Theoretical Study of Effect of Ring Spinning Triangle
Division on Fiber Tension Distribution
Xinjin Liu, Xuzhong Su
Jiangnan University, Wuxi, Jiangsu CHINA
Correspondence to:
Xinjin Liu email: liuxinjin2006@163.com
ABSTRACT
The spinning triangle is a critical region in the
spinning process of staple yarn. Its geometry
influences the distribution of fiber tension at
spinning triangle directly and affects the qualities of
spun yarns. Taking appropriate measures to change
the ring spinning triangle geometry and improve the
qualities of yarn has attracted more and more
interest recently. Spinning triangle division is one of
the most effective measures, such as solospun
technology. Therefore, in this paper, the effect of
ring spinning triangle division on fiber tension
distribution was studied theoretically. The general
case that the spinning triangle is divided into two
parts, primary triangle and final triangle, was
investigated. Firstly, theoretical models of the fiber
tension distributions in the final and primary
spinning triangles are given respectively using the
principle of minimum potential energy. Secondly,
the fiber tension distributions in the spinning
triangle with different shape parameters are
numerically simulated. Finally, the possible effects
of ring spinning triangle division on yarn qualities
are analyzed according to the numerical simulations
and previous results. In addition, the properties of
spun yarns produced by the modified ring spinning
system were evaluated and analyzed.
Keywords: Spinning triangle division; Fiber tension
distribution; Ring spinning
INTRODUCTION
In the ring spinning process, the fiber strand comes
from the draft zone which is flat, and almost all
fibers are parallel to the twisting axis of the strand
in this time. Then, when it is twisted, the fiber
strands rotate around the axis and the width begin to
decrease, and the fibers on both sides of the axis
fold gradually and roll into the center of the spun
Journal of Engineered Fibers and Fabrics
Volume 10, Issue 3 2015
yarn, and the spinning triangle is formed [1]. The
spinning triangle is a critical area in the staple yarn
spinning process and its geometry influences the
fiber tension distribution at the spinning triangle
and affects the properties of spun yarns [1-4].
Taking appropriate measures to change the spinning
triangle geometry and improve the quality of yarn
has attracted great interest recently [3-5]. All the
existing spinning technologies are achieved by
changing the ring spinning triangle in a sense, such
as the sirospun [6], solospun [7], and compact
spinning [8]. Spinning triangle division is one of the
most effective measures, such as solospun
technology, which is implemented by dividing the
ring spinning triangle into several small primary
triangles and one final triangle by using a solospun
roller [7]. Meanwhile, a series of achievements on
solospun technology haven been obtained, such as
Solo-Sirofil Composite Yarn [9], twist amplitude
and breaking strength of Solospun yarns [10], and
Solospun worsted yarns [11] .
The subject of spinning triangle has been one of the
most important research topics and has attracted
increasing attention recently. Fruitful results have
been obtained [3,4,12,13]. A theoretical model of
the symmetrical spinning triangle was developed
using the principle of stationary total potential
energy [13]. However, the spinning triangles are
often asymmetric due to the frictional contacts of
fibers with the bottom roller in real situations.
Therefore, this model was further extended to the
asymmetric spinning triangle by introducing a shape
parameter for describing the skew level of the
geometry of the spinning triangle [1]. Furthermore,
yarn spinning tension often has an obvious angle
with the vertical axis perpendicular to the nip line in
some modified ring spinning systems [4,5].
20 http://www.jeffjournal.org
Therefore, the model was proposed further by
considering the inclination angle of the spinning
tension [3]. Furthermore, the quantitative
relationships between the mechanical performance
of a ring spinning triangle and the spinning
parameters were investigated by using the Finite
Element Method (FEM) [17].

This paper investigates the effects of ring spinning

triangle division on fiber tension distribution. The
general case that the spinning triangle is divided
into two parts, primary triangles and final triangle,
is investigated. Theoretical models of the fiber FIGURE 3. Fiber distribution at middle primary spinning
tension distributions in the final and primary triangle.
spinning triangles are given respectively by using
the principle of minimum potential energy. With the
help of the High-speed camera, the fiber tension
distributions in the spinning triangles with different
shape parameters are simulated numerically.

FIGURE 4. Fiber distribution at right primary spinning triangle.

FIGURE 1. Spinning triangle division with three primary

triangles.

FIGURE 5. Fiber distribution at final spinning triangle.

THEORETICAL ANALYSIS
The spinning triangle with three primary triangles is
shown in Figure 1. Here, h1 and h2 are the
height of the primary and final triangle respectively,
H denotes the height of ring spinning triangle and
FIGURE 2. Fiber distribution at left primary spinning triangle.
satisfies

Journal of Engineered Fibers and Fabrics 21 http://www.jeffjournal.org

Volume 10, Issue 3 2015
 H= h1 + h2
is the division proportion and defined as
= h2 / h1
The primary spinning triangle models are shown in
Figures 2-4 respectively. Here, C1 , C2 and C3
are the twisting points with the load F1 , F2 and
F3 respectively. 2n + 1 denotes the total number
of fibers at the front roller nip line, and
2n1 + 1= (2n + 1) / 3 denotes the total number of
fiber at each primary triangle. 1 and 1 are
inclination angles of spinning tension in the left and
right primary spinning triangles respectively. i is
the angle between the right i -th fiber and the
central fiber, and i is the angle between the left
i -th fiber and the central fiber at each primary
triangle. The final spinning triangle model is shown
in Figure 5. Here, C is the twisting point with the
constant load Fs , 1 is the angle between the
right first substrand and the central substrand, and
1 is the angle between the left first substrand and
the central substrand. Then, we have 1 = 1 and
1 = 1 . Suppose that the width of the spinning
triangle at front roller nip line is w .
Fs 2 ( 2v + 1) AE M j ( M j 1)
=Fj =0,1,,m n
2 M j 1
j =1
m
Fs 2 ( 2v + 1) AE M j ( M i 1)
=Fj=0,1,,m m
j =1
2 M j 1 2
j =1
Journal of Engineered Fibers and Fabrics
Volume 10, Issue 3 2015
(1) For convenience of analysis, we make the following
assumptions [1,3]. The cross section of all fibers is
circular with identical diameters. All fibers are
gripped between the front roller nip and the
convergence point. Fiber slippage and migration
and the frictional contacts between fibers and the
(2) front roller are not taken into consideration. The
velocity of fibers in the spinning triangle is constant.
The fibers stress-strain behavior follows Hooks law
for small strain. The ends of all fibers gripped in the
front roller nip distribute evenly.
In the following, the fiber tension distributions at
primary triangles and the substrand tension
distributions at final spinning triangles will be
analyzed by using the principle of minimum
potential energy. First, the substrand tension at final
triangle will be investigated. For giving the general
substrands tension distribution model, we suppose
that there are 2m + 1 substrands at final spinning
triangle. Then, by using the principle of minimum
potential energy, the substrand tension distribution
at final triangle can be presented in Eq. (3) and Eq.
(4).
Second, according to the primary spinning triangle
models shown in Figures 2-4 and substrand tension
distribution at final triangle shown in Eq. (3) and Eq.
(4), by using the principle of minimum potential
energy, the fiber tension distributions at primary
triangles can be given in Eq. (5) through Eq. (8).
m
j =1
M j + ( 2v + 1) AE ( M j 1) (3)
2
M j + ( 2v + 1) AE ( M j 1) (4)
22 http://www.jeffjournal.org
Where 2v + 1 =
2n + 1 denotes the total number of fibers in each substrand, A is the cross section of fiber, E
2m + 1

is the fiber tensile Youngs modulus, M j =

1 , 1 , wj ( 2n + 1) .
M j = =
tan aa =
tan
cos j 2nH ( 2m + 1)
j
cos j
j

v
Fj AE M i , j ( M i , j 1) + M i, j ( M ij 1)
v

F = i 1 =i 1 M + AE ( M i , j 1) (5)
=i 0,1,
= , v , j 0,1,, m v v i, j

i, j
=i 1 =
M
2

i 1
+ M i, j 2 1

v
Fj AE M i , j ( M i , j 1) + M i, j ( M ij 1)
v

i 1 =i 1 M + AE M 1
= F
i 0,1,
= , v , j
=
0,1,, m v v i, j ( i, j ) (6)

=i 1 =
i , j i, j
i 1
M 2
+ M 2
1

v
Fj AE M i , j ( M i , j 1) + M i, j ( M i, j 1)
v

F = i 1 =i 1 M + AE ( M i , j 1)
(7)
=i 0,1,
= , v , j 0,1,, m v v i , j

M
i , j
=i 1 =
2

i 1
+ M i, j 2 1

v
Fj AE M i , j ( M i , j 1) + M i, j ( M i, j 1)
v

F = i 1 =i 1 M + AE ( M i, j 1)
(8)
=i 0,1,
= , v , j 0,1,, m v v i, j

M
i , j
=i 1 = i 1
2
+ M i, j 1
2

1 1
Here, M i , j = , M i, j =
cos i , j tan a j sin i , j cos i, j + tan a j sin i, j

1 1
M i , j = , M i, j =
cos i , j + tan a j sin i , j cos i, j tan a j sin i, j

wi h1 tan a j wi + h1 tan a j
tan (i, j a j ) = , tan ( i , j + a j ) =
h1 h1

wi + h1 tan a j wi h1 tan a j
tan (i, j + a j ) = , tan ( i , j a j ) =
h1 h1
w w
wi = i , wi = i .
2n 2n

Journal of Engineered Fibers and Fabrics 23 http://www.jeffjournal.org

Volume 10, Issue 3 2015
SIMULATIONS AND DISCUSSIONS
1.5
From Eq. (3) through Eq. (8), we can see that the
fiber tension distribution at the spinning triangle
mainly depends on the spinning tension, the number 1
b=0
of primary triangles, the division proportion, the

Fiber Tension (cN)

b=0.25
b=0.5
fiber tensile Youngs modulus and cross section, and 0.5 b=1
b=2
the height and width of the spinning triangle. For b=3
obtaining the height of the ring spinning triangle H ,
0
High-speed camera should be applied. In the
experiment, a High-speed camera system,
OLYMPUS i-speed3, was set up above the -0.5
0 20 40 60 80 100 120 140
Fiber position (Number)
transparent roller and used to capture the geometry
of the spinning triangle. Combed cotton roving of
600tex (Ne0.97), with the twist of 1.19/inch and FIGURE 7. Fiber tension distribution at spinning triangles with
different division proportions when m = 4 .
containing 40% long-staple cotton and 60%
medium cotton was used as raw material. The fiber
properties are fiber length 32.5mm; fiber strength 1.4

26.8cN/tex; micronaire value 4.1. Then, the Ne32 1.2

cotton yarn will be spun in EJM128K ring spinning 1

frame. The yarn and fiber parameters of Ne32 0.8 b=0

Fiber Tension (cN)

b=0.25
cotton yarn used in the analysis 0.6
b=0.5

are w = 3.3mm , H = 2.9mm , yarn spinning 0.4 b=1

b=2

tension Fs = 20cN , number of fibers in yarn 0.2 b=3

2n + 1 =135 , fiber linear density A = 0.15tex , 0

-0.2
twist direction Z. By using Matlab software, the
-0.4
numerical simulation results of the fiber tensions at
-0.6
spinning triangle were obtained in Figures 6-11. 0 20 40 60 80
Fiber position (Number)
100 120 140

Here, x-axis denotes the serial number of the fibers

at the front roller nip from left to right, y-axis FIGURE 8. Fiber tension distribution at spinning triangles with
denotes the fiber tension, and the negative value different division proportions when m = 7 .
indicates the compressive force whereas the positive
2.5
value indicates the tensile force.
2

2.5 m=0
1.5 m=1
m=4
2 b=0
Fiber Tension (cN)

m=7
b=0.25 1 m=13
1.5 b=0.5 m=22
b=1
b=2 0.5
Fiber Tension (cN)

1
b=3

0.5 0

0 -0.5

-0.5
-1
0 20 40 60 80 100 120 140
Fiber position (Number)
-1

-1.5
0 20 40 60 80 100 120 140
Fiber position (Number) FIGURE 9. Fiber tension distribution at spinning triangles with
different primary triangle number when = 0.25 .
FIGURE 6. Fiber tension distribution at spinning triangles with
different division proportions when m = 1 .
Journal of Engineered Fibers and Fabrics 24 http://www.jeffjournal.org
Volume 10, Issue 3 2015
 Fi of each primary triangle is independent of
2.5
when the number of primary triangle 2m + 1 is
2
fixed. However, with the increasing of , the height
of primary triangle h1 decreases, and shape
1.5 m=0
m=1
parameter m = wm / h2 increases correspondingly,
Fiber Tension (cN)

m=4 which indicates the magnitudes of fiber tensions at

1 m=7
each primary triangle are increased [1,3], where
( )
w ( n m ) / n ( 2m + 1) denotes the width
m=13

0.5
m=22
wm =
of the primary spinning triangle. Meanwhile, with
0
the decreasing of h1 , the angle between the left
-0.5 boundary fiber in the j -th primary triangle and the
right boundary fiber in the j + 1 -th primary triangle
-1
0 20 40 60 80 100 120 140 is increased, which leads to the increase of fiber
Fiber position (Number)
tension fluctuation between two adjacent primary
triangles. Furthermore, as shown in Figures 6-8, the
FIGURE 10. Fiber tension distribution at spinning triangles with total magnitudes of fiber tensions at spinning
different primary triangle number when = 1 . triangles with divisions are larger than that of
common ring spinning. Meanwhile, comparing
Figure 6, Figure 7 and Figure 8, for the same
3 division proportion , with the increasing of primary
2.5
triangle number 2m + 1 , the magnitudes of fiber
m=0
m=1
tensions at each primary triangles are decreased.
2
m=4 According to Eq. (3) and Eq. (4), with the increasing
m=7
1.5
of m , the difference between the load F j and
Fiber Tension (cN)

m=13

1
m=22
Fj +1 is decreased, which lead to the smaller fiber
0.5 tension fluctuation between two adjacent primary
0
triangles according to Eq. (5) through Eq. (8).
Meanwhile, with the increasing of m , the fiber
-0.5
number at each primary triangle 2v + 1 is
-1 decreased, and the load F j also decreased, which
-1.5
0 20 40 60 80 100 120 140
lead to the reduction of fiber tension magnitudes at
Fiber position (Number) each primary triangle.

Fiber tension distributions at spinning triangles with

FIGURE 11. Fiber tension distribution at spinning triangles
with different primary triangle number when
It is easy to see that the distributions of fiber tension
are all symmetrical about the central fiber because
the symmetrical spinning triangle is considered only
in this paper. The asymmetrical spinning triangles
due to the frictional contacts of fibers with the bottom
roller [1] or obvious angle with the vertical axis
perpendicular to the nip line [3] can also be
considered similarly. Fiber tension distribution at
spinning triangles with five different division
proportions in Figure 6 when m = 1 . Here,
= 0 denotes the ring spinning without division. It
is shown that with the increasing of , the
magnitudes of fiber tensions at each primary triangle
are increased. According to Eq. (3) and Eq. (4), it is
obvious that substrand tension in the final triangle is
independent of the division proportion . The load
five different numbers of primary triangle are shown
= 3. in Figures 9-11. Here, m = 0 denotes the ring
spinning without division. As shown in Figures 9-11,
with an increasing of , the fluctuation between two
adjacent primary triangles is increased greatly when
m = 1 , and the total fiber tension magnitudes
increases greatly when m = 22 .
In the following, the possible effects of ring spinning
triangle division on yarn qualities will be analyzed
according to the numerical simulations above and
previous results.
Hairiness is one of the most important properties of
the spun yarn. It has been shown that the larger
tension forces of the outer fibers at spinning triangle
would increase the wrapping of surface fibers and
lead to the reduction of hairiness correspondingly

Journal of Engineered Fibers and Fabrics 25 http://www.jeffjournal.org

Volume 10, Issue 3 2015
[14-16]. According to the numerical simulation time by YG135G evenness tester at a speed of
results above, the tension of outer fibers at each 400m/min, and take the value as the evenness of
primary spinning triangle can be increased greatly this one bobbin of yarn. Finally, the average values
with appropriate spinning triangle division, such as of ten bobbin yarns were taken as the corresponding
m = 1 and = 3 (see Figure 6 and Figure 11). qualities of spun yarn. The measured qualities of the
Meanwhile, lots of hairiness will be rolled into the spun yarns produced are shown in Table I. From the
spun yarn body when the substrands are twisted into measured results, it is also shown that comparing
yarn at the final spinning triangle under the second with common ring spinning; appropriate ring
twisting [7]. Therefore, appropriate ring spinning spinning triangle division can reduce yarn hairiness
triangle division would be benefit for yarn hairiness greatly, and improve yarn strength slightly.
reduction. Meanwhile, yarn evenness changes little with
different spinning triangle divisions.
Yarn strength is another most concerned property in
evaluating yarn performance, which can be
reinforced by increasing the lateral pressure between
fibers [18]. The number of fibers at each primary
spinning triangle is decreased with the spinning
triangle division, which can help to the transfer of the
twist uniformly, and thus benefits for the fiber tension
balance. Then, a more compact core and a densely
packed zone located in somewhere half way from the
centre fiber to the boundary fiber can also be
produced easily. Then, the lateral pressure between
FIGURE 12. The diagram of the Solospun roller.
fibers can be increased.
TABLE I. Yarn properties.
EXPERIMENTS
3mm
For verifying our theoretical results, a number of Hairiness Evenness CV Breaking
Ne32 Ring spun and Solospun yarns were spun on -1 /% strength /cN
/(10m)
the EJM128K machine. The total width of the fiber Common ring
strand coming from the draft zone is equal to 52.3 12.68 325
spinning
4.5mm approximately. The diagram of the Solospun
Solospun with
roller is shown in Figure 12, where d is the width three substrands
41.7 12.43 332
of each groove. For getting the different substrands
(three and five), we take for d1 = 1.5mm and Solospun with
d 2 = 0.9mm respectively. Taking ten bobbin five substrands
39.8 12.52 330

yarns as measuring samples and all the samples

were conditioned for at least 48 hours under
standard conditions before testing (652% RH and
202). Yarn hairiness, strength and evenness were
measured. The test instruments are as follows:
single yarn tester YG063, hairiness tester YG172A,
evenness tester YG135G. For each bobbin yarn, the
hairiness was tested ten times using YG172A
hairiness tester under 100 m/min speed and 5g
pretention, and the test time is 1 minute, and take
the average value of ten tested results as the
hairiness of this one bobbin yarn; the breaking force
was also tested ten times on YG063 fully automatic
single yarn strength tester at a speed of 500 mm/min
with a pretension 1.8 cN/tex, and take the average
value of ten tested results as the breaking force of
this one bobbin yarn; the evenness was done one
Journal of Engineered Fibers and Fabrics
Volume 10, Issue 3 2015
CONCLUSION
In this paper, the effects of ring spinning triangle
division on fiber tension distribution were studied
theoretically. The case that spinning triangle is
divided into two parts, including primary triangles
and final triangle has been studied. Two parameters
including division proportions and number of
primary triangle m , which describes the ratio of the
primary spinning triangle height to the final triangle
height and division number of ring spinning triangle
respectively, were introduced in the analysis. Then,
26 http://www.jeffjournal.org
 by using the principle of minimum potential energy
and based on some assumptions, theoretical models
of the substrand tension distributions Fj =0,1,,m and
Fj=0,1,,m at final spinning triangle and the fiber
tension distributions at all primary spinning triangle
F
=i 0,1,
= , v , j 0,1,, m
, = F
i 0,1,
= , v , j 0,1,, m
F , F have been
=i 0,1,
= , v , j 0,1,, m = i 0,1,
= , v , j 0,1,, m
given. However, some factors such as the ring tensile
friction or the state of the cursor were not considered
in the paper at present, and will be investigated in our
future studies.
The simulation results on Ne32 cotton yarn have
been presented by using Matlab software according
to the proposed theoretical models. The results show
that for the same m , with an increasing of , the
fluctuation of the fiber tension between two adjacent
primary triangles becomes more and more obvious,
and the magnitudes of fiber tensions at each primary
triangles also increases. While for the same , with
an increasing of m , the fluctuation is decreased, and
the magnitudes also decrease. Finally, the properties
of spun yarns produced by the modified ring spinning
system were evaluated, and the theoretical results
obtained verified.
ACKNOWLEDGEMENT
This work was supported by National Natural
Science Foundation of P. R. China [grant number
11102072], Natural Science Foundation of Jiangsu
Province [grant number BK2012254], Prospective
industry-university-research project of Jiangsu
Province [grant number BY2012065], Fundamental
Research Funds for the Central Universities [grant
number JUSRP51301A], and Project Funded by the
Priority Academic Program Development of Jiangsu
Higher Education Institutions.
Journal of Engineered Fibers and Fabrics
Volume 10, Issue 3 2015
REFERENCES
[1] Hua T., Tao X. M., Cheng K. P. S., Xu B. G.
Effects of Geometry of Ring spinning
triangle on Yarn Torque Part I: Analysis of
Fiber Tension Distribution, Textile
Research Journal 2007; 77(11): 853-863.
[2] Hua T., Tao X. M., Cheng K. P. S., Xu B. G.
Effects of Geometry of Ring spinning
triangle on Yarn Torque: Part II:
Distribution of Fiber Tension within a Yarn
, and Its Effects on Yarn Residual Torque,
Textile Research Journal 2010; 80(2):
116-123.
[3] Feng J., Xu B. G., Tao X. M., Hua T.
Theoretical Study of Spinning Triangle
with Its Application in a Modified Ring
Spinning System, Textile Research Journal
2010; 80(14): 1456-1464.
[4] Wang X. G., Chang L. L. Reducing Yarn
Hairiness with a Modified Yarn Path in
Worsted Ring Spinning, Research Journal
of Textile and Apparel 2003;, 73(4): 327-
332.
[5] Thilagavathi G., Udayakumar D., Sasikala L.
Yarn hairiness controlled by various left
diagonal yarn path offsets by modified
bottom roller flute blocks in ring spinning,
Indian Journal of Fiber and Textile Research
2009; 34: 328-332.
[6] An X. L., Yu C. W. Dynamic model of
sirospun process. Part I: theoretical dynamic
model, Journal of the Textile Institute 2010;
101(9):805-811.
[7] Cheng L. D., Fu P. H., Yu X. Y. Relationship
between hairiness and the twisting
principles of solospun and ring spun yarns,
Textile Research Journal 2004; 74(9): 763-
766.
[8] Momir N., Zoran S., Franc L., Andrej S.
Compact Spinning for Improved Quality of
Ring-Spun Yarns, Fibers & Textiles in
Eastern Europe 2003; 11(4): 30-35.
[9] Yang R. H., Wu W. M., Wang H. B., Wan Y.
Q., Xie C. P., Gao W. D. Effects of
Solospun Roller on Properties of
Cotton/Polyester Solo-Sirofil Composite
Yarn, Journal of Engineered Fibers and
Fabrics, 2012; 7(3): 98-102.
[10] Cheng L. D., Fu P. H., Yu X. Y. Relation
between twist amplitude and breaking
strength of Solospun yarns, Textile
Research Journal, 2004; 74(4): 351-353.
27 http://www.jeffjournal.org
[11] Chang L. L., Wang X. G. Comparing the
hairiness of Solospun and ring spun worsted
yarns, Textile Research Journal, 2003;
73(7): 640-644.
[12] Fujino K., Uno M., Shiomi A., Yanagawa Y.,
Kitada Y. A Study on the Twist Irregularity
of Yarns Spun on the Ring Spinning Frame,
The Textile Machinery Society of Japan
1962; 8: 51-62.
[13] Shaikhzadeh N. S. An Analysis of the Twist
Triangle in Ring Spinning, PhD Thesis,
University of New South Wales, Australia,
1996.
[14] Rafael B., Wang L. J., Wang X. G. A
Controlled Experimental on Yarn Hairiness
and Fabric Pilling, Journal of Textile
Research 2007; 77(3): 179-183.
[15] Cheng K. P. S., C. H. L. Li. JetRing Spinning
and its Influence on Yarn Hairiness,
Journal of Textile Research 2002; 72(12):
1079-1087.
[16] Kalyanaraman A. R. A Process to Control
Hairiness in Yarn Journal of the Textile
Institute 1992; 83(3):407-413.
[17] Li S. Y., Xu B. G., Tao X. M. Numerical
Analysis on Mechanical Behavior of a
Ring-Spinning Triangle Using the Finite
Element Method, Textile Research Journal,
2011; 81(9): 959-971.
[18] Yang K., Tao X. M., Xu B. G., Jimmy L.
Structure and Properties of Low Twist
Short-staple Singles Ring Spun Yarns,
Textile Research Journal 2007; 77(9):
675-685.

AUTHORS ADDRESSES
Xinjin Liu
Xuzhong Su
Jiangnan University
Lihu Road 1800
Wuxi, Jiangsu 214122
CHINA

Journal of Engineered Fibers and Fabrics 28 http://www.jeffjournal.org

Volume 10, Issue 3 2015