Indian Journal of Fibre '& Textile Researc!

t\~
..
.
Vol. 15, March 1990, pp. 1-5
,- ~
•..
'

. " ...
•
','

"

Fib~e quality and yarn strength relationships

~
. P Chellamani, IndraDcraiswamy & T V Ratf\aJll,

';• South India Textile Research Association, Coimbatore 641 014, India
. .,-
• Received 12 October 1989; accepted 18 December 1989
, '

A simple empirical relationship between the fibre properties of cottons currently in use and the
CSP of the yarns spun on modem machinery has been worked out in view of the changes 'both in
technology and indigenous cottons. A higher degree of correlation of the order of 0.99 is obtained
between the actual and predicted values of CSP in the modified relationship.

Keywords: Fibre quality index, Span length, Twist multiplier, Yam strength

1 Introduction
changes during the same period. Fibre tended to
Evolving workable prediction formulae to relate
become finer and longer but less mature. There
fibre properties to yam quality has been an inter-
has also been higher incidence of seed coats,
esting field of work for several research workers
trash content and foreign matter.
for more than three decades 1. Lord? developed an
In view of the above changes both in technol-
integrated index using basic fibre properties to
predict the spinning performance. Subramaniam ogy and indigenous cottons, it was considered
worthwhile to take a fresh look at the relationship
et al.3 put forward a fibre quality index by which
between yam quality and fibre properties. This
Bogdan's intrinsic strength parameter of a cotton
paper reports a simple empirical relationship be-
can be predicted. In addition to this, they also
tween fibre properties of cottons currently in use
identified two constants, which, once evaluated
and the CSP of yams spun on modem machinery.
for a given spinning set-up, would enable the lea
strength to be predicted by estimating first the in-
trinsic strength parameter and then the lea
2 Materials and Methods
Nine cottons, viz. J-34, Jayadhar, Jyoti, LRA,
strength from Bogdan's equation. This procedure
MCU-7, H-4, Shankar-4, Varalakshmi and Suvin,
was modified by Dhawan and Subrarnaniam" to
further improve the accuracy of prediction. Two were selected for this study. The important qual-
other papers=" dealing with this subject have been ity characteristics of these cottons are given in
published by Iyer and Iyer. Table 1. Each cotton was spun to 3-6 counts and
each count at 4 different twist multipliers. In all,
Recently, Hunter? has related the properties of about 150 spinnings were made. The yam sam-
fibres measured by HVI and the conventional sys- ples were tested for lea CSP taking 40 readings
tems to the processing waste and yam properties per test.
using multiple regression technique. SITRA deve-
loped an empirical relationship" between import-
ant fibre properties and yam CSP in 1968. 2.1 Analysis of Data
However, in the last decade there has been a The data were analyzed keeping in view the
spectacular improvement in the technological per- simple relationship developed earlier at SITRA.
formance of spinning machinery. This resulted First, the relationship between fibre properties
from a number of factors such as improved open- and maximum strength of yam spun to a given
ing and effective separation of heavy particles in count was studied. Then the optimum TM for
modem blow room lines, better carding and effi- maximum strength was determined for each cot-
cient fibre control during drafting in modem pre- ton and its relationship with fibre properties was
paratory and spinning machinery. Apart from the examined. Finally, the extent to which the yam
technological improvements, indigenous cottons strength at a given TM decreases from the maxi-
have also undergone significant qualitative mum strength at optimum TM was related to the
 INDIAN J. FIBRE TEXT. RES., MARCH 1990

Table 1 - Fibre properties

Property Colton

J-34 Jayadhar Jyoti MCU-7 H-4 LRA ShankaJ:.4 Vara1akshmi Suvin

Effective length, mm 24.8 26.4 26.9 29.7 30.5 31.0 32.8 36.1 39.4

Short fibres, % 17.8 15.1 18.2 14.9 13.5 18.4 14.4 14.9 15.6
2.5% Span length, mm 22.1 22.3 24.2 27.9 28.4 27.6 29.8 33.3 37.8
50% Span length, mm 10.9 11.5 12.0 13.6 14.0 13.8 14.8 15.6 17.7

Stelo strength at
3 mm gauge, g/tex 17.1 17.2 19:6 23.9 20.3 20.6 20.9 24.3 30.3

~g/in 3.80 4.75 4.70 4.40 3.80 4.00 3.65 3.20 2.90

Maturity coefficient
(Caustic soda method) 0.788 0.792 0.770 0.830 0.788 0.844 0.784 0.800 0.758

deviation of the actual 1M from the optimum 2J. 1·2

TM.
20 1·0
3 Results and Discussion
••
.~ a.
-' 0·8 If)
3.1 Fibre Quality and Yarn CSP E 15 u
.~ c
It is well known that when a given cotton is 0.6 ';;'
spun to different counts using the same twist mul- 2' ~--- ....•.... , e
IX l1 o
tipliers for all the yarns, cottons giving strong "0 O.J. ~
yarns not only tend to fall off in strength less ra- ••
a.
pidly but also show high initial values, i.e. high in- tl 5 _ SloP" of K"9r<,ssion lin. 0·2
"t» Crop in C5 P
tercept on the Y-axis. Weak yarns have relatively _ - - 1

high fall in strength with low intercept values. The 01 I!, I , I , I I ,

analysis of data in the study showed that in long- 123J.56789

and extra long-staple cottons ( > 28 mm), the drop Cotton

in esp for a unit increase in count is 0.5%, Fig. 1 - Slope of regression line and % drop in CSP for dif-
whereas in short- and medium-staple cottons the ferent cottons [1 - Jayadhar; 2 - Jyoti; 3 - J-34; 4 - MCU-7;
fall in strength is 0.8%. 5 - H-4; 6 - LRA; 7 - Shankar-4; 8 - Varalakshmi; and
The slope of the regression line relating count 9-Suvin]
and esp was, however, practically the same for
all the cottons studied (Fig. 1). It was 16 for the m = Maturity coefficient measured by caustic soda
best fit. For establishing a relationship between method.
yarn strength and fibre properties, the slope has
been assumed to be equal to this best value. To examine the closeness of fit, the actual maxi-
mum esp values were compared with those pre-
Thus, the yarn csr (Y) is given by: dicted from Eq. (1). A high correlation of 0.99
was obtained between the actual and expected va-
Y= 320(JFQI+1)-16n ... (1) lues (Fig. 2). The standard error of estimate was
about 100 units of esp, which works out to 4.3%
where FQI = LSm/ F, and n is the count. on average values.
L= 50% span length (mm). In the spinning made in this study, the yarn
S= Fibre strength (g/tex) measured by stelometer strength values relating to 14s count were found
at 3 rnm gauge. to be highly inconsistent. In fact, in six cottons,
F= Fineness (micronaire value in mg/in) mea- the esp values for the sante twist multiplier were
sured using airflow method. lower by 4-8% compared to those for 20s count

2
 CHELLAMANI et al.: FIBRE QUALITY AND YARN STRENGTH RELATIONSHIPS

spun from the same cottons. The average CSP for The ratio between the actual and predicted va-
all the nine cottons was 4% less in 14s. In view of lues of yam CSP for various cottons and counts is
this, 14s count was not included in the above fit. given in Table 2, which shows a good agreement
However, the analysis of data including 14s count between the predicted and actual values in all the
also gave a high correlation of 0.98 between the nine cottons and counts, the overall mean devia-
actual and expected CSP values (Eq. 2). tion being only about 3.5%.
A multiple linear regression analysis of the data
Y= 293 + 320 JFOI - 16 n ... (2) (Eq. 3) also gave a fairly high correlation of 0.98.

Eq. (2) is the same as Eq. (1) except for a lower Y=66L+72S-374F+314M-16n+1588 ... (3)
value for constant term (by 27 units). This is be-
where Y is the lea CSP and n, the yam count.
cause generally low values of CSP are observed
The units and measurements for other variables
for 14s. Consequently, by the inclusion of 14s
were the same as for Eq. (1).
count, the standard error of estimate of CSP was
However, Eq. (1) derived from FOI is of a sim-
significantly greater at 120 units.
pler form with ease of application at mill level
apart from being able to characterize by a single
3"X>0 value (FOI) the overall cotton quality from the
point of view of yam strength.
3200
Q.
\II 3.2 Effect of Various Fibre Properties
U

"2 2700
The expression relating fibre properties to CSP
v without taking maturity into account is given by:
.,
;;
Q: 2200 CSP=393+270JLSIF-15n ... (4)
....
1700
The correlation coefficient between the actual
CSP values and those predicted using Eq. (4) was
found to be 0.98. The corresponding standard er-
1700 2200 2700 3200 3700
Actual CSP ror of estimate was about 5%.
Considering the separate contribution of fibre
Fig. 2 - Actual and predicted CSP at optimum twist length, fibre strength and fibre fineness, all of

Table 2 - Ratio of actual and predicted yarn CSP and mean deviation
Cotton 205 305 40s 60s 80s 100s Average Mean
deviation, %
Jayadbar 0.95 0.95 0.95 5.4

Jyoti 1.05 0.97 0.98 1.00 3.4

J-34 1.03 1.05 1.03 1.04 3.4

MCU-7 0.98 0.95 0.97 1.02 0.98 3.1

H-4 1.04 1.00 0.99 1.00 1.01 1.8

LRA 1.03 1.00 1.05 1.03 2.7

Shankar-4 0.93 0.97 1.04 1.01 1.06 1.0 4.3
Vara1akshmi 0.98 1.03 1.04 0.87 0.95 0.98 0.98 5.0
Suvin 0.98 1.04 1.05 0.98 0.93 1.03 1.00 3.8

Average LOO 0.99 1.01 0.99 0.98 1.01 1.00

Mean deviation, % 3.1 3.7 2.7 4.0 6.1 2.8 3.6

3
 INDIAN J. FIBRE TEXT. RES., MARCH 1990

which have about equal effect on yam esp, each 5~r----------------------,

property taken along with the count individually
explains about 75-85% of the total variation, the
correlation coefficient being in the range of 0.85-
",,,,
0.90. Also, any two of these three fibre properties
taken together along with the count gave a very ::I
4.50 .•...
\)'

•..
good fit, the correlation coefficients between the ."
~ 425
actual and predicted values being in the range of
~
0.96-0.98. D:.
For a given cotton, the esp values obtained in
this study were about 13% more in coarse counts
and 7% more in fine counts as compared to the
values obtained earlier at SITRA. 4·00 4.25 4·50 4.75 5.00
Actual TM
The improved opening, carding and fibre con-
trol on modem machines explain the better qual-
Fig. 3 - Actual and predicted TM
ity of yams obtained in the present study.

3.3 Twist for Maximum Strength

It is well known that the twist multiplier for 15
maximum strength is a function of cotton quality
Longer, more uniform and finer cottons have, in
general, lower optimum twist multipliers. Opti-
mum TM for maximum strength was found to be
linearly related to 50% span length and fibre fine-
ness of cottons studied. It is given by the follow-
ing equation: c Y=0·53.- 0·40

c
.2 5
Optimum TM = (50 - L+ F)/9 ... (5) .~
~ .:
c
The optimum TM for maximum strength was
about 5% lower than that observed in an earlier
"
o~--~--~--~--~--~--~
study", Improvement in the fibre cohesion -1 10 15 20 25 30
./. DeoviatTon in TM from Optimum
achieved in modem machinery, perhaps, explains
the attainment of maximum strength at lower le-
velsofTM. Fig. 4 - Relationship between TM and strength
A very close fit was found between the predict-
ed and the actual values of TM, the correlation are empirical and as such can be applied only for
coefficient being 0.95 (Fig. 3) and the mean devi- the range of cottons and counts studied. They will
ation,0.07. nevertheless help mills in evaluating,. with a high
degree of accuracy, their process proficiency in
3.4 Effect ofTM on Strength terms of yam esp with refemce to modem ma-
Eqs (1) to (41) are all based on the maximum chinery. The standard error of the predicted va-
strength of yams at optimum TM. Analysis of the lues is about 5%, and this fact should not be
data showed that the fall in strengh (5%) is fairly overlooked while interpreting data. Also, the rela-
uniform at different levels of twist (Fig. 4). It is tionships are applicable only to the maximum
represented by the following relationship: yam strength. How closely the process proficiency
attained by the mills approaches the one worked
5 = 0.53/- 0.40 ... (6) out from these relationships is being investigated.

where 1 is the % difference in twist multiplier 5 Conclusions

from the optimum. A simple relationship has been established be-
tween yam esp at optimum TM and the square
root of the fibre properties, viz. 50% span length,
4 Limitations strength, fineness and maturity. The correlation
The various relationships given in this paper coefficient between the actual and predicted va-

4
 CHELLAMANI et al.: FIBRE QUALITY AND YARN STRENGTH RELATIONSHIPS

lues has been found to be 0.99, the standard er- Acknowledgement

ror of estimate being about 5%. The authors would like to express their grateful
The CSP at optimum twist decreases at the rate thanks to Mis Lakshmi Machine Works Limited
of 0.8% for every unit increase in count in the and, in particular, to Shri K.B. Krishnan, Assistant
case of short- and medium-staple cottons and General Manager and Shri K P R Pillay, Consult-
0.5% for long- and extra-long staple cottons. The ant, for their cooperation and help in processing
slope of the regression line is, however, almost the samples at their spinning laboratory. They are
same for all the cottons. also thankful to Shri R. Rajamanickam for analyz-
Fibre length, strength and fineness have about ing statistical data in computer and to the staff of
the same effect on yam CSP. Taken along with Physics Division for testing the samples.
the count, they account for 75-85% variation in- References
dividually and 97% variation collectively. The ef- 1 Balasubramaniam P, BaIasubramaniam A & Govindrajan
fect of maturity on yam strength is marginal. R, A survey of the quality of yams- fine counts, SITRA
Yam strength values are about 13% higher in Research Report (SITRA, Coimbatore) 1987.
2 Lord E, Manual of cottonspinning - characteristics of raw
coarse counts and 7% in fine counts than those
cotton (The Textile Institute, Manchester) 1961, 11.
found earlier at SITRA in 1968. The improved 3 Subramaniam T A, Ganesh K & Bandyopadhyay S, J Text
technological efficiency of modem machinery per- Inst, 65(6) (1974) 307.
haps explains the better quality. 4 Dhawan K & Subramaniam T A, Proceedings, 23rd joint
technological conference of ATIRA, BTRA SITRA &
The optimum 1M for maximum strength has a NITRA(Nonhern India Textile Research Association,
linear relationship with 50% span length and fine- Ghaziabadi 1982, 132 .
ness. A very close fit is observed between the pre- 5 IyerJanaki K,Indian Text J, 94(10) (1984) 81.
dicted and actual values of optimum 1M 6 Iyer I K P,J Text Assoc, July 1986, 115.
(r= 0.95) with a mean deviation of only 0.071M. 7 Hunter Laurence, Melliand Textilber, 69( 4) (1988) 123.
8 Seshan K N & Ratnam T V, Proceedings, 10th joint tech-
The optimum 1M for maximum strength is nological conference of ATIRA, BTRA & SITRA (Ahmeda-
about 5% lower than that observed in an earlier bad Textile Industry's Research Association, Ahmedabad)
study. 1968, Section A, 2.1.
9 Seshan K N & Pillay K P R, Cotton assessment and appre-
The fall in strength from the level at optimum ciation, SITRA Mill Control Report No.4 (SITRA, Coim-
1M is fairly uniform at different levels of twist. batore) 1983.