The elastic yarns are complex structures
that are not easily controllable especially at the time of the severe
requests because of their heterogeneous compositions. Elastic yarns are
known as the sheath and spandex or Lycra^{®} as the core or the central part
of the stretch yarn structure ^{[3]}. The external part is often cotton or
another different fibre (PET/Viscose, PET, jute, etc). These core spun yarns
are frequently used especially like a wrap of screen in weaving.
The elastic yarns were studied in 1970 by Balasubramanian and Bhatnagar ^{[1]}.
They examined the effect of the conditions of production on the tensile
strength of core spun yarns. The provision of the filament of Spandex in the
total structure of the thread (PET/Viscose) has been recently the subject of
interest of Babaarslan and Osman ^{[2]}. In this study, statistical methods are
necessary to find out the important parameters, which are most influent on
mechanical and physical properties.
Materials and methods
Materials
Two elastic yarns of 59 tex linear density, but with two different linear
density of Spandex  156 dtex and 78 dtex  are used. The formation of
elastic yarns is made on continuous slip while adding on this one two
rollers in the form of metal tube turning in the same direction located
under the reels of wick. The value of desired draft (drawing) is fixed using
an electronic system, Amsler Spandex^{®}, and this value is in connection with
the percentage of the elastic filament in the yarn. The elastic yarn has
physical properties (breaking strength, RKM, elongation at fracture,
torsion, hairiness etc), which are unverifiable and frequently change value
during later processes of transformation. The authors stipulate that this is
due to several parameters such as the drawing of the filament spandex during
the formation of the core spun yarns, with conditioning (temperature of
transformation and realization and moisture) and especially with the
steaming process. To study the influence of the steaming process on the
characteristics of core spun yarn, the authors proposed to study only one
linear density of 17 Nm with spandex filament counts 78 dtex and 156 dtex by
applying the method of experimental designs.
The equipment used for the study of resistance and elongation at break is
Uster Tensorapid 3, which ensures a tensile test until the rupture and
functions according to the principle of the gradient of constant
lengthening. For a rack of 24 reels, the authors carried out 10 tests for
each reel or spindle. They used a total of 240 tests for each elastic yarn.
For the control of twist we used Zweigle twist machinery which ensures
automatic untwisting and twisting of the yarn. The length of each specimen
measured is 500 mm. According to the selected linear density, they applied
the suitable claim.
Finally, the hairiness was measured using Uster Tester 3. The percentage of
cotton fibre component: Micronaire=4.23, length = 28.85 mm, Tenacity = 29.1
CN/tex, elongation = 7.6 mm, short fibre percentage= 9.6% and maturity=
0.87.The composition of the cotton used is given by Table 1. The physical
properties studied are four: breaking strength expressed by RKM, elongation
at break, twist and hairiness. In order to have a better idea on the effect
of the steaming process, the authors carried out tests on the same core spun
yarn (Nm 17 (78) and Nm 17 (156)). Before steaming process, the
corresponding mean values of RKM, elongation at break, hairiness and twist
are given in the Table 2.
Table 1. Cotton mixed percentages.
Fibre origins 
Syrien fibre 
Spain fibre 
Togolian fibre 
Senegal fibre 
Grec fibre 
Percentage (%) 
21 
11 
25 
17 
26 
Table 2. Mean values of physicomechanical
properties of core spun yarns before steaming.
Ls (dTex) 
RKM (KgNm) 
Elongation (%) 
Twist (T/m) 
Hairiness 
78 
15,61 
9,16 
158,52 
7,02 
156 
15,96 
12,04 
160,87 
7,29 
Method of experimental designs,
Tagauchi
It is a strategy used to facilitate the progressive organization of the
acquisition of knowledge. It can also minimize the number of experiments as
much as possible but it should not sacrifice the quality of the results, on
the contrary, the authors seek the best possible precision ^{[4]}.
Four parameters were used as data base of orthogonal Table 3 presented in
order to study their impact on mechanical properties of core spun yarn after
and before steaming process: duration of steaming process (Ds), duration of
relaxation (Dr), temperature of steaming (Ts) and spandex linear density
(Ls). Table 4 shows the combinations of the regulation points and the
factors and levels used in the mixed orthogonal analysis. The study consists
in testing the input parameters chosen to be the most probable influential
on the mechanical behaviour of core spun yarns. Table 5 shows the results of
each test.
Table 3. Factors and levels used in mixed orthogonal analysis.
Factor level 
Ds (min) 
Ts (°C) 
Dr (hours) 
Ls (dtex) 
1 
20 
60 
3 
78 
2 
30 
70 
5 
156 
3 
40 
80 
7 

Table 4. Experimental mixed design:
N° test 
Ls (dtex) 
Ds (min) 
Ts (°C) 
Dr (hours) 
1 
78 
20 
60 
3 
2 
78 
20 
70 
5 
3 
78 
20 
80 
7 
4 
78 
30 
60 
3 
5 
78 
30 
70 
5 
6 
78 
30 
80 
7 
7 
78 
40 
60 
5 
8 
78 
40 
70 
7 
9 
78 
40 
80 
3 
10 
156 
20 
60 
7 
11 
156 
20 
70 
3 
12 
156 
20 
80 
5 
13 
156 
30 
60 
5 
14 
156 
30 
70 
7 
15 
156 
30 
80 
3 
16 
156 
40 
60 
7 
17 
156 
40 
70 
3 
18 
156 
40 
80 
5 
Principal Components Analysis (PCA)
In order to quantify input and output parameters as well as the correlations
between them, we worked on MATLAB 7.01 and Excel software to show the
influence of input/output parameters. The Principal Component Analysis (PCA)
method makes it possible to summarize in a graph the importance of the
inputs and outputs basis (Mahdi, 2002). The Principal Component Analysis is
a statistical method of description and reduction of parameters. Its goal is
to find correlations between the data by:

An assessment of semblance between
the data: it is a question of seeking the points of resemblance (or of
opposition) between the data input as well as the homogeneous groups of
data.

An assessment of the connections
between the variables: one seeks to find the variables which are
positively dependent, the variation (increase or reduction).
The goal under consideration by the PCA
is to gather all the variables (input, output) presented in the form of a
cloud in a trigonometrically circle. The variables which are set in one of
the rounds are variables known as correlated positively because they
evolve/move in the same way. A variation by one parameter will cause the
variation in the same way of the others. Then the groups of variables which
are opposed by one of the axes of the trigonometrically circle are
correlated negatively.
Indeed, the increase in a variable results causes the reduction in the
other. The importance of the variables is all the more significant as the
corresponding points move away from the centre. Thus, the variables in the
central group do not have a very significant influence for the correlations.
In principal figure of axial vectors determined by PCA method, correlated
groups of dots will be defined on these graphs in the following way:
Results and discussion
Figures 2, 3 and 4 prove that the spandex linear density is an
influential and considerable parameter especially in the variation of the
RKM, elongation at break and twist. This seems to be logical when we accept
that an elastic yarn is only one composite structure and the thicker the
inside filament is made up, the better physical properties it will have. A
linear density of 156 dtex, a low temperature of steaming process (60°C)
and duration of relaxation of 4 hours form a good database to obtain the
best physical properties of an elastic core spun yarn intended for hosiery.
If the elastic yarn is intended in weaving, nonporous, the method of the
experimental designs shows that it is necessary to reach a temperature of
80°C with respect of the other parameters like above.
Figure 4 shows that the porosity increases each time the temperature is
decreased. Table 6 shows the contribution values of each input parameter on
the physical properties. These different values were calculated between high
and low levels of each output parameters. The increase in the twist of
elastic yarn after steaming process is about 31.59 T/m (Table 6). If we
change the elastic yarn instead of 17 Nm (78 dtex) another yarn 17 Nm (156
dtex).
Twist is well in narrow correlation with the spandex linear density used.
This is in agreement with the result found by the PCA method (Figure 6)
where the authors announced that twist is in positive correlation with Ls
since they are in the same group E. On the other hand, the increase in the
temperature of steaming process of 60°C with 80°C decreases the value of
the twist of 12,9T/m. The temperature releases the elastic yarn and acts
primarily on the level of stability by minimizing its twist.
Elongation at break is in remarkable growth when duration of relaxation Dr,
linear density of the spandex Ls, and the duration of steaming process Ds,
take the high levels in our experimental design. On the other hand, one can
notice that elongation at break and the temperature of steaming process are
inversely proportional. Therefore, a deficiency in temperature of steaming
process increases the elastic yarn elongation.
While basing itself on the study of experimental designs and the PCA method
presented by Figure 6, we can minimize the study on a fewer number of input
and output parameters. The study with these two methods enabled the authors
to target the most influential parameters on the most interesting physical
properties. Figure 6 shows that the assembly in groups informs us about the
degree of importance of one or more parameters and also of the nature of
correlation between them: positive or negative. Each time the group of
parameters approaches zero, for example group A, the whole of parameters is
considered of low importance and can be repealed. As much the group
approaches the ends (1 or 1), it is considered very significant, example:
groups E, D and B. The PCA method shows that the two physical properties (RKM
and twist) are gathered in the same group. Indeed, the variation of a
physical property of this group informs us on the variation of the other.
When Ls increases, the two properties increase, the contrary case, when Ls
decreases, the two properties decrease.
The duration of Ds steaming process does not contribute enormously in the
physical properties since this parameter forms a group in the centre.
Indeed, the authors conclude that only Ts, Dr and Ls are significant and
contribute enormously in the elastic yarn physical properties. The hairiness
of yarn is related principally to only one parameter, which is the time of
steaming process. If a yarn is intended for knitting then the temperature
must be weak conversely in the case of weaving where it is to better
increase Ts.
Figure 5 shows that the Ts is the most influential parameter on hairiness.
To increase the temperature of steaming process intuitively amounts
minimizing porosity and thereafter a use of the elastic yarn in weaving. In
the opposite case, the elastic yarn will be intended in knitting. In the
case of elastic yarn, the steaming process is in general recommended if the
destination of the latter is weaving according to DuPont; the steaming
process decreases hairiness and in the majority of case the results of a
steaming process depends on time and the temperature.
Conclusion
The studied parameters are not all influential. Indeed, the present
study enabled the authors to analyze the input parameters and their
influence degrees subjectively by the means of the PCA method and
objectively by experimental Tagauchi designs method. Both methods show that
the linear density is the most influential parameter followed by the
temperature of steaming process and the duration of relaxation.
The PCA method proved that the authirs can minimize the data basis of their
study especially when two or several input/output parameters follow the same
evolution, case of twist and the RKM. To manufacture elastic core spun yarn
intended for weaving, it is necessariy to steam yarn at low temperatures, eg:
60°C. On the other hand, when the elastic yarn is that of a warp thread or
screen and thus with high hairiness level, it is advised to steam at high
temperature so that the yarn is with high hairiness level. As the elastic
yarns are generally used in weaving as being a screen, then we ask for
certain extensibility during insertion it is better to increase the
temperature of steaming process Ts.
The PCA is a method of optimization of the experimental design and it makes
it possible to consolidate the deductions approved by the statistical
method, Tagauchi experimental designs.
References

Balasubramanian N and Bhatnagar V K
(1970): The Effect of Spinning Conditions on the Tensile Properties of
Corespun Yrans, J Text Inst 61(11), : 6567, 148157.

Babaarslan and Osman (2001): Method
of Producing a Polyester/Viscose CoreSpun Yarn Containing Spnadex Using
a Modified Ring Spinning Frame, Text, Res J. 71(4): 367371.

Mahdi SAHNOUN (2002): Thesis,
Caractérisation et modélisation de l'influence des paramètres de
structure des étoffes sur l'évaluation mécanique et sensorielle de
leur toucher: Lycra DuPont, 1999, Producing Core Spun Yarns Containing
LYCRA®, Technical Bulletin L120, Nov: 13. Books

Gilles Sado et MarieChristine Sado,
2000, Les plans d'expériences de l'expérimentation à l'assurance
qualité, Nouvelle édition, pp 8387.
Note: For detailed version of this
article please refer the print version of The Indian Textile Journal
November 2007 issue.
B Jaouachi
Textile Research Unit of ISET Ksar Hellal,
Institut Supérieur des Etudes Technologiques de Ksar Hellal,
Rue El Hadj Ali Soua 5070 Ksar Hellal Tunisia.
Tel: (00 216) 22 92 18 04/ (00 216) 73 47 59 00.
Email: boubaker_jaouachi@ yahoo.fr.
M Sahnoun
Textile Research Unit of ISET Ksar Hellal,
Institut Supérieur des Etudes Technologiques de Ksar Hellal,
Rue El Hadj Ali Soua 5070 Ksar Hellal Tunisia.
Tel: (00 216) 22 92 18 04/ (00 216) 73 47 59 00.
