Quantitative evaluation of air gap thickness and contact area between body and garment

Abstract

The thermal, evaporative and wicking properties of clothing depend not only on the properties of the fabric but also on the thickness of air layers and the magnitude of the contact area and their variation. The aim of this study was to accurately determine the contact area and the air gap thickness between clothing and the human body in detail.

These parameters were measured for a range of typical patterns of garments (tight- and loose fitting) covering either the upper or lower body and made of various types of fabrics (knitted and woven). The method consisted of imposing three-dimensional scans of the nude and dressed manikin and determining the distance between their surfaces by advanced three-dimensional scan post-processing.

Due to this method the distribution of the air gap thickness and the contact area over body parts was obtained and this knowledge can be applied in models of heat and mass transfer in the clothing.

Keywords

air gap contact area 3D body scanning heat and mass transfer in clothing

The heat and mass transfer within the clothing system is a composition of a number of physical processes, such as: dry heat exchange (conduction, convection and radiation), evaporation and condensation, sorption, as well as vapor and liquid water transfer.¹ In addition, factors associated with construction and use of a garment, such as air penetration and compression by wind, body posture and movement and clothing fit, influence these processes significantly.^2–4 That is mainly due to the changing size and shape of the layers of air trapped between the skin and clothing, between clothing layers alone, and in the layer adjacent to the outer surface of clothing. Thus, the thermal, evaporative and wicking properties of clothing depend not only on the properties of the fabric used for the garment but also on the magnitude and the temporal change of the contact area and air layer thickness.

The garment is a three-dimensional (3D) form created from the two-dimensional (2D) pattern on the flat fabric to cover complex geometry of the human body. This fact together with the fabric properties entails draping and sagging of the garment. Effectively, the thickness of the air layer is inhomogeneous and varies over body parts and changes with body shape, posture and movement. The inhomogeneous thickness of the air layers within the clothing system influences the local heat and vapor exchange. The example in Figure 1 illustrates the significance of the air gap thickness for thermal and evaporative resistances of an ensemble calculated according to the model published by Wissler and Havenith.⁵ The air trapped beneath garments and adjacent to the outer layer provides the bulk of both the thermal and evaporative resistances. Moreover, it increases noticeably with only a small change of 5 mm in the air gap thickness (Figure 1). If the air layer is larger than about 8 mm, internal convection will take place.⁶ Therefore, a determination of the extent and location of air layers larger than 8 mm allows the analysis of possible air circulation. Furthermore, when the surfaces of the fabrics and/or the surface of the skin stay in contact, direct exchange of the liquid water can take place.^7–8 Due to the wicking effect, this liquid is distributed over a larger area and, hence, enhances additionally the inhomogeneous heat and mass transfer.⁹

Figure 1.

Thermal (b) and evaporative (c) resistances of the layers in a clothing system⁵ consisting of a cotton fabric (227 g/m², 1 mm thick) separated by an air gap as in scheme (a) for environmental conditions as follows: T_skin = 34°C, RH_skin = 99%, T_ambient = 10°C, RH_ambient = 81%.

On the other hand, existing mathematical clothing models assume a uniform air gap between the body and fabric layers or its full contact.^10–13 Such simplification facilitates the computation process. However, this approach disregards the non-uniform heat, vapor and liquid water transfer, which depend on the presence of contact between surfaces⁹ and on the shape of the air layers trapped within clothing.¹⁴

Several studies have attempted to characterize the air layers in garments.^15–24 In these investigations only an average air gap thickness has been estimated for the whole garment from either the air volume trapped within the garment^{15–16,18,23} or the ratio between clothing and body surface area.¹⁷ The distribution of the air layer thickness was also investigated using a 3D body scanning technique, which combined with the 3D post-processing software offered a high-precision, non-invasive and fast method to digitalize and analyze the spatial form of the dressed body. To obtain the size of the air gap the 3D scans of the nude and dressed body were superimposed and the air gap thickness was calculated from either a selected number of points or from a discrete number of cross sections through the dressed body.^19–22,24 However, none of the devised methods allowed the systematic, accurate and detailed evaluation of the local and average air gap thickness nor did they address the issue of contact area in ensembles.

In this study, we propose a method to accurately determine the air gap thickness and the contact area between clothing and the human body through an advanced analysis of 3D body scans of the nude and dressed body of a male manikin. This method allowed more accurate measurement of the air gap thickness and the contact area than other existing methods. Additionally, both parameters could be obtained for individual body parts. Consequently, this method will contribute to a more realistic evaluation of heat and mass exchange rates through clothing systems and provide more accurate input for ergonomic and comfort design of clothing.

Methods

3D scanner

A 3D body scanner VITUS XXL (Human Solutions GmbH, Germany) that can generate highly precise 3D images of the human body according to ISO20685:2005²⁵ was used for this study. Its operating principle is based on optical triangulation, which is currently the most accurate method for touchless 3D imaging. The measurement uncertainty of the 3D scanner was tested using a cylindrical tube (diameter 110 mm, height 2100 mm, measured at a constant temperature within the range 15–30°C) and was less than 1 mm of average maximal girth error. The single scanning measurement was taken with a point density of 27 points/cm².

Scanning principle

The 3D scanner was used according to the guidelines of the producer, i.e. it was used at a temperature range between 20 and 30°C and calibrated at the beginning of every measurement day and after each 3–4 hours of scanning.

A motionless male manikin with a height of 189 cm, a girth of 97.5 cm at the chest and 74 cm at the waist (measured according to ISO 7250-1:2008) was used in this study. It was equipped with additional locks against turning of body parts and a construction supporting its arms and feet in a fixed position to ensure identical body positioning for all measurements. The manikin was scanned nude and dressed with sample garments using the 3D scanner. To assess the reliability of the proposed technique, the manikin was re-dressed and scanned six times for each garment to allow for random changes in garment drape to analyze the repeatability of the measurements. After putting on a garment all unusual positions of the garment, such as twisting or clinging, were removed and the garment was allowed to rest on the manikin for several minutes before scanning (at least 5 minutes).

Garments

Garments used in this study represented a selection of typical casual clothing covering the upper and lower body. The garments were made of either woven (shirt and trousers) or knitted fabrics (T-shirt and shorts) to obtain a range of different material and fabric properties in order to verify the validity of this method for a variety of clothing (i.e. tight- and loose-fitting clothing). The garments circumferences were measured and compared to the respective manikin’s circumference measurements in selected areas to determine the ease allowances which validated the use of these garments as providing varying degrees of tightness (T-shirt and shorts) or looseness (shirt and trousers) (Table 1 and Figures 3 and 4). All garments were of appropriate size for the manikin and their sizes provided by manufacturers for the European market are given in Table 1. Before the measurements the garments were washed and dried according to ISO6330:2000²⁶ and ironed.

Figure 2.

Division of the manikin body into the individual body parts for which the air gap thickness and the contact area were measured.

Figure 3.

Photographs and corresponding post-processed exemplary single 3D scans (300dpi) indicating the contact area and the air gap thickness of the shirt and the T-shirt.

Figure 4.

Photographs and corresponding post-processed exemplary single 3D scans (300dpi) indicating the contact area and the air gap thickness of the shorts and the trousers.

Table 1.

Garment characteristics including the eases determined at body landmarks according to ISO7250-1.²⁷

Garment type	Garment size	Weave/knit	Fiber content***	Mass	Thickness	Ease at chest*	Ease at waist*	Ease at thigh*	Ease at calf*
Garment type	Garment size	Weave/knit	Fiber content***	g/m²	mm	cm	cm	cm	cm
Shirt	M39/40	plain	100CO	124.7	0.28	10.5	31.0	–	–
Trousers	47	diagonal	60WO/38PES/2EL	202.2	0.42	–	3.0**	10.0	12.5
T-shirt	M	bird’s eye	100PES	156.3	0.72	0.0	14.0	–	–
Shorts	M	bird’s eye	100PES	156.3	0.72	–	0.0**	0.0	–

Notes: *ease allowance was determined as the difference between circumferences of the manikin and the garment at the respective body landmarks; **determined at the waistline of the garment; ***CO stands for cotton, WO for wool, PES for polyester and EL for elastan respectively.

3D scan post-processing

The 3D scans were post-processed using the dedicated software Geomagic Qualify 11 (Geomagic, USA) to obtain the required parameters such as the contact area and the air gap thickness. The post-processing procedure was challenging due to complexity of 3D forms counterbalanced by available computing power. Furthermore, a high precision aligning of the 3D scans despite surface deficiencies and repeatable slicing of the irregular shape of the dressed body into the body parts were also demanding. Consequently, the majority of manipulations of the 3D scans had to be done manually. Generally, the post-processing procedure consisted of the following steps:

Step 1: cleaning of the 3D scan surface by removing scanning artifacts and closing surfaces with deficiencies;

Step 2: super-imposing of 3D scans of nude and dressed manikin in the 3D space using uncovered body parts as the reference shapes;

Step 3: slicing super-imposed manikins into body parts, for which the parameters were sought (Figure 2);

Step 4: computation of the contact area and the distribution of the air gap thickness for each body part;

Step 5: scaling and processing of color maps;

Step 6: processing and statistical analysis of the numerical data.

The division of the manikin body into body parts (Figure 2) corresponded to the boundaries of body coverage for typical garments (neck, wrists, ankles). Additional divisions were made in the areas where the body shape changes and usually induces a different draping pattern of the garments (chest, elbows and knees).

Derivation of air gap thickness and contact area

The air gap thickness was determined in the post-processing phase and was defined as the average distance between points on the surface of nude and dressed manikin. Such averages were determined for each body part (Figure 2) separately. This method also provided an opportunity to determine the surface areas of the manikin that stayed in contact with the garment. To describe this parameter we defined the contact area CA between the garment and the skin as a ratio coefficient

CA = \frac{A_{contact}}{A_{covered}}

(1)

where A_contact is the skin area in direct contact with a garment and A_covered is the skin area covered by this garment. A_covered usually corresponded to the area of the defined body part (Figure 2). However, for the T-shirt and the shorts covering only parts of upper arms and thighs, only the covered area of the manikin surface was considered as A_covered.

The surface area of the body in contact with garment was evaluated based on the distance between the surfaces of the nude and dressed manikin. Theoretically, if the distance between the surface of the dressed manikin and the surface of the nude manikin is equal to the thickness of the fabric (T) used to make the garment, the air gap thickness beneath the garment is zero and it can be classified as contact between skin and garment. However, uncertainties of the individual phases of the evaluation were in the order of magnitude of the fabric thickness and had to be accounted for. These included the inaccuracy of the 3D scans (S) given by the manufacturer of the 3D scanner and the average alignment error (A) resulting from imperfect imposing of scans using uncovered body parts as the reference shapes (Table 2). The average alignment error was provided by the software for the uncovered body parts used as reference. The sum of measurement uncertainties was derived using rules of uncertainty propagation²⁸ (square root of the sum of squares of single uncertainties, such as for the scan of nude manikin S, for the dressed manikin S and for the alignment of these two A). These lumped measurement uncertainties were added to the fabric thickness of the garment (Table 2) and we defined the sum as the distance between the skin and the garment recognized as contact area (T_CA in Table 2).

Table 2.

Inaccuracy of the 3D scans, average alignment error, and the distance between skin and garment recognized as the contact area for studied garments

Garment type	Inaccuracy of the 3D scans, S*	Average alignment error, A**	Distance recognized as contact area, T_CA***
Garment type	mm	mm	mm
Shirt	1.0	0.7	1.9
T-Shirt	1.0	0.6	2.2
Trousers	1.0	0.3	1.9
Shorts	1.0	0.5	2.2

Notes: *value given by manufacturer; **value provided by post-processing software; ***value calculated from errors²⁶ and fabric thickness as described in text.

Validation of the method

Although the 3D scanning technique is the most suitable for investigation of the air layers in clothing, it is prone to various inaccuracies. For example, 3D scanners are usually calibrated based on flat grids or simple forms (such as cylinders) with known dimensions and are assumed to provide the same accuracy when scanning complex forms such as the human body. The studies reported in the literature which involved 3D scanning of clothing did not address this issue and were typically less detailed (fabric thickness not considered, air gap thickness determined manually at discrete number of points).

To prove validity of our procedure a manual analysis of the contact area was conducted for a few chosen garments. The tight-fitting garments, i.e. T-shirt and shorts, were used for the validation of the post-processing method as they were more prone to contact area variation with the change of T_CA. A test in which the contact area between the manikin surface and a semi-transparent tight-fitting garment was drawn on the garment with a soft pencil showed identical patterns and sizes of contact area as those obtained from the post-processed 3D scans. Furthermore, the actual dressed manikin and its photographs were compared visually with the post-processed 3D scans. Numerous characteristic pleats and folds of similar shape and size were found in the post-processed 3D scans and in the photographs (see regions of armholes and chest in Figure 3 and buttocks and thighs in Figure 4). Thus, we concluded that both validation tests indicated the developed post-processing 3D scans method to be both reliable and accurate.

Results

The photographs and post-processed exemplary single 3D scans of garments covering either the upper (shirt and T-shirt) or the lower body (shorts and trousers) are shown in Figures 3 and 4. The distinction between the contact area and the air gap is indicated on the color scale in these figures.

Figure 5 presents the average result of six 3D scans for air gap thickness and its standard deviation for each body part obtained for the studied garments and Figure 6 shows the average result of six 3D scans for contact area and its standard deviation.

Figure 5.

Mean air gap thickness (and standard deviation) of the studied garments covering the upper and lower body obtained for six scanning repetitions and determined for each body part.

Figure 6.

Mean contact area (and standard deviation) of the studied garments covering the upper and lower body obtained for six scanning repetitions and determined for each body part.

The average air gap thickness in loose-fitting garments was typically 40% larger than in the tight-fitting garments. The largest differences were observed for sections of abdomen, lower back, lumbus, and anterior and posterior pelvis (Figure 5) where the loose clothing is usually not tailored to follow the curvature of the body. Interestingly, the air gap thickness in tight-fitting clothing was almost constant for all body parts considered and was included in the range between 6 and 9 mm. Its value exceptionally reached over 30 mm in the lumbus and lower back region due to their clearly concaved form (Figure 5). The air gap in the shirt showed a larger variability over body parts, however, with the same trend of higher values in the lower back and lumbus region. The non-uniform draping of the shirt in the abdomen region was typical for all repetitions and plausibly resulted from the non-symmetrical design of the shirt (the pocket and the overlaps with button holes and buttons on different front sides). For the trousers, the air gap for the shins was always higher than that for the calves.

Discussion

The method to accurately determine the air gap thickness and the contact area between clothing and the human body using an advanced analysis of 3D body scans has been successfully developed. Additionally, both parameters were measured for individual body parts covered by tight- and loose-fitting garments made of various types of fabrics (knitted and woven). Such selection of garments gave a range of air gap thickness and contact area found in modern casual garments, which was necessary for the validation of the method for different clothing patterns.

As expected, differences in the contact area for tight- and loose-fitting garments were observed (Figures 3 and 4). These differences resulted from distinct interactions between the complex shape of the human body and the properties of fabric and garment design. Interestingly, the variability of the contact area amongst six scanning repetitions of randomly draped garments was low. Typically it was approximately 4% (in a range of 1–14%) for tight-fitting garments and 2% (in a range of 1–8%) for loose-fitting garments. Similarly, the low variability of air gap thickness was observed and typically it was approximately 0.7 mm (in a range of 0.2–3.4 mm) for tight-fitting garments and 1.6 mm (in a range of 0.2–3.7 mm) for loose-fitting garments. Effectively, such variability can lead to less than 1% change in the thermal resistance of the air gap beneath the garment.⁵ The observed repeatability of the contact area and the air gap thickness measurement for the setup used in this study (standing stationary position) may allow a reduction of the number of repetitions in further measurements.

Many clothing models are coupled with the models of the human thermal physiology.^10–13 Such a configuration requires individual treatment of each body part in terms of number and properties of the covering layers including the thickness of the air layers. Some mathematical models assume full contact of the underwear with the skin;^5,13 whereas the examples used in this study showed only 4–95% of contact between skin and fabric (Figure 6) with particularly low contact in the lumbus region. The same models also assume that there is no contact between surfaces if there is an air gap between the skin and clothing layers. However, the gravity and the irregular shape of the human body causes, for loose clothing, the observed contact area to be in a range of 1–28%. The contact area in this case originated from the protruding body regions and from random pleats that touch the skin.

In the model of Lotens and Havenith,¹⁰ this problem was partially addressed by mathematical derivation of the thickness of air layers based on the girth of the nude and dressed body part at the same body landmark (fast and affordable method). In reality, however, our results show that the air gap is most probably smaller as the clothing drapes and sags instead of forming a regular cylinder around a body part (see the picture of the shirt in Figure 3). As a consequence, the contact area between the body and the sunken garment part can be created which can promote wicking in clothing layers. Thus, the contact area between the skin and the garment should be included as an additional input parameter in models to enhance simulation of the liquid water transfer in clothing.

The friction interaction between garments and the skin adds on to the draping of clothing;²⁹ however, it is not certain how the difference in friction coefficient between the plastic manikin used in this study and human skin could affect the garment draping.

Conclusions

This study evaluates an advanced post-processing of 3D scanning to measure the air gap thickness and contact area in clothing layers. In this study the thickness of the fabric was considered and the validity of the developed method was evaluated by direct comparison of the post-processing outcome with the actual draping patterns of the garments. The advantage of using this technique over previous techniques is two-fold: (1) the proposed method measures the contact area and the air gap thickness quantitatively and accurately for different parts of the body, and (2) it is relatively quick (less than 6 h per garment scan) and affordable due to the technological advances in 3D body scanning. Moreover, this method allows the distribution of the air gap thickness and the contact area over several body parts to be measured in detail for both tight- and loose-fitting garments. The information provided by this new method can be applied to models that estimate heat and mass transfer in clothing, and as a consequence, enhance simulations of the interaction between a garment and human thermal physiology, comfort and thermal sensation. Further studies are required that investigate a larger selection of clothing ensembles, fabrics, body shapes and body postures in order to provide a better understanding of the governing principles of the air gap thickness and contact area in clothing.

Footnotes

Acknowledgements

The authors wish to thank armasuisse for helping to fund this work and providing the 3D scanner. Our thanks go to the members of the workshop at Empa for their prompt support and to Dr. Veronika Meyer from Laboratory for Protection and Physiology at Empa for her editorial input.

Funding

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

References

Whitaker

. Coupled transport in multiphase systems: A theory of drying. Adv Heat Transfer 1998; 31: 1–104.

Holmer

Nilsson

Havenith

. Clothing convective heat exchange - Proposal for improved prediction in standards and models. Ann Occup Hyg 1999; 43: 329–337.

Parsons

Havenith

Holmer

. The effects of wind and human movement on the heat and vapour transfer properties of clothing. Ann Occup Hyg 1999; 43: 347–352.

Bouskill

Livingston

Parsons . The effect of external air speed on the clothing ventilation index. In: Hanson

(ed.) Contemporary ergonomics 1998. London: Taylor & Francis Ltd, 1998, pp. 540–544.

Wissler

Havenith

. A simple theoretical model of heat and moisture transport in multi-layer garments in cool ambient air. Eur J Appl Physiol 2009; 105: 797–808.

Spencer-Smith

. The physical basis of clothing comfort, Part 2: Heat transfer through dry clothing assemblies. Clothing Res J 1977; 5: 3–17.

Keiser

Becker

Rossi

. Moisture transport and absorption in multilayer protective clothing fabrics. Textile Res J 2008; 78: 604–613.

Weder

Rossi

Chaigneau

. Evaporative cooling and heat transfer in functional underwear. Int J Clothing Sci Technol 2008; 20: 68–78.

Umeno

Hokoi

Takada

. Prediction of skin and clothing temperatures under thermal transient considering moisture accumulation in clothing. ASHRAE Trans 2001; 107: 71–81.

10.

Lotens

Havenith

. Calculation of clothing insulation and vapor resistance. Ergonomics 1991; 34: 233–254.

11.

Holcombe

. Mathematical simulation of heat and moisture transfer in a human-clothing-environment system. Textile Res J 1998; 68: 389–397.

12.

Fan

Luo

. Heat and moisture transfer with sorption and condensation in porous clothing assemblies and numerical simulation. Int J Heat Mass Transf 2000; 43: 2989–3000.

13.

Berger

Sari

. A new dynamic clothing model. Part 1: Heat and mass transfers. Int J Therm Sci 2000; 39: 673–683.

14.

Bouskill

Havenith

Kuklane

Parsons

Withey

. Relationship between clothing ventilation and thermal insulation. Aihaj 2002; 63: 262–268.

15.

Daanen

Hatcher

Havenith

. Determination of clothing microclimate volume. In: Tochihara

Ohnaka

(eds) Environmental Ergonomics - The Ergonomics of Human Comfort, Health and Performance in the Thermal Environment 2005; 361–365.

16.

Havenith

Zhang

Hatcher

Daanen

. Comparison of two tracer gas dilution methods for the determination of clothing ventilation and of vapour resistance. Ergonomics 2010; 53: 548–558.

17.

Kakitsuba

. Investigation into clothing area factors for tight and loose fitting clothing in three different body positions. J Human-Environ Syst 2004; 7: 75–81.

18.

Lee

Hong

. 3D quantification of microclimate volume in layered clothing for the prediction of clothing insulation. Appl Ergonom 2007; 38: 349–355.

19.

Song

. Clothing air gap layers and thermal protective performance in single layer garment. J Indust Text 2007; 36: 193–205.

20.

Wang

Newton

. Ease distribution in relation to the X-line style jacket. Part 1: Development of a mathematical model. J Text Instit 2006; 97: 247–256.

21.

Xu JH and Zhang WB. The Vacant Distance Ease Relation between Body and Garment. ICIC2009: Second International Conference on Information and Computing Science, Vol 4, Proceedings – Modelling and Simulation in Engineering. H. G. Bao, Y. Jiang and C. Grecos, 2009, pp. 38–41.

22.

Zhang ZH and Li J. Relationship between Garment Fit and Thermal Comfort. Textile Bioengineering and Informatics Symposium Proceedings, Vols 1–3. Y. Li, Y. P. Qui, X. N. Luo and J. S. Li, 2010, pp. 1317–1321.

23.

Bouskill

Sheldon

Parsons

. The effect of clothing fit on the clothing ventilation index. In: Hanson

(ed.) Contemporary ergonomics 1998. London: Taylor & Francis Ltd, 1998, pp. 510–514.

24.

Kim

Lee

Corner

. Investigation of air gaps entrapped in protective clothing systems. FireMater 2002; 26: 121–126.

25.

ISO20685: 2005. 3-D scanning methodologies for internationally compatible anthropometric databases.

26.

ISO6330: 2000. Textiles - Domestic washing and drying procedures for textile testing.

27.

ISO7250-1: 2008. Basic human body measurements for technological design. Part 1: Body measurement definitions and landmarks.

28.

Field

. Discovering statistics using SPSS. London: Sage Publications Ltd, 2005.

29.

Luible

. Study of mechanical properties in the simulation of 3D garments. Geneva: University of Geneva, 2008.