Textile Research

http://trj.sagepub.com/
Journal

Relationship Between Fiber and Yarn Strength

Ning Pan, Tao Hua and Ylping Qiu
Textile Research Journal 2001 71: 960
DOI: 10.1177/004051750107101105

The online version of this article can be found at:

http://trj.sagepub.com/content/71/11/960

Published by:

http://www.sagepublications.com

Additional services and information for Textile Research Journal can be found at:

Email Alerts: http://trj.sagepub.com/cgi/alerts

Subscriptions: http://trj.sagepub.com/subscriptions

Reprints: http://www.sagepub.com/journalsReprints.nav

Permissions: http://www.sagepub.com/journalsPermissions.nav

Citations: http://trj.sagepub.com/content/71/11/960.refs.html

>> Version of Record - Nov 1, 2001

What is This?

Downloaded from trj.sagepub.com at LOUISIANA STATE UNIV on September 3, 2014

960 TEXTILE
RESEARCH
JOURNAL

Relationship Between Fiber and Yarn Strength

NINGPAN AND TAOHUA
Division of Textiles arid Clothirig, University of Califoniia, Davis, Califoniia 9.5616, U.S.A.

YIPINGQIU
College of Textiles, North Caroliria State Uriiversily, Raleigh, North Caroliria 27695, U.S.A.

ABSTRACT
A theoretical approach is adopted in this paper based on the existing literature in order
to present an explicit relationship between the strengths of fibers and yams. This provides
a relatively simple method for quick estimation of staple yam strength, while avoiding
some limitations of an empirical method. The CV values of the strengths are given as well.
The effects of important factors such as fiber variation, fiber length and thickness, yam
twist, and yam packing density are considered, and some are discussed in detail. The
results are verified by previously published data.

A fundamental and classical problem in textile mate- the fiber helical paths into his model. However, exclud-
rials science is the connection between strengths of fibers ing the effects of fiber interactions such as interfiber
and yams, and so it is significant both in theory and in friction and lateral constraint in his model seriously
practice to establish this relationship. limits his theory in terms of the applicability and accu-
A given large number of single fibers forms a fiber racy of the predictions.
bundle, and with twist, the bundle becomes a yam. How In this article, we discuss a direct relationship between
the fiber properties translate into yam properties thus the strengths of single fibers and yams based on existing
depends on, among other factors, the twist level. theories. We then compare the theoretical predictions
Because of variations in fiber strength, the breaking with the published data.
stress of a parallel fiber bundle is smaller than that of its
constituent fibers. On the other hand, predicting yam
strength is also different from predicting that of a parallel
Fragmentation Phenomenon and
fiber bundle, because in the latter case, the effect of fiber Critical Fiber Length 1,
interaction is negligible. Also, because fibers are all
We know [12, 191 that the in-sitir properties of the
parallel to the axis of the fiber bundle (the loading
constituent fibers in a yam deviate from those tested
diiection), the fiber obliquity effect is nonexistent. More- before the constituents are incorporated into the struc-
over, the strength prediction of a fibrous structure is ture. The yam structure and its tensile loading situation
unlike its modulus prediction, for the strength of a ma- will affect the fiber mechanical behavior in two ways, the
terial is not a volume average quantity but rather an first because fibers within the yam are under multi-axial
extremum quantity, dictated by the weakest cross section loading conditions of axial tension, lateral compression.
of the structure. This so-called weakest link theorem was This leads, as expected, to a different fiber behavior from
first elucidated by Peirce [13] in 1926 and has since been the uniaxial tension case. The second is due to the
investigated by several authors [8,181. The relationship fragmentation phenomenon, where under the constraint
between fiber strength and yam strength has also been of lateral compression, a fiber in the yam behaves as a
studied by numerous authors [ 5 , 9, 12, 14-17, 20, 211. chain of mechanically independent segments, each of
In 1978, Harlow and Phoenix [6] proposed the concept which possesses mechanical properties differing both
of the chain-of-bundles model of the strength of a fibrous from those of other segments of the same fiber owing to
structure to tackle the statistical nature of an individual the nonuniformity of the fiber and from those of the
filament’s strength, the size (length) effect on filament whole fiber due to the length effect. Because of this, a
strength, as well as the load sharing mechanism during fiber in the yam often breaks many times and, during
structure breakage. Phoenix [ 141 extended their method yam extension, fiber breakage continues until the length
to the analysis of twisted fiber bundles by incorporating of the fiber fragments reaches a minimum value where

Texfile Rex. 1. 71(1 I), 960-961 (2001) 0010-5 175/$2.00

Downloaded from trj.sagepub.com at LOUISIANA STATE UNIV on September 3, 2014

NOVEMBER
2001 961

load can no longer build up to its broken strength [lo, STRENGTH

STATISTICAL OF A PARALLEL
FIBERBUNDLE
191-the well known critical length 1,.
Let us consider a bundle made of a very large number
We must point out that the occurrence of the fragmen-
N of parallel Weibull-type fibers with equal length 1,.
tation process during yarn extension is not inevitable. It
Because it is a parallel bundle without any twist, we can
all depends on the factors such as fiber properties and
expect all fibers to have the same strain upon loading. It
yam structural parameters like fiber strength variations,
is evident that if all the fibers were of the same breaking
twist level, yarn diameter, and fiber length [19]. Also,
strain, the strength of the bundle would be equal to that
there has been clear evidence that the fragmentation
of its constituent fibers. However, since in reality there is
process occurs in thick filament yams [ 10, 191, but seem-
certain dispersion in the breaking strain among the fibers,
ingly no published experiments have yet shown the ex-
this bundle strength will hence approach, according to
istence of the process during fracturing of normal staple
Daniels [4],a normal distribution. Based on Daniels’
yams.
analysis, the expected value of the bundle strength ( a b )
is related to fiber properties as
FIBERSTRENGTH
VERSUSFIBER
LENGTH

On the other hand, Colemen [3], using the weakest-

link theory [ 131, showed that for Weibull-type fibers [23] and the SD value @b of the bundle strength is
of given length If, the mean or the expected value of fiber
strength (af) is

(4)
A comparison of Equations 1 and 3 shows that be-
cause of fiber property dispersion, the bundle strength
and the standard deviation (the SD value) is ( c b ) is lower than the fiber strength (c,).The difference
between the two will diminish when the fiber shape
parameter p 403, strong evidence that this discrepancy
between fiber and bundle strengths is caused by fiber
irregularities.

STATISTICAL OF A TWISTED
STRENGTH YARN
where a is the scale parameter and p is the shape Because of possible fragmentation during yam exten-
parameter, and both of them are independent of fiber sion, a yarn can be treated as a chain of fiber bundles of
length lp Equation 1 clearly shows that when the fiber length 1,. The expected strength of the yam (a,)is then
length, decreases, the statistical strength of the fiber in- obtained by replacing the fiber length lr in Equation 3
creases. with the critical length 1,. Furthermore, for a real yam,
The fiber shape parameter p represents the dispersion the fiber orientation and fiber packing density or volume
of fiber strength along a fiber length, and therefore is a fraction Vr have to be considered. Using the orientation
more interesting variable. A greater p value indicates a efficiency factor vq to reflect the effect of fiber obliquity,
small fiber strength variation or more uniform fibers. and the length efficiency factor q, to reflect the effect of
When p + 03, there will be no variation and the mean definite fiber length, the expected strength (a,) for a
fiber strength will then become independent of its length. staple yarn with normal twist can be expressed as [12]
In general, values of p between 2 and 4 correspond to
brittle fibers, whereas a value of 20 is appropriate for a
ductile material [21.
As stated above, during the yarn extension process, a
fiber works as a chain of segments with each segment Pan [ 121 demonstrated that for practical applications
being tensioned to break. Since the length of the seg- where yam twist exceeds the minimum value so that
ment, as defined by the critical length I,, decreases along q, 3 1, the fiber length effect can be ignored. We then
with the increasing lateral pressure, the in-sittr strength of have
the fiber segments will increase according to the fiber
length-strength connection.

Downloaded from trj.sagepub.com at LOUISIANA STATE UNIV on September 3, 2014

962 RESEARCH
TEXTILE JOURNAL

Statistics theory readily proves that the standard devi- Next, from Pan [l 11, we estimate that at the moment
ation for a twisted yarn 0, is related to @ b of a parallel of yam fracture, the fiber volume fraction V, = 0.6 and
bundle as the fiber orientation efficiency factor qq = 0.8.
First, Equation 8 shows again that the fiber strength is
o y = vjqq@b a (7) always greater than fiber bundle strength. We can plot 9
against /3 as shown in Figure 1, which reveals that, for a
When the yam surface helix angle (I = 0 so that qq= 1, we
given fiber type, the greater /3 value, meaning a more
will have a parallel loose bundle case, and with the fiber-
uniform fiber, leads to a small T value, or a higher fiber
volume fraction V' = 1, this value @,, will become @.,
strength translation efficiency, or a greater fiber bundle
strength.
RELATIONSHIPS BETWEEN FIBER,BUNDLE,
AND YARNSTRENGTHS

It is well recognized that the ratio between the ex-

pected strengths of the fiber and the fiber bundle is
defined as the so-called Coleman factor 9: 1.5

1.4

1.3
The factor 'P, and therefore the p value, specifies the
translation efficiency from fiber strength into bundle 1.2
strength. A greater q' or p value indicates a higher
discrepancy between fiber and bundle strength, i.e., a 10 20 30 40
P
higher ((mf)/(ab))value, or a lower translation effi-
ciency from fiber strength into bundle strength. FIGURE1. Effect of p on the ((u,)/(ub))
ratio.
The relationship between yam strength and fiber bun-
dle strength can be seen'from the ratio
Then we can plot the strength ratio ((m,)/(mb)) in
(9) Equation 9 against the length ratio of ( f f / l c )at three /3
levels, as shown in Figure 2. Note that in the figure, the
vertical axis is located at position (Zf/lc) = 1 , instead of
Obviously, the length ratio ( l f / l c ) represents the effect
the origin, on the horizontal axis. The figure indicates
with or without the fragmentation process. The ratio
that when there is no fragmentation process occuring
between the expected strengths of the yarn and its con-
during yam fracture, so that ( l f / l c ) = 2, the ratio of
stituent fibers can be derived from Equations 1, 6, and 8
((u,.)/(ab)) is below 0.6, a constant independent of p.
as
However, if the fragmentation process takes 'place, a

1.8

1.6
Experiments Data and Discussion
We have adopted experimental data from Sasser et al.
[20], and the details of measurements and the data can be
found in that paper. For discussion, we need to estimate
both the Weibull shape and scale parameters, which can
be done by using the maximum likelihood technique [ 11
based on the experimental data. During the process, we
have found the cotton fibers are brittle enough to be of a 20 40 60 80 100 le
Weibull form, and we have obtained the unbiased esti-
mates of a = 1.11 X lo-" and p = 6.143. ratio at three p levels.
2. Effects of ( + / I c ) on the (uJ)/(ub))
FIGURE

Downloaded from trj.sagepub.com at LOUISIANA STATE UNIV on September 3, 2014

NOVEMBER 2001 963

smaller p value, meaning a more brittle fiber, will pro- I. Prediction and expenmental results.
TABLE
duce a greater strength ratio ((u,.)/(ub)) or a stronger
yam. Note that yarn strength can be greater than fiber
bundle strength, i.e., ((cr,.)/(crb))> 1, if the fragmenta-
tion process takes place and the p value is small enough. Calculated 1.70 0.37 0.53 18.96
Experiment (ring) 1.99 0.37 0.73 16.50
Finally, we plot ((u,.)/(u,))of Equation 10 in Figure Experiment (rotor) 1.99 0.31 0.60 16.50
3. When the effect of lateral pressure is not high enough
to generate the fragmentation process, or if the fragmen-
tation effect is small, say, (Zf/lc) < 3 in Figure 3, yam
strength can be as low as around one-third of the fiber close to the experimental results, given the complex
strength depending on the p value. Also, the greater the structure of the yams. The calculated CV value using the
/3 value or the more uniform the fiber, the higher the yam predicted mean strength and SD value for the fiber is
strength. However, if the lateral pressure in the yam is so close to the experiment as well. We cannot predict the
high that the fragmentation process leads to (l,/lc) > 5 , SD or CV values for both fiber bundle and yarn, since we
then a more brittle fiber (smaller p) will produce a have no knowledge of the number of fibers N in both
stronger yam relative to fiber strength. In the extreme but cases to use Equations 4 and 7. Further, on the one hand,
unlikely situation where the lateral constraints on fibers twisting yams introduces more variability in the form of
are so high that the fragmentation process goes so far that random migration and random slack effects than existed
(Zfllc) becomes very large, such as in composite case in the bundle as suggested by Phoenix [ 141; on the other
[22], then yam strength could be, in theory, greater than hand, twisting may alleviate structure variability by bind-
fiber strength. ing fibers together to eliminate some otherwise weaker
spots. So as shown by Sasser et 01:s experimental data
[20], the yam CV value can be either smaller (in the case
of ring yam CV = 19.30) or greater (rotor yam CV
= 23.30) than the bundle CV value (19.60).
One key factor here is the length ratio (l,-/lc). In
loosely twisted yams or yams of small diameter and thus
low lateral pressure, the fragmentation process will not
occur, so that (l,/lc) = 2 and the translation efficiency of
fiber tenacity to yam tenacity will be low. On the other
hand, if the fiber-fiber cohesion is high enough, either
through high twisting or other means, including applying
size or even providing fiber-fiber chemical bonding, the
fragmentation process will take place, i.e. (l,/lc) 9 2,
0 20 40 60 80 100 Ic and the resultant yam strength will be significantly en-
hanced.
ratio at three p levels.
FIGURE3. Effects of (l,/l,) on the (uY)/(u,)) Note that because we derived the theoretical results
based on the ideal yam model as defined by Hearle et al.
[7], there would be some discrepancies as expected in
In comparison with the experimental results of Sasser comparison with the experimental results of cotton yarns.
et nl. [20], we construct Table I, where for various ratios However, the discrepancies are within the allowable
and CV values, respectively, we calculate the results range, and the general trends of both theoretical and
using the theoretical prediction and the experimental experimental results are in very good agreement, consid-
data. Since there are two kinds of yams in Sasser et al. ering the complexities of actual cotton yams. The theo-
[20], correspondingly there are two results for each ratio retical relationships can thus be used for future predic-
whenever applicable. Sasser et al. [20] reported that the tions.
average strengths of the ring and rotor spun yams are
only 37% and 31%, respectively, of the average fiber Conclusions
tenacity. These values can only be predicted when ( l f / l c )
= 2. That is, the fibers in the yarn only break once and The relationships between the strengths of fiber and
the fragmentation process is nonexistent. We have thus yarn can be predicted using the theoretical approach
determined the value for (f,/lc) for our calculation. From presented in this paper. The translation efficiency of fiber
the table, we see that the theoretical prediction is rather tenacity to yam tenacity is determined mainly by two

Downloaded from trj.sagepub.com at LOUISIANA STATE UNIV on September 3, 2014

964 TEXTILE
RESEARCH
JOURNAL

factors, i.e., the occurrence of the fragmentation process 10. Monego, C. J., and Backer, S., Tensile Rupture of Blended
and the fiber strength variations (p). If the fiber-fiber Yams, Textile Res. J. 38, 762 (1968).
cohesion in the yam is so low that the fragmentation 11. Pan, N., Development of a Constitutive Theory for Short
process does not take place, then the translation effi- Fiber Yams: Mechanics of Staple Yarn Without Slippage
ciency of fiber tenacity to yam tenacity is about one- Effect, Textile Res. J. 62, 749 (1992).
12. Pan, N., Prediction of Statistical Strengths of Twisted Fiber
third.
Structures, J. Mnfer. Sci. 28, 6107 (1993).
There is a critical level of the ( l f / l c ) ratio, below 13. Peirce, F. T., Tensile Tests for Cotton Yams, V: “The
which a brittle fiber will produce weaker yam, and above Weakest Link”-Theorems on the Strength of Long and of
which the opposite is true and a brittle fiber, i.e., the one Composite Specimens, J. Texfile Insf. 17, 355 (1926).
with smaller /3 value, will lead to a stronger yarn. The 14. Phoenix, S. L., Statistical Theory for the Strength of
experimental data adopted in this paper demonstrate that Twisted Fiber Bundles with Applications to Yams and
for similar cotton yams, the fragmentation effect can be Cables, Textile Res. J. 49, 407 (1979).
ignored so that the length ratio can be set as ( l f / l c ) = 2. 15. Pitt, R. E., and Phoenix, S. L., On Modeling the Statistical
With this information and our theoretical resuIts, which Strength of Yarns and Cables Under Localized Load-Shar-
are quite general, we can quickly estimate cotton yam ing Among Fibers, Textile Res. J. 51,408 (1981).
strength in practice. 16. Platt, M. M., Mechanics of Elastic Performance of Textile
Materials, Part 111: Some Aspects of Stress Analysis of
Textile Structures-Continuous Filament Yams, Textile
Literature Cited Res. J. 20, 1 (1950).
1. Bury, K. V., in “Statistical Models in Applied Science,” 17. Ramey, H. H., Jr., Lawson, R., and Worley, S., Jr., Rela-
John Wiley and Sons, NY, 1975. tionship of Cotton Fiber Properties to Yarn Tenacity, Tex-
2. Chou, T. W., “Microstructural Design of Fiber Compos- rile Res. J. 47, 685 (1977).
ites,” Cambridge University Press, Cambridge, U.K., 1992. 18. Realff, M. L., Seo, M., Boyce, M. C.. Schwarts. P., and
3. Colman, B. D., On the Strength of Classical Fibers and Backer. S., Mechanical Properties of Fabrics Woven from
Fiber Bundles, J. Meclr. Plrjs. Solids 7, 60 (1958). Yams Produced by Different Spinning Technologies: Yam
4. Daniels, H. E., The Statistical Theory of the Strength of Failure as a Function of Gauge Length, Textile Res. J. 61,
Bundles of Threads, Proc. R. SOC.A183,405 (1945). 517 (1991).
5. Hamburger, W. J., The Industrial Application of the Stress- 19. Realff, M. L., Pan, N., Seo, M., Boyce, M. C., Backer, S.,
Strain Relationship, J. Textile Insf. 40, 700 (1949). A Stochastic Simulation of the Failure Process and Ulti-
6. Harlow, D. G., and.Phoenix, S. L., The Chain-of-Bundle mate Strength of Blended Continuous Yams, Textile Res.
Probability Model for the Strength of Fibrous Materials, I: J. 70, 415 (2000).
Analysis and Conjectures. J. Contpos. Maferids 12, 195 20. Sasser, P., Shofner, C. K., Chu, Y. T., Shofner, F. M.,and
(1978). Townes, M. G., Interpretations of Single Fiber, Bundle,
7. Hearle, J. W. S., Grosberg, P., and Backer, S., “Structural and Yarn Tenacity Data, Textile Res. J. 61, 681 (1991).
Mechanics of Yarns and Fabrics,” vol. 1, Wiley-Inter- 21. Sullivan, R. R., A Theoretical Approach to the Problem of
science, NY, 1969. Yam Strength, J. Appl. Phjs. 13, 157 (1942).
8. Hussain, G. F. S., Nachane, R. P., Kishna Iyer, K. R., and 22. Watson, A. S., and Smith, R. L., An Examination of
Srinathan, B., Weak-Link Effect on Textile Properties of Statistical Theories for Fibrous Materials in the Light of
Cotton Yams, Textile Res. J. 60, 69 (1990). Experimental Data, J. Muter. Sci. 20, 3260 (1985).
9. Kemp, A., and Owen, J. D., The Strength and Behavior of 23. Weibull, W., A Statistical Distribution Function of Wide
NylordCotton Blended Yams Undergoing Strain, J. Text. Application, J. Appl. Mech. 18, 293 (1951).
Itrsr. 46, T-684 (1955). dlanuscript received 0c:ober 28. 2000; accepted Februaty 5, 2001.

Downloaded from trj.sagepub.com at LOUISIANA STATE UNIV on September 3, 2014