Dr. Yehia E. El Mogahzy, Professor of Textile Engineering, Auburn University
E-mail: yehiae@eng.auburn.edu, Internet: http://www.eng.auburn.edu/~yehiae/

Introduction

As we rapidly enter the era of Internet, information superhighway, and unified currencies, one inevitable situation will be the standardization of the world price of products through the establishment of universal quality standards. When this situation becomes a complete reality, quality will once again drive the consumer demand for any product. In other words, a standardization of product price would certainly mean a disastrous situation for companies producing quality levels below the average universal standards. Certainly, the global yarn market is no exception. In fact, all innovations in the spinning process made in the 1980’s and the 1990’s have been aimed at maintaining a worldwide cost balance of yarn manufacturing through increasing productivity and reducing labor cost (automation & transportation).

In the 21^st century, we expect that only companies that produce yarn compatible with the worldwide standard price will survive. This means that the race for economical compatibility will continue and perhaps at a more forceful level. The future impact of this race on yarn quality remains to be seen. However, one would hope that the outcome will be better than today’s quality situation.

Perhaps, the best way to describe today’s yarn quality is to refer to the recent Uster Statistics published in 1997. A summary of these statistics for 100% cotton yarns is presented in Appendix I (Table I.1 through Table I.7). Examination of these statistics reveals the following points:

• For a given yarn type, the difference between the best and the worst-case yarn quality is quite substantial (30%-100%).
• In both ring and rotor spun yarns, the highest difference is shown to be in the variation of yarn count (50%-80%), and in the total number of yarn imperfections (68%-95%); the lowest difference is in the Uster C.V% irregularity (18%-33%).
• The difference between the best and the worst case scenario of yarn quality tends to be higher in the coarse range yarn than in the medium or fine range yarn count.

The above points clearly indicate that the race for economical compatibility has been simultaneously associated with a significant gap of yarn quality between textile companies around the world. Obviously, the new innovations introduced in the 80’s and the 90’s could not be the cause of this persistent quality gap. In contrast, these innovations have been introduced to minimize human involvement and to produce consistent yarn quality with its ever-developing automation, transportation, process control, and artificial intelligence. The two main factors contributing to this quality gap, in our opinion, are:

1. The lack of complete integration of the fiber quality aspect in the process of yarn manufacturing
2. The Difficulty of handling the variability in fiber properties

The first factor is directly related to the process of cotton fiber selection, and the second is related to cotton fiber blending. The Engineered Fiber Selection (EFS^®) system developed by Cotton Incorporated provides the tools and the algorithms that allow computerized bale management, and efficient implementation of fiber selection and blending strategies. However, the user of the EFS^® system should realize that the key to a successful EFS^® implementation depends on the extent of understanding the spinning system utilized and the level of yarn quality desired by the mill. A successful EFS^® implementation is the one that is based on integrated efforts from the fiber sector to the yarn or the fabric sector. This is due to the fact that the impact of fiber selection and blending can be witnessed in all areas of the textile process and certainly in the quality of the end product. In this paper, we will discuss these two aspects with the hope that the information presented will re-stimulate our thinking of current practice and future developments for optimum yarn quality/cost ratio.  

Cotton Fiber Selection

The primary objective of a fiber selection system should be to provide fibers that exhibit criteria suitable for the spinning technique utilized by the textile mill and the desirable level of yarn and fabric quality. Cotton Incorporated EFS^® system consists of three basic elements that work together to achieve this primary objective:

The first two elements of the EFS^® system involve systematic procedures that make use of the frequency distributions of HVI data of the cotton bale population to establish a category/group scheme suitable for the textile process in question. From the category/group scheme, cotton bales are selected for the cotton mix (or the bale laydown) using a number of blending algorithms that are carefully developed to suite different mill blending strategies. The main outcome of these two elements is a consistent fiber profile of the cotton mix; that is a minimum between-mix variation. For a more detailed reference on the category/group scheme, the reader may refer to two papers written earlier by the present author [1,2].

Our experience with the implementation process of the EFS^® system indicates that most companies mainly implement the first two elements, and that little or no effort is made to implement the third element of the EFS^® system. The assumption here is that a consistent fiber profile fed to the textile process will produce a consistent yarn quality. This assumption is largely true, particularly under optimum processing conditions and when a small range of products is being produced. For companies producing a wide range of yarn types and counts, fiber selection for control of yarn quality and cost becomes an absolute necessity.

Fiber Selection for Control of Yarn Quality

When a company is implementing the EFS^® system for control of yarn quality, the basic question becomes:

What are the fiber properties suitable for the spinning technique used and the level of yarn quality desired?

Attempts to answer this question have been made by many investigators including the present author. These attempts normally provide general answers that can only be made specific through actual implementation in the textile mill. Over the years, we have emphasized through statistical analysis of fiber and yarn data that different spinning systems are associated with different orders of importance of fiber attributes. The work reported by Deussen of Schlafhorst [3] provides lists of fiber attributes in the order of their importance with respect to different spinning systems. Despite these efforts, one can still see textile mills producing a wide range of yarn counts on different spinning techniques using the same cotton mix (same values of fiber attributes). Although this type of practice may seem logistically correct, it is neither quality satisfactory nor cost efficient. In fact, the quality gap shown by the Uster Statistics is largely attributed to this practice.

Different spinning techniques will require different fiber selection strategies. A wide range of yarn count and twist (hard vs. soft) within the same spinning system will require different fiber selection strategies. Perhaps, the best way to emphasize this point is to briefly review the principle of different spinning systems from the fiber viewpoint.

In today’s technology, there are three main spinning systems used by the cotton textile industry: (1) the conventional ring-spinning, (2) rotor (or open-end) spinning, and (3) air-jet spinning. These three systems operate based on the general principle described in Figure 1. As can be seen in this Figure, any spinning system consists of three basic operations: (a) drafting to reduce the size of the input material, (b) consolidation to provide the necessary coherence between fibers, and (c) winding to form the yarn package.

 The general principle described in Figure 1 indicates that any spinning system will require fibers that are:

• sufficiently flexible (to accommodate the continuous arrangement and rearrangement of fibers during drafting, consolidation, and winding),
• of high length/diameter ratio (to permit flexibility, effective consolidation and interfiber coherence), and
• of optimum surface adhesion (not too slippery to allow fiber control and not too clingy to allow smooth drafting)

Accordingly, there are three basic fiber attributes that are required to make a yarn on any spinning system:

 Fortunately, the cotton fiber generally exhibits acceptable levels of these three attributes. It has reasonably good flexibility or ease to deform under tension, bending, and torsion. More importantly, cotton flexibility is inherently optimum so that it offers minimum difficulty during processing, yet it gives the cotton fabric its familiar desirable hand and feeling. Although cotton may seem shorter than many other natural fibers, it has a length/diameter ratio in the order of thousands; this is certainly enough to accommodate processing needs. Cotton fiber has a surface friction that is inherently optimum, and certainly uncontested by any other fiber surface (natural or synthetic.) This optimum friction is naturally induced by the natural wax (content and distribution) and the unique convoluted surface morphology.

Concerns related to the basic fiber attributes may arise, however, when cotton fiber is blended with other types of fibers (e.g. polyester). In this situation, the two fiber types may represent a truly odd couple about to be combined. On average, polyester has more than twice the flexibility of cotton fiber under tension, yet cotton has more flexibility under bending (24%) and also under torsion (65%) than polyester. The aspect ratio of the two fibers can be made equal; however, the increasing trend toward the use of very fine polyester fiber (e.g. microdenier polyester of 0.9 denier or less) creates compatibility problems. The surface friction of the two types of fiber is never the same despite the millions of dollars invested in the spin finish of polyester.

The lack of basic compatibility between cotton and polyester fibers may not be directly reflected on the average values of the blend yarn which will tend to exhibit higher strength and elongation and lower hairiness than those of 100% cotton or of lower percentage cotton. It will, however, be reflected in all variability measures of the blended yarn (C.V% of count, Uster C.V%, C.V% strength, C.V% elongation, and yarn imperfections). This point was clearly demonstrated in Uster Statistics 97 as shown in Appendix II. In these statistics, variability measures were either increased or remain unchanged as the polyester content increases in the polyester/cotton blend. This observation was consistent in all types of spinning (ring, rotor, and air-jet). Traditionally, it has been claimed that the increase of the percentage of synthetic fibers in the cotton/synthetic blend will always reduce yarn irregularity.

It is our opinion that the EFS^® system can be effectively used in the area of cotton/polyester blend. This requires developing category/group schemes of cotton fibers that are compatible with those of polyester fibers. The EFS^® system has the powerful tools that permit such implementation. The effort, however, has to be made by the textile mill.

In earlier discussion, we emphasized the point that different spinning techniques will require different fiber selection strategies. This point can be clearly demonstrated through revisiting the principle of each spinning technique and examining the fiber/machine interaction. In the following sections, we will discuss ring and rotor spinning in the context of fiber attributes.

Fiber Selection for Ring Spinning

Ring spinning is the oldest type of spinning techniques available today. Thus, it has been continuously perfected since its initial development in the 19^th century. Furthermore, the introduction of other types of spinning in the 20^th century has resulted in additional developments and innovative designs in ring spinning to keep pace with the high productivity of the new systems. The true market power of ring spinning lies in its unsurpassed yarn quality and in its diversity. It is true that new spinning techniques can produce yarn at more than 6 times the linear production rate of ring spinning. However, ring spinning is the only system that can produce yarn at virtually any count from 4’s to 240’s and of both soft and hard twist. This point may be the primary reason for the survival of ring spinning particularly in an era in which product-range flexibility has become a significant economical plus. We must point out, however, that such diversity is not a result of the spinning design only but also (and often of more importance) a result of the art of fiber selection.

Principle of Ring Spinning

The general principle of ring spinning is shown in Figure 2.a. A fiber strand called "roving" is fed to the drafting system of the ring-spinning machine. The number of fibers in the cross-section of this strand typically ranges from 3000 to 4000 fibers. The drafting rolls reduce this number down to the number of fibers per yarn cross-section (typically, from 70 fibers for fine yarn to 700 fibers for coarse yarn). This reduction in the number of fibers is achieved mechanically by the speed ratio between the front and the back draft rollers, which results in creating a shift between fibers as they slide against each other in the drafting zone.

The fibers being delivered at the nip of the front roller form a triangle called the "spinning triangle" (Figure 2.a). The bottom end of this triangle represents the point at which fibers are consolidated into a yarn. The consolidation mechanism in ring spinning is twisting. Twist is inserted to the fibers by a traveler rotating around a ring. Each revolution of the traveler inserts one turn of twist into the yarn. The driving mechanism of the traveler is the spindle that carries the yarn bobbin. The amount of twist inserted in the yarn is determined by ratio between the rotational traveler speed and the linear speed of front roll. The fibers being twisted form a balloon shape resulting from the distribution of tension components generated by winding, and yarn/traveler contact. The spindle carrying the yarn bobbin rotates faster than the traveler and drags the traveler behind it. This causes yarn to wind around the bobbin. A building mechanism is used to move the ring vertically so that the yarn can be wound along the bobbin length.

 Fiber/Machine Interaction in Ring Spinning

As indicated above, fibers fed to the ring-spinning machine are in the form of a fiber strand called roving. This fiber strand typically has cotton count equal one (or 3000-4000 fibers/cross-section). Lighter roving can be made from longer and finer fibers. The roving is slightly twisted to allow its handling (winding and unwinding). The level of twist in the roving is typically less than one twist multiplier (twist multiplier = twist per inch)/Ö Ne). Normally, cotton fibers require a higher twist than synthetic fibers, and coarse/short fibers will also need higher twist because they have higher resistance (stiffer) to twisting than fine/long fibers.

The level of twist in the roving is a critical parameter in ring spinning. An optimum roving twist should be selected on the ground that a roving of high twist and light weight can lead to less yarn hairiness, but too high a twist in the roving can impair the drafting process in ring spinning leading to yarn imperfections [4]. In the context of fiber attributes, fiber length and fiber fineness represent determining factors for optimum roving twist.

As shown in Figure 2.a., the drafting system consists of drafting zones each containing two pairs of drafting rolls. The setting between drafting rolls (the length of the drafting zone) is a critical spinning parameter. Too wide a setting (greater than the fiber length) will result in fibers floating in the drafting zone, and too narrow a setting (smaller than the fiber length) will result in fiber breakage. In both cases, the outcome will be yarn defects (thick and thin places). The draft settings are normally determined on the basis of staple fiber length; they are typically slightly larger than the staple length. The bias of settings to the longer length is perhaps a result of the fact that fiber breakage introduces more serious problems than floating fibers (which can be carried by long fibers). Since draft settings are normally fixed, the real burden is on the fibers to accommodate these settings. In this regard, the important criterion is fiber length uniformity; a high variation in fiber length would result in excessive floating fibers and fiber fragments. This will impair both the uniformity and the strength of the yarn. Obviously, a high percent of short fiber content will adversely influence the length uniformity.

Fibers flowing through the drafting system of ring spinning system have to slide smoothly against each other and this requires minimum resistance to the drafting action. Normally, cotton fibers can easily handle themselves in the drafting zone provided that they can be released from the input fiber strand (roving). In this regard, the critical drafting zone is the back zone or the so-called break draft. Normally, the draft ratio in this zone is very small (slightly higher than one) because of the flood of fibers being fed. Fibers must not resist the drafting process; they should slide against one another smoothly. On the other hand, fibers can not be too slippery to maintain the integrity of the fiber flow. Thus, an optimum fiber-to-fiber friction is required.

During drafting, fibers must not lap around the drafting rolls. The chance of lap-up is increased when the number of fibers increases and when high-speed drafting is used. Thus, this problem is probably more pronounced when coarse yarns are being produced. The important fiber criteria here is surface finish in case of synthetic fibers and stickiness in case of cotton fibers.

Upon drafting and just before being twisted into the yarn, fibers form a geometrical zone called the "spinning triangle". In this triangle, fibers are subject to tension generated at the consolidation zone and propagated upward to the triangle zone. Different fibers in the spinning triangle will have different tensions depending on their position in the triangle and on their length with respect to the height of the triangle. Fibers in the center of the triangle are usually slack and those in the outer layers are under maximum tension. When fibers are released from the nip of the front roller, those exhibiting high tension tend to move toward the center displacing the initially central fibers to the outer layers. This phenomenon is called "fiber migration" and its main effect is to enhance fiber cross-linking, and consequently yarn strength.

The dimensions of the spinning triangle (width and height) determine the extent of fiber/machine interaction in this critical zone. Approximately, 80% of yarn end breaks occurs in the spinning triangle. The width of the triangle represents the width of the fiber bundle nipped by the front roller, and the height represents the length from the nip of the front roller to the twisting point. Normally, the width of the triangle is a function of the amount of draft or the total number of fibers delivered (coarser yarns will result in a triangle of larger width), and the pressure on the front roll. The height of the triangle, on the other hand, is quite sensitive to the spinning tension.

If one had the opportunity to magnify the spinning triangle and witness the fiber behavior, one will find that fibers in the triangle are in a continuous battle with the spinning tension (the force pulling the triangle down). In fact, the theory suggests that an end break in ring spinning is often a result of a momentarily increase in spinning tension coincided with a momentarily drop in the strength of the fibers in the triangle. This situation resembles a dynamic tensile test in which fibers are gripped at one end (the nip of the front roller) and being pulled from the other end (the twisting point) by the spinning tension. Although spinning tension is set in such a way that its average will be lower than the average strength of the fibers in the triangle, tension peaks can occur. Furthermore, if a cluster of fibers of unfavorable strength and/or elongation happens to enter the spinning triangle, it will have a weak resistance to the spinning tension, and the yarn will break. In this regard, the most important fiber properties are the variations in fiber strength and fiber elongation. In addition, hard trash particles entering the triangle can act as obstructing elements to the force balance process.

Beyond the spinning triangle, fibers are largely formed into a yarn. Under normal processing parameters (e.g. proper traveler weight and traveler condition), the extent of any further interaction between this yarn and the spinning elements (e.g. yarn/traveler friction) will depend on how fibers have performed in the drafting zone and in the spinning triangle. For example, the extent of yarn hairiness caused by the abrasion between the yarn and the traveler will depend on the surface integrity of the yarn entering the ring/traveler zone. This, in turn, depends on the extent of short fibers in the yarn, and the fiber compactness.

In light of the above discussion, one can develop a list of fiber properties in ring spinning according to the order of their importance. We suggest the following list:

The extent to which these attributes should be considered in fiber selection and blending will largely depend on the quality and the fineness of the yarn to be produced. Obviously, for very fine count yarns, constraints must be imposed on the variation in fiber strength and elongation not only between cotton mixes but also within the mix. Fiber length and length uniformity should also be controlled. In the absence of standard measures of fiber friction, it will be safer to minimize the number of cotton varieties (or sources) in the mix. This is due to the fact that cottons of different varieties and different regions are likely to exhibit significant differences in their wax content and surface morphology. Fiber fineness is a basic criteria for fiber selection (regardless the spinning system) because of its connection with the mix proportion control. Previous work by the present author set values of these fiber attributes that are suitable for different yarn counts [5]. A list of these values is outside the scope of this paper.

Fiber Selection for Rotor Spinning

Ever since its commercial introduction in 1969, rotor spinning has never ceased to develop. The small rotor, which is the key element in rotor spinning, has gradually climbed from a rotational speed of around 30,000-rpm in the initial machine to 150,000 rpm in the present machine. The designer of the rotor-spinning system sees no mechanical limitation to further increases; only the fact that it has to operate with fiber limits its further progress.

Rotor spinning was initially designed with two main objectives in mind: (1) to provide more economical spinning system than the conventional ring-spinning through higher productivity, and (2) to produce a yarn of a quality level that matches or surpasses that of the conventional ring-spinning. The first objective has been fully accomplished. Today’s rotor spinning machine has a linear production rate exceeding the 200 m/min (compared to a maximum of about 40 m/min in ring spinning). Using rotor-spinning means elimination of roving, since it can take a drawn sliver directly, and elimination of after-spinning winding due to the large yarn package and the largely defect-free yarn produced on the system.

Accomplishment of the second objective is still in the make facing some obstacles and limitations. Perhaps, the biggest obstacle facing today’s rotor spinning is the fact that it is limited to coarse/medium yarn count. If one can help overlooking the traditional competition between rotor and ring spinning, one will find that rotor spinning is most suitable (from both economical and quality viewpoints) for coarse to medium yarns (5’s-35’s) while ring spinning excels in the medium to fine counts. (> 35’s).

Principle of Rotor Spinning

The general principle of rotor spinning is shown in Figure 2.b. The input fiber strand is a drawn sliver. A sliver may have more than 20,000 fibers in its cross-section. This means that a yarn of 100 fibers per cross-section will require a total draft of 200. This amount of draft is substantially higher than that of ring spinning. Drafting in rotor spinning is accomplished using a comber roll (mechanical draft) which opens the input sliver followed by an air stream (air draft). These two operations produce an amount of draft that is high enough to reduce the 20,000 fibers entering the comber roll down to few fibers (5-10 fibers). In order to produce a yarn of about 100 fibers per cross-section, the groups of few fibers emerging from the air duct are deposited on the internal wall of the rotor and a fiber ring is formed inside the rotor. The total draft in rotor spinning is, therefore a combination of true draft from the feed roll to the rotor (in the order of thousands) and a condensation to accumulate the fiber groups into a fiber ring inside the rotor. The total draft ratio is the ratio between the delivery or the take-up speed and the feed roll speed. This should approximately amount to the ratio between the number of fibers in the sliver cross-section and the number of fibers in the yarn cross-section.

Consolidation in rotor spinning is achieved by mechanical twisting. The torque generating the twist in the yarn is applied by the rotation of the rotor with respect to the point of the yarn contacting the rotor navel. The amount of twist (turns per inch) is determined by the ratio between the rotor speed (rpm) and the take up speed (inch/min). Every turn of the rotor produces a turn of twist, and a removal of a length of yarn of 1/tpi inches.

The winding operation in rotor spinning is completely separate from the drafting and the twisting operations. The only condition here is that the yarn is taken up at a constant rate. This separation between winding and twisting allows the formation of larger yarn packages than those in ring spinning.

 Fiber/Machine Interaction in Rotor Spinning

As indicated above, the sliver of some 20,000 fibers per cross-section is drafted using a combination of mechanical and air draft. Obviously, a fiber strand that has been carefully prepared by carding and drawing to straighten the fibers will find it unpleasant to be treated by a toothed opening roll as it enters the system. This major fiber entanglement after a long journey of straightening and parallelization provides the first test to the fibers in the rotor spinning system. The comber roll drafts fibers by detaching a fiber beard presented by a feed roll and passing them into the rotor at much higher speed than the advance of the beard. Fortunately, cotton fibers are flexible and tough enough to withstand the comber roll action. Normally, a wire-wound clothing is recommended for cotton and cotton blends where pinned combing rolls are suitable for fragile fibers such as acrylic and rayon [4].

The extent of the opening action imposed by the combing roll will depend on the extent of fiber length. As the fiber length increases, the force acting on the fiber beard increases significantly. This can result in fiber damage. Thus, a moderate fiber length is required for rotor spinning. In fact, long fibers such as pima cotton or Egyptian cotton may suffer waste of fiber fragments if used on the rotor spinning system. Comber noils produced from combing these long staple fibers are more suitable.

As the opened fibers flow around with the combing roll, friction between the fibers and the comber roll metal chamber results in a fiber velocity lower than the surface speed of the combing roll. Those fibers are normally in a disoriented shape. In this regard, fiber attributes such as fiber resilience, fiber/metal friction, crimp, stickiness, and surface finish are of keen importance. The tendency to increase the combing roll speed makes these fiber properties even more critical. This increase is often associated with high yarn hairiness and yarn imperfections.

Although the primary role of the combing roll is to open the fibers, it can also act as a cleaning unit by separating trash particles from cotton. Obviously, this additional function can easily overstress the combing roll making it wear rapidly. It is important, therefore, that the input sliver exhibit a great level of cleanliness. A maximum trash level of 0.1% is typically recommended by the machine maker. In addition, fine trash and dust content can accumulate in the rotor groove leading to yarn defects and end breakage.

Fibers coming out of the comb roller are airborne through an air duct. This zone of draft is of a special significance because of its impact on fiber orientation. Since laminar airflow is hardly a reality, fibers are likely to suffer turbulence as they flow through the air duct adding more disorientation. This factor partially contributes to the weakness of rotor-spun yarn. Long fibers are more vulnerable to air stream disturbance than medium or short fibers. In order to minimize fiber disorientation, the airflow in the duct should have a velocity exceeding that of the surface speed of the opening roll. Investigators suggested speed ratios ranging from 1.5 to 4. To obtain such a fast airflow, the inside of the rotor is run at a vacuum which may be achieved by designing the rotor with radial holes to allow the rotor to generate its own vacuum (self-pumping effect). Alternatively, an external pump can be used as in most modern machines.

Another approach to minimize fiber disorientation in the air duct is by designing it in a tapered shape toward the rotor to allow acceleration of the fibers as they approach the rotor inside surface. This action may also straighten the leading fiber hooks coming out of the opening roll. Fibers emerging from the air duct come into contact with the rotor inside surface, which is typically faster than the fibers. This also assist in straightening the fibers disoriented in the previous zones.

The mass flow per unit time of fibers in the rotor spinning system, particularly from the air duct to the take-up zone provides an interesting insight into the contribution of fibers to the quality of rotor-spun yarn. The product of the fiber mass and fiber velocity can determine this quantity. For a stable process, this mass flow must exhibit a continuity that can be determined by the following simple mass-flow equation:

This above equation indicates that the ratio between the number of fibers in the yarn cross-section (n_fy ) and the number of fibers in the air duct (n_fd ) is governed by the ratio of the fiber velocity in the air duct (V_fd ) and the yarn velocity (V_fy ). The equation summarizes an important phenomenon that is unique to rotor spinning; the doubling effect. As indicated earlier, the effect of the air draft is to reduce the fiber strand down to few fibers (2-10 fibers). These fibers are then landed into the inside surface of the rotor as it takes many layers of fiber to make up sufficient number of fibers per yarn cross-section. As successive layers of fibers are laid into the inside surface of the rotor, a doubling action occurs. This action tends to even out short-term irregularities in the yarn. This doubling action contributes largely to the low irregularity of rotor spun yarn. One should not overlook the fact the elimination of the roving process also contributes greatly to the Low Mass irregularity of rotor yarns.

The number of doubling in rotor spinning can be estimated by the ratio n_fy/n_fd. Thus, an increase in the number of fibers in the yarn and/or a reduction in the number of fibers in the air duct can enhance the uniformity of the yarn. This point partially reveals the critical importance of fiber fineness in rotor spinning. Machine manufacturer commonly states that rotor spinning requires very fine fibers. It is our opinion that this statement should be qualified by a specific value of fiber fineness. Obviously, if the fiber is coarse, less number of fibers will be allowed in the yarn cross-section (for a given yarn count) and the effect of this on yarn strength and irregularity are well known. On the other hand, if the fibers are too fine, the risk of extremely high flexibility (as with microdenier synthetic fibers) and/or low maturity (as with cotton fibers) may arise. In this situation, the benefits of manufacturing fine fibers may be offset by the high tendency of fibers to entangle and disarrange. In a previous study [7], we found that polyester fibers of 0.7 denier provided higher yarn irregularity than those of 0.9 denier. In the same study, we found no significant improvement in the strength of the yarn made from 0.7 denier over that made from 0.9 denier fiber.

If one observes a rotor-spun yarn under a microscope, one will easily notice that along the yarn axis there are many fibers that are not completely tied into the yarn. Those fibers have a free end that wraps itself around the yarn periphery and causes constriction of the yarn. This is an inevitable defect that is peculiar to rotor-spun yarns. It is commonly called "fiber belts" or "wrapper fibers". According to Hunter [8], those fibers are introduced to the yarn in the rotor as a result of fibers that are trapped from the wrong direction, i.e. from the section the yarn has just left, or by fibers that are fed from fibers fed directly onto the yarn-forming point and by the yarn between the doffing tube and rotor groove coming into contact with the airborne fibers.

In connection with the influence of fiber attributes on yarn quality, we should point out that wrapper fibers are largely useless. They fail to contribute to the strength of the yarn and they provide no improvement to any quality aspect. In fact, they should be treated as waste fibers that happen to stick to the yarn body. The inevitability of wrapper fibers, however, has led many machine manufacturers to claim that they may have some merit including improvement of yarn abrasion resistance.

Although wrapper fibers are a result of a technology deficiency, their presence is greatly enhanced by some levels of fiber attributes. For instance, long fibers tend to form wrappers that are so tight that the belt looks more like a thin place. Short fibers, on the other hand, form slack and loose belts. The number of wrapper fibers is often estimated by the ratio between the staple fiber length and the rotor circumference (FL/p d). Other fiber attributes that may contribute to wrapper fibers include fiber stiffness and fiber fineness; stiffer and coarser fibers tend to become wrapper surface fibers.

One important feature that separates ring spinning from rotor spinning is the tighter fiber control in the former due to the higher spinning tension. In rotor spinning, fibers are not firmly gripped at any point of their flow; a differential tension such as that discussed in ring spinning does not exist. Accordingly, no significant fiber migration (to enhance yarn strength) is expected in rotor spinning. This point reflects the importance of fiber/rotor groove friction, and fiber-to-fiber friction. The lack of significant tension also results in some fibers that are only partially twisted leading to inferior yarn strength. These deficiencies can only compensated for by high fiber strength and optimum fiber fineness.

The different aspects of fiber/machine interaction discussed above result in a structure that consists of three layers: a core that is truly twisted (similar to ring-spun yarn), an outer layer that is partially twisted, and fiber wrappers. The true twist in rotor-spun yarns results in a natural curling tendency, similar to ring-spun yarns. However, this torque is partially balanced by a torque caused by the wrapping effect of the wrapper fibers, particularly those that take an anti-clockwise direction. The more such anti-clockwise banding fibers there are, the lower will be the curling tendency in the rotor yarn. Yarns with low curling tendency also display low yarn extensibility by virtue of their "liveliness". These features reveal two important points:

1. In rotor spun yarn, the true amount of twist is difficult to measure
2. The actual twist in rotor spun yarn is typically less than the nominal twist as set by the ratio between the rotor speed and the take-up speed.

With regard to the first point, the inverse connection between the number of fiber wrappers and the curling tendency is normally used to obtain an indirect measure of rotor yarn structure by measuring its curling tendency or the residual twist (difference between the measured yarn twist and the nominal twist). A typical value of residual twist may range from 10% to 40%.

With regard to the second point, rotor-spun yarn always requires higher levels of twist than comparable ring-spun yarns. This means that the yarn will be usually stiffer and will produce a fabric of poor hand. For this reason, some knitters prefer ring spinning over rotor spinning in the medium count range (20’s to 30’s) substituting economical benefits by quality demands. Proper fiber selection can play a major role in producing a flexible rotor-spun yarn. In general, it is well known that fine, long, and flexible fibers provide less resistance to twisting than coarse, short, and stiff fibers. Through optimization of this combination, a rotor spun yarn of acceptable flexibility level can be produced.

Can Rotor Spinning Produce Yarns of Fine Count?

As indicated earlier, rotor spinning has superior economical advantage over ring spinning in the coarse to medium counts. In recent years, there have been many attempts to push rotor spinning further into the area of fine counts When we speak of fine counts, we generally mean yarns of maximum 40’s cotton count. In order to produce fine yarn counts on rotor spinning, two main factors must be addressed: (a) machine-related factors, and (b) material-related factors.

Examining the spinning tension of rotor spinning may summarize the machine-related factor. This is the tension on the yarn, T_y, delivered from the rotor expressed by the following equation:

Where v is the rotor rotational speed in radians/sec, r is the rotor radius, and m is the coefficient of friction between the yarn and the navel surface in contact with the yarn.

The above equation indicates that the spinning tension is highly sensitive to the rotational speed of the rotor and the radius. The product vr is a primary design criterion in rotor spinning; recent trends are to increase rotor speed and reduce rotor diameter so that a balance in spinning tension is always maintained.

In light of the fact that the value of the product vr has virtually reached its technological limit, a reduction in yarn tex will result in a reduction in the spinning tension (which is already low compared to ring spinning). The importance of spinning tension as a controlling factor of the fiber flow was indicated earlier. A reduction in yarn tex will also result in a smaller area of yarn/navel contact. This will reduce the coefficient of friction, m, leading to a further reduction in spinning tension. More importantly, less area for friction heat imposed by the high rotational speed to dissipate. This last point is critical in spinning synthetic fibers or cotton/synthetic blends.

In relation to the material-related factor, the earlier discussion of fiber/machine interaction pointed out the problem of fiber disorientation during spinning, and its impact on yarn strength. This factor, in addition to the loss of fibers through wrapping, makes it difficult to improve the rotor-spinning limit. Furthermore, the need for lower twist level to improve yarn flexibility and fabric hand makes matters additionally complex.

Fine counts are associated with high quality yarn (defect free and certainly trash free). This means that the quality of the fibers must be upgraded to produce fine counts. The sliver fed to the machine should be prepared carefully so that it exhibits the lowest irregularity possible, and the lowest trash level possible. In case of light sliver, inter-fiber cohesion is critical. These criteria indicate that fiber properties such as trash content, short fiber content and inter-fiber friction are extremely important, not only for producing acceptable quality levels, but also for minimizing end breakage during spinning.

In recent years, low level combing (8% comber noil extraction) has been used to upgrade cotton fibers used in producing fine rotor yarns. Combing upgrades the cotton quality by removing neps and short fibers, and by providing better fiber orientation in the fiber strand. The added-cost by combing is justified by lower endsown during spinning , and slight reduction in twist. Bischofberger [9] reported that with optimum noil removal in combing (8-14%), and under similar spinning conditions, the yarn count can be increased from 30's to 36's from the same raw material at a constant rate of endsdown for both counts of 150 endsdown/1000 rotor hours.

Bischofberger also reported the following benefits of using combing for rotor spinning:

• Irrespective of raw material, yarn strength was found to increase by about 10% with combing and strength uniformity was improved.
• Combed rotor-spun yarns yield better filling insertion rates during weaving because of lower rates of filling stops.
• Combed rotor yarns result in better knitting efficiency because of the low fly deposition and the smoothness of yarns. The uniformity and handle of single jersey knitted fabrics were significantly improved as a result of using combed rotor yarns.

In light of the above discussion, one can develop a list of fiber properties in rotor spinning according to the order of their importance. We suggest the following list:

Fiber Selection for Air-Jet Spinning

Principle of Air-Jet Spinning

The fundamental difference between air-jet spinning and rotor-spinning is that air-jet spinning is a false-twist method. While rotor-spinning requires a complete separation of fibers, and ring-spinning requires a complete continuity of fiber flow, air-jet spinning exhibits an intermediate feature in which only a partial separation of fibers is required for the consolidation mechanism.

Similar to rotor spinning, the input strand in air-jet spinning is a drawn sliver that may be carded or combed. Drafting is achieved using high roller drafting to reduce the size of the input sliver down to the desired yarn size. The coherence mechanism in air-jet spinning is achieved by blowing out compressed air through air nozzle holes of about 0.4mm diameter to form an air vortex. The air revolves at high speed (more than 3 million rpm). Thus, the rotating element in air-jet spinning is air. This results in a rotation of the fiber bundle at a rate typically ranging from 200,000 to 300,000 rpm

As shown in Figure 2.c, two air nozzles are used: nozzle 1 and nozzle 2. Nozzle 1 may be called the "end-opening" nozzle, and nozzle 2 may be called "the twisting nozzle". These names imply the specific functions of these two nozzles as explained below.

To simplify the principle of the consolidation mechanism, suppose that only nozzle 2 is at work and that air is rotating in a clockwise direction. This action will result in twisting the fibers fed to the nozzle to form a yarn. When the yarn leaves the nozzle, untwisting takes place. Thus, with one air nozzle, a case of pure false twisting is achieved. In the actual machine, another nozzle (nozzle 1) is positioned between the nip of the front roller and nozzle 2, with air rotating in a counterclockwise direction. Thus, the two nozzles apply air rotation in two opposite directions. However, the air in nozzle 2 has a higher rotational speed than nozzle 1 to avoid complete false twisting.

The fiber strand coming out of the delivery roll forms a spinning triangle similar to that in ring- pinning. However, fibers in this triangle are under much less tension than those in ring- pinning. In other words, the fibers in the triangle are comparatively loose. The rotation of the fiber strand by the air in the two nozzles results in ballooning the fiber bundle between the front roller and nozzle 1, and in turning the balloon in nozzle 2. This balloon has no significant tension, which results in some fibers being raised from the bundle surface and moves freely. This process is called "the end-opening" action. Thus, the opposite rotation of air in nozzle 1 assists in detaching some fibers from the input strand.

The consolidation mechanism results from the two actions discussed above: (i) the false twist action, and (ii) the end opening action. The idea is to transmit the twist inserted by air rotation in nozzle 2 to the fibers at the nip of the front roller, and to detach some fibers from the twisted strand by the rotation of air in nozzle 1 in opposite direction. This end opening action takes place at the moment nozzle 2 twist is imparted. As the strand passes through nozzle 2, it will consist of detached fibers (outer layer) and truly twisted fibers (the core). When the fiber strand exits nozzle 2, the twisted core will immediately tend to untwist and the detached fibers will wrap around the core fibers. These results in a yarn consisting of a core of parallel fibers wrapped at some points along its length with fiber wrappers. The primary source of strength of air-jet spun yarn results from effective fiber wrapping.

In view of the principle discussed above, important fiber attributes in air-jet spinning are as follows:

The New Vortex Spinning (MVS)

In 1998, Murata introduced a new air jet spinning called the "MVS 851". The system is still under development. As in the conventional MJS, the sliver is directly fed to a roller drafting system. The consolidation of fibers is achieved by applying a rapidly spiraling flow of compressed air at a non-rotating spindle tip in the air nozzle. It is claimed that the spun yarn possesses characteristics close to ring-spun yarn. Our limited experience with the system indicates that it is very promising with a range of yarn fineness that can exceeds the rotor spinning range (>40’s).

In the context of fiber attributes, two fiber properties are critical to the new MVS: (I) short fiber content, and (II) dust content. It is our understanding that the system removes a great deal of short fiber. The waste percent can be as high as 6%, mostly short fibers. This gives the yarn a combed-like surface structure that we believe it will be a great plus. The amount of waste is expected to be reduced as the system develops. In this regard, cotton fiber selection and blending is a critical factor. At an expected rate of as high as 350 m/min., we expect this system to dominate the market in a short period of time.

How EFSâ can help in Producing Optimum Yarn Quality

The above discussion provided an overview of the possibilities and the limitations of ring and rotor spinning in the context of the role of fiber attributes. This role should set the criteria for a successful fiber selection and blending strategy. The ultimate goal of such strategy should be to produce an optimum yarn quality; a quality level that is universally superior (5%, 25% Uster Statistics or better) at a minimum manufacturing cost. The EFSâ system has the powerful tools that can allow the production of optimum yarn quality. The key, however, is a well-informed expert interaction with the system. In the following discussion, we will propose a step-by-step plan of action to achieve this goal.

[1] Document the Process and Product Specifications

A successful fiber selection and blending strategy should be based on well-documented information about the process and the product specifications. The more inclusive this information is the better the chance of implementing a successful strategy. Examples of the type of information required are shown in Figure 3. A more detailed outline should be established by the company implementing the EFSâ program.

This step may seem obvious to many textile technologists. However, our experience suggests that many of the information needed for implementing engineered fiber selection are often scattered or difficult to find from one source. Documentation of process and product specifications should be a result of integrated effort that involves personnel from all sectors that will be affected by the fiber selection or the blending strategy. This step can provide an excellent opportunity for technical personnel of different areas to understand how their area may influence or may be influenced by fiber selection and blending.

Quite often, we face a question about how the fiber selection process or the addition of a new cotton type to the cotton mix may affect the dyeing or the finishing performance of fabrics. This type of questions is best answered on the basis of a well-documented information about the specifications of the dyeing and finishing process. Ultimately, the fiber selection and blending criteria should be driven by the level of quality and consistency required in the end product and the processes that suffer the most from improper selection and blending of cotton fibers.

  [2] Evaluate the Fiber Database

Fiber database represents information of the different cotton groups utilized by the mill. It is the information of the parent population from which cotton bales are selected for the bale laydown. Evaluation of fiber database involves grouping of different types of cotton by variety, source, etc., and evaluating the statistical limits of each group (average, standard deviations, C.V%, and frequency distributions). Normally, mills using the EFSâ system understand this step very well. It is a part of the routine process of fiber selection.

[3] Fiber Selection: Select the Average Fiber Profile Suitable for the Process

The essence of a fiber selection strategy is to determine the average levels of fiber attributes required for a particular cotton mix. Most mills rely on general information (from experience or expert recommendations) in determining the average levels of fiber attributes. This approach is generally suitable in the initial process of implementing the EFSâ system. However, conditions such as the type and the range of the technology used, the product range, and the type of cotton utilized are never static. They do change and often at a fast rate. It is important, therefore, that the company utilizes reliable fiber-to-yarn relationships that are suitable for its specific process and product.

The issue of developing fiber-to-yarn models has been discussed repeatedly in previous EFSâ conferences and in many publications [10-14]. The outcome of these discussions may be summarized in the following points:

• The physical fiber/yarn relationship is quite complex and hardly linear.
• Empirical relationships can be developed for some yarn characteristics including yarn strength, yarn elongation, and irregularity.
• A universal empirical relationship can not be established (simply because a universal physical relationship does not exist).
• Developing empirical relationships in the mill environment involves cumbersome and time-consuming activities that are often undesirable by mill personnel.

In recent years, the present author has developed systematic procedures for using empirical fiber/yarn relationships that can be adjusted for individual textile mills without disruption of production. Discussion on these procedures is outside the scope of this presentation. However, the author will be willing to discuss these procedures with interested companies.

The importance of developing fiber/yarn relationships can be realized from the many benefits of such relationships including:

• The ability to predict the yarn quality from a given set of fiber attributes prior to processing
• The ability to determine an optimum combination of values of fiber properties suitable for a particular yarn (the inverse approach)
• The ability to separate material-related factors from process-related factors influencing the yarn quality and the processing performance

These benefits lie in the heart of the process of fiber selection and blending. They are the keys to determine optimum levels of fiber attributes that are suitable for a particular spinning system, yarn count, or end product.

In relation to processing performance and product quality beyond the yarn sector (e.g. weaving, finishing), it is important to establish fiber/fabric relationships that are directly related to the specific process or end product. These relationships are normally of a simple nature (bi-variable) and they should be developed either on the basis of historical data or small mill experiments. Through years of experience with the EFSâ system, we found that fiber properties can in fact influence many processes and quality parameters beyond the yarn sector. For example, the Micronaire value has to be optimized (not too high and not too low) in relation to the beaming efficiency of the denim warp. A low Micronaire value (despite its advantage to yarn quality) can reduce the beaming efficiency substantially. Another familiar example is the adverse effect of wrinkle-free treatments on fabric strength, which necessitates the selection of cotton fibers of especially high strength and elongation.

[4] Fiber Blending: Select a Proper Fiber Blending Strategy

Fiber selection mainly influences the level of quality of the end product. Fiber blending, on the other hand, mainly influences the consistency of the output quality levels. Accordingly, the essence of a fiber blending strategy is to determine the maximum allowable variability within a mix, and to insure that the within-laydown variability falls below this maximum limit. The EFSâ system has several capable methods (algorithms) that can be used to enhance the consistency of fiber profiles both between and within mixes. The user, however, must decide on the maximum allowable level of irregularity in the fiber properties used in the selection process.

The selection of the maximum allowable variability within the mix (C.V_max % ) is truly a challenge because of the fact that a direct relationship between fiber irregularity and yarn or fabric irregularity is yet to exist. Practical experience suggests that a highly irregular cotton mix will produce a highly irregular yarn or fabric unless costly processing effort is made to remedy this high irregularity. It also suggests that even under seemingly normal blending conditions, problems such as excessive yarn irregularities, high rates of endsdown, excessive filling stops, fabric defects, color streaks, and fabric barre are often encountered in the textile process, and in an unpredictable fashion.

To make matters additionally complex, if one attempts to determine the correlation between the irregularity in fiber strength (in the cotton mix) and the irregularity in yarn strength, one will find that this correlation is almost nonexistence. Only at extreme levels of irregularity in fiber strength, one can find some direct correspondence between the two parameters. Even in more straightforward type characteristics such as fiber fineness and fiber length, the correlation between their irregularities and yarn irregularities such as count variation and thickness variation is quite weak and often insignificant.

The lack of direct relationships between fiber irregularity and yarn or fabric irregularity can be attributed to the following main reasons:

1. Parameters used for characterizing yarn or fabric irregularity are not compatible with those used for the irregularities in fiber attributes (different testing techniques, sample size, and sample space).
2. The process-added irregularity is of extremely complex nature
3. Induced attributes such as high short fiber content, excessive neps, and high trash content can largely influence the irregularity of the yarn and end products.

In the absence of a well-established relationship between fiber irregularity and yarn or fabric irregularity, textile mills totally rely on expert recommendations. On the basis of database accumulated over a number of years from many EFSâ mill data, we have established values of maximum allowable variability measures of fiber attributes that largely correspond to yarn variability measures. These values are shown in Figure 4. They represent useful guidelines for the selection of maximum allowable variability in the cotton mix.

 [5] Cost Analysis of Fiber Selection and Blending: Determine the Cost Profile of the Cotton Mix

A complete characterization of a cotton mix can only be achieved through description of both the fiber profile (average fiber properties and variability), and the cost profile of the mix (Figure 5). Determining the cost profile is critical for making proper purchasing decisions of cotton fibers, and for minimizing the cost of raw material with respect to the overall manufacturing cost of yarn. In a previous study [16], the present author developed an integrated program for determining a technological premium/discount scale reflecting the cost of fibers in relation to a particular spinning system and yarn count. The detail of this program is outside the scope of this paper, but the reader is encouraged to refer to the reference indicated.

 Literature

1. El Mogahzy, Y., and Gowayed, Y., Theory and Practice of Cotton Fiber Selection. Part I: Fiber Selection Techniques and Algorithms. Textile Res. J., 65:32-40, 1995

2. El Mogahzy, Y., and Gowayed, Y., Theory and Practice of Cotton Fiber Selection. Part II: Sources of Cotton Mix Variability and Critical Factors Affecting It. Textile Res. J., 65:75-84, 1995

3. Deussen, H., Rotor Spinning Technology, Schlafhorst Inc., Charlotte, N.C., U.S.A., 1993

4. Lord P., The Economics, Science and Technology of Yarn, NCSU, Raleigh, N.C., U.S.A., 1981.

5. El Mogahzy, Y., Fiber-To-Yarn Engineering, A Video-Base Course, Quality Tech. Production, Auburn, AL, U.S.A., 1997.

6. Deussen, H., Some Thoughts on the Role of Cotton in New Spinning Technologies, Presentation to the Cotton Spinners and Cotton Growers Conference in Lubbock, Tx, 1984.

7. El Mogahzy, Y., and Lynch, K. Production of different types of Micro-denier fibers on Open-End and Air-Jet Spinning machinery, Technical report produced for private consultant to a fiber producer, 1990.

8. Hunter, L., The Production and Properties of Staple-Fiber Yarns Made by Recently Developed Techniques. Textile Progress, The Textile Institute, Volume 10, No. ½, 1978.

9. J. Bischofberger, Rotor Yarn-Combed System? Int. Textile Bull. (ITB), 2, PP 30-37, 1990.

10. El Mogahzy, Y., and Broughton, R., Diagnostic Procedures for Multicollinearity Between HVI Cotton Fiber Properties, Textile Research Journal, Vol. 59, No. 8, 440-447,1989.

11. El Mogahzy, Y. E., Broughton, R., and Lynch, W. K., A Statistical Approach for Determining the Technological Value of Cotton Using HVI Fiber Properties, Textile Research Journal, Vol. 60, No. 9, 440-447,1990.

12. El Mogahzy, Y., Selecting Cotton Fiber Properties for Fitting Reliable Equations to HVI Data, Textile Research Journal, Vol. 58, No. 7, 392-397,1988.

13. El Mogahzy, Y., Optimizing Cotton Blend Cost with Respect to Quality using HVI Fiber Properties and Linear Programming. Part I: Fundamentals and Advanced Techniques of Linear Programming. Textile Res. J., 62:1-8, 1992

14. El Mogahzy, Y., Optimizing Cotton Blend Cost with Respect to Quality using HVI Fiber Properties and Linear Programming. Part II: Combined Effects of Fiber Properties and Variability Constraints. Textile Res. J., 62:108-114, 1992.

15. El Mogahzy, Y., Utilization of the Advanced Fiber Information System (AFIS) in the Evaluation of the Textile Process, Meliand, 1-2, E1-E4, 1997.

16. El Mogahzy, Y., Determining the Technological Value of Cotton Fibers, Proceedings of the EFSâ Research Forum, Cotton Incorporated, November, 1996.

APPENDIX I:

Summary Tables of the 1997 Uster Statistics for 100% Cotton Yarns

Table I.1. Comparison between the Best & the Worst Levels of Coarse Yarn Quality

[97 Uster Statistics, 100% Cotton Carded-Ring-Spun, Count Range, Ne = 7’s-15’s]