ABSTRACT. Dani, Nikhil Prakash. The fundamentals of air-jet texturing. (Under the direction of Dr. William Oxenham and Dr. Behnam Pourdeyhimi)

Transcription

1 ABSTRACT Dani, Nikhil Prakash. The fundamentals of air-jet texturing. (Under the direction of Dr. William Oxenham and Dr. Behnam Pourdeyhimi) Air-jet texturing is a well-established filament yarn processing technology that has been around for more than half a century. There is much debate about various aspects of the process in the literature such as the mechanism of loop formation, role of water in air texturing and the influence of process variables on air textured yarn structure. An attempt has been made in the current research to understand the above variables and other aspects of the process. While the present research tends to support some of the earlier findings, this dissertation also presents findings that are contradictory to those in the literature. These include the following. For the yarns and the range of air pressure used in the study ( PSI), there is no significant effect of increase in air pressure on final air textured yarn properties. This is confirmed by image analysis of air-textured yarns, the results of online tension measurements and tensile tests conducted on the yarns. Overfeed is the governing factor in the process of air texturing. The influence of change in overfeed on the final textured yarn properties is greater than that of any other processing parameter or supply yarn parameter. The role of water is to combine filaments, thereby presenting fewer free filaments in the airflow. This leads to better three-dimensional displacement of filaments around the central yarn core, which in turn leads to a more uniform yarn structure. The mechanism of loop formation is redefined based on CFD simulations and highspeed image analysis. As the incoming filaments encounter the plane of air-inlet, they re distributed in one of the six circulation zones and are displaced three dimensionally around the yarn axis as well as pushed out of the nozzle at the same time. Some of the filaments however remain in the non-circulation zone and are carried out of the nozzle and form the core of the air textured yarn structure. The reasons for the disparity between the present conclusions and earlier reported research are discussed.

2

3 BIOGRAPHY Full Name: Nikhil Prakash Dani Born: 22 nd November 1976, Bombay, India (Cusp of Sagi-Scorpio) Brought into this world by: Prakash Ishwerlal Dani & Rekha Prakash Dani Higher secondary education: Baroda High School Alkapuri, Baroda, India Bachelor of Science (BS): Textile Engineering at M.S.U, Baroda, India Masters of Science (MS): Textile & Apparel Technology and Management, NCSU, USA Doctor of Philosophy (Ph.D): Fiber and Polymer Science, NCSU, USA Professional Career: I ll be working for The Clorox Company, in Pleasanton - California from the first week of August 2004 as a Research Scientist in their Consumer Applied Technology Division Other interests: None ii

4 ACKNOWLEDGEMENTS An expression of thanks or a token of appreciation In the order God My parents and family Dr. William Oxenham Dr. Behnam Pourdeyhimi Dr. Hooman Tafreshi Dr. Benoit Maze & Dr. Eunkyoung Shim Dr. Shiffler, Dr. Hinestroza & Dr. Mackenzie Nagendra Anatharamaiah Vignesh Natrajan William Bradford Johnson NCRC Fibrous Material Forum College of Textiles iii

5 TABLE OF CONTENTS LIST OF FIGURES... VIII LIST OF TABLES... XI INTRODUCTION TO THE RESEARCH ) INTRODUCTION TO THE RESEARCH...2 BASICS OF LOOP FORMATION ) FUNDAMENTALS OF LOOP FORMATION ) REQUIREMENTS FOR LOOP FORMATION & RETENTION ) General requirements ) Requirements specific to type of loop...8 OVERALL DESCRIPTION OF THE PROCESS ) INTRODUCTION ) MACHINE COMPONENTS ) PROCESS FLOW ) General description ) Operating principles...13 PRINCIPLE...14 MECHANISM OF LOOP FORMATION ) INTRODUCTION ) REVIEW OF LITERATURE ) Mechanism according to Bock and Lunenschloss [40] ) Mechanism according to Acar, Turton and Wray [23] ) Mechanism according to Sengupta, Kothari and Srinivasan [47] ) Mechanism according to Nosov [13]: ) PROPOSED MECHANISM OF LOOP FORMATION ) Operating sequence of the process ) Mechanism of loop formation in air jet texturing...21 NOZZLES ) INTRODUCTION ) LOCATION OF NOZZLE IN THE PROCESS FLOW ) NOZZLE MATERIAL ) HISTORY OF NOZZLE DEVELOPMENT ) REVIEW OF LITERATURE ) ANATOMY OF THE NOZZLE: ) FUNCTIONS OF THE NOZZLES ) SIMULATIONS OF AIRFLOW INSIDE THE INSIDE THE NOZZLE ) Introduction ) Introduction on compressible flow in nozzles ) Simulation details ) Simulation methodology ) Pre-processing ) Numerical Solution ) Post-processing...35 iv

6 5.8.5) Analysis of simulation images in relation to the mechanism of loop formation...42 ROLE OF WATER IN AIR-JET TEXTURING ) INTRODUCTION ) REVIEW OF LITERATURE ) Issue with the explanations of role of water in texturing provided in the literature ) PROPOSED ROLE(S) OF WATER ) WETTING BEHAVIOR IMAGING ) Interesting findings from the Petri dish experiments ) Interesting findings from single droplet experiments using optical imaging...51 EXPERIMENTAL ) INTRODUCTION ) AIR TEXTURING UNIT ) Auxiliary equipment related to the trials ) Limitations of the machine as related to the experiments performed in current research ) NOZZLES/JETS ) Nozzle maintenance ) SUPPLY YARNS ) AIR TEXTURING TRIALS ) HIGH-SPEED IMAGING ) Limitations ) INDIVIDUAL FILAMENT DIAMETER MEASUREMENTS ) IMAGING OF FINAL AIR TEXTURED YARN STRUCTURES ) Limitations ) IMAGE ANALYSIS ALGORITHM ) Limitations of the above method ) Information obtained from the above procedure...74 STRUCTURE AND PROPERTIES OF AIR TEXTURED YARNS ) INTRODUCTION ) REVIEW OF LITERATURE ) Instability ) Properties of air textured yarns and influence of process variables on them ) Review of literature on the influence on textured yarn properties of change in process variables ) Effect of increase in air pressure ) Effect of increase in overfeed ) Effect of increase in processing speed ) Effect of change in processing conditions Dry v/s Wet ) Review of literature on supply yarn parameters ) Polymer type ) Number of filaments ) Filament fineness or denier per filament (dpf) ) Filament cross sectional shape ) Spin finish ) RESULTS & DISCUSSION: INFLUENCE OF SUPPLY YARNS ON YARN DIAMETER, CORE DIAMETER, NUMBER OF LOOPS & TYPES OF LOOPS ) Dyneema...87 v

7 8.3.2) Kevlar ) Zylon ) Spectra ) Technora ) Vectran ) Dacron ) Comparison between yarns made of same polymer (& same dpf) but having different number of filaments ) Comparison of a single yarn (Dyneema: 440/195) textured using different overfeeds ) Comparison of a single yarn (Dacron) textured using different overfeeds ) Comparison between yarn made of same polymer (Dyneema) textured at different stretch ) Comparison of yarn made of same polymer (Kevlar) textured at different air pressures ) Comparison of yarn made of same polymer (Dacron) textured at different air pressures ) TENSILE TESTING OF AIR TEXTURED YARNS ) Polyester tire cord yarn ) Effect of increase in stretch at 45% overfeed ) Effects of increase in overfeed at 15% stretch ) Effect of increase in air pressure ) Dyneema ) Effects of increase in overfeed at 10% stretch ) Effects of increase in stretch at 15% overfeed ) Effects of increase in overfeed at 15% stretch ) Effect of increase in air pressure ) Kevlar ) Effects of increase in overfeed at 15% stretch ) Effects of increase in stretch at 45% overfeed ) Effect of increase in air pressure ) TENSION MEASUREMENTS ) ANALYSIS OF HIGH-SPEED IMAGE CAPTURES ) Introduction ) Effect of increase in speed ) Effect of increase in air pressure at nozzle exit ) Effect of increase in air pressure at water bath exit ) Effect of increase in air pressure at nozzle entry ) Effect of increase in stretch ) Effect of increase in overfeed ) Effect of increase in effect overfeed at constant core overfeed ) Texturing in dry and wet conditions ) Same yarn textured using different nozzles Polyester tire cord yarn was textured using T100 and T ) DISCUSSION BASED ON THE ABOVE RESULTS ) Introduction )Texturing condition: Dry v Wet ) Yarns of same polymer but having different number of filaments ) Same yarn textured at increasing overfeeds ) Same yarn textured at increasing stretch ) Same yarn textured at increasing air pressures vi

8 CONCLUSIONS AND RECOMMENDATIONS ) CONCLUSIONS ) RECOMMENDATIONS BIBLIOGRAPHY APPENDIX A: EQUATIONS SOLVED IN THE SIMULATIONS APPENDIX B: IMAGE ANALYSIS DATA vii

9 LIST OF FIGURES FIGURE 1: LOOP FORMATION BY BUCKLING...5 FIGURE 2: HORSESHOE LOOP...6 FIGURE 3: LOOP FORMATION BY TWISTING...6 FIGURE 4: CRITICAL GAP FOR FORMATION OF BUCKLED LOOPS...8 FIGURE 5: BUCKLED AND HORSESHOW LOOPS IN AIR TEXTURED YARN...9 FIGURE 6: STAEHLE AIR-JET TEXTURING UNIT AT COLLEGE OF TEXTILES, NCSU...12 FIGURE 7: PROCESS FLOW OF AIR JET TEXTURING...15 FIGURE 8: MECHANISM OF LOOP FORMATION ACCORDING TO ACAR ET AL [23]...18 FIGURE 9: PROCESS FLOW OF AIR JET TEXTURING (REPEATED HERE FOR CONVENIENCE)...21 FIGURE 10: THE PROPOSED MECHANISM IF LOOP FORMATION SHOWING BEHAVIOR OF FILAMENTS INSIDE THE NOZZLE CHANNEL. THE IMAGE ALSO SHOWS THE PLANE OF AIR INLET...23 FIGURE 11: BEHAVIOR OF FILAMENTS AT THE PLANE OF AIR INLET IN DRY TEXTURING CONDITION...23 FIGURE 12: BEHAVIOR OF FILAMENTS AT THE PLANE OF AIR INLET IN WET TEXTURING CONDITION...24 FIGURE 13: METAL CASING WITH BAFFLE (LEFT) AND JET (RIGHT) [41]...26 FIGURE 14: OLD DUPONT NOZZLE [14]...27 FIGURE 15: MIRLAN CYLINDRICAL JET [14]...27 FIGURE 16: EVOLUTION OF TASLAN NOZZLE [14]...28 FIGURE 17: TYPICAL AIR JET DESIGN [7]...31 FIGURE 18: MASS FLOW RATE...31 FIGURE 19: COMPARISON OF HIGHER (LEFT) AND LOWER (RIGHT) DENIER YARNS BEING TEXTURED USING SAME NOZZLE...32 FIGURE 20: AIR TEXTURING NOZZLE (T-351) TOP VIEW...36 FIGURE 21: AIR TEXTURING NOZZLE (T-351) SIDE VIEW...36 FIGURE 22: GRID DISTRIBUTIONS ON THE NOZZLE WALL...37 FIGURE 23: VELOCITY VECTORS...38 FIGURE 24: VELOCITY VECTORS CLOSE TO NOZZLE AXIS AT THE AIR INLET PLANE...39 FIGURE 25: VIEW OF 6 CIRCULATION ZONES...40 FIGURE 26: CONTOURS OF VELOCITY MAGNITUDE...41 FIGURE 27: HYPOTHESIZED BEHAVIOR OF PERIPHERAL FILAMENTS IMPACTED DIRECTLY BY AIRFLOW AT THE AIR-INLET PLANE...43 viii

10 FIGURE 28: YARN UNDER TENSION TILL THE PLANE OF AIR INLET...47 FIGURE 29: YARN EMERGING OUT OF THE WATER BATH SURROUNDED BY A SHEATH OF WATER (LEFT) AND YARN JUST ENTERING THE WATER BATH (RIGHT)...48 FIGURE 30: WET YARN JUST ENTERING THE NOZZLE...48 FIGURE 31: ZYLON PRODUCED UNDER WET TEXTURING CONDITIONS...49 FIGURE 32: EFFECT OF DROP OF WATER MIGRATING (FROM RIGHT TO LEFT) IN A LOOSE MULTIFILAMENT YARN...51 FIGURE 33: REGIONS OF AIR TEXTURING PROCESS THAT WERE IMAGED USING HIGH-SPEED PHOTOGRAPHY (IMAGES AT DIFFERENT MAGNIFICATIONS)...64 FIGURE 34: STEP FIGURE 35: STEP FIGURE 36: STEP FIGURE 37: STEP FIGURE 38: STEP FIGURE 39: STEP FIGURE 40: STEP FIGURE 41: STEP FIGURE 42: STEP FIGURE 43: STEP FIGURE 44: STEP FIGURE 45: STEP FIGURE 46: STEP FIGURE 47: YARN DIAMETER AND CORE DIAMETER AS DEFINED IN THE CURRENT RESEARCH...85 FIGURE 48: IMAGE ANALYSIS ALGORITHM...86 FIGURE 49: DYNEEMA COMPARISON OF YARN TEXTURED IN DRY AND WET CONDITIONS...87 FIGURE 50: KEVLAR COMPARISON OF YARN TEXTURED IN DRY AND WET CONDITIONS...98 FIGURE 51: ZYLON COMPARISON OF YARN TEXTURED IN DRY AND WET CONDITIONS FIGURE 52: SPECTRA COMPARISON OF YARN TEXTURED IN DRY AND WET CONDITIONS FIGURE 53: TECHNORA COMPARISON OF YARN TEXTURED IN DRY AND WET CONDITIONS FIGURE 54: VECTRAN COMPARISON OF YARN TEXTURED IN DRY AND WET CONDITIONS FIGURE 55: DYNEEMA EFFECT OF INCREASE IN OVERFEED ON YARN AND CORE DIAMETER FIGURE 56: KEVLAR EFFECT OF INCREASE IN AIR PRESSURE ON YARN AND CORE DIAMETER 117 FIGURE 57: POLYESTER TIRE CORD YARN: EFFECT OF INCREASE IN STRETCH AT CONSTANT OVERFEED ix

11 FIGURE 58: POLYESTER TIRE CORD YARN: EFFECT OF INCREASE IN OVERFEED AT CONSTANT STRETCH FIGURE 59: POLYESTER TIRE CORD YARN: EFFECT OF INCREASE IN AIR PRESSURE FIGURE 60: DYNEEMA: EFFECT OF INCREASE IN OVERFEED AT CONSTANT STRETCH FIGURE 61: DYNEEMA: EFFECT OF INCREASE IN STRETCH AT CONSTANT OVERFEED FIGURE 62: DYNEEMA: EFFECT OF INCREASE IN OVERFEED AT CONSTANT STRETCH FIGURE 63: DYNEEMA: EFFECT OF INCREASE IN AIR PRESSURE FIGURE 64: KEVLAR: EFFECT OF INCREASE IN OVERFEED AT A CONSTANT STRETCH FIGURE 65: KEVLAR: EFFECT OF INCREASE IN OVERFEED AT A CONSTANT STRETCH FIGURE 66: KEVLAR: EFFECT OF INCREASE IN AIR PRESSURE FIGURE 67: TENSION MEASUREMENT ZONES FIGURE 68: CHANGE IN SPEED FIGURE 69: EFFECT OF INCREASE IN SPEED FIGURE 70: CHANGE IN AIR PRESSURE FIGURE 71: CHANGE IN AIR PRESSURE FIGURE 72: CHANGE IN AIR PRESSURE AT WATER EXIT FIGURE 73: CHANGE IN AIR PRESSURE AT NOZZLE ENTRY FIGURE 74: CHANGE IN STRETCH FIGURE 75: EFFECT OF INCREASE IN STRETCH AT CONSTANT OVERFEED FIGURE 76: EFFECT OF INCREASE IN OVERFEED AT CONSTANT STRETCH FIGURE 77: KEVLAR: EFFECT OF INCREASE IN OVERFEED AT CONSTANT STRETCH FIGURE 78: POLYESTER TIRE CORD: EFFECT OF INCREASE IN OVERFEED AT CONSTANT STRETCH FIGURE 79: CHANGE IN EFFECT OVERFEED FIGURE 80: INCREASE IN EFFECT OVERFEED FIGURE 81: DRY V WET CONDITION AT NOZZLE ENTRY FIGURE 82: TEXTURING USING DIFFERENT NOZZLES x

12 LIST OF TABLES TABLE 1: PRINCIPLES AND COMPONENTS OF AIR JET TEXTURING...14 TABLE 2: SPECIFICATIONS OF NOZZLES USED IN CURRENT RESEARCH...30 TABLE 3: YARNS USED IN THE CURRENT RESEARCH WITH RELEVANT DETAILS. THE YARNS ARE LISTED FROM TOP TO BOTTOM IN THE COLUMNS STARTING FROM HIGHEST BENDING RIGIDITY (SPECTRA ) TO LOWEST BENDING (TECHNORA )...58 TABLE 4: TRIAL TABLE 5: TRIAL TABLE 6: YARNS USED FOR HIGH-SPEED IMAGING...62 TABLE 7: PACKING DENSITY OF TEXTILE STRUCTURES (FROM PILLER ET AL [12])...76 TABLE 8: DYNEEMA YARN DIAMETER: DRY V WET. ALL VALUES IN THE TABLE ARE IN MM TABLE 9: DYNEEMA CORE DIAMETER: DRY V WET: ALL VALUES PRESENTED IN TABLE ARE IN MM...96 TABLE 10:DYNEEMA: NUMBER AND TYPES OF LOOPS IN DRY AND WET CONDITIONS...97 TABLE 11: KEVLAR YARN DIAMETER: DRY V WET. ALL VALUES PRESENTED IN THE TABLES ARE IN MM...99 TABLE 12: KEVLAR CORE DIAMETER: DRY V WET. ALL VALUES PRESENTED IN THE TABLES ARE IN MM...99 TABLE 13:ZYLON YARN DIAMETER: DRY V WET. ALL VALUES PRESENTED IN THE TABLE ARE IN MM TABLE 14: ZYLON CORE DIAMETER: DRY V WET. ALL VALUES PRESENTED IN THE TABLE ARE IN MM TABLE 15: SPECTRA YARN DIAMETER: DRY V WET. ALL VALUES PRESENTED IN THE TABLE ARE IN MM TABLE 16: SPECTRA CORE DIAMETER: DRY V WET. ALL VALUES PRESENTED IN THE TABLE ARE IN MM TABLE 17: TECHNORA YARN DIAMETER: DRY V WET. ALL VALUES PRESENTED IN THE TABLE ARE IN MM TABLE 18: TECHNORA CORE DIAMETER: DRY V WET. ALL VALUES PRESENTED IN THE TABLE ARE IN MM TABLE 19: VECTRAN YARN DIAMETER: DRY V WET. ALL VALUES PRESENTED IN THE TABLE ARE IN MM xi

13 TABLE 20: VECTRAN CORE DIAMETER: DRY V WET. ALL VALUES PRESENTED IN THE TABLE ARE IN MM TABLE 21: DACRON YARN DIAMETER: DRY V WET. ALL VALUES PRESENTED IN THE TABLE ARE IN MM TABLE 22: DACRON CORE DIAMETER: DRY V WET. ALL VALUES PRESENTED IN THE TABLE ARE IN MM TABLE 23: YARN DIAMETER COMPARISON FOR SAME POLYMER YARN HAVING DIFFERENT NUMBER OF FILAMENTS TABLE 24: CORE DIAMETER COMPARISON FOR SAME POLYMER YARN HAVING DIFFERENT NUMBER OF FILAMENTS TABLE 25: THE EFFECT OF INCREASE IN OVERFEED ON YARN DIAMETER. ALL VALUES IN THE TABLE ARE IN MM TABLE 26: THE EFFECT OF INCREASE IN OVERFEED ON CORE DIAMETER. ALL VALUES IN THE TABLE ARE IN MM TABLE 27: THE EFFECT OF INCREASE IN OVERFEED ON YARN DIAMETER. ALL VALUES ARE IN MM TABLE 28: THE EFFECT OF INCREASE IN OVERFEED ON CORE DIAMETER. ALL VALUES ARE IN MM TABLE 29: THE EFFECT OF INCREASE IN STRETCH ON YARN DIAMETER. ALL VALUES ARE IN MM TABLE 30: THE EFFECT OF INCREASE IN STRETCH ON CORE DIAMETER. ALL VALUES ARE IN MM TABLE 31: THE EFFECT OF INCREASE IN AIR PRESSURE ON YARN DIAMETER. ALL VALUES ARE IN MM TABLE 32: THE EFFECT OF INCREASE IN AIR PRESSURE ON CORE DIAMETER. ALL VALUES ARE IN MM TABLE 33: THE EFFECT OF INCREASE IN AIR PRESSURE ON YARN DIAMETER. ALL VALUES ARE IN MM TABLE 34: THE EFFECT OF INCREASE IN AIR PRESSURE ON CORE DIAMETER. ALL VALUES ARE IN MM TABLE 35: TENSION VALUES IN THE THREE ZONES FOR THREE DIFFERENT YARNS WITH INCREASING OVERFEED TABLE 36: TENSION VALUES IN THE THREE ZONES FOR THREE DIFFERENT YARNS WITH INCREASING AIR PRESSURES xii

14 TABLE 37: TENSION VALUES IN THE THREE ZONES FOR THREE DIFFERENT YARNS WITH INCREASING STRETCH TABLE 38: COMPARISON OF YARN AND CORE DIAMETERS AND UNIFORMITY IN WET AND DRY CONDITIONS TABLE 39: COMPARISON OF NUMBER OF LOOPS IN WET AND DRY CONDITIONS xiii

15 Section I Introduction to the research 1

16 1.1) Introduction to the research Air-jet texturing is a well-established filament yarn processing technology that has been around for more than half a century. Researchers starting from the late 1970s have attempted to study various aspects of the process. Acar et.al [2,14,22,23,24,34] published a comprehensive analysis of the process in a series of papers in 1986, which, remain till the current time the only concentrated research publications in air texturing. The limited range of end use of air-textured yarns has not facilitated continued interest in the understanding of the process. Most of the previously reported work was mainly done on Polyester, Polypropylene & Nylon yarns, and although the texturing of Glass fibers was and is common, very little published work exists elucidating the various aspects of the process or the final yarn properties in relation to their texturing. There is also little or no published work relating to the air texturing of organic high modulus fibers, either in terms of the process or the final yarn properties. Due to the wide differences in physical and mechanical properties of organic high modulus fibers compared to ordinary textile yarns (apparel grade yarns), the understanding of the process currently available is inadequate to optimize and control the process. Primarily, it does not point to any means for precise control of the process in terms of: Loop type, size & density Core-effect component selection Nozzle selection/ design optimization Such control is vital for air texturing of high-modulus yarns. High modulus yarns such as Kevlar, Nomex & Vectran are used in applications where their very high strength, flame resistance and other specific properties are of considerable importance. However most of these yarns have low work of rupture, low adhesion to matrix (leading to delamination in composites) & non-cotton like feeling, among other disadvantages. Furthermore, they don t lend themselves to be combined easily at the spinning stage due to complex issues of rheology. Air texturing addresses the above issues with some degree of success. However, the air textured yarn structures currently available are relatively non-uniform and some 2

17 means of controlling them is desirable. To obtain a better understanding of the process as it relates to the air texturing of organic high modulus fibers (OHMF from here on) the fundamental aspects of the process, as listed below, were analyzed: Loops: o Types of loops o Mechanisms of loop formation Functions of the nozzle: o Precise definition of the function(s) of the nozzle o Airflow inside the nozzle Influence of change in process variables: o Effects of change in overfeed, stretch, speed & air pressure Role of water: o Lubricant? Cohesive agent? Supply yarn parameters o Linear density of yarn/individual fiber & Number of filaments o Mechanical and physical properties High-speed imaging, image analysis, & mechanical testing were combined with established concepts in fiber science to define and supplement the theories proposed in the research. 3

18 Section II Basics of loop formation 4

19 2.1) Fundamentals of loop formation Definition of a loop: A length of line, thread, ribbon, or other thin material that is curved or doubled over making an opening [46] The mechanism of loop formation for a single fiber is discussed below. The same mechanism may be used to understand loop formation for a group or bunch of fibers. Figure 1: Loop formation by buckling Consider the two ends of a fiber (figure 1) approaching each other due to application of a force at the two ends 1. Assuming that the fiber is not very weak (so that there is no local crinkling or wavy behavior) the fiber will start buckling (arching) and if enough strain energy is developed in one or both parts of fiber then it will eventually form a crossover point (to 1 The same mechanism may apply while considering only one end of the fiber to be moving towards the other 5

20 dissipate the strain energy and achieve minimum energy state) and form a cross-over loop or as classified in this research, a buckled loop. If enough strain energy is not developed the arc will not form a crossover point and will be retained in one of the many possible shapes, all classified in this research as horseshoe loops (Figure 2). Figure 2: Horseshoe loop Figure 3: Loop formation by twisting A third class of loops, known as twisted loops, as classified in the current research, require an initial loose length and are formed by the action of a high twisting force on one or both ends of a fiber. The twist is transmitted along the fiber axis as strain energy and causes the 6

21 fiber to arc up, which then crosses over by dissipation of the developed strain energy and forms a loop having a twisted base. As with buckled loops, if the twist is not transmitted for some reason, the resulting loop is retained as a horseshoe loop. These loops can be formed from individual fibers (if they are stiff enough) or from a bunch (or group) of fine fibers that combine to achieve that critical loop forming stiffness. Accordingly, this research proposes another classification of loops: Individual & Multi-fiber. The mechanisms involved in loop formation using bunched fibers are the same as that for individual fibers. 2.2) Requirements for loop formation & retention 2.2.1) General requirements Based on the above discussion, the following can be listed as the general requirements for loop formation & retention: Critical stiffness: A certain critical stiffness of the fiber is essential for bending/buckling without crinkling or having localized waviness. In case of very fine and long fibers, this critical stiffness may be achieved by the grouping of several fibers together as a bundle Force: To bend/buckle/twist the fiber(s) Structural integrity: The fibers must have sufficient strength to withstand the forces and not break on being bent or twisted Core (Supporting structure): It is not only sufficient to form the loops but also to retain them. Thus, a supporting structure that helps in stabilizing and retaining the loops is a requirement. In air-textured yarn structures, the same elements that form the loops also form the core at different points along the length of the yarn. Thus the intensity of intermixing of the core structure, combined with the geometric (cross sectional shape, length of the fiber involved) & surface characteristics (amount of spin finish, frictional properties) may greatly effect the retention of the formed loops 7

22 2.2.2) Requirements specific to type of loop As discussed above there are two main types of loops: Individual and Multi-fiber. Further, either type can be buckled, horseshoe or twisted. This section discusses the specific requirements for the formation of these three types of loops. Buckled loops o Critical gap between two ends of the fibers approaching each other (or one stationary and other moving): This gap may be characteristic for a particular fiber type and may be considered chiefly a function of fiber stiffness o Lack of obstruction to the migration of two ends towards each other i.e., minimum resistance to the migration of the two ends of the loops forming fibers is desirable. The fiber to fiber frictional properties and the density of packing (number of filaments & their arrangement) may thus play an important role. This may also indicates that the fibers on the periphery of a yarn cross section are more likely to form buckled loops than inner ones. Figure 4: Critical gap for formation of buckled loops Twisted loops o Loose initial length: As shown in figure 3, a twisted loop cannot be formed without a loose initial length. This can be verified by twisting a fiber under tension o Very high twisting force: The torsional rigidity of a fiber is much higher than the bending rigidity. In other words, it is much harder to twist the fibers than to 8

23 bend them. The lack of any significant amount of twisted loops in the textured yarn structure indicates that along with the obstruction to twist transference along fiber axis, which is inherent in the process of air texturing, the nozzles may not be providing enough twisting force to facilitate twisted loop formation o Lack of obstruction to the transference of twist along fiber axis Horseshoe loops o Horseshoe loops are formed purely as a result of bending of the fibers. Thus the only requirement is that of a bending force. If the conditions for the formation of buckled and twisted loops are not met or are met only partially, then the resulting structure is a horseshoe loop It may be hypothesized that each of the three loop types has associated bulk and stability. The twisted loops maybe the most stable but have the least amount of bulk, while the horseshoe loops maybe the least stable but have the most amount of bulk. Buckled loops may exhibit provide properties somewhere in the middle. Thus the frequency and distribution of each type may determine the final textured yarn properties (figure 5). Buckled Horseshoe Figure 5: Buckled and horseshow loops in air textured yarn Though they have not provided any evidence, Piller et al [12] has claimed that the loop shape depends on the fineness (dpf) of the filaments and their rigidity. 9

24 Section III Overall description of the process 10

25 3.1) Introduction This section describes the overall working of the process without going into details about individual aspects, which have been covered in other sections of the dissertation. The flow of the process and the place of individual components in the overall picture is described. Finally, a slightly more detailed description of the operating principles of the process is presented. 3.2) Machine components Figure 6 shows a Staehle RMT-D single spindle air-texturing unit at the North Carolina State University, College of Textiles. This machine is used widely in the industry and was used to conduct all the experiments in this research. Descriptions of various other machines used in the air texturing industry can be found in Kothari [36]. The yarn runs from the creel (not shown in figure 6) at the back of the machine to the drafting system and into the texturing box, which contains the air texturing jet and the wetting system. From there the yarn runs through the stabilizing zone into the heater box and finally via the oiling rollers to the take-up. Details of each zone are as below: Creel: One or several packages can be mounted on the creel that is located on the side or back of the machine. Tensioning in this zone is very critical to the success of the process, especially for high modulus yarns, which normally have high linear densities and large number of filaments Drawing zone: The rollers that form this zone are heated godets, facilitating the draw texturing of partially oriented yarns (POY) Texturing zone: The texturing box, which is usually insulated for sound, comprises this zone and contains the wetting system and the air jet Stabilizing zone: Within the stabilizing zone, the yarn is tensioned by rollers running with a positive differential i.e., the front roller moves at a faster speed than the back roller Heat-setting zone: The yarn is overfed into the heat-setting zone and thus under low tension and after setting its tendency towards shrinkage is reduced. The mechanical stability i.e., the retention of surface loops under application load, of the yarn is 11

26 improved further and the size of the loops is again reduced. This zone was not utilized in the current research Winding zone: The winding tension and package build can be selected in such a manner that packages with optimum unwinding characteristics can be produced Nozzle Water Figure 6: Staehle Air-jet texturing unit at College of Textiles, NCSU The Staehle air-texturing unit and its features and limitations as related to the current research have been described in greater detail in

27 3.3) Process flow 3.3.1) General description The most widely quoted general description of the process is from Acar [16]. It has been reproduced below. When the overfed filaments enter the texturing nozzle, they are carried along through the nozzle, blown out from the texturing end, and are formed into loops which are mutually trapped in the yarn structure by the effect of the supersonic and turbulent air stream and forms a textured yarn structure. The supply yarn is normally wetted just before it is fed into the texturing nozzle by passing it through a wetting unit. Wet texturing improves the quality of textured yarn produced. Textured yarn is taken up at right angles to the nozzle axis by the delivery rollers located after the nozzle. Another set of take-up rollers, running at slightly higher speeds than the delivery rollers, may be used before the high-speed winders to apply tension to the textured yarn in order to stabilize the loops formed during the process. The textured yarn is then wound up by means of a high-speed winding unit. Heaters can optionally be used to impart further desired properties to thermoplastic yarns, but this is not essential for the process. Texturing nozzles are usually enclosed in a chamber; not only to reduce the noise created by the air jets, but also to collect the used water and the spin-finish washed off from the filaments during the process. Some texturing nozzles have an impact element at the nozzle exit, to be used in certain cases as recommended by the supplier. These impact elements, which can be of different shapes, e.g., cylindrical, flat, or spherical, are believed to improve the process stability and yarn quality in texturing finer yarn [16]. The above description is very accurate and sufficient to describe the basic process flow and requires no alteration ) Operating principles Air textured yarns have surface features unlike any other textile yarn structure. Though closely resembling and performing essentially the same function as the hairs on the surface of a spun yarn, the surface elements of an air textured yarn are closed arcs or as they are popularly called - loops. To understand the process it is essential to understand the 13

28 Nature & types of these loops Mechanisms of loop formation Influence of loop type/distribution on final yarn property Understanding the process in terms of loops may point towards pre and post process modifications for optimizing the size, type, distribution & evaluation of loops thereby facilitating better control of the final yarn structure. This may lead towards the process finding acceptance in an area of critical functionality such as high modulus yarns where very high uniformity and performance prediction are imperative. The air texturing process is designed on principles & components that address the formation and control of loops such as show in table 1. Table 1: Principles and components of air jet texturing Principle Function(s) Component(s) Overfeeding Provides excess length that Rollers (negative speed can be bent/twisted to be differential let off roller formed into loops i.e., the moves faster than take up new surface element (bulk) roller) as well as the intermingled core Bending, rotating, displacing, opening, - Transforming the excess Nozzle/Jet length into loops translating (forcing out) - Distributing the loops three dimensionally around the fiber axis - Forming a central core that supports the surface loops by mixing the fibers Wetting -Imparts cohesion thereby Water bath system facilitating the formation of stable core -Combines fibers thereby 14

29 Stretching helping very fine/long fibers achieve critical stiffness to form into loops rather than being locally crinkled -Removes spin finish - Removes any slack that may be present in the core structure thereby minimizing the instability of the structure Roller (positive speed differential take up roller moves faster than let off roller) Each of the above principles and components is covered in greater detail in subsequent chapters. Figure 7: Process flow of air jet texturing 15

30 Section IV Mechanism of loop formation 16

31 4.1) Introduction The main purpose of the air texturing process is to take flat multifilament yarns and convert them into a three dimensionally stable bulked structures by forming continuous filaments into loops and core that supports these loops. This section describes the mechanism of loop formation in air jet texturing. A critical review of currently accepted mechanisms of loop formation is presented, followed by the proposed mechanism of loop formation. 4.2) Review of literature Many hypotheses have been put forward regarding the mechanism of loop formation in airjet texturing process. A review of the literature appears to show that there is no consensus among the researchers regarding any particular hypothesis. Earlier hypotheses were based on the assumption that the supply yarns needed to be pre-twisted and they do not apply to current processing technology where zero twist supply yarn is used, and hence will not be discussed here ) Mechanism according to Bock and Lunenschloss [40] Bock and Lunenschloss analyzed the mechanism of texturing using a Taslan type XIV jet. They suggested that the multi-filament yarn is opened in the nozzle by turbulence and/or gradients of the flow velocity, and places itself in a stream of high kinetic energy below the nozzle axis. With right-angled draw-of after the nozzle, an interlacing point forms above the axis, at the interface between two zones of the different flow states. The filaments blown through below this interlacing point pass through a zone of high air turbulence, and are decelerated by the subsequent drop of the dynamic pressure. When the filaments interlace, loops projecting from the yarn are formed by the differently sized filament bends ) Mechanism according to Acar, Turton and Wray [23] Acar et al disagreed with the above hypotheses by suggesting that though shock waves exist in free, undisturbed flows, they are destroyed by the presence of filaments in the actual texturing process. They also suggest that shock waves, their strength varying according to the particular nozzle type, are at least partially destroyed by the presence of filaments in the nozzle during the texturing process.. As nozzles providing varying degrees of shock strength 17

32 are apparently equally effective in producing commercially viable textured yarns, it was concluded by them that the effect of pressure waves on the filament s motion is negligible and that any loop formation mechanism based on the presence of such waves is probably invalid. The authors [23] observed that although flow from the texturing jet at the usual air pressures used in texturing is supersonic, turbulent, slightly asymmetric and non-uniform in profile. The mean velocity of the flow and degree of its non-uniformity increases with increasing air pressure. Their explanation of the mechanism of loop formation described below. At any instant, some of the filaments will be moving at faster speeds than others owing to the relatively greater fluid forces acting on them The free excess lengths provided by overfeeding the filaments enable the faster moving filaments to slip and be displaced longitudinally with respect to the slower moving filaments. The degree of these longitudinal displacements is affected by local drag and frictional forces instantaneously acting on the filaments and also by the overfeed. The textured yarn is withdrawn from the nozzle at right angles to the nozzle axis at the texturing speed. Since the filaments are entangled and formed into loops, the resultant textured yarn is shortened, and a tension is created of a magnitude determined by the effectiveness of the texturing. Figure 8: Mechanism of loop formation according to Acar et al [23] Thus, on one hand, the emerging filaments are blown out of the nozzle along the direction of the air stream at much greater speeds than the yarn-texturing speed; on the other hand, the tension created in the yarn pulls the leading ends of the emerging filaments in the direction of the yarn delivery. Whereas the trailing ends of the filaments are held within the core of the much more slowly moving textured yarn and are pulled downward while being kept fairly close to the nozzle 18

33 exit plane. The emerging filaments are, therefore, forcibly bent into bows and arcs by the fluid forces acting on them. These are then entangled with other emerging filaments, which are formed into fixed stable loops within the textured yarn. The increased tension resulting from loop formation and subsequent entanglement causes loop forming filaments to migrate towards the lower part of the nozzle (for the downward delivery) since they will assume the shortest possible path between the trailing and the leading ends. Hence, the filaments in the nozzle change their positions ) Mechanism according to Sengupta, Kothari and Srinivasan [47] According to these authors, turbulent, asymmetric fluid forces in association with intermittent compression shock waves open up the filaments and blow the overfed lengths out of the texturing nozzle at speeds much greater than the delivery speed of the yarn. The difference in the speeds between the leading and trailing ends, of the section of filaments under the action of fluid forces, causes the bending of the filaments in the form of loops and arcs ) Mechanism according to Nosov [13]: According to Nosov the loop formation begins at once on exit of the yarn from the nozzle, when the length of the elementary filaments becomes different due to non-identical tension and position in the yarn cross section. Since the place of each elementary filament in the section is not constant, migration of these takes place from the center to the periphery and vice-versa, and loops are formed in various sections of the starting yarns. More detailed reviews on mechanisms of loop formation can be found Acar [14] and Kothari [36]. Though all researchers attribute an improvement in quality of textured yarns by pre-wetting of the yarns, none of the above researchers have considered the impact of water on the mechanism of loop formation. Water is always used while air texturing in the industry and though most of it is blown away from the yarn surface before it enters the nozzle channel, water may play a very important role in the mechanism of loop formation by influencing the structure & stability of the core. 19

34 Apart from stating that the basic function of the nozzle is to form a supersonic, turbulent airflow stream, none of the above researchers elaborate on the functions of the nozzle as related to the loop formation mechanism. 4.3) Proposed mechanism of loop formation Among all the aspects of the process, the most critical ones for understanding the mechanism of loop formation are Operating sequence of the process Functions of the nozzle (Section 5) 4.3.1) Operating sequence of the process The sequence of operation on an air-texturing machine can be described as follows: Flat yarn unwinds from package (tensioning during unwinding is critical for the stability of the process, especially the start of the process) Yarn is overfed into the texturing box Overfed yarn enters the water bath in the texturing box Wet yarn (from which most of the water is blow away by the secondary flow of the nozzle) enters the nozzle (which is also in the texturing box) Looped (textured) yarn is stretched outside the texturing box Stretched yarn is wound on to the package OPTIONALLY: When using POY (partially oriented yarn) the yarn passes through a drawing zone before it is enters the texturing box. Also, to stabilize the yarn the textured yarn may be run through a steam unit to stabilize the surface loops before being wound on to the package to minimize yarn shrinkage 20

35 Figure 9: Process flow of air jet texturing (repeated here for convenience) ) Mechanism of loop formation in air jet texturing Along with producing a supersonic & turbulent airflow field inside the nozzle, a further function of the nozzle is to push the yarn out of the nozzle channel at high speeds. Due to this the yarn is under tension right up to the plane of air inlet (figure 10), where it is believed that the effect of overfeed takes place. As can be seen from the simulations (section 5.8.1), there is a sudden expansion of the airflow at the plane of air inlets and this leads to a turbulent, supersonic regime being created in and around that region. There are six circulation zones and a central zone where there is little or no circulation component. It may be hypothesized that as the filaments enter the primary airflow of the nozzle channel, they are not stable, but are being blown around randomly across the cross-section of the 21

36 channel (Figures 11 & 12). This may lead to some of them encountering one of the six circulation zones and be reoriented three dimensionally around axis of the yarn from their initial position, while the others may remain in the central non-circulation zone. Due to this random re-distribution of the filament across the yarn channel, and the large magnitude of forces involved in the airflow, some of the filaments may be bent and formed into loops which get stabilized due to the obstruction created by the other reoriented or intermixed filaments that don t allow the bent lengths to be straightened-out again. As the filaments speed away from the plane of air inlet, the geometry of the nozzle dictates that they converge back together around the central core. This mechanism may also point towards the reason, why most of the loops observed in air-textured yarns are horseshoe type loops, which are formed by pure bending forces. The properties of supply yarns that determine the loop formation are the bending rigidity of the filaments, their fineness, their surface properties (spin finish, surface roughness), number of filaments in the yarn and cross sectional shape of the filaments. It has been proved that the use of water while air texturing leads to a yarn of better quality. A very small amount of water is required for processing in air texturing. The water acts to keep the filaments together as a cohesive bundle and thus leads to fewer filaments being randomly redistributed around the yarn axis as they encounter the primary airflow (figure 12). In other words, in wet texturing, there is less obstruction to the three-dimensional reorientation of the filaments that are blown apart, as fewer free filaments would be present. This leads to their being redistributed more uniformly around the yarn axis, resulting in a more uniform air textured yarn structure. A more detailed discussion on the role of water in air jet texturing is presented in section 6. 22

37 Plane of air i l t Yarn Figure 10: The proposed mechanism if loop formation showing behavior of filaments inside the nozzle channel. The image also shows the plane of air inlet Dry Texturing Figure 11: Behavior of filaments at the plane of air inlet in dry texturing condition 23

38 Wet Texturing Figure 12: Behavior of filaments at the plane of air inlet in wet texturing condition 24

39 Section V Nozzles 25

40 5.1) Introduction The nozzle is the heart of the air texturing process. Increase in production speeds in air texturing over the years have been attributed to the development of newer more efficient nozzle designs. These newer designs have also lead to the production of more stable and uniform yarn structures. 5.2) Location of nozzle in the process flow The nozzle is usually placed just after the wetting unit inside the texturing box on an airtexturing machine. The jet (nozzle) is housed inside a metal casing, which also holds the baffle element as shown in the figure 13. Figure 13: Metal casing with baffle (left) and jet (right) [41] 5.3) Nozzle material The jet cores are usually made from sintered material, which is either high-grade ceramic or tungsten carbide [1]. There is no information available in the literature explaining the influence of jet material on textured yarn quality. 5.4) History of nozzle development Detailed history of the development of nozzles can be found in Acar et al [14] and Hoffsommer [8]. The main points are mentioned below. Acar et.al classified all texturing nozzles into two groups according to their structure: 26

41 1. Converging-diverging type i.e., a converging-diverging nozzle is situated at the yarn exit end of the nozzle assembly, e.g., Dupont s Taslan type 14 nozzle (Figure 14) Figure 14: Old Dupont nozzle [14] 2. Cylindrical nozzle, i.e., one or more air inlets opening at an angle to a cylindrical straight uniform main-flow duct of the nozzles, e.g., Heberlein s HemaJet (Figure 15) Figure 15: Mirlan Cylindrical Jet [14] This is essentially the same design as Heberlein Jets. According to Hoffsommer [8] the evolution of jet configurations can be divided into three groups: 1. Yarn entering into the air stream at an oblique angle (typified by Dupont 9 ) 2. Yarn entering into the air stream on the same axis as the exiting air stream (Dupont type 10 & 11 ) 3. The addition of external plates and baffles into the area of the air stream exhausting out of the venturi: Dupont type 14 & 15, EMAD Scorpion & Cobra, and 27

42 Heberlein Jets ( Mirlan, a Czech nozzle is thought to be the basis for design of Heberlein) (Figure 16) Figure 16: Evolution of Taslan nozzle [14] 5.5) Review of literature Boch and Lunenschloss carried out most of the earlier research on understanding the design and function of nozzles [21] and this was followed by Acar and his colleagues who did most of their work on airflow and nozzle [2]. Since then not much has been published in the domain of nozzle design and function. There have been some publications on simulations of airflow inside an air-texturing nozzle including Rwei [7] in early Some important aspects touched upon by these researchers are discussed below. Acar et.al [2] studied the airflow inside the nozzle using a scaled-up model of Heberlein s HemaJet nozzle. They used shadowgraph & interferometer techniques to study The effects of diverging trumpet-shaped exit on the airflow, which they claimed increased the non-uniformity of air flow outside the nozzle exit 28

43 Pressure distribution inside the nozzle Turbulence: Though not succeeding in directly measuring the turbulence, the authors hypothesized that the turbulence and movements of the filaments mutually contribute to each other s activity. They confirmed the work done by earlier researchers [21] that the airflow inside the nozzle was: Supersonic Turbulent: This is a critical criterion for the texturing of continuous filament yarns. According to Smelkov [5], in laminar airflows the gas layers move relative to each other without mixing. Only in turbulent flows particles are intensively mixed. Therefore, he concluded that air textured yarns can only be produced with turbulent flows Of non-uniform velocity distribution: The degree of this non-uniformity in the velocity distribution increases with increase in air pressure. They have also highlighted the importance of non-uniformity of air velocity profile at the jet exit. Considerable variations in the air velocity lead to the different filaments being blown out at different speeds. They claimed that an increase in air pressure would result in increased longitudinal displacements. They concluded that the trumpet shaped divergent exit was responsible for this non-uniformity at the exit They initially claimed that [2] the staggered position of the air-inlet bores induces a swirl to the flow inside the nozzle. This was supported by an argument stating that this swirl can be observed as the twist inserted in a bundle of stationary filaments placed in the nozzle. Using Fluent simulations the current research shows that though there is a swirl induced in the yarn as Acar et.al have indicated, there is no swirl in the airflow without the yarn. This may be due to the fact that even though the air inlet bores are longitudinally staggered, they are not inclined at an angle to the radial direction. The authors have acknowledged this correction in a subsequent publication [3], which also elaborates more on the points discussed above. 29

44 5.6) Anatomy of the nozzle: Demir et.al [3] described the anatomy of an air texturing nozzle as below. The commercial texturing nozzles consists of a cylindrical main duct into which one inlet hole (Heberlein T-100 W ) or three inlet holes (T 314, T 315, T370 ) of smaller diameter are obliquely opened. The compressed air is admitted into the main duct through this (these) inlet hole(s), while they yarn follows a straight path in the nozzle. The yarn inlet section of the cylindrical duct is called the secondary flow section and the other end of the duct is the primary flow section usually with a trumpet-shaped divergent exit. The two most common features of all cylindrical nozzles are that the ratio of the sum of inlet hole areas to the main duct area is always less than unity and the inclination of the main duct is at the same angle. The compressed air that is supplied to the nozzle flows through the inlet holes and is divided into two main streams. The primary flow, which has a large momentum due to the inclination of the inlet hole, flows vigorously and creates a supersonic, asymmetric, non-uniform and turbulent airflow. The secondary flow, having a smaller momentum, flows in the opposite direction to the primary flow and is always subsonic and uniform. Air consumption results by Versteeg et al [4] have shown that the flow is choked at the nozzle inlets and that subsequent area enlargement ensures that flow characteristics in all texturing nozzles correspond to those of convergent-divergent nozzles. Table 2 gives the geometrical details of the nozzles used in the current research. Figure 17 helps define the parameters of the table visually. Table 2: Specifications of nozzles used in current research Inner Diameter (mm) Outer Diameter (mm) Air Inlet Diameter (mm) Oblique inlet length (mm) Channel Length (mm) Volume of cylinder (mm 3 ) Heberlein Nozzles Area (mm 2 ) Area (mm 2 ) T T T T

45 Figure 17: Typical air jet design [7] Different nozzles are required for different denier yarns because of consideration of process stability. Mass flow rate = density x area x velocity (Kg/sec) Figure 18: Mass flow rate 31

46 Figure 19: Comparison of higher (left) and lower (right) denier yarns being textured using same nozzle According to the above equation and as shown in figures 18 & 19, for the same diameter nozzle channel, a larger denier yarn will present more surface area to the incoming airflow, thereby being pushed further away from the nozzle exit due to its higher velocity and adversely affecting the process stability. This may also the same reason why processing is more stable in wet texturing. Water causes the filaments to become more cohesive, thereby reducing the overall diameter of the yarn and thus the surface area presented to the airflow. 5.7) Functions of the nozzles When supplied with compressed air, a complex airflow pattern is produced inside the nozzle that along with rest of the air texturing machine components produces a high bulk yarn structure. The functions of the nozzle are: To push the yarn out of the main nozzle channel, thereby assisting in tensioning the yarn up to the plane of air inlet and maintaining process stability 3D spatial displacement of discrete as well as multi-filament groups around the axis of the yarn and within the central core. The former gives rise to the 3D surface bulk of an air textured yarn, while the later helps form the intermingled core that not only 32

47 assists in supporting the surface loops against being pulled out, but also forms the barrier regions into which subsequent filaments run into, forming loops To open & separate discrete elements as well as multi-filament groups from the main body of the yarn thereby assisting in distributing them three-dimensionally around the yarn axis Along with water, nozzle helps remove spin finish from the filament surface. Since the position of individual filaments of the yarn within the nozzle channel may be random along the cross section of the nozzle channel (due to the airflow), it may be assumed that this removal of spin finish may be non-uniform along the filament length as well as between different filaments from periphery to the inner core of the yarn. This non-uniform removal of spin finish may influence the fiber to fiber frictional properties in the final air textured yarn and may be one of the contributing factor to the instability of the yarn. This fiber-fiber frictional relationship may also be responsible for the overall non-uniform structure of air textured yarns and may also serve to explain the difference between wet textured and dry textured yarns 5.8) Simulations of airflow inside the inside the nozzle 5.8.1) Introduction One of the main challenges in understanding the process of air texturing is to visualize what goes on inside the nozzle. A few attempts have been made by researchers to look inside the nozzle. Acar et.al [45] using partial glass models and high-speed photography attempted to visualize the airflow inside the nozzle, but were not successful. Rwei [7] using CFD simulations defined a texturizability parameter. However, their simulations were never related to other aspects of the process, nor were there any attempt at using the understanding obtained to support the mechanism of loop formation. Simulations using Fluent were carried out in this research to Understand the airflow inside the currently available industrial nozzles Use this understanding to supplement the proposed mechanism of loop/core formation 33

48 Though the simulations were carried out for the geometry of nozzle T 351, which has three air inlet holes, the findings may be used to understand the behavior of single air-inlet hole nozzles, with minor modifications ) Introduction on compressible flow in nozzles A supersonic expansion can be obtained by imposing a pressure ratio less than P / P 0.5 b 0 across a convergent nozzle driving an isentropic flow expansion where P 0 and P b are the stagnation and the background pressures, respectively. The expansion through a convergent nozzle will always take place in a subsonic regime regardless of the amount of the applied pressure ratio. Outside the converging nozzle, depending on the pressure ratio, the flow will supersonically expand to pressures even much lower than the background. The flow in the current nozzle system used in the research was far from the classical onedimensional compressible inviscid (non-viscous) flow traditionally employed for nozzle flow problems. A system with sudden turns having sharp edges and a nozzle with a small diameter-to-length ratio can be simulated more realistically using viscous models. A steady turbulent compressible model was been assumed for the flow regime inside the system. The airflow field wass obtained using finite volume method for solving the Navier-Stokes equations together with the ideal gas equation of state (FLUENT 6) ) Simulation details The airflow in an air-texturing nozzle is very complex due to viscosity effects, which increase the difficulty of the simulations. If a non-viscous flow had been assumed the simulations would have been relatively simpler. However, the flow in the nozzle system used in the current research was far from the classical one-dimensional compressible inviscid (nonviscous) flow traditionally employed for nozzle flow problems. A system with sudden turns having sharp edges and a nozzle with a small diameter to length ratio can be simulated more realistically using viscous models. A steady turbulent compressible model was assumed for the flow regime inside the system. The airflow field was obtained using finite volume method for solving the Navier-Stokes equations together with the ideal gas equation of state. 34

49 The convergence criteria for these simulations were defined in terms of the residuals. The residuals provide a measure of the degree to which each of the conservation equations is satisfied throughout the flow field. The residual for each flow variable gives a measure of the error magnitude in the solution in iteration. A three orders of magnitude reduction in the residuals is normally required for a solution to be considered converged. It is recommended however, to consider several orders of magnitude further reductions in the residuals to ensure that the simulation is not trapped in any intermediate local convergence. The state of convergence is achieved once a reasonable answer is obtained and when its properties do not change by additional iterations ) Simulation methodology ) Pre-processing At this stage the geometry of the nozzle (Figure 20 & 21) for simulation was defined using Gambit and the simulation domain was defined and subdivided into a large number of cells (Figure 22) using the same software ) Numerical Solution The governing equations (Appendix A) in each cell zone were solved using Fluent CFD code ) Post-processing After the simulation was converged, the velocity and pressure contours were produced to show the flow field (see figures 23-26) 35

50 Figure 20: Air texturing nozzle (T-351) Top view Figure 21: Air texturing nozzle (T-351) Side view 36

51 Figure 22: Grid distributions on the nozzle wall 37

52 2 Darker colors indicate higher velocities Figure 23: Velocity vectors 2

53 Figure 24: Velocity vectors close to nozzle axis at the air inlet plane 3 3 Darker colors indicate higher velocities 39

54 Figure 25: View of 6 circulation zones 4 4 Darker colors indicate higher velocities 40

55 Figure 26: Contours of velocity magnitude 41

56 5.8.5) Analysis of simulation images in relation to the mechanism of loop formation The mechanism of loop formation in air texturing is a very complex phenomenon and as has been noted in the literature review section, there seems to be no general agreement regarding how it works. Figure 23 shows the velocity vectors of the airflow field inside the nozzle and also the velocity vectors in the plane of the three air-inlet holes. The contours of velocity magnitude shown in figure 26 indicate that the maximum velocity of the airflow is obtained in the immediate vicinity of the exit of the air-inlet holes and gradually decreases towards the nozzle main channel exit. In this region of the main channel (where the velocity is lower) there is also a radial gradient, with maximum velocities in the center and lower velocities towards the outside. Figures 24 and 25 show that there are distinct circulation zones (six) in the plane of the air inlet holes. It can also be see that there is a central zone where there is no circulation. The turbulence is maximum at the plane of air inlet and decreases as one goes away from it towards the nozzle exit. Now, assuming that there is no significant variation in the airflow patterns in the presence of the yarn, a multifilament yarn entering the nozzle stream may experience the following: At any given time a certain portion of the continuous filament yarn is at the plane of air inlet, while it immediate front region is located in the primary airflow stream and the immediate back region in the secondary airflow stream. As the yarn travels away from the plane of airinlet, the geometry of the nozzle dictates that the filaments converge back together. So the filaments that have not felt the full impact of the air inlet (i.e., ones that are more towards the center of the yarn than on the periphery) get relatively more compact and form the core, while the filaments on the periphery of the yarn that have had the full impact of the airflow are not only three dimensionally redistributed around the yarn axis, but also are pushed out at greater speeds forming the surface bulk. Their behavior is as shown in figure

57 Figure 27: Hypothesized behavior of peripheral filaments impacted directly by airflow at the air-inlet plane Further use of the understanding obtained using these simulations is made in the section on the effect of processing parameters on the air textured yarn properties, 43

58 Section VI Role of water in air-jet texturing 44

59 6.1) Introduction It is widely accepted that the use of a small quantity of water (According to Acar [42] amounts as low as 0.1 liter/hr is sufficient) helps improve the quality of air-textured yarns considerably. The supply yarn is normally wetted just before it is fed into the texturing nozzle by passing it through a water bath or a wetting unit, which can either be separate or integrated in the nozzle. The most recent systems use separate wetting heads that are positioned just before the nozzle inside the texturing box as shown in figure 6. In these newer systems, fresh clean water is always used, and hence only the washed off spin finish is drained away with the used water, reducing the likelihood of nozzle contamination. After presenting a review of the current literature, this section describes the proposed theory on the role of water. High-speed imaging and optical microscopy were used for this purpose. 6.2) Review of literature A detailed and thorough review of theories proposed to explain the role of water in air texturing can be found in Chand [27] & Dani [39]. Acar et al [26] proposed, the most widely accepted explanation of the role of water. They used on-line measurement of yarn tension, visual assessment and tensile testing results of air-textured yarns to understand the role of water. They claimed that water acts as lubricant and reduces the friction between the filaments and other contacting parts. This results in an increase in the resultant forces acting on the filaments. Furthermore, the reduction in inter-filament friction results in larger longitudinal displacements between filaments, facilitating better loop formation and entanglement formation, thus generating improved textured yarns. They also suggested that the addition of suitable wetting agents into the water might change the frictional properties further and aid in texturing. The same would result from a slight increase in the water temperature. 45

60 The above researchers used higher measured yarn tension values in wet texturing as an indicator of the reduced fiber friction in wet condition. However, they have not explained why this is the case. It may be that in the presence of water, there is relatively more removal of spin finish than in dry conditions, which may lead to higher inter-filament frictional values. This increased friction may help form a more tightly entangled core structure that may be causing the higher tension values. Acar et al [26] also touched upon two problems that may be associated with the use of water, namely, the contamination of the nozzle by the removed spin finish & other impurities, and undesirably high moisture content in the final yarn. The first issue becomes very significant when texturing high modulus yarns due to the relatively large amount of spin finish. Due to the hydrophobic nature of most industrial yarns, the second issue may not be that significant. Schwarz [28] stated that by wetting the incoming yarn the process velocity can be raised significantly. However, he has not provided any proof. The current research shows that wetting enhances the stability of the process by forming a more cohesive structure. This presents a smaller surface area to the incoming air and prevents the yarn from being pushed too far away from the nozzle exit. This may result in a more stable process and hence higher production speeds. None of the researchers above have discussed the bunching effect as a result of water. As mentioned in section 2, one of the important criterions for loop formation is a critical stiffness, below which the filaments exhibit local crinkling and waviness. To overcome this problem a group of filaments combine to achieve this critical stiffness and hence become capable of loop formation. This same bunching effect may also cause problems when dealing with yarn composed of ultrafine denier filaments ) Issue with the explanations of role of water in texturing provided in the literature Most of the earlier researchers have considered the role of water from the viewpoint of final yarn properties, and have pointed out an improvement of textured yarn quality in the presence of water. The problem with this is that there is the lack of any definition of yarn quality in the textured yarn domain. The researchers have not pointed out whether this 46

61 improvement in quality relates to an increased overall yarn diameter, improved stability or a more uniform yarn structure. 6.3) Proposed role(s) of water The primary role of water is to stabilize the process by imparting cohesion to the yarn structure as it encounters the airflow of the nozzle. The majority of the tension in the yarn is due to the airflow (section 8.5), which is constantly pushing the yarn out at higher speeds than it is being taken up (Figure 28) but water assists in maintaining this tension to the core structure once the yarn is inside the nozzle. Even though most of the water is blown-off from the yarn, the compactness of the core (due to application of water) is not destroyed. Section below which gives the results from the single-droplet experiments clearly show that even if one drop of water is retained in the structure, it will keep the individual filaments together and maintain the core. Figure 29 and 30 show the yarn at the water bath and the nozzle entry respectively. It can be clearly seen that there is a coat of water as the yarn exits the water bath, but most of the water is blown away by the secondary airflow by the time the yarn reaches the nozzle entry. Also, as indicated in the simulation section of this research, the non-circulation zone present in the nozzle facilitates the retention of this core structure. Figure 28: Yarn under tension till the plane of air inlet 47

62 Water bath exit Water bath entry Figure 29: Yarn emerging out of the water bath surrounded by a sheath of water (Left) and yarn just entering the water bath (right) Nozzle Figure 30: Wet yarn just entering the nozzle The secondary albeit a very important role of water is to create multi-fiber groups that achieve the critical stiffness required for bending/buckling and hence loop formation. This effect primarily manifests itself as increased number of buckled loops (table 40) in a wet textured structure compared to a dry textured structure. As has 48

63 been noted before, the formation of buckled loop requires a critical stiffness that may not be achieved by individual filaments that are too fine and have a tendency to locally crinkle. However, this bunching may be undesirable if the yarn is composed of very few ultra fine (<= 1 denier) filaments, as can be seen in the case of Zylon. As shown in the figure 31, the cohesive/bunching effect of water prevents the efficient separation of filaments from the surface and bends and twists the whole yarn Figure 31: Zylon produced under wet texturing conditions Another important effect of the use of water is the removal of spin finish. It can be explained as follows. Combined with the influence of air, water helps remove the spin finish from the surface of the filaments undergoing texturing. It has been reported [39] that in wet texturing a significant portion of the spin finish on the supply yarn gets removed. The removal may be non-uniform due to different amount of water being retained 5 in the core of the yarn than on the periphery. In other words, different amount of water + spin finish is blown from fibers on the outer periphery than those that are at the core. The impact of this non-uniform removal may be seen at the textured yarn testing stage, where due to the variable spin finish content along the length as well as the diameter of the yarn, the frictional properties of the fibers may vary considerably giving a stick-slip kind of appearance to the stress-strain curve. Based on the current hypothesis, this stick-slip appearance may be more of a characteristic of wet textured yarn than of dry textured ones. 5 However, the application of water is same for all the filaments as they are wholly submerged in the water bath as they pass through it 49

64 6.4) Wetting behavior imaging The relationship between water and filament yarns was explored using imaging techniques. Initial experiments were conducted by placing the filament yarn in a Petri dish filled with water and observing their behavior. However, it was very difficult to image the yarns in this set up. It was thus decided to study the interaction of water and yarn under an optical microscope system. The yarn was put on top of a slide and backlit. Using a dropper, one drop 6 of water was placed at one edge of the yarn. Using a motion capture feature of the customized image analysis software, a sequence of images was captured in bitmap format. These images were subsequently combined in Macromedia Flash to produce a movie ) Interesting findings from the Petri dish experiments A Petri dish was filled with water and a multifilament yarn loosely held at two ends was gradually submerged in water. It was observed that when the yarn first touched the water surface, it instantaneously spread open, but as it was submerged further, all the filaments combined together. This started as soon as the last part of the yarn lost contact with the atmosphere and was completely under water. This compact condition of the yarn was not altered on being further submerged When the submerged yarn was pulled out of the Petri dish from one end, due to the effects of surface tension, it still maintained its cohesive structure and did not spread open. This is exactly the condition observed in the water bath system of an airtexturing unit (figure 32) All fibers tested with the exception of Polypropylene exhibited similar behavior Similar behavior was also observed when the yarn was dipped in a low surface tension fluid (Methyl alcohol), a non-polar fluid (lighter fluid) and when the Zeta potential effects of water was reduced by the addition of salt Further research into the spreading of yarns on surface of fluids was not conducted as it was out of scope for the present research. 6 In all cases, the size of the drop was roughly equal to the size of yarn 50

65 6.4.2) Interesting findings from single droplet experiments using optical imaging As mentioned above, using a dropper, one drop of water, approximately of the same diameter as the yarn, was placed at one end of the yarn. As shown in figure 32, as the water droplet migrated from right to left along the yarn length, it pulled the entire yarn together Figure 32: Effect of drop of water migrating (from right to left) in a loose multifilament yarn However, as more and more drops of water were added, the yarn eventually spread open. Thus it may be said that, although not to the same degree, the effect of one drop of water (of the same size at the yarn) on the multifilament yarn is the same as that of the yarn being completely submerged in water, while when more water is available for the water to interact with (i.e., on addition of more amount of water) it migrates across and through the filaments to merge with the other water molecules, forcing the yarn open. When similar one-drop experiment was carried out with a surfactant instead of water, there was no change in the yarn profile when only the drop of surfactant 51

66 alone was applied to the yarn, but on addition of a drop of water to the surfactant, similar results as above were observed the fibers in the yarn again combined together. Further research into this activity was out of the scope of this study and hence was not pursued. 52

67 Section VII Experimental 53

68 7.1) Introduction This section describes the main experimental aspects of the research. These include experimental components such as the texturing machine and analysis instruments, along with the supply yarns and techniques used for air textured yarn analysis. 7.2) Air texturing unit The air-texturing machine used in this research was a Staehle RMT-D single spindle airtexturing unit, which is widely used and popular in the industry. The main features of the machine are: Four heated godets two for single end texturing and then two for supplying the effect yarn during core-effect texturing. Since these are heated godets, POY feed can be used for air texturing Texturing box that houses the nozzle and water bath system Drain system to take away the excess water as well as the spin finish removed along with it Two rubber covered rollers that form the stretch zone to prevent unnecessary slippage of the yarn All rollers are driven by a direct motor thereby affording precise control of the speeds Precision cone winding unit to give uniform packaging Post heating zone A vertical creel at the back end of the machine that can feed up to 16 different packages. The creel designed so as to have good tension control and be convenient ergonomically for yarn threading Maximum speeds: ~1000 m/min according to company brochure 7.2.1) Auxiliary equipment related to the trials Compressed air supply: The maximum air pressure attainable through the main supply system at school was 70 PSI, which was too low for the purposes of air texturing high strength yarns, and hence a compressed air supply system was 54

69 installed. The maximum attainable air pressure with this new Kaeser compressed air system was approximately 180 PSI Tension measurement system: A three-pin tension measurement device was used during the trials to measure the on-line tension in the yarn as it traveled through various regions of the air texturing system. Tension values obtained in various zones were used to correlate the image analysis and tensile testing results ) Limitations of the machine as related to the experiments performed in current research Maximum speeds: Though the machine manufacturers claim that the unit could go as high as 1000 mpm, when air texturing high modulus yarns (i.e., higher linear density, number of filament yarns), the machine could barely handle 300 m/min. The main problem was getting the process started This machine was not designed to run high strength yarns like Kevlar, Dyneema etc and there was wear of machine components, especially the yarn guide at the winding unit. Also, the yarn guide size was also not sufficiently big enough to handle the larger sized yarns once they were textured. The spin finish content of high strength yarns is comparatively higher than for ordinary yarns and there was accumulation of spin finish on the godet rollers as well as the guides, which resulted in frequent stoppages for cleaning The creel did not have a suitable tensioning system to impart sufficient tension to high strength yarns. Spring-loaded tensioners had to be added for both core and effect side feeds. Furthermore, it was found that for texturing high modulus yarns, a vertical package unwinding was more preferable than the inclined one supplied by the creel The post-heating system was not used due to lack of a suitable threading device. Also the use of only fully drawn yarns removed the need for use of post-stabilizing heat Lag: Due to the design of the air-texturing box, the machine could not be threaded when it was running. This caused problems in starting the process because of a certain amount of lag between the starting of various components. In other words, it took a few seconds for the entire machine to come to some kind of dynamic 55

70 equilibrium. This problem was exacerbated when trying to run the machine at higher speeds. Number of wraps: The number of wraps as recommended by the machine manufactures (8 in the feed zone and 4 in the stretch zone) did not apply to high modulus yarns and had to be revised according to the yarn used Frequent stoppages: Due to lack of a drainage system, the machine had to be stopped frequently to dispose of the water from the water bucket that collected the blown of water and spin finish from the texturing box 7.3) Nozzles/Jets Details on the nozzles used in this research have been described in section ) Nozzle maintenance The main issue with nozzle maintenance was the accumulation of spin finish. High strength yarns typically have a greater amount of spin finish compared to their ordinary counterparts. To clean this accumulated spin finish and maintain consistency of nozzle performance over several trials, the nozzle was cleaned using an ultrasonic bath periodically. The ultrasonic bath was filled with water and soap solution and the nozzle was placed in the bath for approximately 30 minutes, as per Heberlein s recommendations. 7.4) Supply yarns Table 3 shows all the specifications of the yarns used in current research.. The yarns are listed in the columns from the highest to the lowest bending rigidities. An explanation of each column follows: Denier/Number of filaments: They represent the supply yarn (or feed yarn) parameters. The numbers are as supplied by the manufacturer. However, the denier was verified by using the cut and weight technique. The number of filaments was indirectly verified by using the diameter values of individual fibers (which was measured using an optical microscope) and using that to calculate individual fiber denier (dpf) and then dividing the total denier by the dpf value 56

71 Density: From textbook [43] Individual filament diameter: Measured using optical microscope system Continued on page 59 57

72 Table 3: Yarns used in the current research with relevant details. The yarns are listed from top to bottom in the columns starting from highest bending rigidity (Spectra ) to lowest bending (Technora ) Polymer Type Trade Name Denier Number of Filaments Denier per filament Density g/cm3 Diameter (microns) Single Fiber Bending Rigidity (x Nm 2 ) Single Fiber Torsional Rigidity (x 10-9 Nm 2 ) Tensile Modulus (GPa) Shear Modulus (GPa) UHMWPE Spectra Polyester Vectran PBO Zylon UHMWPE Dyneema UHMWPE Dyneema Aramid K Aramid Kevlar Aramid Technora

73 Tensile Modulus (Individual fibers): An attempt was made to obtain individual filament tensile modulus by using a Dynamic Mechanical Analyzer. However, the results were far from satisfactory due to several issues with the DMA as discussed below o Handling of samples before testing: The DMA had a resolution as high as 5 nanometers and the way the samples were handled before testing could easily change what the DMA was measuring o Straight alignment of fibers in the clamps: It was almost impossible to verify the straightness of alignment of the fiber, because of the extremely small sizes of the fibers, the nature of the clamps and also due to the ergonomics of the DMA itself o Clamps: The DMA clamps were not suited for the fine size of fibers that were being tested. Further, since the clamping mechanism was manual, as opposed to pneumatic or hydraulic, more errors may have been introduced o No break: No filament breaks were observed on completion of testing. This may mainly be because of the high strength nature of the fibers and the low testing range of the equipment. This was observed for a bundle (yarn) as well as for single fibers o Slippage: It was very difficult to detect any fiber slippage in the clamps o Submersion testing: While attempting to measure the moduli of individual fibers in wet condition, the main problem was again the clamps. It was seen that no matter what fiber was used, the readings were exactly the same for wet and dry conditions. The submersion clamp in the Q800 series DMA was a complicated set up and worked only while using films, but not while using single fibers o Lack of familiarity with the equipment: In summary, the issue was the lack of familiarity and trust of the results from the equipment. The machine did not have suitable clamps for use with single fibers, and did not have enough range for use with bundles, specifically for high modulus fibers Therefore, the values of the individual fibers tensile modulus seen in the table 3 were taken from earlier research findings [39] and published data from manufacturers. 59

74 Shear Modulus: The shear modulus was calculated using the Poisson s ratio and tensile modulus Bending & Torsional rigidities: These were calculated using the tensile and shear modulus values mentioned above 7.5) Air texturing trials The above described equipment and supply yarns were used to conduct air-texturing trials with an aim to understand the fundamental aspects of the process such as: Mechanism of loop formation Influence of water Effect of change in processing parameters on final yarn properties Over the course of the research several trials were conducted. Trial1: Polyester tire cord yarn, Kevlar and Dyneema yarns were textured under different conditions as shown in the table 4. The main aim was to produce yarns under different processing conditions and tested for tensile. The geometry of the nozzle is the same for all the types used in the experiments and their function is just different in degree rather than in its fundamental effect. Table 4: Trial 1 Varied Kept Constant Overfeed 7 : 15%, 25%, 35%, 45%, 55% Stretch: 10% & 15% Air pressure: 100 PSI Speed: 100 mpm Stretch 8 : 6%, 8%, 10%, 12%, 14% Overfeed: 15% & 45% Air pressure: 100 PSI Speed: 100 mpm 7 Overfeed: Negative speed differential of rollers: Back roller moves faster than front roller 8 Stretch: Positive speed differential: Front roller moves faster than back roller 60

75 Air pressure (PSI): 100, 120, 140, 160, 180 Stretch: 15% Overfeed: 35% Speed: 100 mpm Speed (mpm): 100, 200, 300, 400, 500 Stretch: 15% Overfeed: 35% Air pressure: 100 PSI Trial 2: The same yarns were textured again under identical conditions as in table 4. This retrial had to be done, as all the tensile testing data was lost due to a computer crash. Trial 3: A series of different high modulus yarns as shown in the table 5 were air textured. These were the yarns on which all the image analysis was done. Yarn trade name Denier/Number of filaments Spectra 650/120 Vectran 1500/300 Dyneema 440/195 Technora 1520/1000 Kevlar 850/835 Zylon 166/120 Table 5: Trial 3 Overfeed Stretch Speed Air Nozzle (%) (%) (mpm) pressure (PSI) T T T T T T 351 Apart from the trials listed in table 5, Kevlar (850/835) was produced at different air pressures for image analysis. Also, to understand the influence of different filament numbers for yarns with similar denier per filament; Dyneema (780 filaments) and Dyneema (195 filaments) were produced. Further Dyneema yarn (440/195) yarn was produced under different conditions of overfeed, stretch, air-pressure and stretch. 61

76 Trial 4: Finally, in order to see whether all the findings applied to apparel grade yarns also, Dacron (Polyester) yarn was textured under different conditions of overfeeds, stretch, airpressure and stretch and used for image analysis. Trial 5: For high-speed imaging, several yarns were textured under different conditions as shown in the table 6. Table 6: Yarns used for high-speed imaging Yarn Conditions imaged Polyester tire cord yarn Dry and wet conditions Effect of stretch Effect of air pressure Effect of change in speed Different types of yarns Monofilament Polypropylene two ply yarn Single end texturing (To observe swirl) Kevlar Dry and wet conditions Effect of air pressure Effect of overfeed Polyester tire cord and Dyneema - Coreeffect Different effect overfeeds 440 denier high strength polyester Wetting unit Dry and wet conditions None of the yarns in table 6 were used for tensile tests. Due to the very large number of images obtained using high speed capture, only a partial image analysis (section 8.5.5) was carried out on the high-speed image captures in various zones, under varying conditions, to support the other experimental and theoretical work. 62

77 7.6) High-speed imaging High-speed imaging was utilized to support the proposed theories. The images were combined later into a movie and they helped better understand the behavior of filaments/forces in several different zones as below (figure 33): Water bath entry Water bath exit Nozzle entry Nozzle exit Just outside the texturing box On the stretch rollers Stretch zone (between the stretch rollers) Before being wound on package Redlake Motionscope 8000S series high-speed image acquisition system was used for high-speed imaging at a capture rate of 2000fps and a play back rate of 2000fps with a shutter speed of ms at a magnification of 2X. Imaging was done in above zones for different conditions of overfeed, stretch, dry, wet, air pressure, speeds, number of filaments, and processing type. High-speed movies from all the zones were coupled together and analyzed to get an overall picture of the process. The images were converted into movies using Macromedia Flash & AnimaPro software s. The conversion from images to movies in Flash lead to some loss in the quality of the movies, as the very large sized tiff files had to be first converted to a Macromedia Flash compatible png file. However, there was no such problem with AnimaPro that made movies straight out of tiff files. 63

78 Yarn outside texturing box Yarn in stretch zone Textured yarn just before it is wound on package Yarn on stretch roller Yarn leaving nozzle Figure 33: Regions of air texturing process that were imaged using high-speed photography (Images at different magnifications) Yarn entering nozzle Yarn entering water bath Yarn leaving water bath Supply Yarn 64

79 7.6.1) Limitations Halogen lamps used for lighting the texturing box were burning the rubber surrounding the texturing box. This prevented imaging the region for an extended period The design of the texturing box (which is made out of plexi glass and mists up in presence of water), the intensity of the halogen lights and the size of the motion capture unit did not allow much closer proximity to the texturing jet, which in the absence of powerful lenses was the only option to get an even closer look at the process While imaging in wet condition, there was substantial amount of mist generated that prevented higher magnification imaging The design of the texturing box and the of the machine overall prevented the lighting of the region of interest with enough intensity to be able to capture really high magnification/high frame rate high speed movies The three dimensional rotational movement of the yarn during process meant that the yarn went in and out of focus while imaging. There was no easy solution to this problem, but the overall imaging was very effective Size of the movies: The movies from the Motionscope system could be saved in different formats; the best among them was to save them as tiff images. However, these required the largest amount of storage space. Processing them into movies would have required an even larger amount of storage space and faster processing computers. Due to this, among other considerations each movie capture was restricted to five seconds, giving around 2500 images 7.7) Individual filament diameter measurements The diameters of individual filaments were measured using an optical microscope system. Twenty images were taken and ten readings on each image were taken to give an average value of the diameter. Individual filaments were cut from the yarn bundle and put on a slide, and then an oil of different refractive index was put on the filament. Finally the filament was covered with a cover slip and imaged. 65

80 7.8) Imaging of final air textured yarn structures The yarns produced under different conditions were imaged using an optical microscope system to evaluate their structure and properties. The yarn package was taken and placed next to the microscope platform, from where it was unwound and placed under the lens of the microscope. It was then backlit and imaged. Once this section was imaged, they yarn was pulled ahead and the next section was imaged and so on. All images of a particular yarn were obtained under the same magnification, however the magnification had to be varied based on the yarn diameter ) Limitations The main limitation is that since the air textured yarn has a three-dimensional structure, and a non-stereoscopic imaging technique was used, all the analysis was based on a two dimensional structure. However, the air textured yarn structure is similar around the fiber axis and the random nature of sampling may avoid creating any bias that would have affected the analysis. 66

81 7.9) Image analysis algorithm A customized algorithm as described below was developed to analyze the images. Step 1: The image(s) (bitmaps) were imported into Photoshop (Figure 34) Figure 34: Step 1 Step 2: The image was centered using the line tool and the angle of rotation menu (Figures 35-37) Figure 35: Step 2 67

82 Figure 36: Step 2 Figure 37: Step 2 Step 3: The Posterize feature of Photoshop was used to reduce the vinyeting effect. The impact of posterizing can be seen by comparing figures 39 & 40. Figure 39 shows lower level of posterizing, where there are only black and white pixels. Figure 40 shows higher levels of posterizing where a greater range of shades between black and white can be seen. At lower values of posterizing, a few loops and other features present on the yarn surface are lost. 68

83 Figure 38: Step 3 Figure 39: Step 3 69

84 Figure 40: Step 3 Step 4: The grid option in Photoshop was activated to help split the image in three equal sections (Figures 41 & 42) Figure 41: Step 4 70

85 Figure 42: Step 4 Step 5: The top & bottom Minimum Core boundaries were established using the guides as shown (using high magnification) (Figure 43) Figure 43: Step 5 71

86 Step 6: The top & bottom Maximum core boundaries were established using the guides (using high magnification) Figure 44 Figure 44: Step 6 Step 7: The maximum and minimum core diameters were measured using the line tool and shift key. The shift key ensured a perfectly straight line. Figure 45: Step 7 72

87 Step 8: Individual and bunched loops were counted for all three zones Step 9: The number of buckled loops in each zone were counted and added up to give a single count Step 10: At high magnification the yarn diameter was measured along the length of the yarn as shown in figure 46. A total of seventy-five readings were taken for each image analyzed Figure 46: Step ) Limitations of the above method Slow: As the image analysis was carried out manually, it was really slow. This restricted the number of images that could be analyzed. However, the purpose of this method was to evaluate a pattern rather than identify numerical significance of particular parameters The maximum and minimum core positions for the top and bottom of the yarn were determined visually It was visually difficult to separate out and count the multi-fiber bunches. To account for this, only the most clearly distinguishable fiber bunches were counted. Individual filament counting was easier 73

88 7.9.2) Information obtained from the above procedure The following yarn structure properties were obtained: Overall yarn diameter: As determined by step 10 Best Core diameter: As determined by steps 6 & 7 Number and type of loops The statistical mean and standard deviation were used to describe the yarns from the above data. Though a similar procedure was used to analyze high-speed images, due to the very high number of images that had to be analyzed, only the core diameter was measured. 74

89 Section VIII Structure and properties of air textured yarns 75

90 8.1) Introduction From the viewpoint of end use, the main aspects of interest in air-textured yarns are its bulk (yarn diameter) and instability, which means the retention of the texture (loops) under applied load such as those encountered during further processing of the yarns. The bulk may be considered a characteristic of the number, type & frequency of loops on the yarn surface, while the instability, may be dependent on the integrity of the core structure, which in turn may be dependent on the level of intermixing of the yarn, fiber-to-fiber frictional relationship and the extent of removal of the surface finish. Higher bulk and lower instability values have traditionally been used as indicators of the quality of air textured yarns. However, there is no agreement about the definition of quality in the published literature. A further complication is that, what may be considered good quality for one end use may not be acceptable quality for another end use. Inspite of this disagreement, researchers have consistently used the term improvement in quality when trying to understand the influence of processing parameters on air textured yarns. In most literature no evidence has been provided to support the claims of improvement in quality and at best indirect indicators have been used. 8.2) Review of literature Over the years several attempts have been made to quantify air-textured yarn structures. Piller et al [12] have presented packing density values for different textile structures in an effort to define the voluminosity or the bulk of air textured yarns. They inferred that lower the packing density, higher the bulk. Table 7: Packing density of textile structures (from Piller et al [12]) Textile structure Packing density Monofilament 1 Highly twisted continuous filaments Combed cotton yarn Air textured filament yarn

91 They also derived geometric equations describing the structure of an air-textured yarn based on the loop height, loop length, loop radius and other features. Though there was no visual or other evidence provided, they have described the air textured yarn structure as below: Two essentially different structures, one compact and one loose, are continuously connected in the yarn itself by a transition structure, whereby the latter with its percentage contributes to the properties of both structures mentioned. It can be converted by subsequent processes into a more compact form (for example by compression of the yarn during twisting process, in the intersecting points of a knit fabric or a woven fabric) They have further claimed, that an air textured yarn structure is more uniform than a staple yarn with the same linear density. Seidel [10] defined bulk as increase in denier. Kothari [18] measured the bulk of air-textured yarn using the modified Dupont method. In this method a package is wound under a fixed tension for a defined time period and the physical bulk is calculated using the formula: Physical bulk (%) = [Density of parent yarn package in g/cm 3 ] x 100 [Density of textured yarn package in g/cm 3 ] The following methods have also been used to evaluate air-textured yarns: Wilson [29] wound the textured yarn on a black card and observed & marked on basis of uniformity of loop size, number of open places where the filaments were insufficiently interlaced, number of thin places and number of slubs. 50mm long specimens were mounted on a microscope slide in a tensionless state and the yarn and core diameters were measured at sections of 0.6mm wide. For loop frequency, specimens were flattened between two microscopic slides to flatten the loops into one plane. After defining a section of 2.5mm, the number of loops was counted to determine the loop frequency Dupont method: The overall diameter (D) and core diameter (d) was determined optically. Loop size is defined as (D-d)/2 77

92 Piller and Lesykova [12] described the geometrical structure using an idealized air textured yarn structure Wray at Loughborough University improved on the Dupont technique by using a microprojector and mounted the yarn by clamp and weight, and pulled it through to be imaged by using a movable stage and toggle device. He also devised a graphical recording device for assessment of core diameter and overall diameter. The overall diameter was recorded by making a mark on paper tape on the outermost limits of the loops on either side of the yarn from its magnified image projected on to a screen. Similar method was used for core diameter Bock used an opto-electronic instrument to asses the loop size and frequency, wherein the shadow of the textured yarn was projected onto a line of diodes and the corresponding light impingement on the individual diodes was evaluated electronically Kollu used a microdensitometer technique Acar et al [29] used shadowgraphy technique to analyze the yarn. Details can be found in the same publication Mukhopadyay [30] used image analysis to evaluate the physical bulk of yarn by using projected image of yarn to obtain core and total projected area of air textured yarn. The specific volume of the textured yarn was derived from the projected area of air-textured yarn and its linear density. They used standard deviation of the textured yarn as a measure of physical irregularity of the yarn. Standard deviation has also been used as a factor for determining textured yarn quality in the current research Electronic Inspection Board (EIB) at ITT Charlottesville [31]: In this instrument, the yarn is scanned between the light source and the camera at the rate of 2 scans/mm when operating at 100 m/min. The information from the camera is digitized and transmitted to the computer. The yarn profile test mode produces an image of the yarn and also a graph showing the variation in yarn diameter. The variation in yarn diameter due to the present of loops in the air-textured yarn is detected. They developed a bulk-index by incorporating loop frequency, size and total number of loops present in the air-textured yarn 78

93 8.2.1) Instability Instability refers to the behavior of the textured yarn under applied load. In other words instability refers to the retention of texture (loops) under the application of load such as those encountered in processing the yarns into fabric. Most instability tests involve applying a known tension to the textured yarn, measuring the extension of the yarn, and expressing the instability in terms of percent extension. Demir et al [32] have presented a detailed review of all available instability tests. They concluded that the instability tests results alone provide misleading information on the quality of air-textured yarns. This leads back to the basic issue of the meaning of quality. None of the researchers in the literature has related instability to the variation in overall yarn or core diameter on the application of a constantly applied load over time. Most researchers have attributed the instability of the yarn to the surface loops. As mentioned earlier, the same filaments that form the loop also form the core at a different point along the length of the yarn. This indicates that the instability may be more a function of the core and the frictional behavior of filaments within the core, rather than of the loops, which may be in turn be related to the non-uniform removal of spin finish along the fiber length. Though not quite similar to instability, another feature of the air-textured yarn that may be crucial for its end-use may be the squashing of the loops when in the final fabric structure. There is no mention of this aspect or methods of evaluating it in the literature. If most of the loops do get squashed once inside the fabric structure, then it may be suggested that the property of air textured yarn structure of interest may be the increase in core diameter rather than the surface loops. Sankhe [37] has stated that parallel and core-effect structures have greater stability than single-end textured structures, the reason being the large number of filaments. However, no evidence has been provided. Artunc [38] stated that stability of an air textured yarn shows up in the visual appearance of the yarn. According to him, a stable single yarn is more compact than an unstable yarn. 79

94 8.2.2) Properties of air textured yarns and influence of process variables on them It is widely known that the physical, mechanical and aesthetic characteristics of an airtextured yarn are governed by the type, frequency and size of loops projecting from the surface of the yarn. Fischer [11] and others have provided evidence to claim that there is no change in the physical or mechanical properties of individual filaments comprising the air textured yarn structure and so properties such as dyeing characteristics such as levelness and uptake are almost unchanged. The main properties of the supply yarn that are affected by texturing are: Linear density: The linear density of an air textured yarn is always higher than that of the supply yarn Elongation: The elongation of an air textured yarn is always lower than the supply yarn Tenacity: The tenacity of an air-textured yarn is always lower than the supply yarn. Demir et al [34] have used this lowering of tenacity as an indicator of the quality of the texturing and have stated that the higher the reduction in tenacity, the better the textured yarn structure Yarn diameter: The overall yarn diameter of the textured yarn is always higher than the supply yarn. This may be considered as a statistical feature of the yarn and the comparison with supply yarn is usually based on an idealized supply yarn diameter Appearance: Air textured yarns are always flatter in appearance compared to the supply yarns Feel: An air textured yarn always feels softer compared to a flat yarn As mentioned earlier, the properties of individual filaments remain virtually unaltered. The main processing variables in air texturing are overfeed, stretch, operating condition (i.e., dry or wet, though wet texturing is the established standard in the industry probably due to the improved stability that it affords), production speed, and air pressure. The variation of these parameters is believed to influence the final air textured yarn structure to varying degrees. This research shows that the changes in the structure on changing the process parameters may not be as large as is commonly thought. 80

95 8.2.3) Review of literature on the influence on textured yarn properties of change in process variables Several researchers have considered the effects of processing parameters on air-textured yarns. Kothari et al [36] have presented an overview of the classification of air textured yarns and their properties ) Effect of increase in air pressure Acar et al [23,24,32] used visual inspection (SEM photomicrographs), on-line tension measurements and tensile testing to understand the influence of processing parameters on air textured yarn properties. They concluded that with increasing air pressure, the flow velocity and hence the ability of the nozzle to open and entangle the structure increases because of increase in the turbulence of the airflow, which in turn results in better texturing. They have not defined what they mean by better texturing, nor have they provided any evidence to quantify the increase in turbulence of the airflow. With regards to the processing properties of air-textured yarns, they have claimed that the tenacity and elongation decrease with increase in air pressure. Sengupta et al [19] have claimed that increase in the air pressure increases the bulk, but also reduces instability, which they calculated using Dupont instability method. They claimed that at higher air pressures, there is more migration and entanglements of filaments, which results in more contact points and hence higher frictional forces to resist deformation, decreasing instability. With respect to the air-textured yarn properties they claimed that the stiffness of the yarn increases with increasing air pressure. Mukhopadyay [30], used image analysis and values of standard deviation to claim that as the air pressure increases, the core of the textured yarn becomes more regular, initially with formation of loops of variable size, and higher air pressure leads to formation of greater number of loops with less variability ) Effect of increase in overfeed Acar et al [23,24] have claimed that with increasing overfeed; there is an increase in the loop size, loop frequency and linear density. However they have provided definitions for loop size and frequency. With regards to air textured yarn properties they have claimed that yarn 81

96 instability increases with increasing overfeed. While tenacity and elongation both decrease with increase in overfeed. Demir et al [34] using tensile testing results have claimed that As the overfeed increases up to 20% there is an enhancement of texturing quality as indicated by the lower tenacity values, but above 25% there is no such improvement in quality. Sengupta et al [19] have claimed that with an increase in overfeed, large number of loops and arcs are formed, thereby decreasing the cohesion between the fibers, and giving rise to larger instability. With regards to the processing properties of air textured yarns they have claimed that the stiffness decreases with increasing overfeed. Mukhopadyay [30] has stated that at higher levels of overfeed, the physical bulk of textured yarn and yarn structural irregularity are higher ) Effect of increase in processing speed Acar et al [23,24] have claimed that as the processing speed is increase the number of stable loops will be reduced. They have not provided any criterion for the stability of the loops, i.e., they have not defined stable loops. Mukhopadyay [30] has claimed that with increase in texturing speed, the textured yarn physical bulk increases up to a certain extent, but the yarn structural irregularity increases steadily at higher texturing speeds ) Effect of change in processing conditions Dry v/s Wet Demir et al [34] have used instability alone as a measure of quality (They later stated the unsuitability of using instability as a measure of quality [32]) and claimed that the instability increases with both wet processing and use of spherical impact elements {they have later stated in another publication [24] that impact elements have little or no effect on the processing in air texturing}, but that these are within acceptable limits, and as such are indicative of improved yarn quality ) Review of literature on supply yarn parameters The main supply yarn parameters to consider in ascertaining their suitability for air texturing are denier per filament (dpf), number of filaments, cross sectional shape, and their mechanical/physical properties. 82

97 Several researchers have attempted to understand the influence of supply yarn parameters on texturing and texture yarn properties ) Polymer type As noted previously, most of the work in air texturing has only been done on Polyester, Polypropylene and Nylon yarns and very little published work exits exploring the air texturing of a wide range of polymers. Demir et al [34] compared polyester and polyamide yarns and found that polyamide exhibited lower instability, slightly lower linear density increase, and lower percentage tenacity decrease and thus concluded that polyamide textured poorly compared to polyester ) Number of filaments Acar et al [33] have claimed that the optimum number of filaments that can be textured depends on the nozzle that textures them. Demir et al [34] have claimed that with an increase in the number of filaments there is an improvement in texturing quality up to a certain maximum, but that further increase in filament number results in lowering the yarn quality due to two reasons: o More disturbed airflow due to increased number of filaments (same as increase in overfeed) o Increased number of filaments means that there is more number of loops and consequently an increase in the chance of the loops being pulled out In their conclusion, they stated, yarns suitable for texturing should have a filament dpf less than 2dtex (1.8 denier) ) Filament fineness or denier per filament (dpf) According to Mukhopadyay et al [30] yarns having finer dpf filaments, possess greater physical bulk upon being textured along with higher structural irregularity than coarser dpf filament yarn. Acar et al [33] used momentum flux equations and showed that a supply yarn composed of finer filaments should texture more satisfactorily than coarser filament yarns. Demir et al [17] have claimed that for an equal total yarn linear density, yarns consisting of 83

98 finer filaments require smaller fluid forces to displace and entangle than those consisting of coarser filaments, due to their reduced bending and twisting rigidities, thus facilitating their better texturing. Kothari et al [19] conducted experiments on 125/100 & 126/34 filament yarns and have claimed that the yarn composed of finer individual filaments produced inferior results to yarn produced from the coarser individual filament components. They state that the 125/100 yarn produced a large number of smaller size loops at greater frequency and a compact core ) Filament cross sectional shape Using established equations of bending and torsional rigidity, but no experimental evidence, Acar et al [33] have shown that from viewpoint of drag forces, elliptic or hollow circular filaments are better suited for air texturing than filaments having a circular cross section. It may be hypothesized that the most significant effect of the change in cross sectional shape may be the amount of spin finish removal and the feel and look of the final air textured yarn. Based on second moment of area formula they have claimed that larger forces and torques are required to deflect the hollow fibers than those required for solid circular fibers, which may indicate their low suitability for texturing. This claim contradicts those made while considering different cross sectional shapes. They have further stated that filaments with smaller diameter have lower bending and torsional stiffness and facilitate loop the loop formation ) Spin finish According to Dani et al [39] & Acar et al [45] 90% of the spin finish gets removed during wet texturing. According to Acar et al [33] this removal of spin finish during air texturing depends only on the type of spin finish applied on the yarn. 8.3) Results & Discussion: Influence of supply yarns on yarn diameter, core diameter, number of loops & types of loops In the current research the different yarns, and yarns textured under difference processing conditions were compared using the following parameters: 84

99 Yarn diameter: The overall diameter of the yarn i.e., surface loops + core was designated as the yarn diameter. An average value of the yarn diameter as described in the algorithm was taken as the final yarn diameter value while comparing different yarns. Yarn Diameter Core Diameter Figure 47: Yarn diameter and core diameter as defined in the current research Core diameter: The core diameter was defined as in figure 47 and calculated using the algorithm as described in the experimental section of this dissertation. Average value of core diameter (termed the best-core diameter) was used while comparing different yarns Mean & Standard deviation: Statistical descriptors of mean and standard deviation were used to describe the yarn structure as well as to compare yarns under different processing conditions Types of loops: As mentioned in previous chapters the loops have been classified based on two systems. Firstly they have been divided into individual group and bunched groups. Furthermore, the individual or bunched loops may be either twisted, buckled or horseshoe loops. It has been observed that there are very few if any twisted loops in the structures observed Five mm length of yarn was imaged and then the yarn was pulled through and another five mm was imaged and so on as shown in figure 48. Consecutive sections of the yarn were avoided so that a larger length of yarn could be analyzed. Fifteen images were analyzed for 85

100 each processing parameter. Seventy-five measurements of yarn diameter were taken for each image. 5mm Position Observations Average Standard deviation A B C D E F G H I J K L M N O Figure 48: Image analysis algorithm The values for yarn diameter, core diameter and number of loops are presented in their full for Dyneema to show the data collection method. For the rest of the yarns only fifteen values and that too only for one position are presented in the tables. However, the standard deviation and mean values quoted in the dissertation are based on the calculations on all positions and all data points. The discussion below attempts to point towards probable trends in the behavior of texturing of different filament yarns and yarns textured under different processing conditions. The above parameters are presented individually for each yarn and an attempt is made at the end of the discussion to identify a trend. 86

101 8.3.1) Dyneema Comparison between yarns produced in dry and wet texturing conditions: o Yarn diameter: Yarn diameter is 45% smaller in wet condition than in dry condition. The standard deviation at the same time is 45% lower in wet condition than in dry condition, indicating that the yarn has less variability in the diameter along its length in wet condition o Core diameter: Core diameter is 5% smaller in wet condition that in dry condition, as expected. The standard deviation is 40% lower in wet condition than in dry condition, indicating a more uniform core in wet condition o Individual: 15% more individual loops in wet condition. o Bunched: 33% more bunched loops in wet condition o Buckled: 50% more number of loops in wet condition than in dry condition Dyneema Dry Dyneema Wet Figure 49: Dyneema comparison of yarn textured in dry and wet conditions 87

102 Table 8: Dyneema Yarn diameter: Dry V Wet. All values in the table are in MM. Position A A B B C C D D E E Observations Dry Wet Dry Wet Dry Wet Dry Wet Dry Wet

103

104 Mean SD F F G G H H I I J J K K Dry Wet Dry Wet Dry Wet Dry Wet Dry Wet Dry Wet

105

106

107 L L M M N N O O Dry Wet Dry Wet Dry Wet Dry Wet

108

109 Dry Wet Average Yarn diameter Standard deviation

110 Table 9: Dyneema Core diameter: Dry v Wet: All values presented in table are in MM Observation Dry core Wet core Mean SD

111 Table 10:Dyneema: Number and types of loops in dry and wet conditions 97

112 8.3.2) Kevlar Comparison between yarns produced in dry and wet texturing conditions: o Yarn diameter: Yarn diameter is 4% smaller in wet condition than in dry condition. The standard deviation at the same time is 15% smaller in wet condition than in dry condition, indicating that the yarn has more variability in the diameter along its length in dry condition o Core diameter: Core diameter is 2% smaller in wet condition, indicating no difference. The standard deviation is 50% lower in wet condition than in dry condition, indicating a larger non-uniformity of core diameter in dry condition o Individual: 25% fewer loops in wet condition o Bunched: Similar number of bunched loops in both conditions o Buckled: 20% more number of loops in wet condition than in dry condition Kevlar Dry Kevlar Wet Figure 50: Kevlar comparison of yarn textured in dry and wet conditions 98

113 Table 11: Kevlar Yarn diameter: Dry v Wet. All values presented in the tables are in MM. Observation Dry Wet Table 12: Kevlar Core diameter: Dry v Wet. All values presented in the tables are in MM. Observation Dry Wet

114 8.3.3) Zylon Comparison between yarns produced in dry and wet texturing conditions: o Yarn diameter: Yarn diameter is 13% larger in wet condition than in dry condition. The standard deviation at the same time is 24% higher in wet condition than in dry condition, indicating that the yarn has more variability in the diameter along its length in wet condition o Core diameter: Core diameter is 10% smaller in wet condition than in dry condition. The standard deviation is 15% lower in wet condition than in dry condition, indicating a smaller non-uniformity of core diameter in wet condition o Individual: 15% fewer loops in wet condition o Bunched: Similar number of bunched loops in both conditions o Buckled: 60% fewer buckled loops in wet condition than in dry condition Zylon Dry Zylon Wet Figure 51: Zylon comparison of yarn textured in dry and wet conditions 100

115 Table 13:Zylon Yarn diameter: Dry V Wet. All values presented in the table are in MM. Observation Dry Wet Table 14: Zylon Core diameter: Dry V Wet. All values presented in the table are in MM. Observation Dry Wet

116 8.3.4) Spectra Comparison between yarns produced in dry and wet texturing conditions: o Yarn diameter: Yarn diameter is 3% smaller in wet condition than in dry condition. The standard deviation at the same time is 6% lower in wet condition than in dry condition, indicating that the yarns have similar uniformity o Core diameter: Core diameter is 23% smaller in wet condition than in dry condition. The standard deviation is 15% lower in wet condition than in dry condition, indicating a smaller non-uniformity of core diameter in wet condition o Individual: Similar number of loops in both conditions o Bunched: Similar number of bunched loops in both conditions o Buckled: None in either condition Spectra Dry Spectra Wet Figure 52: Spectra comparison of yarn textured in dry and wet conditions 102

117 Table 15: Spectra Yarn diameter: Dry V Wet. All values presented in the table are in MM. Observation Dry Wet Table 16: Spectra Core diameter: Dry V Wet. All values presented in the table are in MM. Observation Dry Wet

118 8.3.5) Technora Comparison between yarns produced in dry and wet texturing conditions: o Yarn diameter: Yarn diameter is 20% smaller in wet condition than in dry condition. The standard deviation at the same time is 25% lower in wet condition than in dry condition, indicating that the is more uniform in wet condition o Core diameter: Core diameter is 40% smaller in wet condition than in dry condition. The standard deviation is 55% lower in wet condition than in dry condition, indicating a smaller non-uniformity of core diameter in wet condition o Individual: 30% more loops in wet condition than in dry condition o Bunched: 50% more loops in wet condition o Buckled: No buckled loops formed in either condition Technora Dry Technora Wet Figure 53: Technora comparison of yarn textured in dry and wet conditions 104

119 Table 17: Technora Yarn diameter: Dry v Wet. All values presented in the table are in MM. Observation Dry Wet Table 18: Technora Core diameter: Dry v Wet. All values presented in the table are in MM. Observation Dry Wet

120 8.3.6) Vectran Comparison between yarns produced in dry and wet texturing conditions: o Yarn diameter: Yarn diameter is 5% smaller in wet condition than in dry condition. The standard deviation at the same time is 25% lower in wet condition than in dry condition, indicating that the yarns is more uniform in wet condition o Core diameter: Core diameter is 3% larger in wet condition than in dry condition. The standard deviation is 37% higher in wet condition than in dry condition, indicating a larger non-uniformity of core diameter in wet condition o Individual: 18% fewer loops in wet condition than in dry condition o Bunched: Similar in both conditions o Buckled: 50% more buckled loops in wet condition Vectran Dry Vectran Wet Figure 54: Vectran comparison of yarn textured in dry and wet conditions 106

121 Table 19: Vectran Yarn diameter: Dry v Wet. All values presented in the table are in MM Observation Dry Wet Table 20: Vectran Core diameter: Dry v Wet. All values presented in the table are in MM Observation Dry Wet

122 8.3.7) Dacron Comparison between yarns produced in dry and wet texturing conditions: o Yarn diameter: Yarn diameter is 24% larger in wet condition than in dry condition. The standard deviation at the same time is 33% higher in wet condition than in dry condition, indicating that the yarn has greater variability in the diameter along its length in wet condition o Core diameter: Core diameter is 20% smaller in wet condition that in dry condition, as expected. The standard deviation is 10% higher in wet condition than in dry condition, not indicating a significant difference in the uniformity of the core o Individual: 33% more individual loops in wet condition. Standard deviation in wet condition is 40% lower than in dry condition o Bunched: No bunched loops were found in dry condition. Though a few bunched loops were present, the number of bunched loops in wet condition was not significantly different o Buckled: Same number of buckled loop in either condition Table 21: Dacron Yarn diameter: Dry V Wet. All values presented in the table are in MM. Observation Dry Wet

123 Table 22: Dacron Core diameter: Dry v Wet. All values presented in the table are in MM Observation Dry Wet

124 8.3.8) Comparison between yarns made of same polymer (& same dpf) but having different number of filaments Dyneema yarn having 780 filaments and 195 filaments was used for this analysis: Yarn diameter: 780 filament yarn diameter was ~20% larger than 195 filament yarn diameter. The standard deviation values indicate that the 780 filament yarn had an almost 50% more uniform structure than the 195 filament yarn Core diameter: 780-filament core diameter was 50% larger than 195-filament core diameter. The 780 filament yarn standard deviation is 24% greater than 195 filament indicating a more non-uniform core diameter structure along yarn length Number of loops: o Individual: 15% more loops in 780 filament yarn than in195 filament yarn o Bunched: 30% more loops in 780 filament yarn than in 195 filament yarn o Buckled: 33% more loops in 195 filament than in 780 filament yarn Table 23: Yarn diameter comparison for same polymer yarn having different number of filaments Observation 780 fils 195 fil

125 Table 24: Core diameter comparison for same polymer yarn having different number of filaments Observation 780 fils 195 fil ) Comparison of a single yarn (Dyneema : 440/195) textured using different overfeeds Yarn diameter: As the overfeed is increased to 15%, 25%, 35%, and 45%, the increase in yarn diameter (for every step of 10%) was 25%. The difference in yarn diameter between 15% and 45% was 60%. The standard deviation varied from 20-46% as the overfeed was increased from 15-45%, indicating a increasingly nonuniform structure along the length of the yarn with increasing overfeed. The nonuniformity increased at high at 75% when comparing only yarns textured at 15% and 45% Core diameter: As the overfeed was increased from 15-45%, there was a decrease in core diameter by 16% and an increase in standard deviation by 20% indicating a more non-uniform core with increasing overfeed Number of loops: o Individual: 37% more loops were present at 45% than at 25% o Bunched: 50% more bunched loops at 45% than at 25% o Buckled: There are 32% more buckled loops at 35% than at 25%. However, there are 23% fewer buckled loops at 45% than at 35% 111

126 Figure 55: Dyneema Effect of increase in overfeed on yarn and core diameter Table 25: The effect of increase in overfeed on yarn diameter. All values in the table are in MM Observation 15% OF 25% OF 35% OF 45% OF

127 Table 26: The effect of increase in overfeed on core diameter. All values in the table are in MM Observation 15% OF 25% OF 35% OF 45% OF ) Comparison of a single yarn (Dacron ) textured using different overfeeds Yarn diameter: 24% higher yarn diameter value at 45% overfeed. The standard deviation is 33% lower at 25% than at 45% indicating a more uniform structure along the length of the yarn Core diameter: The core diameter at 45% overfeed is 20% smaller than at 25% overfeed. The standard deviation is 10% lower at 25% overfeed than at 45%, indicating a more uniform core along the yarn length at 25% overfeed Number of loops: o Individual: 60% more loops were present at 45% than at 25%. The standard deviation was almost double for 45% indicating a much less uniform distribution of loops along the yarn length than at 25% o Bunched: Same number of bunched loops at both overfeeds o Buckled: There are 40% more buckled loops at 45% than at 25% 113

128 Table 27: The effect of increase in overfeed on yarn diameter. All values are in MM. Observation 25% OF 45% OF Table 28: The effect of increase in overfeed on core diameter. All values are in MM. Observation 25% OF 45% OF

129 8.3.11) Comparison between yarn made of same polymer (Dyneema ) textured at different stretch Yarn diameter: 8% decrease in overall yarn diameter at 15% stretch than 10%. The standard deviation indicates that there was 40% more non-uniformity along yarn length at 10% stretch Core diameter: 16% higher value of core diameter at 15% stretch, while the standard deviation was almost same Number of loops: o Individual: 20% more loops at 10% stretch. The standard deviation is higher by 24% for 10% stretch indicating a less uniform distribution of loops along yarn length o Bunched: 17% more loops are present at 10% stretch o Buckled; Almost the same number is present at both stretch values Table 29: The effect of increase in stretch on yarn diameter. All values are in MM. Observation 10% ST 15% ST

130 Table 30: The effect of increase in stretch on core diameter. All values are in MM. Observation 10% ST 15% ST ) Comparison of yarn made of same polymer (Kevlar ) textured at different air pressures Yarn diameter: As the air pressure increases from 100 to 160 PSI in increments of 20 PSI, the overall yarn diameter decreases by 17%. However, as the air pressure is increased to 180 PSI, the yarn diameter increased by 24%. If one compares only yarns textured at 100 PSI and 180 PSI, the yarn diameter increases by 10%. The standard deviation is similar for all air pressures, indicating similar uniformity for all air pressures Core diameter: As the air pressure is increased from 100 PSI to 180 PSI, there is a 10% decrease in the core diameter. But when the air pressure is further increased to 180 PSI, there is an increase in core diameter by 16%. If one compares yarns at 100 PSI and 180 PSI, there is a 7% increase in the core diameter. A 30% lower standard deviation value with increase in air pressure from 100 to 160 PSI indicates that the yarn gets more uniform in with an increase in air pressure in this standard deviation. As the air pressure is further increased from 160 PSI to 180 PSI, there is a 30% increase in the standard deviation value indicating a more non-uniform structure. If 116

131 only 100 PSI and 180 PSI are compared, there is only a 4% reduction in the standard deviation value, indicating no change in the uniformity Number of loops: o Individual: 9% decrease with increase in air pressure. o Bunched: 10% fewer loops at 180 PSI than at 100 PSI o Buckled; 5% fewer buckled loops at 180 PSI than at 100 PSI Figure 56: Kevlar Effect of increase in air pressure on yarn and core diameter Table 31: The effect of increase in air pressure on yarn diameter. All values are in MM Observation 100 PSI 120 PSI 140 PSI 160 PSI 180 PSI

132 Table 32: The effect of increase in air pressure on core diameter. All values are in MM Observation 100 PSI 120 PSI 140 PSI 160 PSI 180 PSI ) Comparison of yarn made of same polymer (Dacron ) textured at different air pressures Yarn diameter: 20% higher yarn diameter at 80 PSI than at 180 PSI. The standard deviation is 10% larger at 80 PSI, indicating a comparatively less uniform yarn structure Core diameter: Almost exactly the same core diameter at both air pressure and also the standard deviation was the same. Number of loops: o Individual: 34% more number of loops at 80 PSI. The standard deviation was exactly the same indicating that the loops were distributed uniformly along the yarn length o Bunched: Same number of bunched loop in both condition o Buckled; 30% more loops at 80 PSI than at 180 PSI 118

133 Table 33: The effect of increase in air pressure on yarn diameter. All values are in MM Observation 80 PSI 180 PSI Table 34: The effect of increase in air pressure on core diameter. All values are in MM. Observation 80 PSI 180 PSI

134 8.4) Tensile testing of air textured yarns Tensile testing of air-textured yarns was carried out to further understand the influence of processing parameters on the final structure of air-textured yarns. Polyester tire cord, Dyneema and Kevlar yarns were produced under different processing conditions and tested. The following were the test parameters: Gage length: 1 Test speed: 0.1 inch/min Number of specimens for each sample: 6 (All curves shown below are average curves of six specimens for each sample) Software: TestWorks 8.4.1) Polyester tire cord yarn ) Effect of increase in stretch at 45% overfeed The stretch values were varied between 6-12% in increments of 2%. There is a 30% increase in strain as the stretch is increased from 6% to 8%. There is no significant difference in the tenacity for different stretch values Polyester tire cord yarn: Effect of increase in stretch at a constant overfeed of 45% Tenacity (gpd) % St ret ch 8% st ret ch 10% st r et c h 12% st r et c h Strain (%) Figure 57: Polyester tire cord yarn: Effect of increase in stretch at constant overfeed 120

135 ) Effects of increase in overfeed at 15% stretch As the overfeed is increased from 15% to 55%, there is a 38% reduction in tenacity and a 46% increase in the strain Polyester tire cord yarn: Effect of increase in overfeed at a constant stretch of 15% Tenacity (gpd) % Overfeed OF 15% Stretch 15% OF 55% Stretch 15% Strain (%) Figure 58: Polyester tire cord yarn: Effect of increase in overfeed at constant stretch ) Effect of increase in air pressure As the air pressure increases from PSI, there is a 11% increase in the tenacity of the yarn, while the strain increases by 34% 121

136 Polyester tire cord yarn: Effect of increase in air pressure Tenacity (gpd) PSI 160 PSI 120 PSI 140 PSI 160 PSI PSI Strain (%) Figure 59: Polyester tire cord yarn: Effect of increase in air pressure 8.4.2) Dyneema ) Effects of increase in overfeed at 10% stretch There is a 60% reduction in tenacity as the overfeed is increased from 15%-45%, while there is a 35% increase in strain 122

137 Dyneema: Effect of increase in overfeed at a constant stretch of 10% % Overf eed Tenacity (gpd) % Overf eed 25% Overf eed 15% Overf eed 25% Overf eed 35% Overf eed 45% Overf eed 5 45% Overf eed Strain (%) Figure 60: Dyneema: Effect of increase in overfeed at constant stretch ) Effects of increase in stretch at 15% overfeed As the stretch increases from 6-10% in increments of 2%, the tenacity increases by 20%, while the strain increases by 28% Dyneema: Effect of increase in stretch at a constant overfeed of 15% % St r et c h 15 Tenacity (gpd) % St ret ch 6% st ret ch 8% St ret ch 10% St r et c h 0 8% St ret ch Strain (%) Figure 61: Dyneema: Effect of increase in stretch at constant overfeed 123

138 ) Effects of increase in overfeed at 15% stretch As the overfeed increases from 15%-45% in increments of 10%, there is a 48% reduction in tenacity, while there is a 36% increase in the strain value also increases by 27% Dyneema: Effect of increase in overfeed at a constant stretch of 15% % Overf eed 15 25% Overf eed Tenacity (gpd) % Overf eed 15% Overf eed 25% Overf eed 35% overf eed 45% Overf eed 35% Overf eed Strain (%) Figure 62: Dyneema: Effect of increase in overfeed at constant stretch ) Effect of increase in air pressure There is no significant difference in the tenacity or strain across the pressure standard deviation 124

139 Dyneema: Effect of increase in air pressure PSI Tenacity (gpd) PSI 100 PSI 120 PSI 140 PSI 160 PSI 180 PSI PSI PSI 100 PSI Strain (%) Figure 63: Dyneema: Effect of increase in air pressure 8.4.3) Kevlar ) Effects of increase in overfeed at 15% stretch There is a 58% reduction in strength with increase in overfeed from 15%-45%. There is a 56% increase in strain at the same time. 125

140 Kevlar: Effect of increase in overfeed at a constant stretch of 15% % Overf eed 7 Tenacity (gpd) % Overf eed 35% Overf eed 15% Overf eed 15% St ret ch 25% Overf eed 35% Overf eed 45% Overf eed 3 2 5% Overf eed Strain (%) Figure 64: Kevlar: Effect of increase in overfeed at a constant stretch ) Effects of increase in stretch at 45% overfeed There is a 12% reduction in tenacity with increase in stretch, while there is a decrease of 40% in the strain Kevlar: Effect of increase in stretch at constant overfeed of 45% Tenacity (gpd) % St ret ch 15% St r et c h 15% st r et c h 20% st ret ch 25% st ret ch % St ret ch Strain (%) Figure 65: Kevlar: Effect of increase in overfeed at a constant stretch 126

141 ) Effect of increase in air pressure There is no significant change in the tenacity or strain with increase in air pressure from PSI Kevlar: Effect of increase in air pressure PSI Tenacity (gpd) PSI 140 PSI 160 PSI 100 PSI 120 PSI 140 PSI 160 PSI 180 PSI PSI Strain (%) Figure 66: Kevlar: Effect of increase in air pressure 127

142 8.5) Tension measurements On-line tension was measured during the processing of air-textured yarns, using a three-pin tensiometer in three zones as shown in figure 66. The tension was measured just before the yarn entered the wetting bath (Before-delivery zone), just outside the nozzle exit (Delivery zone) & between the stretch rollers (Stretch zone). Figure 67: Tension measurement zones 128

143 Table 35: Tension values in the three zones for three different yarns with increasing overfeed. Polyester tire cord yarn Increasing overfeed % Before-delivery zone (gm) Delivery zone (gm) Stretch zone (gm) Dyneema Increasing overfeed % Before-delivery zone (gm) Delivery zone (gm) Stretch zone (gm) Process break down Kevlar Increasing overfeed % Before-delivery zone (gm) Delivery zone (gm) Stretch zone (gm) Process break down It can be seen that with an increase in overfeed, there is a reduction in the delivery zone tension, which is an indication that more filaments are wildly moving around inside the nozzle and are more likely to be in one of the six circulation zones, as opposed to the filaments in the central zone where there is no circulation. 129

144 Furthermore, the tensioning function of the nozzle can be confirmed by noting the tension values in the before delivery zone region, which are not altered with increasing overfeed. Table 36: Tension values in the three zones for three different yarns with increasing air pressures. Polyester tire cord yarn Increasing Air Before-delivery Delivery zone Stretch zone pressure PSI zone (gm) (gm) (gm) Dyneema Increasing Air Before-delivery Delivery zone Stretch zone pressure PSI zone (gm) (gm) (gm) Kevlar Increasing Air Before-delivery Delivery zone Stretch zone pressure PSI zone (gm) (gm) (gm)

145 As is seen from table 37, the tension values do not change significantly in the delivery zone for all three yarns with increasing air pressure. These correlate very well with the image analysis and tensile testing results, which show very little change in the air textured yarn structure with increase in air pressure. The increasing tension values in the before-delivery zone with increasing air pressure confirm the fact that the nozzle imparts tension to the yarn right up to the plane of air inlet, and that this tension increases with increasing air pressure. Table 37: Tension values in the three zones for three different yarns with increasing stretch. Polyester tire cord yarn Before-delivery Delivery zone Stretch zone Increasing Stretch % zone (gm) (gm) (gm) Dyneema Before-delivery Delivery zone Stretch zone Increasing Stretch % zone (gm) (gm) (gm) Kevlar Before-delivery Delivery zone Stretch zone Increasing Stretch % zone (gm) (gm) (gm)

146 As expected with increasing stretch, the tension in the stretch zone increases, but in the other two zones remains unaltered. These results correlate very well with the image analysis and tensile testing data. 8.6) Analysis of high-speed image captures 8.6.1) Introduction The high-speed images were analyzed using the same algorithm described in the section above. However, due to the very large number of images that had to be manually analyzed, only the core diameter was measured. Since the scale factor related to the magnification at which these images were captured was unknown, the data points of the analysis were not converted to any known unit, such as mm or cm, but the Photoshop unit of points was retained. The image analysis was carried out only for the yarn at the nozzle exit. The images from the other zones of texturing process shown below have been presented for the sake of completion. Along with the image analysis, a visual interpretation for each zone is presented. The two interpretations correlate well ) Effect of increase in speed: Kevlar yarn was imaged at 60 and 120 mpm Visual Interpretation: The process stability seems to improve as the speed is increased from 60 mpm to 120 mpm. 60 mpm 120 mpm Figure 68: Change in speed 132

147 Kevlar : When the speed is doubled from 60 mpm to 120 mpm, there is a 50% reduction in core diameter, while there is 74% more uniform core at 120 mpm. There is no significant change in core diameter or uniformity with further increase in the speed Kevlar: Effect of increase in speed Core diameter (points) SD mpm 60 mpm SD Distance along yarn length (points) Figure 69: Effect of increase in speed 8.6.3) Effect of increase in air pressure at nozzle exit: Polyester tire cord yarn was imaged at 80 and 120 PSI. 80 PSI 120 PSI Figure 70: Change in air pressure 133

148 Visual Interpretation: At 80 PSI, the yarn is much more open as it comes out of the nozzle, while at higher air pressure, a distinctive core is seen. This confirms to the findings from the analysis of the final yarn structure Kevlar : With an increase in air pressure the core diameter decreases by 36%, while at the same time it is 66% more uniform based on standard deviation values Kevlar: Effect of increase in air pressure Core diameter (points) SD PSI 120 PSI SD Distance along yarn length (points) Figure 71: Change in air pressure 8.6.4) Effect of increase in air pressure at water bath exit: Kevlar yarn was imaged at 100, 140 and 180 PSI. 100 PSI 140 PSI 180 PSI Figure 72: Change in air pressure at water exit 134

149 Visual interpretation: As the air pressure is increased from 100 PSI to 180 PSI, the magnitude of secondary flow increases considerably, which helps blow of the water closer and closer to the water bath exit, as opposed to the nozzle exit. This may not have a significant impact on the final yarn properties, because as mentioned, only a very small quantity of water plays any role in the air texturing process ) Effect of increase in air pressure at nozzle entry: Kevlar yarn was imaged at 100, 140 and 180 PSI. 100 PSI 140 PSI 180 PSI Figure 73: Change in air pressure at nozzle entry Visual Interpretation: With increase in air pressure, less amount of water is dragged on the yarn surface up to the nozzle entry, because most of it is blown away due to the very high magnitude of air pressure ) Effect of increase in stretch: Polyester tire cord yarn was textured at 4% and 12% stretch. 4% Stretch 12% Stretch Figure 74: Change in stretch Visual interpretation: There seems to be a more stable structure with a better defined core at higher stretch 135

150 Polyester tire cord yarn: There is a 62% reduction in core diameter with increase in stretch from 4% to 12%, while the core is 83% more uniform at 12% stretch, based on standard deviation values Polyester tire cord yarn: Effect of increase in stretch at constant overfeed Core diameter (points) SD % Stretch 12% Stretc h SD Distance along yarn length (points) Figure 75: Effect of increase in stretch at constant overfeed 8.6.7) Effect of increase in overfeed: Kevlar yarn was textured at 10 and 15% overfeeds. 10% overfeed 15% overfeed Figure 76: Effect of increase in overfeed at constant stretch 136

151 Visual interpretation: The Kevlar yarn textured at different overfeeds, was not a very good selection for imaging due to its very large number of filament. No useful information could be obtained from the above high speed movies. Kevlar : 6% reduction in core diameter at higher overfeed, while its uniformity is 20% higher at higher overfeeds. As noted above, the analysis of Kevlar yarn high-speed images may not very accurate. Polyester tire cord yarn: 4% decrease in core diameter at higher overfeed, while the core diameter is 25% more uniform at the same time, based on the standard deviation values Kevlar: Effect of increase in overfeed Core diameter (points) SD % Overfeed 15% Overfeed SD Distance along yarn length (points) Figure 77: Kevlar: Effect of increase in overfeed at constant stretch 137

152 Polyester tire cord yarn: Effect of increase in overfeed Core diameter (points) SD CE 6% S4% CE 15% S4% SD Distance along yarn length (points) Figure 78: Polyester tire cord: Effect of increase in overfeed at constant stretch 8.6.8) Effect of increase in effect overfeed at constant core overfeed Core and effect yarn was produced using Dyneema as the core and the photoluminscent PA/PP 124 stripes yarn as the effect. The core overfeed was kept constant at 10%, while the effect overfeed was 20, 30 and 50% for the three trials. 20% effect overfeed 30% effect overfeed 50% effect overfeed Figure 79: Change in effect overfeed Visual interpretation: There seems to be no difference in terms of processing performance in the three scenarios. 138

153 Polyester tire cord yarn (core) and Dyneema (effect): As the effect overfeed is increased from 20 to 50%, there is no significant change in the core diameter, however it get 20% more uniform at the same time Increase in effect overfeed Core diameter (points) SD: C10E20 C10E30 C10E Distance along yarn length (points) Figure 80: Increase in effect overfeed 8.6.9) Texturing in dry and wet conditions Polyester red tire cord yarn and Dyneema yarn were parallel textured in dry and wet conditions. Dry Wet Figure 81: Dry v Wet condition at nozzle entry 139

154 Since the yarn is under tension in this zone, there is no significance influence of water at this stage ) Same yarn textured using different nozzles Polyester tire cord yarn was textured using T100 and T370. T100W T370 Figure 82: Texturing using different nozzles Visual Interpretation: T370 has got three air holes, while T-100 has got one air inlet hole. It can be expected that the T370 would produce a more defined core due to its heavier intermixing action in the air-inlet plane. 8.7) Discussion based on the above results 8.7.1) Introduction The last few sections have presented data from image analysis, tensile testing and on-line tension measurements, along with a visual interpretation of the high-speed imaging. The discussion below aims to combine the understanding obtained from all three data sets with the theories proposed in this work and presents the main findings )Texturing condition: Dry v Wet When comparing the yarns textured in dry and wet conditions, there is a decrease in the yarn diameter for all yarns, in some cases significant (Dyneema /Technora ), in some not 140

155 as significant (Kevlar, Spectra, & Vectran ). The uniformity of the yarn as indicated by the standard deviation is higher for all yarns in wet condition. In other words, wet texturing improves the uniformity of the air textured yarn structure. This can be explained by considering figure 12. Due to the cohesive effect of water, there is less obstruction to the migration of free filaments around the fiber axis during loop formation. This results in a more uniform distribution of filament length around the yarn axis and along its length. The only exception to the rule being Zylon, which has a slightly larger yarn diameter in wet condition and more non-uniformity in wet condition. Zylon is composed of very fine dpf (~1 dpf) filaments as well as very few filaments. As can be evidenced from figure 31 in the presence of water the entire structure remains together and the individual filaments do not separate, causing the whole yarn to bend and twist. This may point towards the fact that yarns composed of very few and very fine individual filaments texture better in dry condition than in wet condition. Furthermore, the core diameter is smaller for all yarns in wet texturing, confirming the cohesive effect of water. The lower obstruction to the migration of filaments and the bunching effect of water (i.e., when fine fibers combine to achieve critical stiffness) is also confirmed by the fact that there are substantially more number of buckled loops in wet texturing than in dry (Table 40). As has been noted previously, the main criterions for formation of buckled loops are critical stiffness and lack of obstruction to the migration of filaments. Table 38: Comparison of yarn and core diameters and uniformity in wet and dry conditions Yarn type Spectra (650/120) 5.4 Denier per filament (dpf) Yarn diameter 3% smaller in wet Yarn diameter SD 6% lower in wet Core diameter 23% smaller in wet Core diameter Standard deviation 15% lower in wet Vectran (1500/300) 5 Zylon (166/150) 1.11 Dyneema (440/195) 2.25 Kevlar (850/835) % smaller in wet 13% larger in wet 45% smaller in wet 4% smaller in wet 25% lower in wet 24% higher in wet 45% lower in wet 15% lower in wet 3% larger in wet 10% smaller in wet 5% smaller in wet 2% smaller in wet 37% larger in wet 15% lower in wet 40% lower in wet 50% lower in wet Technora (1520/1000) % smaller in wet 25% lower in wet 40% smaller in wet 55% lower in wet 141

156 Table 39: Comparison of number of loops in wet and dry conditions Yarn Denier per filament Spectra (650/120) 5.4 Vectran (1500/300) 5 Zylon (166/150) 1.11 Dyneema (440/195) 2.25 Kevlar (850/835) 1.02 Technora (1520/1000) 1.52 Individual loops Bunched loops Same number Same number 16% fewer in wet 15% fewer in wet 15% more in wet 25% fewer in wet 30% more in wet Same number Same number 33% more in wet Same number 50% more in wet Buckled loops None in either condition 50% more in wet 60% fewer in wet 50% more in wet 20% more in wet None in either condition 8.7.3) Yarns of same polymer but having different number of filaments When comparing yarns of same polymer having different number of filaments, the yarn with larger number of filaments has larger yarn diameter, as expected. It is also more uniform along the yarn length compared to the yarn with less number of filaments, both being textured under identical texturing conditions. This is because the yarn with more filaments is relatively more stable in the airflow as it has a higher probability of having some of its filaments in the non-circulation zone and hence participating in the formation of core, then the yarn with fewer filaments. The yarn with more filaments also has more number of individual and bunched loops. However, the yarn with fewer filaments has more number of buckled loops. This is due to the lower obstruction presented to the bending of the filaments ) Same yarn textured at increasing overfeeds With an increase in overfeed, there is a significant increase in the yarn diameter, however, the uniformity of the yarn is considerably lower. The core diameter at the same time decreases. As the overfeed increases, the tension on the filaments is reduced (tables 36, 142

157 37, 38), which results in more opening of the filaments and fewer of them being present in the core structure. Though the number of loops increases with increase in overfeed, the non-uniformity of the structure also increases. This is because at higher overfeeds, more length of the filaments is present in the airflow and hence the probability of filaments obstructing each other s migration is increased. This correlates well with the tensile testing data wherein with increase in overfeed, the tenacity decreases and the strain increases 8.7.5) Same yarn textured at increasing stretch With an increase in stretch, the yarn diameter reduces, but the yarn becomes more uniform. At the same time, the core diameter increases. The number of loops decreases with increasing stretch as can be expected. An increase in stretch serves to increase the tension in the yarn being presented to the incoming airflow, as shown in tables 36, 37, & ) Same yarn textured at increasing air pressures With an increase in air pressure, the yarn diameter reduces by a very small amount, and the uniformity of the structure remains essentially the same. The core diameter also reduces by a very small amount, but it become more uniform across the length of the yarn. This finding correlates very well with the tensile testing results, which also indicate no significant change in the air textured yarn structure with increasing air pressures. This is further corroborated by the on-line tension measurements, which show that there is no change in the delivery zone tension with increasing air pressure i.e., the air textured yarn structure at the exit of the nozzle is essentially the same at different air pressures. 143

158 Section IX Conclusions and recommendations 144

159 9.1) Conclusions An in depth analysis of the air jet texturing process has been presented as related to processing of high modulus yarns. Even though all the work was done using high modulus yarns, the findings were confirmed by texturing apparel grade yarns. High-speed imaging, image analysis and concepts of fiber science were utilized for the research. The major findings are presented below: Nozzles & Mechanism of loop formation: The functions of the nozzle have been defined more precisely than in the available literature. This combined with the simulations of the airflow inside the nozzle may point towards different nozzle designs for producing yarns with unique surface properties o Nozzle functions: To push the yarn out, to redistribute the filaments around the central axis to create the 3D structure of an air textured yarn, to remove spin finish and to open individual and groups of filaments o Simulation: The simulation helps understand the airflow inside the nozzle and also elucidates the precise influence of increase in air pressure and other factors along with providing a more accurate analysis of the mechanism of loop formation. The presence of a non-circulation zone, also helps explain the formation of a core of an air-textured yarn structure Loop classification: Though classification schemes have been presented before, this research for the first time presents an attempt to relate the loop classification to the process and mechanism of loop formation. Also, the concept of multi-fiber loops is introduced for the first time in air texturing literature which not only helps explain one of the roles of water in texturing, but also has significance in terms of yarn instability and end use properties, especially in high-modulus yarn applications such as composites and ballistics Role of water: This research shows that the most probable role of water may be to impart cohesion to the filament bundle and therefore help form a more stable core and hence improve the stability of the process. A secondary role of water may also be to combine fibers to form a bunch of fibers that may attain a critical stiffness and hence be able to form loops rather than locally crinkle or be wavy. Another role of water that was not investigated in this research but has been investigated earlier [39] is to remove the spin finish from the surface of the yarn. Though no experimental 145

160 investigation has been carried out this research proposes that the spin finish may be removed non-uniformly along the surface of the yarn as well as between different filaments (between those at the center and those on the periphery), thereby affecting the fiber-to-fiber frictional characteristics, which may show up in the stress-strain curves as stick-slip effect. The influence of the non-uniform removal of spin finish only shows up post processing stage, rather than during the texturing process. During the process, the influence of water is mainly be to impart stability to the process by forming a more cohesive structure Influence of increase in air pressure: One of the most significant findings of this research has great practical value. It has been found that for the yarn used in this research and the air pressure range tested, the structure and properties of airtextured yarns do not change to any significant degree with an increase in air pressure. This is corroborated by image analysis, on-line tension measurements and tensile testing results, along with visual interpretation of high-speed images Governing factor Overfeed: The chief governing factor in controlling the final air textured yarn structure, as well as the process performance is overfeed. The findings of this research show that no other factor, including, processing condition (dry v wet), stretch, air pressure, supply yarn polymer or process speed have the same impact as does overfeed Quality parameter of air textured yarns: This research suggests the use of standard deviation of air-textured yarns as a measure of quality rather than the subjective better or improved quality. Further the classification of loops presented in this research also helps define the quality of the yarn more objectively, since the type and frequency of the loops changes with processing condition High-speed imaging: An exhaustive visual library has been provided for the process and all its important zones under a wide range of processing conditions. A more detailed statistical analysis of these image captures may provide more precise insight into the mechanisms of air texturing 9.2) Recommendations Image analysis: The image algorithm used in this research may be programmed and developed into an automated imaging system 146

161 Simulation: More precise simulations of the airflow may be attempted for several different nozzle geometries and airflow conditions to have a better understanding of the process Flexi-glass nozzle: To have better understanding of the behavior or filaments inside the nozzle, flexi glass nozzles may used along more efficient lighting systems Different cross-section filament yarns: Yarns with different filament cross sections such as elliptical, hollow, triblobal etc may be textured to confirm or contradict the findings Spin finish mapping: The role of the non-uniform removal of spin finish on the properties of air textured yarn at the end use state may be substantial and needs to be investigated in greater detail, especially for high-modulus yarns Core-effect and parallel processing: Though the findings of this research may be sufficient to explain the process in core-effect and parallel processing, to confirm or deny them, more experimental work may need to be done in core-effect and parallel processing Role of water: In order to confirm the role of water as a cohesion agent that brings the bundle of filaments together, any liquid that does the opposite may be used in air texturing and its influence studied 147

162 Bibliography 1. Hearle J.W.S, Hollick L, Wilson D.K (2001) Air jet texturing. Yarn texturing technology Acar M, Turton R.K, & Wray G. R (1986) An analysis of the air jet yarn texturing process Part II: Experimental investigation of the airflow Journal of textile institute. No. 1, Demir A, Acar M, Turton K. R (1988) A basic understanding of the airflow in texturing nozzles. Melliand English. No. 4, E126-E Versteeg H.K, Bilgin S, Acar, M (1994) Effects of geometry on the flow characteristics and texturing performance of air jet texturing nozzles Textile research journal. 64(4), Smelkov D, Kogan A (1996) Description of airstreams in the air jet nozzles for air jet fancy yarns. Chemical Fibers International. V Kothari V.K, Sengupta A K, Sensarma J K (April 1996) Air-jet textured yarns. The Indian textile journal Rwei P, Pai H (2002) Fluid simulation of airflow in texturing jets. Textile Research Journal 72(6) Hoffsomer K. P (1980) Air texturing for the 1980s: A look forward and backward. Fiber producer 9. Kothari V.K, Sengupta A.K (1996) Air jet texturing: A review. The Indian textile journal (March 1996) Siedel L.E (1978) Another look at air textured yarns. Textile industries Fischer K. I (1980) Air jet texturing International Textile Bulletin Spinning , Piller. B, Lesykova E (1981) Structure and economic aspects of air textured yarns 20 th International synthetic fiber symposium in Dornbirn, September Nosov M.P, Tarasenko N. K, Nazarenko T.S, Dmitrieva I.A (1985) Preparation and properties of looped viscose-kapron yarns Khimicheskie Volokna No.1, Acar. M, Wray G.R, (1986) An analysis of the air jet yarn texturing process Part 1: A brief history of developments in the process Journal of textile institute No.1, Simmen C (1987) A.T for fine filaments Textile Horizons Acar M (1989) Use of air jets in yarn texturing processes International Fiber Journal 148

163 17. Demir A, Acar. M & Wray (1987) Spun-like Polypropylene filament yarns produced by air-jet texturing Polypropylene fibers and textiles IV 21/1-21/5 18. Kothari V.K. & Yadav V.K (1999) Air-jet texturing of filament feed yarns of different shrinkage potential Indian Journal of Fiber & Textile Research Sengupta A.K, Chatopadhyay R, & Sensarma J.K (1992) Air-jet texturing of Sirospun yarn Textile research journal Mukhopadhyay A & Kothar V.K. (2001) Selection of air jet texturing process variables The Indian Textile Journal Bcok.G & Lunenschloss. J (1984) The intermingling action during air jet texturing Chemiefasern/Textilindustrie Acar M, Turton R.K, & Wray G. R (1986) An analysis of the air jet yarn texturing process Part III: Filament behavior during texturing Journal of textile institute. No. 4, Acar M, Turton R.K, & Wray G. R (1986) An analysis of the air jet yarn texturing process Part III: The mechanism of loop formation Journal of textile institute. No. 6, Acar M, & Wray G. R (1986) An analysis of the air jet yarn texturing process Part VII: The effects of processing parameters on yarn properties Journal of textile institute. No. 6, SASMIRA Technical Digest (1988) Texturing: Part Acar M, & Wray G. R (1986) An analysis of the air jet yarn texturing process Part VII: The effect of wetting the yarns Journal of textile institute. No. 6, Chand. S (1995) The role of water in air0jet texturing (a critical review) Journal of textile institute Schwarz.E (1993) Air-jet texturing process for the manufacture of a multifold of structures from filament yarns Chemiefasern/Textilindustrie E42-E Acar M, King T.G, & Wray. G.R (1986) Textured yarn quality Textile Asia Mukhopadhyay A, Kaushik R C D & Kothari (2000) Effect of air-jet texturing process variables on physical bulk obtained by image analysis method Indian journal of fiber & textile research Ghosh S, Grindle R & Hill M. (1999) Practical approach to air textured yarn bulk measurement in relation to process parameters Chemical fibers international

164 32. Demir A, Acar M, & Wray G.R (1986) Instability tests for air-jet textured yarns Textile research journal Acar M (1988) Factors governing the choice of supply yarns suitable for air-jet texturing Chemiefasern/Textilindustrie E Demir A, Acar M & Wray (1988) Air jet textured yarns: The effects of process and supply yarn parameters on the properties of textured yarns Textile research journal Denton M.J (1989) Air jet considerations Textile Asia Kothari V.K, Sengupta A.K, Sensarma J.K (1996) Air-jet texturing: A review The Indian textile journal Sankhe M.D (2000) Air texturing- Process, product and applications Manmade textiles in India Artunc H, Bocht B & Weinsdorfer. H (1979) Air-jet texturing process with integrated drawing and shrinking zones Chemifasern/Textilindustrie E118-E Dani N (2000) Air texturing of high modulus yarns Masters thesis at North Carolina State University, College of textiles 40. Lunenschloss, G. B. a. J. (1982). Textile Machinery: Investing for the future", the Textile Institute, Manchester. The textile Institute Acar, Demir. (1989). Yarn Wetting Mechanism and Suitable Supply Yarn for Air Jet Texturing. Proceedings of Air Jet Texturing and Mingling/Interlacing, Second International Conference, Loughborough (27-29 September): Textile science Steve Warner 44. Acar, M., Turton, R.K. and Wray, G.R., ''Air flow in yarn texturing nozzles, Transactions of ASME Journal of Engineering for Industry, 109(3), 3rd August 1987, pp Acar M, Bilgin S, Versteeg H.K, Dani N & Oxenham O (Accepted for publication by Textile Research Journal 2004) The mechanism of the air-jet texturing: the role of wetting, spin finish and friction in forming and fixing loops Sengupta, Kothari, Srinivasan (1990). Man Made Fiber Year Book of Chemiefasern/Textileindustrie:

165 Appendix A: Equations solved in the simulations 151

166 152 The governing equations written in vectorial form are as follows: Conservation of mass: = 0 + V Dt D ρ ρ (A1) Conservation of linear momentum: ( ) ( ) V V P Dt DV + = η η ρ 3 / 4 (A2) Conservation of energy: T k Dt DP Dt DT c p 2 + Φ + = ρ (A3) where, z v y v x v V z y x + + =, z y x + + =, z v y v x v Dt D z y x + + =, and z x y y x z x y z e y v x v e z v x v e z v y v V ˆ ˆ ˆ + =. Φ which is called viscous dissipation term-characteristic of high speed flows, is as follow: Φ = x v z v y v z v x v y v z v y v x v z v y v x v z x z y y x z y x z y x η η η In the above equations x v, y v, z v, x ê, y ê, and z ê are the components of velocity field vector V and the unit vectors in the x, y, and z direction, respectively. T, P, ρ, k, η, and p c represent the temperature, pressure, density, thermal conductivity, viscosity and the specific heat of air, respectively.

167 Appendix B: Image analysis data 153

168 Table 1: Dyneema yarn diameter values in dry texturing

169 Table 2: Dyneema yarn diameter values in wet texturing 155

170 Table 3: Dyneema - Number and type of loops in dry and wet texturing 156

171 t= 558 sdev= 0.484E-02 degrees of freedom = 28 Dyneema: Dry v Wet texturing - Student t-test results for average values of yarn diameter Group A: Number of items= Mean = % confidence interval for Mean: thru Standard Deviation = 6.826E-03 Hi = 1.60 Low = 1.58 Median = 1.59 Average Absolute Deviation from Median = 5.529E-03 Group B: Number of items= Mean = % confidence interval for Mean: thru Standard Deviation = 5.293E-04 Hi = Low = Median = Average Absolute Deviation from Median = 3.609E-04 Based on t-table values, there is a significant difference between average values of yarn diameters 157

172 t= 151 sdev= 0.624E-02 degrees of freedom = 28 Dyneema: Dry v Wet texturing: Student t-test results for standard deviation values of yarn diameter Group A: Number of items= Mean = % confidence interval for Mean: thru Standard Deviation = 8.753E-03 Hi = Low = Median = Average Absolute Deviation from Median = 6.851E-03 Group B: Number of items= Mean = % confidence interval for Mean: thru Standard Deviation = 1.101E-03 Hi = Low = Median = Average Absolute Deviation from Median = 6.379E-04 Based on t-table values, there is a significant difference between values of standard deviation between the two conditions 158

173 Dyneema: Dry v Wet texturing: Student t-test results for core diameter t= 1.37 sdev= degrees of freedom = 28 Group A: Number of items= Mean = % confidence interval for Mean: thru Standard Deviation = Hi = 1.83 Low = 1.22 Median = 1.32 Average Absolute Deviation from Median = Group B: Number of items= Mean = % confidence interval for Mean: thru Standard Deviation = Hi = 1.54 Low = 1.12 Median = 1.30 Average Absolute Deviation from Median = 8.811E-02 Based on t-table values, there is no significant difference between the core diameters 159

174 Dyneema: Dry v Wet texturing: Student t-test results for number of individual loops t= 2.02 sdev= 4.03 degrees of freedom = 88 Group A: Number of items= Mean = % confidence interval for Mean: thru Standard Deviation = 4.27 Hi = 21.0 Low = 5.00 Median = 13.0 Average Absolute Deviation from Median = 3.47 Group B: Number of items= Mean = % confidence interval for Mean: thru Standard Deviation = 3.76 Hi = 20.0 Low = 4.00 Median = 12.0 Average Absolute Deviation from Median = 3.11 Based on t-table values, there is no significant difference between number of individual loops in either condition 160

175 Dyneema: Dry v Wet texturing: Student t-test results for number multi-fiber loops t= 3.35 sdev= 1.10 degrees of freedom = 88 Group A: Number of items= Mean = % confidence interval for Mean: thru Standard Deviation = 1.25 Hi = 6.00 Low = 1.00 Median = 3.00 Average Absolute Deviation from Median = 1.04 Group B: Number of items= Mean = % confidence interval for Mean: thru Standard Deviation = Hi = 5.00 Low = 1.00 Median = 2.00 Average Absolute Deviation from Median = Based on t-table values, there is a significant difference between number of multi-fiber loops in either condition 161

176 Dyneema: Dry v Wet texturing: Student t-test results for number of buckled loops t= 4.39 sdev= 1.46 degrees of freedom = 28 Group A: Number of items= Mean = % confidence interval for Mean: thru Standard Deviation = 1.51 Hi = 6.00 Low = 2.00 Median = 4.00 Average Absolute Deviation from Median = 1.20 Group B: Number of items= E E E Mean = % confidence interval for Mean: thru Standard Deviation = 1.40 Hi = 4.00 Low = 0.000E+00 Median = 1.00 Average Absolute Deviation from Median = 1.07 Based on t-table values, there is a significant difference between number of buckled loops in either condition 162

177 Table 4: Kevlar yarn diameter values in dry texturing 163

178 Table 5: Kevlar yarn diameter values in wet texturing 164

179 Table 6: Kevlar: Number and type of loops in dry and wet texturing 165

180 t= 13.8 sdev= 0.123E-01 degrees of freedom = 28 Kevlar: Dry v Wet texturing: Student t-test results for average values of yarn diameter Group A: Number of items= Mean = % confidence interval for Mean: thru Standard Deviation = 1.751E-03 Hi = 1.76 Low = 1.76 Median = 1.76 Average Absolute Deviation from Median = 1.489E-03 Group B: Number of items= Mean = % confidence interval for Mean: thru Standard Deviation = 1.733E-02 Hi = 1.76 Low = 1.69 Median = 1.69 Average Absolute Deviation from Median = 5.909E-03 Based on t-table values, there is a significant difference between average values of yarn diameters 166

181 t= 14.0 sdev= 0.209E-01 degrees of freedom = 28 Kevlar: Dry v Wet texturing: Student t-test results of standard deviation values of yarn diameter Group A: Number of items= Mean = % confidence interval for Mean: thru Standard Deviation = 2.147E-03 Hi = Low = Median = Average Absolute Deviation from Median = 1.493E-03 Group B: Number of items= Mean = % confidence interval for Mean: thru Standard Deviation = 2.949E-02 Hi = Low = Median = Average Absolute Deviation from Median = 9.839E-03 Based on t-table values, there is a significant difference between values of standard deviation between the two conditions 167

182 Kevlar: Dry v Wet texturing: Student t-test results for core diameter t= 0.988E-02 sdev= degrees of freedom = 28 Group A: Number of items= Mean = % confidence interval for Mean: thru Standard Deviation = Hi = 1.07 Low = Median = Average Absolute Deviation from Median = Group B: Number of items= Mean = % confidence interval for Mean: thru Standard Deviation = Hi = 1.31 Low = Median = Average Absolute Deviation from Median = Based on t-table values, there is no significant difference between the core diameters 168

183 Kevlar: Dry v Wet texturing: Student t-test results for number of individual loops t= 6.98 sdev= 3.48 degrees of freedom = 88 Group A: Number of items= Mean = % confidence interval for Mean: thru Standard Deviation = 3.87 Hi = 30.0 Low = 12.0 Median = 20.0 Average Absolute Deviation from Median = 2.62 Group B: Number of items= Mean = % confidence interval for Mean: thru Standard Deviation = 3.03 Hi = 20.0 Low = 10.0 Median = 15.0 Average Absolute Deviation from Median = 2.38 Based on t-table values, there is a significant difference between number of individual loops in either condition 169

184 Kevlar: Dry v Wet texturing: Student t-test results for number multi-fiber loops t= 1.08 sdev= degrees of freedom = 88 Group A: Number of items= Mean = % confidence interval for Mean: thru Standard Deviation = Hi = 5.00 Low = 1.00 Median = 2.00 Average Absolute Deviation from Median = Group B: Number of items= Mean = % confidence interval for Mean: thru Standard Deviation = 1.06 Hi = 5.00 Low = 1.00 Median = 2.00 Average Absolute Deviation from Median = Based on t-table values, there is no significant difference between number of multi-fiber loops in either condition 170

185 Kevlar: Dry v Wet texturing: Student t-test results for number of buckled loops t= 1.25 sdev= 6.16 degrees of freedom = 28 Group A: Number of items= Mean = % confidence interval for Mean: thru Standard Deviation = 7.47 Hi = 24.0 Low = 2.00 Median = 12.0 Average Absolute Deviation from Median = 6.20 Group B: Number of items= Mean = % confidence interval for Mean: thru Standard Deviation = 4.47 Hi = 22.0 Low = 9.00 Median = 16.0 Average Absolute Deviation from Median = 3.80 Based on t-table values, there is no significant difference between number of buckled loops in either condition 171

186 Table 7: Zylon yarn diameter values in dry texturing 172

187 Table 8: Zylon yarn diameter values in wet texturing 173

188 Table 9 Zylon: Number and type of loops in dry and wet texturing 174

189 t= 348 sdev= 0.131E-02 degrees of freedom = 28 Zylon: Dry v Wet texturing: Student t-test results for average values of yarn diameter Group A: Number of items= Mean = % confidence interval for Mean: thru Standard Deviation = 1.426E-03 Hi = 1.18 Low = 1.17 Median = 1.18 Average Absolute Deviation from Median = 7.465E-04 Group B: Number of items= Mean = % confidence interval for Mean: thru Standard Deviation = 1.182E-03 Hi = 1.34 Low = 1.34 Median = 1.34 Average Absolute Deviation from Median = 8.915E-04 Based on t-table values, there is a significant difference between values of standard deviation between the two conditions 175

190 t= 203 sdev= 0.184E-02 degrees of freedom = 28 Zylon: Dry v Wet texturing: Student t-test results for standard deviation values of yarn diameter Group A: Number of items= Mean = % confidence interval for Mean: thru Standard Deviation = 2.030E-03 Hi = Low = Median = Average Absolute Deviation from Median = 1.030E-03 Group B: Number of items= Mean = % confidence interval for Mean: thru Standard Deviation = 1.622E-03 Hi = Low = Median = Average Absolute Deviation from Median = 1.157E-03 Based on t-table values, there is a significant difference between values of standard deviation between the two conditions 176

191 Zylon: Dry v Wet texturing: Student t-test results for core diameter t= sdev= degrees of freedom = 28 Group A: Number of items= Mean = % confidence interval for Mean: thru Standard Deviation = Hi = Low = Median = Average Absolute Deviation from Median = Group B: Number of items= Mean = % confidence interval for Mean: thru Standard Deviation = Hi = Low = Median = Average Absolute Deviation from Median = 9.797E-02 Based on t-table values, there is no significant difference between the core diameters 177

192 Zylon: Dry v Wet texturing: Student t-test results for number of individual loops t= 1.14 sdev= 6.28 degrees of freedom = 88 Group A: Number of items= Mean = % confidence interval for Mean: thru Standard Deviation = 4.58 Hi = 22.0 Low = 6.00 Median = 16.0 Average Absolute Deviation from Median = 3.78 Group B: Number of items= Mean = % confidence interval for Mean: thru Standard Deviation = 7.61 Hi = 27.0 Low = 2.00 Median = 14.0 Average Absolute Deviation from Median = 6.40 Based on t-table values, there is no significant difference between number of individual loops in either condition 178

193 Zylon: Dry v Wet texturing: Student t-test results for number multi-fiber loops t= sdev= 1.54 degrees of freedom = 88 Group A: Number of items= E E Mean = % confidence interval for Mean: thru Standard Deviation = 1.79 Hi = 12.0 Low = 0.000E+00 Median = 1.00 Average Absolute Deviation from Median = Group B: Number of items= E E E E E Mean = % confidence interval for Mean: thru Standard Deviation = 1.24 Hi = 4.00 Low = 0.000E+00 Median = 2.00 Average Absolute Deviation from Median = Based on t-table values, there is no significant difference between number of multi-fiber loops in either condition 179

194 Zylon: Dry v Wet texturing: Student t-test results for number of buckled loops t= 2.67 sdev= 2.94 degrees of freedom = 28 Group A: Number of items= E E Mean = % confidence interval for Mean: thru Standard Deviation = 3.64 Hi = 9.00 Low = 0.000E+00 Median = 7.00 Average Absolute Deviation from Median = 3.07 Group B: Number of items= E E E Mean = % confidence interval for Mean: thru Standard Deviation = 2.00 Hi = 7.00 Low = 0.000E+00 Median = 1.00 Average Absolute Deviation from Median = 1.40 Based on t-table values, there is a significant difference between number of buckled loops in either condition 180

195 Table 10: Spectra yarn diameter values in dry texturing 181

196 Table11: Spectra yarn diameter values in wet texturing 182

197 Table 12 Spectra: Number and type of loops in dry and wet texturing 183

198 t= 29.6 sdev= 0.536E-02 degrees of freedom = 28 Spectra: Dry v Wet texturing: Student t-test results for average values of yarn diameter Group A: Number of items= Mean = % confidence interval for Mean: thru Standard Deviation = 2.240E-03 Hi = 2.13 Low = 2.13 Median = 2.13 Average Absolute Deviation from Median = 1.671E-03 Group B: Number of items= Mean = % confidence interval for Mean: thru Standard Deviation = 7.249E-03 Hi = 2.09 Low = 2.06 Median = 2.07 Average Absolute Deviation from Median = 4.584E-03 Based on t-table values, there is a significant difference between values of standard deviation between the two conditions 184

199 t= 23.4 sdev= 0.357E-02 degrees of freedom = 28 Spectra: Dry v Wet texturing: Student t-test results for standard deviation values of yarn diameter Group A: Number of items= Mean = % confidence interval for Mean: thru Standard Deviation = 1.761E-03 Hi = Low = Median = Average Absolute Deviation from Median = 1.290E-03 Group B: Number of items= Mean = % confidence interval for Mean: thru Standard Deviation = 4.733E-03 Hi = Low = Median = Average Absolute Deviation from Median = 3.546E-03 Based on t-table values, there is a significant difference between values of standard deviation between the two conditions 185

200 Spectra: Dry v Wet texturing: Student t-test results for core diameter t= 3.06 sdev= degrees of freedom = 28 Group A: Number of items= Mean = % confidence interval for Mean: thru Standard Deviation = Hi = 1.36 Low = Median = Average Absolute Deviation from Median = Group B: Number of items= Mean = % confidence interval for Mean: thru Standard Deviation = Hi = 1.04 Low = Median = Average Absolute Deviation from Median = Based on t-table values, there is no significant difference between the core diameters 186

201 Spectra: Dry v Wet texturing: Student t-test results for number of individual loops t= sdev= 3.59 degrees of freedom = 88 Group A: Number of items= Mean = % confidence interval for Mean: thru Standard Deviation = 3.98 Hi = 23.0 Low = 7.00 Median = 12.0 Average Absolute Deviation from Median = 2.98 Group B: Number of items= Mean = % confidence interval for Mean: thru Standard Deviation = 3.16 Hi = 20.0 Low = 6.00 Median = 12.0 Average Absolute Deviation from Median = 2.49 Based on t-table values, there is no significant difference between number of individual loops in either condition 187

202 Spectra: Dry v Wet texturing: Student t-test results for number multi-fiber loops t= sdev= degrees of freedom = 88 Group A: Number of items= Mean = % confidence interval for Mean: thru Standard Deviation = Hi = 4.00 Low = 1.00 Median = 2.00 Average Absolute Deviation from Median = Group B: Number of items= Mean = % confidence interval for Mean: thru Standard Deviation = Hi = 4.00 Low = 1.00 Median = 2.00 Average Absolute Deviation from Median = Based on t-table values, there is no significant difference between number of multi-fiber loops in either condition 188

203 Table 13: Technora yarn diameter values in dry texturing 189

204 Table 14: Technora yarn diameter values in wet texturing 190

205 Table 15 Technora: Number and type of loops in dry and wet texturing 191

206 t= 0.180E+04 sdev= 0.121E-02 degrees of freedom = 28 Technora: Dry v Wet texturing: Student t-test results for average values of yarn diameter Group A: Number of items= Mean = % confidence interval for Mean: thru Standard Deviation = 1.192E-03 Hi = 3.98 Low = 3.97 Median = 3.98 Average Absolute Deviation from Median = 8.329E-04 Group B: Number of items= Mean = % confidence interval for Mean: thru Standard Deviation = 1.229E-03 Hi = 3.18 Low = 3.18 Median = 3.18 Average Absolute Deviation from Median = 9.090E-04 Based on t-table values, there is a significant difference between values of standard deviation between the two conditions 192

207 t= 491 sdev= 0.112E-02 degrees of freedom = 28 Technora: Dry v Wet texturing: Student t-test results for standard deviation values of yarn diameter Group A: Number of items= Mean = % confidence interval for Mean: thru Standard Deviation = 1.548E-03 Hi = Low = Median = Average Absolute Deviation from Median = 1.009E-03 Group B: Number of items= Mean = % confidence interval for Mean: thru Standard Deviation = 3.265E-04 Hi = Low = Median = Average Absolute Deviation from Median = 2.314E-04 Based on t-table values, there is a significant difference between values of standard deviation between the two conditions 193

208 Technora: Dry v Wet texturing: Student t-test results for core diameter t= 6.03 sdev= degrees of freedom = 28 Group A: Number of items= Mean = % confidence interval for Mean: thru Standard Deviation = Hi = 3.96 Low = 1.81 Median = 2.70 Average Absolute Deviation from Median = Group B: Number of items= Mean = % confidence interval for Mean: thru Standard Deviation = Hi = 2.11 Low = 1.15 Median = 1.61 Average Absolute Deviation from Median = Based on t-table values, there is a significant difference between the core diameters 194

209 Technora: Dry v Wet texturing: Student t-test results for number of individual loops t= 5.32 sdev= 5.00 degrees of freedom = 88 Group A: Number of items= Mean = % confidence interval for Mean: thru Standard Deviation = 4.48 Hi = 25.0 Low = 6.00 Median = 14.0 Average Absolute Deviation from Median = 3.24 Group B: Number of items= Mean = % confidence interval for Mean: thru Standard Deviation = 5.46 Hi = 29.0 Low = 11.0 Median = 22.0 Average Absolute Deviation from Median = 4.53 Based on t-table values, there is a significant difference between number of individual loops in either condition 195

210 Technora: Dry v Wet texturing: Student t-test results for number multi-fiber loops t= 2.25 sdev= degrees of freedom = 88 Group A: Number of items= E E E Mean = % confidence interval for Mean: thru Standard Deviation = Hi = 4.00 Low = 0.000E+00 Median = 2.00 Average Absolute Deviation from Median = Group B: Number of items= Mean = % confidence interval for Mean: thru Standard Deviation = Hi = 4.00 Low = 1.00 Median = 2.00 Average Absolute Deviation from Median = Based on t-table values, there is a significant difference between number of multi-fiber loops in either condition 196

211 Table 16: Vectran yarn diameter values in dry texturing 197

212 Table 17: Vectran yarn diameter values in wet texturing 198

213 Table 18 Vectran: Number and type of loops in dry and wet texturing 199

214 t= 41.5 sdev= 0.111E-01 degrees of freedom = 28 Vectran: Dry v Wet texturing: Student t-test results for average values of yarn diameter Group A: Number of items= Mean = % confidence interval for Mean: thru Standard Deviation = 1.123E-02 Hi = 4.11 Low = 4.06 Median = 4.10 Average Absolute Deviation from Median = 4.672E-03 Group B: Number of items= Mean = % confidence interval for Mean: thru Standard Deviation = 1.094E-02 Hi = 3.97 Low = 3.93 Median = 3.93 Average Absolute Deviation from Median = 4.404E-03 Based on the t-table values, there is a significant difference between the yarn diameters in the two conditions 200

215 t= 128. sdev= 0.527E-02 degrees of freedom = 28 Vectran: Dry v Wet texturing: Student t-test results for standard deviation values of yarn diameter Group A: Number of items= Mean = % confidence interval for Mean: thru Standard Deviation = 3.658E-03 Hi = Low = Median = Average Absolute Deviation from Median = 2.032E-03 Group B: Number of items= Mean = % confidence interval for Mean: thru Standard Deviation = 6.495E-03 Hi = Low = Median = Average Absolute Deviation from Median = 4.167E-03 Based on the t-table values, there is a significant difference between the yarn diameters in the two conditions 201

216 Vectran: Dry v Wet texturing: Student t-test results for core diameter t= sdev= degrees of freedom = 28 Group A: Number of items= Mean = % confidence interval for Mean: thru Standard Deviation = Hi = 1.66 Low = Median = 1.25 Average Absolute Deviation from Median = Group B: Number of items= Mean = % confidence interval for Mean: thru Standard Deviation = Hi = 2.02 Low = Median = 1.27 Average Absolute Deviation from Median = Based on the t-table values, there is no significant difference between the core diameters in the two conditions 202

217 Vectran: Dry v Wet texturing: Student t-test results for number of individual loops t= 4.40 sdev= 5.87 degrees of freedom = 88 Group A: Number of items= Mean = % confidence interval for Mean: thru Standard Deviation = 6.54 Hi = 40.0 Low = 20.0 Median = 25.0 Average Absolute Deviation from Median = 5.42 Group B: Number of items= Mean = % confidence interval for Mean: thru Standard Deviation = 5.12 Hi = 35.0 Low = 15.0 Median = 23.0 Average Absolute Deviation from Median = 4.16 Based on the t-table values, there is a significant difference between the core diameters in the two conditions 203

218 Vectran: Dry v Wet texturing: Student t-test results for number multi-fiber loops t= 1.28 sdev= 1.07 degrees of freedom = 88 Group A: Number of items= Mean = % confidence interval for Mean: thru Standard Deviation = Hi = 5.00 Low = 2.00 Median = 3.00 Average Absolute Deviation from Median = Group B: Number of items= Mean = % confidence interval for Mean: thru Standard Deviation = 1.30 Hi = 6.00 Low = 1.00 Median = 4.00 Average Absolute Deviation from Median = 1.04 Based on the t-table values, there is no significant difference between number of multifiber loops in the two conditions 204

219 Vectran: Dry v Wet texturing: Student t-test results for number buckled loops t= 2.36 sdev= 4.25 degrees of freedom = 28 Group A: Number of items= Mean = % confidence interval for Mean: thru Standard Deviation = 2.91 Hi = 9.00 Low = 1.00 Median = 4.00 Average Absolute Deviation from Median = 2.47 Group B: Number of items= Mean = % confidence interval for Mean: thru Standard Deviation = 5.26 Hi = 16.0 Low = 1.00 Median = 8.00 Average Absolute Deviation from Median = 4.53 Based on the t-table values, there is a significant difference between the core diameters in the two conditions 205

220 Table 19: Dacron yarn diameter values in dry texturing 206

221 Table 20: Dacron yarn diameter values in wet texturing 207

222 Table 21:Dacron: Number and type of loops in dry and wet texturing 208

223 t= 118. sdev= 0.285E-02 degrees of freedom = 28 Dacron: Dry v Wet texturing: Student t-test results for average values of yarn diameter Group A: Number of items= Mean = % confidence interval for Mean: thru Standard Deviation = 3.955E-03 Hi = Low = Median = Average Absolute Deviation from Median = 2.984E-03 Group B: Number of items= Mean = % confidence interval for Mean: thru Standard Deviation = 7.690E-04 Hi = Low = Median = Average Absolute Deviation from Median = 3.988E-04 Based on the t-table there is a significant difference in the values of yarn diameter between the two conditions 209

224 t= 14.8 sdev= 0.216E-02 degrees of freedom = 28 Dacron: Dry v Wet texturing: Student t-test results for standard deviation values of yarn diameter Group A: Number of items= Mean = % confidence interval for Mean: thru Standard Deviation = 2.668E-03 Hi = Low = Median = Average Absolute Deviation from Median = 1.874E-03 Group B: Number of items= Mean = % confidence interval for Mean: thru Standard Deviation = 1.481E-03 Hi = Low = Median = Average Absolute Deviation from Median = 6.200E-04 Based on the t-table there is a significant difference in the values between the two conditions 210

225 Dacron: Dry v Wet texturing: Student t-test results for number of individual loops t= 5.45 sdev= 2.67 degrees of freedom = 88 Group A: Number of items= Mean = % confidence interval for Mean: thru Standard Deviation = 2.91 Hi = 12.0 Low = 1.00 Median = 7.00 Average Absolute Deviation from Median = 2.38 Group B: Number of items= Mean = % confidence interval for Mean: thru Standard Deviation = 2.40 Hi = 14.0 Low = 4.00 Median = 9.00 Average Absolute Deviation from Median = 1.98 Based on the t-table there is a significant difference in the values between the two conditions 211

226 Dacron: Dry v Wet texturing: Student t-test results for number of multifiber loops t= 11.0 sdev= degrees of freedom = 88 Group A: Number of items= E E E E E E E E E E E E E E E E E E E E E E E E+ Mean = % confidence interval for Mean: thru Standard Deviation = Hi = 2.00 Low = 0.000E+00 Median = 0.000E+00 Average Absolute Deviation from Median = Group B: Number of items= E E Mean = % confidence interval for Mean: thru Standard Deviation = Hi = 4.00 Low = 0.000E+00 Median = 2.00 Average Absolute Deviation from Median = Based on the t-table there is a significant difference in the values between the two conditions 212

227 Dacron: Dry v Wet texturing: Student t-test results for number of buckled loops t=0.117 sdev= 3.12 degrees of freedom = 28 Group A: Number of items= E E Mean = % confidence interval for Mean: thru Standard Deviation = 2.90 Hi = 9.00 Low = 0.000E+00 Median = 6.00 Average Absolute Deviation from Median = 2.33 Group B: Number of items= Mean = % confidence interval for Mean: thru Standard Deviation = 3.33 Hi = 11.0 Low = 1.00 Median = 4.00 Average Absolute Deviation from Median = 2.33 Based on the t-table there is no significant difference in the values between the two conditions 213

228 Table 22: Dacron: Effect of air pressure (80 PSI) on average values of yarn diameter 214

229 Table 23: Dacron: Effect of air pressure (180 PSI) on average values of yarn diameter 215

230 Table 24: Dacron, different air pressures, type and number of loops 216

231 t= 143 sdev= 0.349E-02 degrees of freedom = 28 Dacron: Student t-test results for average values of yarn diameter at 80 PSI and 180 PSI Group A: Number of items= Mean = % confidence interval for Mean: thru Standard Deviation = 3.410E-03 Hi = 1.04 Low = 1.03 Median = 1.04 Average Absolute Deviation from Median = 2.647E-03 Group B: Number of items= Mean = % confidence interval for Mean: thru Standard Deviation = 3.572E-03 Hi = Low = Median = Average Absolute Deviation from Median = 3.002E-03 Based on the t-table there is a significant difference in the values between the two conditions 217

232 Dacron: Student t-test results for standard deviation values of 80 PSI and 180 PSI t= 105 sdev= 0.263E-02 degrees of freedom = 28 Group A: Number of items= Mean = % confidence interval for Mean: thru Standard Deviation = 2.855E-03 Hi = Low = Median = Average Absolute Deviation from Median = 2.072E-03 Group B: Number of items= Mean = % confidence interval for Mean: thru Standard Deviation = 2.373E-03 Hi = Low = Median = Average Absolute Deviation from Median = 1.422E-03 Based on the t-table there is a significant difference in the values between the two conditions 218

233 Dacron: Student t-test results for core diameter values at 80 PSI and 180 PSI t=-0.790e-01 sdev= degrees of freedom = 28 Group A: Number of items= Mean = % confidence interval for Mean: thru Standard Deviation = Hi = Low = Median = Average Absolute Deviation from Median = 9.257E-02 Group B: Number of items= Mean = % confidence interval for Mean: thru Standard Deviation = Hi = Low = Median = Average Absolute Deviation from Median = 8.244E-02 Based on the t-table there is a significant difference in the values between the two conditions 219

234 Dacron: 80 PSI v 180 PSI: Student t-test results for number of individual loops t= 5.62 sdev= 3.68 degrees of freedom = 88 Group A: Number of items= Mean = % confidence interval for Mean: thru Standard Deviation = 3.66 Hi = 22.0 Low = 7.00 Median = 15.0 Average Absolute Deviation from Median = 2.91 Group B: Number of items= Mean = % confidence interval for Mean: thru Standard Deviation = 3.70 Hi = 21.0 Low = 5.00 Median = 10.0 Average Absolute Deviation from Median = 2.89 Based on the t-table there is a significant difference in the values between the two conditions 220

235 Dacron: 80 PSI v 180 PSI: Student t-test results for number of multifiber loops t= sdev= 1.02 degrees of freedom = 88 The probability of this result, assuming the null hypothesis, is 0.36 Group A: Number of items= Mean = % confidence interval for Mean: thru Standard Deviation = 1.21 Hi = 6.00 Low = 1.00 Median = 2.00 Average Absolute Deviation from Median = Group B: Number of items= Mean = % confidence interval for Mean: thru Standard Deviation = Hi = 4.00 Low = 1.00 Median = 2.00 Average Absolute Deviation from Median = Based on the t-table there is no significant difference in the values between the two conditions 221

236 Dacron: 80 PSI v 180 PSI: Student t-test results for number of buckled loops t= 6.35 sdev= 2.76 degrees of freedom = 28 The probability of this result, assuming the null hypothesis, is less than.0001the probability of this result, assuming the null hypothesis, is less than.0001 Group A: Number of items= Mean = % confidence interval for Mean: thru Standard Deviation = 3.55 Hi = 18.0 Low = 5.00 Median = 9.00 Average Absolute Deviation from Median = 2.73 Group B: Number of items= Mean = % confidence interval for Mean: thru Standard Deviation = 1.64 Hi = 6.00 Low = 1.00 Median = 4.00 Average Absolute Deviation from Median = 1.33 Based on the t-table there is a significant difference in the values between the two conditions 222

237 Table 25: Dacron yarn diameter at 25% overfeed 223

238 Table 26: Dacron yarn diameter at 45% overfeed 224

239 Table 27: Dacron effect of overfeed, number and type of loops 225

240 t= 120 sdev= 0.557E-02 degrees of freedom = 28 Dacron: 25% Overfeed v 45% Overfeed PSI: Student t-test results for average values of yarn diameter Group A: Number of items= Mean = % confidence interval for Mean: thru Standard Deviation = 6.629E-03 Hi = Low = Median = Average Absolute Deviation from Median = 4.763E-03 Group B: Number of items= Mean = % confidence interval for Mean: thru Standard Deviation = 4.258E-03 Hi = 1.04 Low = 1.03 Median = 1.03 Average Absolute Deviation from Median = 3.526E-03 Based on t-table there is a significant difference in the values of yarn diameter in the two conditions 226

241 Dacron: 25% Overfeed v 45% Overfeed PSI: Student t-test results for standard deviation values of yarn diameter t= 46.0 sdev= 0.495E-02 degrees of freedom = 28 Group A: Number of items= Mean = % confidence interval for Mean: thru Standard Deviation = 6.456E-03 Hi = Low = Median = Average Absolute Deviation from Median = 3.780E-03 Group B: Number of items= Mean = % confidence interval for Mean: thru Standard Deviation = 2.705E-03 Hi = Low = Median = Average Absolute Deviation from Median = 2.249E-03 Based on t-table there is a significant difference in the values of yarn diameter in the two conditions 227

242 t= 8.57 sdev= 2.96 degrees of freedom = 88 Dacron: 25% Overfeed v 45% Overfeed PSI: Student t-test results for individual loops Group A: Number of items= Mean = % confidence interval for Mean: thru Standard Deviation = 2.43 Hi = 17.0 Low = 5.00 Median = 9.00 Average Absolute Deviation from Median = 1.93 Group B: Number of items= Mean = % confidence interval for Mean: thru Standard Deviation = 3.42 Hi = 22.0 Low = 7.00 Median = 15.0 Average Absolute Deviation from Median = 2.53 Based on t-table results, there is a significant difference between the values in the two situations 228

243 Dacron: 25% Overfeed v 45% Overfeed: Student t-test results for multifiber loops t= 1.58 sdev= 1.07 degrees of freedom = 88 Group A: Number of items= E E E E E Mean = % confidence interval for Mean: thru Standard Deviation = 1.06 Hi = 4.00 Low = 0.000E+00 Median = 2.00 Average Absolute Deviation from Median = Group B: Number of items= Mean = % confidence interval for Mean: thru Standard Deviation = 1.07 Hi = 5.00 Low = 1.00 Median = 2.00 Average Absolute Deviation from Median = Based on t-table results, there is no significant difference between the values in the two situations 229

244 t= 4.73 sdev= 3.55 degrees of freedom = 28 Dacron: 25% Overfeed v 45% Overfeed PSI: Student t-test results for buckled loops Group A: Number of items= Mean = % confidence interval for Mean: thru Standard Deviation = 2.91 Hi = 11.0 Low = 2.00 Median = 4.00 Average Absolute Deviation from Median = 1.93 Group B: Number of items= Mean = % confidence interval for Mean: thru Standard Deviation = 4.09 Hi = 18.0 Low = 4.00 Median = 12.0 Average Absolute Deviation from Median = 3.13 Based on t-table results, there is a significant difference between the values in the two situations 230

245 Table 28: Dyneema average value of yarn diameter at 10% stretch 231

246 Table 29: Dyneema average value of yarn diameter at 15% stretch 232

247 Table 30: Dyneema Effect of stretch on number and type of loops 233

248 t= 18.6 sdev= 0.218E-01 degrees of freedom = 28 Dyneema: 10% stretch v 15% stretch: Student t-test results for average values of yarn diameter Group A: Number of items= Mean = % confidence interval for Mean: thru Standard Deviation = 2.605E-02 Hi = 2.07 Low = 1.97 Median = 2.02 Average Absolute Deviation from Median = 1.968E-02 Group B: Number of items= Mean = % confidence interval for Mean: thru Standard Deviation = 1.654E-02 Hi = 1.91 Low = 1.84 Median = 1.87 Average Absolute Deviation from Median = 1.075E-02 Based on t-table results, there is a significant difference between the values in the two situations 234

249 t= 46.4 sdev= 0.172E-01 degrees of freedom = 28 Dyneema: 10% stretch v 15% stretch: Student t-test results for standard deviation values of yarn diameter Group A: Number of items= Mean = % confidence interval for Mean: thru Standard Deviation = 2.285E-02 Hi = Low = Median = Average Absolute Deviation from Median = 1.936E-02 Group B: Number of items= Mean = % confidence interval for Mean: thru Standard Deviation = 8.419E-03 Hi = Low = Median = Average Absolute Deviation from Median = 5.910E-03 Based on t-table results, there is a significant difference between the values in the two situations 235

250 Dyneema: 10% stretch v 15% stretch: Student t-test results individual loops t= 2.25 sdev= 3.80 degrees of freedom = 88 Group A: Number of items= Mean = % confidence interval for Mean: thru Standard Deviation = 4.32 Hi = 20.0 Low = 3.00 Median = 10.0 Average Absolute Deviation from Median = 3.42 Group B: Number of items= Mean = % confidence interval for Mean: thru Standard Deviation = 3.18 Hi = 18.0 Low = 5.00 Median = 8.00 Average Absolute Deviation from Median = 2.24 Based on t-table results, there is a significant difference between the values in the two situations 236

251 Dyneema: 10% stretch v 15% stretch: Student t-test results multifiber loops t= 1.45 sdev= 1.89 degrees of freedom = 88 Group A: Number of items= Mean = % confidence interval for Mean: thru Standard Deviation = 2.22 Hi = 11.0 Low = 2.00 Median = 6.00 Average Absolute Deviation from Median = 1.84 Group B: Number of items= Mean = % confidence interval for Mean: thru Standard Deviation = 1.48 Hi = 9.00 Low = 3.00 Median = 6.00 Average Absolute Deviation from Median = 1.22 Based on t-table results, there is no significant difference between the values in the two situations 237

252 Dyneema: 10% stretch v 15% stretch: Student t-test results buckled loops t= sdev= 2.08 degrees of freedom = 28 Group A: Number of items= Mean = % confidence interval for Mean: thru Standard Deviation = 1.60 Hi = 7.00 Low = 1.00 Median = 3.00 Average Absolute Deviation from Median = 1.13 Group B: Number of items= E Mean = % confidence interval for Mean: thru Standard Deviation = 2.47 Hi = 9.00 Low = 0.000E+00 Median = 4.00 Average Absolute Deviation from Median = 2.00 Based on t-table results, there is no significant difference between the values in the two situations 238

253 Table 31: Dyneema: Effect of 15% overfeed on yarn diameter values 239

254 Table 32 :Dyneema: Effect of 45% overfeed on yarn diameter values 240

255 Dyneema 15% and 45%: Student t-test results for average yarn diameter values t= 298 sdev= 0.108E-01 degrees of freedom = 28 Group A: Number of items= Mean = % confidence interval for Mean: thru Standard Deviation = 1.359E-02 Hi = 2.03 Low = 1.98 Median = 2.01 Average Absolute Deviation from Median = 8.489E-03 Group B: Number of items= Mean = % confidence interval for Mean: thru Standard Deviation = 6.939E-03 Hi = Low = Median = Average Absolute Deviation from Median = 5.347E-03 Based on t-table values, there is a significant difference in the average yarn diameter with increase in overfeed 241

256 Dyneema 15% and 45%: Student t-test results for standard deviation values t= 150 sdev= 0.109E-01 degrees of freedom = 28 Group A: Number of items= Mean = % confidence interval for Mean: thru Standard Deviation = 1.045E-02 Hi = Low = Median = Average Absolute Deviation from Median = 7.005E-03 Group B: Number of items= Mean = % confidence interval for Mean: thru Standard Deviation = 1.138E-02 Hi = Low = Median = Average Absolute Deviation from Median = 7.213E-03 Based on t-table values, there is a significant difference in the standard deviation with increase in overfeed 242

257 Dyneema 15% and 45%: Student t-test results for core diameter values t= 1.94 sdev= degrees of freedom = 28 Group A: Number of items= Mean = % confidence interval for Mean: thru Standard Deviation = Hi = Low = Median = Average Absolute Deviation from Median = Group B: Number of items= Mean = % confidence interval for Mean: thru Standard Deviation = Hi = Low = Median = Average Absolute Deviation from Median = 8.580E-02 Dyneema, effect of overfeed, individual, multifiber and buckled loops, No loops of any kind formed at 15% so no comparison. 243

258 Table 33: Kevlar: 100 PSI 244

259 Table 34: Kevlar: 180 PSI 245