175
Twist Structure of Plied Yarns
*
E. R. Schwarz
THE SPINNING OF YARN is a major ob- found effect upon the load-extension behavior and
jective of the textile industry. The details of upon the ultimate strength of the ply.
yarn structure should therefore be of major impor- , c
It is necessary only to realize how seriously the
tance to manufacturer and research worker alike. tensile strength of a plied yarn may be altered by
Yet, although yarns have been spun for centuries, changing simply the number of turns per inch in the
it has been only within the past few years that the single yarns composing it by even small amounts.
mathematics of yarn structure has been seriously A change of much less than one turn per inch in the
studied. Naturally, since twist is essential to yarn singles can throw a plied yarn badly out of balance.
Such facts are relatively widely appreciated in both
structure, the measurement of twist has occupied
the attention of many of the investigators. the laboratory and the mill. What is not at all
Devices have been perfected for untwisting yarns widely appreciated, although it was pointed out by
in such fashion that the number of turns per unit H. J. Woods ~1~, is the fact that the twist in a yarn
may be considerably altered simply by distorting
length could be determined conveniently. It was
realized almost at once that when the final yarn was the axis of the yarn into a three-dimensional struc-
ture and without the rotation of either end of the
untwisted, each of the elements composing it was
also rotated about its axis. Hence, the twist in the yarn. This change in twist was called &dquo;tortuosity&dquo;
single yarn is not the same when it lies in the ply as by Woods and was denoted by the Greek letter r,
it was before plying. whereas the twist as ordinarily measured was de-
The usual test, as presently conducted, recog- noted by the letter &dquo;T.&dquo;
nizes the fact that at least three conditions of twist A moment’s consideration will show that when a
are of interest-namely, (1) turns per inch in the plied yarn is made, the majority-or all-of the
plied yarn, (2) turns per inch in the single yarn as it single yarns composing it will be distorted into
was before plying, and (3) turns per inch in the helices, and because of this fact tortuosity must be
taken into consideration. It is not necessarily true
single yarn as it lies in the ply. that the helices developed will be strictly circular,
It should have been obvious that if the properties
of a plied yarn were to be studied intelligently, they but if they are considered to be so, the mathematics
can be considerably simplified without undue
ought to be related to the yarn as it existed during
the tests. Turns per inch in the single as the single sacrifice of precision.
lies in the ply, therefore, become important, for Work by Chow C2] is of importance in the study
of the geometry of yarn structure. It is of particu-
certainly the single yarn would not be expected to
behave in the plied state as it would just as it left lar interest to note that the formula derived by
the spinning frame. Furthermore, the single, when Woods and further studied by Schwarz ~3~ is a
it becomes a part of the ply, in many cases takes the special case of Chow’s rigorously derived general
form of a helix. In the laboratory, however, the equation. It should also be noted that the general
properties of single yarns are determined on straight equation emphasizes the necessity for careful set-
strands or on loops but not on strands whose axes ting of the yarn on the microscope stage whenever
are helical. It becomes difficult to relate single- microanalysis of twist structure is attempted,
yarn properties to the properties of a plied yarn, for
Furthermore, the use of the geometry of the torus
in the plied yarn stresses are distributed in complex is believed to be important, particularly when the
torus is so distorted that it forms a helix about a
fashion, both in the single yarns and among the
fibers composing them. It is very probable that helical axis. Again, the use of trigonometry and of
concentrations of stress exist which will have pro- calculus makes possible the mathematical analysis
and once more demonstrates the growing need for
*
Professor of Textile Technology and in charge, Textile the study of these subjects by students in the field
Division, Massachusetts Institute of Technology. of textile technology.
Downloaded from [Link] at UNIV OF ILLINOIS URBANA on March 10, 2015
176
For the purpose of the present analysis, three solidforming the locus of the axes of the outermost
mutually perpendicular axes of reference may be yarns in a ply or of fibers in a single yarn. The
taken in the directions xi, x2, and x3, respectively, transverse, in turn, is a line normal to the generator
with X3 representing the direction assumed by the and joining it with the axis of the yarn under con-
axis of the plied yarn. A point in space-as on the sideration. Twist is then defined as the rotation of
surface of one of the outermost singles in the ply- such a transverse per unit length of the axis, or
may then be located with respect to this set of
reference axes and so also can the direction of an
infinitesimal distance along the fiber (ds) or along
the single yarn (dS) ...
The angle Q is taken to be the angle between the The usual concepts of normal, n (a transverse),
axes of the single and the plied yarns, while 4, is
and binormal, b (a perpendicular to both the normal
considered to be the angle between the fiber and and the tangent), will be employed, together with
the single yarn as the single lies in the ply, which in- the necessary direction cosines referred to axes xi,
cludes the effects of tortuosity. The very useful x2, and x3. The angle 0 is the angle of rotation of a
transverse joining the ply axis (x3) with a single-
concept of transverse and generator as defined by
Woods will be employed. Thus, a generator is any yarn generator. Similarly, 0 is defined as the angle
convenient line on the surface of a yarn or of the of rotation of a transverse joining the axis of the
single yarn with a fiber generator (Figures 1 and 2).
It should be noted that single-yarn helix radius
(a) and ply-yarn helix radius (r), rather than the
usual yarn diameters, are used until the final steps
of the derivation.
While Chow gives the preliminary mathematics
concerning the definition and properties of a cylin-
drical helix, a circular helix, the tangent to a circu-
FIG. 1. FIG. 2.
Downloaded from [Link] at UNIV OF ILLINOIS URBANA on March 10, 2015
177
lar helix, the principal normal and the binormal of n-axis and varies parametrically. Thus, the vector
the circular helix, and the helical torus in full, the to the point referred to the same origin as that for
present treatment will note only those portions the circular helix is defined by
which aid in clarifying the study.
The angle 4, is of primary interest because it is
the angle between the direction taken at any in- where 5B is the unit vector in the direction of xi.
stant by a fiber and the axis of the single yarn, or The vector of the point A is rO cot Q.
between the elementary lengths ds and dS.
For a circular helix (see Figure 1) with equations:
The equation of a helical torus is:
and
or
which leads to
the differential curve ds assumes the direction of
the unit tangent to the circular helix, the direction
cosines of which are
where 0 =
X0.
Differentiating:
and
The direction cosines of dS are
Squaring each differential and summing:
Now a helical torus involves a surface developed
around a circular helix so that a section in the nor-
mal plane of the circular helix is a circle with radius
a, where a is less than r, the radius of the helix. At which simplifies to
any point on a circular helix the unit vectors of the
principal normal, binormal, and tangent constitute
a right-handed system of rectangular coordinates
corresponding to xi, X2, and x3, respectively, with
the origin at point A (Figure 2) on the helix. Now
Thus,
if xi be the coordinates of the point B on the helical
torus, the vector connecting the points A and B will
assume a form
where cb is the angle between the vector AB and the
Downloaded from [Link] at UNIV OF ILLINOIS URBANA on March 10, 2015
178
or
Substituting equation (6) in equation (5) :
and of where P
But, in terms angle P, =
~r/2 - Q,
so that Now, if the microscope is set so that the yarn being
measured is properly placed, then Xe 7r and =
But, by definition,
Substituting equation (9) in equation (8):
Now, if T’ and T&dquo; are the turns per unit length in
the ply and in the single as it lies in the ply,
respectively, then or
and The equation for the tortuosity, T, is
Since
And if
then
then
which is the formula obtained by Woods.
Many plied yarns are made up so that any tend-
ency for the ply to untwist is checked by the re-
sistance of the single yarns to being twisted. Such
a yarn is said to be balanced.
Since equations (4) and (2) when combined give
Substituting equation (4) in equation (2) :
and when X0 =
7r,
For a circular helix:
and
The single fibers are inclined to the axis of the ply
so that at an angle equal to (~ + 0), which examination of
*
For derivation see Appendix to this paper.
Downloaded from [Link] at UNIV OF ILLINOIS URBANA on March 10, 2015
179
balanced yarns shows must equal zero if the yarn and the tortuosity of the curve is given by
is balanced. Thus, .
and
Solving for T&dquo;/T’ gives
For a circular helix,
Also, t
where Dip =
ply diameter, Ds =
single diameter, I
and D fiber diameter. The significance of the
=
minus sign is simply that T&dquo; and T’ must be in
opposite directions.
Particular attention should be paid to the fact
that this simple relationship is obtained only when
tortuosity in the single yarn has been taken into
consideration.
Appendix
Derivation of ~c (the Tortuosity of the Curve)
The derivation is based upon the direction cosines
of the unit vector of the binormal * of a twisted Thus,
curve.
In general,
where k is the curvature :
Literature Cited
1. Woods, H. J., J. Text. Inst. 24, T317 (Sept. 1933);
J. Text. Sci. 4, 33 (June 1931).
2. Chow, Tseng-yeh, unpublished Master’s thesis,
M.I.T., Textile Division (June 1948).
3. Schwarz, E. R., J. Text. Inst. 27, T237 (Oct. 1936).
*
The binormal is perpendicular to both the tangent and the
normal to the curve at the given point. (Manuscript received September 15, 1949.)
Downloaded from [Link] at UNIV OF ILLINOIS URBANA on March 10, 2015
