Indian Journal of Fibre & Textile Research

Vol. 30, September 2005, pp. 252-257

A theoretical model to predict the thermal resistance of plain woven fabrics

D Bhattacharjee & V K Kothari"
Department of Textile Technology, Indian Institute of Technology, New Delhi 1 1 0 0 16, India

Received 8 October 2004; accepted 3 December 2004

A single cell woven fabric model has been developed to predict the thermal resistance in a non-convective environ­
ment. The values from the model have been compared with those obtained from Alambeta in which the fabric is confined
between two plates at a pressure of 200 Pa, thus having the negligible natural convection and no forced convection. The
model gives good correlation with the experimental values.
Keywords: Alambeta, Conduction, Cotton, Thermal resistance, Woven fabric
IPC Code: Int. CI. 7 D06H3/00

1 Introduction searchers, depending upon various approaches'>.

Heat and mass transfer through textiles affect their
comfort characteristics. Most of the work done in this
field is mainly related to (i) understanding different
modes of heat transfer; (ii) study of different fabric
parameters which are i mportant for effective fabric
insulation; (iii) use of new materials which facilitate
the use of garments for extreme as well as moderate
climatic conditions; and (iv) development of new
methods and mathematical models to determine ther­
mal properties of textiles.
Many researchers 1 -4 have tried to obtain the contri­
bution of various modes of heat transfer for deter­
mining the thermal insulation of fabric. Most of them
have purported that the heat transfer through textiles
takes place by conduction, convection and radiation.
The consideration of each mode is different from one
researcher to another. A few have combined two dif­
ferent modes of heat transfer and considered them as
one2 .4 . A number of workers 5-7 have stated that the
most important fabric factor which determines the
thermal insulation of the fabric is fabric thickness; the
magnitude of the thickness depends upon the pressure
applied while carrying out the measurements. B lack
and Mathew 8 have investigated the plain woven fab­
rics with different thread densities and found that the
thermal insulation decreases with increasing thread
density. To build a mathematical model, the fabric is
considered as a porous structure. The effective con­
ductivity for packed beds was predicted by many re-
To whom all the correspondence should be addressed.
U
ChanglO has given a theoretical model based on com­
bined series and parallel conduction for the effective
thermal conductivity of wire mesh. A unit cell model
was employed to simulate the mesh and the solid-fluid
lumped volumes were taken to predict the effective
thermal conductivity. Luikov et al. " have considered
an elementary cell to be a system of resistances and
by using the analogy of electrical system they pre­
dicted the effective conductivity of the porous system.
In all the cases, the fabric has been considered either
as a porous layered sheet or a fibrous batting. In this
study, an attempt has been made to find out a theo­
retical prediction of thermal resistance for plain
woven fabrics in a non-convective environment.
2 Materials and Methods
2.1 Theoretical Model
Thermal transfer from the skin to environment
through the fabric takes place individually by con­
duction, convection and radiation. The total heat
transfer of the fabric is the algebraic sum of the
amounts of heat transfer through each of these modes,
which is given by the following equation :
Q tolaI =Q con d uctlon
' , + Q r�d'mllon
+ Q convection ,
... (I)
If the fabric is kept i n a confined environment un­
der a fixed pressure, the effects of forced convection
can be neglected. The air inside the fabric can be as­
sumed to be another insulating material with thermal
conductivity and density values same as that of air.
Therefore, the heat transfer through conduction and

Phone: 2659140 1 ; Fax :+9 1 - 1 1 -26581 1 03 ;

E-mail: kothari @textile.iitd.ernet.in radiation only is considered.
 B HATIACHARJEE & KOTHARI: THERMAL RESISTANCE OF PLAIN WOVEN FABRICS 253

2.1.1 Heat Transfer Through Plain Woven Fabric Structure

Consider a single cell of the plain woven fabric
(Fig. 1). When observed closely, it can be found that . . . (4)
there are mainly three areas from which conduction
takes place: (i) through warp and weft yarn; (ii)
through interlacements; and (iii) through air pores.
The basic equation of conductivity as per the Fou­
rier's Law is given below: . . . (5)
dQ aT
= -kA . . . (2)
dr ax

where dQ is the quantity of heat conducted in time dr; where 1 , 2 suffixes are meant for warp and weft re­
T, the temperature; x, the thickness of the wall per­ spectively except for T, and T2 which mean hot and
pendicular to the direction of heat; A, the area of the cold plate temperatures ; p, the thread spacing; a, the
heated surface; and k, the coefficient of thermal con­ flattened diameter; b , the flattened thickness; e, the
duction. weaving angle; and t, the thickness of fabric.
When Eq. (2) is integrated over a time 't, then un­
Conduction through interlacement
der the steady state conditions
(a,a2 )(I; - T2 )
(
I; - T
�nduclion = kA --2
t
) . . . (3)
. . . (6)

where T, and T2 are the hot and cold temperatures re­

spectively; and t, the thickness of the material. Equa­
tion (3) can be used to calculate the total heat transfer
due to conduction.
It can be observed from Fig. 2 that the interlace­
ment regions consist of warp and weft yarns as well as
t
air. They can be depicted in terms of series resis­
tances. The yarns are also flattened (Fig. 3) so a single
plain woven cell can be assumed to have a warp, a
weft, an interlacement and an air gap. B ased on
Eq. (3), the conduction through various sources is

tances between yarns and air [1 - warp/weft + air, and 2 - warp +

given by the following equations : Fig. 2 - Transverse view of the fabric seen as system of resis­

weft + air]

Warp

Air Gap

Weft

Fig. 1 - Single cell plain woven fabric model [ I -conduction

through warp and weft yarns, 2-conduction through interlacement
and 3-conduction through air pore; R, -radiation from skin through
air pore, R2-radiation from the area occupied by yarns; and Fig. 3 - Top view of the elementary cell of the fabric from which
Rrradiation heat exchange between yarns and skin] heat transfer takes place
254 INDIAN 1. FIBRE TEXT. RES., SEPTEMBER 2005

Conduction through air pore

kajr [(PI - al )(P2 - a2 )] ... (11)
Q,ond.pore = ' " (7)
t
Total heat transfer due to conduction
Qcond.tot = Qcond.warp + Qco�d.weft + Qeond.!n + Qeond.pore Therefore, the heat transfers due to radiation
through the plain woven cell are given as:

[( ]
Conduction per unit area Through airpore
a [a1 ( P2 - a2 ) + Pl� ] (r. - T2 /
Qcond.tot (8)
Qconduetion - ---
_

Qrad.yarn -

J
( PIP2 ) _

'" ( 2)
1 - Eskin
Hence, the resistance of the fabric due to conduction
only is given by the following equation: ESkin

Rconduction =
(r. T2 )
-
. . . (9)
Qconduction
. . . ( 1 3)
2.1.2 Radiative HeatTransfer
If the same single cell model of fabric is consid­
ered, there are three ways in which beat transfer due
to radiation takes place: (a) from skin to environment
through air pores; (b) from yams to environment; and
(c) between yams and skin (Fig. 1 ).
The basic equation of radiation heat transfer be­ . . . (4)
tween two plates is
a(T; - T/
Qradiation = . . . . ( 0) Total heat transfer due to radiation
R.
Qrad.total = Qrad.pore + Qrad.yarn + Qrad.skyarn
lJ
OS)
where Rij is the total resistance encountered during net
radiation transfer through two surfaces. This value has Total heattransfer due to radiation per unit area
following two components:
Qrad.totaf
)
(

(i) surface resistance (Rsurface) which depends upon Qradiation -

_
. . . ( 6)
the emissivity and the area of the surface, as ( PIP2
given by Rsurface = ( 1 - Ei ) / f\ ; and Hence, the resistance of the fabric due to radiation
(ii) space resistance (Rs ace) which depends upon the
p only is given by the following equation:
view factor of the two surfaces, as given by
(7; - Tz )
Rspaee = 1 / f\ F;j . Rradiation = . . . (7)
Qradiation
where Gi is the emissivity of the surface i; Ai, the area
of the surface; and Fij, the view factor between sur­ If the modes of heat transfer are made analogous to a
faces i and j. parallel circuit system, then the overall resistance of
In case of radiation heat transfer directly to the at­ the fabric is given by the following relationship:

RtolJJ.l = Rconduetion r l + (Rradiation rl

[( J-1
mosphere, the environment can be assumed to be an
( 1 8)
all-absorbing black surface whose temperature is
equal to the ambient temperature 12 . In that case, only
Rsurface is relevant. Also, the view factor in case of par­ 2.1.3 Effective Yam Diameter
allel planes is 1 . When single yams are woven into fabric, the yarn
Considering all these factors, Eq. ( 1 0) now be­ during interlacement tends to flatten. The woven fab­
comes rics were observed under microscope with a magnifi-
 BHATTACHARJEE & KOTHARI: THERMAL R ESISTANCE OF PLAIN WOVEN FABRICS 255
,

.�.
cation of x 40. It can be observed in Figs 4a and 4b '.� \ t:
that the yarns do not have the same flattened diameter
throughout the fabric. The diameter is more at the , .�
-4

crown of the interlacement and less in the rest of the , i\ �t;

body. It has been assumed that the effective diameter
in this case is the average of flattened diameter and 1f.
t
�,
normal diameter, as given below: �.

= d [�]
ya
m 2e
. . . ( 1 9)

where

e= -- = --
dyam byam
. . . (20)
Fig. 4 - Difference in the diameters of warp yams through their
ayam dyam
course in the fabric [a - 2-ply warp, and b - single ply warp]
where deff is the effective diameter of the yarn; dyam,
the theoretical diameter of the yarn; ayam, the flattened
diameter; byam, the flattened thickness; and e, the de­
gree of flattening.
In case of two-ply yarns, the resultant diameter, as
per convention, is equal to 1 .44d where d is the di­
ameter of individual yarn. When a yarn is woven into
a dense fabric, the yarns try to push each other to ac­
commodate more yarns per unit space. This leads to
rolling of the individual yarns on top of each other in
a two-ply yarn. The diameter of the yarn then be­
comes equal to the diameter of individual yarns.
Hence, Eq. ( 1 8) is followed in case of two-ply yarns
as well except that the value of dyam is equal to the
individual ply diameter.

2.1.4 Effective Thermal Conductivity of Yarn Fig. 5 - Solid-fluid matrix of a twisted yam as a system of series
From Fig. 5, it can be assumed that the solid-fluid resistances
matrix in the yarn complies with the system of series
arrangement. In that case, the effective conductivity is weft linear densities, thread densities, thickness and
given bl mass per unit area, are given in Table 1 . These fabrics
� were tested for thermal resistance on Alambeta. A line
kf diagram of the Alambeta instrument is given in Fig 6.
. . . (2 1 ) The fabric is kept between hot and cold plates; the hot
plate comes in contact with the fabric sample at a

..
where ke i s the effective conductivity; k the conduc­
tivity of the solid; kf' the conductivity of the fluid; and
¢J , the porosity of the yarn.
2.2 Experimental
Twelve plain woven fabrics of different construc­
tions were taken. The fabric parameters, like warp and
pressure of 200Pa. As soon as the hot plate touches
the surface of the fabric, the amount of heat flow from
the hot surface to the cold surface through the fabric
is detected by heat flux sensors 1 3. There is also a sen­
sor to measure the thickness of the fabric. These val­
ues are then used to calculate the thermal resistance of
the fabric. This instrument also gives the transient
heat transfer as well as Qrnax values .
256 INDIAN J, FIBRE TEXT. RES., SEPTEMBER 2005

3 Results and Discussion model were compared with the resistance values ob-
3.1 Comparison of Theoretical Model with Experimental
tained from A lambeta (Table 2). The average devia-
Values tion between Alambeta and the theoretical model is
The values obtained from the single cell fabric 7 .92% with a minimum and maximum deviation of
1 .45 % and 14.77% respectively . It can be seen from
Table I -Specifications of plain woven cotton fabric samples Fig. 7 that the R2 value is high which indicates a
strong correlation between the two values.
Sample Warp Weft Endsl Picksl Thick- Fabric In this case, convective heat transfer is not taken
No. count count inch inch ness weight
Ne Ne mm g/m
2 into consideration as the two plates are separated only
by the thickness of the fabric which is of the order of
1 39.0 2/28 1 42 56 0.32 173 1 mm and therefore, there is negligible natural con-
2 38.0 30.0 140 80 0.26 150 vection. Furthermore, as there was no open air space
3 39.0 38.0 140 94 0.22 143 above the fabric sample, forced convection could also
4 36.0 2170 1 44 92 0.24 144 be ruled out.
5 37.0 1 9.0 1 36 68 0.3 1 161 It can be seen from Fig. 8 that the thermal resis-
6 37.0 20.0 140 66 0.3
tance values of some samples obtained by the experi-
1 66
mental data are lower than those obtained from the
7 38.0 36.0 1 32 70 0.25 1 26
model. These are mainly fabrics with 2-ply warp
yarns. This is because the dyam in Eq. ( 1 8) is obtained
8 2/36 2/60 1 22 56 0.39 191
9 2/38 28.0 120 56 0.37 1 92
from theoretical individual ply diameter. But when
10 2/38 2/40 1 18 50 0.4 1 197
these fabrics are observed under the microscope, the
II 2/38 20.0 1 14 50 0.41 1 97 diameters are much larger than deff at various places.
12 2/38 2/30 1 16 42 0.43 213 Hence, the conductive heat losses are much higher

Switch 1 -+- 1 :-----

Processing and
Filter unit

1
I Display
I 1
I Diagnostics Output to
computer
� nt
J L
I
·cb
Temp.
r �
1 � 0�
Controller
f- Motor

en CD--
_I 1
�

J
J 61 1 H J+- Fulcrum

I Heat flow
Optical
sensor
sensor
Top plate Heater
Bottom Th eaded
plate Specimen shaft
Temp senso r

Fig. 6 - Line diagram of Alambeta

 BHATIACHARJEE & KOTHARI: THERMAL RESISTANCE OF PLAIN WOVEN FABRICS 257

0.012 -,-------,
Table 2-Errors between the thermal resistance values obtained
from Alambeta and the theoretical model

. .. .
. .. . . . ..
0,01 1

Sample Thermal resistance, K.m2/W , .

{ 0,01 . . .•.
E
No. Alambeta Theoretical Deviation"

'7
i
..
model model %

k
0,009

I 0.0088 0.0093 5.68

0.0083 0.0075 9.64
� 0,008

I
2
i
3 0.0072 0.0062 1 3.89
.'
0,007

.ct
rr:

iI '
4 0.0069 0.007 1 .45
5 0.0094 0.009 4.26 I- 0,006

6 0.009 1 0.0087 4.40

0,005

. . . .. . . Model
7 0.0088 0.0075 14.77 -+- Alam beta

8 0.0095 0.0 1 06 1 1 .58

9 0.0098 0.0 1 02 4.08 2 4 5 6 7 8 9 10 11 12

10 0.0098 0.0 1 1 12.24 Sample

II 0.0 1 02 0.0 1 1 7.84 Fig. 8 - Comparison of thermal resistances obtained from the
12 0.01 14 0.0 1 2 5.26 theoretical model and Alambeta

"Average, 7.92; min, 1 .45; and max, 14.77 References

I Rees W H, Physical factors determining comfort perform­
0.0135 -,------, ance, Proceedings, The Third Shirley Conference on Com­
y = 1 .3682x · 0,0033
fort, edited by W H Rees (Shirley Institute, Manchester,
R' = 0,8558
0.0125 UK), 1 97 1 .
�
�" clothing. J Text Illsl TrailS, 34 ( 1 943) T4 1 -T59.
, .
2 Baxter S, and Cassie A B D, Thermal insulating properties of

:;;:
0,01 1 5
#
. .

thermal insulation, J Appl Phys, 38( 12) ( 1 967) 4663-4668.

, 3 Hager N E & Steere R C, Radiant heat transfer in fibrous

I
•

�
0,0105 ,

'�
sulation, Text Res J, 53 ( 1 983) 7 1 7-725,
0,0095
4 Farnworth B, Mechanism of heat flow through clothing in­

.
1
. ,

5 Morris G J, Thermal properties of textile materials, J Text

cr:
.
'
0.008S #

F= Illst, 44 ( 1 953) T449-T476.

,
0,0075 • • 6 Hoge H J & Fonseca G F. The thermal conductivity of a

.I...- #:...-
. ..I. .. --'---....-
.. -....-
... -....- � tions, Text Res J, 34 ( 1 964) 40 1 -4 1 0.
• # multilayer sample under a variety of experimental condi­
0, 0065
.. -

Thermal Resistance (Model) of fabrics, Text Res J, 35(2) ( 1 965) 95- 103.
0,0065 0,0075 0.0085 0,0095 0,0105 0,01 1 5 0,0125
7 Morris M A, Thermal insulation of single and mUltiple layers

in relation to clothing: Part 3 - Heat insulation by fabrics

Fig. 7 - Correlation between the values obtained from Alambeta 8 Black C P & Mathew J A, The physical properties of fabrics

used as body clothing, J Text Inst Trans, 25 (1 934) T249-

and the theoretical model

than the theoretical approxi mation, because of the T276.

increased surface area. Therefore, the thermal resis­ 9 Kaviany M, Prillciples of Heat Transfer in Porous Media.
n
2 d edn (Springer - Verlag, New York), 1 995.
tance in these cases is less.
10 Chang W S, Porosity and effective thermal conductivity of
wire screens, ASME J Heat Trans. 1 1 2(5) ( / 990) 5-9.
4 Conclusions

Thermal conductivity of porous systems, Int J Heat Mass

I I Luikov A V, Shashkov A G , Vasiliev L L & Fraiman Y U E,
The single cell plain woven model gives a good
prediction of the thermal resistance of plain woven Transfer, I I ( 1 968) 1 1 7- 1 40,
fabrics. The correlation obtained with experimental 12 Jaluria Y & Torrance K E, Computational Heat Transfer, 2nd
values is found to be high. Convective heat transfer edn (Taylor and Francis, New York), 2003.
has not been taken i nto account in this case as the fab­ 1 3 Operator Manual, Alambeta (Sensora Instruments and Con­
ric is held between two plates at a pressure of 200 Pa. SUlting, L iberec, Czech Republic), 2002.