Characterization and Control of Fabric Properties inTextile and Apparel Manufacturing

PIS: P. Banks-Lee, N. C. State; D. R. Buchanan, N. C. State; T. G. Clapp, N. C. State;J.W. Eischen, N. C. State; T. K. Ghosh, N. C. State; B. S. Gupta, N. C. State;T. J. Little, N. C. State:

File: S92C3, Team Leader: T. G. Clapp

RELEVANCE TO NTC GOALS:The apparel segment of the U.S. textile industrial complex is under tremendous pressure

from foreign competition. The apparel industry is trying to respond to this pressure by changingits production philosophy to accommodate the consumer demand for ever increasing stylevariation and retailer demands for more frequent Just-In-Time (JIT) deliveries of smallerorders. The apparel industry must be able to respond to these demands to remain competitive inthe U.S. and abroad.

Apparel manufacturing in a Quick Response (QR) environment must have the ability toquickly optimize its processes when faced with rapidly changing fabrics with variable materialproperties. This variability is inherent (statistical) within a given fabric and between differenttypes of fabric. In order to deal successfully with this challenge, it is necessary to understandhow important material properties permeate through the manufacturing chain. We must developtechnologies to measure these important properties quickly and reliably (not dependent ontraditional laboratory techniques and constraints). It is important to lay the groundwork fordealing with property variability in the design of apparel fabrics as well as in the manufacturingand finishing of these products.

An interdisciplinary team of researchers from Textile Engineering (TE), ElectricalEngineering (EE), Mechanical Engineering (ME), and Textile Technology (IT) are conductingresearch to collectively address the following research themes: 1) model fabric behavior, 2)on-line characterization of material properties, and 3) on-line manufacturing control. Newdevelopments in computer modeling and sensor technology are being applied to address thesethemes.

GOALS:LONG TERM: Develop new technologies and methods to characterize fabric properties on-line(during processing) and control textile and apparel manufacturing processes to optimize qualityand productivity in a Quick Response or Just-in-Time manufacturing environment. Educategraduate and undergraduate students to help strengthen and lead U.S. fiber, textile, and apparelcompanies into the 21st century.SHORT TERM; Using common fabrics as a basis, 1) validate 3-D fabric deformation models,2) refine and validate a theoretical model for predicting visco-elastic behavior of fabrics in rollform, 3) develop and validate a theoretical model for measuring fabric stiffness on-line, 4) de-velop sensor technology to characterize fabric on-line during high speed sewing, and 5) quan-tify fabric frictional behavior and mechanical properties as a function of orientation. Involve atleast three undergraduates in research activities, and graduate at least two, Ph.D. and two Mas-ter’s students.

S92C3,lNational Textile Center Annual Report: Septembcr30,1993 177

TECHNICAL QUALITY AND ACCOMPLISHMENTSAn interdisciplinary team of researchers from Textile Engineering, Electrical Engineering,

Mechanical Engineering, and Textile Technology have collectively addressed the followingresearch themes: 1) model fabric behavior, 2) on-line characterization of material properties,and 3) manufacturing control. The technical quality of the research is reflected by the numberof theses (5), papers (7) and presentations (8) submitted in 1992-1993. These are listed below.A brief summary of each of the five research tasks is also provided.

Thesis Completed and in Preparation1. Rim, Y. G., “Fabric Manipulation Simulation Including Material Non-linearity and

Contact”, Ph.D Dissertation, Completed June 10, 1993.2. Timble, N., ” Structural Factors Affecting Interfacial Forces Between Fabrics, Ph.D.

Thesis”, submission in Final Form expected by October 25, 1993.3. McDevitt, T., “Flexible Fabric Mechanics Analysis Using Large Deflection Beam

Theory”, MS Thesis, final defense scheduled August 17, 1993.4. Kong, D., “Development of a Control System for Fabric Roll-Making Operation”, MS

Thesis final defense schedualed August 19,1993.5. Li, H., “Simulated Roll-Like Loading in Laboratory and it’s Effects on Fabric Properties”

in preparation.

Dissemination:Papers published (1992-1993):1.

2.

3.

4.

5.

6.

7.

Eischen, J. W., and Rim, Y. G., “Optimization of Fabric Manipulation During Pick/PlaceOperations,” Proceedings of the 4th Annual Academic Apparel Research Conference, Ra-leigh, NC, Feb. 8-9, 1993 and published in International Journal of Clothing Science andTechnology, Vol. 4, No 5, 1993.Clapp T. G., Timble, N and Gupta, B. S., “Structural Factors Affecting Interfacial Forcesin Fabrics.” Poster Paper, the First Annual National Textile Center Forum, Auburn, AL,February 12,1993.EL-Mogahzy, Y. E. and Gupta, B. S., “Friction in Fibrous Materials, Part II: ExperimentalStudy of the Effects of Structural and Morphological Factors.” Textile Res J., 63(4): 219-230 (1993).Gupta, B. S., Leek, F. J., Barker, R. L., Buchanan, D. R., and Little, T. J., “DirectionalVariations in Fabric Properties and Seam Quality,” International Journal of Clothing Sci-ence and Technology, 3 (2/3): 7 l-78 (1992).Clapp, T. G., Barrett, G., “Characterization of Sewing Machine Dynamics Using ForceSensors”, Proceedings of the 93 ASME Annual Conference, San Francisco, Dee 1992.Barrett, G., Clapp, T. G., “On-Line Characterization of Sewing Machine & Fabric Dynam-ics”, Proceedings of the 4th Annual Academic Apparel Research Conference, Raleigh, NC,Feb 8-9, 1993Eishen, J. W., McDevitt, T., “Simulation of Fabric Draping and Manipulation with Arbi-trary Contact Surfaces and Adaptive Meshing,” Proceedings of the 4th Annual AcademicApparel Research Conference, Raleigh, NC, Feb. 8-9, 1993.

Invention Disclosure (1992-1993):Barrett, G., Clapp, T. G., “On-Line Pressure Foot Force Controller for Sewing Machines”.

s92c3,2

178 National Textile Center Annual Report: Scptnnba 30,1993

Presentations (1992-1993):1. Eischen, J. W., “Computer Simulation of Fabric Drape and Manipulation,” invited seminar

at DuPont Experimental Station, Wilmington, DE, April 6,1993.2. Gunner, M., Clapp, T., have made various trips and presentations to Sara Lee, Russell Cor-

poration and Union Special3. Gunnner, M., gave presentations at Bellow Machine Co., De Montford University, Brad-

ford University and the University of Hull, all England.

Conferences and Exhibitions Attended (1992- 1993):1. Clapp, T., attended the ATME Exhibition, Greenville, SC, April 22, 19932. Gunner, M., attended the Japanese International Apparel Machinery (JIAM) Exhibition,

Tokyo, Japan, May 9 - 12 1993, also visiting Brother and Pegasus factories.3. Gunner, M., Clapp, T., attended an AMTEX meeting, Velvet Cloak Inn, Raleigh, NC, Nov

19-20, 1992.4. NTC Conference, Auburn, AL, 11-13 February 1993.5. Gunner, M., attended the Technology Modernization Conference Defense Personnel Sup-

ply Center, Philadelphia, June 3, 1993.6. Gunner, M., attended the CIM Workshop, “CIM defining the future”, NCSU, June 20-26,

1993.

RESEARCH SUMMARIES:

Models for Predicting Fabrics Friction and Directional Variationin Fabric Mechanical and Visco-elastic Properties.

Using a series of model fabrics constructed in house, surface frictional behavior was2evalu-ated using a specially set up device and eight (8) different normal pressures (3-160 g/m ). Theresults were fitted to the model (F/A) = C (N/A)” with correlation coefficient exceeding 0.98.

The values of the friction constants C and n, thus estimated, could be used to constructpredictive models in which the effects of fiber type, yarn type, yarn twist, weave tightness(picks per inch) and fabric finish could be examined. Two such models giving the effects offilling yarn twist (T) and picks per inch (P) with R2 greater than 0.9 are given below.

n = 0.894 + 0.049(T) + 0.001 (P)

C = 0.306 - 0.0240 - 0.002 (P)

These models show that the value of n increased and that of C decreased as the picks countand yarn twist increased. Using the classical equation, p = (F/A) / (N/A), the values of thecoefficient of friction, CL, were calculated. It was observed that the coefficient of frictionincreased with increase in yarn twist and decrease in picks count.

In order to rationalize the trends, a preliminary mathematical model was developed toestimate (1) the relative area of contact between contacting fabric surfaces, and (2) the relativeheight of the yarn crowns protruding from the fabric plane. The calculations showed that thearea of contact between the protruding yarn crowns and fabric surface increased with increasein T and decrease in P which supports the trends obtained for the value of the coefficient offriction. The same calculations could be used to support the trends obtained for n and C.

S92C3,3National Textile Center Annual Jkport: Septemba30,1993 179

On the effects of yarn type, it was found that the spun yarn fabrics had high C and low nwhile the continuous Nament yam fabrics had low C and high n values. The test in which afabric rubbed against a metal surface (stainless steel) showed that the fabric-to-metal frictionwas not influenced significantly by the fabric stru~turai factors. The present effort is beingconcenuated on developing sound nationals for the various results, writing of the doctoraldissertation, and writing of NVO papers.

For studying and modeling directional variations in fabric mechanical and v&o-elasticproperties, experimental work was completed on a plain woven fabric used by our team as abase fabric. It involved laser cutting strips at different angles to warp, from 0” to 90°, measuringtensile properties, removing warp and weft yarns and measuring their dimensional and tensileproperties, and pulling single warp and weft yarns from given widths .of fabric to determine thefrictional resistance exerted at the cross-over points. The results of tensile tests on fabrics areshown in Table 1 and an example of pull out frictional test on warp yarn is shown in Figure 1.The solid line in the figure is the theoretically predicted curve using a modification of thecapstan equation. The focus at present is in modeling the tensile behavior of the fabric using afinite element model and geometrical model. Both approaches have provided encouragingresults which will be reported in the next report.

Anglo MeanPeak MeanPeak MeanPeak M.an M*atl hiaonOf warp Load Elong slms~ Peak Peak Modulus

KGI cm KGUgrrVcm Slmln E n e r g y (KGl/gnu’cm)Y. KG&cm

“__” 18 26.91 5.649 306110

-- .-, 0.4525 1 316.13 lz63 2sn 23szu36301 I 0.6597 ( 24261 3298 2404 u773.3

11 31.71 2149 1149.12

-- .___,., .-. 00 61.66 9.167 1308.19

7p I 22s113t lJ27 ( SM.02 sbu 9.28 164.931

1.089 1 i96.16 54.43 1263 210415

Table 1. Tensile Properties of Fabric Measured atDifferent Angles to Warp.

Figure 1. Frictional Resistance at Cross-OverPoints of warp Yams.

Computational/Experimental Determination ofFabric Drape and Motion

Y. G. Kim- Fabric Manipulation Including Material Nonlinearitv and ContactThe primary objective of this research was to develop a computer simulation tool to allow

modeling of various fabric manipulation processes that occur during manufacturing. Thephysical model of the fabric is a very flexible elastic beam (or strip) with a nonlinearmoment-curvature relationship. The governing equilibrium equations are solved using the finite

s92c3,4

National Tcxtilc Center Annual Report: Septemba 30,1993180

element method. The nonlinear moment curvature response is measured directly with theKawabata Test System, or with a simpler drape test.

Realistic manipulation processes involve interaction of fabric parts with other objects suchas: work surfaces, robot manipulators, other fabrics. Therefore, the ability to model contact hasbeen implemented. A key aspect of this research has been the development of a capability todetermine optimum ways to manipulate fabric parts while minimizing sliding of the fabric orforces generated in the fabric. This feature will play an important role in developing automatedequipment to handle fabric parts while maintaining positional control and minimizing wrinklingor buckling. Exper+nents were conducted to verify the computer generated fabric drape shapesand excellent agreement was obtained. Furthermore, this research has shown it is crucial to usethe actual nonlinear relationship between moment and curvature, i.e. using the linear B valuegenerated by the Kawabata test is not sufficient.

T. McDevitt- Flexible Fabric Mechanics Analvsis Using Large Defection Beam TheorvThe primary objective of this research was to develop a computer method to simulate the

quasi-static motion of fabric parts during certain manufacturing processes in the apparelindustry. Fabric parts are modeled as very flexible elastic beams that can accommodatestretching and bending in a single plane. Again, the equilibrium equations are solved using thefinite element method. When solving for drape shapes and fabric configurations duringmanipulation, numerical problems are encountered when the fabric begins to buckle or wrinkle.In this project, a technique known as arc-length control has been implemented to treat thesedifficulties, and be able to simulate very complex load deflection paths. Attention has also beenfocused on generating finite element meshes that adapt to the severity of the deformation.Generally speaking, a numerical method requires more refinement (nodes) in the areas wherethe state variables (displacements) are varying the most rapidly. A method to distribute thefinite element nodes to those regions of the fabric undergoing the most curvature has beenimplemented. Figure 2 shows a simulation of a strip of fabric passing through a “bottom feedermachine”. This machine was sent to the College of Textiles at NCSU by Ark Inc. in hopes ofimproving the picking operation. The computer simulations allows critical dimensions on themachine to be adjusted for optimum performance over a range of fabric stiffnesses.

S. Deng- 3D Fabric Draue MechanicsThis project is an ongoing extension of our capability to model three dimensional fabric

drape over complicated surfaces. The use of large displacement shell theory to model thefabrics behavior has proved to be very successful. A very interesting aspect of this work is thediscovery of multiple numerical solutions for seemingly simple fabric shapes. The solutions ofthe highly nonlinear equilibrium equations is non-unique. The challenge is to determine whichof the several solutions is the physical realizable one, or whether multiple physical solutions arepossible. Figure 3 shows two drape modes for a square sheet of fabric draping over a block. Weare also working with the North Carolina Supercomputer Center to generate high quality 3Dcomputer graphics animations of fabric drape.

s92c3,5

National Textile CenterAnnual Report: September30,1993 181

Figure 2. The Bottom Feeder (Ark- Inc.) Figure 3. Multiple Drape Shapes

Develop on “Intelligent” Closed Loop System to Control the Roll Making Process

A feed-back control system for fabric roll-making has been developed. The controlsystem, in principle, would allow production of fabric rolls with optimal levels of internalstresses. The system is also designed to minimize variation of stresses.

Traditionally, fabric rolls are produced without any particular concern for the windingtension or roll-size. Their levels are mainly dictated by the unit process or the experiential.knowledge of the operator. Generally, the winding tension remains somewhat constant duringany process. It has been demonstrated in this research that in case of constant tension windingthe fabric inside the roll may actually buckle and may cause permanent deformation of thefabric. This is particularly undesirable for easily deformable or delicate fabrics such as certaintypes of fine woolen or knit fabrics. To demonstrate the levels of stresses developed in fabricrolls and their variation strain measurement systems for both compressive and tensile strainshave been developed. A miniature compression load cell has been used to measure the radialcompressive stresses within a roll. Figure 4 shows the results of such a measurement. In thiscase the compressive stress has been measured at the core of the roll at various sizes of the roll.The result shows initial rapid increase in compressive stress followed by a leveling off. Thissuggests that after a certain number of layers on the core the hoop stresses are not transmitted tothe core from further layers. Maybe the intermediate layers together act as a non-deformablerigid cylinder. To measure the in-plane tensile stresses foil-type strain gages have been used. Itwas necessary to develop special mounting techniques to obtain any meaningful measurementof strains in fabric layers. Figure 5 shows results from a typical measurement of in-planetension. The plot shows state of strain measured at various layers with increasing number oflayers in the roll. The results show that in some layers the fabric is actually under in-planecompressive stresses, even though, the web tension in the fabric at the time of winding waspositive. These results substantiate earlier results of theoretical analyses during this project.

In the control system developed in this project the winding tension is used as the controlparameter. In this, the vertical position of a control cylinder is precisely controlled to generatevarious levels of wrap angle which in turn determines the outgoing tension in the fabric. Themeasured tension in the fabric web is used to control the position of the control cylinder. Thecontrol cylinder is driven by an electric cylinder. Software for the control system has been

182 National Textile Cmter Annual Report: September 30,1993

S92C3,6

developed by using a graphical programming language LabView2. Using this programwinding tension can be controlled as function of time.

the

Figure 4. Variation of Radial COITpeSSiVeStress Measured at the Empty Tubewith Various Numbers of LayersWound on the Tube.

Figure 5. Variation of Tensile Stress at Various RadialPositions with Roll Diameter

Develop a Systems for Measuring Fabric Stiffness on-Line

The three principal fabric mechanical properties of interest to this research are tensile,bending, and compression. Intuitively, on-line measurement of tensile and compressionalproperties pose lesser of a challenge than measurement of bending behavior. So, it was decidedto explore the possibilities of measuring bending behavior and demonstrate the principlesthrough theoretical analysis.

- Measurement or estimation of bending behavior of textile fabrics under static conditionscan be carried out by using pure bending testers (Kawabata’s) or by ex amining shapes of loops(Peirce’s ‘Heart” or “Pear”) or beams (FAST: Cantilever). None of these methods can be usedreadily under dynamic conditions. However, types of loops can be generated and theirinstantaneous shapes can be examined under dynamic conditions. These shapes are determinedprimarily by the bending rigidity of the fabric in addition to few other known parameters. Therelationship between the loop-shape and these parameters can be derived theoretically. So, inprinciple, once the loop-shape is determined the theoretical relationship can be used todetermine/estimate the bending rigidity of the fabric. Figure 6 shows the shape of fabric loopsbeing examined for this research. A theoretical model has been developed which relates theloop-shape to fabric weight per unit length, bending rigidity, linear speed of the fabric, etc.Solution of this theoretical model requires solution of boundary value problems with movingboundaries and they are quite complex. The process of obtaining theoretical solutions forvarious fabrics under typical conditions of speed has started. A prototype apparatus is being

s92c3,7

National Textile Center Annual Report: September30,1993 183

developed to validate the theoretical model and demonstrate the viability of the measurementsystem.

Figure 6. Shape of Loop Being Analyzed for Bending Rigidity Measurement

Develop a system to Monitor Fabric Characteristics and Automatically Control HighSpeed Sewing Machines.

Analysis of current sewing techniques indicated that higher quality apparel assemblyrequired research into the fundamental aspects of sewing control. From this analysis, the twocourses of research to be followed were 1) the determination of methods for characterizing thedynamics of high speed sewing machine and fabric interactions, and 2) the determination ofmethods of adaptively controlling the sewing system based on the on-line evaluation of thefabric/machine interactions.

The first course of study, the on-line characterization of fabric and machine interactions,focused on the development of a modeling technique to describe the presser foot forcedynamics during high speed sewing. Fast Fourier Transforms provide the frequencyinformation of force oscillations, and settling time analysis provides transient decay rateinformation. A model of the natural system response was generated by imposing the decayenvelope and initial conditions on the force oscillations.

The characterization technique discussed in the fast study produced models which,although sufficiently accurate, lacked specific application. It was desired that a relationshipcould be found between the dynamics of the feeding system and particular fabric properties. Toaccomplish this, a more general modeling technique was developed Again, Fourier Transformswere used to provide frequency content of sewing transients, however this second methodIooked at the decay of each frequency. Using this method, much more revealing characteristicscan be extracted from the force transients. This characterization method reveals the dominantsystem eigenvalues that describe the natural system response of the presser foot and fabricinteractions.

Another experiment was attempted to identify fabric properties based on across-correlation of the presser foot force dynamics without fabric to those presser foot forcedynamics with fabric. This experiment looked at the properties of the resulting cross-correlationwaveform. These properties such as the peak correlation lag, magnitude, and main lobe widthcan provide information concerning the alteration of force dynamics due to fabric properties.

184S92C3,8

National Textile Center Annual Report: September 30,1993

The strongest cotrelation found (p.99) was that of the cross-correlation main lobe width withfabric thickness. This method then, provides a reliable method of predicting fabric thickness

from the presser foot force wavefomCurrent research into the needle bar forces of high speed sewing have generated promising

results. Frequency analysis of the needle bar forces show that excellent correlations (r=.98) canbe found with fabric thickness.(See Fig. 7) Even more promising results arise when examiningthe average power of the needle bar force waveform. A fabric property other than thicknessseems to be evident in the data. As seen in figure 8, the strongest correlation (s.997) is not tofabric thickness but to some other fabric dependent property. This property is currentlyundetermined but is believed related to fabric density and yarn friction.

The second course of research pursued was the determination of methods of adaptivelycontrolling the sewing system based on the on-line evaluation of the fabric/machineinteractions. The fundamental knowledge gained from these investigations has led to a fullreport on a presser foot force controller and patent disclosure to the University. This controlmethod stabilizes the force applied to a fabric during feeding. By maintaining an optimalapplied force, fabric may be less likely to pucker during sewing. A prototype to demonstrate thefundamental operation of this device is currently being designed and built.

Figure 7. Surface Plot of Power Density ofNeedle Force Waveform as a Functionof Stock Thickness (660 spm).

q

T2,! O.lS-

‘..5 rabdc

e 0 rabric0.10 - /=

_ -’ - ,it

: /E 0 , --- rit

B 9’/af 0.0s -

: 0

/ 0

/@

0.00

0.0 0 . 1 1 . 0 l.s 2.0

Figure 8. Average Needle Force Signal Powerper Stitch for 1.2, and 3 Plys of Fabrics1 and 4 (660 spm)

s92c3,9

National Textile Center Annual Report: Septcmbtr 30,19!#3185

RESOURCE MANAGEMENT

The management structure of the project has been established for several years. Thisstructure is based on the establishment and pursuit of common goals and objectives by a teamof interdisciplinary researchers while maintaining a sense of individual ownership andresponsibility to meet those goals. The management team is coordinated by Dr. T. Clapp. Hecoordinates reporting, communications, and resource allocation for the project.

The project is divided into five parallel tasks to insure an organized execution of the workand a logical progression of knowledge. Each CO-PI, Task Leader, is responsible for managingand conducting one of the tasks. Other team members collaborate with the task leader toprovide the necessary interdisciplinary, resources required to complete stated goals. Theprimary means of communication is E-mail. Meetings are called as necessary to discuss overallproject managements and to disseminate results of specific tasks.

The bulk of the work is conducted by graduate students. The education of these students isconsidered to be of primary importance to meeting NTC goals. A list of students participatingin this project shows the diversity of backgrounds and interests.

Post-Doctorate Research Associates: Dr. M. Gunner(EE), Dr. F. SunGraduate Students: G. Barrett(EE), S. Deng(ME), Y. G. Kim(ME), D. Kong(TAM),H. Li(TAM), T. McDevitt(ME), N.Timble(FPS)Under graduate Students: W. Crowe(EE), G. Duncan( M. Spense(TE)

COLLABORATIONAs the project has matured, our team has sought to develop additional research partners

and transfer technology to industry. A list of these collaborative discussions is given below:

1.

2.

3.

4.

5.

6.

7.

8.

9.

10.

186

Eischen, J. W., provided technical papers to Unilever Research Port Sunlight Laboratoryin England.Eischen, J. W., provided technical papers to Mechanical Engineering Department atCarnegie Mellon University in Pittsburgh, PA.Eischen, J. W., provided on-camera remarks for University of North Carolina Center forPublic Television program “Search”, documentary on the textile industry.Gupta, B. S., N.C.State University, and El-Mogahzy, Y. E., Auburn University, collabo-rated in research and writing a paper.Gunner, M., Clapp, T., have held collaboration talks with Dr. L. Dorrity, Dr. H. Olson, Dr.Jayaramen, and Dr. M. L. Realff all at Georgia Institute of Technology.Gunner, M., held collaboration talks with Dr. S. Smith from the University of South-WestLouisiana.Clapp, T., Gunner, M., held talks with Dr. D. Boyd of PNL and Vorhees, M., of (TC)2resulting in a proposal: “Sensors for the Apparel 1ndustry”as part of the AMTEX initiative.Clapp, T., Gunner, M., held collaboration talks with Dr. P. Taylor of the University ofHull, England.Clapp, T., Gunner, M., discussed possible collaboration with Jet Sew Technology, NewYork.Gunner, M., Clapp, T., have written 5 proposals for research with Russell Corporation.

s92c3.10National Tedle Center Annual Report: September 30,1993