working Life, Solna, Sweden Introduction According to the European standard ENV 342-97 (ENV 342, 1997) the heat-shielding parameter of an ensemble for protection against cold, called the resultant total thermal insulation (Itr, ºCm2/W) is determined with a heated physical model (manikin). This ade- quately reflects the conditions of heat exchange with the environment. According to the Russian standard GOST R12.4.185-99 (Standart R 12.4.185-99, 1999) thermal insulation of an ensemble for protection against cold is determined directly on a person. Undoubt- edly each of these methods has both advantages, and disadvantages (Higenbottam et al., 1997). A comparison of the methods would be useful from positions of mutual use of the data with the purposes of manufacturing clothes with proper heat-shielding parameters. Methods Four clothing ensembles (I-IV) were used for determination of their thermal insulation on a manikin (ENV 342, 1997) and on a person (Standart R 12.4.185-99, 1999). The ensem- bles included a jacket and trousers, having various thickness of thermal insulation. As top and lining the fabric "Pertex" was used and Thinsulate™ manufactured by the company 3M was used as thermal insulation. A volunteer (man in the age of 40 years, surface of a body is equal to 1.8 m2) and a manikin were dressed in a cotton linen, half-woolen sports costume, cotton and half- woolen socks, heated jacket and trousers, helmet, mittens, heated boots. Before the experiment according to (Standart R 12.4.185-99, 1999) the volunteer, wearing an ensemble of “indoor” clothes, sat within 30 minutes in a comfortable micro- climate: air temperature (t a ) 23.0±0.5 ºC, relative humidity 40±5 % and air velocity (V) less than 0.1 m/s. The sensors for measurement of “dry” heat flow (q, W/m2) and skin temperature (t s , ºC) (Nilsson, 1997) were placed on the skin surface to the following locations: forehead, chest, back, stomach, waist, shoulder, back of the hand, upper and bottom of a hip, shin, and back of the foot. The subject donned the clothing ensemble (jacket, trousers, helmet, mittens, footwear) and entered the climatic chamber. Air tem- perature was kept at 8.0±0.5 ºC. The value of t a was selected so as to provide the subject with comfort. The tests were repeated three times in a relatively calm air. The volunteer was standing in the climatic chamber for 60 minutes. The values of t s and q were registered with a pe- riodicity of 5 minutes. On the bases of local values of t s and q were calculated the average 189 weighted skin temperature (t aws ) and the "dry" heat flow (Afanasieva, 1977, Afanasieva, 1994). The thermal insulation of the ensembles, determined on the person, was calculated ac- cording to the equations 1-3 (Table 1). Table 1. The equations used for calculation of a thermal insulation of clothing ensembles, I e , ºC·m2/W, by results of tests on the person (1 - 3) and the manikin (4 - 5). Nr. Equation Note 1 I e = (t aws - t a ) /q aws Ÿ [( 6 fi · tsi) - ta] /6fi·qi Ÿ Ÿ [( 6 fi · tsi) - ta] · A/ 6Qci GOST 12.4.185-99 SSBT 2 I e = 6 fi · Ri Ÿ 6fi · [(tsi - ta) /qi] Ÿ Ÿ 6 fi · [(tsi - ta) · ai /Qci] {ºC·m2/W} ENV 342-1997 3 I e = 1/ 6 fi · Ki Ÿ 1/ 6fi · [qi / (tsi - ta)] GOST 12.4.185-99 SSBT 4 Itr = [(6fi · Tsi) - Ta] /6Hci Ÿ Itr = [(6fi · tsi - ta)] · A/6Qci {ºC·m2/W } PrEN 342: 1995 5 Itr = 6fi · [(Tsi -Ta) · ai/ Hci] {K·m2/W} Ÿ Ÿ Itr = 6fi · [(tsi - ta) · ai /Qci] {ºC·m2/W} ENV 342-1997 Where: fi - factor of area of a section of a manikin or human body part; fi = ai/A; ai - area of surface of a section of a manikin or human body part, m 2; A - common area of surface of a manikin or human body, m2; T si = tsi - local temperature of a surface of a section of a manikin or human body part, ºC; T a = t a – air temperature, ºC; H ci - local power of heating of a section of a manikin, W; qi - density of a “dry” heat flow of a site of a human body, W/ m 2; Q ci - heat flow of a site of a human body surface (manikin), W; Ki - heat transfer coefficient, W/ºC·m 2 . The change of a thermal insulation of the clothing ensemble under influence of wind and walking of the person was determined on the bases of the empirical equation ob- tained before (Afanasieva, 1977, Afanasieva, 1994) with reference to the given sort of clothing (jacket and trousers): C = (0.07 · B + 2.0) · V + 5, where: C - lowering the I s value, %; B - air permeability of a package of clothing materials, dm3/ m2·s; V - Wind speed, m/s. The above described clothing ensembles were investigated also according to (prEN 342, 1995; ENV 342, 1997) on a manikin. The conditions for the standing manikin were the following: air temperatures of 10.6±0.2 ºC or 5.1±0.3 ºC and air velocity <0.2 m/s, and at t a =4.1±0.1 ºC and V=1 m/s. The conditions for the moving manikin were the fol- lowing: walking 0.8 m/s (2.8 km/h), t a =4.8±0.1 ºC and V<0.2 m/s, and ta = 4.2 ºC and V=1 m/s. The calculation of the resultant thermal insulation (Itr) of the clothing ensemble by the results of the tests on the manikin were conducted according to the equations 4 (prEN 342, 1995) and 5 (prEN 342, 1995; ENV 342, 1997) (Table 1). The equations 1-3 and 4-5 are identical in their heat physical meaning. 190 Table 2. Thermal insulation of a clothing ensembles [I e , °C·m2/W (clo) *]. Ensemble Average weighted On the person On the manikin number thickness of material Equation, on which was calculated I e package, G , cm 1 2 3 4 5 I 4.49 0.750 0.808 0.749 0.704 0.928 (4.87) (5.21) (4.83) (4.5) (6.0) II 3.47 0.648 0.676 0.647 0.636 0.839 (4.18) (4.36) (4.18) (4.1) (5.4) III 2.23 0.557 0.581 0.557 0.555 0.660 (3.59) (3.75) (3.59) (3.58) (4.30) IV 1.28 0.519 0.535 0.519 0.548 0.667 (3.35) (3.45) (3.35) (3.5) (4.40) * 1 clo = 0.155 ºC·m2/W; Results and discussion The values of thermal insulation of the clothing ensembles calculated by various ways according to the test results from a person and a manikin are given in Table 2. According to the data, the value of I e calculated with a person by the equations 1 and 3 are practi- cally identical (difference 0-0.7 %); and those calculated by the equations 1 and 2 differ by 3-6 %. The difference between values of I e and Itr defined by the tests on the person and the manikin and calculated according to the equations 1, 3 and 4 are within the limits of 3-6 %; the differences in values defined by the equations 2 and 5, 1 and 5 are more essential (12-22 %; 15.6-22.8 %). Thus, the analysis of obtained data indicates that the difference in values of a thermal insulation of a clothing ensembles measured on a person and a manikin "at rest" under relatively calm air are defined by the way of their calculation. The least differences are observed with calculations according to the equations 1 and 3 (Standart R 12.4.185-99, 1999) and 4 (prEN 342, 1995); the greatest - by the equations 1 and 5 (ENV 342, 1997). Accordingly, a conclusion can be made, that for determining of heat-shielding properties of clothing ensembles the various methods (studies on the person and manikin) can be used. However, only when the equations 1 (or 3) and 4 (respectively for the person and manikin standing still) are used for the calculation of a thermal insulation of clothing en- sembles, the statistical differences are minimal. The studies carried out under conditions of wind and walking have shown that under influence of these factors a greater reduction of the thermal insulation of a clothing en- semble is observed when testing them on a manikin than on a person. The thermal insu- lation of an ensemble during walking under the effect of wind was reduced 32.6 % and 10 %, respectively. The most relevant values of a thermal insulation of clothing ensembles are obtained by calculation according to the equations 1 (Standart R 12.4.185-99, 1999) and 4 (prEN 342, 1995). The essential differences in values of thermal insulation of a clothing ensemble ob- tained under the effect of wind and walking on the person and the manikin suggest that additional studies on the influence of these factors are carried with allowance for varia- tion in its construction and thermo-physical parameters of materials. 191 References Afanasieva R F (1977) Hygienic principles of clothing projection for cold protection. Legkaya Industriya, Moscow, USSR. Afanasieva R F (1994) Physiological and hygienic clothing requirements for protection against cold, methods determination of thermal resistance, assessment. In: Holmér I eds. Work in cold environments. Invest. report. 1994:31. Pp 96-103, Sweden. Nilsson H (1997, Wednesday, February 12) Analysis of two methods of calculating the total insulation. In: Thermal Manikin Testing. Proceedings of a European Seminar at the National Institute for Working Life. Pp 17-22. Higenbottam C, Neale M S & Witney W R (1997, Wednesday, February 12) Thermal manikin measurements for human thermoregulatory modelling: the needs and options. In: Thermal Manikin Testing. Proceedings of a European Seminar at the National Institute for Working Life. Pp 6-16. Kolesnikov P A (1965) Heat Protection Clothing Properties. Legkaya Industriya, Moscow, USSR. Kolesnikov P A (1971) Design Bases of Heat Protective Clothing. Legkaya Industriya, Moscow, USSR. prEN 342. (1995). Protective clothing. Ensembles for protection against cold. ENV 342. (1997). Protective clothing. Ensembles for protection against cold. Standart R 12.4.185-99 Occupational safety standards system. Personal protective equipment against cold. 192 Evaporative resistance of various clothing ensembles measured on standing and walking manikin Krzysztof Blazejczyk1, Ingvar Holmér2 1 Institute of Geography and Spatial Organization, Warsaw, Poland 2 National Institute for Working Life, Solna, Sweden Introduction Evaporative heat loss (E) and evaporative resistance of air layer and clothing (Rt) are im- portant components of human heat balance models (Wenger 1972). They can be esti- mated according to ISO standard (ISO 7933). However, measured values for E and Rt obtained with thermal manikins are rare (Blazejczyk & Holmér 2000, McCullough et al. 1989, Meinander 1994, Kuklane & Holmér 1998, Liu & Holmér 1995, Tamura et al. 1994). The purpose of the paper is to present the results of experimental studies of evapo- ration rate (dm) and evaporative resistance (Rt) with a sweating manikin when wearing various clothing ensembles in different ambient conditions. Materials and methods The measurements were made in climatic chamber. Air temperature is regulated from 0 up to 35 °C. However, air humidity depends on air moisture outside the chamber. Air movement is generated by 3 wind turbines and can be regulated precisely from 0.2 to about 2 m s-1. At ”no wind” conditions an air motion of 0.1 m s-1 was assumed. The hu- midity in the chamber was determined by psychrometric measurements. Three pairs of thermometers (dry and wet bulb) - located on the altitude of 30, 130 and 180 cm over the floor - were used. A standard thermal, moveable manikin (TORE) was used as the anthropometric model for the sweating system. The manikin has 17 separately controlled segments. For the wet measurements the whole manikin was covered with thin plastic foil to protect tempera- ture sensors and heated wires against water infiltration. The skin surface was simulated using a tight fitting, cotton, long underwear. Feet and hands were dressed in thin socks and gloves. The water supply system consisted of a GILSON  peristaltic pump Minipuls 3, which supplied water simultaneously through 8 PVC tubes. The water was distributed to 6, left side segments of the manikin (upper arm, lower arm, hand, thigh, calf and foot). Special Y-shaped devices multiplied the number of watered points on the skin surface (up to about 20). The other segments of the manikin were kept dry. The volume of water sup- plied was settled at a level similar to evaporation observed in subjects at the same ambi- ent conditions. We have excluded the problem of water dripping and the wettedness of skin was relatively constant during every run of the experiment (36-46 % of total water capacity of skin textile). The MENEX model was used for simulations (Blazejczyk 1994). The water loss from the skin surface was observed on the nude manikin as well as un- der two types of clothing: tennis (T-shirt+shorts+socks) - basic I cl of 0.32 clo and stan- dard (long sleeve shirt+trousers+socks) -I cl of 0.59 clo. The tests were made on stationary 193 and walking (v’ = 0.8 m s-1) manikin as well as at two wind speeds (v): 0.1 and 1.0 m s-1. The studied combinations of ambient conditions are shown in Table 1. Table 1. The studied ambient conditions. Ta (°C) vp a (kPa) RH (%) v (m s-1) v’ (m s-1) 20 27 34 0.9-1.3 1.1-1.5 1.4-1.9 40-55 30-45 25-40 0.1 & 1.0 0.1 & 1.0 0.1 & 1.0 0.0 & 0.8 0.0 & 0.8 0.0 & 0.8 The I cl values for standing manikin in still air were assumed as refereed insulation of examined ensembles. However, clothing insulation observed on dry manikin in various combinations of wind and walk is shown in Table 2. Table 2. Clothing insulation at various combinations of wind and walk. Clothing Air movement conditions ensemble v=0, v’=0 v=1, v’=0 v=0, v’=0.8 v=1, v’=0.8 tennis 0.324 0.133 0.116 0.130 standard 0.593 0.492 0.418 0.426 v and v’ in m s-1 Total evaporative resistance (Rt in kPa m2 W-1) was calculated as follows: Rt vp vp A dm dt s a = - · · ( )  where: dt - time rate (in seconds), dm - rate of water evaporated (in grams), O - latent heat of va- porisation (= 2426 J), A - studied area (m2), vp s - vapour pressure on the surface (kPa), vp a - air vapour pressure (kPa). The total evaporative resistance consists of vapour resistance of boundary air layer (R a ) and vapour resistance of clothing (R cl). The Rt calculated for nude manikin is equal to Ra values. Thus, clothing evaporative resistance (R cl) can be determined. Results The water loss from the manikin (dm) decreased according to the rise in basic insulation of clothing ensembles. The trend is observed at all combinations of ambient conditions and varied from about -40 g per hour for an increase in insulation of 1 clo at an air tem- perature of 20 °C up to -55 g per hour - at T a of 34 °C (Figure 1). Evaporative resistance of clothing ensembles (R cl) were derived from Rt values calcu- lated for clothed and nude manikin. It clearly depends on air and/or manikin movement as well as on basic insulation of clothing. The biggest R cl was found in still air (0.006 kPa m 2 W-1 for standard and 0.0025 kPa m2 W-1 for tennis clothing). Air and/or manikin movement reduced R cl values to about 0.0005 - 0.0015 kPa m 2 W-1 (Figure 2). The biggest differences in R cl for standard and tennis clothing were noticed at still air on standing manikin. Considerably high were also differences in R cl between compared clothing ensembles at air temperature of 34 °C. 194 0 50 100 150 200 250 300 Ta 20 °Cdm (g h-1) v=1, v'=0.8 v=0, v'=0 v=1, v'=0 v=0, v'=0.8 0 50 100 150 200 250 300 Ta 27 °Cdm (g h-1) 0 50 100 150 200 250 300 0 0.1 0.2 0.3 0.4 0.5 0.6 Ta 34 °C Icl (clo) dm (g h-1) Figure 1. Evaporation rate (dm) as function of clothing insulation (I cl) at various air temperature (T a ) and air movement/walking (v, v’). Conclusions 1. The trends in dm in relation to clothing insulation have negative values, i.e. increase in I cl reduces amount of water evaporated from the manikin surface. 2. Evaporative resistance of clothing (R cl) was highest in still air on standing manikin. The air and/or manikin movement reduces R cl values to about 40 % for tennis and about 20 % for standard clothing. 195 3. The probable reasons of R cl reduction due to air movement around the manikin is the increased convection in microclimate and boundary layers by wind and movements. 0.00000 0.00100 0.00200 0.00300 0.00400 0.00500 0.00600 0.00700 0.00800 Tennis Standard standing, no wind walking, no wind standing, wind walking, wind Rcl (kPa m2 W-1) 20 27 34 20 27 34 20 27 34 20 27 34°C Figure 2. Evaporative resistance of clothing ensembles at various combinations of air tempera- ture (20, 27 and 34 °C) and air movement on standing and walking manikin. References Blazejczyk K (1994) New climatological-and-physiological model of man-environment heat exchange (MENEX) and its applications in bioclimatological studies. Zesz.IGiPZ PAN, 29, 28-58. Blazejczyk K & Holmér I (2000) Evaporation through clothing in various ambient conditions. Proc. 9th Int.Conf. on Envir.Ergon. (in press). ISO 7933 (1989) Hot environments - Analytical determination and interpretation of thermal stress using calculation of required sweat rate. Int.Org.Stand., Geneva. Kuklane K & Holmér I (1998) Effect of Sweating on Insulation of Footwear, Int.J. Occup. Safety and Ergon., 4, (2) 123-136. Liu X & Holmér I (1995) Evaporative heat transfer characteristics of industrial safety helmets, Applied Ergonomics, 26, 135-140. McCullough EA, Jones BW & Tamura T (1989) A data base for determining the evaporative resistance of clothing, ASHRAE Transactions, 95, 316-328. Meinander H (1994) Influence of sweating and ambient temperature on the thermal properties of clothing. Proc. Fifth Int.Symp. on Performance of Protective Clothing Improvement through Innovation, San Francisco. Tamura T & Nomiyama I (1994) Effects of wet condition of skin surface on evaporative heat transfer through clothing. Proc. Second Int.Cong. on Physiological Anthropology, Kiel, German Society of Physiological Anthropology, 334-337. Wenger CB (1972) Heat of evaporation of sweat: thermodynamic consideration. J.Appl.Physiol., 32, 4, 456-459. 196 Rain tightness of protective clothing – Prenormative interlaboratory tests using a manikin Peter Heffels Zentrum für Sicherheitstechnik (ZS), Bau-Berufsgenossenschaft Rheinland und Westfalen, D – 40699 Erkrath, Germany Introduction Current European and national standardisation of protective clothing against foul weather requires water-tightness tests for materials and seams only, as for these items, appropriate test standards and extensive test experience are available. In addition to material testing, standards should require testing of the complete garments in order to verify their rain tightness. In order to achieve this aim, a test standard to determine the rain tightness of garments has yet to be developed. In recent years, test institutes have applied various rain test methods in the context of research activities, product development and quality assurance in the framework of pro- duction control measures. These test methods need to be compared by interlaboratory tests involving various test institutes in order to prepare the ground for the development of a European standard for a rain test. This paper outlines different rain test methods, ex- plains the methodology applied for the interlaboratory tests and presents their essential results. The tests were conducted in the framework of a research project carried out jointly by EMPA (Swiss Federal Laboratories for Materials Testing and Research), St. Gallen, Switzerland, and ZS (Centre for Safety Engineering), Erkrath, Germany. Rain test methods In order to check the rain tightness of garments, three different rain test methods are used in Europe, which will be briefly described in the following. They mainly differ in the way in which the artificial rain is produced and the type of manikin. Rain test with drop nozzles (Method 1) In order to produce the artificial rain (approx. 450 l/hm²), a tub containing drop nozzles spaced 3.4 cm apart is fixed approx. 10 m above a manikin dressed with the test clothing. The diameter of the water droplets of approx. 4 mm complies with EN 29865 (rain- shower test for textile fabrics). An anthropometric manikin made of synthetic material is used for testing. In order to visually check and inspect the test clothing for any signs of water penetration after the exposure to artificial rain, the manikin is dressed with a detec- tion undershirt made of material with high water absorbing qualities, such as cotton. In addition, several sensors are fixed on different parts of the manikin in order to collect in- formation on when and where during the rain shower test wet areas occur. 197 Rain test with full cone nozzle (Method 2) The artificial rain (approx. 100 l/hm²) is produced by a full cone nozzle (with a spray angle of 90°), which is placed approx. 6 m above a manikin and sprays out the water droplets to obtain a diameter of approx. 0.5 mm. The protective clothing against foul weather is worn by an inflatable manikin, so that almost the entire inner face of the test garments, such as a jacket or a pair of trousers, comes into contact with the inflatable manikin. The manikin used for this rain test method is also dressed with a (cotton) detec- tion undershirt, which, where appropriate, is provided with humidity sensors. Rain method with drop-spray-nozzles (Method 3) The manikin is positioned in the centre between four vertical tubes containing several nozzles which are placed at a defined height. The nozzles produce bow-shaped drop sprays (droplet diameter approx. 1 mm), wetting the test clothing from the sides. In addi- tion, the test clothing is wetted from behind by drop-spray-nozzles placed above the manikin's head. The intensity of the artificial rain is approx. 240 l/hm². Humidity sensors are fixed at the manikin, which is also dressed with a cotton detection undershirt. The dimensions of the rain test device are approximately 1 m x 1 m x 2.5 m. In the following, some meteorological data on the amount of precipitation and droplet size will be given, which will allow you to compare the artificial rain as generated by these three methods with natural rain. Amount of precipitation: The mean amount of precipitation in many places in Central Europe varies between approx. 500 and 1000 l/m² per year. Heavy rainfall is called a cloudburst if a particularly intensive rain of at least 1 l/min m² falls for a very short pe- riod of time. During such a cloudburst, an essential part of the annual precipitation in Central Europe can come down within one hour. In other parts of the world, e.g. in tropi- cal regions, values may exceed 300 l/hm² in special cases. Type of droplets: The following types of droplets can be distinguished: a) cloud drops – diameter less than 0.1 mm – drop velocity < 0.25 m/sec. b) drizzle – diameter 0.1 to 0.5 mm – drop velocity 0.2 to 2 m/sec. c) rain drops – diameter 0.5 to approx. 5 mm – drop velocity 2 to 9 m/sec. These data show that all three rain methods correspond to a cloudburst with small or large rain drops. As the rain test method with drop nozzles involves a large droplet mass m and a high drop velocity v, the kinetic drop energy (E=0,5 mv²) is accordingly much higher than the values obtained when using the other methods. This has to be taken into account when comparing the test results. How the tests were made Three different test jackets were used for comparison: a) green jacket – approx. 990 g – PUR coating – without hood b) blue jacket – approx. 1000 g – micro fibres – without hood c) red jacket – approx. 480 g – double layer laminate – with a hood sewn onto it. In order to be able to ascertain whether the test results obtained for the three rain methods used show significant differences, the rain test methods applied should neither lead to a dry nor to a completely wet detection underwear. This is why, e.g. for testing the green and the blue jacket, their non-watertight collar seams as well as the head of the 198 manikin were covered with a plastic bag in order to protect the jackets against excessive water penetration during the tests. The comparison tests on the three jackets were carried out by five test institutes. At each test institute, the individual jackets were tested twice in accordance with one or more rain test method(s). After the tests, the position of visible wet areas as produced by a leakage in the test jackets was entered into diagrams. In addition, the test jackets and the detection underwear were weighed in their wet and dry condition and the results recorded. Test results The essential test results as achieved by the test institutes involved were analysed by means of figures and tables (see Table 1). The results came from a total of 42 rain tests, each carried out for a period of 1 h (without preliminary and additional tests). Table 1. Summary of wet areas detected. Test institute Method Test Test jacket green blue red A 1 12 - - - - - - - - - B 2 12 - - +1) + - - C 3 12 - - +1) +1) - - D 3 12 - - + + - - E 1 12 - - - - - - - - - - E 2 12 - - + + - - E 3 12 - - + + - - + no wet area - large wet area or several wet areas +1) only a small wet area - - very large wet area (> 50 % of the total jacket surface) Some essential conclusions drawn from the results of the tests using the three rain methods:  For one test jacket, the size of the wet area that resulted from water penetration varied depending on the rain method used. The rain tests showed the following sequence in terms of increasing wet spots on the jackets under test:  red, blue, green jacket for the rain method with drop nozzles;  blue, green, red jacket for the two other rain methods. The material of the blue and the green jacket proved not to be sufficiently tight when exposed to large rain drops with high kinetic energy as produced in the rain tests with drop nozzles. When exposed to smaller rain drops, the blue and the green jacket only showed much smaller wet areas or none at all. Because the red jacket was not water- tight, testing carried out on this item led to wet areas on the detection undershirt, re- gardless of the rain method used.  The tests generally showed a greater need for the material in the upper region of the jackets, especially in the shoulder area, to be water proof when the tests were carried out with the rain method with drop nozzles rather than with one of the other methods. 199 The rain tightness of the shoulder area is of particular importance e.g. for protective clothing against foul weather for use on construction sites, since the shoulder area is prominent when rain exposure is concerned. Fabrication flaws such as can be found in seams and the design of fasteners in the lower part of the jacket can be more easily detected in the rain test with full cone nozzle or with lateral drop-spray-nozzles than in the rain test with drop nozzles. A notice indicating such deficiencies may be useful when work has to be carried out under special marginal conditions (e.g. implying the use of water jets).  For the evaluation of rain tests as carried out on jackets, it is useful to use sensors to get information on when and where wet areas occur as well as data on the water ab- sorption of the detection underwear and the test jacket (see Table 2). The water ab- sorption of the detection underwear and the test jacket should in general be as low as possible. The water absorbed by a test jacket, which is e.g. provided with outer pockets or made of multi-layered material, can amount to more than 500 g (disad- vantage: increase in weight and prolonged drying times).  For the first time ever, the comparison tests allowed, for all three test methods, pre- liminary information to be gained on the repeatability of the test results. The test re- sults show that for each of the three rain methods, the same sequence could be estab- lished for the water tightness of the jackets under test, when the tests were repeated by the same institute or carried out by another. In addition, the occurrence of wet areas was generally quite similar. In order to increase the repeatability of the rain test results, it is necessary – in view of the number of influencing parameters that have to be taken account of in the test specifications – to establish detailed specifications of the test device (e.g. rain and manikin characteristics), the test procedure (e.g. posi- tioning of the manikin), the recording of the test and evaluation of test results. Table 2. Examples of water absorption of jackets (ja) and detection underwear (de). Garment Water absorption [g] Method 1 Method 2 Method 3 ja - green 611 105 270 de 429 - 47 ja - blue 452 40 57 de 120 - 16 ja - red 208 65 74 de 72 - 22 The test results achieved from the present investigation form the basis for discussions and decision-making in the process of standardising a rain test method. A work item for the development of a European standard is already in preparation, so that a generally available European test basis will be specified in the near future. References Heffels P, Hospach F, Rossi R & Weder, M (1999) Wetterschutzkleidung für den Bau (WEBAU) – Untersuchung zur Wasserdichtheit unter Berücksichtigung des Tragekomforts; Abschlußbericht, ZS und EMPA ENV 343 (1998) Protective clothing - Protection against foul weather; European Committee for Standardization; CEN, Brussels EN 29865 (1993) Textiles - Determination of water repellency of fabrics by the Bundesmann rain-shower test; European Committee for Standardization; CEN, Brussels 200 Development of the research and technology group flammability manikin systems James D. Squire Textile Technologist working for The Research and Technology Group of The Defence Clothing and Textiles Agency, Flagstaff Road, Colchester, Essex, CO1 2RS, UK Introduction The Research and Technology (R&T) Group is responsible for research and development programmes aimed at optimising the performance of textile equipment used by United Kingdom military forces. These programmes are actively investigating the following important aspects of mili- tary equipment.  Durability.  Comfort.  Protection from the environment.  Protection from battlefield hazards. Providing the military with the best available materials is essential to ensure they can carry out their duties in the most effective way possible. However, it is also important to minimise the physiological burden imposed by the equipment. R&T Group is leading programmes of work which use smart and responsive materials which allow protection to be provided when required, whilst keeping the physiological burden to a minimum dur- ing normal use. The flame and heat protection programme One of R&T Groups major items of work is the optimisation of flame and heat protection. UK military carry out many varied tasks in a wide range of environments. The range of threats from flame and heat are correspondingly varied and can include very simple threats like waste paper basket and fuel spillage fires as well as very complex and severe threats like explosive thermal pulses and Naval below decks fire-fighting. Personnel can be adequately protected from simple threats by provision of flame retardant working dress for military users or CE marked personal protective equipment for civilians working as contractors on military bases. Complex threats can only be adequately dealt with using well designed and, in many cases, specifically developed equipment; the key properties being protection from the clearly identified range of risks and compatibility and design of each item of a protective ensemble. 201 Testing for flame and heat protection As well as the range of threats being complex and varied, the number of test methods available are also bewildering in their variety. Over recent years, R&T Group have ob- tained a world class range of apparatus allowing us to characterise important aspects of flame and heat protection relevant to the military environment. This apparatus range allows us to measure the properties of materials, starting with sample sizes of a few tenths of a gram right through to the whole ensemble of equipment, and includes the following test methods:  Differential Scanning Calorimeter, Thermo-mechanical and Thermo-gravimetric analysis.  Resistance to convective and radiative heat sources.  Resistance to small ignition source.  Cone calorimetry.  Flammability manikin. The flammability manikin has been in use for the last five years and during this time many hundreds of clothing ensembles have been tested. A great deal of useful informa- tion has been produced allowing many improvements in protective clothing to be made available to the user. Why was the flammability manikin developed? Manikins have been used for many years by R&T Group. Their usefulness was first identified during simulated nuclear thermal exposures in New Mexico, in a series of tests carried out by the US military in which we co-operated. Deploying instrumented steel manikins produced much useful information relating to the measures necessary to protect men in the open from these very severe events. Development of a laboratory based full body manikin was completed in 1995, which gave us the capability to measure protection when using a less severe challenge than the nuclear pulse. The system was designed to measure two important performance charac- teristics; the effect of fit and design of equipment on protection and the protective properties different types of textile material. Features of the manikin system The system is made up of an instrumented manikin, an array of burners and a com- puterised data acquisition system. The manikin is constructed from reinforced polyester resin and is instrumented using 10 mm diameter copper disc calorimeters, set flush with the surface of the manikin. K type thermocouples are welded onto the inside of the copper discs that measure the temperature of the calorimeter. The manikin is jointed at the neck, shoulders, elbows, wrists, hips, knees and ankles, allowing the manikin to be posed into various positions adding an extra dimension to the range of test conditions available. The manikin is surrounded by an array of thirty propane cup burners that engulf the manikin in flame, with the energy delivered to the surface being, on average, 75 kW/m2. Simple manual controls are used at present to pilot and run these burners, fed by four 202 large propane cylinders, connected in series to allow an even pressure to be delivered throughout the length of any exposure. Information from the 120 calorimeters over the surface of the manikin is collected by a computerised data acquisition system. One reading is taken per second from each calo- rimeter. On completion of an exposure, a unique programme, designed specifically for this system, is used to calculate the number of calorimeters that reached the equivalent of first, second and third degree burn. Also the most severely damaged sensor and the ear- liest time that each of the levels of burn injury was reached is calculated. All the results are presented in the form of an injury curve, a pictogram of damaged sites and the injury values along with other relevant test parameters such as exposure duration and a descrip- tion of the ensemble. What is the manikin used for? The manikin produces results that allow two major aspects of performance to be assessed;  The effect of garment design.  The effect of material design. Exposure of complex ensembles of protective equipment regularly shows up weak- nesses in design, for example a very costly flame protective garment will be of little use if parts of the garment trim are ignitable. This type f failure of performance is obvious when manikin testing is carried out, but can be very easy to overlook if only looking at the performance of component materials. A disadvantage to testing ensembles on the whole body manikin was found to be that it is impossible to realistically dress the manikin with equipment worn about the head and neck, such as headovers, helmets and respiratory protection. This is due to wiring leaving the manikin via a loom attached to the neck and a hook in the top of the head, which sup- ports the weight of the manikin. Development of the instrumented head form Due both to the problems described above and the increasing importance of measuring the performance of head protection, particularly for the Royal Navy, we embarked on the development of an instrumented manikin head. The main features of this system were;  An increased density of calorimeters over the head surface (30 in total, compared to only 18 on the full body system).  Supported in a way which allows head protective equipment to be donned realisti- cally, without cutting seams or drilling holes.  For the first time, protective garments for the head can be tested in a situation repre- sentative of real-life. The burner and data acquisition systems are fundamentally the same as for the whole body. The results are presented on a specially adapted pictogram, mapping damaged areas of the head, consequently equipment and material design weaknesses can be accu- rately identified and corrected and the effect these corrections have on the performance can be measured realistically. 203 Areas of use for the instrumented head form As well as identification of optimised head protection for Royal Navy fire-fighters, we are now in a position to offer this innovative new test apparatus for other purposes, namely;  An optional additional method to support draft European Standards of performance for fire-fighters head protection.  A commercial facility available to equipment designers to assist in the complex deci- sion making processes involved in measuring the relative performance of the many types of protective clothing currently on the market.  A design tool for aiding in the optimisation of the available performance of equipment. We currently have an extensive programme of testing underway to characterise the protective capabilities of new and used in-service military equipment. This programme will, for the first time, make available a comprehensive body of data: This can be used to assist in the risk taking approach which is becoming increasingly necessary when pur- chasing protective equipment when faced with shrinking budgets. Also it can be used to help to build the confidence of the user in the performance of his equipment – their life and future well being depend on the ability o their equipment to protect them and it is essential that this factor remains uppermost in the designers mind. Summary Research and Technology Group’s international reputation as designers of military tex- tiles is based on our unbiased and objective assessment of the performance of a huge range of equipment. We are able to back up our assessment work with unrivalled measurement facilities. In the case of flame and heat protection, the full body manikin is now complemented by the world’s first head form, a unique piece of equipment that has already proved to be an invaluable tool for our design team. It is our intention to extend the use of our flammability measurement apparatus to al- low it to be exploited commercially. The addition of the head form in particular will be- come an accepted additional test method to help civilian designers and users decide what equipment is most appropriate for their end use. 204 Thermal properties of protective gloves measured with a sweating hand Harriet Meinander Textile Group, VTT Chemical Technology, PO Box 1402, FIN-33101 Tampere, Finland Introduction The protective gloves form an important part of the personal protection of a worker. The gloves protect the hands against cold, heat, chemicals, mechanical threats, radiation, etc. An unrestricted performance of the hands is essential for many working tasks, and the need for finger dexterity is often in conflict with the need for protection of the hands. From the thermal balance point of view the hands react differently from the central body parts, cooling rapidly in cold conditions and being an area with high heat loss and sweating in the warm conditions. The thermal comfort of the hands is essential for the general work efficiency and safety. This is not generally considered in the choice and classification of protective gloves. In the European standards for protective gloves, the general requirement standard EN 420 states that “Where practicable, protective gloves shall allow water vapour trans- mission” and gives a recommended value of 2 mg/cm² for leather gloves, tested accord- ing to the leather standard IUP 15. EN 420 also states that “Where the protection level of the glove inhibits or excludes water vapour transmission, then the glove shall be designed to reduce the effect of perspiration as much as possible”. The water absorption of leather gloves should be at least 8 mg/cm²·8 h. In the standard for protective gloves against cold, EN 511, the thermal resistance is determined using a thermal hand model, in which how- ever the influence of sweating on the thermal insulation and comfort is not considered. Methodology A sweating hand was constructed for the measurement of simultaneous heat and moisture transmission through gloves. The same basic principle was used for the construction as in our previous instruments, the sweating cylinder, foot and manikin (Meinander, 1992). The hand surface is divided into 10 heating sections, which are separately controlled (5 fingers, palm, knuckle, wrist, 2 lower arm sections). Liquid water is supplied to 30 sweat glands underneath the skin material, where it evaporates. Figure 1 shows the sweating hand with some of the sweat glands, photographed before the application of the skin ma- terial. In the first test series with the sweating hand the thermal insulation and water vapour transmission properties of a set of different types of protective gloves were measured. The gloves were acquired from Finnish producers and importers, and are listed in Table 1. The measurements were performed in the climatic chamber, where the temperature was set to –15 °C. First a dry test was done, duration 2 hours, and directly after that the sweating 200 g/m²·h was turned on, duration 3 hours. The gloves were conditioned in 65 % RH / 20 °C for 24 hours before the tests, and weighed before and directly after the 205 tests. The parts of the lower arm, which were not covered by the glove, were insulated with a thick woollen knit and a quilted fabric. Two replications of each test were done. Figure 1. The sweating hand without skin material. Some of the sweat glands are indicated with arrows. Table 1. Tested gloves Sample no Product Type 1 M150 cold protection 2 M3202 cold protection 3 3040 heat protection 4 TKK 36 NC heat protection 5 JokaOiler chemical protection 6 JokaTop chemical protection 7 JokaTherm chemical protection + thermal insulation Results Each test gives the following set of results:  a summary table with the results at the end of the test (set values, total condensation, total heat supply, heat of evaporation, thermal insulation (mean and local), weight in- crease in gloves)  graphs of water supply / condensation / evaporation, local temperatures, local heat supply values, and local thermal insulation values during the tests. Most important are the thermal insulation in dry and sweating conditions, and the evaporation and condensation of water, which are shown in Figures 2 and 3. Discussion The dry thermal insulation R c of all the tested gloves is somewhat on the same level, between 0,35 and 0,40 m²·°C/W, Figure 2. Even the chemical protective gloves have fairly high insulation values, due to their loose design. The decrease in thermal insulation 206 after the sweating test is between 0,069 and 0,118 units, being smallest for the cold pro- tective glove #2 and largest for the chemical protective glove #5. Thermal insulation of the tested gloves, dry and sweating (corrected) values 0 0.1 0.2 0.3 0.4 0.5 1 2 3 4 5 6 7 Th er m al in su la tio n, m ²· °C /W dry Rc sweating Rc,corr Figure 2. Thermal insulation values of the tested gloves in the end of the dry and the sweating tests. Condensated and evaporated water during the sweating tests 0 5 10 15 20 1 2 3 4 5 6 7co n de ns at ed / ev ap or at ed w at er , g/ 3h condensated, mc evaporated, me Figure 3. Measured water condensation and evaporation values in the sweating tests. The supplied amount of water during the 3 hour test is 22,8 g (the sweating area is 0,038 m²). A part of the water is transmitted as water vapour through the glove to the en- vironment, transmitting heat of evaporation; a part evaporates but recondensates to liquid water in the glove; and a part remains as liquid water in the skin material of the hand. A high evaporation and a low condensation are always aimed at. The heat protective glove #4 has the highest evaporation, 11,8 g or 52 % of the supplied water. The chemical pro- tective gloves #5, #6 and #7 have the lowest evaporation values, 1,7 g or 8 %, 2,2 g or 10 % and 2,1 g or 9 %, respectively. However, for impermeable gloves it could have been expected that no evaporation at all can take place, and it can be assumed that the water vapour in this case is transferred through the opening of the glove. The tests were performed in a low temperature, which caused a relatively low water vapour transmission even in those gloves which do not have a moisture barrier. Based on earlier studies of protective clothing ensembles it can be assumed that in a higher envi- ronment temperature, the difference between “breathable” and “non-breathable” gloves would be even more significant. 207 Conclusions Within this limited project it was shown that the thermal comfort properties (thermal in- sulation and water vapour transmission) of protective gloves can be assessed using a sweating thermal hand. Differences between different types of gloves can be defined, which should be important information for the development and choice of gloves for spe- cific wear situations. Acknowledgements This project was carried out at VTT Chemical Technology, Materials Technology, in Tampere, Finland. Financial support of the Finnish Work Environment Fund is gratefully acknowledged. My colleagues, Mr Unto Mäenpää, Mr Ralf Österlund and Mrs Anja Kesti, who have participated in the project, are also gratefully thanked, as well as the companies who kindly supplied the test gloves. References Meinander H (1992) Coppelius – a sweating thermal manikin for the assessment of functional clothing. Fourth Scandinavian Symposium on Protective Clothing against Chemicals and Other Health Hazards (NOKOBETEF IV), Kittilä, Finland. EN 420:1994. General requirements for gloves. Brussels: European Committee for Standardisation. 14 p. EN 511:1994. Protective gloves against cold. Brussels: European Committee for Standardisation. 10 p. 208 Manual performance after gripping cold surfaces with and without gloves Qiuqing Geng1, 2, Eva Karlsson1, Ingvar Holmér1 1 Programme for Respiratory Health and Climate, National Institute for Working Life, S - 171 84 Solna, Sweden 2 Department of Human Work Sciences, Luleå University of Technology, S - 971 87 Luleå, Sweden Introduction Manual performance is a combination of many kinds of abilities that require, for instance, good tactile sensitivity, hand dexterity, force capability and motor co-ordination etc. and these skills are all influenced by hand cooling. Previous studies mainly focused on the influence of hand cooling during cold air exposure (Mackworth, 1953; Morton et al., 1960; Tanaka et al., 1983; Rogers et al., 1984; Daanen, 1993; Geng et al., 1997) or im- mersion in the cold water (Provins et al., 1960; Bensel et al., 1974) on the manual per- formance. In practise, workers often use bare hands or gloved hands to grip cold surfaces. Cold contacts with bare hand or insufficient protective hand wear add to hand cooling when gripping, thereby reducing the manual performance in cold operations. In addition, a risk of cold injury through gripping the cold surface exists. The present study concerns hand cooling response and the subsequent manual performance after gripping different cold rods with bare hands and gloved hands in the cold. Methods Ten subjects (5 females and 5 males) age 22 - 35 years, weight 56-90 kg without any history of peripheral vascular disease or cold injury volunteered in the experiments. The instruction and explanation of the experiment were given to the subjects. Subjects were allowed to discontinue the performance of the experiment whenever they felt uncomfort- able in the cold situation. All subjects were right-handed and non-smokers. Two types of industrial gloves were selected for the experiment. Their characteristics are described in Table 1.The selection of protective gloves against cold was based on the contact cold materials at various cold temperatures. Table 1. Gloves used in the experiment. Glove type Insulation (m2 °C/W) Thickness (mm) Weight (g) Material Usage in the test 1 0.06 1.00 15.00 Cotton and Rubber Gripping Al, steel and stone rods at –10 °C; gripping nylon rod at –10 and –20 °C 2 0.12 2.70 59.40 Leather (pig) and Cotton Gripping nylon rod at –20 °C; gripping Al and steel rods at –10 and –20 °C; Experiments were carried out in a cold climatic chamber at various cold ambient tem- peratures, which were adjusted from –20 to -10 °C to equilibrate with the surface tem- perature of the rods tested. Four rods with size  40400 mm of different materials (steel, aluminium, stone and nylon) were selected as the gripping objects. Their thermal 209 properties are listed in Table 2. The rods were mounted in a counter balance system to support a constant weight of 0.5 kg for gripping force. Thermocouples were used to measure hand/finger-cold surface interface contact temperature (TC). Two standard instruments, namely, Semmes-Weinstein filaments for the pressure tac- tile sensitivity and O'Connor model 32021 for the finger dexterity, were utilised to evalu- ate and analyse the manual performance after gripping the cold rods. Table 2. Materials used for gripping cold contact. Material Thermal conductivity, O, (Wm-1K-1) Specific heat, C, (J kg-1K-1 ) Density, U, (103 kg m-3) Penetration coefficient (J m-2 s-1/2 K-1) Nylon-6 0.34 1484.00 1.20 778.12 Stone 2.07 750.00 2.80 2084.95 Steel 14.80 461.00 7.75 7271.64 Aluminium 180.00 900.00 2.77 21183.48 Experiments were carried out under combined conditions of different materials and surface temperature. Pilot studies had been performed to determine optimum experimen- tal procedure. During gripping with or without gloves, the TC was measured on the palm side (under index finger) of the dominant hand. The duration of gripping depended on either extensive pain or the contact temperature (stopped when TC < 1 °C). The subject was asked to rest between trails in order to fully re-warm at room temperature (~ +20 °C). Also, each subject was asked to perform the manual tasks after gripping. The pres- sure tactile sensitivity test was performed using Semmes-Weinstein filaments, which touched on the palm side (under index and little fingers and pad of the middle finger). In the experiment, the subject gave the actual response within 3 seconds without seeing. The filament’s size 2.36 represents that a press force of 15 mg was used as the lightest force in all tests (Tomancik, 1987). For the finger dexterity, the subject was required to fill the first line with 3 pins per hole from left to right as quickly as possible. As a result, the time needed to complete the task and number of mistakes (incorrect pins were filled or fell down) were recorded. To avoid a cold injury during gripping the cold metallic sur- faces at –10 °C, the cooled hands were treated immediately by supervisor’s warm hand. This could influence the results of the manual performance tests. Results and discussion Hand cooling on cold rods with or without gloves Figure 1 shows the contact temperature at the end of gripping the cold metallic and non- metallic rods without or with gloves. It is seen that the hand cooling is related to both the material of the rod, ambient temperature as well as type of gloving. As shown in Figure 1a, significant temperature differences (at 95 % LSD Intervals) appeared for the case of without or with gloves and for the case of glove type when gripping the metallic rods at – 10 °C. Clearly, a thicker glove with higher thermal insulation value (type 2) gives a higher TC (>21 °C), compared to a thinner glove (Type 1) (<14 °C) or no glove (<9 °C). Gloves lead to a decreased rate of heat transfer from hand to the cold surface during grip- ping (Table 1). In addition, Figure 1b shows a significant glove effect on the TC at the end of gripping the cold nylon rod at –20 °C. The TC was significantly lower at the end of gripping without gloves. This emphasises that hands must be protected with gloves during gripping non-metallic materials like nylon at cold temperature of –20 °C and 210 lower. However, there exists no statistically variation on the TC between glove types 1 and 2 in this case. It is noticed that the variation of the TC appeared to be significantly related to the pres- ence of hand protection. Averaged for all conditions, the presence of a glove increased hand contact temperature by 4-10 °C for gripping the cold rods, compared to bare hand. Figure 1. Mean TC at the end of gripping different cold rods with bare hand (0), gloves (1) and (2): a) for aluminium and steel at-10 °C and b) for nylon at –20 °C. Manual performance after gripping the cold surfaces The manual performance (SWP sensitivity and O’Connor fingers dexterity) after gripping with and without gloves at –10 °C is shown in Figure 2. It is obvious that manual per- formance after gripping the cold non-metallic rods with gloves was better than with bare hand. Type of glove also has a significant effect on the manual performance for the case of gripping the cold metallic rods. Hands after using glove 2 with higher thermal insulation value give a subsequent better performance. The resultant differences that are due to the hand cooling after gripping the cold rods with or without protective hand wear to varying degrees. It is known that the hand cooling reduces the dexterity due to the influence of cold on muscles and joints of finger/hand (Havenith, et al., 1995). Finger dexterity decreased sharply when the finger skin temperature fell below 15 °C has been found by some earlier studies (LeBlanc, 1956; Daanen, 1993). Also, the loss of tactile sensitivity of fingers/hands at cold environment may be attributed to changes in the properties of the skin or to the effects on biochemical processes at nerve or receptor level (Geng, et al., 1997). The results reflect that glove 1 with lower thermal insulation value was ineffective to protect the hand during gripping the cold metals at –10 °C, but it seems satisfactory for the case of gripping on the non-metallic rods even at –20 °C (Figure1b). Conclusions  Hand contact cooling was affected both by type of material and their surface tem- perature as well as by the use of the protective hand wear. It is suggested that hands must be protected with gloves during gripping metallic materials at -10 °C or lower and while gripping non-metallic materials like nylon at temperatures of –20 °C and lower.  A significant hand cooling after gripping cold rods with bare hands or with insuffi- cient protective hand wear caused a significant decrease in manual performance. 211 0 20 40 60 80 100 120 140 Nylon Stone Steel Alum. M e a n SW P fo rc e (m g) No Glove Glove 1 Glove 2 34 36 38 40 42 44 46 Nylon Stone Steel Alum. Material M a e n tim e (S e c. ) No Glove Glove 1 Glove 2 Figure 2. Effects of gripping material type on the SWP sensitivity and O’Connor fingers dex- terity after hand gripping with and without gloves at –10 oC. References Bensel CK & Lockhart JM (1974) Cold-induced vasodilatation onset and manual performance in the cold. Ergonomics, Vol. 17, 6, 717-730. Daanen HAM (1993) Deterioration of manual performance in cold and windy climates. AGARD Conference Proceedings; “The support of air operations under extreme hot and cold weather condition”, NATO, May, 540. Geng Q, Kuklane K & Holmér I (1997) Tactile sensitivity of gloved hands in the cold operation. Applied Human Science, Vol. 16, No. 6, 229-236. Havenith G, Heus R & Daanen HAM (1995) The hand in the cold, performance and risk. Arct. Med. Res. 54: Suppl. 2: 37-47. LeBlanc JA (1956) Impairment of manual dexterity in the cold. J. Appl. Physiol., 9, 62-64. Mackworth NH (1953) Finger numbness in very cold winds. Journal of Applied Physiology, 5, 533-543. Morton R & Provins KA (1960) Finger numbness after acute local exposure to cold. Journal of Applied Physiology, 15 (1), 149 –154. Provins KA & Morton R (1960) Tactile discrimination and skin temperature. Journal of Applied Physiology, 15 (1), 155-160. Rogers WH & Noddin EM (1984) Manual performance in the cold with gloves and bare hands. Perceptual and Motor Skill, 59, 3-13. Tanaka M, Tochihara Y, Yamazaki S, Ohnaka T & Yoshida K (1983) Thermal reaction and manual performance during cold exposure while wearing cold protective clothing. Ergonomics, Vol.26, 2, 141-149. Tomancik L (1987) Directions for using Semmes-Weinstein Monofilaments. North Coast Medical, New Jersey. 212 Cold protective gloves in meat processing industry - product development and selection Hannu Anttonen, Piritta Pietikäinen, Hannu Rintamäki and Sirkka Rissanen Oulu Regional Institute of Occupational Health, Oulu, Finland Background One of the major problems in food industry is cooling of hands causing discomfort, acci- dents and further health problems. Workers in the cooled factories face many health risks due to repetitive manual work, cold environment, draught and moisture. Griefahn et al (1996) have presented evidence that working in moderately cold working places may constitute a health risk e.g. symptoms of white finger disease. Upper limb overuse symptoms are common in the meat processing industry (Viikari-Juntura et al, 1991). There is not so many studies concerning the development of gloves in this field of in- dustry even though the effect of thin gloves has been known to be very important on cooling (Havenith et al, 1992). The main aim of this project was to test and develop gloves to decrease cold induced discomfort and hand cooling among the workers and test the meaning of clothing on fin- ger temperatures. Material and methods Workers in various departments used several different types of gloves. Cool or cold envi- ronment (0-10 C) as well as cold and wet products were common for all the workers. In the prestudy 11 workers were interviewed concerning gloves used today. Six different kind of commercial or self-developed knitted gloves were tested, meas- ured and evaluated by measurements and interviews of users. Materials tested were: PES (100 % polyester), CO/PES (50/50 cotton/polyester), CO/PAN (50/50 cotton/acrylic), CO (100 % cotton), Polycolon (100 % polypropylene), Thermastat (100 % polyester; hollow fibre). 19 workers tested the new materials; 10 in an elimination test for all six materials and 9 for the best two. Most common used gloves were two-layer system: rub- ber glove with knitted underglove. Because thicker gloves than used today were not pos- sible, different materials to decrease sensation of cold and moisture were tested. Conduc- tive and convective (prEN 511) thermal insulation were measured in laboratory. Also the cooling of hands were measured by the field tests. The overall heat balance of the work- ers was also assessed by the measurements and guidelines for protective clothing were revised. The project was conducted in co-operation with glove manufacturers, which pro- vided the new type of gloves for the companies. Results In the prestudy (n=1117) 90 % of workers reported about harmful cooling of hands and fingers. 40 % informed about pain experienced in hands. So the hand protection against cooling is very important. 213 Workers in the case companies were interviewed in purpose to clear up how the pre- sent gloves function in those cold conditions. Results from the interview are presented in Table 1. Table 1. Results from the interview of the workers relating to the used cold protective gloves in the case companies. Answer (%) (n=11) Question yes no Do you suffer from hand cooling during work? 100 0 Do you suffer from moisture in gloves? 91 9 Do the gloves fit? 55 45 Does your manual dexterity remain while wearing gloves? 45 55 In the physical measurements of contact cooling the differences between the materials and gloves were maximally 3 C in 25 min test period. Contact cooling of tested mate- rials is presented in Figure 1. Materials were tested as dry and moist with weight of 0.1 kg. 2 4 6 8 10 12 14 16 18 20 22 0. 02 15 0 27 0 39 0 51 0 63 0 75 0 87 0 99 0 11 10 12 30 13 50 14 70 time (s) su rf ac e te m pe ra tu re ( C ) PES CO/PES CO Polycolon CO/PAN Thermastat Figure 1. Contact cooling for tested dry (beginning temperature 22 qC) and moist (beginning temperature 20 q C) materials. Different materials’ possible differences in convective insulation were tested with four new kind of glove materials. Test results are presented in Figure 2. Convective insulation of the tested materials did not differ greatly from each other, but when much thicker gloves were tested for comparison the range of variation was larger, 0.051-0.110 m2K/W. The differences in insulation of the gloves in the field measurements were similar to the laboratory tests even though the conditions varied to some degree in every test. 214 0.00 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09 0.10 Thermastat PES Polycolon CO/PAN material co n v ec tiv e th er m a l i n su la tio n (m 2 K /W ) dry / wind 0.4 m/s dry / wind 4 m/s moist / wind 0.4 m/s Figure 2. Convective cooling for dry and moist materials (wind 0.4 m/s or 4 m/s). By combining the results from laboratory and field tests four and later two superior materials were chosen to be further tested in the companies. Material based on polypro- pylene and thermastat were found to be better than the other tested materials according to the measurements and questionnaires. Results of interview concerning comparison of the old and the two best glove materials are presented in Table 2. Comparing Polycolon and Thermastat, all the workers preferred Polycolon. Because Polycolon as a moist transferring material was better than other tested materials it shows the meaning of moisture on cold sensation. Table 2. Results from the interview concerning comparison of the old and the two best glove materials. Answears (%) (n=9) Material better than the old one no difference worse than the old one Polycolon 89 11 0 Thermastat 67 22 11 The use of additional clothing, down vest (0.5 clo), made the finger skin temperatures either 4-5 C higher or the cooling of the fingers became slower. In conclusion, by combining the test results from laboratory tests and from comparison field tests done by workers in the case companies, best materials are in superiority order: Polycolon, Thermastat, CO/PAN, PES. Discussion The present study confirms the earlier findings (Griefahn et al, 1996; Viikari-Juntura et al, 1991; Havenith et al, 1992) by showing a high prevalence of complaints of fingers in moderately cold food processing industry. Workers’ cooling is a marked problem in the cold food processing industry and causes unpleasant cold sensations, pain and numbness of fingers and hands. In this study the cooling by convection, conduction and subjective sensations of cool- ing with different gloves were measured. In conclusion, the differences between tested 215 materials and gloves were quite small but distinct. The effect of thickness of the material was small compared to the effect of material itself. In the other hand increasing the thick- ness of glove wasn’t possible because decreasing manual dexterity. Wetness of gloves wasn’t so big problem as informed beforehand measured as amount of water in glove. However it effected very much on cold sensations, but with the moisture transferring material situation became much better. The results in different tests were along the same line and the subjective evaluations supported the measurement results. Also the meaning of body clothing insulation on finger temperature was clear accord- ing to the field measurements and that should be taken care better than nowadays. References Griefahn B, Mehnert P, Bröde P & Forsthoff A (1996) Health risks related to work in cold in the food industry. 25th Int. Congress on Occupational Health, Stockholm 1996, 22. Havenith G, Van de Linde E, Heus R (1992) Pain, thermal sensation and cooling rates of hands while touching cold materials. European Journal of Applied Physiology, 65, 43-51. Viikari-Juntura E, Kurppa K, Kuosma E, Huuskonen M, Kuorinka I, Ketola R & Könni U (1991) Prevalence of epicondylitis and elbow pain in the meat-processing industry. Scandinavian Journal of Work, Environment and Health, 17, 38-45. 216 Protective gloves against mechanical and thermal risks Doina Toma, Eftalea Carpus The National R-D Institute for Textile and Leather, Bucharest, Romania Introduction A significant part in preventing professional diseases and work accidents is held by pro- tective gloves which have to respond to the following requirements:  a perfect conception, without faults  comfort in wear  adapted the purpose of wearing  made of which should provide protection to all parameters involved in the specific safety field The choice of a solution concerning the accomplishing of protective g1oves involves knowing the risks, the properties of the fabrics their behaviour under risk conditions. The poster will present the results of researches concerning:  the risk levels depending on the fields of utilisation  designing and accomplishing the woven structures meant for protective gloves  protective gloves against mechanical and thermal risks The textile structures have been accomplished by using fibres with performant characteristics, such as  Twaron para-aramidic fibres  Kermel polyamid-imidic fibres Criteria Essential criterion is to protect the human life. Other criteria are:  damages and level of risk assessments  limits of the main characteristics, specific and additional properties  establishing the criteria of fiber selection  designing the textiles depending on the performance demands of end-uses  testing in conformity to European standards  designing and accomplishing the individual protective clothing  IPC certifying General requirements for designing and accomplishing:  terms of choosing textile raw materials  elements of designing  technological solutions The other requirements are:  Requirements for environmental resistance  Requirements for designing  Requirements for an efficient marketing 217 Caloric protective gloves “CALOR TW” Structure Layers:  exterior: 100% twaron fabric  intermediary: 100% kermel nonwoven  interior: 100% viscose FR Utilization fields:  glass and ceramic industry  metallurgic industry Characteristics Mechanical risks:  Abrasion resistance: > 4000 cycles, 3rd performance level (SR EN 388)  Blade cut resistance: cut index >2.5, 2nd performance level (SR EN 388). Thermal risks resistance:  behaviour to fire: after-flame time 0 s, after-glow time 0 s (after 15 s fire exposure), 4th performance level (SR EN 407);  contact heat resistance: break point time 20 s (350 °C contact temperature), 3rd performance level (SR EN 407);  convection heat resistance: increasing time* 36 s (HTI>18 s), 4th performance level (SR EN 531). *time to increase internal layer temperature with 40 °C, at direct flame,caloric flux of 80 kW/ sqm Knitted protective gloves Structure Layers:  exterior: 100% twaron knit  interior: 100% viscose FR Utilization fields:  glass and ceramic industry  machine -building industry  physic - mechanical laboratories Characteristics Mechanical risks:  Abrasion resistance: > 100 cycles, 3rd performance level (SR EN 388).  Tear strength: >380 N, 4th performance level (SR EN 388). Thermal risks:  heat resistance 250 °C, 6 s;  contact temperature 100 °C, 15 s. 218 A case study on the selection and development of cut resistant protective gloves for household appliance assembly industries Jaime Lara, Chantal Tellier IRSST, 505 de Maisonneuve Blvd. West, Montreal, Quebec, Canada H3A 3C2 Introduction Statistics of the Quebec Occupational Health and Safety Board demonstrated that in the metal and electrical appliances industries, close to 30 % of injuries are to the hands (Commission de la santé et de la sécurité du travail du Québec, 1999). Most injuries are cuts, punctures and lacerations caused by tools, metal pieces, small screws and wire ma- nipulations. These industries are characterized by rapid working operations and the high level of dexterity needed to perform the assembly operations. A test method to evaluate the cut resistance of materials used in protective clothing was recently developed and has been adopted as ISO Standard 13997 (ISO 13997, 1998). To evaluate dexterity when us- ing gloves, laboratory test methods exist (Bensel, 1993; Robinette et al., 1986; Johnson & Sleeper, 1986; Plumber et al., 1985). However, to our knowledge, no study exists that attempts to correlate the laboratory results and the workers’ perception of dexterity in real working situations. The goal of this study was to select or develop a protective glove for household appliance assembly operations that provides good dexterity and has a good cut resistance. Methodology This study had two major parts: I) Identifying the characteristics of a glove adapted to the type of work performed by assembly line workers; II) Selecting or developing a glove that fulfills the requirements identified in the first part of the study Identifying the characteristics of the protective gloves A Primary analysis. The goal of this part of the study was to establish the main re- quirements for a protective glove acceptable to workers, by analyzing the long-term in- jury register, by systematic questioning of workers, and by observing real operations in the field. A household appliance manufacturing company collaborated on this study. B Detailed analysis. After the first analysis of the information extracted, a detailed analysis was carried out on the hand injuries that occurred in a two-month period. For this, a detailed questionnaire specifically prepared for this study was used to obtain the workers’ description of the accident within two days of its occurrence, so that they would clearly remember what had happened. One hundred and forty accidents involving 93 workers were analyzed. This step in the study enabled us to assess the level of risk and the workstation on the assembly line with the highest risk of injury. 219 Analysis of workstations. The operational modes and working positions on the assembly lines for the workstations identified as the highest risk were systematically observed and filmed. This identified the risk of cut injuries and the dexterity level required to perform this work. The reasons for using protective gloves or not were identified as well. The workers’ opinions of the desired characteristics for protective gloves were also obtained. The identified characteristics for the protective gloves were the following;  cut resistant  allow operations needing good dexterity  good adhesion to metal pieces contaminated with oil or grease  glove material must have a capacity to absorb oil or grease. Selecting or developing a glove Glove distributors and manufacturers were asked to propose protective gloves with the above mentioned characteristics. Thirty gloves were proposed and tested for cut resis- tance using standard test method ISO 13997, and for dexterity using an in-house method simulating field operations involving the handling of screws and wires. Results From the thirty gloves proposed, twelve gloves were selected for more detailed tests. Re- sults of these tests are presented in Table 1. Furthermore, the glove manufacturer that participated in this project agreed to develop two prototypes with the above-mentioned desired characteristics, which are also presented in the same table. Cut resistance values reported in Table 1 represent the force necessary to cut through a material with a 20-mm blade displacement. Dexterity values are a subjective classification from 1 (low dexterity) to 5 (very good dexterity). Table 1. Preliminary cut resistance and dexterity results. Gloves (commercial products) Material Cut resistance F (g) Dexterity 1- Jomac 1814 Kevlar/nitrile 1 043 4.5 2- Jomac 1805 NBC Kevlar/cotton/ nitrile 1 268 4.0 3- Jomac 1800 NPC Kevlar/cotton/Nitrile 1 131 3.5 4- Levitt GKVL20AR-100 Kevlar 1 324 3.0 5- Dupont Golden Needles Kevlar 1 232 3.0 6- Superior Gloves S13K Kevlar 774 4.0 7- Superior Gloves LL100 Cotton 528 4.5 8- Superior Gloves SSL-C13G Kevlar/leather 700 4.5 9- Superior Gloves Cotton 556 4.5 10- Superior Gloves Cotton/rubber 472 4.5 11- Ansell-Edmont 47-200 Cotton/Nitrile 418 4.0 12- Ansell-Edmont 72-025 Spectra 1 269 4.5 Glove prototypes Superior Gloves Prototype 1 Kevlar/cotton/lycra 616 4.5 Superior Gloves Prototype 2 Spectra/cotton/lycra 1 050 4.5 As a result of these series of tests to select the best candidates, only three gloves from the commercial products were selected. These were glove numbers 1, 2 and 3, because they best met the desired specifications. Also, the two prototypes prepared by the glove manufacturer that participated in this project were selected. These gloves were used for 220 field testing. Workers tested each glove for one week on a voluntary basis. The gloves were identified by a number and associated with a worker, which allowed a controlled follow-up of the field testing. A specially developed protocol included a questionnaire on the workers’ dexterity, comfort and durability perceptions after they finished the test. The questionnaire included the following information: how easy it was to handle screws, tools, metal parts, wires, etc.; an evaluation of comfort, glove fit, thermal comfort, grip, skin irritation, etc.; perception of cutting resistance; and an overall assessment. Table 2 presents the results of worker perception of dexterity and comfort. The marks for the answers were as follows:  a positive answer +1  a negative answer -1 The numbers reported in the table represent the average of the answers for all items in the questionnaire for all workers participating in the field testing and for each glove. Gloves are represented by the number given in Table 1. The results in this table show that prototype 2 was the glove most appreciated by the workers for dexterity and comfort. Glove 3 was the least appreciated. Table 3 presents the durability results for the gloves tested. They were obtained by cal- culating the time the worker used the glove before deciding to replace it with a new one. The results represent an average glove durability in relation to the gloves used prior to this project. The results in this table show that all gloves except glove 3 are more durable than the gloves used at the time this project was started. The most durable glove was Prototype 2, with an average durability 13 times longer than the previous glove. Table 2. Results of field testing of gloves for dexterity and comfort. Gloves Dexterity Comfort Prototype 1 (Kevlar-cotton-lycra) 0.30 0.35 Glove 1 (Kevlar) 0.24 0.25 Glove 2 (Kevlar-cotton) 0.43 0.35 Prototype 2 (Spectra-cotton-lycra) 0.86 0.64 Glove 3 (Kevlar) -0.6 -0.38 Table 3. Durability ratio between the gloves tested in this project and the gloves used prior to this project. Prototype 1 Glove 1 Glove 2 Prototype 2 Glove 3 Durability ratio 2 to 16 times longer 2 to 8 times longer 2 to 16 times longer 1.5 to 80 times longer Not available Average 6.2 3.5 6.1 13.1 Number of workers who answered this question 9 9 13 19 3 Table 4. Overall worker assessment of the protective gloves. Rating Prototype 1 Glove 1 Glove 2 Prototype 2 Glove 3 Unacceptable 0 0 0 0 3 Poor 2 2 3 0 0 Average 2 3 5 0 1 Good 4 3 4 9 0 Very good 1 1 1 10 0 Number of workers answering the questionnaire 9 9 13 19 4 221 Table 4 present the results of the overall worker assessment of glove quality based on the characteristics that the gloves should have, namely cut resistance, dexterity, dura- bility, grip and comfort necessary for this type of work. The results in this table show that Prototype 2 stands out from all the other gloves tested, with 9/19 workers finding the gloves good, and 10/19 finding them very good. This glove is made of knitted on Spectra™, Lycra and cotton fibers. Spectra™ is a cut- resistant material, Lycra provides elasticity to obtain a good glove fit, and cotton pro- vides the capacity to absorb oils and greases, which is important for a good grip. Fur- thermore, to improve the grip, PVC dots are placed on the glove palm. Conclusion This study demonstrated that no commercially available glove was adapted to the type of work performed in electrical household appliance assembly operations. As a result of this study, a new protective glove was developed that meets the specifications established in the first part of this study, namely good cut resistance, grip and dexterity. The results also demonstrated that the gloves are well accepted by the workers, which is essential in en- suring that they will be worn. This new glove is commercially available, and it has been demonstrated that it is useful for other types of work operations, such as mirror manu- facturing and cutting, and any kind of glass or metal operations requiring good dexterity. An approach has been developed in this study for assessing the characteristics that gloves must have for working operations. Furthermore, the study combined the expertise of researchers, a glove manufacturer and users in developing, in an iterative exchange, a new glove adapted to the workers. References Bensel CK (1993) The effects of Various Thicknesses of Chemical Protective Gloves on Manual Dexterity. Ergonomics, Vol. 36, No 6, p 687-696. Commission de la santé et de la sécurité du travail du Québec (1999) Fichier sur les lésions professionnelles, 1996-98. Montréal, Québec, mis à jour en septembre 1999. ISO 13997 (1998) Protective Clothing - Mechanical Properties - Determination of Resistance to Cutting by Sharp Objects. International Standard. Johnson RF & Sleeper LA (1986) Effects of Chemical Protective Handwear and Handgear on Manual Dexterity. Proceedings of the Human Factors Society – 30th Annual Meeting. Plumber R, Stobbe T, Ronk R, Myers W, Kim H & Jaraiedi M (1985) Manual Dexterity Evaluation of G Gloves Used in Handling Hazardous Materials. Proceedings of the Human Factors Society – 29th Annual Meeting. Robinette KM, Ervin C & Zehner GF (1986) Dexterity Testing of Chemical DefenseGloves. Tech. Rep., AAMRL-TR- 86-021, Harry G. Amstrong Aerospace Medical Research Laboratory, Wright Paterson Air Force Base, OH. 222 Issues and challenges in chemical protective clothing Jeffrey O. Stull International Personnel Protection, Inc., 10907 Wareham Lane, Austin, Texas 78739, USA Introduction Significant advances have been made in the field of chemical protective clothing. These advances have included new materials offering tremendous improvements in chemical resistance and the development of several standards covering a range of clothing applica- tions. The industry is now maturing but faces new issues and challenges particularly with regard to overuse of permeation data and human factor considerations. Historical progression of chemical protective clothing Starting in the 1970’s, the primary materials used in chemical protective clothing were elastomers such as Neoprene rubber or thermoplastics represented by polyvinyl chloride. With the evolution of emergency response teams and the greater awareness of toxic ef- fects from skin exposure to chemicals in the early 1980’s, the need for broader chemical resistance became important. This trend coincided with the first use of permeation testing to demonstrate the barrier effectiveness of clothing materials. Many material suppliers and clothing manufacturers then began working on new materials with increased chemi- cal resistance. Some of the early developments included chlorinated polyethylene, im- proved versions of polyvinyl chloride, and new laminates such as Viton/butyl rubber. While these materials provided greater resistance, they were usually susceptible to per- meation by one or more classes of chemicals. The first successes for broad-based chemical resistant materials were made in the of- fering of Teflon products and multilayer films. In the United States, ChemFab Corpora- tion applied material technology used in weather-resistant radar covers to clothing mate- rial. At nearly the same time, solubility theory of “like dissolves like” was applied to the construction of multilayered plastic films to achieve what is known today as the 4H mate- rial used in gloves. Both materials were able to provide broad resistance to a number of chemical classes. However, the ChemFab product was relatively expensive and the mul- tilayer plastic laminates did not have the elongation and other properties normally asso- ciated with rubber to allow three-dimension hand forms required for good glove hand function. The establishment of inexpensive plastic film technology led to further development of alternative clothing and glove materials. The introduction of inexpensive nonwoven fab- rics as substrate materials created a new class of lightweight, limited uses clothing products. With specific engineering, these materials were developed to provide the de- sired range of chemical resistance and physical strength. Permeation resistance became an increasingly important factor in chemical protective clothing selection. As limited use fabrics gained importance, a different type of material development was undertaken. These materials were not permeation-resistant, but were designed to prevent the penetration of liquid splashes. While liquid splash protective clothing had been avail- 223 able for many years, primarily in the form of polyvinyl chloride rainwear, this new group of materials offered something different – breathability or water vapour transport. The use of microporous film technology was adapted from filter applications to clothing, by laminating microporous films to nonwoven or woven textile fabrics. To demonstrate the appropriate barrier technology, penetration tests were developed that involved a determi- nation that gross amounts of liquid did not pass through the materials. Key issues and challenges The history of the chemical protective clothing market appears to have met the original industry goals in producing materials that have broad chemical resistance at the molecu- lar level. The market has also provided alternatives for offering varying levels of per- formance to different types of exposure. In Europe, a type classification system for chemical protective clothing has been established that creates a hierarchy of protection from the most extreme hazards to the less innocuous chemical exposures. Despite these achievements, there remain key issues and challenges that remain unaddressed. These include:  overuse or inappropriate use of permeation data  distinguishing between limited use and reusable products  lack of attention to ergonomic issues in clothing design  protection from multiple hazards Use of permeation data The permeation breakthrough time has become the standard measurement by which end users judge material barrier performance. Breakthrough times are now used to classify material chemical resistance and determine product acceptance. A material with good chemical resistance is one that has no reported breakthrough or high breakthrough times against a battery of different chemicals. Implicit in the term breakthrough time is the prevention of any exposure to a given chemical, but in reality, the breakthrough time represents a threshold equal to a specific permeation rate where some chemical passes through the material. With the exception of a few chemicals, the amount of acceptable chemical permeation is generally unknown, as permissible dermal exposure limits are not set in industry as they are for respiratory pro- tection. Presumably, the indication of “no breakthrough” is perceived as appropriate for any clothing (or glove) use up to the breakthrough time. Consequently, any information related to the permeation rate (if reported) is neglected and not applied to clothing use decisions as it would be for the analogous selection of respirators. The majority of permeation testing is conducted with neat organic liquids or con- centrated inorganic acids, bases, and salts. Most testing against gases is performed with the gas challenge at 100 % concentration. In the majority of circumstances, this testing represents a “worst case” exposure with liquid or gas at full “strength” continuously in contact with the clothing over the duration of exposure. In most work situations, employ- ees attempt to minimise their exposure and in some cases are instructed to exit the work area and remove clothing when contaminated. Under these circumstances, it would ap- pear that permeation testing includes a large safety factor. There are also arguments that permeation testing does take into consideration work- place factors. For example, permeation testing is performed at ambient temperatures (21 ºC in Europe and 25 ºC in North America) and does not account for changes in the mate- 224 rials introduced by wear (abrasion and flexing). Many temperatures may occur at ele- vated temperatures. Furthermore, materials in contact with the body will be raised to temperatures near the skin, some 10 degrees warmer than the test temperature. Given the temperature dependence of permeation, increased permeation could be expected. In addi- tion, permeation testing is performed on static samples that do not take into account changes in material resistance produced by wear. Despite the limitations of permeation test data, the information provided by testing is useful if properly applied. Concerns arise when permeation data are used for situations where the type of barrier performance is not consistent with the clothing design or func- tion. For example, the use of permeation data for a clothing item with non-sealed seams against a volatile chemical with known skin absorption points to misuse of the data and perhaps the selection of the wrong type of clothing. Similarly, applying permeation data for clothing products where the main intent is to keep liquid off the wearer’s skin does not make sense and could result in overprotecting the individual. Limited use versus reusable clothing The trend towards limited use or disposable products has attempted to offer an economic alternative and convenience for an end user. However, this trend has also raised the ques- tion as to the appropriate reasons for replacing clothing and acceptable levels of dura- bility. In the marketplace, the distinction between reusable and disposable clothing is based on the purchase price of the clothing. This approach fails to take into account other important factors. Decisions for deciding between reusable and limited use clothing should take into ac- count the clothing life cycle cost, the durability of the clothing, and the ease and effec- tiveness of decontamination. Purchase cost is insufficient for comparing clothing. Costs associated with the maintenance and care of clothing, such as those involved in cleaning, testing, storing, and disposing of clothing must also be considered. Durability is another important factor but is more difficult to assess. Material physical properties such as breaking strength, tear strength, abrasion resistance, and flex fatigue are typically used for evaluating material durability, but may not determine the durability of the overall clothing. Furthermore, different end uses will dictate different levels of clothing dura- bility. The ease of decontamination is also a difficult clothing attribute to assess. Deter- mining effective decontamination requires destructive evaluation of the clothing, but a material with good surface chemical resistance is less likely to be affected when com- pared to a material that might readily absorb chemicals. In addition, when the chemical contaminants and exposure circumstances are well understood, organisations can have greater confidence in ensuring acceptable decontamination. Ergonomic issues Improvements in clothing design have not kept pace with the advances in material tech- nology. The industry now has tests for evaluating the integrity of clothing for exposure to different type of chemical exposure, whether as vapours, liquids, or particles. These tests create the need for interfaces and other design features that restrict wearer movement and reduce comfort. In some cases, ill-fitting and uncomfortable clothing actually creates hazards in the form of reduced vision, tripping hazards, and heat stress. Gloves, footwear, respirators, and other equipment may be incompatible with selected garments. The limited availability of clothing sizes and lack of form-fitting designs create poor fit for end users. Tape is often used to incorporate sizing adjustments and seal off interfaces. 225 Clearly, design improvements must be sought that provide a reasonable balance between protection and wearer function and comfort. Another aspect of clothing comfort can be affected by clothing design and material choices. Overspecification of clothing, particularly in using clothing with more protection than is needed may create more of a hazard to the wearer than the exposure to the chemi- cals present. In particular, if the chemical hazards are not severe, the situation is well characterised, and experience shows exposure is not likely, a lower level of clothing can be selected. For example, breathable penetration-resistant fabrics can be used in place of coated, non-breathable fabrics in a “splash” (liquid-resistant) suit for chemicals that pose no dermal vapour exposure hazards. Table 1 shows a hierarchy of clothing performance based on both clothing integrity and material barrier performance. Table 1. Hierarchy of chemical protective clothing performance. Type of protection Needed garment integrity Needed material barrier performance Vapour-protective Gas-tight Permeation resistance Liquid-protective with vapour dermal hazard Gas-tight Permeation resistance or penetration resistance Liquid protective Jet-tight Penetration resistance Splash protective Spray-tight Repellence Particle protective Particle-tight Particle hold-out Protective clothing against multiple hazards The increasing specialisation of clothing and applications where special protection is needed has grown. For example, while emergency response teams engage in hazardous materials incidents, it is often impossible to segregate different hazards in the incidents they respond to. The most common dual incident is having clothing that offers flash fire protection in combination with chemical protection. Most products must achieve this protection by employing aluminised clothing over chemical suits. This practice increases the bulk of clothing on the wear and requires the sacrifice of a relatively expensive outer- suit. Another example exists for clothing used for abrasive blasting involving harmful skin-toxic particle debris. Special reinforcements of this clothing are necessary to provide the combination of clothing abrasion and particle holdout. Continued improvement of protective clothing designs and materials are necessary for identifying the optimum balance of attributes for appropriate levels of protection. Conclusions Chemical protective clothing is not the final solution in avoiding chemical exposure. Nevertheless, advances in chemical protective clothing have enabled relatively high levels of protection to end users in many applications. Continued improvement of chemi- cal protective clothing and attention to the issues addressed in this paper will further al- low end user additional protection with balanced consideration of comfort, function and other ergonomic needs. 226 Sweat effects on adsorptive capacity of carbon- containing flannel Hubin Li, Jiangge Liu, Lei Li, Zhiqiang Luan Research Institute of Chemical Defence, P.O.Box 925 (West Building), Beijing 100083, P.R.China Introduction Permeable chemical protective suit (PCPS) protect against toxic gases (vapors) mostly depends on activated carbon on PCPS. Because activated carbon on PCPS is contami- nated by human sweat in hot environment, adsorptive capacity for toxic gases is reduced. Sweat effects on adsorptive capacity of chinese carbon-containing flannel were studied. Carbon-containing flannel, inner layer of chinese permeable chemical protective suit, is made of flame-resistance cotton flannel, one surface of it is finished with oil-repellent agent, and the other surface finished with the emulsion of activated carbon and poly- acrylate. Methods Determination of adsorption isotherm Determination by Head-Space Gas Chromatograph (HS-GC), its apparatus is shown in Figure 1. Figure 1. Head-Space gas chromatograph system. Adsorbate: carbon tetrachloride(CCl4) Adsorbent: treated and untreated carbon-containing flannel with sweat and its indi- vidual constituents Adsorption equilibrium time: more than 48 hours, enough to static adsorption equilibrium. 227 Determination of breakthrough curve A schematic of the apparatus for the study of dynamic adsorption of carbon-containing flannel is shown in Figure 2. Figure 2. Schematic of vapor test apparatus. Experimental conditions: Temperature: 20 ºC Relative humidity: 50 % Test agent: CCl4 Inlet concentration of test agent: 0.05~0.20 mg/l Specific velocity of air flow: 2.5~00 cm/min Results and discussion The effects of sweat and its individual constituents on static adsorptive capacity of carbon-containing flannel The effects of the principal constituents of sweat on CCl4 static adsorption capacity of carbon-containing flannel were determined. The data shown in Figure 3 indicate that or- ganic constituents of sweat poison carbon-containing flannel most severely, lactic acid and amino acid are most severe in inhibiting total amount of CCl4 adsorbed and reduce static adsorption capacity of carbon-containing flannel by 24.29 % and 21.37 % with respect to conditioned material. Glucose treatment, surprisingly enhances static adsorp- tion capacity by 8.45 %. 228 Figure 3: Adsorption isotherm of CCl4 on treated and untreated carbon-containing flannel at 42 ºC. Table 1. Comparison of static adsorption capacity. a ( l/g) conditioned sweat lactic acid amino acid inorganic salt glucose urea Cb/Cs=0.175 57.97 43.38 43.89 45.58 57.00 62.87 55.92 Degradation percent (%) -25.17 -24.29 -21.37 -1.67 8.45 -2.05 Protection time analysis Table 2. Comparison of protection time. Condition t0.1(min) C (mg/l) V(cm/min) W(mg/cm2) conditioned sweat lactic acid amino acid 2.45 5 8.45 28.64 19.74 21.37 23.63 Degradation percent (%) -31.08 -25.38 -17.49 Table 2 shows that protection time of carbon-containing flannel contaminated by hu- man sweat is seriously decreased to about 31.08 %. Adsorption rate constant In our laboratory, 54 breakthrough curves of treated and untreated carbon-containing flannel were measured. Adsorptive rate constant k ads can be calculated with Wheeler’s equation, its value as follows: Table 3. Comparison of adsorption rate constant. treatment mode conditioned sweat lactic acid amino acid k ads . V-0.5 1.06 1.46 2.38 2.15 Above-mentioned table shown, the effects of sweat and its individual constituents on adsorption rate of carbon-containing flannel are just a little. 229 Conclusions The experimental results show that organic constituents of sweat are mainly responsible for poisoning action, lactic acid and amino acid are most severe in inhibiting total amount of CCl4 adsorbed and reduce static adsorption capacity of carbon-containing flannel by 24.29 % and 21.37 % with respect to conditioned material. Protection time of CCl4 on carbon-containing flannel is decreased to about 31.08 %, the effects of sweat and its indi- vidual constituents on adsorption rate of carbon-containing flannel are just a little. The cause on protection time of carbon-containing flannel decreased is in sweat reducing its static adsorption capacity. References Liu J (1997) Proceedings of the fifth Scandinavian Symposium on protective clothing, ISBN 87-89895-17- 7,138  142. Aneja A (1975) The effects of perspiration on adsorption dynamics of productive clothing materials, N76-30808. Ferrell JK, Rousseau RW & Aneja AP (1972) Perspiration poisoning of protective clothing materials, AD—A100234, AD—A100235, AD—A098793. Pytlewski LL (1977) Mechanism of activated carbon degradation by perspiration, AD—A063587, AD— A059872, AD—A038144. Hart JA (1979) Sweat resistant gas protective material, US4153745. Schoene K (1983) Proceedings of the First International Symposium on Protection Against CWA, FOA Report C-40171 C2, C3, 215-220. 230 Dynamic elongation test to evaluate the chemical resistance of protective clothing materials Jaime Lara1, Gérald Perron2, Jacques E. Desnoyers2 1 Institut de recherche en santé et en sécurité du travail du Québec, 505 de Maisonneuve Blvd. West, Montreal, Quebec, Canada H3A 3C2 2 INRS-Énergie & matériaux, 1650 Lionel Boulet Blvd., P.O. Box 1020, Varennes, Quebec, Canada J3X 1S2 Introduction To evaluate the resistance of a protective material to chemicals, standardized permeation test methods are generally used. These techniques determine quantitatively the permea- tion of the chemical through the material by measuring the amount of chemical vapors that reach the inside face of the tested material and then evaporate. The breakthrough times and permeation rates can thus be determined. These values characterize the resis- tance of a material to a chemical. The higher the breakthrough time, which usually cor- responds to a low permeation rate, the higher the material’s resistance to the chemical. To obtain reliable permeation results, the chemical must be sufficiently volatile to be detect- able when breakthrough occurs. However, this method does not allow non-volatile (e.g., chemicals with boiling points of 160 °C or higher) or water insoluble chemicals (e.g., pesticides) to be investigated. Despite the work that has been done to develop a method to evaluate a material’s chemical resistance to non-volatile chemicals, no reliable test is currently available (Ehnholt et al, 1988; Bromwich, 1997). In this paper we describe a new test method to characterize the chemical resistance of protective materials by measuring the changes in length of a piece of polymeric mem- brane immersed in the chemical liquid which is followed over time in a specially de- signed cell. This method is suitable for characterizing the chemical resistance of protec- tive materials to chemicals of low volatility. Furthermore, the elongation data can also be used to obtain solubility parameters as demonstrated in this paper. Figure 1. Schematic representation of the elongation test cell. Description of the dynamic elongation test Dynamic elongation tests are carried out by placing a rectangular piece of the sample material, 50 mm long x 4 mm wide, in a glass filter column (Ace-Glass Michel-Miller) containing the solvent (see Figure 1). This cell is 150 mm long x 8 mm inner diameter and closed at both ends ( 7 adapter injection port ACE-thread). A Michel-Miller Teflon® injection port adapter is used at each end. The sample is held in place by two small 10 231 mm x 5 mm metallic sheets attached at the ends of the rods. One rod is fixed while the other can be moved to keep the polymeric band fully extended during the swelling ex- periment. Care must be taken to avoid stretching the material. To start the experiment, the solvent is introduced into the cell through the opening in one of the stoppers placed in one end of the cell. Air in the cell is evacuated through the opening in the other end of the cell. The holes into the Teflon® injection port are enlarged to allow easy movement of the rod inside the cell when the material swells, and also to allow free air and solvent circulation. The membrane sample is completely immersed in the solvent, thus allowing the solvent to diffuse into the entire sample surface. The timer is started as soon as the solvent comes in contact with the polymeric membrane. By measuring the displacement of the mobile rod used to keep the sample extended, the change in length of the sample is followed over time with a precise caliper (millimeter square graph paper can also be used). The initial reference point of the rod corresponds to the extended material without any solvent. Tetra lin - N eoprene (0 .4m m ) 0 50 100 150 200 250 300 T im e (m inutes) 5.0 5.5 6.0 6.5 7.0 7.5 Le n gt h (cm ) 0 0 .04 0.08 0.12 0.16 dL /d t (c m m in - 1 ) 0 50 100 150 200 250 300 350 400 Tim e (m inutes) 5 6 7 Le n gt h (cm ) 0 0.02 0.04 0.06 dL /d t ( cm m in - 1 ) C yclohexane - N eoprene 0 50 100 150 200 250 300 350 400 0 50 100 150 P g m in - 1 cm - 2 A S TM A B 0 50 100 150 200 250 300 350 400 T im e (m inutes) 5.0 5.5 6.0 6.5 7.0 7.5 le n gt h (cm ) 0 0 .01 0.02 0.03 0.04 dL /d t (c m m in - 1 ) Tetra lin - Butyl (0.78m m ) C D Tetralin-Butyl Tetralin-Neoprene Cyclohexane-Neoprene Permeation Elongation Figure 2. Curves A and B are a comparison of results obtained with the ASTM F 739 permea- tion test and the dynamic elongation test on a neoprene glove material with cyclohexane. Curves C and D represent the dynamic elongation tests performed with tetralin on butyl and neoprene glove materials, respectively. Results and discussion Figures 2A and 2B show a comparison of the results obtained with a volatile solvent, cyclohexane, and a neoprene glove material using the ASTM F 739 standard permeation test and the dynamic elongation test. Figure 2A shows that breakthrough occurs at close to 50 minutes. Figure 2B shows that maximum material elongation is reached at close to 40 minutes, which is a shorter time than the observed breakthrough time with the per- meation test. This is because in the elongation test, the material’s overall surface is in contact with the solvent, whereas in the permeation test, only one of the surfaces is in contact with the chemical. A derivative value dL/dt curve, calculated using a second or- der polynomial that fitted the results obtained from the first part of the experiment to the time just before the plateau is reached, is represented on the same figure. The dL/dt values extrapolated to 0 correspond to the time where the plateau of the dynamic elonga- 232 tion curve is obtained. This value was compared to breakthrough values obtained with the permeation tests. Some examples are presented in Table 1. Curves C and D show the results obtained with the dynamic elongation test with a low volatility solvent, tetralin (b.p. 207.2 °C) tested with butyl and neoprene. These curves show that for butyl, it takes close to 100 minutes to reach the maximum elongation, while for neoprene, the maximum is reached in close to 30 minutes. This demonstrates that the solvent penetrates neoprene faster, so that this material has a lower resistance than butyl to tetralin. Table 1 shows some examples of results obtained with the permeation and dynamic elongation tests. Breakthrough times obtained with the ASTM permeation test method are compared to the extrapolated dL/dt values as described above. These results demon- strate that breakthrough time and extrapolated dL/dt values are comparable. This means that the chemical resistance of protective clothing materials can be determined with the dynamic elongation test method. The principal advantage of the dynamic elongation test is that the results are not affected by chemical volatility or by detector sensitivity, which is the main problem with permeation test methods. Table 1. Comparison of breakthrough times obtained with the ASTM permeation test and extrapolated dL/dt values obtained from the dynamic elongation test. Elongation ASTMSolvent Material/Company Min Min Acetone Neoprene/Fairprene 9 7.2 a 12 b 9.5 c Cyclohexane Neoprene/Ansell-Edmont 29-870 40 57 d m-Cresol Neoprene Ansell-Edmont 29-870 220 245 d Nitrobenzene Neoprene/Fairprene 22 22 d aASTM Standard Test Method for Resistance of Protective Clothing Materials to Permeation by Liquid or Gases under Conditions of Continuous Contact F 739-96. bJ. Lara Report IRSST R-104 La résistance des gants aux mélanges de produits chimiques, 1995. cD. Bromwich The validation of a permeation cell for testing chemical protective clothing, American Industrial Hygiene Association Journal, 59:842-851, 1998. dK. Forsberg, L.H. Keith, Instant Gloves+ CPC Database, 1999. Hansen solubility parameters calculated from elongation data The solubility of an organic solvent into a polymeric material has been associated with the resistance of the protective material to chemicals. The higher the solubility of the sol- vent into the protective material, the lower the resistance of the protective material to the chemical. Solubility parameters have therefore been used to predict the resistance of protective materials to chemicals, and consequently, to select the best skin protection. The solubility of an organic solvent into polymers is well described by the use of Hansen three-dimensional (3-D) parameters (Hansen, 1967). From Hildebrand (Hildebrand & Scott, 1950), the solubility parameters for a volatile solvent are repre- sented by the square root of the cohesive energy density =(EV/V)1/2, where E is the va- porization energy, and V, the molar volume. The 3-D solubility concept considers the contribution of three components to the total cohesive energy, namely dispersion, dipole- dipole, and hydrogen bonding forces. The square of the total solubility parameter is rep- resented by the sum of the squares of the Hansen parameters as follows: 2  = 2 D + 2 P + 2 H (1) 233 where subscripts D, P and H refer to partial solubilities corresponding to dispersion, dipole- dipole, and hydrogen bonding forces respectively. If the 3-D Hansen solubility parameters for a polymer and a solvent are known, equa- tion 2 (Hansen and Skaarup, 1967) can be used to determine the degree of solubility of the solvent into the polymer. A = [4( PD - SD )2 + ( PP - SP )2+ ( PH - SH )]1/2 (2) Indices P and S in the equation refer to the solubility values for the polymer and sol- vent respectively. The smaller the A value, the greater the solubility of the solvent into the polymer, and consequently, the lower the resistance of the protective material to the chemical. Hansen 3-D solubility parameters have been published for close to 700 solvents (Hansen, 1999). For polymers, solubility parameters can be experimentally obtained from data on solvent-polymer solubilities (polymer swelling or weight changes) using a battery of solvents with a broad range of 3-D Hansen parameters. In the recent studies of Zellers et al. (1996) different methods were compared to determine the Hansen parameter ex- tracted from the experimental data. In this paper, we report data obtained from the elongation test with a battery of 50 sol- vents for butyl, nitrile, neoprene, latex, and Viton® glove materials. The inverse value for maximum material elongation replaces the A value in equation 2. Thus, 1/!L = A. Using SOLO software, iterative calculations were performed to obtain the best values for the partial 3-D solubility parameters for the five commercial gloves. These values are com- pared to those reported by Zellers et al. (1996). The results are presented in Table 2. Table 2. Comparison of Polymer 3-D solubility parameters determined with three different methods by Zellers et al 1996 and from the elongation data from this study. Butyl Graphic Method Weighted Average* Regression* This study Regression Gd 19.7 17.1 18.4 17.5 Gp 3.5 2.1 - 5.0 2.7 Gh 3.6 2.6 - 0.8 3.7 Nitrile Gd 16.9 17.3 18.9 17.1 Gp 9.7 8.2 8.6 8.0 Gh 8.6 6.2 6.0 7.2 Neoprene Gd 17.0 (6.3-13) 17.5 19.4 17.5 Gp 9.3 4.8 3.6 4.9 Gh 7.6 4.7 4.2 5.5 Latex Gd 17.0 17.2 18.4 17.7 Gp 6.0 2.9 - 3.7 3.8 Gh 5.2 3.4 1.4 3.7 Viton Gd 17.0 17.0 Gp 10.3 8.7 Gh 6.1 8.2 Zellers studied two methods for determining the 3-D solubility parameters as alterna- tives to the standard graphical method. With one of the methods, the solubility parame- ters are determined from the weighted average of the solubility parameters of the sol- vents, where the weighting factor is the product of the solvent molar volume and the 234 fractional uptake of the solvent measured by immersion testing. In the other method, Zellers uses a multiple non-linear regression to estimate the 3-D solubility parameters. These two methods are presented as better alternatives to the arbitrary graphical solution. The results presented in Table 2 demonstrate that the calculated 3-D solubility parameter values for the five commercial products compare well to the ones obtained by Zellers using the weighted average method. Conclusion The study demonstrated that the chemical resistance of materials to low volatility sol- vents can be easily characterized by the dynamic elongation test method. This technique can be used with solvents where standard techniques based on the analysis of solvents that vaporize after breaking through the membrane cannot be applied. Furthermore, the results obtained from the elongation test can be used to calculate the Hansen 3-D solu- bility parameters. These values can be used to estimate the resistance of the polymeric materials used in protective clothing to chemical attack. References Bromwich DW (1997) The Design of Permeation Cells for Testing Chemical Protective Clothing. Proceedings of the Sixteenth Annual Conference of Australian Institute of Occupational Hygienists, Melbourne: the Australian Institute of Occupational Hygienists, pp. 33-39. Ehntholt DE, Almeida RF, Beltis KJ, Crundolo DL, Schwope AD, Whelan RH, Royer MD & Nielsen AP (1988) Test Method Development and Evaluation of Protective Clothing Items Used in Agricultural Pesticide Operations. Performance of Protective Clothing: Second Symposium, ASTM STP 989, Mansdorf SZ, Sager R & Nielsen AP (eds.), American Society for Testing and Materials, Philadelphia, pp. 727-737. Hansen CM (1967) The Three Dimensional Solubility Parameter - Key to Paint Component Affinities I. J. Paint Technol., 39(505), 104-117. Hansen CM (1999) Hansen Solubility Parameters - a User's Handbook, CRC Press LLC, Boca Raton, Florida. Hildebrand J & Scott RL (1950) The Solubility of Nonelectrolytes. 3rd Ed., Reihnold, New York. Lara J, Roberge B, Velazquez A & Nelisse H (1991) Chemical Permeation Test of Commercially Available Gloves Using the ASTM F 1001 Standard. IRSST Report R-050, Montreal, QC, Canada. Zellers ET, Anna DH, Sulewski R & Wei X (1996) Improved Methods for Determination of Hansen's Solubility Parameters and the Estimation of Solvent Uptake for Lightly Crosslinked Polymers. Journal of Applied Polymer Science, Vol. 82, 2081-2098. 235 Physiological strain and wear comfort while wearing a chemical protective suit with breathing apparatus inside and outside the suit in summer and in winter Raija Ilmarinen1, Harri Lindholm1, Kari Koivistoinen2, Petteri Helistén2 1Finnish Institute of Occupational Health, Vantaa, Finland 2Finnish Emergency Services College. Kuopio, Finland Introduction Rescue work while an impermeable chemical protective suit (CPS) is being worn is physically and psychologically one of the most demanding tasks for a fire fighter. The use of CPS in operational work may impose even greater physiological strain in fire fighters than the use of turnout suit when entering into smoke-filled enclosures, which is well documented (Ilmarinen et al., 1994; 1997; 1998). Most studies on the effects of an impermeable chemical, gas or NBC protective clothing are conducted under controlled laboratory conditions at temperate or hot environment using test protocols which are very far from real operational rescue work. The effects of CPS under cool or cold conditions are much less studied (Smolander et al., 1984; Cortili et al., 1996; Rissanen 1998). The purpose of the present study was to investigate the physiological responses in fire fighters while wearing an impermeable chemical protective suit with self-contained breathing apparatus (SCBA) outside and inside the suit during a simulated chemical acci- dent at an outdoor processing plant both in summer and winter. Additionally, we at- tempted to find out the effects on the work performance, the wear comfort and function of the suit systems at work. Methods The voluntary subjects comprised 8 experienced healthy male fire-fighting trainers with an average age of 38.6 (31-44) years, height of 183.5 (178-190) cm, weight of 88.3 (72- 110) kg, body fat of 14.6 (9.2-18.9) %, body surface area of 2.1 (1.9-2.3) m2, BMI of 26.2 (22.6-31.9), and V O2max of 51.6 (46-60) ml/kg min-1. Chemical protective equipment system (CPES) consisted of pants, cotton underwear with long sleeves and legs, polyester fleece sweat shirt and trousers, wool underhood, wool socks, cotton undergloves, helmet and an impermeable CPS. Two types of the CPS were studied: The SCBA (Dräger PA 90/6 l, approx. 16 kg ) was carried either outside (SuitA ) or inside (SuitB) the CPS. The material of both CPS was butyl rubber. SuitA weighted 5.5 kg and SuitB 7.8 kg. Correspondingly, air flow rate of air supply to CPS were 4 l/min and 2 l/min. The total mass of the CPES averaged 25,5 kg for SuitA system and 27 kg for SuitB system. Experimental procedure and measurements: The test protocol was developed to repro- duce situations encountered in an actual chemical accident. The drill was conducted out- doors in the rescue training area of the Finnish Emergency Services College in Kuopio at air temperature ranging from 13 to 20 C in summer and from -11 to -20 C in winter. The work task was 'sealing the leak in the flange in a chemical processing simulator'. The 236 drill was divided into two consecutive work sessions (WS) with a 20-min rest between the sessions for body cooling (partly doffing the CPS for ventilation of underclothing), drinking ad libitum and chancing the air container. In order to perform the demanded task the fire fighters have to walk and search, carry a pressurized hose (length of 80 m, weight of 96 kg) climb stairs and ladders, carry two 35 kg heavy cans about 100 m, seal the flunge and finally decontaminate themselves. The drill was performed in fire fighter pairs with their own speed. The CPS system was randomized for each pair. The performance time was recorded. Physiological registration included heart rate (HR), rectal (T re ) and skin temperatures (T sk) sweat loss and blood pressure (BP). The change in heat storage for heat exposure time was calculated from changes in mean body temperature using 0.97 Wh/kg·°C for specific heat of the body. Subjective ratings of per- ceived exertion and thermal comfort were requested and a questionnaire on the wear comfort and function of CPS at work was fulfilled. The measuring and evaluation methods are published elsewhere (Ilmarinen & Lindholm, 2000). The protocol was approved by the Institutional Ethics Committee and the written in- formed consent of the subjects was obtained before the experimental sessions. The test drill was terminated if one of the following criteria was met: 1) emptying of the air con- tainer 2) T re  39.5C with subjective signs of severe discomfort or fatigue, chest pain or intense muscle pain, and 3) objective signs of exhaustion and exertional dyspnea or dizziness. Statistics. Means SD, ranges and medians were used for description of the data and non- parametric tests were used to calculate differences between test conditions. The < 0.05 level of probability was accepted as significant. Results and discussion All test drills were completed. The average performance times for all the drills are in Figure 1. The interindividual variation in performance times was only some minutes. In summer the average performance time for SuitB (38 min) was significantly longer than for SuitA (24 min). In winter the difference between the suit systems was smaller (7 min) but still significant. Snow, partly even ice covered the ground and ladders, which ham- pered movements thus lengthening the time to complete the task with both suit systems. The difference between summer and winter was significant only for SuitA. 0 5 10 15 20 25 30 35 40 45 SuitB Winter SuitA SuitB Summer SuitA Working time, min 1. WS 2. WS 1. WS + 2. WS 1 + 2 Figure 1. Average performance times for the WSs and completed drills while wearing CPS with SCBA outside SuitA and inside SuitB. 237 The higher work speed for SuitA resulted in higher average ventilation rates of 80 l/min in summer and 70 l/min in winter than corresponding average ventilation rates for SuitB being 65 l/min both in summer and winter. The differences were not signifi- cant. After rapid increase from the beginning of work HR fluctuated at the near submaximal level during the WSs. The higher work speed in summer also resulted in considerably higher HR levels than in winter. The highest values were measured by using SuitB. On average, median HR (including both WSs and rest) was in summer 131 bpm for SuitA and 145 bpm for SuitB, and correspondingly, in winter 118 bpm and 126 bpm. The average peak HR in summer was 163 (149-173) bpm for SuitA and 168 (136-188) bpm for SuitB. In winter the respective values were 153 (135-173) bpm and 161 (148-178) bpm. The circulatory load was expressed as a time fraction of HR >75% of the individual maximal HR (HR75%). In summer HR75% was on average 21 (9-37) % of total working time for SuitA and 37 (3-54) % for SuitB. The difference was statistically significant. In win- ter the time fractions were 13 (1-31) % and 24 (11-33) %. The difference was not significant. There were no differences in systolic BP between the suits or between the seasons. Diastolic BP was higher for both suits in winter than in summer but the difference was not significant. Individual variation in T re responses was considerable. At the start of working T re ranged from 36.2 C to 37.7 C. In summer the work resulted in average 0.8 C increase in T re regardless of the suit system. In winter the average increase was 0.2 C for SuitA and 0.4 C for SuitB. Average T sk for SuitB fluctuated during the drill in all conditions at higher levels than for SuitA, on average but the difference between the suits was not significant. Median values for T sk (including both WSs and rest) averaged 33.9 C for SuitA and 34.7 C for SuitB, and correspondingly, in winter 31.6 C and 32.0 C. The individual rates of heat storage varied in summer from 16 to 41 W/m2 and in win- ter from -4 to 23 W/m2 which resulted greater heat gain in SuitB both in summer and winter (Figure 2). The difference was significant only in summer. In winter the heat gain was significantly smaller for both suit systems than in summer. No significant differences were found in sweat rates between the suit systems. How- ever, the average sweat rates for both suit systems were significantly greater in summer than in winter being of 657(463-912) g/h m-2 and of 233 (72-461) g/h m-2 for SuitA and correspondingly, 558 (430-898) g/h m-2 and 305(157-453) g/h m-2 for SuitB. -10 0 10 20 30 40 50 Summer WinterC ha ng e in h ea t s to ra ge , W /m 2 SuitA SuitB Figure 2. Mean (rSD) changes in body heat storage in summer and in winter while wearing CPS with SCBA outside SuitA and inside SuitB. 238 Donning and doffing SuitB without help was impossible for experienced fire fighters contrary to SuitA, which everyone managed to don themselves. Restricted movement and especially the loss of vision caused by the misting of the visor imposed an additional stress for the wearer by using SuitB and the work was perceived as being 'hard' both in summer and winter, on average. Some fire fighters perceived the physical work as being 'very very hard'. Correspondingly, physical work by using SuitA was reported 'somewhat hard', on average. Only some fire fighters reported 'hard'. In summer SuitB was con- sidered significantly warmer and significantly more uncomfortable than Suit A. Also in winter SuitB was significantly warmer. There was no difference in average ratings of skin wettedness between the suits but the individual variation was considerable. The skin was felt to be form 'clammy' to 'wet' in summer and from 'nearly dry' to 'wet' in winter. Conclusions The results indicate longer performance time, significantly greater thermal and cardio- vascular strain as well as more intense subjective discomfort in fire fighters both in sum- mer and winter while working dressed in fully encapsulating impermeable CPS with SCBA worn inside the suit. The performance of fire fighters and the success of rescue operations are adversely affected by all these facts. Based on the results it is concluded that a CPS with SCBA worn outside the suit is a more ergonomic, safer and more suitable personal protective clothing for use in Nordic climatic conditions. Acknowledgements The authors appreciate financial support from the Nordic Council of Ministers and the technical assistance of the personnel of the Finnish Emergency Services College and the Dept. of Physics of the Finnish Institute of Occupational Health. References Cortili G, Mognoni P, Saibene SD (1996) Work tolerance and physiological responses to thermal environment while wearing protective NBC clothing. Ergonomics 39(4): 620-633. Ilmarinen R, Mäkinen H, Griefahn B, Künemund C (1994) Physiological responses to wearing a fire fighter's turnout suit with and without a microporous membrane in the heat. In: Frim J, Ducharme MB, Tikuisis P, eds. Proc. of the Sixth Int. Conf. on Environ. Erg. Montebello, Canada. North York: Defence & Civil Institute of Environmental Medicine,:78-79. Ilmarinen R, Louhevaara V, Griefahn B, Künemund C (1997). Thermal responses to consecutive strenuous fire-fighting and rescue tasks in the heat. In: Shapioro Y,. Moran DS , Epstein Y eds. Environmental Physiology - Recent Progress and New Frontiers. Freund Publishing House, Ltd. London and Tel Aviv, 295-298. Ilmarinen R & Koivistoinen K (1998) Heart rate and thermal responses in prolonged job-related firefighting drills. Hodgdon JA & Heaney JH eds. ICEE 8, Program and Abstracts, San Diego, California, 38. Ilmarinen R & Lindholm H (2000) The influence of age and gender on thremoregulatory control during heat stress. Proc. of 9th ICEE Ruhr 2000, Bochum, Germany (in press). Rissanen S. Quantification of thermal responses while wearing fully encapsulating protective clothing in warm and cold environments.1998 Acta Univ. Oul. D 486, Doctoral thesis. Smolander J, Louhevaara V, Tuomi T, Korhonen O, Jaakkola J. 1984 Reduction of isometric muscle endurance after wearing impermeable gas protective clothing. Eur J Appl Physiol; 53: 76-80. 239 Performance criteria for PPE in agri- and horticulture Torsten Hinz1, Eberhardt Hoernicke2 1 Bundesforschungsanstalt für Landwirtschaft (FAL), Braunschweig, Deutschland 2 Biologische Bundesanstalt für Land- und Forstwirtschaft (BBA), Braunschweig, Deutschland Introduction At their working places farmers are affected by various kinds of load which result from the ambient conditions (climate) and the work process itself e.g. by emitting noise or air- borne contaminants. These can be gases or particles, which form a risk potential for the farmers’ health and welfare by possibly harmful substances like pesticides. Also the bio- logical content of aerosols – germs, bacteria, and fungi – must be taken into account. All possible loads have to be kept on low levels minimizing risk. If load can’t be re- duced means of protection have to be taken at least PPE (personal protective equipment) at the end of the chain of labour protection. According to the EU directive 89/686 EWG all PPE components on the European Market have to be labelled and certified that means that the equipment has to be tested with standard procedures and graded by given re- quirements on the performance. Those requirements will be derived from kind and situa- tion of exposure time and level. Limit values can be given by national or international responsibilities. After a risk assessment defining necessary values of protection in agri- and horticulture additional questions of performance must be answered e.g. thermal com- fort, prize and acceptance by the farmers and their environment. Here the situation is quite different to industrial conditions for the use of PPE, which is mostly understood as a mean for accidental prevention. In the following performance criteria for PPE in agri- culture will be given for particular cases of load and protection. Outgoing from the example of risk assessment in pesticide (plant protection products, PPP) use it will be shown how these requirements will be stated. Methodology / Application of PPE Food and agricultural non-food production is characterised by various kinds of working places. It is to distinguish between work indoor e.g. in livestock-buildings or greenhouses and outdoor stationary or mobile that means on propelled harvesters or tractors with / without mountings. Apart from climate particular stress factors result from particular work and requires specific means of protection. Table 1 gives an overview about PPE and examples of its application in agriculture, prescribed in Germany (accident prevention regulation, 2000). A special case form airborne contaminants in agri- and horticulture, which include pos- sible risks by accidents and by long termed effects. Particles and gases may influence farmers health and welfare by dermal or respiratory uptake. Gases are to consider during mixing manure, handling silage and during gas application for conservation, or control- ling pests. Particles emitted in livestock-buildings or in field operations like harvesting influence mostly respiration by their possible allergic potential (Hinz et al., 1990). 240 Table 1. PPE and examples of application in agriculture. PPE protecting work field respiration use of pesticides, disinfectants, work in dusty atmosphere head handling with lifted goods, operating on trees foot handling agric. mobiles, contact with claw and hoof animals eye and face work with saws and cutter, handling chemicals hearing work on non capsulated mobiles, with motor saw, feeding pigs hand work with saws cutter, handling chemicals body work with motor saws, handling chemicals, pesticide application Table 2. Main sources of airborne contaminants in agri- and horticulture. source contaminant effect manure gas respiratory silage, bio-gas gas respiratory greenhouse aerosol / gas respiratory / dermal animal house aerosol / dust respiratory / dermal field operating dust respiratory plant protection aerosol / gas / dust respiratory / dermal A higher risk potential will be given by the use of chemicals – disinfectants and espe- cially pesticides. These are signed according to the hazardous substances ordinance (1986) if they are high-toxic (T+), toxic (T), caustic, harmful or irritating. The distribution of pesticides in Germany is given in Figure1 and shows, that the ratio of risky pesticides is much lower than normally expected, only 6 % of the pesticides are classified with T+ or T which will be used mostly for rodenticides. harmful 27% irritating 16% nothing 51% toxic 3% high toxic 3% caustic 0,1% Figure 1. Distribution of pesticides regarding to possible effects to men. But the influence of those possibly hazardous substances is only one factor load to- gether and must be seen with exposure level and time. Because of a probably remaining portion of risk for all pesticides which shall be distributed in Germany a special risk as- sessment has to be presented to the body of authorisation. For this purpose a risk man- agement model has been developed which is discussed to be a European harmonized model at the present time (Lundehn et al., 1992). Depart from the more simple idea of accident prevention this model shows some peculiarity e.g. by the introduction of dose effect functions instead of the break through criteria. The main idea of the model is the calculation of exposure whereby all kinds of work and possible ways of uptake are to consider: undiluted pesticide while mixing and loading - short termed, the process of ap- plication - long termed and additional possible jobs of maintenance on the field, before and after spraying. The oral, respiratory and the dermal path of uptake must be deter- mined. So calculated over all exposure must be compared with a relevant toxicological limit value which results from special tests and will be prescribed mostly by national 241 authority. In case of pesticides this value is the acceptable operator exposure level (AOEL) which is required by an EU directive. In case that the ratio of calculated expo- sure and the AOEL is greater one, means of load reduction e.g. using other procedures of application or less harmful agents or PPE are to be fixed. One key point of the model is the commitment of reduction factors to maximum allowable values of penetration through the protective material. These values are given in Table 3 with e.g. 5 % for suits and 1 % for gloves. Table 3. Reductions coefficients of performance of PPE. protective mean reduction dermal coefficient inhalation universal protective gloves (plant protection) 0.01 standard protective garment (plant protective) and sturdy footwear 0.05 protective clothing against chemicals; type 1 0 broad-brimmed headgear of sturdy fabric 0.5 hood and visor 0.05 particle filtering half-mask FF2-SL or half-mask with particle filter P2 0.8 0.08 half-mask with combination filter AIP2 0.8 0.02 Plant protection products are only authorized, if the calculated exposure is tolerable for instance by use of PPE. In case of severe special kinds of exposure e.g. while application in greenhouses or in stock protection (fumigation) high protecting suits of type 1 chemi- cal protective clothing may be required. Besides of the performance requirements of protection more measures are requested for gloves and especially suits, which will be used while handling pesticides. Looking to the long time job of spraying thermal comfort is an important factor. To ensure it water vapour resistance must not pass over a maximum a value, just like the mechanical properties. One really important point concerning the acceptance of these PPE by the farmer is the design and the colour of the equipment. It must be taken into account that the work is in the outdoor area, which is mostly used for vacation of the town people. Means which may represent high degree of protection by view will never been worn in sensitive regions of fruit production or vineyards. These criteria of performance are given in Table 4. Table 4. Additional criteria of performance of PPE. criterion measure limit value strength tensile strength tear resistance longitudinal 600 N cross 400 N 25 N thermal comfort water vapour resistance 20 m2Pa/W acceptance design, colour, prize availability Depending on the load PPE will be worn in agri-horticulture for the different types of work as given in Table 1. According to the EU directive those means have to been tested by standard procedures of harmonized standards so far available. In other cases national norms or guidelines are to use e.g. as done in Germany with the BBA guideline for PPE in plant protection (BBA, 1993). Table 5 shows the network of existing standards of testing various parameters of PPE performance. 242 Table 5. CEN standards for testing performance of PPE. performance property test method standard respiration EN 136, EN 140, EN 146, 147, 149 eye/face protection EN 166 foot protection EN 345 hearing EN 352, EN 458 hand protection EN 374, EN 388, EN 381 - 7 body EN 340, EN 465, EN 470, draft DIN 32780 – 300, prENISO 13982 – 2, EN 468, EN 463, EN 368, EN 369 Conclusion Many working places in agriculture and horticulture effect farmers’ health and welfare in various kinds of load. Examples are mechanical or heat stress as well the exposure by airborne contaminants, which may harm or irritate skin or breathing. In such cases the possible risks have to be limited by reducing load or by the use of different types of PPE protecting head, foot, body or respiration. This equipment must be tested and certified according the EU directive 89/686 EWG using CEN standard so far available. In other cases national standards or guidelines may be used as done in Germany with the defini- tion of PPE to protect the applicator of plant protection products (PPP). At the present time a work item in CEN is defined to introduce the test procedure as a European standard. References BBA 1993 (1993) Richtlinien für die Prüfung von Pflanzenschutzmitteln im Zulassungsverfahren Teil I 3- 3, Kennzeichnung von Pflanzenschutzmitteln – Gesundheitsschutz. Biologische Bundesanstalt für Land- und Forstwirtschaft der Bundesrepublik Deutschland. Hinz T, Krause K-H, Stalder K (1990) Gesundheitsgefährdungen durch Stäube beim Einsatz von Mähdreschern und Möglichkeiten technischen Arbeitsschutzes. Jahrestagung der Deutschen Gesellschaft für Arbeitsmedizin, Frankfurt-Höchst, 28./31.05.1990, Abstracts Band. Lundehn et al. (1992) Einheitliche Grundsätze zur Sicherung des Gesundheitsschutzes für den Anwender von Pflanzenschutzmitteln Mitteilungen aus der Biologischen Bundesanstalt für Land- und Forstwirtschaft, Heft 277. Unfallverhütungsvorschrift (accident prevention regulation) der landwirtschaftlichen Berufsgenossenschaften (2000). Verordnung über gefährliche Stoffe. Gefahrstoffverordnung (hazardous substance ordination) vom 26.08.1986 (1986) (BGBl. S. 1470 und Änderungen). 243 Limits of recycling in protective apparel Serhiy Zavadsky 1 Kevlar“ Marketing, DuPont Engineering Fibres, Du Pont de Nemours International S.A., 2, chemin du Pavillon, P.o.Box 50, Le Grand-Saconnex, CH-1218, Switzerland Recycling has grown significantly during the last 10 years in Europe. Different compa- nies have created technologies and established commercial supply chains for collecting and so-called “recycling” or “regenerating” fibres and aramids in particular. Globally speaking this is a positive initiative. DuPont is committed to sustainable growth as a matter of fundamental policy and supports reasonable activities in this area. Our own DuPont Engineering Fibres group is working to introduce a special program to collect and recycle aramid waste (which can occur in processing textiles or at the end of the useful life of garments). Our aim is reliable quality of the resulting fiber products and finished articles that may be achieved. Together with our partners, several interesting opportunities for utilization of recycled para-aramid fibers have been identified. These applications do not require the high level of performance of original (virgin) KEVLAR® brand fibre and can accept a significant variation of properties. It is clear that, after recycling, the properties of the regenerated fibres vary significantly more when compared to original fibres. Variability includes cut length, impurities in the bale, but also fundamental strength, modulus, cut resistance, heat performance properties, etc. The virtues of WELL CONTROLLED recycling of aramids can be easily understood and are generally accepted as responsible good business practice. DuPont would like to invite the market to join in an industry-wide consensus to apply reasonable discipline in the collection, recycling and processing of “second life” fibre. We all need to stop the use of ‘recycled’ or ‘regenerated’ aramid products in applications which are critical for human life and nature, such as ballistic and stab resistant vests and other armor, protective apparel, and safety equipment. We strongly believe and have evidence that improper use in these fields could lead to se- rious accidents, injuries and human suffering. Unfortunately, nowadays, a visitor to fairs related to protective apparel will easily find several stands with gloves and garments produced from so-called ‘regenerated aramid’. There are several problems in allowing the existing situation to go on or grow. 1. End-users are not informed that these products are made of ‘regenerated’ aramids and hence cause dangerous differences in performance. 2. Some manufacturers even disguise the fact by using the terminology of virgin fibre marketing, misrepresenting their offering by fraudulent use of trademarks and fibre brand names. 3. Workers and specifiers do not know or understand the difference in performance between ‘regenerated’ and original (virgin) aramid fibres in products made out of them and generally trust ‘regenerated’ product as much as they trust virgin fibres. 4. Safety engineers put themselves, and the people they are responsible for, at risk. For even if the product will pass the relevant European Norms, nobody can guarantee that properties do not vary significantly from one individual item or garment to another. This could lead to serious accidents. 244 5. The much lower cost and price of products made out of ‘regenerated’ aramids reduces significantly the perceived value of products made out of virgin aramid fibres. The failure of “regenerated” products is ascribed to the “virgin” as well, because nobody so far is asking about differences in performance as soon as it is stated: “Made of ‘aramid’ fibre”. 6. Beyond the waste aramid problem there are other examples of cheating in the market. People sell yellow cotton gloves under the name KEVLAR®, DuPont's registered trademark and brand, hoping to avoid a court case they may use 1 g of aramid sewing thread, for example. You can imagine how many possibilities for cheating, imitating and copying one can have with the un-controlled use of regenerated fibres. You can probably cite more examples of the problem. But even with above 6 examples, we consider the existing situation with ‘recycled’ aramids as very dangerous for the PPE market, the workforce and the general consumer. DuPont are undertaking immediate action in the marketplace and in the legal arena. DuPont already started, together with our partners, to establish its own chain of col- lecting and recycling KEVLAR® brand fibres and other aramids.  We will control strictly where recycled aramids end up.  We are putting in place an intensive campaign to inform safety engineers throughout Europe about the hazards associated with waste in safety products.  We will publish, with the agreement of companies negatively affected by un- controlled products, real cases of accidents, which resulted in injuries.  We are pursuing before the court of law those unprincipled enough to mislead the public through the unqualified and un-authorized use of our trade names.  We also invite you to support our efforts in managing a proper use of high-value re- cycled aramids and avoid their misuse in safety applications. 245 Protective clothing and survival at sea Hilde Færevik Department of Health and Work Physiology, SINTEF Unimed, N-7465 Trondheim, Norway Introduction During the past 50 years, the increase in maritime and intercontinental air traffic and the introduction of helicopter transport to offshore oil platforms have increased the risk of being exposed to accidental immersion in cold water, while a sharper focus on occupa- tional health and safety has brought this matter to more general attention. Offshore oil workers, fish-farm workers, fishermen and military personnel are all, through their occu- pations, at risk of falling into the water. A wide variety of protective clothing and equip- ment has therefore had to be designed to meet the range of requirements of these user groups. The area of survival at sea is a multidisciplinary, bringing together expert from a wide variety of fields including policy-making, manufacturing, industry and science. In order to ensure the best possible performance of protective clothing and equipment, a better understanding of the impact of cold water on human physiological responses is es- sential. In the following paragraphs, hypothermia and other critical events during acci- dental immersion in cold water are briefly reviewed. The importance of maintaining heat balance during exposure to cold water is discussed in terms of thermodynamic laws and physiology. The design and development of personal protective clothing and equipment for aircrew are also discussed in order to illustrate the difficulty of reconciling thermal protection in water with other user requirements. When the Titanic sank in 1912, 1498 people lost their lives. All deaths were attributed to drowning and there was no mention of hypothermia in reports of the disaster. Even though the importance of heat loss from the body when immersed in cold water has been known since the work of Lefèvre at the end of the last century (Lefèvre., 1898), not until after the Second World War did it become clear that cold was also a major cause of death after shipwrecks (Keatinge, 1969). The large number of ships and aircraft lost at sea forced the authorities of the countries at war to develop methods of protecting crews against cold water immersion. Research was done on designing protective equipment and on determining survival time in relation to water temperature. Much research during this period focused on the body insulation factor and how it is influenced by body measures. Using the US Navy’s well-documented information on shipwrecks between 1942 and 1945, Molnar (1946) was the first to publish a tolerance curve based on the temperature in the water and the total time of immersion. His research also revealed wide individual variations in the length of time that people could survive in the water. In the course of the past 10 years a number of mathematical models that estimate human survival time under various conditions have been outlined. Such models are useful, but must be viewed with caution since reliable data concerning deep hypothermia are unavailable (Tikusis, 1997). Data obtained from case reports of accidental immersion can only provide a posteriori proof of the value of protective equipment. It is therefore necessary to assess the process of heat exchange between the body and its environment by studying both general physiological reactions and individual variations in the study of tolerance to cold-water immersion. Recently, much research on accidents at sea has focused not only on the 246 cooling of the body, but also on secondary physiological responses associated with im- mersion in cold water (Tipton, 1989). When accidents occur, the chances of survival are critically dependent on the performance of protective equipment, including protective clothing and buoyancy devices. Although protective equipment has indeed saved lives, there are numerous reports of equipment malfunctioning during emergency situations. A recent example is the Sleipner accident off the coast of Norway, in which 16 of the 88 passengers of a high-speed ferry lost their lives. The surviving passengers reported se- rious problems in donning life-vests, and the rafts carried on board were not functional. This emphasises that there is still much to be done to improve safety when immersion accidents occur. Direct effects of the cold; hypothermia The innate protection mechanisms of the human body against a cold environment are very limited. This is explained physiologically by the fact that man is a homeothermic organism and that our thermoregulatory system is optimised to maintain a deep body temperature of around 37 °C. In contrast, reptiles and amphibians are able to tolerate large fluctuations in their body temperature, avoiding freezing by behavioural avoidance or physiological adaptation (Withers, 1992). Some mammals and birds utilise a regulated lowering of the body temperature to lower their daily energy expenditure. A human being can tolerate a variation of only about 4 °C in deep body temperature without impairment of his physical and mental performance. Any greater change in body temperature will affect cellular structures, enzyme systems and a wide range of temperature-dependent chemical reactions that occur in the body (Åstrand & Rodahl, 1986). Hypothermia is defined as a condition in which the deep body temperature falls below 35 °C. It usually occurs accidentally, but can also be induced deliberately as part of a therapeutic regime, e.g. open heart surgery. There are three physiologically distinct types of accidental hypothermia; immersion hypothermia, exhaustion hypothermia and urban hypothermia. Immersion hypothermia is the type we must consider in accidents at sea, and it includes the most severe cold stress. This hypothermia is generally divided into three distinct types according to the degree of body cooling; mild (body temperature of 34-35 °C), moderate (body temperature of 30-34 °C) and deep hypothermia (body temperature < 30 °C). Mild hypothermia is characterised by changes in peripheral resistance due to vasoconstriction, thermogenic shivering and tachycardia (Lexow, 1989). The victim is usually conscious and respon- sive. Peripheral vasoconstriction leads to an increase in central blood volume, which in turn induces diuresis. This cold diuresis is mediated through hormonal responses (e.g. antidiuretic hormone, atrial natriuretic peptide, aldosterone) that reduce the reabsorption of water and sodium by the kidneys (Vander et al., 1994, Ganong, 1997). As a conse- quence of cold diuresis and the redistribution of water from vascular space to extracellu- lar space, the hypothermic victim will often be dehydrated (Popovic, 1974). Thermogenic shivering is normally most intense at 35 °C and causes a three- to five-fold increase in heat production. At moderate hypothermia shivering gradually decreases and heat production declines. The victim’s consciousness is clouded and there is increased muscular rigidity with the result that muscular co-ordination is impaired (Bristow, 1984). The cold affects the myo- cardial conduction system, inducing gradual cardiac slowing. Below 32 °C central body temperature cardiac arrhythmia develops and ventricular fibrillation may occur if the 247 heart is irritated (Bristow, 1984). There is a gradual fall in blood pressure because of bra- dycardia and a fall in peripheral vascular resistance. Deep hypothermia is a life-threatening condition. The victim is usually comatose, the skin is pale and the pupils are dilated and unresponsive to light. There is pronounced bra- dycardia and at 18-20 °C the heart usually stops. Respiration and pulse are difficult to register below 20 °C deep body temperature, and it is impossible to measure blood pres- sure. However, due to a decrease in the metabolic rate, the oxygen requirements of the brain are greatly lowered, which implies that hypothermia actually offers some protection against hypoxia. Indirect effects of the cold Hypothermia is not always the direct cause of death, but may it induce other lethal ef- fects. Immediately after immersion in cold water, a so-called “cold-shock response” is induced. Tipton (1989) has excellently reviewed this response. The “cold shock” is de- fined as the range of the initial responses when the victim falls into cold water, including cardiovascular and respiratory disturbances that may cause drowning and other fatal ef- fects. The “cold shock response” is probably responsible for the majority of the 400-1000 open-water deaths that occur annually in the UK (ROSPA, 1988). The cardiovascular responses include tachycardia, increased cardiac output and pe- ripheral vasoconstriction (reduced peripheral bloodflow). As a result of both cardiac and vascular responses there is a dramatic increase in the blood pressure of the victim. The rapid increase in cardiac output and blood pressure increases the workload of the heart. This may cause greater ventricular irritability, cardiac irregularities and on rare occasions ventricular fibrillation (Tipton, 1989). The respiratory responses of cold shock are a greater threat to survival than the initial cardiovascular responses to cold water immer- sion. The sudden decrease in cutaneous temperature when exposed to cold water induces an inspiratory gasp reflex followed by hyperventilation (Mekjavic et al. 1987). Hyper- ventilation causes respiratory alkalosis and hypocapnia, but the most dangerous conse- quence is that the victim is not able to control his breathing. This, together with the powerful discomfort of the cold, which causes the victim to swim very clumsily, is be- lieved to be one of the major causative factors in the mechanisms of cold-water swim- ming failure (Golden & Hardcastle, 1982). It has been demonstrated that even good swimmers are not able to swim for more than a few minutes in cold water (Golden & Hardcastle, 1982). The maximum breath-holding times for normally clothed individuals are reduced to less than 10 sec when exposed to cold water (Tipton & Vincent, 1989). A reduction of as much as 30-60 % in maximal breath-holding time has been reported (Hayward et al., 1984). This is extremely critical if the immersion occurs in choppy water or includes submersion from a ditched helicopter or a vehicle. The loss of both breath- holding time and voluntarily control over breathing increases the chances of aspirating water and drowning. Victims of accidental immersion may collapse and die during the process of rescue or shortly afterwards. Several hypotheses exist regarding the cause of this phenomenon. Deaths have been attributed to an after-drop in observed core body temperature following removal from cold water. However, Golden & Hervey (1981) found no after-drop in the temperature of the central venous blood of pigs after cold water immersion. As an alter- native hypothesis they suggested that in the hypothermic individual, the loss of the hydrostatic assistance to circulation (up to 30 % of cardiac output) on removal from the water may lead to the collapse of arterial pressure and, as a consequence, cerebral ische- 248 mia, coronary insufficiency and myocardial hypoxia (Golden & Hervey, 1981). These effects can be counteracted by lifting victims from the water in a horizontal position. Thermal balance when exposed to cold water The above sections have emphasised the importance of maintaining body temperature in humans in order to avoid the lethal effects of cold. The time that elapses before critical lower body temperatures are reached is dependent on the rate of heat loss. This in turn is dependent on the insulation factor provided by clothing and body fat together with envi- ronmental factors such as wind, waves, water and air temperature. Heat exchange between the body and the cold water follows the normal laws of thermodynamics. To keep the body in heat balance, heat production and heat gain must equal heat loss, ac- cording to the equation M + R + C + K + S + E=0 (where M is metabolic heat produc- tion, R is radiation, C is heat loss by convection, K is heat loss by conduction, S is stored body heat and E is evaporative heat loss). Evaporative heat loss is of no significance in water, so total heat loss is determined by radiation, convection and conduction. Heat pro- duction can be increased in two ways; either involuntarily by shivering or voluntarily by muscular exercise. It is generally acknowledged that the rate of body cooling in cold wa- ter is increased by physical activity, which increases heat loss more than heat production (Keatinge, 1984). Physical activity increases blood flow in the extremities, reducing the internal insulation and thereby increasing convective heat loss (Hayward & Eckerson, 1984). However, leg exercise has been shown to be beneficial in maintaining heat balance during cold water immersion when subjects are wearing insulated survival suits (Reinertsen et al., 1993). Physical activity should therefore be considered as a means of improving endurance when protective equipment and procedures for survival at sea are being designed. Convective heat loss from the body rises during cold-water immersion, since water has a heat-removing capacity 20 times as high as that of air. Heat loss is dependent on the temperature difference between the skin and the environment, and this in turn is largely dependent on insulation. Circulatory adjustment by vasoconstriction increases the body’s insulation by shunting the blood from the peripheral vessels to the core of the body, keeping the central temperature high (Veicsteinas et al., 1982). Once vasoconstriction has set in, the ability to maintain body temperature is largely dependent on the thickness of the subcutaneous fat layer (Hayward & Keatinge, 1981). A thin person will lose heat to the surroundings faster than a fat person. A thin person will thus become hypothermic faster, while a fat person will be able to stabilise his body temperature better. Fat is also an advantage in cold-water immersion because it provides buoyancy. Another important determinant of heat loss is the ratio of surface area to mass. Therefore, as shown by Sloan & Keatinge (1973), children are at greater risk of hypothermia than adults because of their relatively high ratio of surface area to mass. Last but not at least, the impacts of en- vironmental factors are significant in determining heat loss from the body. It has been demonstrated that accidental immersion in rough seas is associated with significantly shorter survival times than have been estimated from calm-water studies (Steinman et al., 1987). This is of importance in the work of developing test methods and standards for protective clothing. 249 Development of protective equipment; the compromise between different user requirements The above discussion has demonstrated that the subject of survival at sea is truly complex. The basic principle is that the development and design of protective equipment should focus on all the potential hazards discussed in connection with cold-water immer- sion. To protect victims from the many different hazards, it is likely that more than one type of equipment will be required. To mention just a few: a survival suit should provide sufficient insulation to prevent the lethal effects of accidental hypothermia and attenuate the cold-shock response by reducing the rate of fall in skin temperature and prevent leakage. Gloves should protect the hands and ensure that the victim is able to swim and clamber on board a dinghy. Life jackets or other buoyancy aids should prevent deaths from initial respiratory responses by keeping the victim’s head above water. Emergency underwater breathing apparatus would be advantageous when escaping from a ditched helicopter, by preventing inhalation of water. However, these protection requirements often interfere with performance and comfort requirements, complicating the process of development of protective equipment. One example of this is the difficulty of reconciling good thermal protection in water with other functional requirements associated with flying. In the Royal Norwegian Air Force, wearing a survival suit is compulsory if the water temperature is below 10 °C. Aircrew personnel in Norway’s northernmost squadrons are required to use a survival suit all year round because they are operating in areas where the water temperature may be as low as 0 °C. These suits are designed to extend survival time in case of cold-water immersion, and are therefore made of a material that has low water permeability and pro- vides good thermal insulation. Paradoxically, this leads to a situation in which aircrew are subjected to considerable thermal stress during flight, with possible negative effects on cognitive abilities and mental performance. The survival suit must be compatible with the wearing of other aeronautical equipment and must not hinder flight control operations by restricting movement. If an accident does occur, the survival suit must have good mo- bility and low buoyancy, thereby facilitating escape from aircraft that may float upside down. In the water, dinghies will help keeping the victim afloat and offer some thermal protection. In the worst-case scenarios, victims may be unable to enter the dinghy and survival suits must keep them alive for up to 12 hours under conditions in which waves continuously wash over them. Finally, important requirements such as fire retarding properties, maintenance and cost must be taken into consideration. It is understandable that producing protective equipment that fulfils all these requirements will necessarily be a compromise. In view of the functional requirements in a working situation, the thermal protection offered by an acceptable suit in water will necessarily be limited. The testing programme should include an analysis of all critical phases in which the equipment might be used. First, procedures for testing personal protective equipment must include an analysis of its usability in the aircraft as well as of the thermal condition of the aircrew during flight. Secondly, the analysis of thermal protection in water must focus on proce- dures to maintain thermal balance under the environmental conditions against which it is intended to provide protection, and the duration required for such protection. A questionnaire of Sea-King personnel at five helicopter squadrons in Norway showed that pilots and crewmembers have quite different priorities to the survival suit. All per- sonnel categories put highest priority on safety for eventual emergency situations. Com- paring the other requirements for the working situation, thermal comfort was most im- portant for pilots, while flight engineers put a higher priority on keeping the suit clean. The rescue men were most concerned with maximum freedom of movement, while sys- 250 tem navigators put highest priority on thermal comfort of the feet. The differing priorities of pilots and crewmembers reflect their different working tasks. This emphasizes the im- portance of involving the end users in order to acquire a better understanding of the working environment and the problems experienced. The method of involving the end- users when developing protective clothing will also be applicable in other user groups that have quite different requirements. For example, thermal protective aids for ferry pas- sengers should focus on the importance of easy donning of the life west’s. Offshore oil workers are easy to localize and may not need protection for a very long time. Protective clothing for fish farmers should give a better protection for the sudden exposure to cold water, than long time immersion to cold water. In contrast, fishermen require a survival suit that will only be used in case of an emergency, and should protect against cold water immersion for several hours. In conclusion, the development and testing of protective equipment should focus on all hazardous responses resulting from cold water immersion and take into consideration the operational requirements of the user. References Bristow G (1984). Accidental hypothermia. Can anaesth Soc J. 31: 52-55. Ganong WF (1997). Review of medical physiology. Appleton & Lange. Stanford, Connecticut. 18 ed. Golden FStC & Hervey GR (1981). The after-drop and death after rescue from immersion in cold water. In: Hypothermia ashore and afloat, pp. 37-56. Ed. Adam. J.A. Aberdeen University press, Aberdeen. Golden, FStC & Hardcastle PT (1982). Swimming failure in cold water. Journal of Physiology (London), 330: 60-61. Hayward JS & Eckerson JD (1984). Physiological responses and survival time prediction for humans in ice-water. Aviation, space and evironmental medicine: 55. 206-212. Hayward JS, Hay C, Matthews BR, Overweel CH & Radford DD (1984). Temperature effects on the human dive response in relation to cold water near-drowning. J Appl Physiol, 56: 202-206. Hayward JS, Keatinge WR (1981). Roles of subcutaneous fat and thermoregulatory reflexes in determining ability to stabilize body temperature in water. J Forensic Sci. Jul: 26(3):459-61. Keatinge WR, Hayward MG (1981). Sudden Death in cold water and ventricular arrhythmia. Journal of forensic sciences. JFSCA. Vol. 26. No.3: 459-461. Keatinge WR (1984). Hypothermia at sea. Med Sci Law. 1984 Jul:24(3):160-2. Keatinge WR (1969). Survival in cold water : The physiology and treatment of immersion hypothermia and of drowning. Blackwell Scientific Publications. Oxford, Edinburgh. Lefevre J (1898). Analyse des Phènomènes Thermiques qui Prèparent, Accompagnent et Suivent la la Mort par Rèfrigèration. Arch. Physiol. Norm. et Pathol. 10(5):685-697. Lexow K (1989). Aksidentell hypotermi. Tidskr Nor Lægefor. 30. 109: 3105-3107. Mekjavic IB, La Prairie A, Burke W, & Lindborg B (1987). Respiratory drive during cold water immersion. Respiration physiology, 70: 121-130. Molnar GW (1946). Survival of hypothermia by men immersed in the ocean. JAMA, 131, 1046-1050. Popovic V, Popovic C (1974). Hypothermia in biology and medicine. New York: Grune and Stratton. Reinertsen RE, Volla TT, Sandsund M, Eid T, Bakkevig MK (1993). Comparison of thermal responses between rest and exercise during cold water immersion. In: Life in the cold III (ed. C.Carey). Westview press. New York. USA. Royal society for the prevention of accidents (1988). Drownings in the U.K. ROSPA, Birningham. Sloan REG & Keatinge WR (1973). Cooling rates of young people swimming in cold water. J.Appl. Physiol. 35: 371-375). 251 Steinman AM, Hayward JS, Nemiroff MJ, Kubilis PS (1987). Immersion hypothermia: comparative protection of antiexposure garments in calm versus rough seas. Aviat Space Environ Med. Jun: 58(6): 550-8. Tanaka M (1991). Accidental hypothermia and death from cold in urban areas. Int J Biometeorol. Mar:34(4):242-6. Tikuisis P (1997). Prediction of survival time at sea based on observed body cooling rates. Aviation, Space and Environmental Medicine, Vol. 68. No. 5. Tipton MJ & Vincent MJ (1989). Protection provided against the initial responses to cold water immersion by partial coverage wet suit. Aviation, Space and Environmental Medicine, 60: 769-773. Tipton MJ (1989). The initial responses to cold-water immersion in man. Clin Sci: 77: 581-588. Vander AJ, Sherman JH, Luciano DS (1994). Human physiology, the mechanisms of body function. McGraw-Hill. USA. Veicsteinas A, G Ferretti, DW Rennie (1982). Supeficial shell insulation in resting and exercising men in cold water. J Appl. Physiol. Jun: 52(6):1557-64. Withers PC (1992). Comparative Animal Physiology. Saunders College Publishing. USA Åstrand PO & Rodahl K (1986). Textbook of work physiology. 3 ed. McGraw-Hill, Inc. USA. 252 Current and future standards of survival suits and diving suits Arvid Påsche SINTEF Unimed, Trondheim, Norway International standards for survival suits was first included in the International Maritime Organizations Resolution A.689(17) on Life Saving Appliances, which covered two types of survival suits: insulated suits and uninsulated suits, both of a dry suit type. The insulated suit are commonly referred to as a 6-hour suit, a name which has arrived from the thermal testing of suits requiring a test of 6 hours in cold water (<2 °C). The uninsulated suit is accordingly referred to as a 1-hour suit, as this type of suit is thermally tested for 1 hour. Recently the IMO Regulations has been expanded, and now include a third suit, an anti-exposure suit, which can be considered as a work suit with specific integrated sur- vival functions. With regards to thermal properties the requirements for the anti-exposure suit are identical to those for the IMO-uninsulated suits. The necessity of such life saving appliances onboard commercial sea going vessels have been demonstrated in numerous sea accidents, in which such suits have saved a large number of human lives, in particular in the fisheries. However, recent accidents with passenger transporting ships have given clear indica- tions that the number of survival suits onboard should be sufficient, also for the passen- gers, not only for the crew as the situation is today. This is particularly important for sea activities in northern areas and in cold water, where the standard IMO-lifejacket is insuf- ficient to provide any thermal protection of significance. As a consequence of this realisation the Norwegian Maritime Directorate are proposing a standard for a passenger survival suit which would fulfil the requirement for an im- proved thermal protective aid (TPA) which is a standard life saving appliances in liferafts and lifeboats. The Norwegian Maritime Directorate is proposing this standard to be implemented in the IMO Regulations for life saving appliances. This would be an addi- tion to the thermal protective lifejacket which was presented at the Subcommittee on Ship Design and Equipment in 1999. The Norwegian Maritime Directorate has further suggested to differenciate the re- quirement for personal life saving appliances based upon the water temperatures where the ship or vessel is active. The proposal is as follows: In the 1990’s the initiative was taken to prepare an European standard for immersion suits. During the work this has developed to become a joint EN and ISO standard, which covers two types of immersion suits: Constant wear suits, and Abandonment suits The requirements including safety for these two suits are presently covered in two documents: prEN ISO 15027-1 (Constant wear suits), and prEN ISO 15027-2 (Abandonment suits) 253 The test requirements and test methods for these suits are presented in a third document: prEN ISO 15027-3 These three documents are currently submitted to the formal vote among the CEN member countries. The constant wear suit presented by prEN ISO 15027-1 also cover the helicopter transit suit. However, in 1999 the Civil Aviation Authorities in the countries with offshore activi- ties in the North Sea (UK, Netherlands, Denmark and Norway) started the preparation of a joint standard for helicopter transit suits. The prEN ISO 15027-1 represent the basis for this new standard, as it outlines the minimum requirements for such suits. The Working Group for the Civil Aviation Authorities have prepared a draft standard for “Helicopter Crew and Passenger Immersion Suits”. This Joint Technical Standard Order gives the requirement which immersion suits for use on helicopters operating to or from helidecks located in a hostile sea area (as defined in JAR-OPS 3.480) must meet. These immersion suits must be combined with a JTSO lifejacket. The Working Group for the Civil Aviation Authorities are also preparing a separate standard for an “Integrated Immersion Suit for Helicopter Crew and Passengers”. In this suit the lifejacket functions are integrated as a permanent part of the immersion suit. Much of the standardisation work for immersion suits/survival suits have taken place during the last 10-20 years. It can therefor be foreseen that these standards will be ad- justed and modified with regards to requirements and test methods as experience are being gained for the standards in their present form. One area has been of particular concern, the evaluation of the thermal properties of the suit. The majority of such evaluation are presently being performed as manned tests using human subjects. To reveal any problems with the thermal properties of the suits the tests have to include exposures which could be considered extreme. Several countries have expressed concern and reluctance for performing such tests on ethical grounds. Proposed changes to the suit standards have been prepared, recommending such evaluation being performed using thermal manikins. Unfortunately a convincing relation between the re- sults obtained on manikins and human subjects for the same suits has not been estab- lished. Thermal manikins for wet environments are presently available only in a very low number. The replacement of human thermal testing with manikin testing has consequently been delayed. Even though the testing performed as defined in the standards may be considered extreme, experience from real accidents at sea indicate that these test conditions are far from being as demanding as the conditions in many real accidents. The thermal evaluation is a good example on this, as sea accidents in cold environ- ments have demonstrated that the cooling of the victims’ bodies have been three times as fast as measured in the approval tests. The main reason for this difference is due to the fact that the approval test does not ask for low air temperatures, high wind speeds, waves and water spray, which are normally occurring in rough water, and which are affecting the body cooling of the victims. It is therefor strongly warned against trying to extrapo- late the laboratory test results to predict survival times in real sea accidents. Considering the extreme conditions against which the diving suits shall protect the wearer. Such protective suits have received surprisingly little attention. Notified bodies have, however, been faced with the problem of lacking common standards for the 254 evaluation and certification of the diving suits. Comparing the tests performed and evaluation criteria used for certifying the suits by different notified bodies reveal signifi- cant differences. Finnish Institute of Occupational Health (FIOH) took the initiative to propose a working group for developing a common European standard for diving suits. This group was established as WG12 in TC162, and started its work 2 years ago. The working group is preparing 4 standards for 4 different groups of diving suits: I Wet suits II Dry suits III Actively heated or cooled suits IV One atmosphere suits The working group has recently presented the first working documents for these diving suits standards. The final standards is expected to be presented for formal voting within a two year period. 255 Heat preservation behavior of diving suit Zhongxuan Luo , Edward Newton, Yi Li, Xiaonan Luo Institute of Textiles and Clothing, The Hong Kong Polytechnic University, Hong Hum, Kowloon, Hong Kong Introduction Heat preservation is one of the important functional attributes to design diving suit for the survival of divers. Many researchers investigated the mechanism of heat and moisture transfer in porous media such as textiles (Ogniewicz & Tien, 1981; Motakef & El-Masri, 1986; Henry, 1939; Luo et al., in press). The aim of this paper is to simulate the dynamic distribution of heat and water condensate in a diving suit consisting of two layers of fabrics with different porosity in diving environment. In simulating the dynamic heat and moisture transfer in a diving suit, we focus on comparing numerical results computed by using fabrics with different porosity due to the time limit for completing the paper for this conference. Model and methods Governing Equations The mechanism and mathematical model to heat and moisture transfer in a porous fabric can be described by the following equations (Luo et al., in press): ° ° ° ° ° ¯ ° ° ° ° ° ® ­ w w * w w  w w  w w * w w  w w  w w w w w w 2 2)(),())(1()( ),( ),( ),(),())(1()),((),( 4 4 x a CxaDtxx t a C x txTF x F txTF x F txtxx x F x F x T txk xt T txC L L R R LR v W H HH EVE EVE OH where  is the absorption constant,  is the Stefen Boltzmann constant, and ),( tx" is the (de)sorption rate, condensation rate or freezing rate by fibers in the case of gas, liquid or solid, respectively. Details of all nomenclatures are given in (Luo et al., in press). When there is no condensation on the surface of a fiber in a fabric, ),( tx" is in the form of t txC tx f   " ),(),( to describe the sorption/desorption of moisture by the fibers, which obeys Fick’s law of diffusion. The water content of the fiber is determined by /),(),( txCtxW fc  . When the relative humidity exceeds 100 % at a position and/or water content in a fiber reaches the saturation, condensation occurs. In the condensation region, the liquid water and water vapor reach thermodynamic equilibrium. The vapor Dept. of Appl. Math., Dalian University of Technology, CHINA. The work is completed during the period in which author worked as a Research Fellow in HKPOLYU. 256 concentration is saturated and solely determined by temperature, i.e., )),((** txTCC aa  . Under such circumstances, the water condensation rate ),( tx" can be determined by the mass balance equation, i.e. ),),(),(()(1 )(),(),( * 2 *2 t txC x txCD x x x txC tx aaa f w w  w w  w w * WH H where ).,(/10)),((5.216),( 6* txTtxTVaptxCa uu With ),( tx" , the water content can be calculated by  * t c dttxtxW 0 .),( 1),( U When regainx x txW fiber water c   ))(1()(),(    , water sorption reaches its maximum at the point x under the assumption of that water is immobile in the fabric. Boundary conditions To simulate heat transfer and water distribution in the diving suit in dividing environment, we assume that heat flux between diver’s skin and the suit decrease exponentially with time, i.e., ° ° ¯ ° ° ® ­   1 1 0 |)(),( |)(),( R TT dx xdT txk ce dx xdT txk bb Lx pt x and ° ° ¯ ° ° ® ­  w w  w w 1 1 0 00 0 || || w CC x C D w CC x CD Lxaab Lx a a abxa x a a H H , where C and p are an initial heat flow and attenuation coefficient of the heat flow respectively. Results and conclusion We assume that a diving suit consist of two layers - cotton and wool fabrics, which is coated at the inner and outer surfaces. The dynamic heat and moisture transfer processes in the diving suit is simulated for the situation when a diver wears a diving suit and moves suddenly from an initial warm condition to a cold environment of deep sea. The initial external condition is assumed to be 20 °C and 80 % RH, and the cold condition is assumed to be 0 °C and -20 °C, 100 % RH, respectively. The inner surface of the diving suit is assumed to be close to the human skin and the microclimate next to skin is assumed to be initially at 33 °C and 96 % RH. In our numerical experiments, we will take large parameter 1w (resistance to water vapor transfer) for the outer surface of the diving suit according to the design of the suit. In computation, the values of the material parameters are listed as follows: 3/1310 mKg U (density of wool), 3/1550 mKg ( density of cotton); meRf 503.1  (mean radius of wool), me 566.0  (mean radius of cotton); ° ¯ ° ® ­    regionfreezinginkgKJ regionwetinKGKJ regiondryinKgKJe cW w /)3332260( /2260 /0.25225.1602 72.11 O ° ¯ ° ® ­    regionfreezinginkgKJ regionwetinKGKJ regiondryinKgKJe cW c /)3332260( /2260 /0.25229.1030 39.22 O KmKJTWC cv 3/221.4.46613.373  (Volumetric heat capacity of wool fabric); KmKJWWC ccv 3/)1/()41841663(  (Volumetric heat capacity of cotton fabric). The following parameters are also used in the computations: 257 2.1 W (Effective tortuosity of the wool and cotton fabric) mkwK a /025.0 (Thermal conductivity of still air) smeD a /5.2 25 (Diffusion coefficient of water vapor in the air) mL 04.0 , 9.021 ee , ms /400 . The thermal conductivity in our computation is calculated as the same as in (Luo et al., in press). The resistance to heat transfer of the inner ( 0R ) and outer ( 1R ) fabric was taken to be WKm /003.0 2 and WKm /00001.0 2 , and the water vapor resistance of the inner ( 0w ) outer ( 1w ) fabric was taken to be ms /10 4 and ms /10 4 . The attenuation coefficient p and constant c are taken the values of 0.5 and 0.05 in our computation, respectively. Figure 1 and Figure 2 show that the distributions of temperature and water condensate accumulation with time in the diving suit with cotton fabric of 60 % porosity and wool fabric of 80 % porosity. The numerical results imply that the temperature in the suit decrease rapidly at the outer surface to 0 C, followed gradually by the inner layers of the suit. Different levels of moisture condensations are predicted in the two layers of fabrics of the suit during the first 5 hours of simulation. The distribution of water vapor concentration in the suit, however, shows a pattern of rapid decrease across the thickness of the suit (Figure 3). Figure 4 compares temperature distributions after 5 hour in the sea among four different combinations of porosity in the cotton and wool fabrics: 1. %5.92%,9.91 wc HH ; 2. %80%,60 wc HH ; 3. %60%,80 wc HH ; 4. %90%,60 wc HH . It can be seen from Figure 4 that the suits consisting of higher density in the inner cotton fabric and lower density in the outer wool fabric has better heat preservation behavior than those with lower density in the inner cotton fabric and higher density in the outer wool. Figure 1. Figure 2. 0 5 10 15 20 25 30 0.0 0.2 0.4 0.6 0.8 1.0 1 2 3 4 Figure 4.Figure 3. oC 258 Figure 5. To investigate and simulate the heat and moisture transfer in a diving suit under a very cold deep-sea environment, the outer boundary condition is assumed as -20 °C and the attenuation coefficient 6 5 p . Rapid temperature drop is predicted across the thickness of the suit as shown in Figure 5, indicating that the survival time of the diver is much shorter in such an environment than in a warmer environment. Acknowledgement We would like to thank Mr. Burley Wang, for his useful help in the completion of the paper. We would like to acknowledge The Hong Kong Polytechnic University for the funding of this research. References Ogniewicz Y & Tien CL (1981) Analysis of condensation in porous insulation. J. Heat Mass Transfer, 24, pp 421-429. Motakef S & El-Masri MA (1986) Simultaneous heat and mass transfer with phase change in a porous slab, J. Heat Mass Transfer, 29(10), pp 1503-1512. Henry PSH (1939) Diffusion in absorbing media, Proc. Roy. Soc., 171(A), pp 215-241. Luo ZX, Fan J & Li Y (in press) Heat and Moisture Transfer with Sorption and Condensation in Porous Clothing Assemblies and Numerical Simulation. Int. J. Heat Mass Transfer. 259 The effect of the distribution of insulation in immersion suits on thermal responses Randi Eidsmo Reinertsen Division of Health and Work Physiology, SINTEF Unimed, N-7465 Trondheim, Norway Introduction The IMO’s thermal requirements for survival suits are linked to core temperature changes in test subjects. However, although the core temperature is relatively protected, cooling the peripheral regions of the body may cause discomfort and pain. Furthermore, severe cooling of the hands and feet affects physical performance and impairs the ability to move (Enander, 1987). In an emergency situation this may be lethal. When a subject is protected with an insulated survival suit, only some parts of the body are immersed. The anterior part of the body with most of the head, hands and toe area is exposed to air, while the posterior part is exposed to water. Heat is lost to the environ- ment mainly by conduction and convection. The rate of heat loss depends on the heat conducting properties of air and water together with the temperature gradient between the body in its survival suit and the surrounding air or water. Because heat conduction in water is about 25 times as great as in air, the potential for heat loss is greatest for the re- gions of the body exposed to water, i.e. the posterior part (Smith & Hames, 1962). Furthermore, water pressure compresses the air between clothing layers inside the suit, resulting in poorer clothing insulation. Heat loss also depends on the surface to volume ratio, which means that the feet, hands and legs will cool faster than the torso. In con- junction with a developing negative body-heat balance this produces vasoconstriction and decreased blood flow to the hands, legs and feet. The thermoregulatory responses to temperature changes differ between different body regions. This presentation summarises some of our results that demonstrates the effects of increased regional insulation during cold immersion on the following parameters: local skin temperature, mean body temperature, core temperature, metabolic response to heat loss and subjective evaluation of thermal sensation and thermal comfort. Results and discussion Insulation of the hands Six subjects dressed in neoprene survival suits underwent two hours of immersion in cold water (water temperature slightly above 0 C, air temperature -5 C and wind speed 5 m·s 1). The core temperature fell by approximately 0.3 C per hour, while the skin tem- peratures of the neck, abdomen and lower back also decreased slowly from between 32 and 36 C to between 28 and 34 C. All these temperatures were sensed as thermally comfortable. Skin temperatures of the posterior thigh and leg, hand, heel and toe fell much more rapidly, reaching temperatures between 10 and 20 C. The thermoregulatory responses we can deduce through these changes in body tem- perature involve vasoconstriction that reduces the temperature gradient between the body and its surroundings. The blood flow to muscles and subcutaneous fat is also reduced and this contributes to the protection of the body core temperature. Metabolic heat production increased three times the resting metabolic rate. 260 At the end of such a two-hour immersion period, the body core shrinks in terms of the proportion of tissue which is able to maintain a 37 C blood supply. The requirements of the IMO regulations regarding core temperature are still met, but the thermal status of the body is not satisfactory. Leg and toe temperatures below 10C are painful. Hand and heel temperatures are falling rapidly towards the same low values. In order to delay the cool- ing of the hands, the insulation of this body region was improved by a knitted woollen glove. At the end of the two-hour immersion the hand skin temperature had stabilised at 18 C, resulting in a significant positive effect on local thermal sensation in the hand re- gion. Furthermore, the ability to perform manual tasks was maintained. Insulation of the posterior region of the body In a further attempt to avoid rapid cooling of the peripheral body regions, an extra layer of neoprene was applied to the posterior of the suit, covering the area between the lower scapula and the knees. Figure 1 shows the development of posterior thigh temperatures in nine test subjects during six-hour of immersion in 2 C water. Air temperature was 10 C. Thigh skin temperature of subjects wearing the standard suit decreased 12 C during the six-hour immersion period. The extra layer of insulation produced a 6.5 C reduction in the drop in temperature during the same immersion period (Table 1). There were no sig- nificant differences in development of rectal temperatures between the two conditions. 20 25 30 35 0 60 120 180 240 300 360 Time (minutes) Te m pe ra tu re (° C) Figure 1. Time course of thigh temperature during six-hour immersion for the two conditions, standard suit (lower) and suit with extra insulation (upper). Values are means of nine subjects with standard deviation indicated. Table 1. Body temperatures at the start and end of six-hour immersion. Thigh = posterior thigh. Values are means r SD of nine subjects. Condition Measuring site Temperature (qC) start Temperature (qC) end Thigh 34.1 r 0.4 22.3 r 0.7 Standard suit Mean skin 32.8 r 0.6 27.0 r 0.9 Rectal 36.8 r 0.1 36.2 r 0.4 Thigh 33.4 r 0.5 26.9 r 0.8 Extra insulation Mean skin 33.2 r 0.3 28.2 r 0.7 Rectal 36.7 r 0.2 36.2 r 0.1 261 Thirteen skin temperatures were measured with thermistors positioned at the forehead, chest, abdomen, upper arm, forearm, dorsal hand, anterior and posterior thigh, shin, big toe, heel, upper back and neck. Mean skin temperature was calculated from the un- weighted average of these sites. There was a significant difference between the drop in mean skin temperature for the standard suit and the suit with the extra insulation. The improveded insulation also reduced total heat loss and thus the need to increase metabolic heat production during the immersion (Table 2) (Kuhnen & Jessen, 1988). Similar rectal temperatures and higher skin temperatures in the suit with extra insulation, would produce more afferent input from the thermosensitive reseptors in the standard suit condition (Jessen, 1990; Mittelmann & Mekjavic, 1991). Total, peak and increased heat production from resting value to peak value were all significantly different between the two conditions. This effect of the extra layer of insulation on metabolic heat production may enhance the potential of sustained heat production, and thus increase chances of sur- vival in an emergency situation (Hesselberg et al., 1995). The condition with the extra layer of insulation also resulted in significant differences between subjective ratings of thermal sensation and thermal comfort. All nine subjects wearing extra insulation reported a thermal state closer to "neutral" and "comfortable" than under the standard suit condition (Gagge et al., 1967). Table 2 Heat production (HP) during a resting period before immersion, peak heat production, maximum increase of heat production and total heat production during the six-hour immersion period. Values are means r SD for nine test subjects. Condition Start HP (W˜m-2) Peak HP (W˜m-2) Increased HP (W˜m-2) Total HP (W˜m-2) Standard suit 46.0 r 3.7 81.4 r 12.5 35.33 r 13.7 23.781 r 2.101 Extra insulation 49.3 r 4.5 69.4 r 10.1 20.2 r 9.9 20.930 r 1.954 Conclusions The effects of improving insulation in selected regions of the body, may well include in- creased chances of survival and an improved ability to perform essential manual tasks. If the requirements of international standards focused on the total thermal condition of subjects, including comfort, and the skin temperatures of hands and feet as well as meta- bolic thermal responses, manufacturers could easily produce survival equipment that could satisfy such requirements and thus provide better protection in emergency situations. References Enander A (1987) Effects of moderate cold on performance of psychomotor and cognitive tasks. Ergonomics, 30:1431-45. Gagge AP, Stolwijk JAJ & Hardy JD (1967) Comfort and thermal sensation and associated physiological responses at various ambient temperatures. Environ. Res., 1:1-20. Hesselberg Ø, Waag T & Reinertsen RE (1995) Metabolic changes during cold water immersion. Arct. Med. Res., 54:65-69. Jessen C (1990) Thermal afferents in the control of body temperature. In: Thermoregulation, Physiology and Biochemistry (eds. E Schönbaum & P Lomax), Pergamon Press, N.Y. Kuhnen G & Jessen C (1988) The metabolic response to skin temperature. Pflügers Arch., 412:402-408. Mittelmann KD & Mekjavic IB (1991) Contribution of core cooling rate to shivering thermogenesis during cold water immersion. Aviat. Space. Environ Med, 62:842-848. Smith GB & Hames EF (1962) Estimation of tolerance time for cold water immersion. Aerosp. Med., 33:834-840. 262 Lifevests - what is the value of the certification? Arvid Påsche Sintef Unimed, Trondheim, Norway Abstract An accident on the coast of Norway last November, with the fast speed ferry SLEIPNER, in which 16 people died has resulted in a number of questions regarding the certification of life jackets. The life jackets onboard this ferry was manufactured only a year prior to the accident and were marked to be approved according to SOLAS 74/83 Res. IMO A689(17). However, several of the survivors from the SLEIPNER reported problems with the life jackets. When retested on the request of the Norwegian Maritime Directorate it was found that the life jackets did not fulfil the IMO requirements for strength, for donning and for the water performance. Investigation connected to life jackets, and particular the life jackets onboard SLEINER, has revealed differences between Notified bodies with regards to performing, witnessing or accepting tests performed by other institutions as a basis for issuing certifi- cates for such survival equipment. The discussion around the certification of life jackets has also resulted in intensified discussions about some of the acceptance tests in the IMO Regulations. In particular the tests for righting properties and free board have been focused upon. The tests are per- formed using human test results and with considerable differences in test results from one test subject to another for the same life jacket. A test less dependent of the test person and with more reproducible results is clearly preferable. Improvements both with regards to test methods and practice for certification are needed. 263 Pass/fail criteria to evaluate the strength of buoyancy aids (50 N) and lifejackets (100 N) in accordance to EN 393:1993, EN 395:1993 and the A1:1998 Hanna Koskinen, Raija Ilmarinen Finnish Institute of Occupational Health, Dept. of Physics, Vantaa, Finland Introduction Personal flotation devices (PFD)  lifejackets and buoyancy aids  must be sufficiently comfortable to wear and attractive to use, to ensure that they are not just purchased but also actually used. However, in accidental immersions, and especially during rescue operations in rough waters, not only good in-water performance but also sufficient strength of PFD is a vital requirement for survival. The European standards EN 395:1993 for Lifejackets 100 N (for use in relatively sheltered waters) and EN 393:1993 for Buoyancy Aids 50 N (for use in sheltered waters when help is at hand and the user is a swimmer) require a horizontal load test to measure the strength of an assembly. This test is mainly for the body strength of PFD. The amendments EN 393/A1:1998 and EN 395/A1:1998 introduced also a vertical load test as a normative test. This is used mainly for testing the strength of the shoulder seams. However, the requirements are written in very general terms and thus clear pass/fail crite- ria are not available to a technical appraiser of notified bodies to assess the conformity of the tested products. In EN 393 and EN 395, clause 4.16.6 only states "Strength of as- sembly shall be tested according to Annex A for (5.0 0.1) min in both wet and dry con- ditions. No damage shall result which would result in the device failing to function in accordance with this standard." Annex A in standards EN 393 and EN 395 instruct the appraiser to "examine the buoyancy aid or lifejacket for any resultant defects" and to "measure any adjustment device slippage". Pass/fail criteria have also been discussed at the test house experience exchange meeting (Vertical Group 8, Lifejackets, personal buoyancy aids, immersion suits) held in Cologne in April 1999. The meeting came to the conclusion that some damage could be accepted, and that if the jacket functioned after the test, then it would pass. This could be verified by additional subject tests. What does this mean? Should in-water performance tests be carried out on all test items that have been torn in the load tests? This would bring extra expenses for the manufacturers. The meeting also concluded that the damage was impossible to define in terms of length of the torn seam, and the decision was left up to the test houses. However, the vague pass/fail criteria have caused difficulties because the notified bodies in different countries have interpreted the given test results differently. This handicap has become apparent when the marketing controls on PFDs have been car- ried out. Many CE-marked PFDs have failed in the inspection tests. This study summarizes the results of strength tests for buoyancy aids 50 N and life- jackets 100 N carried out during EC type-examinations conducted by notified body 0403, and during marketing control projects in Finland. The results are meant as a basis for further discussion in the EN/ISO standardization group for revision of the PFD standards (prEN ISO 12402 - Part 1-9, requirements and test methods) and in the test house meetings to exchange experiences. 264 Material and methods The data have been collected from the PFD mechanical approval test reports of the accredited Finnish test house T013. The samples have been tested in 1998-2000, ac- cording to EN 393/A1:1998 and EN 395/A1:1998. A vertical load test is always per- formed after a horizontal test, and correspondingly, a wet test after a dry test on the same PFD sample. This study includes 31 buoyancy aids and 33 lifejackets of 23 different designs and in several sizes from numerous manufacturers. The numbers of tested PFDs types (overall, jacket, and vest) are given in Table 1. Table 1. Types of tested PFDs Sample type Overall Jacket Vest Total Buoyancy aid 6 9 16 31 Lifejacket - - 33 33 Total 6 9 49 64 All tested PFDs contained inherently buoyant material, but they were of very different designs, materials and seam types: i.e. from very crude and simple vests to very bizarre and purpose-built overalls. The most used fabrics were 100% Nylon and 100% Polyester. A zipper with /without a placket was the primary closure system in 59 samples, and an adjustable belt with a quick-release buckle of the waist in five (all buoyancy aids) samples. The most common secondary closure systems were fastening cords at the neck and hem. Join stitching without a topstitch was the most commonly used seam type. In some of the samples the shoulder or side seams were sewed with a single or double topstitch. In a few samples there was also an overlock stitch. Some samples were constructed with side- pieces instead of a side seam. Results 33 samples (52 %) of the tested PFDs were damaged in some way in the load tests, most often during vertical loading (Figure 1). Vest type PFDs were the worst in the load test. 0 10 20 30 40 D ry W et D ry W et % o f t es te d sa m pl es V est Jacket Overall Horizontal load te st Ve rtical load test Figure 1. The number of damaged samples (%) in the load tests. In the horizontal load test only zippers or belt buckles broke, the fabric and seams re- mained intact. In vertical load test, on the contrary, also the fabric tore in five samples, in three of which (2 / dry and 1 / wet) the side ripped open to the tightening cord (Figure 2a). 265 The fabric tore also down from the neck opening (Figure 2b, shown with Î ). The vertical load test were carried out for 51 samples, which passed the horizontal load test. The side or armhole seams tore in eight PFDs during the vertical dry load tests. Three of the tears were so severe that the samples were not tested in wet conditions. The shoul- der seam torn completely in one vest type buoyancy aid (Figure 2b). The tears in the side seams between the side and back or side and front ranged from 5 to 12 cm, and cor- respondingly, in the armhole seams from 1.5 to 12 cm (Figure 2c). In some cases the side (seam or fabric) was damaged so badly that only the belt kept the sample from falling off the test cylinder. a. Pass/fail? b. Fail c. Pass Figure 2a, b and c. Some examples of damaged PFDs after the vertical load test. Also the components turned out to be fairly weak. Some belts slid through the buckles as much as 19 cm in the horizontal load test, and 8 cm in the vertical load test. Also a few belt buckles broke totally, and some belt loops were torn off. The slide fastener opened or broke down totally in both the horizontal and vertical load tests even in some vest type PFDs in which the zipper was the primary closure. The final condition of the tested PFDs after the load tests is given in Table 2. Table 2. The number of the tested PFDs classified according to thin final condition after the load tests. Sample type No damage Material, shoulder or side seam tore 1-5 cm Material, shoulder or side seam tore completely or > 5.1 cm Other type of damage occurred Overall 4 - 1 1 Jacket 7 1 1 - Vest 20 - 12 4 Total 31 1 14 5 Discussion The results of the load tests carried out during type-testing clearly demonstrated that the strength of the fairly cheap and simple design Lifejackets 100 N and Buoyancy aids 50 N, which are very popular in Scandinavian wear conditions in lake areas, is poor. These PFDs do not guarantee survival. Their construction is not strong enough to allow the use of the PFD for lifting a victim out of water during a rescue operations. Surprisingly, also many CE-marked products that have been tested and approved by different test houses and notified bodies failed in the marketing control inspections. This means that despite the efforts of notified bodies to pursue identical interpretation of the requirements given in the standards, the technical appraisers come to different decisions on passing or failing. It is understandable that the diversity of the damages makes the de- 266 cisions difficult when unique criteria are lacking. This allows poorly constructed products to be put on the market. We recommend, on the basis of our results, that, as long as there are no harmonized pass/fail criteria, at least a small-scale (with a limited number of test subjects) in-water performance test must be conducted on the PFDs damaged in the strength tests. If the de- vice passes this test, it functions and conforms with the standard. All the conducted tests also demonstrated the need to improve the test apparatus for the load tests, so as to be more realistic: i.e. the test cylinder defined in the EN standards for PFD ought to be replaced by a standardized human-like torso. Conclusion The pass/fail criteria of the PFD standards should be revised so that ambiguities are re- moved from the text. Meanwhile, active discussion is encouraged between the test houses and notified bodies in comparing the test results and their interpretations. They should be encouraged to give a statement if application of the standard in its present form leads to confusion at a homogenous performance level. Furthermore, we recommend strict market control of PFDs to ensure the safety of the end users. References EN 393. (1993). Lifejackets and personal buoyancy aids - Buoyancy aids - 50 N [European Standard]. Brussells: Comité Européen de Normalisation. EN 393/A1. (1993/1998). Lifejackets and personal buoyancy aids - Buoyancy aids - 50 N [European Standard]. Brussells: Comité Européen de Normalisation. EN 395. (1993). Lifejackets and personal buoyancy aids - Lifejackets -100 N [European Standard]. Brussells: Comité Européen de Normalisation. EN 395/A1. (1993/1998). Lifejackets and personal buoyancy aids - Lifejackets -100 N [European Standard]. Brussells: Comité Européen de Normalisation. prEN ISO 12402 - Part 1-9 (1998). Personal flotation devices: Safety requirements and test methods. N217. Brussells: Comité Européen de Normalisation. 267 The effect of protective clothing on thermoneutral zone (TNZ) in man Drude Markussen1, Gro Ellen Øglænd1, Hilde Færevik2, Randi E. Reinertsen2 1 Norwegian University of Science and Technology, NO-7491 Trondheim, Norway 2 Division of Health & Work Physiology, Sintef Unimed, NO-7465 Trondheim, Norway Abstract Personnel operating the Sea king helicopters are required to wear survival suits (clo-value = 2.20 C m2/W) during flights. This may lead to pilots and crew suffering from heat stress during flight and cold stress when immersed in cold water in case of an accident. Gradual accumulation of heat in the body caused by personal protective equipment (PPE) during long exposure, gradually decreases cognitive performance (Enander, 1989). TNZ is a range of T a at which metabolic rate is minimal and constant (Withers, 1992). The aim of this study was to define TNZ in the cockpit for these pilots. We hypothesised that the ambient temperature in the cockpit today is above the TNZ for pilots wearing a survival suit and they may easily be exposed to heat stress. Eight volunteer subjects participated in the study; their heart rate, rectal (T re ) and 13 skin temperatures (T sk), metabolic heat production (VO2), humidity and subjective evaluation of thermal sensation and thermal comfort were measured during one hour for five different ambient temperatures, 0, 10, 14, 18 and 25 C respectively. The results show that 10 and 14 C (T a ) fulfil the criteria of thermal neutrality, where mean skin tem- perature (MST) is 33,6-34,1 C, VO2 at its lowest (0,331  0,05 sV`O2 (l/min)) and the subjects were comfortable. The conclusion of the study is that the ambient temperature in the cockpit is above the TNZ for pilots wearing the survival suit. 268 Passenger survival suits - a new emergency equipment Arvid Påsche Sintef Unimed, Trondheim, Norway Abstract Considerable effort has been put into the work of improving the safety for people en- gaged in activities at sea. For passenger transporting ships the recent strategy has been that dry evacuation should be possible. However, several accidents, also recent accidents have demonstrated, that people evacuating such ships very often are immersed in water before being rescued. Immersed in cold water protected only by personal clothing and a life jacket, people would by at a high risk for experiencing cold shock and hypothermia with fatal consequences. The Norwegian Maritime Directorate has taken the initiative to look into the possi- bility of developing a simple passenger survival suit with low production cost, shall packing volume, easy donning, preventing unacceptable water ingress, and providing thermal protection against the cold water. The manufacturer Helly Hansen Spesialprodukter has responded to this invitation, and has manufactured a passenger survival suit. This suit has a packing volume less than 2.0 dm3, and was able to meet the recommended acceptance levels for water ingress. When tested by six test subjects in water temperature below 5 °C for two hours, none of the six test subjects had rectal temperatures exceeding 2 °C. When dressed by test subjects with no earlier knowledge about the suit it was demon- strated that the donning time was less than 2 minutes, including the time to put on and secure the life jacket. 269 Aspects of firefighter protective clothing selection Mandy Stirling Leicestershire Fire & Rescue Service and Loughborough University, UK Introduction This paper discusses aspects for consideration by fire brigades when selecting new pro- tective clothing (specifically tunic and trousers) for firefighters. The major difficulty is the range of situations and environments in which firefighters are expected to perform in one set of protective garments. These include: sub-zero conditions at night-time; high levels of humidity, ambient air and radiant temperatures during operational and training heat exposures; confined space work at road traffic accidents (RTA’s) and in collapsed structures; heath and scrub fires; and flooding and rescue. Due to cost implications, stowage on fire appliances, and time limits within which a fire appliance must attend an incident, one set of firekit must be suitable for use in all these scenarios. The procurement procedure within most fire brigades in the UK at present involves purchasing ‘off-the- peg’ fire-kit, the choice of which is usually based upon subjective views of past per- formance, or perhaps the advice of another brigade. However, since the costs involved in procurement are so high, both in terms of PPE (personal protective equipment) expense and the preservation of firefighters’ lives, this approach is no longer acceptable. Therefore, this paper attempts to bring together a number of the many issues faced by fire brigades during the PPE procurement process: garment design; garment testing; the procurement process; compatibility; garment care; and finally, litigation issues. Garment design Material combination Firefighting tunics and trousers are constructed around a layered approach: the outer ma- terial layer, a vapour permeable membrane, a thermal barrier and a liner. There are many available combinations of these layers, each of which can be provided in various weights and weaves. Typical ambient temperatures during firefighting range from 38oC to 66oC according to Abeles (1973), but other studies report far more extreme temperatures (Foster & Roberts, 1994; Stirling & Parsons, 1999). Therefore, high insulative properties are traditionally provided in firekit. However, maximal physical work performance has been reported to be impaired during long-term heat exposure by thick and heavy clothing materials with high insulating properties that have a vapour barrier, which limits body cooling through evaporation (Ilmarinen et al, 1994; Mäkinen et al, 1995; White & Hodus, 1987). Therefore, although such garments are designed by manufacturers to pro- vide effective insulation from the external environment, no consideration is given to the dissipation of heat from the body. One study (Louhevaara et al, 1995) investigated the effects of a multi-layer firefighting turnout suit designed to fulfil European standard EN 469 (1994). When it was used over standardised clothing with SCBA, the study found an average decrease in the maximal power output, in terms of maximal working time and walking speed, of 25% compared to the control of standardised clothing only. The authors concluded that all possible means to decrease the mass of both the fire-protective 270 clothing system and the SCBA for maintaining sufficient power output in physically de- manding firefighting and rescue tasks need to be considered. This emphasises the impor- tance of the balance between the provision high levels of thermal insulation and the degree of disability afforded by this. Essentially, if a wearer considers the consequences of wearing the PPE to be worse than the likely consequences of not doing so, then the PPE will not be worn (Graveling, 1999). Garment cut Traditional ‘bunker’ style tunics are becoming a thing of the past and many brigades now wear shorter, more tailored tunics, both for ease of movement and aesthetics. EN 469 states that an overlap of 30 cm must be present between the tunic and trousers (to prevent exposure). Therefore, if the bottom of the tunic is higher, the top of the trousers must be as well. The result is a 30 cm high double layer of protection further up the body, covering a high proportion of the torso. Some of the less controversial advances in firekit design include action-back slits and under-arm gussets, which increase the range of movement without compromising protection, as the tunic is less likely to rise up. Removable liners There are two reasons for considering the incorporation of removable liners in fire- fighting kit. Firstly, a laundering and replacement issue: for example, if the outer layer becomes soiled and requires either cleaning or disposal, yet the layers beneath are still fit for purpose, only the outer layer need be dealt with. Secondly, firefighters can experience hot ambient conditions in summertime but with no immediate risk of fire, for example, attending an RTA. Such a garment construction would allow the thermal barrier to be removed, reducing the likelihood of thermal discomfort. This, however, is also a management and risk assessment issue. Sizing It may well be that in anthropometric terms, a population of firefighters differs signifi- cantly from the general population, and the author is unaware of any such work docu- menting this or otherwise. However, it has been shown that a poor design and fit of fire- protective clothing or the shoulder harness of SCBA may decrease the mechanical effi- ciency of moving and breathing, as well as cause discomfort during both submaximal and maximal work (Louhevaara et al, 1984). Therefore, correct sizing of firekit is vital in or- der to provide the thermal protection it was designed for. There is an increasing propor- tion of female firefighters in UK fire brigades who also need to be catered for. If firekit fits tightly around certain areas of the body, thermal protection will be compromised. This will be the case if female firefighters do not have specifically designed firekit. Garment testing In order for garments to be manufactured and sold as firekit, they must pass EN 469. However, none of the tests in EN 469 incorporate the garments being worn by humans carrying out firefighting activities. Heus and colleagues (Heus et al, 1992) developed a method which allows firekit to be evaluated in this way. Leicestershire Fire & Rescue 271 Service (UK) have further developed this method and recently carried out extensive tests on firekit, incorporating many aspects of the job of a firefighter. The methodology includes tests of radiant heat protection and simulated structural fire- fighting work in hot ambient conditions. A battery of ergonomics tests considers garment cut, fit, comfort, fastenings, pocket placement, interfacing with other items of PPE, and how well the wearer can reach, jump, run and bend in the garment. There is a confined space test which identifies issues such as elbow and knee pad placement, and items at- tached to the outside of the firekit, such as torches and personal lines, getting caught. The method allows for the conspicuity of the firekit with regard to reflective tape type and configuration to be assessed. In addition, a water protection test evaluates the effective- ness of the firekit at preventing water absorption and leakage. This is not only necessary for comfort, but in the prevention of steam burns. Procurement Once a brigade has made a decision as to the material combination and garment cut of the firekit they wish to purchase, they then enter the procurement process. This entails pro- viding the exact specifications of the garment, right down to the type of stitching used. Unless fire brigades are to become experts in the manufacturing industry, they require some guidance in the specification and procurement procedure if they are to make a sound purchase. Compatibility Since the financial consideration of procuring new fire-kit is often the priority for most brigades, it is common practise to purchase fire tunics and over-trousers first, followed by other items of PPE as and when the budget allows. A vital consideration here is that all the components of the firekit interface correctly; and this is unlikely if they are produced by various manufacturers who do not communicate. Mutual compatibility is a require- ment under Regulation 5 of the PPE Regulations, yet the end result of problematic PPE in this respect may be that one or other item is not worn correctly, reducing the effective level of protection from that intended (Graveling, 1999). Garment care It has now been recognised that home cleaning of firekit is not acceptable. Instead, spe- cial care and expert cleaning is required, for example, to reactivate and reapply fluorocar- bons on the outer material layer, to prevent shrinkage, and to deal with substances such as oil and body fluids effectively and safely. EN 469 states that garments must pass the standard after they have been cleaned five times. However, this does not include any wearing in the meantime, and there is no information available on the performance of firekit after a controlled number of wears and cleans. Therefore, apart from subjective opinion, brigades do not know after how many wears a garment should be cleaned, and after how many cleans a garment should be replaced. Time is not a useful indicator, due to the variation in the number and type of calls on different fire stations. Bar coding and batch testing may provide an answer to this problem over time, although such a system would require close control and would be an extensive project. 272 Litigation Although finance is always a major issue for fire brigades, due to justification to local authorities, it is very difficult to exonerate a decision to buy cheaper firekit that does not protect or perform as well when a fatality occurs. Also, in less serious but significant in- cidents, a brigade should be able to categorically state the number of times a garment had been worn and cleaned when the incident occurred, and was it ‘fit for purpose’? Conclusions Manufacturers and cleaners of firekit need to work more closely with fire brigades to de- termine the needs of a firefighter at work. An evaluation procedure such as that de- veloped by either TNO or Leicestershire F&RS would contribute to this process, and also would be a valuable addition to current firefighting clothing standards. References Abeles FJ, Del Vicchio RJ & Himel VH (1973) A firefighter’s integrated life protection system: Phase 1. Design and Performance Requirements, Gruman Aerospace Corporation, New York. EN 469 (1994) Protective clothing for firefighters[European Standard]. Brussells: Comité Européen de Normalisation. Foster JA & Roberts GV (1994) Measurements of the firefighting environment. Home Office Fire Research and Development Group, CFBAC / JCFR Research Reports 61, London: Home Office. Graveling R (1999) Ergonomics and personal protective equipment. The Safety & Health Practicioner, July 1999, suppl. pp. 29-30. Heus R, Wammes LJA & Lotens WA (1992) A comparison of six firefighter garments against a reference suit; simulated tests for general use and comfort. TNO Human Factors Research Institute, Report No: 1992-C12. Ilmarinen R, Griefahn B, Künemund C & Mäkinen H (1994) Physiological responses to wearing a firefighter’s turnout suit, with and without microporous membrane, in the heat. In: Frim, Ducharme & Tikuisis (Eds), Proceedings of the Sixth International Conference on Environmental Ergonomics, Government of Canada, Montabello, pp. 78-79. Louhevaara V, Tuomi T, Korhonen O & Jaakkola J (1984) Cardiorespiratory effects of respiratory protective devices during exercise in well-trained men. Eur. J. Appl. Physiol. Occup. Physiol., 52: 340-345. Louhevaara V, Ilmarinen R, Griefahn B, Künemund C & Mäkinen H (1995) Maximal physical work performance with European standard based fire-protective clothing system and equipment in relation to individual characteristics. Eur. J. Appl. Physiol, 71: 223-229. Mäkinen H, Ilmarinen R, Griefahn B & Künemund C (1995) Physiological comparison of firefighters’ turnout suits, with and without a microporous membrane, in the heat. In: Johnson JS & Mansdorf SZ (Eds), American Society for Testing and Materials, Philadelphia. Stirling MH & Parsons KC (1999) The effect on hydration state of exposure to extreme heat by trainee firefighters. Contemporary Ergonomics 1999, (Ed) M.A.Hanson, Taylor & Francis Ltd, London, pp 380-384. White MK & Hodus TK (1987) Reduced work tolerance associated with wearing protective clothing and respirators. Am. Ind. Hyg. Assoc. J., 48: 304-310. 273 Investigating new developments in materials and design via statistically designed experiments Dave J. Stewardson1, Shirley Y. Coleman1, John Douglass2 1Industrial Statistics Research Unit, University of Newcastle upon Tyne, UK 2Draeger Ltd. Northumberland, UK Introduction This paper discusses the improvement of a major product line at Draeger Ltd. Draeger is a German owned producer of high quality, high tech breathing equipment with a large market share in carbon composite, fully wrapped compressed air cylinders. The charac- teristics of the carbon fibre / resin matrix used to wrap the seamless aluminium liner and the conditions of the wrapping process, are known to be important factors in obtaining a consistently strong cylinder. A systematic experiment was designed to examine their combined effects in detail. The main features of this experiment are presented here. The usual way of assessing the strength of a cylinder is via a destructive test, however, an ad- ditional non-destructive method was also investigated. The relationship between the two tests was examined. Both sets of results are presented and the conclusions discussed. The potential of statistically designed experiments for understanding processes and making more robust, higher quality products is enormous and a number of other applications are discussed. ‘Design of Experiments’ Methodology When investigating processes, in order to gather information quickly and effectively it is possible to experiment over a small balanced sub-set of the total number of possible combinations of factors without a serious loss of information. The sub-sets of possible combinations are often referred to as ‘orthogonal’ designs. These methods (DoE) have been known for some time, were first described by (Fisher 1960), and have been made popular amongst engineers recently following work by the engineer (Genichi Taguchi 1986). A number of books and papers describe the basic theory behind these methods, for example (Montgomery 1991), and (Cochran & Cox 1992). In the context of product per- formance optimisation, as in this work, the idea is often to minimise the number of ex- periments required to establish which parameters are important and why. There are number of advantages to using DoE.  The number of experiments can be minimised, giving speedier and cheaper results.  The results can be used to predict outcomes within the entire experimental range.  We can identify the most important factors over a range of conditions.  The effect of changing several factors settings at the same time can be estimated.  The effect of changing one factor setting in relation to the others can be estimated.  We can estimate levels of background uncertainty (experimental error).  We can accurately estimate the cost of the experimental programme in advance.  We can often estimate the effects of factors not actually included in the design, pro- vided that these are also monitored and measured. 274  When there are many responses, we don’t need to know which is critical at the outset.  We can overcome human errors such as the incorrect setting of factors.  The method is ‘robust’ in the sense that lack of control over the factors is not fatal.  We can investigate the effect of new factors if we want at a later date. The Problem and the design The investigation is of the cycle life of the pressure cylinder. The standard prEN 12245, requires that one cylinder per batch of 200, should be tested over a pressure cycle ranging from 0-450 bar at a maximum frequency of 15 cycles per minute for up to 7500 cycles. The pass point is 3750 minimum cycles without leakage. The second test is for perma- nent expansion of a cylinder after Auto-frettage. This is tested by weighing the water dis- placed by the cylinder, using an electronic scale. If there is significant correlation between the results the second test may be used to predict the first. The factors thought to influence results are given in Table 1 and show the ranges tested over. Table 1 also designates the factor settings -1 or +1 as explained below. Resin tack level is an estimate of the degree to which the epoxy resin contained within the pre-impregnated fibre is “advanced”. Greater advancement can result in varying resin flow during the cure process. Table 1. Experimental factors. Factor Label Low (-1) High (+1) Carbon Fibre UTS UTS 5.4 Gpa 5.85 Gpa Resin Tack Level RT Low High Winding Tension WT 3.6 kg 4.5 kg Auto-Frettage pressure AF 580 bar 600 bar With four factors tested at two levels the total number of possible combinations is (24) = 16. The design used was a ‘Half fraction’ of 8 runs as shown in Table 2. This sub-set, of the 16 runs, is balanced in that every combination of settings is represented equally. The 'interactions' are used to estimate the effect of changing combinations of factor set- tings together. In this design each interaction column is the sum of two potential effects ‘confounded’ together. If such a sum is seen to be significant, we may have to run further tests to untangle the confounding. The results of the tests are also shown. Table 2. Orthogonal array with results. Factor settings Interactions Results Run UTS RT WT AF UTSxRT WTxAF UTSxWT RTxAF RTxWT UTSxAF Cycle Life Exp’n 1 -1 -1 -1 -1 1 1 1 5595 54.7 2 1 -1 -1 1 -1 -1 1 6200 55.3 3 -1 1 -1 1 -1 1 -1 6517 64.3 4 1 1 -1 -1 1 -1 -1 6210 54.9 5 -1 -1 1 1 1 -1 -1 6334 51.5 6 1 -1 1 -1 -1 1 -1 4935 41.5 7 -1 1 1 -1 -1 -1 1 8004 50.7 8 1 1 1 1 1 1 1 5528 54.7 275 Results and discussion The effects of factors and interactions are determined by comparing the difference between the average of the results of all the +1 and -1 settings in each column. The re- sultant values show the average effect of changing from the high to the low setting of a factor (main effect), or of an interaction effect. The effects from both tests are given in Table 3. We can compare these effects with the background or experimental uncertainty using a Half-Normal plot (Grove & Davis 1992). Those effects that fall away from a straight line passing through the origin are seen to be significant ones. Table 3. Factor effects. Factor UTS RT WT AF UTSxRT WTxAF UTSxWT RTxAF RTxWT UTSxAF Effect Exp’n -3.7 5.4 -7.7 6 1 0.7 0.8 Effect Cycle Life -894 799 70 -41 -497 -1043 333 It can be see from Table 3 and Figure 1 that all the Main Effects have a significant ef- fect on permanent expansion. An increase in permanent expansion may follow a reduc- tion in WT or use of carbon fibre with lower UTS. Higher AF or high RT will also cause an increase in Permanent Expansion. The situation with Cycle life is not so simple. 0 1 2 3 4 5 6 7 8 0 0.5 1 1.5 2 Normal Scores Ef fe ct in g ra m m es WT AF RT UTS Figure 1. Half-normal plot of permanent expansion effects. 4500 5000 5500 6000 6500 7000 7500 0 Ef fe ct in C yc le s WT High WT Low UTS High UTS LOW Figure 2. Interaction diagram of UTS and WT for cycle life. It is clear from Table 3 that we need to do further test of cycle life to establish which of the confounded interactions are the significant ones. What this means is that although 276 both the carbon fibre and Resin tack levels are important we need to take into account the way settings of one or more factors are effected by the settings of others. For example Figure 2 shows the interaction plot for cycle life assuming that the UTSxWT interaction is the important one. In that case then winding tension does have an effect on cycle life if it is set high. The higher the Winding tension (WT) the longer the cycle life provided that carbon fibre (UTS) is low. If WT is set low then UTS has little effect. It is also clear, from Table 3, that changes in UTS and WT settings effect the cycle life and permanent expansion similarly. Once we know the effect of all interactions on cycle life we should be able to predict this via the expansion test by choosing optimum factor settings. Untangling the interactions can be achieved by either completing the other half of the Full factorial, or by running just a few of them. In fact by running four more trials, as a balanced sub-set of the remaining eight, we can untangle the three confoundings. Conclusions By testing only 8 cylinders we have established the effect of changing the settings of any of four factors on Permanent expansion, and found much of the same information about cycle life. We will soon be able to optimise the four factor settings to maximise cycle life and get a good match between the two tests. Thus we will not have to rely on a single test when assessing the general fatigue capability of an entire batch of cylinders. An im- proved understanding of the factors that are critical to cycle life and permanent expansion after auto-frettage presents an opportunity for quality improvement. We may introduce Statistical Process Control to help minimise variation in these important factors. By do- ing this, a more consistent and reliable product may be produced. These same DoE methods can be used in any situation, dynamic or otherwise. Material strength, fireproofing, impact resistance and the like can all be maximised using simple systematic experimentation in this way. More complicated factors with more than two levels can also be accommodated in other types of orthogonal array and these are dis- cussed in several of the references quoted below. References Cochran WG & Cox GM (1992) Experimental Designs 2nd ed. Wiley, New York. Fisher R (1960) Design of Experiments 8th ed. Oliver & Boyd, Edinburgh. Grove DM & Davis TP (1992) Engineering, Quality & Experimental Design Longman, London. Montgomery DC (1991) Design and Analysis of Experiments 3rd ed. Wiley, New York. Taguchi G (1986) Introduction to Quality Engineering Asian Productivity Organisation UNIPUB, New York. 277 Design of UK firefighter clothing Richard Graveling, Margaret Hanson Human Sciences Section, Institute of Occupational Medicine, Edinburgh, EH8 9SU, UK Introduction The design of protective clothing for firefighters presents a challenging quandary. A high level of thermal protection is required against the extremely adverse environments to which the wearer might be exposed. However, this protection also severely disrupts the avenues for heat loss from the body, causing potential for considerable heat retention, particularly where the firefighter is engaged in heavy physical activities. This paper outlines the findings of a major study of UK firefighter clothing. The main purpose of the study was to determine the physiological and subjective responses to wearing various forms of protective clothing that met industry specifications. Based on both laboratory and simulated firefighting activities, it compares different types of, and approaches to, clothing design (one-piece; two-piece bunker coat etc.), and makes recommendations for the design of these garments. Methods The study collected data on the use and acceptability of different forms of firefighter garment through three questionnaires, and a series of laboratory tests and simulated and live fire exercises to determine subjective and physiological responses. Firstly, all UK Fire Brigades completed a questionnaire concerning the garments they currently used. Information was collected on the types of garments used as well as the policies and procedures in relation to inspection garments for damage, laundering and other administrative issues. A second questionnaire was developed and sent to individual firefighters to obtain their views on the garments that they wore, covering issues such as comfort, effective protection and interaction with other PPE. A third questionnaire was developed to collect information on accidents (including burn injuries) that occurred during operational duty. In order to examine differences between garments in more detail, subjective and physiological wearer trials were conducted on thirteen different firefighter’s clothing en- sembles grouped into five clusters based on the nature of the ensemble (including dif- ferences in clothing worn underneath). A total of 96 subjects took part in these trials, drawn from 18 UK based Fire Brigades. A standardised set of activities (e.g. dressing in a tender, climbing ladders, crawling and shovelling) were undertaken to examine the wearability of the clothing. Subjects completed a questionnaire following these activities to report on any discomfort, restriction of movement or other problems they had expe- rienced with the garment. Physiological trials were also conducted to estimate the energy cost associated with the different garments; subjects walked on a treadmill in ambient conditions in control clothing (shorts and T shirt) and in the firefighter’s clothing en- semble. Two simulated fire exercises were also undertaken and the physiological impact of the environment and the garments estimated based on core temperature and heart rate re- 278 cordings. The first exercise involved moderate physical work in a warm-humid environ- ment (typically 34 C WBGT); the second, light physical work while exposure to radiant heat (10 KWm-2 source, typically 79 C globe temperature). They performed each test twice, once wearing self contained breathing apparatus (SCBA) and once without, as it was thought that wearing SCBA could interact with the effect of the clothing. Finally, a live fire exercise was undertaken in which subjects entered a burning building and extin- guished the fire, again wearing SCBA. As earlier work by the IOM (Love et al, 1996) had demonstrated significant differences in physiological impact between different forms of SCBA, it was decided to standardise on a single, widely-used make of SCBA, and to select Brigades which used that make. Results At the time of the survey, UK Brigades used tunics provided by seven different manu- facturers. However, the postal questionnaire showed that the vast majority of UK Fire Brigades (82 %) were using the same form of tunic (from the same supplier), so few comparisons could be made between them. Altogether, 746 firefighters completed the postal questionnaire concerning the us- ability of the garments; over 80 % reported that they were comfortable to wear, although significant minorities (typically just under 20 %) reported that they caused some restric- tion of movement and most (80 %) stated that they became very sweaty when working in the garments. One third of respondents reported that a gap formed at the wrist when reaching up, although nearly 90% regarded cuff protection as adequate. Only 11 (1 %) of the firefighters reported ever having to withdraw from an incident early as a result of wearing a damp tunic. A question on this issue was included as it was seen to be a par- ticular concern by some Brigades. Over 700 completed questionnaires were received concerning accidents during opera- tional duty. Nearly a quarter of these accidents related to burns injuries; the majority of these occurred at the head / neck and hands / wrists, areas that are not usually covered by PPE. Burns to head and neck were taken into account in subsequent research, investi- gating the value of firehoods (Johnstone et al 1996). Although some anecdotal reports had previously been received, there were few burns through or underneath the clothing; where these did occur, it was usually due to prolonged contact with hot material. Further to this, wearability and fit were examined systematically for each ensemble during the wearer trials where subjects commented on these issues. The majority of ad- verse comments were minor, mainly relating to fit, although some design issues were also identified. Fit was particularly an issue with the one-piece garment which, by its na- ture, was closer fitting. The two-piece ensemble with short jacket and dungaree style leg- gings was generally preferred to the conventional two-piece with a ‘bunker’ coat. What- ever style was used, it was clear that good sizing was important. The results of the physiological trials showed that, on average, the clothing imposed a 15 % increase in physiological cost over the control conditions. One ensemble showed a significantly lower physiological cost than others, although this finding was not repeated when the same tunic was worn with other leggings. No other consistent differences in physiological response were found between garments. During the two simulated fire exercises and the live fire exercise, many of the partici- pants experienced near-maximal heart rates and, in many cases, considerable increases in aural temperature, mirroring findings from earlier studies (Love et al, 1996). Few sys- tematic differences between clothing ensembles emerged, although there were some indi- 279 cations that there could be advantages in utilising a cotton coverall instead of the standard uniform trousers and T-shirt under the firekit. Also, in some situations there may be a slight advantage in using a one-piece style overgarment, although this was not a strong effect and was not apparent in all test conditions. When wearing SCBA the work rate in the hot and humid conditions was reduced to ac- count of the increased metabolic load due to SCBA; no significant physiological dif- ference was therefore found when wearing SCBA. However there was a statistically sig- nificant clothing x SCBA interaction. Discussion Modern firefighters’ clothing appears to be very effective in fulfilling its primary pur- pose, that of protecting the firefighter against the direct effects of the severe environ- ments in which he or she may have to work. Few instances were reported of burn injuries to firefighters where the clothing protection appeared to be inadequate, most reported burn injuries incurred on the skin areas not covered by the clothing. The research has, however, demonstrated and quantified the negative aspect of this protection: that the clothing itself increases the physiological cost of working whilst wearing it; and the clothing can create a risk of heat stress through its considerable dis- ruption of the thermoregulatory pathways. At the onset of the research, firefighter’s clothing in the UK was regulated by a UK Home Office standard (A26), superseded during the study by a European Standard (BSEN 469). All garments used in these trials conformed to the UK Standard, and there- fore incorporated a vapour-permeable fabric layer as required. There has been some de- bate in the scientific literature regarding the value of such layers, in particular in relation to their efficacy at facilitating heat loss through the evaporation of sweat. Despite the presence of this vapour-permeable layer, the questionnaire responses from serving firefighters indicated that becoming ‘very sweaty’ whilst wearing their standard issue garments was a very common problem, clearly suggesting some disruption of the evaporative heat loss pathway. There is evidence that, although all garments tested during the study complied with the requirement for vapour permeability, the extent of that permeability varied considerably between the fabric layers used in the different makes of garments tested. Despite this, no clear advantage between similarly styled garments could be shown for one fabric combination over another. There must therefore be some question as to the value of such fabrics, endorsing the views expressed by others (e.g. White and Hodous, 1988; Goldman, 1990). The study showed that there was some scope for improvement in wearability/fit of firefighters’ clothing. For example, garment sleeves could tend to ride up with reaching or stretching movements, potentially creating an unprotected gap at the wrists. This problem had been recognised by a number of manufacturers who had incorporated thumb loops into the knitted wrist cuffs of their garments to aid retention. Unfortunately, inade- quate allowance had sometimes been made for arm extension and, as a result, some fire- fighter’s found their arm movements to be restricted if they used the thumb loop. This occurred in activities such as lifting ladders from tenders; climbing ladders; or other ac- tivities where arm extension was required. Other problems encountered included leggings with insufficient provision for the expansion in thigh diameter in squatting or kneeling; insufficient leg length; knee padding incorrectly located in leggings; and insufficient or excessive body length in one-piece garments. Whilst many of these problems are not 280 unique to firefighters’ clothing, the extreme environments in which they are to be worn throws any deficiencies into stark relief. Finally, given the level of protection required of such garments, a degree of heat re- tention and potential heat stress seems unavoidable. Therefore, recommendations were also made regarding the development and introduction of other risk control measures, including administrative measures such as steps to enhance heat tolerance as well as technical solutions possibly including pre-cooling of fighters. Conclusions The main conclusion was that no garment style and no fabric combination in the same basic style offered any meaningful advantage over any other in terms of physiological load. All created significant accumulation of body heat, even in the warm-humid envi- ronment where conditions were by no means extreme. In summary, during the laboratory trials, standard firefighter clothing typically in- creased physiological cost (oxygen consumption) by 15 % over control sessions. In simulated firefighting exercises and other trials at elevated temperatures, no consistent differences were identified between different styles or fabrics. The study supported ear- lier US studies in concluding that there appeared to be little scope for reduction in the risk of heat stress through garment/fabric design (e.g. vapour-permeable fabrics) in these conditions, although attention to other ergonomic aspects of clothing design could be beneficial. Based on the study, recommendations were made concerning firefighter’s clothing. While there is little significant physiological difference between clothing and fabrics, it is thought that there is scope for improvements in wearability. In particular, more appro- priate sizing would be advantageous, particularly where garments are bulky. The provi- sion of styling details such as loops over the thumb may also help protect the wrists, al- though these can cause discomfort; sleeve length must also be appropriate if they are to be used. Other means of managing and controlling heat stress should also be considered. References Graveling RA, Johnstone JBG, Butler DM, Crawford J, Love RG, Maclaren WM, Ritchie P (1999) Study of the degree of protection afforded by firefighters’ clothing. London: Home Office, FRDG research report 1/99. Goldman R (1990) Heat stress in firefighting. The relationship between work, clothing and environment. Fire Engineering May: 47-52. Johnstone JBG, Graveling RA, Butler DO, Butler MP, Cowie H, and Hanson MA (1996) The effectiveness and safety of fire hoods. London: Home Office. (FRDG Publication No. 18/96). Love RG, Johnstone JBG, Crawford J, Tesh KM, Graveling RA, Ritchie PJ, Hutchison PA, Wetherill GZ (1994) Study of the physiological effects of wearing breathing apparatus. Edinburgh: Institute of Occupational Medicine (IOM Report TM/94/05) White MK, Hodous TK. (1988) Physiological responses to the wearing of firefighter’s turnout gear with neoprene and Gore-Tex barrier liners. American Industrial Hygiene Association Journal, 49, 521- 530. 281 Effects of clothing design on ventilation and evaporation of sweat Emiel A. den Hartog Thermal Physiology Group, TNO Human Factors Research Institute, Soesterberg, The Netherlands. Introduction An important aspect of thermal comfort in protective clothing is the possibility to dissi- pate heat by evaporating sweat. If the produced sweat is not transported away from the body, it just drips off and is inefficient in cooling the body. To have optimal cooling, the evaporated sweat should encounter no impediments in flowing away from the skin. How- ever, while wearing protective clothing, the flow of sweat is impeded by the clothing. Analogous to the heat resistance, or insulation, the vapour resistance denotes the resis- tance to evaporated sweat due to clothing. The resistance to heat and water vapour de- creases due to wind and body movements. Wind and body movements move the enclosed air beneath the clothing, thus reducing its insulation, called the pumping effect. The amount of vapour that is transported away depends on: air-permeability of the fabric, the presence of openings, the amount and location of enclosed air and maybe other factors. Considering these factors it can be assumed that the design and fit of clothing will also influence the transportation of sweat from the body. In a number of previous studies, the effects of air permeability of (protective) clothing materials on vapour transfer trough clothing and on heat stress has been shown (Havenith et al., 1997, Den Hartog et al., 1998). In this paper, the results of pilot experiments on the effects of clothing fit and design on vapour transfer through clothing are reported. Methods Ventilation measurements The vapour resistance was determined in a series of experiments on three subjects by de- termination of the ventilation of a tracer gas (Argon) at room temperature. The method has been described extensively by Lotens and Havenith (1988, 1990). The subjects wore a harness of polyethylene tubes which blew air, enriched with 10 % Argon under the garments. A similar harness was used to suck out air at the same rate. Both harnesses were connected to a pump which provided the constant air flow. The concentrations of Argon in the air flow to the suit (Cin), coming out of the suit (Cout) and in the surrounding air (C air) were measured with a mass spectrometer. From these concentrations the ventila- tion under the suit was calculated by:     1 1Vent V C C C C lpump in out out air ˜   ˜  ' ( ) ( ) min In which V’pump is air flow generated by the pump. To calculate the ventilation only the relative differences of the concentrations are used. Therefore, it was not necessary to 282 calibrate the output signal of the mass spectrometer. Via the ventilation (Vent) the vapour resistance (d, in mm air equivalent) can be calculated by:    2 60 106d D A Vent mmAr D  In which DAr is the diffusion constant of Argon, AD is the body surface area and 60#10 6 is a constant to transform the vapour resistance to mm air equivalent. From equation (2) it becomes clear that the vapour resistance of the garment is dependent on the ventilation of the garment. Experiments Two pilot experiments were performed with different aims. In both experiments one subject participated, wearing the harness mentioned above. In the first experiment the subject wore three cotton coveralls over the ventilation har- ness, all made from the same highly air permeable cotton fabric. The first coverall was a commercial “off the shelf” product (CO, size 40). The second coverall was adapted to the body length and waist circumference of the subject (MC), in fact the waist and hip widths were enlarged for this subject. The third coverall was tailor made to the subjects body dimensions (MM). The ventilation in all three coveralls was determined: 1) while stand- ing still, 2) walking on a treadmill and 3) standing still and waving his arms. The coverall were measured in the order MM - CO - MC - MM. In each condition the ventilation was determined four times. In the second experiment two protective clothing garments were used (A and B) which were equal in insulation and vapour resistance, as measured on a thermal manikin, but different in design, especially in the length of the coat and the trousers. The garments were comparable to fire fighters’ garments in terms of thickness and insulation. Both garments were tested in one subject while doing various movements of arms and legs. In each condition the ventilation was determined three times. In both experiments the statistical analysis was performed by a repeated measures ANOVA, with p<0.05 as a significant difference. Results In figure 1 the results of the ventilation measurements on the three coveralls are pre- sented. The statistical analysis showed that the ventilation in the MC coverall was sig- nificantly larger than in the MM coverall (p=0.002) and the CO coverall (p=0.02) in all conditions. The difference between the CO coverall and the MM coverall was not signifi- cant over all conditions (p=0.08). The effect of body movement on ventilation is clearly demonstrated, as it increases from about 170 l/min standing still, to about 300 l/min during movements of arms and legs. In figure 2 the results of the experiments on the protective clothing garments are pre- sented. The tests showed that while standing still the ventilation of garment A was 17 l/min whereas, the ventilation in garment B was 12 l/min (71 %), leading to values for the vapour resistance (d) of 102 mm and 144 mm (garments A and B respectively). There- after, measurements were performed during various movements, such as walking, run- ning, waving arms and kneeling/standing up (figure 2). The ventilation of garment A during movements was on average about 100 l/min (d = 17 mm), whereas the ventilation of garment B was about 50 l/min (d = 34 mm), a decrease of 50 %. Furthermore, for both 283 garments the increase in ventilation due to arm movements was larger than due to leg movements. MM1 C O M C MM2 CONDITION Ve nt ila tio n [l /m in] 0 50 100 150 200 250 300 350 400 Standing sti l l Walk ing Waving arms Figure 1. Results of the ventilation measurements in the three different coveralls. The results are shown in order of measurement, MM1 is the ‘made-to-measure’ coverall measured for the first time. CO is the confection coverall. MC is the confection coverall adapted for body length and waist circumference. MM2 is the made to measure coverall measured for the second time. The symbolsdenote the median values, the bars denote the 25th percentiles and the whiskers denote the ranges. 0 50 100 150 200 250 300 350 Sta nd ing sti ll Wa lkin g s tra igh t le gs Wa lkin g Ru nin g Wa vin g a rm s Kn ee ling /sta nd ing Vir tua l-s tai r c lim bin g Ve nt ila tio n [l/ mi n] Garment A Garment B Figure 2. Results of the ventilation measurements in to protective clothing. On the x-axis the various movements are indicated, during which the ventilation has been measured. In all condi- tions the ventilation in garment A was higher. The bars indicate the errors from the repeated measures. Discussion Analogous to previous studies, the first result of all ventilation measurements is that the ventilation and, therefore, the vapour resistance, is strongly decreased by any movement of the subject. The overall ventilation, as obtained from the measurements, is a summa- tion of the ventilation through the fabric and the ventilation through the openings of the 284 garment (sleeves, collar, etc.). In both experiments performed in this study the materials were the same between garments, but the design of the garments differed. Thus, we ob- tained information about the effects of clothing design and fit on the ventilation, and thus, on the vapour resistance to evaporated sweat. The coveralls of the first experiment were made of normal cotton, and had normal sleeves and collars. The overall ventilation levels were high in all conditions, but the ventilation in the MC coverall was significantly larger than in the other coveralls. On average the lowest ventilation, corresponding to the highest vapour resistance, was found in the MM coverall. The explanation can be found from the design and the sizing of the coveralls. The hip and waist widths of the MC coverall were largest (990 and 923 mm). The CO coverall had widths of 973 mm and 915 mm, and the MM coverall was smallest (855 and 910 mm). As the front of the coveralls consisted of a closure with buttons, the arm and leg movement caused most ventilation to occur in the waist area. The MC cover- all enclosed the largest air layer at the waist, this probably led to the largest ventilation values during movements. However, one should realize that the measurements were performed at only one sub- ject, which limits the statistical power. Moreover, the differences in ventilation will be too small to cause differences in thermal comfort and physiological parameters as heart rate, sweat rate or skin temperature (Havenith et al., 1998). On the other hand, it is sur- prising that any differences in ventilation are found in these highly air permeable cover- alls. It leads to the idea that physiologically significant may be found in thicker or less air permeable (protective) clothing. In the second experiment two protective clothing garments, similar to fire fighters’ garments, were tested. In these garments, the absolute ventilation levels were much lower than in the coveralls. Both garments had similar characteristics as measured on a thermal manikin. Due to design differences the ventilation in suit B was 40 % to 50 % lower during body movements. The vapour resistance during movements was computed as 12 mm for garment A and 23 mm for garment B. This difference may very well lead to physiological differences in heat strain, especially during moderate work under normal environmental conditions. In conclusion, the design of protective clothing may have a significant influence on the ventilation through the clothing. Differences in ventilation as large as found in this study may lead to differences in thermal comfort while wearing the garments. References Lotens WA & Havenith G (1988). Ventilation of rain wear determined by a tracer gas method. Env. Ergonomics; Mekjavic I.B., Bannister B.W., Morrison J.B., Eds, Taylor & Francis, London 162- 176. Den Hartog EA, Wammes LJA & Havenith G (1998), Influence of air-permeability of nbc-suits on vapour resistance and evaporation of sweat. Proc. 8th International Conference on Environmental Ergonomics, San Diego, USA. Havenith G & Lotens WA (1990). Clothing ventilation, vapour resistance and permeability index: changes due to posture, movement and wind. Ergonomics, 33: 989-1005. Havenith G, Zeijl A & Wammes LJA (1998). Heat strain in NBC-protective clothing in relation to the air- permeability of the outer layer. Report TNO-HFRI: TM-98-A008 (In Dutch). 285 Physiological load during tunnel rescue Ulf Danielsson1, Henri Leray2 1 Defence Research Establishment, SE-172 90 Stockholm, Sweden 2 Södertörns Brandförsvarsförbund, SE-136 25 Haninge, Sweden Introduction Accidents in tunnels may cause dangerous and time-consuming work situations for the rescue personnel. Yet, the number of tunnels and underground constructions are steadily increasing and so is the number of people working or being transported in such installa- tions. The rescue problems are associated with long distances where the work often must be manual because of debris and other obstacles. Such rescue conditions may imply a substantial physical load to the people involved. Trials have revealed low endurance at self-paced rescue work. Two groups of six, all well-trained firemen, had to stop after less than one hour because of exhaustion despite of favourable environmental conditions. Smoke and high temperatures make the work much more dangerous and stressful. To re- duce the risk of injury, individual protection must be used most of the time. If self- contained breathing apparatus is needed longer than 20 minutes, compressed air can not be used because of its limited capacity. For longer periods extra air bottles must be brought, a situation that may increase the physical load further. An alternative solution is to use a re-breathing device using oxygen as respiration gas. The purpose of this investigation was to study the physical load of protective en- sembles and respiration devices used, on firemen performing a simulation rescue work in a rock tunnel. The aim of the study was also to predict what length of work period can be endured during various climatic conditions and work loads. Methodology The investigation was performed in a rock tunnel, being one part of an underground naval dockyard. The mission was to rescue victims from the scene of an accident to the tunnel mouth, a distance of 540 m. Six full-time employed firemen, five men and one woman, participated. Their average age, mass and height were 33 years, 77 kg and 181 cm, respectively. They wore protective turnouts and used breathing apparatus. Each fireman participated four times wearing either a thick fire-protective turnout (RB90) or a navy combat uniform (SSD93) and breathed compressed air (Interspiro) or oxygen (Dräger BG4). The total mass of ensemble and breathing apparatus was roughly 25 kg. The fire- men pulled a rescue cart loaded with compressed air bottles at the speed of 1.0 m/s. The cart was designed also for transportation of victims lying on stretchers. The weight of the cart, without victim, was about 250 kg. The climate along the rescue path was recorded by data loggers attached to the cart. During almost two hours four missions were com- pleted with one victim (mass approx. 90 kg) rescued each time. Hence, the total distance covered was 4.3 km. The order of used ensemble and breathing apparatus was different for each participant. During each mission the two firemen pushed and pulled the cart in turns. The power demand was measured and calculated from four methods; by measuring lung ventilation rate and oxygen fraction in expired air minute by minute (separate test), 286 by recording the pressure in the oxygen bottle at every start and stop of activity and by weighing the oxygen bottle before and after each mission (main study) and finally by a continuos measurement of the force needed to pull the cart along the rescue route and then add the measured power demand when walking without pulling the cart (separate test). Furthermore, rectal temperature, nude and clothed body mass were recorded before and after completed mission whereas heart rate was collected continuously. Perceived exertion and temperature were rated at each stop. The RB90 ensemble was composed of a jacket with hood and trousers, briefs, stockings, long underwear, helmet, boots and gloves. The SSD93 ensemble was composed of field jacket (without hood) and trousers, briefs, stockings, long vest, sweater, helmet, boots and gloves. The main difference between the two ensembles in respect of thermal properties was that RB90 is con- siderably thicker and has a hood. Results The climate in the rock tunnel is very stable during the year and with only minor varia- tions along the rescue route. The average temperature was 12 °C and the relative hu- midity was 45 %. The radiation temperature from the rock walls and the paved road was similar to the air temperature. There was no external wind. The mean metabolic power was 462 W and there were only small differences (about 20 W) between the techniques used. The most strenuous part of the rescue work was to reach the victim, lying on a staircase, place him on the stretcher and then carry him to the cart, a total distance of 50 m. The power demand of this procedure was roughly 700 W. Highest average heart rate for the whole mission was obtained for RB90+air (130 beats/min) while the lowest was for SSD93+oxygen (117 beats/min). Breathing oxygen seems to reduce heart rate with at least 5 beats/min compared with compressed air irrespective of ensemble worn. The average heart rate increased slowly throughout the trial. The individual values did not vary much along the route with one exception, the transportation of the victim to the cart. Then heart rate reached 160-180 beats/min. RB90+air produced greatest body mass loss (1.9 kg) while SSD93+air caused a loss of 1 kg. Both the ensemble and respiration gas contributed significantly to the mass loss. The difference between the various combina- tions in respect of water evaporation was minor but for a given ensemble breathing oxy- gen resulted in significantly greater evaporation than using compressed air. About 20 % of the evaporation were lost from the lungs. Including this part about 50 % of the body mass loss evaporated when SSD93 was worn whereas 35-40 % was evaporated from RB90. The differences between ensembles as well as respiration devices were significant. Rise in rectal temperature was significantly higher for RB90 than for SSD93. For a given ensemble the oxygen device produced greater rise in rectal temperature but this dif- ference was not significant. However, oxygen apparatus produced about 4 °C higher face temperature than compressed air did. Breathing temperature was also rated significantly warmer with the oxygen device. RB90 was considered significantly warmer than SSD93 in respect of body temperature. For a given ensemble oxygen breathing tended to give higher perceived (body) temperature but this difference was not significant. Rated per- ceived exertion did not display any significant differences between the various combinations. 287 Discussion The metabolic rate measurements showed that the load was shared approximately equal between the subject pulling the cart and the one pushing it. The power demand on each fireman was about 460 W of which roughly 40 % referred to walking, carrying clothing and breathing apparatus while 40 % referred to pulling/pushing the cart. The remaining 20 % refer to the basic metabolic rate. The power demanded can be estimated to about 650 W if only one person pulls the cart at 1m/s. The grade of the rescue route was ± 1° so with steeper slopes the load can rapidly become exhaustive. The estimated demand on two people sharing the job is 600 W per person if the speed is 1.35 m/s. This was the average speed chosen by the two groups failing during the initial test. A power demand of about 500 W during eight hours should be endured as the participating firemen has a maximum work capacity exceeding 1050 W, assuming favourable heat dissipation con- ditions. This was not the case as the work was performed in protective garments. To re- duce the risk of heat exhaustion the speed was strictly kept at 1.0 m/s. Yet, heart rate continued to rise during the whole mission. The dehydration rate was not alarming, less than 2.5 % on average, but yet one individual exceeded 4 %. The RB90 ensemble caused the greatest heat load, an expected result, as this ensemble is considerably thicker than SSD93. The hood certainly added to this difference. The oxygen breathing apparatus also contributed more than the compressed-air device to the heat load, an effect of warmer respiration gas giving a warmer face. This effect was however smaller than the effect of ensemble. In spite of favourable tunnel climate and a relatively low work rate, a steadily increasing heat storage followed. The almost two-hour mission could be sustained but the 4-hours capacity of the oxygen-breathing device could probably not have been utilised. The firemen estimated that another hour could have been endured in the RB90 ensemble and about 1.5 hours wearing SSD93. This illustrates two problems arising in long rescue work, the intensity of activity and the level of protection chosen. Even though most of the participating subjects were experienced firemen two groups from the initial test did not realise that they worked too hard and consequently became exhausted. The two groups who could accomplish the mission chose a lower speed, about 1.1 m/s. Obviously, during long rescue activities the work rate must be guided. This should also concern the protec- tion level chosen. In case of high temperatures or open fire a thick whole-body-covering turnout must be used to prevent from burn injury and "external" heat exhaustion. But, there is a number of situations where these conditions are not expected to occur. Still, firemen are most often wearing their thick turnouts with very little ambition to adjust their protection level. Rescue work in tunnels is associated with great physical and men- tal load on the personnel. The environmental conditions are not always that favourable as in this study. A fire in a tunnel causing a temperature of 500 °C at the scene of accident may cause a temperature of 80 °C, 500 m from the seat of fire if the tunnel is dry or 35 °C and 100 % rh if wet (Wolki, 1991). A simulation program, INSULA, was used to estimate the physiological load based on the work conditions prevalent during the tests. Calculations showed e.g. that the rise in core temperature and heart rate were very close to those measured for the various combi- nations of ensemble and breathing apparatus. Calculations also predicted well the heart rates reached during the initial test. These agreements gave support to predictions of the effects of various work rates and environmental conditions on the thermal load. Figure 1, left panel, shows the accumulated body heat (kJ/kg) after 1.5 hours of rescue work in a climate similar to that during the initial test. The figure shows that 600 W, corresponding to a walking speed of 1.35 m/s, would give a heat storage exceeding 10 kJ/kg, a level that is associated with heat exhaustion. This explains why the two failing groups only 288 managed to work about one hour. The two successful groups had a speed of about 1.1 m/s giving a metabolic rate of about 500 W. Estimated heat build is around 7 kJ/kg, a level that well-trained firemen are expected to manage. Another calculation was done, assuming that a vehicle was used for the transportation, reducing the metabolic rate to 175 W during this phase (Figure 1, right panel). The other activities and the personal equipment were unchanged but the air temperature could adopt different values. Figure 1, right panel, shows that the exhaustive level of 8 kJ/kg is reached after 1.5 hours when the air temperature is about 36 °C, a situation that hardly can be considered as extreme, espe- cially if there is a fire in the tunnel. These two simulations indicate that long rescue ef- forts are associated with a significant heat load if protective ensemble is worn. The situa- tion is improved if heat production is kept as low as possible, but this will be insufficient at elevated ambient temperature and humidity unless the rescue people has access to ex- ternal cooling. If the fireman can be connected to a cooling system during the vehicle transport, the accumulated body heat reaches only 1.2 kJ/kg after 1.5 hours of rescue work if 5 litres of dry cool air (+20 °C) is allowed to ventilate the turnout every second. These simulations indicate that the rescue capacity could increase considerably if a motor vehicle is used to reduce physical load. Such vehicle should also deliver cool air for en- semble ventilation to reduce the risk of heat exhaustion. 0 2 4 6 8 10 12 100 200 300 400 500 600 700 B od y he at s to ra ge (k J/k g) Power (W), pulling/pushing air temperature: 15°C air humidity: 45% Speed: 1,35 m/s Speed: 1,0 m/s 0 2 4 6 8 10 12 14 10 15 20 25 30 35 40 45 50 B od y he at s to ra ge (k J/k g) Air temperature (°C) Air humidity: 45% Figure 1. Calculated body heat storage reached after 1.5 hours of rescue work including move- ments of the victim (metabolic rate, MR=700 W) and short rests (MR=150 W). Fire protective turnout and oxygen apparatus are used. Left panel shows the effect of metabolic rate (speed) when the cart is pulled/pushed. External wind is the same as that of the walking speed ( § 1 m/s) Right panel shows the effect of air temperature assuming that a motor vehicle, MR=175 W, is used for the transportation. Air speed inside the vehicle is 0 m/s. Conclusions Rescue in tunnels may cause heat exhaustion, also in cool climates, if the work period is sufficiently long. The protective clothing is the main cause of the elevated risk although the work intensity is of importance too. Because of that, the advantages of the oxygen re- breathing system cannot be utilised unless work rate is low and/or personal cooling equipment can be used. A motorised vehicle can provide these demands and this would improve the rescue capacity substantially. References Wolki, J (1991) Method for predicting airway climatic conditions during a fire and its validation using reduced scale experiments. Mining Science and Technology, 12. 289 Effectiveness of a light-weight ice-vest for body cooling in fire fighter’s work Juhani Smolander, Kalev Kuklane, Désirée Gavhed, Håkan Nilsson, Eva Karlsson, Ingvar Holmér Programme for Respiratory Health and Climate, National Institute for Working Life, S - 171 84 Solna, Sweden Introduction Fire-fighter’s work, especially smoke-diving, often involves exposure to heavy physical work and heat stress. This combination of stress factors reduces efficient work time and productivity and may increase the risk of heat-related illnesses. Several different auxiliary personal body cooling devices have been developed for the industry (Kamon et al., 1986; Muir et al., 1999). They are, however, often of limited value for fire fighters because of their weight. An ideal body cooling device for fire fighters should be rather light-weight, should not interfere with the job performance, and it should be feasible and easy to use during the fire alarms. The aim of the present study was to examine, in a systematic way, the effects of wearing the ice-vest on physiological and subjective responses in fire fighter’s work. The experiments were carried out in a climatic chamber, in a container under extreme radiant heat, and during simulated smoke diving. In addition, the physical cooling effect of the ice-vest was measured with a thermal manikin. Subjects and methods Subjects: Four experienced fire fighters participated in the experiments. They were healthy, moderately physically active, and all had a normal blood pressure. Their physical characteristics were; mean age 36 years (range 32-39), height 1.86 m (1.85-1.88) and weight 87 kg (78-98). Clothing and icevest: In all tests, the subjects wore the standard clothing for Swedish fire fighters (RB90) with self-contained breathing apparatus (AGA Divator). During the labo- ratory tests, the face mask was not used because this would interfere with the measure- ment of oxygen consumption. The total weight carried was 21-23 kg. The ice-vest was developed in co-operation between the Swedish Rescue Board and Flexy-Ice AB. The material is cotton. The inside consists of two removable flat plastic containers, which have several small pockets for water/ice. Five different vests were used in the tests. Their weight varied slightly (1.0-1.1 kg). The ice-vest covered most of the trunk area and it was worn over the underwear. The vests were kept in a freezer at -20 °C over night before the experiments. Laboratory experiments: Laboratory experiments were carried out in a climatic chamber, where the air temperature was 45 °C, and relative humidity 30 %. In the tests, the sub- jects walked on a treadmill for 30 minutes twice at a moderate exercise intensity (4 km/h, 0 degrees), and twice at heavy exercise intensity (4 km/h, with an inclination of 4 de- 290 grees). At each work intensity, one test was done without and one with the ice-vest. A 5 minute rest period preceded each test. For each subject, only one test was done in one day. Measurements: Rectal temperature was measured with a thermistor probe (YSI401, USA) inserted 10 cm beyond the anal sphincter. Skin temperatures were measured with thermistors (StowAway) taped to the skin. The sites for skin temperature measurements were forearm, upper arm, chest, back, thigh, and calf. In all tests, heart rate was measured once a minute with the telemetric SportTester sys- tem (PolarElectro, Kempele, Finland). Oxygen consumption was measured with a port- able gas analysing system (Metamax, Cortex, Germany) twice during each test (10-15 min and 25-30 min). For that purpose the subjects wore a half-face mask during the tests instead of the full-face mask of AGA Divator. The amount of evaporated sweat and the amount of sweat in the clothing items were estimated by weighing (Mettler-Toledo, KC 240, 2 g). During the tests, the subjects rated their perceived exertion (RPE), thermal sensations, and comfort with standard scales. Results and discussion Oxygen consumption was, on average, very similar between the tests with and without the ice-vest; 1.17 and 1.18 l/min at moderate work and 2.23 and 2.25 at heavy work. The variability in metabolic rate was due to differences in body weight between the subjects. During the first 15 minutes of walking at the moderate work level the average heart rate was similar between the tests with and without the ice-vest. However, during the last 15 minutes of walking the heart rate was lower with the ice-vest. At the end of walking the heart rate was ca. 10 beats/min lower with the ice-vest compared to walking without it. At the heavy work level the heart rate increased continuously during all tests. As shown in Figure 1, throughout the walking the average heart rate remained about 10 beats/min lower when the ice-vest was worn compared to walking without the vest. 60 80 100 120 140 160 180 200 0 5 10 15 20 25 30 Tim e (min ) H e a rt r a te (b e a ts /m in ) without vest with vest Figure 1. The average heart at the heavy work load with and without the ice-vest. At both work levels, rectal temperature increased continuously during the tests, reach- ing higher values at the heavy work level (Figure 2). During both tests, the resting level of average rectal temperature was slightly lower with the ice-vest. Therefore, the increase in rectal temperature was rather similar between tests with and without the ice-vest. 291 36.5 37.0 37.5 38.0 38.5 39.0 0 5 10 15 20 25 30 35 40 T im e (m in ) T em pe ra tu re (°C ) without vest with vest Figure 2. The average rectal temperature during walking in the heat at the heavy work load with and without the ice-vest. The walking started at minute 10. During tests with ice-vest the back skin temperature decreased during the first 10-15 minutes over 10 oC compared to tests without the vest (Figure 3). Then it gradually started to increase due to warming up of the melted water. The decrease in chest skin temperature with the ice-vest was much less than in the back. This difference was very likely caused by the pressure of the gas cylinders on the back causing reduced insulation between vest and skin. In other skin areas the temperature differences were rather small. 20.0 25.0 30.0 35.0 40.0 0 5 10 15 20 25 30 35 40 T im e (m in ) B a c k sk in te m pe ra tu re (° C ) without vest with vest Figure 3. The average back skin temperature during walking in the heat at the heavy work load with and without the ice-vest. The walking started at minute 10. At both work levels the amount of sweat was less with the ice-vest compared to tests without the ice-vest (Table1). At the moderate work level the reduction in sweat produc- tion was 17 %, and at the heavy work level 9 %. Table 1. Amount of sweat produced and evaporated sweat during the tests with and without ice- vest at the moderate and heavy work level. The values are the means (range) for 4 subjects. Total sweat (g) Evaporated sweat (g) Work level Without ice-vest With ice-vest Without ice-vest With ice-vest Moderate 443 (385-485) 369 (250-447) 183 (146-206) 135 (99-187) Heavy 608 (497-693) 555 (505-598) 210 (175-250) 195 (183-217) The amount of sweat in clothing was slightly less when the ice-vest was used (Table 1). Also, the absolute amount of evaporated sweat was slightly less. But expressed in per- 292 centage of sweating, the amount of evaporation did not differ between tests with and without the ice-vest. At the moderate work level the rating of perceived exertion (RPE) was similar between tests with and without the ice-vest. At the heavy work level, RPE was during the first 10 minutes very similar between the two test conditions, but at 20 and 30 minutes of walk- ing the RPE was about 1 unit lower with the ice-vest. At both work levels, the subjects felt cooler in the whole body when wearing the ice- vest. Naturally, the cooler sensations were felt under the ice-vest on chest and back. Also, thermal sensations were slightly lower in other areas with the ice-vest (Table 2), espe- cially at the heavy work level. In addition, the subjects felt more comfortable at the heavy work level when the ice-vest was used compared to walking without the vest. Table 2. Thermal sensation of the whole body during tests with and without the ice-vest at the moderate and heavy work level at minutes 10, 20, and 30. The values are the means (standard deviations) for 4 subjects. See table 1. Work level Time Without ice-vest With ice-vest Moderate 10 1.2 (0.5) 0.0 (2.2) 20 1.8 (0.5) 0.3 (1.7) 30 2.0 (0.0) 1.3 (0.5) Heavy 10 1.5 (0.6) 0.3 (1.7) 20 2.0 (0.0) 1.0 (0.8) 30 3.0 (0.0) 1.5 (1.0) Conclusions It can be concluded that the light-weight ice vest reduces circulatory, thermal, and sub- jective strain during moderate to heavy work in a hot environment. References Kamon E, Kenney WL, Deno NS, Soto KI, Carpenter AJ (1986) Readdressing personal cooling with ice. Am Ind Hyg Assoc J, 47:293-298. Muir I, Bishop P, Ray P (1999) Effects of a novel ice-cooling technique on work in protective clothing at 28 degrees C, 23 degrees C, and 18 degrees C WBGTs. Am Ind Hyg Assoc J, 60(1):96-104. 293 Fire fighter garment with non textile insulation Michael Hocke1, Lutz Strauss2, Wolfgang Nocker3 1 Institute of Applied Ergonomics, D-81543 Munich 2 Professional Fire Fighter Brigade of Berlin 3 W.L. Gore GmbH, D-85640 Putzbrunn Introduction Fibres and yarns are not the real thermal insulators of a garment. It is the locked still standing air. Fibres conduct the heat 10 to 20 times better than still standing air. This was the idea to substitute the traditional textile insulation by an air cushion. GORE-TEX® Airlock® is a combination of moisture barrier and thermal protection: Heat stable „spacers“ of foamed silicone on the GORE-TEX® moisture barrier create the insulating air buffer. Hot plate measurements showed that at similar heat resistance we gained a 40 % lower water vapour resistance with the new system whereas the water vapour absorption was reduced by 60 %. Transient measurements showed that the Airlock® system features a higher transportation rate but lower absorption rate for liquid water. Consequently a combination with Airlock® causes a significantly shorter drying time. Conclusion out of this physical data: For wear situations with strong sweating, as expe- rienced by fire fighters, a material combination with an Airlock® liner can be expected to show a better physiological performance than a material combination with a conventional textile insulation. In a controlled wear test on a treadmill in the climatic chamber we tried to proof this. Methods Five professional fire fighters wore the following ensembles at 30°C / 50 % RH for 95 min:  Leather jacket with lining but without moisture barrier, Paris style, GORE-TEX® fire fighting trousers  Textile jacket (Nomex, Aramid lining, GORE-TEX® barrier), Berlin style, GORE- TEX® fire fighting trousers  Textile jacket (Nomex, Airlock®, lining), Paris style, modified Airlock® fire fighting trousers Long-legged underpants and long-sleeved undershirt of a functional material (Ullfrotté: 60 % wool; 25 % PES; 15 % PA) have been worn underneath. To be close to practice each test subject wore its boots, gloves, helmet, belt and breathing apparatus (without respirator mask). The load regimen was based on former studies and consisted of work and rest cycles: 10 min 4 km/h 0 % 10 min rest 5 min 5 km/h 5 % 10 min rest 294 5 min 5 km/h 10 % 10 min rest 5 min 5 km/h 8 % 40 min rest Parameters measured: Core temperature (rectal), heart rate, skin temperatures, relative humidity between underwear and suit, weight loss of subject, weight gain of clothing. At the end of each cycle, the test subjects rated their heat and moisture perception as well as the wear comfort. Results and discussion Core temperature: Up to the 25th test minute, the temperature increases of the body core lay within a relatively narrow temperature band and on a comparatively low level. From that time on, however, distinct increases were observed. In the Leather fire fighting jacket, the body core temperature rose by 0.9 K, in the Berlin GORE-TEX® jacket by 0.65 K, whereas an increase of only 0.45 K was measured in the jacket Airlock®-Paris. In the Leather jacket the body core temperature was continuously rising, even after the load phase had ended (55th test minute), and kept on rising until the end of the test in the 95th test minute. In the two other jackets, the body core temperature kept gradually fal- ling from the 75th test minute on. Mean skin temperature: The temperature curves lie within a band of approximately 2 to 1.3 Kelvin; Leather on the top (more than 37 °C) and Airlock®-Paris the lowest skin temperature (nearly 36°C). Berlin GORE-TEX® was in the middle. Heart rates: They rose in a way, which was typical for the load regimen of the test and lay within a physiologically plausible bandwidth. The small differences between the mean value curves reveal that the individual jacket types have less influence on the heart rates. It was only in the fire fighting jacket Airlock®-Paris that the heart rates returned nearly to the starting levels from the 65th test minute on. Weight change: The weight losses of the test persons were, on an average, approximately 1 kg in the Airlock®-Paris jacket, 1.3 kg in the Berlin GORE-TEX® jacket and 1.7 kg in the Leather jacket. The water uptake of the garment ensembles coincided with the weight losses: 0.5 kg in the Airlock®-Paris jacket; 0.7 kg in the Berlin-GORE-TEX® jacket and 1 kg in the Leather jacket. Relative humidity: The humidity measured in the Leather jacket ranged about 5 to 10 % above the levels measured in the two other jackets, throughout the test. It was especially in the Airlock®-Paris jacket that the humidity reached a steady state slightly above 80 % RH. Heat perception: It was ranked on a scale from 0 to 7 (0=comfortable; 1=slightly warm; 2=warm; 3=very warm; 4=hot; 5=very hot; 6=uncomfortable; 7=intolerable). Up to the end of the last walking period (55th min) the Airlock®-Paris jacket was perceived as warm to very warm; Berlin GORE-TEX® as hot and Leather as very hot. This perception continued till the end of the test, whereas the two other jackets showed an improvement. 295 Moisture perception: It was ranked on a scale from 0 to 7 (0=dry; 1=chest or back slightly moist; 2=chest or back moist; 3=body moist; 4=body moist with clothing partly sticking to the body; 5=perspiration is running down at some spots; 6=perspiration is pouring down the body in many areas; 7=intolerable). Up to the end of the last walking period (55th min) Airlock®-Paris ranked 3rd and Berlin-GORE-TEX® and Leather 5th. Wear comfort: It was ranked on a scale from 1 to 6 (1 = excellent; 2 = good; 3 = satis- factory; 4 = uncomfortable; 5 =very uncomfortable; 6 = extremely uncomfortable). Throughout the test period, Airlock®-Paris was perceived as excellent; Berlin GORE- TEX® as good and Leather as very uncomfortable. The outcome of the study was that for all parameters, except for the heart rate, a clear rank can be assigned to each jacket type tested (rank 1 = best, rank 3 = worst per- formance): 1. Airlock®-Paris 2. Berlin GORE-TEX® 3. Leather Conclusions In the fire fighting suits with the new combination of thermal protection and liquid bar- rier very favourable thermophysiological conditions prevailed. Such suits can be ex- pected to produce less heat stress at the wearer. Fire fighting suits with Airlock® fulfil EN 469 and had been successful in thermo-man-tests. With the new concept the bulki- ness of insulation could be reduced while maintaining the same level of heat protection. Due to minimal moisture absorption and high moisture vapour transfer the risk of injuries by scalding should be reduced. High flexibility and reduced weight of such suits in- creases the wear comfort. References Schopper-Jochum S, Schubert W & Hocke M (1997) Comparative evaluation of the wear properties of fire fighting jackets (phase1). Arbeitsmed., Sozialmed., Präventivmed. 32, 138-144. Wagner S, Will N & Schuckmann F (1995) Investigations on the strain to which the heart and circulatory system of professional fire fighters are exposed. Ergometrics and test range. In accordance with the principles of state of the art preventive examinations in occupational medicine for persons wearing a breathing apparatus. Arbeitsmed., Sozialmed., Präventivmed. 30, 169-173. INDIANAPOLIS-study (1998). Field evaluation of protective clothing, effects on fire fighter physiology, predictive capability of total heat loss test. Final report to the IAFF. 296 Assessing fire protection afforded by a variety of fire- fighters hoods James R. House1, James D. Squire2, Ron Staples2 1 Institute of Naval Medicine, Gosport, PO12 2DL, United Kingdom. 2 Defence Clothing and Textiles Agency, Colchester, CO2 7SS, United Kingdom. Introduction Fire fighting crews on Royal Navy (RN) warships wear multi-layered Proban® treated Flame-Retardant (FR) clothing, breathing apparatus (BA), a helmet, gloves and boots. They also wear a single-layered FR treated cotton `anti-flash' hood to protect the areas of the head not covered by the BA mask or helmet. Previous work has demonstrated that this hood provides insufficient protection and serious head burns are probable if the fire fighter was accidentally engulfed in flame for more than four seconds (House, 1998). This study was undertaken to assess how better to protect the head from flame to explore the effect of using up to four material layers in a prototype hood. The study was designed to test the hypotheses that: a. protection will increase with the number of layers added; b. the greatest improvement will occur when doubling layers from 1-ply to 2-ply; c. using three or more layers affords little extra protection. Methods Four prototype hoods of similar design (1, 2, 3 and 4-ply) were constructed and these were tested in comparison to the current hood, a RN aircraft-crash fire-fighting hood and a US Navy hood. The level of protection against fire provided by the hoods was assessed using a flame manikin head form as described earlier in these proceedings (Squire et al., 2000). The head incorporates 30 thermocouples distributed over it’s surface and is en- gulfed in flame using 30 propane cup burners that produce a mean incident heat flux on the head of 53 kW·m-2, peaking at 85 kW·m-2 (Squire et al., 2000). During the tests, head “skin” temperatures were recorded every second for sixty seconds from the time of the initial flame exposure. Heat flux incident upon the head was calculated from these data. Predictions of pain and tissue damage could be made from heat flux data using a model developed from human experiments (Stoll & Greene, 1959; Stoll & Chianta, 1971). Further details about the development of the manikin testing method, the method of use and the prediction of burn injury are published earlier in these proceedings (Squire et al., 2000). Predicted burn injuries were described as first (1°), second (2°), and third (3°) degree burns. This classification describes the depth of damage to the skin. In simple terms: 1° degree burn is superficial reddening of the skin, which is painful but does not cause blistering; 2° degree burn damages some of the skin layers with resultant blistering but does not cause scarring; 3° is a full thickness burn, through the skin to the tissues below and results in scarring. 297 The fire-fighting hoods tested are shown in Table 1. Table 1. Description of the hoods tested. Hood Type Identifier No. of layers Hood fabric RN anti-flash RNAF 1 Proban® cotton RN prototype RNP-1 1 50 %/50 % Kermel / FR Viscose RN prototype RNP-2 2 50 %/50 % Kermel / FR Viscose RN prototype RNP-3 3 50 %/50 % Kermel / FR Viscose RN prototype RNP-4 4 50 %/50 % Kermel / FR Viscose RN aircraft-crash RNAC 1 Proban® cotton US Navy USN 1 80 % Rayon / 20 % PBI In each test, the hood was placed on the manikin head and engulfed in flame for speci- fied periods of between one to ten seconds. 2 to 5 seconds being the most likely duration of fuel explosion events and ten seconds being considered the maximum likely possible survival time in full flame exposure (House, 1998). The hoods were normally challenged initially for four seconds so that the duration of subsequent challenges could be deter- mined. If no burns were predicted after a four-second challenge then there was no need to conduct shorter duration tests and likewise if severe burns were predicted there was no requirement to conduct longer duration challenges. At least two examples of the hoods were tested at each flame challenge to ensure reproducibility of the results. Whilst the `proper’ RN fire fighting teams, like almost all civilian fire-fighters, wear a helmet and BA mask on the head, those providing the initial response to the fire on a ship may be wearing only the hood, or the hood and BA mask on their head. Therefore, the protection afforded by the hood should be tested both with and without a BA mask, and a BA mask & helmet. During a pilot study it was shown that in all cases none of the “skin” directly under the helmet or the BA mask recorded a burn injury even after ten seconds of flame challenge. Therefore, it was possible to test the hoods alone and to predict the extra protection afforded if a BA mask or BA mask & helmet had been worn thereby reducing the number of destructive tests required. Results Although 1° burn injuries were predicted they are not reported here. These type of inju- ries, although painful are not serious and heal quickly without permanent injury. All re- sults quoted refer to the total predicted 2° and 3°injuries, which are more serious, rapidly debilitating, often cause permanent injury and are potentially fatal. Figure 1 shows the burn injury when the prototype hoods were challenged with flame for up to 10 seconds. It can be seen that the level of injury was reduced as the number of hood layers increased. It is also apparent that the greatest improvement in protection occurred when increasing the hood layers from 1 to 2, with diminishing improvements in protection as subsequent layers were added. E.g. after 4 seconds of flame engulfment the predicted burn injuries were 78 %, 24 %, 11 % and 8 % with 1-ply, 2-ply, 3-ply and 4-ply hoods respectively. Similar results were seen when the effects of adding the BA mask and the helmet were added to the predictions, although the difference between the hoods was reduced as a smaller surface area of the head was exposed to flame. Figure 2 shows the burn injuries for the current hood (RNAF) in comparison with the RNAC, USN and RNP-2 hoods. It is clear that the RNP-2 hood provided more protection than the current RN AF hood whilst the USN hood provided the best protection. Of interest was the finding that the 298 RNAC hood provided much less protection than the RN AF hood despite being made from exactly the same weight of Proban® treated FR cotton. 0 25 50 75 100 0 2 4 6 8 10 Flame duration (seconds) Pe rc en ta ge b ur ns (% ) RNP-1 RNP-2 RNP-3 RNP-4 Figure 1. Percentage of head with a 2° or 3° burn injuries when the prototype 1 to 4-ply hoods were engulfed in flame. 0 25 50 75 100 0 2 4 6 8 10 Flame duration (seconds) Pe rc en ta ge b ur ns (% ) RNAF RNAC RNP-2 USN Figure 2. Percentage of head with a 2° or 3° burn injuries when the fire hoods were engulfed in flame. Conclusions As expected, the level of protection from fire afforded to the head increases as the num- ber of layers of FR material in a hood is increased. As hypothesised, the majority of the improvements in protection occur when a single layer is increased to a double layer with subsequent improvements being less significant. For the RN, the introduction of a new fire fighting helmet and communications system which included a skull conduction microphone required that the number of layers in the hood be restricted to two. Whilst the RNP-2 hood offered increased protection compared to the current RNAF hood, the USN (2-ply) hood provided the best level of protection; the increased protection is assumed to be due mostly to the material differences between the RNP-2 and the USN navy hoods, as their designs were essentially similar. Consequently the RN is introducing the USN 299 hood to fire fighters as an interim measure whilst the RNP-2 (which fits more easily around the BA mask) is manufactured in the same materials as the USN hood. An interesting finding was that the RNAC hood provided much less protection com- pared to the RNAF hood as both of these were constructed from the same weight of Proban® treated cotton. The difference was due to the RNAC hood having perforations at the ears on each side of the hood, to avoid interfering with hearing and supposedly to al- low fire fighters to `detect the heat of a fire'. No evidence has been found to show that fire hoods interfere with hearing to an extent significant enough to reduce safety. It is further suggested that allowing fire fighters to `detect the heat' by leaving small areas of the head unprotected (usually the ears) when the rest of the body is enclosed in multi- layer FR materials is inappropriate. If fire fighters have to detect dangerous levels of heat, it is suggested that this could be achieved easily, and more safely, by using cheap temperature sensitive indicating strips that are readily available. References House JR (1998) Balancing the fire and flash protective clothing needs of Royal Navy personnel against the debilitating effects of heat strain. The Institute of Marine Engineers', Fourth International Naval Engineering Conference and Exhibition, HMS Sultan 15-17 April 1998 Proceedings Part II, 21-29. Stoll AM & Greene LC (1959) Relationship between pain and tissue damage due to thermal radiation. Journal of Applied Physiology 14(3), 373-382. Stoll AM & Chianta MA (1971) Heat transfer through fabrics as related to thermal injury. Transactions New York Academy of Sciences 649-670. Squire JD (2000) Development of the Research and Technology Group flammability manikin systems. Proceedings of the 1st European Conference on Personal Protective Equipment, Stockholm, Sweden May 7-10. 300 Fire fighters’ views on ergonomic properties of their footwear Helena Mäkinen1, Susanna Mäki1, José S. Solaz2, Dave J. Stewardson3 1 Finnish Institute of Occupational Health, Finland 2 Institute of Biomechanics of Valencia, Spain 3 Industrial Statistics Research Unit, University of Newcastle upon Tyne, England Introduction The foot is a complex support structure formed by 26 bones and has the highest con- centration of bone structure in the human body. The bones of foot have to admit con- tinuous variations in the load. Ideally all shoes should meet the following ergonomic re- quirements; fit properly both in length and width, grip the heel firmly, there should be a straight line from the heel to the end of the big toe to prevent turning of the big toe, and there should be enough room for the toes. There should also be some way of fastening across the top of the foot to prevent the shoe slipping on and off while walking (Steemson 1988). Clumsiness, heaviness and poor contact sensitivity were the most fre- quent complaints about safety boots among firefighters (Louhevaara, 1995; Mäkinen, 1991; Marr, 1990). Biomechanical aspects of footwear have been studied, a.o., from the point of view of poor fitting (Audemars, 1978; Oakley, 1984; De Moya, 1982), needs for shock absorbency (Audemars, 1978) and unhindered mobility and ease of movement (Oakley, 1984). In laboratory experiments leather boots were found to have advantages with regard to biomechanical aspects (Neevees et al., 1989). Hawes et al. compared dif- ferences in forefoot shape and dimensions between North American and Japanese or Korean males, and concluded that the shape of these two groups differ. Thus different shoe lasts for each population are required for optimal shoe comfort (Hawes et al., 1994). Ergonomics of PPE is one essential requirement of PPE directive 89/686/EEC. So far some ergonomic requirement only exists in a few harmonised EN standards. A research project on fire fighters’ protective footwear supported by The EU Commission started last year. Part of this project aims to clarify fire fighters' perceptions of needs in their footwear. The aim of this paper is to identify fire fighters' perceptions of the ergonomic and biomechanical properties of their footwear when engaged in different tasks. Methods An inquiry was made with selected fire brigades in six European countries, Austria, Britain, Finland, France, Spain and Sweden. Fire fighters opinions on the ergonomic and biomechanical properties of their footwear, in use, were asked. The questions concerned the following general properties: restrictions caused by the footwear during various ac- tivities, discomfort caused by the footwear. Biomechanical properties investigated were: sizing and fitting problems, weight, flexibility and stability, and the thermal comfort of footwear. Some of the questions allowed yes/no answers, while others required a degree of agreement to various statements using the following scale: strongly disagree, disagree, not sure, agree, strongly agree. The questionnaire also had two open questions where the 301 fire fighters were asked to specify what boot properties have been sensitive to wear and ageing effects, and what are thought to be the causes if the fit was not good. Analysis of variance (ANOVA) was used to analyse the data. Some data manipulations were necessary to make different responses comparable and to give them similar struc- ture. Means comparisons and General Linear Models were used for testing differences. Results and discussion Answers from 754 fire fighters were received. More leather boots and shoes than rubber boots were in use in Finland (61 %), Sweden (55 %), in Spain (86 %), and in France (100 %). It might have been useful if the answers had been classified according to leather or rubber material. This was not done because many fire fighters have both types in use simultaneously. Safety footwear has caused some restrictions in fire fighters work e.g. 58 % of Finnish fire fighters, 52 % of Swedish, 32 % of French, 31 % of English and 27 % of Spanish fire fighters. There were differences between countries in activities where footwear caused restrictions (Figure 1). 0 5 10 15 20 25 30 35 Stepping (up/down) stairs Driving Running Walking Walking on slippery surface Jumping Climbing Ladders Crouching Kneeling Stepping in and out of fire vehicles Other Pe rc en ta ge Finland UK Spain Sweden France Austria Figure 1. The comparison between countries of the restrictions caused by the footwear. 36 % of Finnish fire fighters, 11 % of Swedish, 33 % of English, 34 % of Spanish, 14 % French, and 24 % of Austrian fire fighters reported some discomfort caused by foot- wear. The cause was reported as the material in 32-50 % of answers, and wrong size or design in 34-67 % of answers. 57 % (N=80) of Finnish firemen, 12 % (N=41) of English, 33 % (N=80) of Spanish, 23 % (N=35) of Swedish, 80 % of French (N=21), and 39 % of Austrian firemen had ob- served some loss of protection due to wear and ageing of your footwear. The reported causes were due to upper or lining material, outsoles, seams, toe cap, inserts, water proofing and heat. 15-29 % of fire fighters in different countries reported about some problems obtaining correct footwear sizes. There were differences between countries in obtaining the correct sizes. In Finland and Sweden the main reason was the height of the metatarsal area, in Britain the width and in France and Spain the length of the footwear. 302 0 10 20 30 40 50 60 70 80 Width Height of toe area Length Height of metatarsal area Girth Pe rc en ta ge Finland UK Spain Sweden France Austria Figure 2. The problems with sizing of the footwear by answered fire fighters in different countries. The areas, where the fire fighters had problems obtaining the correct size of boots, are shown in Figure 2. These results indicate that there are differences in different areas of Europe; north or south. 17 % of the Finnish fire fighters, 28 % of English, 27 % of Spanish, 21 % of Swedish, 46 % of French, and 14 % of Austrian fire fighters disagreed that their footwear has good fitting properties. Swedish and Finnish fire fighters in particular reported that. The main causes were due to material, size, width, instep, general fitting, loose fitting, fastenings etc. The answers concerning biomechanical properties are summarized in Table 1. More Spanish fire fighters complained than others, Austrian fire fighters complained less than fire fighters in other countries. Feet sweating and weight and flexibility of footwear were the most important biomechanical properties in Spain. The flexibility of footwear was more important in Finland than in other countries. For French fire fighters foot stability was more important than in other countries, but feet sweating and flexibility of footwear were not as important for them. Sweating was the biggest concern in all countries except France. Ankle flexion was of little importance in most countries, being more clearly of lower importance in Spain and Austria than in the other countries. Table 1. Fire fighters' opinions about the problems in biomechanical properties of their foot- wear, significance was determined, in relation to sample size, by country or property. Country Weight % Flexibility % Ankle flexion % Foot stability % Sweating % Finland (N=102) 35 37* 27 20= 62" Britain(N=47) 21 34* 23 25 57" Spain (N=132) 61*" 36* 27" 23" 79*" Sweden (N=69) 25 12 10 23 49" Austria (N=180) 22 10¯ 8¯ 18¯ 38¯" France (N=150) 28 7¯ 19 60* 15¯ * This biomechanical property is significantly more important in this country than in other countries ¯ This biomechanical property is significantly less important in this country than in other countries " This biomechanical property is significantly more important, than other properties, in this country = This biomechanical property is significantly less important, than other properties, in this country There is less than a 5% chance that any of the significant differences indicated are not true. 303 Conclusions In general there were differences between general and biomechanical properties with bio- mechanical being the most important. 27-58 % of fire fighters though that their footwear had caused some restrictions in their work specially when driving, running, walking on slippery surfaces and jumping. 11-36 % of fire fighters think that footwear causes some discomfort in their work. Only about 27 % of fire fighters reported that they had ever had problems with boot size. There were differences between the countries with regard to footwear measurements, the width and girth of the footwear being the most important. Further research is needed to explain the reasons why fire fighters in different countries have different types of sizing problems; is the reason different foot anatomy, or the use of different footwear design methods in different countries? References Audemars P (1978). Underfoot cushioning in working footwear - A question of comfort. Protection. 15 (9) 11-14. De Moya RG (1982). A biomechanical comparison of the running shoe and the combat boot. Military Medicine. (147), 380- 383. Hawes MR, Sovak D, Miyashita M, Kang S-J, Yoshihuku Y and Tanaka S (1994) Ethnic differences in forefoot shape and the determination of shoe comfort. Ergonomics (37), 187-196. Louhevaara V, Griefahn B, Künemund C, Ilmarinen R, Mäkinen H (1995) Maximal physical work performance with European standard based fire-protective clothing system and equipment in relation to individual characteristics. Eur. J. Appl. Physiol. (71) 223-229. Marr S (1990) Firefighters' safety footwear: an investigation into problems reported by firefighters. J Occup Health Safety - Aust NZ 4, 297-301. Mäkinen H (1991) Analysis of problems in the protection of fire fighters by personal protective equipment and clothing - development of a new turnout suit. Tampere University of Technology/Finnish Institute of Occupational Health, Helsinki 1991, 201 pp. Oakley E, Howard N (1984) The design and function of military footwear: a review following experiences in the South Atlantic. Ergonomics. 28 (6) 631-637. Neeves RE, Barlow DA, Richards JG, Provost-Craig M & Castagno P (1989) Physiological and biomechanical changes in fire fighters due to boot design modifications. Sports Science Center, University of Delaware, Newark, Delaware, August 1989, 38 pp. Steemson J (1988) Foot work. Occupational Safety and Health. (March) 12-18. 304 Participant list Abeysekera John Afanasieva Rallema Alfano Gaetano Luleå University of Technology Res.Inst.of Labour Med. Russian Acad.of Med Sc DETEC, University of Napoli Div. of Industrial Ergonomics Budennogo prospekt 31 Piazzale Tecchio, 80 971 87 Luleå 105 275 Moscow 80125 Napoli Sweden Russia Italy Phone 46-920-91407 7-095-366 07 92 39-81-7682298 Fax 46-920-91030 7-095-366 05 83 39-81-2390364, 81-7682323 John.Abeysekera@arb.luth.se 257366-10@mtu-net.ru alfanoga@unina.it Andersen Johnny Anttonen Hannu Appelblom Marja Danish Standards Association Oulu Regional Institute of Occupational Health Försvarets Materielverk Kollegievej 6 Aapistie 1 FMV.KC Gem DK 2920 Charlottenlund 90220 Oulu 115 88 Stockholm Denmark Finland Sweden Phone 45-39 96 61 01 358-8-5276086 46-8-7826935 Fax 45-39 96 61 02 358-8-5276000 46-8-7826167 ja@ds.dk hannu.anttonen@occuphealth.fi maapp@fmv.se Asserfors Mikael Bartels Volkmar T. Bartkowiak Grazyna W.L. Gore & Associates Scandinavia AB Forschungsinstitut Hohenstein Central Institute for Labour Protection Neongatan 9, Box 268 Schloss Hohenstein 48 Wierzbowa St. 431 23 Mölndal D - 74357 Bönnigheim 90-133 Lodz Sweden Germany Poland Phone 46-31-7067842 49-7143271-611 48-42 6781963 Fax 46-31-864226 49-7143271-8781 48-42 6781915 asserfors@wlgore.com v.bartels@hohenstein.de grbar@ciop.lodz.pl Bentley Mark Bessonova Nina Busmundrud Odd Centre for Human Sciences, DERA Res.Inst.of Labour Med. Russian Acad.of Med Sc FFI Norw. Def. Res.Establ. Div. for Protection Building F 138 Budennogo prospekt 31 P O Box 25 Farnborough GU14 0LX 105275 Moscow NO-2027 Kjeller UK Russia Norway Phone 44-1252394402 7-095-366 07 92 47 - 63 80 78 90 Fax 44-1252393789 7-095-366 05 83 47 - 63 80 75 09 mbentley@dera.gov.uk bessonov@gazeta.ru odd.busmundrud@ffi.no Carlsson Dan Carlsson Birgitta Coudrez Michel Swedish Rescue Services Agency National Board of Occup. Safety and Health Lainière de Picardie Dept Proline BP 89 651 80 Karlstad 171 84 Solna 80202 Peronne Cedex Sweden Sweden France Phone 46-54-104213 46-8-730 9483 33-3-22838361 Fax 46-54-103344 46-8-730 1967 33-3-22838385 dan.carlsson@kd.srv.se birgitta.carlsson@arbsky.se mcoudrez@chargeurs- interlining.com 305 Curtis Nicholas J. Danielsson Ulf den Hartog Emiel Lion Apparel Inc. FOA 5 Thermal Physiology Group, TNO Human Factors 6450 Poe Ave PO Box 13576 P.O. Box 23 Dayton, Ohio 172 90 Stockholm 3769 ZG Soesterberg USA Sweden Netherlands Phone 937-898-1949 46-8-7063214 31-346 356 288 Fax 937-898-9204 46-8-7063309 31-346 353 977 nickc@lionapparel.com udanielsson@sto.foa.se denhartog@tm.tno.nl Drinkmann Margret Eichinger Helmut Fatah Alim A. Sympatex Technologies Du Pont de Nemours International S.A. Nat.Inst.of Standards & Technology Kasinostrasse 19 - 21 2, Chemin du Pavillon 100 Bureau Drive, Stop nr 8102 42103 Wuppertal CH 1218 Le Grand Saconnex, Geneva Gaithersburg, MD 20899-8102 Germany Switzerland USA Phone 49-202-323228 41-22-7175886 301-975-2757 Fax 49-202-322488 41-22-7176164 301-948-0978 margret.drinkmann@sympatex.de helmut.eichinger@che.dupont.com alim.fatah@nist.gov Færevik Hilde Foss Nina Frodsham Dave SINTEF Unimed, Dept of Health and Work phys. Police procurement service W.L. Gore & Associates Ltd Postbox 124 Blindern Pbox 112 Kirkton Campus Livingston 0314 Oslo 2714 Jaren EH54 7BH Livingston Norway Norway Scotland Phone 47-225-89415 47-61336710 44-1506-678213 Fax 47-220-67909 47-61336711 44-7867-906261 Hilde.Ferevik@Unimed.Sintef.no polmatj@online.no frodsham@wlgore.com Gavhed Désirée Geng Qiu Qing Giedraityté Lina National Inst. for Working Life National Inst. for Working Life Luleå University of Technology Respiratory Health and Climate Respiratory Health and Climate Dept of Human Work Sci 112 79 Stockholm 112 79 Stockholm 971 87 Luleå Sweden Sweden Sweden Phone 46-8-730 9200 46-8-730 9056 46-920-910 000, 920 723 27 Fax 46-8-730 9226 46-8-730 9226 46-920-910 30 desiree.gavhed@niwl.se qiuqing.geng@niwl.se lina.giedraityte@arb.luth.se Gooderson Chris Y. Gonzalez Jose A. Grime Lars A. FRDG Room 401, Horseferry House Dept of Clothing and Textiles, Univ of Manitoba Police procurement service Dean Ryle Street 35 Chancellor´s Circle P.O. Box 112 SW1P 2AW London Winnipeg, MB R3T 2N2 2714 Jaren UK Canada Norway Phone 44-171-2178367 204-474-8065 47-61 33 67 10 Fax 44-171-2178254 204-474-7592 47-61 33 67 11 Chris.Gooderson@ homeoffice.gsi.gov.uk gonzale0@ms.umanitoba.CA polmatj@online.no 306 Haase Jürgen Hanson Margaret Havenith George Sächsisches Textilforschungsinst. e.V. Inst. of Occup. Medicine Human Thermal Env. Lab. Loughborough Uni Annaberger Str 240 8 Roxburgh Place Dept. of Human Sciences, D-09125 Chemnitz Edinburgh EH8 9SU LE11 3TU Germany UK UK Phone 49-371 5274 171 44-131- 6675131 44-1509-223031 Fax 49-371 5274 153 44-131- 6670136 44-1509-223940 haase@stfi.de Margaret.Hanson@IOMHQ.org.uk G.Havenith@lboro.ac.uk Heffels Peter Heinry Didier Hinz Torsten Zentrum fur Sicherheitstechnik (Z S) Paul Boye Inst. of Biosyst Eng Fed Agricult Res Centre Klinkerweg 4 Site Carnot - Tecnosud Bundesallee 50 406 99 Erkrath 66100 Perpignan 38116 Braunschweig Germany France Germany Phone 49-2104.3101 33-29-4-68-556863 49-531-596-317/456 Fax 49-2104.31 33-29-4-68-556850 49-531-596-369 ZS-Erkrath@T-Online.de pbtech@mnet.fr torsten.hinz@fal.de Holmér Ingvar House James R. Howie Robin M. National Inst. for Working Life Institute of Naval Medicine Robin Howie Associates Respiratory Health and Climate Crescent Road, Alverstoke 12 Morningside Road 112 79 Stockholm P012 2 DL Gosport GB 699 8542 Edinburgh Sweden UK Scotland Phone 46-8-730 9288 44-23 9276 8032 44-131 447 5161 Fax 46-8-730 9226 44-23 9250 4823 44-131 447 5161 ingvar.holmer@niwl.se yj72@dial.pipex.com robin.howie@btclick.com Hu Junyan Hyun Ki-Ja Hyttinen Pekka ST 713, Inst. of Textiles and Clothing Environmental Health, Nara Women´s University Leijona-Pro Oy The Hong Kong Polytechnic University Kitawuoyanishimachi Kaartilantie 12 Kowloon Hong Kong Nara 57200 Savonlinna China Japan Finland Phone 852-2766-6509 81-630-8509 358-15 259 621 Fax 852-2773-1432 81-742203469 358-15 259 620 tcjxhu@polyu.edu.hk wan.htun@cc.nara-wu.ac.jp pekka.hyttinen@leijona-pro.fi Ilmarinen Raija Jacobsen Peter Karkkula Sanna Finnish Inst. of Occup. Health Hälleflundra gatan 12 Finnish Inst. of Occup. Health Laajaniityntie 1 Laajaniityntie 1 FIN-01620 Vantaa 426 79 Västra Frölunda FIN-01620 Vantaa Finland Sweden Finland Phone 358-9-47472660 46-31- 69 80 60 358-9-47472659 Fax 358-9-47472805 46-31- 69 82 22 358-9-47472805 Raija.Ilmarinen@occuphealth.fi peter.jacobsen@albright-wilson.se Sanna.Karkkula@occuphealth.fi 307 Karlsson Eva Konarska Maria Korhonen Eero National Inst. for Working Life Centralny Instytut Ochrony Pracy Finnish Inst. of Occup. Health Respiratory Health and Climate ul. Czerniakowska 16 Laajaniityntie 1 112 79 Stockholm 00-701 Warsaw FIN-01620 Vantaa Sweden Poland Finland Phone 46-8-730 9626 48-226233283 358-40-5453821 Fax 46-8-730 9226 48-226233695 358-9-2947627 eva.karlsson@niwl.se makon@ciop.pl eero.korhonen@occuphealth.fi Koskinen Hanna Kuklane Kalev Lara Jaime Finnish Inst. of Occup. Health Dept of physics National Inst. for Working Life IRSST Laajaniityntie 1 Respiratory Health and Climate 505 de Maisonneuve, Blvd. West FIN-01620 Vantaa 112 79 Stockholm Montreal, H3A 3C2 Quebec Finland Sweden Canada Phone 358-9-47472686 46-8-730 9248 514-288-1551 Fax 358-9-47472805 46-8-730 9226 514-288-9399 Hanna.Koskinen@occuphealth.fi kkuklane@niwl.se lara.jaime@irsst.qc.ca Li Yi Lindqvist Ann-Mari Linné Anders Inst. of Textiles and Clothing National Inst. for Working Life Norrbotten Regiment Hong Kong Polytechnic Univ. Box 9105 Hung hom Kowloon Hong Kong 112 79 Stockholm 961 00 Boden China Sweden Sweden Phone 852-2766-6479 46-8-730 97 60 46-921 - 69000 Fax 852-2773-1432 46-8-730 98 97 46-921 - 59366 tcliyi@polyu.edu.hk aml@niwl.se anders.linne@mekb19.mil.se Magnusson Per-Henrik Manek Thomas Markussen Drude Snickers Europe AB AUVA HUB Norvegian University of Science and Technology Box 989 Adalbert Stifter Str 65 191 29 Sollentuna A 1200 Wien NO-7491 Trondheim Sweden Austria Norway Phone 46-8-92 51 37 43-1 331 11 587 47-739 43383 Fax 46-8-96 87 30 43-1 331 11 901 47-735 91005 Ph.magnusson@snickers.se thomas.manek@auva.sozvers.at drudem@stud.ntnu.no Marszalek Anna McCullough Elizabeth A. McKeown David Central Institute for Labour Protection Institute for Environmental Research SATRA House ul. Czerniakowska 16 Kansas State University, 64 Seaton Hall Rockingham Road Kettering 00-701 Warsaw Manhattan, KS 66506-2907 Northamptonshire NN16 9JH Poland USA UK Phone 48-22-6233280 785 532-2284 44-1536410000 Fax 48-22-6233695 785 532-6642 44-1536520093 anma@ciop.pl liz@ksume.me.ksu.edu davem@satra.co.uk 308 Meinander Harriet Mellgren Christer Morrison Iain VTT Chemical Technology Melltex AB Rhodia Consumer Spec Ltd PO Box 1402 P.O.Box 7870 FIN-33101 Tampere 112 49 Stockholm B69 4 Hx Oldbury, West Midlands Finland Sweden UK Phone 358-3-3163550 46-8-6461006, 070-7207370 44-121 541 3366 Fax 358-3-3163517 46-8-6545262 44-121 541 3442 Harriet.Meinander@vtt.fi melltex@swipnet.se iain.morrison@eu.rhodia.com Mustonen Sanna Mäkelä Esa Mäkinen Helena Finnish Inst. of Occupational Health Finlayson Forssa Oy Finnish Inst. of Occup. Health Dept of physics Laajaniityntie 1 P O Box 43 Laajaniityntie 1 FIN-01620 Vantaa FIN-30101 Forssa FIN-01620 Vantaa Finland Finland Finland Phone 358-9-47472678 358-3 4123250, 358-50 3785804 358-9-474 72 764 Fax 358-9-47472805 358-3 4123444 358-9-474 72 115 sanna.mustonen@occuphealth.fi esa.makela@finlayson.fsp.fi Helena.Makinen@occuphealth.fi Nicholson Annemarie Niinimäki Keijo Nilsson Håkan O. WLGore & Associates Ltd Leijona-Pro Oy National Inst. for Working Life Kirkton Campus, Livingston Kaartilantie 12 Respiratory Health and Climate EH54 7 BH West Lothian 57200 Savonlinna 112 79 Stockholm Scotland Finland Sweden Phone 44-1506 460123 358-15 259621 46-8-730 9789 Fax 44-1506 420004 358-15 259620 46-8-730 9226 annemarie_nicholson@wlgore.com keijo.niinimaki@leijona-pro.fi hakan@niwl.se Nocker Wolfgang Nordh Birgitta Nyström Ulf W.L. Gore & Associates GmbH BNB Agenturer Trelleborg Protective Products AB Hermann-Oberth-Str 24 Kaptensgatan 12, 3 tr P.O.Box 1520 D-85640 Putzbrunn 114 57 Stockholm 271 00 Ystad Germany Sweden Sweden Phone 49-89-4612-2621 46- 8- 232940 46-411-67956 Fax 49-89-4612-2610 46- 8- 663 29 82 46-411-15285 wnocker@wlgore.com birgitta@bnbagenturer.se ulf.nystroem@trelleborg.com Parsons Ken C. Persson Louise Poschenreithner Gerhart Human Thermal Env. Lab. Loughborough Uni 3M Svenska AB AUVA HUB Loughborough Gullbergs strandgata 6 Ad. Stifterstrasse 65 Leics LE11 3TU 411 04 Göteborg A 1200 Wien UK Sweden Austria Phone 44-1509 223023 46-31-631049, 070-673 13 31 43-1-33111 522 Fax 44-1509 223940 46-31-631010 43-1-33111 901 K.C.Parsons@lboro.ac.uk lpersson@mmm.com gerhart.poschenreitner@ auva.sozvers.at 309 Pylsy Auli Påsche Arvid Reinertsen Randi E. TEVASTA SINTEF Unimed SINTEF Unimed P.O. Box 50 Div. Health and Work Physiology Div. Health and Work Physiology 332 11 Tampere N - 7465 Trondheim N - 7465 Trondheim Finland Norway Norway Phone 358-3-388-9127 47-739 68 511 47-735 96 989 Fax 358-3-388-9120 47-739 68 514 47-735-91 005 auli.pylsy@tevasta.ttliitot.fi Arvid.J.Pasche@unimed.sintef.no randi.e.reinertsen@ unimed.sintef.no Reinikka Leila Risikko Tanja Rossi René Opastinsilta 12 B, PL 73 Finnish Inst. of Occup. Health EMPA St Gallen Aapistie 1 Lerchenfeldstrasse 5 00521 Helsinki FIN-90220 Oulu CH - 9014 St. Gallen Finland Finland Switzerland Phone 358-204 442 232 358-8-5276127 41-71-274-7765 Fax 358-204 442 449 358-8-5276121 41-71-274-7762 leila.reinikka@tieh.fi tanja.risikko@occuphealth.fi rene.rossi@empa.ch Rydholm Jonas Sahlgaard Åke Sawada Shin-ichi Taiga AB W.L.Gore & Associates Scand. AB National Institute of Industrial Health Box 20 Box 268 21-1 Nagao 6 chome, Tama-ku Varberg 431 23 Mölndal Kawasaki, Kanagawa 214-8585 Sweden Sweden Japan Phone 46-340-666900 46-31-7067843 81-44-865-6111 Fax 46-340-666922 46-31-31864226 81-44-865-6116 jonas.rydholm@taiga.se sahlgaard@wlgore.com sawada@niih.go.jp Scott Richard A. Shaw Anugrah Shishoo Roshan Defence Clothing & Textiles Agency University of Maryland Eastern Shore IFP Research AB Flagstaff Road Richard Henson Center Room 2113 Box 104 Colchester, Essex CO2 7 SS Princess Anne, MD 21853 431 22 Mölndal UK USA Sweden Phone 44-1206-782353 410-651-6064 46-31-706 6300 Fax 44-1206-782337 410-651-6207 46-31-706 6334 rascott@dcta.demon.co.uk ashaw@mail.umes.edu roshan.shishoo@ifp.se Sjöblom Mårten Skjäl Lundgren Jennifer Skoglund Kjell-Åke Ma-Ru OY Aerotech Telub AB Fristads AB Selininkatu 5 FIN-33240 Tampere 581 82 Linköping 513 80 Fristad Finland Sweden Sweden Phone 358-3-2237500 46-13-231233 46-33-202209 Fax 358-3-2238390 46-13-231788 46-33-200280 ma-ru@yritys.tpo.fi jennifer.skjal- lundgren@aerotechtelub.se kjell-ake.skoglund@fristads.se 310 Skullman Johan Soltynski Krzysztof Squire James D. Swedish Armed Forces, Combat School Centralny Instytut Ochrony Pracy Defence Clothing & Textile Agency Box 120 ul. Czerniakowska 16 Flagstaff Road 590 30 Borensberg 00-701 Warsaw CO2 7SS Colchester, ESSEX Sweden Poland UK Phone 46-13-19 52 13 48-226233276 44-1206 782358 Fax 48-226233695 44-1206 782337 serefenix@hotmail.com krsol@ciop.pl jsquire@dcta.demon.co.uk Stare Ann Stewardson Dave J. Stirling Mandy IFP Research AB ISRU MMME Stephenson Building Leicestershire Fire and Rescue Headquarters Box 104 University of Newcastle upon Tyne Anstey Frith, Leicester Road, Glenfield 431 22 Mölndal Tyne & Wear GB - NE1 7RU LE3 8 HD Leicester Sweden UK UK Phone 46-31-7066300 44-1912226244 44-116-232-4481 Fax 46-31-7066363 44-1912227365 44-116-231-1180 stare@ifp.se D.J.Stewardson@ncl.ac.uk stirloin_steaks@hotmail.com Stull Jeffrey O. Sullivan Janette Tietäväinen Riitta International Personnel Protection, Inc. SSL International Research & Development The Polytechnic of Southwest Finland 10907 Wareham Lane 205 Cambridge Science Park Vehmaantie Austin, Texas 78739 CB4 0GZ Cambridge FIN 23100 Mynämäki USA UK Finland Phone 512 288 8272 44-1223 423232 358-2-430 6211 Fax 512 288 8272 44-1223 423310 358-2-430 5232 intlperpro@aol.com janette.sullivan@ligplc.com riitta.tietavainen@lsktol.fi Tokura Hiromi Toma Doina Träff Mona Dept of Environmental Health The National R-D institute for Textile and Leather Aerotech Telub AB Nara Women´s University Str.Lucretiu Patrascanu nr 16:3 630-8506 Nara Bucuresti 746 74 581 82 Linköping Japan Romania Sweden Phone 81-742-20 -3469 40-1-3404928 46-13-23 15 96 Fax 81-742-20 -3469 40-1-3405515 46-13-23 17 88 tokura@cc.nara-wu.ac.jp centex@ns.centex.ro mona.traff@aerotechtelub.se Wallén Anna Wickman Bengt Wigardt Lisskulla Försvarets Materielverk CB Gedda AB Försvarets Materielverk FMV : KC Flyg Box 19009 FMV : KC Gem 115 88 Stockholm 40012 Göteborg 115 88 Stockholm Sweden Sweden Sweden Phone 46-8-7824507 46-31- 81 30 10 46-8-7 825 817 Fax 46-8-7825988 46-31- 20 62 79 46-8-7 826 167 anwal@fmv.se bengt.wickman@cbgedda.se Liwig@fmv.se 311 Wiik Ulla Vilander Seija Withey W. R. OY PD-Service Ltd Finnish National Road Administration Headquarters Centre for Human Services, DERA P.O.Box 101 P.O. Box 73 FIN-67101 Kokkola FIN-00521 Helsinki Farnborough GU14 0LX Finland Finland UK Phone 358-6-8319636, 358-400-562842 358-204 44 21465 44-1252-393535 Fax 358-6-8319687 358-204 44 2804 44-1252-392097 ulla.wiik@pd-service.kpnet.com seija.vilander@tieh.fi von der Bank Nicola Zimmerli Traugott Zimmermann Dorit BIA, Berufsgenossenschaftliches Institut fuer Arbeitssicherheit EMPA St Gallen Kommission Arbeitsschutz und Normung.KAN Alte Heerstr. 111 Lerchenfeldstrasse 5 Alte Heerstr 111 D 53754 St. Augustin CH-9014 St. Gallen 53757 Sankt Augustin Germany Switzerland Germany Phone 49-224 123 12508 41-71-2747878 49-2241-231-3474 Fax 49-224 123 12234 41-71-2747150 49-2241-231-3464 nicola.von.der.Bank@hvbg.de traugott.zimmerli@empa.ch Zimmermann@kan.de Øglænd Gro Ellen Österlund Lennart Norvegian University of Science and Technology Stockholm Fire Dept NO-7491 Trondheim 111 83 Stockholm Norway Sweden Phone 47-73-598908 46-8-4548775 Fax 47-9-591005 46-8-4548903 groog@stud.ntnu.no brand@brand.stockholm.se 312 Author index A Abeysekera, J............................................. 62, 67 Afanasieva, R. F. ........................................... 188 Anttonen, H. .................................... 17, 179, 212 B Bahima, J. ...................................................... 141 Bartels, V. T. ................................................... 53 Bartkowiak, G.................................................. 98 Bentley, M. ...................................................... 21 Bessonova, N. A. ........................................... 188 Blazejczyk, K. ............................................... 192 Bouskill, L. M.................................................. 21 Burmistrov, V. M........................................... 188 Burmistrova, O. V. ........................................ 188 C Cáceres, I. ...................................................... 141 Carpus, E. .................................. 88, 94, 114, 216 Chan, J. F. L. ................................................... 30 Cohea, I. .......................................................... 88 Cohen, E. ............................................... 141, 154 Coleman, S. Y................................................ 273 D Danielsson, U. ............................................... 285 den Hartog, E. A. ........................................... 281 Desnoyers, J. E. ............................................. 230 Dhir, A........................................................... 119 Douglass, J..................................................... 273 Dowson, M. ..................................................... 30 F Færevik, H. ............................................ 245, 267 G Gavhed, D............ 67, 71, 75, 167, 171, 175, 289 Geng, Q. Q..................................................... 208 Gonzalez, J. A................................................ 119 Graveling, R. ................................................. 277 Gross,R. ........................................................... 12 H Haase, J.......................................................... 123 Hanson, M. ............................................ 159, 277 Hassi, J............................................................. 44 Havenith, G................................................ 21, 26 Heffels, P. ...................................................... 196 Helistén, P. .................................................... 235 Heus, R. ........................................................... 26 Hinz, T................................................... 154, 239 Hocke, M. ...................................................... 293 Hoernicke, E. ................................................. 239 Holmér, I. ..... 17, 67, 71, 75, 167, 171, 175, 179, ........................................... 188, 192, 208, 289 House, J. R. ....................................................296 Howie, R. .......................................................163 Hu, J. ..............................................................102 I Ilmarinen, R. ..........................................235, 263 K Karkkula, S. ...................................................150 Karlsson, E.................................67, 71, 208, 289 Kausch, F. ..................................................12, 48 King, M. W. ...................................................119 Klein, N............................................................41 Koivistoinen, K. .............................................235 Konarska, M.............................................38, 183 Korhonen, E. ..................................................135 Koskinen, H. ..................................................263 Kuklane, K. ..67, 71, 75, 167, 171, 175, 188, 289 L Lara, J. ...........................................145, 218, 230 Leray, H. ........................................................285 Li, H. ..............................................................226 Li, L. ..............................................................226 Li, Y. ..................................................8, 102, 255 Lindholm, H. ..................................................235 Liu, J. .............................................................226 Luan, Z...........................................................226 Luo, X. N. ..................................................8, 255 Luo, Z. X....................................................8, 255 M Markussen, D. ................................................267 Marszalek, A. ...................................................38 Massé, S. ........................................................145 McCullough, E. A. ...........................................90 Meinander, H. ................................................204 Mihai, C. ........................................................114 Milici, M. .......................................................114 Mustonen, S. ..........................................127, 131 Mäki, S...........................................................300 Mäkinen, H. ...........................127, 131, 150, 300 Mäkinen, T. M. ................................................44 N Newton, E. .................................................8, 255 Nieminen, K. ..................................................131 Nilsson, H. O. ..................................17, 179, 289 Nocker, W......................................................293 P Parkin, N. .........................................................30 Parsons, K. .......................................................34 Peel, M.............................................................30 Perron, G........................................................230 Pietikäinen, P. ................................................212 313 Pyryt, J........................................................... 183 Påsche, A. ...................................... 252, 262, 268 R Raivo, S. ........................................................ 150 Rajamäki, E. .................................................. 150 Reinertsen, R. E. .................................... 259, 267 Rintamäki, H.................................................. 212 Risikko, T. ....................................................... 44 Rissanen, S. ................................................... 212 Rossi, R. .................................................... 12, 48 S Sawada, S. ....................................................... 57 Scarlat, N. ...................................................... 114 Scott, R. A. .................................................... 108 Shaw, A. ........................................................ 154 Shim, H............................................................ 90 Shishoo, R........................................................ 79 Smolander, J. ........................................... 38, 289 Sobolewski, A.......................................... 38, 183 Solaz, J. S. ..................................................... 300 Soltynski, K. ............................................ 38, 183 Squire, J. D. ........................................... 200, 296 Staples, R....................................................... 296 Stewardson, D. J. ............................. 30, 273, 300 Stirling, M...................................................... 269 Strauss, L. ......................................................293 Stull, J. O. ......................................................222 T Tellier, C. .......................................................218 Toivonen, L......................................................44 Tokura, H. ................................................25, 107 Toma, D. ....................................88, 94, 114, 216 U Umbach, K.-H. .................................................53 W Weder, M. ..................................................12, 48 Visileanu, E..............................................94, 114 Withey, W. R. ..................................................21 Vogel, C. ........................................................123 Z Zavadsky, S....................................................243 Ziegenfuß, B. ...................................................41 Zimmerli, T. .......................................................1 Zimmermann, D. ............................................137 Ø Øglænd, G. E. ................................................267