Random winding is widely used for the preparation of yarn packages used in a
..... The following tapers are widely used for winding cones: 3°30', 4°20', 5°57', ...
CROSS WINDING OF YARN PACKAGES
Submitted in accordancewith the requirementsfor the degreeof Doctor of Philosophy by
GUNGÖR DURUR
being an account of work carried out under the supervision of
Dr. M. P.U. Bandara
School of Textile Industries The University of Leeds Leeds, LS2 9JT
July 2000
The candidateconfirms that the work submitted is his own and that appropriate credit hasbeen given where reference hasbeen made to the work of others.
Dedicated to: My wife, My son, Ihsan Umay and
My daughter, Ceren
111
ACKNOWLEDGEMENTS
I would like to express my sincere gratitude to my supervisor, Dr. M. P.U. Bandara for his continuous interest, invaluable supervision, thoughtful suggestions and help
throughoutthecourseof this work. I am also thankful to Mr John K. Clarke for his interest, technical assistanceand helpful discussion during the work.
Thanks also to Mr. Harbhajan Rait, Mr. John Peckham, and Mr Azim Abadi for their technical assistance.
My sincere gratitude to the Ministry of National Education of The Republic of Turkey for awarding me a scholarship to make this study possible.
My thanks to my friend, Mr Ian Lenton Roberts for his support.
Finally I would like to express my love and heartfelt gratefulness to my family, my wife, Cemile, my son, Ihsan Umay and my daughter, Ceren for their patience, continual
encouragement and moralsupport.
iv
ABSTRACT
Random winding is widely used for the preparation of yarn packages used in a variety of textile processes. Ribboning caused by overwinding of yarn turns at certain places within the package is a basic problem encountered in this process. Standard random winders incorporate some means of ribbon breaking, which strictly speaking have limited effectiveness. The research reported in thesis was undertaken to realise a random winder capable of detecting the occurrence of ribboning, and taking the ribbon breaking action at these precise times, and thereby achieve 'active' ribbon breaking. Such a winder was realised by the addition of suitable transducers to a standard random winder, which could be controlled by a PC. Winding trials carried out using this equipment established the effectiveness of the concept and also the greater freedom from ribboning the method achieved in comparison with other available methods. Unwinding trials carried out by suitably modifying the winder for measuring yarn unwinding tension were also used to establish the effectiveness of active ribbon breaking. Preliminary experiments carried out on the above apparatus showed the possibility of constructing a Angle of Double Traverse (ADT) diagram on the VDU of the PC during winding operations, which could serve as a useful aid to follow the progress of the winding operation. It was very useful for showing the occurrence of major and minor ribboning in the wound package, and also for visually indicating the effectiveness of the ribbon breaking procedures. The diagram was of further use in studying the nature of package driving by the grooved drum. The case of driving a deformable body such as a yarn package on a comparatively non-deformable grooved drum is important for understanding how the package rotates without any obvious slipping. By varying the winding conditions, it was in fact shown that some measurable slip occurs, and these relationships were experimentally established.
Computer simulation is useful to determine the stresseswithin a wound package, particularly as thesecannot be measuredusing conventional techniques.The results of such a simulation carried out on a randomwound packageand the comparisonof the resultswith thoseobtainedfor a precisionwound packagearealso presented.
V
CONTENTS Page Title Page Dedication
ii
Acknowledgements
iii
Abstract
iv
Contents
v
List of Figures
xiv
List of Tables
xxi
CHAPTERONE Part I WINDING FUNDAMENTALS
1
1.1 Introduction
1
1.2Typesof woundpackages
1
1.2.1 Types of cross wound package
1.3 Basiccrosswinding terminology
2
3
1.3.1 Yarn orientation in cross winding
3
1.3.2 Wind ratio and Traverse ratio
4
1.3.3Traverselength
4
1.3.4 Gain
4
1.3.5Ribboning(Patterning)
5
1.3.6Yarn speed
6
1.3.7 Yarn tension
7
1.3.8Pressure
7
1.3.9Commonfaults in packages
8
1.3.9.1Hard edges
8
1.3.9.2Cob-webbing(crossthreading)
8
1.3.9.3Bulging
8
1.3.9.4Cauliflowereffect
8
1.3.9.5Sloughingoff
8
1.4Typesof winding
9
V1
1.4.1 Precisonwinding
9
1.4.2 Random winding
10
1.4.3 Comparison of the two packagewinding methods
12
1.4.4 Modern techniques of ribbon free winding
13
1.4.4.1 Steppedprecision winding
13
1.4.4.2 Ribbon free random winding (RFR)
15
1.5 The wind angle at the reversal point of cylindrical package
17
1.6 Random winding machine
22
1.6.1 Conventional winding machine
22
1.6.2 Modern developments in random winding machines
22
1.7 Packageperformance analyzer
Part II LITERATURE
REVIEW AND AIMS OF THE RESEARCH
27
28
1.8 Anti-ribboning techniques
28
1.9 Deformation of yarn packages
29
1.10 Slippage between the drum and the package
38
1.11 Aims of the presentresearch
39
CHAPTER TWO
41
INSTRUMENTATION OF THE RANDOM WINDER
41
2.1 Introduction
41
2.2 The random winding machine
41
2.2.1Winding unit
41
2.2.2Tensionassembly
43
2.3 Descriptionof the apparatus 2.3.1 Yarn compression (thickness) tester 2.3.1.1 Description of the yarn thickness tester
43 43 44
2.3.1.1.1 Non-contact displacement measuring system
44
2.3.1.1.2Circuit diagram
45
2.3.1.1.3Powersupply
46
2.3.1.2Calibrationof the yarn thicknesstester
46
vii
2.3.2 Yarn tension measurement
49
2.3.2.1 Yarn tension probe
49
2.3.2.2 Calibration of the yarn tension probe
50
2.3.3 Shaft encoder
52
2.3.3.1 Absolute shaft encoder
52
2.3.3.2 Incremental shaft encoder
53
2.3.4 Measurementof packagediameter and compression
54
2.3.4.1 Linear variable differential transformer
54
2.3.4.2 Potentiometer
55
2.3.5 Interface card
58
2.3.6Calibrationof theLVDT andPOT
59
2.3.7 Frequencyinverter
64
2.3.8 Stroboscope
65
2.3.9 Oscilloscope
66
2.3.10 Tachometer
66
2.3.11 Electronic circuits
67
2.3.11.1 Frequencyto voltage converter for checking yarn
67
packagespeed 2.3.11.2 Amplifier card 2.3.11.2.1 Zero adjustmentand level shifting for POT
68 68
and LVDT 2.3.11.2.2 Counter for incrementalshaft encoder
69
2.3.11.2.3Voltageregulatorfor POT 70 2.3.11.2.4Steppermotordrive cardfor lifting the package71 2.3.12Brushlessservomotorandcontrollercard
71
2.3.13Low voltagepowersupply
72
2.3.14 Actuators
72
2.3.14.1 Stepper motor drives
72
2.3.14.2 Safety switches
72
2.4 PC hardware
74
2.5 Software
74
viii
2.6 Noise on the signals and other problems
74
CHAPTER THREE
76
MEASUREMENT
OF PACKAGE DENSITY, SLIPPAGE
AND DEFORMATION
76
3.1 Introduction
76
3.2 Physical properties of yarn
76
3.2.1 Elastic modulus of yarn
78
3.2.1.1Results
79
3.2.2 Traverse modulus of yarn
80
3.2.2.1Measurementof yarn thickness
80
3.2.2.2Results
82
3.3 Preparing yarn packages
87
3.3.1Pressure
88
3.3.2 Yarn winding tension
88
3.3.3 Oscillation of drum
89
3.3.4 Packagediameter
89
3.3.5 Speedof winding drum
89
3.4 Density of random wound packages
90
3.4.1 The method of finding the density of random wound packages
90
3.4.2The resultsof the densityof therandomwound packages
92
3.4.3Discussionof density
95 95
3.5 Slippage 3.5.1Introduction
95
3.5.2The methodof measurement of slippage
95
3.5.3Provisionof thedrageffect
96
3.5.3.1.The pulley
96
3.5.3.2The calculationof weight hung
97
3.5.4Resultsof slippage 3.5.4.1Resultsof slippagewith yarn drageffect 3.5.5 Comparison of the two experiments
98 10: 10:
ix
3.5.6 Discussion of slippage 3.6 Deformation of a wound package at the contact point
105 105
3.6.1 Introduction
105
3.6.2 The method of measuring packagedeformation
105
3.6.3 Resultsof deformation measurement
106
3.7 Discussion
107
CHAPTER FOUR
109
RIBBONING ON THE WINDER
109
4.1 Introduction
109
4.2 Definition of ribboning
109
4.3 Analysisof ribboningon the randomwinder
110
4.3.1Configurationof experimentalwinder
110
4.3.2The methodof following yarn on the drum andthe package
111
4.3.2.1 The program for reading shaft encoder on the drum
111
4.3.2.2 Experiments to observe ribbon formation in winding
111
4.3.3 Graphic representation of progressof winding
112
4.3.4 Packageformation
116
4.3.5 Observing ribboning with stroboscope
119
119
4.4 Analysisof randomwoundpackage 4.4.1The numberof doubletraversesin a layer 4.4.2The angledifferencesaspackagediameterincreases
119
4.5 Analysis of random wound package under various winding conditions
121
with ADT diagram 4.5.1Variation of level of drag on thepackage
119
121
4.5.1.1 Without drum oscillation
122
4.5.1.2With oscillation
124
4.5.1.3Evaluationof the drageffect
126
4.5.2 Variation of packagearm pressure
127
4.5.2.1Without oscillation
127
4.5.2.2With oscillation
130
X
4.5.2.3 Evaluation of pressureeffect on the ADT diagram
132
4.6 Discussion
132
CHAPTER FIVE
134
ACTIVE RIBBON BREAKING
134
5.1 Introduction
134
5.2 Standard Ribbon Breaking Methods
134
5.2.1 The system for ribbon breaking originally provided
134
5.2.2Effect of normalribbon breaking
135
5.3 The method of interrupting motor speed
136
5.3.1Systemsetup
136
5.3.2Choiceof speeds
137
5.3.3 Results
142
5.4 Methodof ribbon breakingby lifting the package
142
5.4.1 System set up
142
5.4.2 Running and programming
146
5.4.3Results
150
5.5 Comparison of ribbon breaking methods
150
CHAPTER SIX
153
THE UNWINDING
PROPERTIES OF RANDOM WOUND PACKAGES
153
6.1 Introduction
153
6.2 Unwinding Performanceof the packages
153
6.2.1 Apparatus for yarn unwinding trials
154
6.2.2 Tension measurementand recording of data
155
6.2.2.1 Calibration of tension measurement
156
6.2.3Preparationof packages
157
6.2.4Testing
158
6.2.5Test results
159
6.3 Discussion
162
xi
CHAPTERSEVEN
163
SIMULATION
163
7.1 Introduction
163
7.2 The approach to theoretical solution of the random wound cheese
165
7.3 The equation of compression of the random wound cheese
165
7.3.1 Number of double traversesas a function of package radius
167
7.3.2 The number of yarn in the element
168
7.3.3 The addition of a layer at R
170
7.4 Valuesof q,z,Z,Q
174
7.5 Integration of the differential equation
175
7.5.1Boundarycondition
175
7.5.2Procedurefor solvingthe equation
176
7.5.3 Interpolation of the valuedu/dr at the core radius
176
7.5.4Euler's modified methodfor integratingtheequation
177
7.5.5Error in thevaluesof R andU at r
177
7.6 Theoretical results
178
7.6.1 Values of the variables
178
7.6.2 Forcesthroughthe facesof the addedelement
178
7.6.3 Changesdue to addition of a layer at outer radius
179
7.6.4 Results for a completed cheese
182
7.7 Discussion
190
CHAPTER EIGHT
192
SUMMARY, CONCLUSION AND FUTURE WORK
192
8.1 Summary and conclusion
192
8.1.1ADT Diagram
194
8.1.2 Slippage
194
8.1.3 Studiesin ribbon breaking
195
8.1.4Simulation
196
8.2 Future work
197
xii
REFERENCES
199
APPENDICES
202
Appendix A Instrumentation
202
A. 1 Absolute shaft encoder
202
A. 2 Incremental shaft encoder
203
A. 3 Technicalspecificationsof frequencyinverter
204
A. 4 Drum speedgivenby computerasdecimal
205
A.5 Steppermotor
205
A. 5.1 Technicalspecifications
205
A. 5.2 Wiring detailsof steppermotor andthe control card
206
Appendix B Experimental results
209
B. 1 All ResultsOf ExperimentFor Density
209
B.2 All ResultsOf ExperimentFor SlippageAnd Deformation
210
Appendix C Software Programs
211
C.1 Checkingthe diameterof thepackageduringwinding
211
C.2 Drum speedchecking
211
C.3 Countingthedrum andpackagerevolution
212
C.4 For packagedeformation
213
C.5 Eliminatinga falsereadingattheabsolute shaftencoder
214
C.5.1Thewayof creatingtwo loops
214
C.5.2 The way of comparingfour reading
214
C.6 ForADT diagram
215
C.7 The basicprogramfor disturbingthe drum speed
216
C.8 All programsfor disturbingthe drum speed C.9 A programfor lifting the package
216
C.10 For packageperformanceanalyser
221
218
X111
C.10.1Readvaluefrom tensionprobe
221
C.10.2Analysisof PPA data
221
Appendix D Computer simulation of random wound packages
223
D. 1 Computer program for solving the differential equation
223
D. 1.1 Notation
223
D. 1.2 The program structure
224
D. 2 Flowchart of simulation program
225
D.3 Computerprogram
227
Appendix E Publications
232
E. 1 "The measurementof yarn thickness as applicable to cross wound packages" P. Bandara,G. Durur Proceedingsof the 3rdInternationalConferenceon Novelties in Weaving Researchand Techology,September,23-24,1999, Mauritor, Slovenia, pages74-82
232
E. 2 "Improving the effectiveness of theribbon breakingin randomwinding" P. Bandara,G. Durur Submittedto `Mechatronics',ElsevierSciencesLtd.
242
E.3 "The use of the Angle of Double Traverse diagram to study package formation in random winding and the effectiveness of different techniquesof ribbon breaking" P. Bandara,G. Durur PaperID: M2000-156 International Conference`Mechatronics2000', 6-8- September2000,
Atlanta,Georgia,USA ( Acceptedfor presentationandpublicationin proceedings)
253
xiv
LIST OF FIGURES Page Fig 1.1
(a)Cheese(b)cone (c)constant-tapercones (d)accelerated-tapercones 2
Fig 1.2 Fig 1.3
Terminologyapplicableto yarn lay Successivecoils of yarn
5
Fig 1.4
Speedrelations in winding
6
Fig 1.5
Layout of the precisionwinder
9
Fig 1.6
The variationof wind anglewith packagediameterfor
Fig 1.7
precisionwinding Layoutof the randomwinder
Fig 1.8
Thevariationof wind ratio with packagediameterfor 12
Fig 1.9
randomwinding DigiconeWinder
Fig 1.10
Principle of the Digiconer
14
3
10 10
14
Fig 1.11 RFR Winder
16
Fig 1.12 Principleof RFR
16
Fig 1.13 Forcesat theyarn reversal(the sideview of the wound package) Fig 1.14 Forcesat theyarn reversal(top view of thewound package)
18
Fig 1.15 Reversalgeometryandstrokeat the package Fig 1.16 Conventionalrandomwinder
20
Fig 1.17 Autoconer338- SchlafhorstAutotenseyarn tensionregulator 25 Fig 1.18 SSM- schematicfunction of the Preciflex
19 23
26
Fig 1.19 SchlafhorstAutoconer338 with ATT direct drive (ATT :Auto TorqueTransmission)
26
Fig 1.20 PackagePerformanceAnalyser(PPA)
27
Fig 2.1
Configurationof the winding unit
42
Fig 2.2
The tensionassembly
43
Fig 2.3
Configurationof the tester
45
Fig 2.4
Circuit diagramof the yarn thicknesstester
46
Fig 2.5.
Powersupplyfor the yarn thicknesstester
47
xv
Fig 2.6
The calibration curve of tester
48
Fig 2.7
Wheatstonebridge arrangement of thetensionprobe
50
Fig 2.8.
Set-upto calibratethesingleyarn tensionmeasurement device
50
51 Calibrationof yarn tensionmeasurement device 51 Fig 2.10 TheOp-Amp for smoothsignalfor yarn tensionmeasurement Fig 2.9.
Fig 2.11
Schematic diagram of the LVDT
Fig 2.12 The methodof measuringdeformedwound packageradius
54
55
The method of measuring un-deformed wound package radius
55
Fig 2.14 L.V. D.T excitationandsummingjunction circuit diagram(1) Fig 2.15 L.V. D.T. excitationandsummingjunction circuit diagram(II)
56
Fig 2.16a ElectronicandMechanicalconnectionsof devices
60
Fig 2.16b A view of winding machine Fig 2.17 Computerinterfacecircuit diagram
61
Fig 2.18 The original tubewith a metalsleeve
63
Fig 2.19a Calibrationgraphof LVDT
64
Fig 2.19b Calibration graph of POT
64
Fig 2.13
57
62
Fig 2.20
Control connections for operation as a speedcontroller
65
Fig 2.21
The connection of FN converter
68
Fig 2.22
The schematic diagram of amplifier for POT and LVDT
69
Fig 2.23
Conversion of signal from the incremental shaft encoder
to 8 bit binary code
70
Fig 2.24 Thevoltageregulator Fig 2.25 Basic configurationfor the lifting the package
70
Fig 2.26 Low voltagepowersupply Fig 3.1 Variation of thedensityof cheesewith radiusin precisionwinding
73
71
77
Fig 3.2
InstronLoad-Extensioncurve
79
Fig 3.3
Variation of z and y with load
86
Fig 3.4
Variation of E with load
87
Fig 3.5
Sideview of a package
91
Fig 3.6
The densitydistributionin termsof yarn tensionandpressure
93
xvi
Fig 3.7
Thedensitydistributionat thepressure2 kg
94
Fig 3.8
The densitydistributionat the pressure1.25kg
94
Fig 3.9
The density distribution at the pressure0.7 kg
94
Fig 3.10 a) thedimensionof thepulley b) thepulley on the machine
96
Fig 3.11 The dragdueto cord with weight
97
Fig 3.12 Theslippageat pressure2 kg Fig 3.13 Theslippageat pressure1.25kg
101
Fig 3.14 The slippageat pressure0.7 kg Fig 3.15 Deformationof packageat pressure2 kg
101
Fig 3.16 Deformationof packageat pressure1.25kg
108
Fig 3.17 Deformationof packageat pressure0.7 kg
108
Fig 4.1
Configurationof thewinder
Fig 4.2
Flowchartfor eliminatingfalse values
101 108
110
a) nestedloopsb)comparingfour readings
113
Fig 4.3
ADT diagram
114
Fig 4.4
Flowchartof the programof ADT diagram
115
Fig 4.5
Traverseratio 3.5 ( 3.60- 3.48)- no ribbon breaking
117
Fig 4.6
Traverseratio 3.3 ( 3.41-3.31)-no ribbon breaking
117
Fig 4.7
118
Fig 4.8
Traverseratio 3.0 ( 3.06-2.97)- no ribbon breaking Traverseratio 2.75 ( 2.80-2.73)- no ribbon breaking
Fig 4.9
The numberof doubletraversein 1 mm thicknessof diameter againsttraverseratio.
118
120
Fig 4.10 The angledifferencesbetweeneachdoubletraverse againsttraverseratio
120
drag Fig 4.11 Rotationof constantdiameter packageatincreased ( diameter:80 mm, traverseratio:3.4, no-weight)
122
Fig 4.12 Rotationof constantdiameterpackageat increaseddrag
( diameter:80 mm, traverseratio:3.4,weight50 g) Fig 4.13 Rotationof constantdiameterpackageat increaseddrag ( diameter:80 mm, traverseratio:3.4, weight 100g)
123 123
xvii
Fig 4.14
Rotation of constant diameter package at increaseddrag
( diameter:80 mm, traverseratio:3.4, weight 150g)
124
Fig 4.15 Rotationof constantdiameterpackageat increaseddrag with drum oscillation(diameter:80 mm, traverseratio:3.4, no-weight, oscillation is on)
124
Fig 4.16 Rotationof constantdiameterpackageat increaseddrag with drum oscillation(diameter:80 mm, traverseratio:3.4, weight 50 g, oscillation is on)
125
Fig 4.17 Rotationof constantdiameterpackageat increaseddrag with drum oscillation (diameter :80 mm, traverse ratio: 3.4,
weight 100g, oscillationis on) Fig 4.18 Rotationof constantdiameterpackageat increaseddrag with drum oscillation(diameter:80 mm, traverseratio:3.4, weight 150g, oscillationis on)
125
126
Fig 4.19 Rotationof constantdiameterpackageat increasedpressure ( diameter:80 mm, traverseratio:3.4, scale1)
127
Fig 4.20 Rotationof constantdiameterpackageat increasedpressure ( diameter:80 mm, traverseratio:3.4,scale2)
128
Fig 4.21 Rotationof constantdiameterpackageat increasedpressure ( diameter:80 mm, traverseratio:3.4,scale3)
128
Fig 4.22 Rotationof constantdiameterpackageat increasedpressure ( diameter:80 mm, traverseratio:3.4,scale4)
129
Fig 4.23 Rotationof constantdiameterpackageat increasedpressure ( diameter:80 mm, traverseratio:3.4,scale5)
129
Fig 4.24 Rotationof constantdiameterpackageat increasedpressure with drum oscillation(diameter:80 mm, traverseratio:3.4, scale1, oscillationis on)
130
Fig 4.25 Rotationof constantdiameterpackageat increasedpressure with drum oscillation(diameter:80 mm, traverseratio:3.4,
scale2, oscillationis on)
130
xviii
Fig 4.26 Rotationof constantdiameterpackageat increasedpressure with drum oscillation(diameter:80 mm, traverseratio:3.4, scale3, oscillation is on)
131
Fig 4.27 Rotationof constantdiameterpackageat increasedpressure with drum oscillation (diameter :80 mm, traverse ratio: 3.4, scale 4, oscillation is on)
131
Fig 4.28 Rotationof constantdiameterpackageat increasedpressure with drum oscillation (diameter :80 mm, traverse ratio: 3.4,
scale5, oscillationis on)
132
Fig 5.1
Original ribbon break mechanism
135
Fig 5.2
The effect of original ribbon breakingmethodat wind ratio 2
136
The speed-time graphfor motorspeedinterruption Fig 5.4a Flowchartof theprogramfor interruptingmotor speed
137
Fig 5.4b Flowchartof theprogramfor interruptingmotor speed
140
Fig 5.5
138
Fig 5.6
Checkingof packageanddrum speedsundermotor speedcontrol Outputof packagespeedandwinding drum speed
Fig 5.7
Effect of disturbingmotor speedat wind ratio 2
142
Fig 5.3
139
141
Fig 5.8a Arrangementfor lifting thepackage
143
Fig 5.8b Packagelifting arrangement
144
Fig 5.9
145
Connectionof switch
Fig 5.10 Positiontime graphfor packagelifting
146
Fig 5.1la Flowchartof the program Fig 5.1lb Flowchartof the program
148 149
Fig 5.12 ADT diagramof 750 doubletraversesaroundwind ratio of 2
151
Fig 5.13 ADT diagramof 1500doubletraverses aroundwind ratioof 2
151
Fig 6.1
Configurationof the unwindingapparatus
155
Fig 6.2
Calibrationgraphof tensionmeasurement
157
Fig 6.3
A periodiccourseof yarn tensionsignal Fig 6.4a Thebasicresultsof analysisof storedtensiondataobtained
with the3 differentanti-ribboning methods
159
161
xix
Fig 6.4b The resultsof analysisof storedtensiondataobtained 161
Fig 7.1
with the3 different anti-ribboningmethods An elementof thecrosswound package
Fig 7.2
Layer of yarn 1 mm thick
167
Fig 7.3
The measurednumber of double traverses as a function of area
168
Fig 7.4
Forceson theyarn
170
Fig 7.5
Pressureand circumferential forces at the element
172
Fig 7.6
Pressureat the outer radius
175
Fig 7.7
Forcesin axial andcircumferentialdirection at r=R
178
Fig 7.8
u- the incremental radial deformation
179
Fig 7.9
z- the changein axial componentof theforce
180
Fig 7.10 q- the changein circumferentialcomponentof the force Fig 7.11 t- thechangein yarn tension
180
Fig 7.12 p- the changein radial pressure Fig 7.13 Pressuredistributionin a completedrandomwound package
166
181 181 182
Fig 7.14 Total radial distancemovedby the element in a randomwoundpackage
183
Fig 7.15 Yarn tensiondistribution in a completedrandomwound package
183
Circumferentialcomponentof theforce distribution in a completedrandomwound package
184
Fig 7.16 Fig 7.17
Axial componentof the forcedistribution in randomwound package 184
Fig 7.18 Forcesin axial andcircumferentialdirectionat r=R in a precisionwound package
185
in a precisionwoundpackage 185 Fig 7.19 u- theincremental radialdeformation Fig 7.20 z- thechangein axial componentof the force in a precisionwound package
186
Fig 7.21 q- the changein circumferentialcomponentof the force in a precisionwound package
186
Fig 7.22 t- the changein yarn tensionin a precisionwoundpackage
187
in a precisionwoundpackage Fig 7.23 p- thechangein radialpressure
187
xx
Fig 7.24 Pressure distributionin acompletedprecisionpackage Fig 7.25
Total radial distancemoved by the element
in a completedprecisionpackage Fig 7.26 Yarn tensiondistributionin a completedprecisionpackage Fig 7.27
188 189
Circumferential component of the force distribution
in a completedprecisionpackage Fig 7.28
188
189
Axial component of the force distribution
in a completedprecisionpackage
190
xxi
LIST OF TABLES Page Table 2.1
Specifications of the probe
45
Table2.2
Addressesusedby the interfacecard
59
Table 2.3
Calibration data of LVDT and POT read from computer
62
Table 2.4.
Technical specifications of the tachometer
66
Table3.1
Yarn specifications
77
Table 3.2
Thickness measurements
81
Table3.3
85
Table3.4
Resultsof analysis Densityof randomwound packages
92
Table3.5
Theresultsof theslippagetest
99
Table3.6
The resultsof slippagetestwith hungweight The resultsof comparingtwo testsof slippage
102
107
Table4.1
Deformationof packages Relationof decimalof traverseandnumberof curves
Table 6.1
Calibration of tension measurement
156
Table3.7 Table3.8
102 116
1
CHAPTER ONE
PART 1 WINDING FUNDAMENTALS 1.1Introduction Winding is the processof transferringa yarn from one packageto form anotherpackage
thatis moresuitablefor subsequent processing of the yarn.Themostimportantaspects of winding are the type of package,the methodof winding andthe type of yarn.
1.2Types of wound package There are a vast number of different types of wound packagesavailable to meet the requirementsof knitting, weaving, dyeing, texturing and the production of sewing threads,tapes,string, etc. There are two main types of package:flanged packagesand flangelesspackages.The advantageof the flangelesstype is that it allows over-end withdrawal of yam during unwinding in the next process.In unwinding a flanged package,the whole package must be rotated.In textile industries,the flangelesspackageis preferredcommonly and is more widely used. The wound packageis also divided into three categoriesin terms of how the yarn is laid on the package.
-
theparallelwoundpackage
-
thenear-parallel woundpackage thecross-wound package
On the first oneof these,a multitudeof yarnsarelaid parallelto oneanotherso that to thepackageaxis.The useof fnges on thebobbin eachyam is woundperpendicular for thepackageto be stable.On the sated category,oneor more or beamis necessary yarnsarelaid nearlyparallelto oneanotherowing to the slow traverseusedto fill the foil length of the flangedbobbinused.The third category,the-cross'wound
2
usually consists of a single thread which is laid on the package at an appreciable angle so that layers of yarn cross one another giving stability. The near-parallel wound package and the cross-wound package require a traversing mechanism on the winding machine incorporating a reciprocating action actuated by a guide or cam for each spindle, or a grooved drum.
1.2.1Types of crosswound package There are two basic types of cross-woundpackage:those that have the surfaceof the packagetaperedto its axis, and those that have the surface parallel to the axis. The taperedpackageis called a cone,whereasthe packagewith surfaceparallel to its axis is a known a tube, or cheese.The size andshapeof a cheeseis specifiedwhen its diameter and traverselength are known, althoughto specify a cone requiresthreeparameters;the basediameter,the taper,and the traverselength (Fig 1.1).
a- roper diameter
be$*
ofchoue
amew
i
nose
_die
a
ä
-..
length
averse length
(a)
(b)
(c)
(d)
cones(d) Accelerated-taper cones Fig 1.1(a)Cheese(b) Cone(c) Constant-taper Yam can be withdrawnfrom this kind of packageby pulling it over one end of the package.If theyarnis withdrawnat a low speed,yarnwill dragon thepackagesurface, and the withdrawaltensionis largely influencedby friction, particularlyas one,,cpil movesover others,which arenot yet beingunwound.If the yarn withdrawalspepc{gis
3
sufficiently high, the yarn forms a rotating balloon' which rotatesaboutthe axis of the package.The unwinding tension is higher due to the centrifugal effect of balloon rotation, andundergoessomefluctuationas theunwinding point movesbetweenthe two edgesof the package.When the packageis a cone,withdrawal is from the noseside of it, and unwinding tensionis less fluctuating due to reduceddrag of the yarn over the surface of package.Evennessof unwinding tension is so important that cones are in generalmorepopular than cheesesfor most applications. The following tapersare widely usedfor winding cones: 3°30', 4°20', 5°57', 9°15'. It should be noted here that constant-taperconesare straight-endedbut accelerated-taper coneshaveconvex(base)andconcave(nose)ends(Fig 1.1). 1.3 Basic cross-winding terminology Certain standardterms apply to crosswinding, which should be defined so asto be able to explain the winding processclearly. 1.3.1Yarn orientation in crossWinding
Theseare Thereareseveralanglesformedby thelay of yarnsin a crosswoundpackage. identified in Fig 1.2.
Fig 1.2Terminology applicableto yarn lay: wind angle (A), crossingangle(B), reversal angle(C) andcoil angle(D)
The windangleis definedasthe angle(A) betweenthedirectionsoft#te.yarnlay on the packagesurfaceand any plane perpendicularto the packageaxis. B is called the
4
crossingangle, and C is called the angle of reversal. D is the coil angle betweenthe direction of the yarn on the packageand the direction of the traverselength [Textlnst 95]. In random winding (see section 1.4.2), the wind angle is independentof the diameterbecauseit is producedby surfacedrive to the packageusing the grooveddrum. On the other hand,the wind angle in precisionwinding (seesection 1.4.1) dependson the packagediameter.In this casethe wind angle decreases continuouslyasthe package builds up. 1.3.2Wind ratio and traverse Ratio The wind ratio is the numberof revolutions madeby the packagewhile the yarn guide
makesa singletraversefrom oneend of the packageto the other.(i.e. the numberof coils laid on the package) The traverse ratio is the numberof coils laid on the packageduring a doubletraverseof the yarn guide. It is twice the wind ratio. It is also referred to as Wind per Double Traverse.This parameteris frequentlyindicatedasw.
This is oneof theimportantspecifications In randomwinding,the of a woundpackage. wind ratio decreaseswhile the packagediameter increaseswhereasit has a constant value in precisionwinding for all stagesof winding. 1.3.3Traverse length Traverse length or traverse is defined as the distancebetween extreme positions of a reciprocatingthread-guidein one cycle of its movement.In both precision and random winding, traverselength is basically constant.This is true for themajority of cheeseand cone packages.A pineapplepackage uses a traverse which gradually reducesas the packageis built up. 1.3.4Gain Gain is the angulardisplacementof the yarn at the beginning of a double traverse,with respectto the correspondingposition for the previousdoubletraverse.it is important for the yarn to be displacedat the endof a pattern repeat,so that the next turn wt11not causO.
5
over winding. The new yarn coil should fall slightly to one side or theother of the first coil. The gain to be usedis a minimum gain if the coils are to be laid sideby side so as to increasewinding density. The minimum gain is dependenton the diameter of the yarn andthe diameterof thebobbin used.
Fig 1.3 Successivecoils of yarn
Fig 1.3showsthepackagesurfacewith two coilsof yarnlying sideby sideat theendof the package.The displacementof two successivecoils is x, and this gives rise the definition of gain as the angle z at the centre of the cross-sectionof the package.The use of the term is mainly confinedto precisionwinding, asdiscussedlater.
1.3.5Ribboning(Patterning) If yarn is repeatedlylaid on top of or along the samepath asthe previously wound yam, this duplication of yam path on the packagecreatesa defect known as ribboning or patterning. This condition occurswhenthe traverseratio becomesan integervalue. It is evident that someribboning also takesplacefor valuesof traverseratios that differ from an integerby 0.5,0.25 etc.
In practiceit is observedthat ribboningis mostpersistentat small valuesof wind per double traversesuch as 3,2 or 1. At thesevalues,asthe packageis relatively large, the rate of changeof packagediameteris relatively small, and henceribboning persistsfor longer. Such ribboning action is "normally referred to as major ribboning. Stagesof ribboning that persist over a relatively small number of package rotations may be termedminor ribboning or occasionallypatterning.
6
1.3.6 Yarn speed A basic winding machine rotates the yarn package about a horizontal axis. Yarn being added is supplied in a direction that is perpendicular to the package axis. Some form of a yarn guide is used, which guides the yarn to form helical coils on the package surface. The yarn winding speed V, results from the vector-sum of the surface speed Vs and the traverse speed VT. D is the coil angle as in Fig. 1.4.
VT D
VS
U
Fig 1.4Speedrelationsin winding "
Winding Speed(V. )
by thefollowingequation. Windingspeedcanbe calculated
`w
"
=
VzT + ui
s
(1.1)
Surface Speed(Vs)
Surfacespeedof theyarnpackagecanbegivenby Vs=n*DP*Np
(12)
7
Where Dp- Packagediameter
NP-Packagerotational speed(rpm)
"
Traverse Speed(VT)
Traversespeedcan be calculatedby theequation VT=2.L.Nd/k
(1.3)
Where
L- Traverselength Nd-Rotationalspeedof the driving roller or cam drum (rpm) k-driving roller revolutionsper doubletraverse Note: Nd/k- double traversesper minute 1.3.7Yarn tension Tension is appliedto a yarn during winding for removing weak placeson the yarn and building a stable packageof the required density. Yarn tensionis introducedby means of a disc tensioneror a gatetensioner[Oxtoby 87]. As a rule of thumb,winding tension should be about 10 - 15 % of the yarn mean-breakingload. With natural fibres this meansthat weak placeswill be removedduring winding, but for strong syntheticsand blends this doesnot necessarilyfollow.
1.3.8Pressure In precisionwinding,pressureis appliedat the winding point using a smoothroller, which is either drivenor which is free to rotatewith the package.The level of such pressureapplied is a significant factor affecting the density of the packagewound. A satisfactorypackagecan be wound over a rangeof pressuresettings dependingon the required densityof the package.For instance,a soft packagesuitablefor dyeing requires a lower pressurethan for a hard packageswhich is to contain more yam in a given volume. In randomwinding the grooveddrum servesthis function.
8
1.3.9Common faults in packages There are certain typesof commonfaults in wound packages.A major fault applicable to random winding is ribboning or patterning,but this definition of fault has already beengiven in the previoussection.The other possiblethe faults are the following. 1.3.9.1Hard edges More yarn is found to be laid at the reversalpoints of yarn at the edgesof a crosswound package than is theoretically desirable. This occurs due to the slowing down and eventualreversalof the yarn at eachend of the package.This leadsto someunavoidable differencesin packagedensityalong the traverselength. 1.3.9.2Cob-webbing (cross-threading) Yarn coils may slip at reversalsof traverseif the yarn is not sufficiently tensionedto form a stablepackageor if the traverseguide is at the wrong distancefrom the surface of the package.The fault involves yarn at the end of the packagetaking a short-cutfrom one point on the circumferenceto a point further round, insteadof following the curved circumferential path. Such a fault can cause tension peaks or end-breakagesduring unwinding.
1.3.9.3Bulging The soft inner layers may be squeezedout at the ends of the packagedue to different hardnessof layersfrom the inside to the outsideof a package.
1.3.9.4Cauliflower effect The fault identifiesthe distortionproducedon the yarn at the intermediate layersof a package,brought about by the pressureexerted on them by the outer layers, if the tensionin theselayersdropsto zero, this leadsthemto buckle [Wegener69].
1.3.9.5Sloughing-off Too low a yarn tensionor drum pressuremay permit the yam to fall off the edgeof the packageduring unwinding causingthe fault known as sloughing off. This is more of a problem in the caseof continuousfilament yarns.
9
1.4Types of winding There are two main methodsfor producingcrosswound packages;randomwinding and precision winding. Thesehave different specificationsand havecertain advantagesand disadvantages.There are also some relatively recent winding techniquesthat combine the best features of precision winding and random winding, such as Step Precision Winding (Digicone) and Ribbon FreeRandomWinding (RFR). The two basic winding processesaredescribedbelow. 1.4.1Precision winding The layout of a precision winder is given by Fig. 1.5. The packageis mounted on a spindle which is positively driven. The traversing of the yarn is provided by a yarn guide which is driven by cylindrical cam, which is coupled to the packagespindle at a constantratio. The packagespindlein precisionwinding normally hasa constantspeed, which causesthe surface speedof the packageto gradually increase,as the package becomeslarger.
package M `ý-;
-; -------------
4-I--
*M-motor Fig 1.5Layoutof theprecisionwinder The packagerotatesa certainnumberof timesduringeachdoubletraverseof the yarn guide. Hencethetraverseratio (w) for precisionwindingremainsthesame.Howeverit from Al to A2 as decreases canbe seenfrom Fig 1.6thatthewind angleprogressively duringwinding. thediameterof packageincreases
lU
wind angle Al
d2
A2
packagediameter
Fig 1.6The variation of wind anglewith packagediameterfor precisionwinding The main advantageof precision winding is that, the gain in this type of winding is a constant which when adjustedto a suitable value, enablesthe production of a crosswound packagewith no ribboning. Integer valuesof wind ratio are of courseavoided. 1.4.2Random winding The other main method of producing cross wound packagesis random winding. In randomwinding, the packagespindle is not positively driven. Instead,the packageis driven by surfacefriction from the winding drum, which incorporatesan endlessgroove to provide traverseto the yarn. package
--------------
package M MR
ýC X
.......
ýf " ý1C" JI -.
drum "M- motor
_. _. _
(a)
Fig 1.7Layout of the randomwinder
cam (b)
11
The simplicity of the abovearrangementis that a cam type traversearrangementis not required. However considerablerubbing is causedon the yarn by the groove, so this type of winder is only suitablefor spunyarns. The constructionof Fig. 1.7b is sometimesemployed.Here the packageis driven by a smoothcylinder which is rotatedat a constantspeed.A separatecamdrive is coupledto the driving drum asshown.This mechanismbehavesin an identical mannerto the basic randomwinder shownby Fig. 1.7a. One further advantageof the grooved drum is that the yarn being wound, due to the winding tension,falls readily into the groove,without the needfor any threadingaction to be carriedout. This is a distinct advantagein such operationsas yarn clearing where defective portions of yarn have to be cut off, the yarn rejoined, and the winder set in motion again.The freedomfrom having to repeatedlyrethreadthe yarn throughthe yarn guide hasthereforeresultedin the useof thegrooved drum winder in theconstructionof automaticwinders. As thepackageis driven by frictional contactwith drum, the surfacespeedsof drum and packageare equal.As the drive speedis constant,this enablesa constantwinding speed to be maintainedduring the whole period of winding a package. The wind angle of the yam on the packageremainsthe same at all points, as this is decided by the resultant velocity of yarn on the surface of the package,which is constant.This also contributesto a uniform packagedensity. On the other hand, the traverse ratio decreasesfrom the core to full diameter of the package as seen in Fig 1.8. This has a great significance for the random winding process. As the traverseratio (which is twice the wind ratio) gradually diminishes,it passesthrough a seriesof integervalues.How long the numberlingers on at an integer dependson the rate of change of packagediameter as winding continues.,Obviously, this rateis higher at small packagediameters.
12
traverse ratio wl
WZ
package diameter
Fig 1.8The variation of wind ratio with packagediameterfor randomwinding
The significanceof the traverseratio attainingintegervaluesis thatat theseintervals, successiveyarn coils are coincident,giving rise to overwinding of yarn turns known as ribboning or patterning as identified before. This effect will be explained in greater detail in Chapter5.
Ribboning or patterning affects the package unwinding characteristics and the consistencyof dyeing in subsequentmanufacturingprocesses.Avoidanceof ribboning is of utmost importance in random winding. Thus, all random winding machines incorporatesomemethodof ribbon avoidance.
1.4.3Comparisonof the two packagewinding methods
Precisionwinding Patternzones Packagedensity
+
Densityof wind
Freeof atternzones densityis notconstant + Package
Decreasesfrom core to outer radius
Wind angle +
Package geometry + Unwindin property + Technically
Randomwinding
+
constant
Lowe2!!!e
Highpackagedensity Preciseyarndisplacement God unwindingproperties Complicated system
Anti-patterningis necessary Uniformpackage ensty
+
Irregularyarnlayingfrom layerto layer Problemo(patternzom. Simplesystem
13
1.4.4Modern techniquesof ribbon free winding As indicatedin the previoussectioneachwinding methodhassomeadvantagesand disadvantages. With the introductionof microprocessorbasedcontrol, superiortypesof winding machineshavebeenrealisedby combiningthe betterfeaturesof the two basic Thesemethodsare known asSteppedPrecisionWinding (Digicone) winding processes. and RibbonFreeRandomwinding (RFR). These are described below.
1.4.41 Steppedprecision winding The digicone is the namegiven to a type of precisionwound packagewhich is produced by a processtermed steppedprecision winding. The main positive characteristicof randomwinding is the constantwind angle and that of precision winding is constant winding ratio which prevents pattern formation. These positive characteristicsof randomand precisionwinding are combinedin the digiconeprocess(steppedprecision winding) which achievesa patternfree packagewith a near-constantwinding angle. Fig. 1.9 showsthe arrangementusedin a Digiconer. The drum is driven at a constant speedby the motor so that a constantwinding speedis achieved.The motor also drives the variable ratio drive V which is used to vary the speedof the traversecam. At the beginningof winding a,package,the traverseratio is high and set at a value (w1) which gives the requiredamountof gain. As winding begins the wind angle has the value Al correspondingto curve 1. As winding proceeds,the traverseratio is held at this value but the wind angle graduallyincreasesto the value A2 correspondingto the curve 2. At this point, which is determinedusing the shaft encodersplaced on the two spindlesas shown, the ratio of the coupling is changedso that the wind angle is again set at the lower value Al. This results in a lower value of traverse ratio (w2), which avoids winding at an integervalue of traverseratio, and still ensuresthe samegain as required. Winding at this traverse ratio continues,as the wind angle becomesexcessiveagain. The above procedureis repeatedtill the packageis wound to the required diameter. Thus package consists of several intermeshedconcentric layers, each wound with a different traverseratio but with very little variation in wind angle.
14
Package _. _. _. _. _. _
M wv Drum
V- variable ratio drive
Fig 1.9 DigiconeWinder
wl I
o w2 Ew3 w4
9 2
packagediameter(d) Fig 1.10Principle of the Digiconer
15
1.4.4.2Ribbon Free Random winding (RFR)
This methodis rather similar to the basic randomwinding method.The only difference is in changingthe wind ratio, soas to avoid winding at critical valuesof traverseratio. The RFR winder is very similar in constructionto that usedin the digiconer as shown by Fig 1.11.The differencein operationis achievedby the mannerin which the control unit varies the variable ratio drive (V) to achieve two traversecam speeds,nDT1and nDT2.The ratio of packagespeed(npac)to one of the double traversespeedsupplies a constant wind angle as in normal random winding. As a result the yarn package producedhas two wind angles.The winding is switchedbetween them so as to avoid winding at integervalues of traverseratio. As shown by Fig 1.12, winding is started with wind angle A2. As the packagediameterincreases,the traverseratio reachesthe critical value w2. At this stage,the speedof yarn guide is instantaneouslychangedby the control unit and the traverseratio drops to w3, correspondingto wind angle Al at this diameter.The packageis built up at thesamewind angleuntil the new critical point (w4) is reached. The control unit incorporating microprocessortechnologychecksthe speedof packageand yarn guide at all times, andundersoftware control, determinesthe points of change. As in the Digiconer a packagewith intermeshedlayers is produced,but in this casethe layers havethe two alternatingvaluesof wind angle. The developmentof the microprocessorcontrol has meant that the above design is
furthersimplifiedby dispensingwith the mechanical variableratio drive altogether,as the desired speedscan be obtained now directly through frequency controlled motor drives.
16
Package
C C
8 --------------
DTI MV
nDT2
Drum V- variable ratio drive
Fig 1.11 RFRWinder
w1 w2 .. 3
w5
0
w3
aý
w4 A2
.r
packagediameter (d) Fig 1.12 Principle of RFR
17
1.5The wind angle at the reversal point of cylindrical package The problem of hard edges on cross wound packages is caused by the inability of the traversing mechanism to sharply reverse the direction of traverse at the edges of the package. As such the change of direction of the yarn involves a lower speed of traverse, and hence more yarn is deposited at the reversal points than is ideally required. It appears that if a cam were available which is capable of very fast reversal of traverse, the problem will be solved.
However the problem also dependson the behaviourof yarn at a point of reversalof traverse,asit is the tensionof the yarn that holds it on the packagesurfaceand keepsit from slipping from the position it is depositedat by the yarn guide. At a point of reversalof traverse,the tensionof the yarn tends to straightenthe yarn, which in other words meansthat the yarn slips back some distancebehind the yarn guide as the latter changesdirection. So it is evident that, even if a perfecttraversereversalis achievable, the stability of the yarn at the reversalpoint will anyway lead to the formation if a curvaturein the yarn at the point of reversal.
The following sectionconsidersthis on the basisof yarn tensionandfrictional forces betweenthe yarn and thepackagesurfaceat a point of reversalof traverse. The calculations involved are basedon the equilibrium of forces acting on the yarn curve at the reversalpoint [Hebberling 691.Two distributed loadsact on the yarn at a reversalpoint on the packagesurface 9 Normal load dueto the friction betweenthe yarn and thepackagesurface
"
Tangentialload asa component of theyarntensionat theturn
As the wind angleis increased,the radiusof the turn becomessmaller.The wind angle hasan uppervalue at which equilibrium is no longer possible.The last wind angle at which equilibrium still exist is called the maximumwind anglea..
The corresponding
radiusof yarn curvatureat the reversalpoint is pd.. The following assumptionsare madefor simplifying thecalculation. " The packageis cylindrical (cheese).
18
"
The yarn tensionis assumedto be constantalong the reversalcurve andyarn path is assumedto be circular at the reversal.
The yarn reversalis in equilibrium until the radiusof curvaturep reachesits minimum valueat the point of reversal.In this conditionthe product of normal tensionload P and coefficient of fibre friction .t must equalthe tangentialload Q. (1.4)
µ. P=Q
The geometricrelations are shown in Fig 1.13 and Fig 1.14.All notationsusedin this calculationare given below. a-
wind angle
degree
amax T-
maximumwind angle
degree
yarn tension
g
P-
distributednormalload
g/cm
Q-
distributedtangentialload
g/cm
1-
lengthof the yarn
cm
9-
radial angle
degree
peripheralangle
degree
p-
radiusof curvature
cm
µ-
coefficient of friction
-
E-
strokeratio
-
D-
diameterof the package
cm
cp
-
dl
-al--/
dei T
de/ä
T
D
Fig 1.13Forcesat theyarnreversal(thesideview of thewoundpackage)
9/z T
19
Normal tensionload P can be calculatedconsideringthe reversalarc with referenceto Fig 1.13. 2.T.sin (dA/2)= P.dl
(1.5)
dl= (D/2)
(i)
.de
sin(dO/2)=dO/2
(ii)
P= [2.T.sin (dA/2)]/ dl
(iii)
Substituting(i) and (ii) in Equation(iii)
(1.6)
P=2.T/D
T
dl
ýcý
dß2
--
-"---------
T
-D-
dl
d(plVl T
Fig 1.14 - Forcesat the yarn reversal(top view of the wound package)
Thetangentialload0 canbe calculated usingthegeometricrelationsof Fig 1.14. 2.T.sin (d(p/2)=Q.dl
(1.7)
dl=p. dcp
(iv)
sin(d(p/2)=dcp/2
(v)
Q=[2.T.sin(d(p/2)]/d1
(vi)
20
Substituting(iv) and (v) in Equation(vi); (1.8)
Q=T/p
As mentionedbefore,the reversalis in equilibrium and p is a minimum, so
P=Pmin
SubstitutingEquation (1.6) and Equation(1.8) in Equation(1.4), µ [2.T/D]=T/pmin
(vii)
Pmin=D/(2. µ) = R/µ
(1.9)
Under the assumedconditions, the minimum reversal radius dependson the radius of packageand the fibre friction as seenby Equation 1.9 whereasyarn tension doesnot affect theminimum reversalradius.
T
11a.
SI
Fig 1.15`-Reversalgeometryandstrokeat thepackage
21
In thefollowing solution,yarn reversalis assumedto be a circular arc. Referringto Fig 1.15,cosa canbe written as (1.10)
cosa=PPQ
The strokeratio is a measureof the reductionof the traversestroke.It can be defined as;
SI $o
where
So- theoreticallength of package S1- actuallengthof package $p Q-
-S,
22E)
SubstitutingEquation(1.11) in Equation(1.10) andsolving for p,
(1 ýcosa)(1 21 EE)
(1.12)
Substituting Equation (1.9) in Equation (1.12) and solving for µ when a=a..
and
P=Pmin"
D
(1-cosam, x
S, cosa.
aox = cos'
1- e
(1.13)
(1.14)
' (1ýE)+1
In conclusion,Equation(1.14)showsthatthemaximumwind:ang is4a;,nctionof } S1,D ande. If thewind anglewereoverthe maximumwindangle,it wouldcauseyarn to slip over packagesurfaceat the reversal.Therefore,the yarn traversestrokeat the
22
packagedecreases.Speedof winding does not count in this calculation. It is assumed thatwinding speedis low. 1.6 The random winding machine
1.6.1Conventional winding machine The basicconstructionof the randomwinder hasnot changedmuch over the years.The path of the yarn on the machineis very much the sameas before. New developments havebeenmadeon individual sections,like different types of yarn guide and improved compensationof yarn packageholder pressureas packagebuilds up. The winding head of a winding machinehas generallythe samearrangementof featuresfor handling the yarn on the way to the package.These are supply spindle, unwinding accelerator, tensionassembly,yarn guide cylinder (with lap guard, traversedisplacementand antipatterning), packagecradle (with packageweight compensationdevice, and package size control), linkage housing (with performancecontrol, and linkage of all stop and start motions) asshownby Fig 1.16.
in randomwinding machines 1.6.2Modern Developments The new developmentsin random winding machine are aimed at improving yarn properties (i.e. by clearing), package quality, unwinding behaviour of package and winding speed. The winding machine has clearly benefited from microprocessor technology.The information aboutthe processtaken from various points of the winding machineis evaluatedby the microprocessor,and then the necessaryactionwill be taken automatically. Yarn tension is maintainedat a constantlevel in order to minimize variation of density at the package.Companiesthat produce winding machinesuse different methodsfor maintaining constant tension. A new winding machine by the Schlafhorst company [Schlafhorst] has a tension sensor connected to the winding head control unit for ensuring uniform yarn tension during process.It is situatedin the yam path after the clearerto registerthe actualyarn tensionat the package(Fig 1.17).The measuredvalues
23
Package cradle
Package
Yarn Guide
0
Clearer -* Knotteror Splicer Tension device -f Pre-clearer
feeler of yarn Guide plates
UnwindingAccelerator
4-- Spindle
Fig 1.16 Conventionalrandomwinder
24
are transmittedto the tensioner,wherethe pressureis increasedor loweredaccordingto the information received. Some winders incorporate double section yarn tensioner [SSM]. It is believedthat the double sectionyarn tensionersuppliesa uniform winding tension,sincetensionis appliedat two points. Today, on most machinesthe yarn guide drum of the individual headsis directly driven by a dedicatedservomotor.The motor runs under the electronic control as seenin Fig 1.17. These bring some advantagesto the modified random winding machine. It is claimed that the combinationof winding headcomputerand direct drum drive facilitates smooth,absolutelyjerk-free start-up, slip-controlled acceleration,high winding speeds and improved anti-patterning.Obviously, the effectivenessof anti-ribboning can be improved due to slip control. As one of the well known anti-ribboning methodsis the fluctuation of the drive to causeslippage between package and drum, electronically controlled direct drum drive can achievesuch slippageundergreatercontrol. In one type of precision winding machine producedby SSM, the packageis directly connectedto a drivemotor, and a servomotoras shown by Fig 1.18 which drives the yarn guide.They are connectedto a control unit which collects all information about the process. At the beginning of winding, the information about speedand wind ratio is collectedby the control unit. Fig 1.19 shows the direct drive systemof the drum of the SchlafhorstAutoconer 338. Eachwinding headis equippedwith the servomotorfor the drum drive in this machine. The drum is mounted directly on the shaft of the drive motor, so that the motor transmitsthe torquedirectly.
25
Yarn package
ý/ý ý/Z
Controlled
/\
Servomotor
Speed control
Winding Head computer
Yarn tension sensor
ý III Tension Control
Yarn tensioner
Fig 1.17 Autoconer338 - SchlafhorstAutotenseyarn tensionregulator
26
Positron 'DWI M rdlio. SDBBC
-_ýli
Info-wo,,
i'wuWon
Co.
c
I
i
Dos,DOnfpm
nmoef r Nnpfr of moans
Fig 1.18 SSM- schematic function of the Preciflex
Yarn guide drum Servo Motor
Winding head computer
Fig 1.19 Schlafhorst Autoconer 338 with ATT direct drive (ATT
Transmission)
:Auto Torque
27
1.7 Package Performance Analyser PPA (Package Performance Analyzer), as shown by Fig 1.20 is an instrument for the assessmentof package unwinding performance for a range of yarn package types, including many industrial and spun yarn packages [Rieter-Scagg]. The assessmentis an analysis of tension measurements made whilst yarn unwinds from a package. The output signal from a purpose designed fast response transducer is sampled at 1000Hz. The tension data is processed and stored in the form of a cumulative tension distribution. The stored data is analysed after completion of a test run and the package performance factor calculated by consideration of the peak tension with respect to general characteristics of the distribution.
The required operating speed and test length is entered at the keypad with sample identification data. A package is accommodated on a creel while balloon length and guide wrap angles can be set to simulate the unwinding conditions of subsequent processes. The yarn is threaded throughout the instrument into waste suction and test started. The results can be printed as PPF (Package Performance Factor), mean and peak tensions with the probability distribution of 100 equal subsections of the test.
Fig 1 20 PackagePerformance Analyser (PPA) .
28
CHAPTER ONE
PART II
LITERATURE REVIEW AND AIMS OF THE RESEARCH
1.8 Anti-ribboning techniques As mentioned briefly in Part I, the random winding process is prone to producing ribboning on the wound package.The traverseratio attainedat any particular stageof winding determinesthe spacingbetweensuccessivedouble traversesof yarn. When the wind ratio becomesan integer,yarn coils producedover a numberof double traverses coincide, and causethe defect known as major ribboning. Similar overwinding occurs when the wind ratio is 0.5 ,0.33,0.25 etc, but the amountof ribboning is less intense (minor ribboning). Commercial random winding machinescombat ribboning by creating abrupt slippage between the package and the winding drum. This helps to disperse the yarns and minimise ribboning. Schlafhorstwinding systems[Schlafhorst]for randomwinding usesthe ribbon breaking techniqueof switching on and off the drive motor. This createsslippagebetween the packageand the drum. They have indicated that ribbon breaking methodsprovided on their conventional winders produce comparativelysmall differencesof speed,but new models of random winder, Autocore 238 or 338, produce much bigger speed differences.The length of the anti-ribboning cycle dependson the packagediameter.It is smaller at small packagediameters.As packagediameterincreases,the length of the cycle is increased.However,the action runs at regular intervalsduring winding and can hencebe called `passive'. Other winding machine manufacturingcompaniesalso use similar methodsof ribbon breaking.
29
Kyoto [Kyoto,89] reportedanothermethodfor winding packagescalled superwinding The method is basedon using a new type of grooveddrum called the `superdrum'. The superdrum hasone groovein one direction andtwo groovesin the other direction.Yarn can be selectively guided into one or the other of these grooves as necessary.It was claimed that the methodprovides packagesof evendensity and favourableshapewhich are the advantagesof the conventionalrandom winding. During winding, yarn is bypassedto the additional groove,when approachingthe packagediameterat which major ribboning is expected.A yarn-path switching device is used on the machine for changingthe numberof windings. Packagesproducedby this methodwere compared with those produced by conventional random winding, using unwinding trials. The numberof yarn breaksexperiencedon the packageswound with the new methodwere very few comparedwith thoseexperiencedwith conventionallywound packages.It was claimed that the packagesproducedby the new method do not have any ribboning. However there is no evidence of the industry having adopted this new concept of randomwinding.
1.9Deformationof yarn packages A wound yarn packagehas a number,of specifications associatedwith it. The wind angleis of basic importance.While for parallel wound packagesthis angleis practically zero, for crosswound packagesit may be about 160.In the presentwork, cross-wound packageproducedby randomwinding are specifically considered.The random wound packageis investigatedin terms of its behaviourduring the winding operation,and its propertiesafter being wound to full size. The instrumentationof the winding machine facilitatesthe understandhow the packagebehavesduring winding. A yarn packageis an assemblyof a yarn on a suitable tube or bobbin. A cross wound packageis consideredin this context, and hencethere are no flanges, which constrain the yarn being wound in any way.
The yarn is associatedwith elasticpropertiesin its length direction as well as its transversedirection.Howeverthe elasticmoduli in thesetwo direction arediffe ieft dueto the structureof theyarn.The 'tensile'modulusin the axialdirectionalis usu y direction. modulusin thetransverse ` greaterthanthe'compressional'
30 Although a crosswound packageis built up by adding a single yarn with traverse,it is convenientto think of 'layers'of yarn which are addedup to form thefinal package.The layer conceptis easierto adopt in the caseof a precisionpackage.This can be the taken asthe set of yarn turns involved in the numberof double traverseswhich is numerically equalto the reciprocalof thegain used.The packagecan be consideredthe collection of suchlayersrequiredto completethepackageto its full size. This conceptis still useful for the considerationof a randomwound package,but there is the difficulty that spacesbetween successivedouble traverses of yarn are not constant.This is on accountof the constantlydiminishing wind ratio. However thewind ratio only changesby a small amount over a small increment of diameter of the package.So if the packageis divided into a numberof annuli, eachof a small thickness, it is possibleto assumethe traverseratio to havea constant(mean)value corresponding to a given annulus.An effective ribbon breaking techniquewill help maintain some reasonablyevenspacingbetweenthe turns, so in practice,it seemsuseful if the number of double traversesof yarn required to cover each such annulus from the core to the surfaceof the packagecouldbe determined. In winding a package,yarn is suppliedto the winder at a constanttension.It is desirable to have a constantresidual tension in yarn inside the package.This is on account of viscoelasticityof textile yarns, as the tensionof the yarn within the packageinfluences how the yarn relaxesonce it is unwoundfrom the package.Unevenrelaxation of yams may causesuchproblemsasvariation of checkwidth in a woven fabric. A yarn packagecan be consideredas having beenformed by adding layers of yarn on top of the previously wound layers. Each addedlayer will imposepressureon the layer beneathit. The layer that is subjectedto this pressureis compressedradially towardsthe centreof the core and movesto a position of smaller circumference.Thus, eachlayer of yarn deforms radially. Also the movementof the layer due to the pressurecausesthe yarn in that layer to contract(i.e. undergoa reduction of radiuscomparedto its valueas it is wound) and henceits tensionto drop. So as eachnew layer is;.. addedat: the otter radius of the package,the pressureinside the packagewill increase:ýHowever :ems layer from core to outer radius will not be under the samepressure,:saw number-
31 layers existing above and below it are different, dependingon its radius within the package.The amountof radial movementtowards the core will be different for every layer and so will the amountof reduction of yarn tensionin every layer. It is possible that a layer at a certainradius inside the packagemay lose all its tensionand will begin to be under compression.The layer next to the core of the packagewill not undergo radial movementand not lose its tension,as the core is generallyincompressible.These layerscan only lose someof their tensiondue to Poisson'seffect. However for a cross wound packagethe Poisson'seffect is likely to be much smaller than for a parallel wound package. This basic concept of deformation has been used by several researchersfor cheese packageswhich are precisionwound.A randomwound packageis more complexthan a
because it involvesmorevariables. parallelwoundor precisionwoundpackage, Catlow and Walls [Catlow 62] have derived equations for determining the stress distribution in a pinn formed by parallel winding. In the solution, the pinn is considered as having layers of yam of infinitesimal thickness.It is assumedto be isotropic and homogenous.Another assumptionis that it is parallelwinding so that winding tensionis acting perpendicularto the axis of the package.A layer at the outer radius is considered to impose additional pressure to the layers below. The changes in radial and circumferentialstresscausedby this areobtainedassecondorder differential equations. The necessaryboundary conditions are that the core is incompressibleso the radial deformationof the pim at the core is zero and the pressureimposedby the addedlayer at the radiusof pirn is known from the yarn tensionand the thicknessof that layer. The equationcan be integratedanalytically to give the expressionsof incremental,radial and circumferentialstressesat any radiusof the pinnfor eachlayer addedat the outside.The total changesat any radius for final outer radius of the pirn are obtainedby integrating contributionsof all the layersadded.
The valuesof the ratio of residualcircumferentialstressto the winding stressat the different pirn radii for differentsize of the pirn andfor differentvaluesof Poisson's ratiowereobtainedasoutlinedabove.If the Poisson'sratio is zero,theminimumvalue occursin the mid-radiusand the residualtensionat the core is equalto the winding tension.By increasingthe Poisson'sratio, theminimumvalueis found to moveclose
32 to the core radius. Finally, the minimum value occurs at the core radius, when Poissons'sratio increasesup to 0.5. Beddoe [Beddoe 67] used the same approach to establish a theory of winding anisotropicelastic yarn on a thick-flanged tube in order to predict pressureson the tube and its flanges. He found that the modulus ratio, which he has defined as the ratio of Young's Modulus in thecircumferentialdirection to that for radial and axial direction in the wound yarn, could be as high as 20. He calculatedand constructedgraphs with severaldifferent modulusratios in the range1-20. One of the graphsshowsthevariation of yarn radial pressureagainstto radius, the other is variation of yarn tension against radius.The pressureimposedby the wound yarn in the isotropic case(modulusratio=1) is twice as much as when the modulus ratio is 20. Additionally, the maximum pressure at the tube to the minimum pressureat the outsideof the completedbeamin the caseof high modulus ratio (20) is also twice asmuch as in the isotropic case.He found that the tension in the yarn, or distribution of circumferential stress,changesas the modulus ratio is increased.In the isotropic case,yarn tensionincreaseslinearly from a minimum value at the tube radius to the maximum value at the outer radius of the beam. At a modulusratio of 20 (non-isotropic),the value of the ratio of the residualcircumferential to the winding stressvaries from 0.12 to 1. The minimum value (0.12) occurs at the radiusof 1.3. The value of the ratio at the tube radiusis about 0.75. The Poisson'sratio chosenfor thesecalculation is 0.3. The yarn loses its tension at the tube radius due to Poisson's effect if the tube cannot be deformed radially. In the isotropic case, the pressureat the tube is highest, therefore the loss of yam tension at the tube is also highest.
Nakashimaet al [Nakashima68] used a similar approachto Beddoe to find the stress distribution inside a flanged beam. They assumedthat the yarn layer is a anisotropic elasticbody. The modulusratio is defined as the ratio of Young's modulusfor the layer in the circumferential direction to that in radial direction. They calculatedforceson the core and flangesof beam,tension distribution in the yarn packageand the distribution of radial stressin the beam.Theseresultswere similar to thoseobtainedby Beddoe.
33 Jhalani [Jhalani 70] carried out a computer simulation by combining the approachof Catlow and Walls with the geometryof the precisionwound package.There are certain clear differences between a cross wound package and a parallel wound pinn. In a parallel wound packagethe axial componentof the tension in the yarn is fairly small and can be consideredto be zero. The circumferential stressof the addedlayer at the outer radiusis constantregardlessof the radius.In the crosswound packagethe tension in the yarn hastwo components,the circumferentialandthe axial due to the wind angle. In his study, he assumedthat the cheeseconsistedof yarn layers of small thickness. Each layer is supportedby a similar layer beneathit. Interlacing of the two layers is ignored and the contactbetweentwo layers is establishedat a multitude crossingpoints madeby the threadsof the two layersthus forming a cylindrical trellis. Two endsof the cheeseare different from the centralpart of cheesedue to the direction of threadat the two ends.Hencehis theoreticalsolution was aimedat the centralpart of the cheese. He consideredthe yarn packageas anisotropicand nonhomogenous.Poisson'sratio is assumednegligible becausethe yarn is not close-packedas in parallelwinding. Yarn is treatedas elastic. The tensionin the yarn is the only force acting on the yarn and acts along a tangentto the cheese.He madethe solution in two stages.On the first stage,he assumedthe modulus of compressionof the cheeseE and the elasticity of yarn in extensionEY as constant,but in secondstage,he treatedthem as variable particularly when the interyarnpressureis small. A secondorder differential equationwas developedto give the radial deformationof the cheesedue to the pressureimposed by the added layer at the outer radius of the package.The equation was numerically integrated to give radial deformation at any radius of the package from core to outer radius. The boundary conditions for the solution are that the compressionat the core which is assumedto be incompressibleis zero and the pressureimposedby the addedlayer at the outer radiusis known. As a new layer is added to the package,a new solution is made on the computer using the equation. The total compressionof the cheeseat any radius is equal to the combined compressionof thecheeseat that radiusdue to the addition of all the layers.
34 He plotted graphsof the dependentvariablespressurewithin thepackage,tensionin the yarn layers and the movement of layers for each of the independentvariables considered.The variables used are winding tension, spacing at the adjacentwraps of threads,traverseper wind, the modulusof the compressionE and the elasticity of yarn EY. The value of modulus ratio which is the elasticity of yarn in extensionEY to the modulusof compressionof the cheeseE is a measureof the anisotropyof the yarn.The modulusratio hasa considerableeffect on the behaviourof the crosswound package.A high valueof the ratio brings aboutsmallervaluesof the pressure,greaterchangesin the tension of the yarn, and the yarn may acquire 'negativetension'. Packagesproduced from low modulusratio yarn showeda continuousincreasein pressurewith the increase in outer radiusof the packageand the value at the core is considerablyhigher than with the high modulustype. The tensionwould not reduceso much asin the former type and is not likely to acquirenegativevalues. He also found that the residual tension in the yarn, like pressure,dependson the modulus ratio of the yarn. With a high modulus ratio of yarn, the residualtensionin the yarn falls sharplywith radius near the core and quickly becomesnegativeshowing the yarn in compression.After the initial fall it changeslittle with radius till it rises again near the outsideof the package.With a low modulus ratio of yam it falls graduallywith radiustill it startsrising from near mid radius of the cheese.Changesin the tensionare comparativelysmallerand the yarn tensionis not likely to becomenegative. In the secondstageof simulation, the value of compressionmodulus and of elasticity modulus in extensionare not constantespeciallywhen the loads are small, though at high loads the variation in them may be small. He took into accountthe variation of them in his modified theoretical solution. First he determined the relation between modulus of compression and pressurein the cheeseand between the modulus of elasticity and the tension in the yarn. The second order equation was integrated numerically using a different computerprogramwhich was written in KDF 9 Algol.
His resultsin this stageweredividedinto threesections,eachcontainingthreesolutions E is assumed for threevaluesof thewindingtension.First,themodulusof compression to be constantandthe modulusof elasticityin extensionEY is variablewith winding
35 tension. The value of the modulus ratio at a given inner radius reduces as outer radius increasesfrom the inner radius and reachesits minimum value as the tension in the yarn continues to fall. Then the modulus ratio at inner radius increases for some increases in outer radius due to the yarn acquiring negative tension as the compression of the cheese at inner radius continues with outer radius. Finally the modulus ratio becomes constant as further increase in outer radius does not effect the cheese at inner radius.
Next, a solution was obtained as the modulus of compressionand the modulus of elasticity were assumedvariable with pressureand tension respectively.He admitted that the program ran very slowly due to many trial solutions required. Quadratic interpolation was used insteadof linear interpolation which was used in all previous solutions in the program.Another reasonfor slow progresswas that many trial solution fail due to the valueof themodulusof compressionbecomingvery small or the valueof incrementaldeformationbecomingvery large.He attemptedto improve the solution by reducing the thicknessof added layer, but a different solution to the problem resulted. He acceptedthat there wereerrors,which might be proceduralratherthan mathematical. And also the modulusof compressionwas setto a lower limit in the programin order to avoid the situationwith near-zeroforce acting on near-zeroresistancedue to the lower values of pressureand the compressionmodulus at outer radius. The staring value of modulus ratio was very high and thereforethe effect of the addedlayer was limited to fewer layers immediately beneathit. This preventedthe build up of high values of pressure,the modulus of compressionand radial compression,the tension in the yarn would not becomenegativeat higher radius. Last, he made solution, starting with a low initial value of the compressionmodulus. The starting values of the compressionmodulus were raised and this reduced the starting modulusratio (EY/E) causingthe tensionto persistfurther into the packagebut then to changemore rapidly and becomenegative.Also the values of pressurein the package,the radial compressionand the variation in yarn tensioninside the packageare not proportional to the winding tension in the yam due to the changein modulus ratio with increasedin winding tension.
36 Jhalani commentedthat the theoretical analysis is not applicable to a random wound cheesebecauseof basic differencesin the two types of packages.The differencesare that winding angle,the numberof yarn in a given axial width, thereforethe numberof crossingpoints in the elementwhich doesnot vary with a radiusfor the precisonwound cheese.He addedthat it is not also applicableto a conical package,whetherprecision wound or random wound due to different geometricalshapeand to a parallel wound package.In a parallel package,the contactbetweenthe wrapsof yarn is line contactand changesto a surfacecontactin the deformedstate. Wegenerand Schubert[Wegener,68,69] give expressionsfor determiningthe pressure on the coresfor precision and randomwound cheese.The expressiongive the pressure at the someintermediateradius of the cheese.They assumedthe tensionin the yarn at that radiusis the same.Adding a layer at the outer radius did not changedthe tensionat the particular elementof the layer. But this assumptionwould be possibleif there was no compressionof the cheeseat that radius.To find the pressureat the core radius, the expressionis integratedbetweenthe limits of the core radiusand the final outer radius of the cheese.The expressiongave a rough idea of the pressuresfor calculation of the deformationat thecylindrical package. Ursiny [Ursiny 86] attempteda calculationof the deformation of the packageat a cross wound cylindrical packages produced on rotor open-end spinning machines. The packagecan be consideredidealised as a compact pressurevessel so that the laws of linear elasticity apply and can be characterisedby means of the constant average Young's modulus elasticity and the constant coefficients of lateral contraction (Poisson'seffect). He assumeda constantpackagetension in the yarn to calculate the relative radial tension and the relative tangential tension which are a function of the radiusaccordingto the equationhe derived.
He defineda degreeof anisotropywhich is square,root of the young's modulusof elasticityin thetangentialdirectionto thatin theradialdirection. k=
E` Eý
where Eý- theYoung'smodulusof elasticityin tangentialdireotio
37
CLOTHWORKERS' LIBRARY UNIVERSITY OF LEEDS
E, - the Young's modulusof elasticity in radial direction. The results were plotted and this showed that radial tension increases from zero at the outer radius of package to the maximum at the core. On the other hand, the tangential tension falls at certain intervals from the core down to negative values and increases to the outer radius of the package. The results are similar to those reported by other researchers.
While the above researchershave tried to find a suitable theoretical approachedfor
pressure anddeformation at a cylindricalpackage, someof themhavedesigned methods for measuringradial and axial deformationof a crosswound package. Wegenerand Schubert[Wegener69] insertedaluminium foil strips inside the package at different chosen radii when the packagewas prepared for measuring the radial deformation of a cheese.They measuredthe position of each foil strip and recorded them.This is repeatedduring winding and then during unwinding. Analysing the record gavethe total and permanentdeformationat the cheese.When the radial deformationof the cheeseplotted againstthe radius of the cheese,the maximum total deformation of about 1.4% occursat a radiusof about9 cm. The permanentdeformationof about0.7% at the radiusof 9 cm is more than half of the total deformation. Jhalani [Jhalani 70] has also usedsimilar method but in a different way for measuring the radial deformationsof the package.A resistancestrain gauge wire placed on the circumferenceof the cheeseshortensalong with the circumferenceof the cheeseat a radiuschosenon his experimentas the cheesebuilt up. The changein the length of the wire was estimatedby the changein the resistanceof the wire. A slotted paperwhich hasdifferent crimping behaviourcomparingthe ordinary paperwas wrappedaroundthe central region of the package.The gaugewire was wound on top of it. Two strips of copperfoil were solderedto the endsof the gaugewire asthe leadsto be connectedto the measuring unit. The winding was continued, the resistanceof the gauge was measuredafter 10 mm increasein radiusof the package.The changein resistanceof the gaugewas calculatedand it correspondedto the amountof compressionreceivedat that
taken fob b were,,, particular radius where the gaugewas placed.Measurement, .; the. The that radial unwinding of permanent package. winding and resultsshowed
38 deformationof the cheeseat any radius is nearly half of the total radial deformation of the cheeseat that radius as the cheesewas built up to a given outer radius. Initial winding at that radius gives a quick, large initial compressionof the package.As winding continuedfurther, thereis little increaseof compressionand thereis a tendency for it to stay constant.These results were in agreementwith Wegenerand Schuber's results. They used a different method to find radial deformation, and different yarn materials. Both above investigationsagree that permanentradial deformation at any radiusis approximatelyhalf the total deformationat that radius. Law [Law 80] set up similar methodswith Wegenerand Schubertfor measuringthe radial and axial deformationof the cheeseduring the winding. In his research,several kinds of yarn were usedand their propertiessuch as the modulus ratio were measured. His results from the radial deformation tests for types of yarn were of similar nature. For any radius of the package, initial winding at that radius gave a quick initial compression,this compressionbecamevery little as the winding carried on. The magnitude for radial and axial deformation varied with different types of yarn. The greatestdeformation took place at the region about midway between the core and the outer radius.Theseresultswere of similar shapeas thoseby Jhalani.Finally, he claimed' that the radial,deformation alwaysshows a reduction in radius.The greatestvalue of it occurs at any radius soon after initial winding at that radius, then gradually becoming constant.The axial deformationmostly concentratednearthe packageends.
1.10Slippagebetweenthe drum and the package As mentioned before, a random wound package is driven by a winding drum. The packageis rotated solely by virtue of friction which is presentat the area of contact between the two bodies. It seemsthat some slippage between these two bodies -in contact is inevitable. Very few researchershave reportedstudy of slippagein random winding. Osawa and Koyama [Osawa 721 studied the slippage between grooved drum and forming packageswhile cotton yarns were wound into a cylindrical cheese.-JIM measuredthe slippageon their experimentalrandom winding machine by varying yarn tension,the pressurebetweendrum and packageetc.
39 They admittedthat thereis slip betweenthemand the slip is usually within 5%. The slip is causedby the resistanceagainstthe driving torque of the drum. It is affectedby the yarn tension,the friction at the spindle-bearing,the air resistance,the resistanceof the inertiaof thepackageand thedeforming resistanceof thepackage. Their finding was that yarn tensionduring winding affectsthe slippagewhencomparing the slippagebetweenwith feeding the yarn to the packageand without feeding any yarn. The slippagewas also dependenton the pressurebetweenthe drum and the package.At the very low value of pressure(less than 0.5 kg), slippageis more. They also indicated that the deformationof the package,which is the decreasein the radius of packageat the point of contact,should affect the slippage.Their later work [Osawa76], concerned two types of slippage,sliding slip and elastic slip, that occur simultaneously.Sliding slip is the relative speedwhen two rigid bodies contact each other and their surfaces move at different speed.Elastic slip is the other type of speeddifference that occurs between two bodies moving in contact becauseof their elastic deformation. On the randomwinding, the drum is acceptedasa rigid body but the yarn wound packageis a deformable cylinder of fibrous material. They set up an apparatusand method of experimentfor analysingslippageand resistancetorqueof the package.There is relation ship between resistancetorque and slip. As the resisting torque increases,the slip between the drum and the package increases.Moreover, there is relation between resisting torque (M) of packageand slip rate(n) at different values of RPM of drum. This was plotted on a graph.The inclinations of M-ij lines changeat M=30 g.cm(=M1). Torque M1 is the transient point from elastic to sliding slip They claimed that the packagedeformation at the contactpoint would be maximum at that M1 of resistance torque. When the value of the resistancetorque is below M1, the slippagewould be elastic slip and when the resistancetorque is aboveM1 , the slippagewould be sliding slip.
1.11Aims of the presentresearch The different winding processesare quite importantfor the subsequentprocessesof the textile industry. Better quality of wound packagesare aimed at;-by researchersand
winding machinemanufacturers. Most of the reportedwork hasbeenon the precision wound packageperhapsbecausethe structure of the random wound packageis more
40 complex than that of the precision wound package.As the randomwound packageis usedwidely, thereis a casefor carrying out researchon it. The project was startedwith the intention of applying mechatroniccontrol to the basic randomwinder to realisea systemwhich achieves'active'ribbon breaking as identified in the following chapters.As the apparatusconceived was capableof carrying out considerablemeasurementsconcerningthe winding process,a number of objectives were identified aspart of the study. Aims of the presentwork :
-
analysisof the structureof randomwound packages investigatingtheslippagebetweendrum and package
-
evaluationof ribboning or patterningduring winding
-
active avoidanceof ribboning
-
examiningthe effectivenessof different anti-ribboningprocedures
-
computersimulation of the randomwound package.
-
The structureof the packagerelatesto the mannerin which successiveturnsof yarn are depositedand such physical properties as the resulting density distribution within the package.The rotation of the yarn packageby friction by the grooved drum naturally posesthe questionabout the natureof any slip that exists betweenthe elementsduring the progressof winding of a package.Oncethe experimentalwinder was equippedwith shaft encodersand these were interfaced to the PC, it was realized that the above determinationscould be carried out with considerableeaseunder program control. So the first two objectivesabovearoseout of preliminary work with the apparatus. The next three objectivesmay be consideredof primary interestas thework was mainly directed at achieving a better ribbon breaking method. During this early work it was also clear that the apparatuscould be usedto determinethe numberof double traverses of yam wound at eachlayer (thickness) of thepackage.This information would make it possible to extend the work on the simulation of packagewinding carried out by an earlier worker applicableto precision winding, to the caseof randomwinding. This is how the last objective was identified.
41
CHAPTER TWO
INSTRUMENTATION OF THE RANDOM WINDER 2.1 Introduction To be able to effectively analyse a random wound packages, it is important to be able to wind such packages under controlled conditions. The basic winder as described below has hardly any instrumentation provided on it as standard. This project therefore required the addition of a range of sensors and actuators to configure the winder into an apparatus that was capable of carrying out the planned experiments as described in the following chapters.
This additional instrumentationprovided includes some equipment that are mounted
directlyon thewinder,andsomethatarestandaloneor of an auxiliarynature The instrumentation thatwas devisedis describedin this chapter.Additionaldetailson the electronicsusedare providedin Appendix A. 2.2 The Random winding machine Generalinformation aboutrandomwinding and randomwinding machinesare given in
Chapter1. The windingmachineusedin this work is Autoconertype GKN produced by Schlafhorst.The machinehad beenacquired by the School of Textile Industries in 1967.The machinebasically hasa winding unit anda tensionerassembly.
2.2.1Winding unit The winding unit consistsof a grooved drum, a control housing and a packageholder. The drum is driven from the main shaft of the machine through an intermediateroller and can be displacedlaterally to give soft edgesto the packages.The drum is 90 mm in
diameterandits lengthis 125mm, andthedrumconstant(numberof drumrotationsper doubletraverse)is 3. This machineavoids patterningby changingthe speedratio between the grooved drum and the package at random intervals. Basically, the intermediateroller is pulled away from the drive shaft roller in order to interrupt the
42
drive to the drum every 2 seconds.The drum losesspeedduring this, and the slippage that occursbetweenthe drum and thepackageasthe drive is re-engageddispersesturns of yarn,therebyeffectively diminishing the intensityof any patterning. The packagecradleholds the tube,and keepsit in contactwith the grooved drum under some pressure. Increasingpackageweight is compensatedby a relief system on the packageholder for uniform andconstantpressurethroughoutthewinding process.The pressurebetweenthe packageand the winding drum is oneof the factors that determine the package hardness,the other being the yarn tension. There is a screw for the adjustmentof the pressureused. The winder had been originally equippedto wind cones. For this work, the package cradle was modified to allow the mounting of cylindrical tubes, so as to be able to producecylindrical packages.This modification mainly amountedto the provision of an adapterto enlarge the tail side of the tube holder. The control housing contains the controls for the stop motion, lateral displacement,anti-patterning and performance control. Mostly the systemruns mechanically.
-
Package
Pecked Holder
00 Yarn GuideCylinder
ControlHousing
Fig 2.1 Configurationof thewindingunit
43
2.2.2Tension Assembly The tensionassemblyconsistsof the following partsin the yarn path as shownby Fig. 2.2: threading plate, pre-clearer, gate feeler, tension device, yarn guide plate. The threadingplate assiststhe yarn to be guided through the other elementsof the tension assembly.The pre-clearer strips coarse and loose impurities from the yarn without damagingit. The slot width in the pre-cleareris adjustablefrom 0.5 mm to 1.5mm. The gate feeler feels the presenceof the yarn. The tensioneris important in that it applies sufficient tension to the yarn so as to break weak places of it, and give the desired packagedensity.The tensionis createdby pressingthe tensionshoethat holdsthe yarn againsta tensiondisc under the predeterminedpressureof the spring to suit the yarn being wound. There is a small electric motor that drives the tensiondisc at a speedof roughly 375 rpm to achieveself cleaningof the tensionerduring winding.
yarn guide plate
tension device
r
gate feeler
Yarn direction
pre-dearer threading plate
Fig 2.2 the tensionassembly
2.3Descriptionof the apparatus 2.3.1Yarn compression (thickness) tester
Early in this work, it was intendedthat a computersimulationof yarn'package `mOdulu` formationwould be carriedout, andthis necessitated knowingthetransverse of the yarn,in additionto its longitudinalmodulus.Howeverthetransverse modulusil
44
dependentof the transversepressureacting on the yarn, and hence required the constructionof a suitableapparatusto measureit. A yarn thicknesstester constructedusing a linear eddy current proximity sensorwas usedto measurethe thicknessof layers of yarn so placedon the testeras to provide a numberof intersectingpoints. The resultsof the measurements were usedto obtain the relationshipof load per intersectingpoint vs. deformation from which yarn transverse moduluscanbe determinedover a rangeof loadingvalues. 2.3.1.1Description of the yarn thickness tester The thicknesstesterwas constructedusing an inductive proximity sensor[Distec Instr], which was readily available.The sensorhas an associatedamplifier module giving a voltageoutput which varies directly as the distancebetweenthe face of the sensorand a steel targetplacedbefore it. The sensorusedhas a measuringrangeof 2 mm, and this was consideredadequatefor the purpose. The testerasshownby Fig. 2.3 has the proximity sensormountedon the top surfaceof a die-cast metal box, chosenfor convenienceof use and its sufficient rigidity. To prevent the sensitivity of the probe from being reducedby the metal of the housing itself, the sensorwas placedat the centreof a 25 mm diameterhole madeat the centre of the top of the housing. This necessitatedthe adoption of a loading plate 'target', shapedas shown in the figure. The rectangularbearing surfacesof the plate give two equal areasof 10 cm2 each.The actual area is not of consequencefor yarn thickness measurement,but simply the yam intersectionsconcernedshouldlie within theseareas.
2.3.1.1.1Non-contactdisplacementmeasuringsystem The proximity sensor is a device [Distec Instr.], supplied by Graham&White InstrumentsLtd, consistedof two parts, probe and signal conditioning unit. The probe measuresthe gap between its tip and the surface of the target plate. The signalconditioning unit provides the probe with a constanthigh frequencysigoal. The output could be fed into a readout equipment such as an oscilloscope or a:voltmeter; r'i necessarypower for the,signal-conditioningunit is -18 V. The specificationsOfprow involved, type 375-1, is given in Table 2,1
45
Load Plate
Housing
Proximity sensor
iýI
Side View
Yarn
Spring Plain View
Fig. 2.3 Configuration of the Tester
Specification of probe
Type 375-1
Typical Displacement Range
0.006"
Temperature Range Coil Material
C to +80 C -29 Copper
Probe Tip Diameter
0.300"
Face Material
Fibreglass
Body Material
StainlessSteel
ThreadSize
3/8" U.N.F.
Probe Length
1" standard
Table 2.1 Specifications of the probe 2.3.1.1.2 Circuit Diagram The output of the signal conditioning
unit required further
measurement purposes, so an additional amplifier
amplification
for
was built around the LM348
operational amplifier. This IC consists of four independent, low power operational amplifiers which are similar to the familiar 741 operational amplifier. The IC requires a
46
bipolar power supply giving +15 The circuit diagram adopted is given ,0 and -15 volts.
by Fig. 2.4.
The circuit basically involves a variablegain non-inverting amplifier stage,which is fed by a buffer amplifier providing the output signal from the signal conditioner,and a DC level shifting stage,which helpsto offset the DC componentof the input signal. 2K
Input from proximity se
own
+18V
L"wl ShiMng
Fig 2.4 Circuit diagramof the yarn thicknesstester 2.3.1.1.3Power Supply A power supply was built to provide the necessarypower to the signal conditioning unit and the amplifier circuit. It produces+15 V, -15 V, 0V and-18 V. The circuit diagram of power supply is given in Fig.2.5. The -18 V line is for the signal conditioning unit whereas±15 volts are for LM348.
2.3.1.2Calibration of the yarn thicknesstester The output signal voltage is level shifted and amplified so as to give the calibration curve given by Fig. 2.6. The calibration was carried out using a number of steel shim strips of precisethicknessplacedunder the targetplate. A numberof steel shim strips in the rangeof No. 5 (0.05 mm) to No. 45 (0.45 mm) were employed to give different separationsbetweenthe sensorand the target plate. The output correspondingto each shim strip assemblywas read four times using a digital voltmeter, and the average taken.Using linear regression,the equationof the relationshipbetweenthicknessand
>
v>i +0
u>i
a>o
LL
6ýy1 .ti
r. + Ir
d
a H fý
3 ä id)
N
w
N
48
the output voltagewas obtained.This formula is valid for all yarn thicknessexperiments carriedout. The formula is asfollows. Thickness(mm) = 0.006 + 2.9363E-2* Voltage(volt) The thicknessof shimmy (lmm=100) 45
The averageoutput voltage (volt) 15.325
The thickness (mm) 0.45
The averageoutput voltage (milivolt) 15325.0
40
13.7025
0.4
13702.5
35
11.545
0.35
11545.0
30
9.7175
0.3
9717.5
25
8.17
0.25
8170.0
20
6.4675
0.2
6467.5
15
4.8975
0.15
4897.5
10
3.1175
0.1
3117.5
5
1.5925
0.05
1592.5
0
0.05
0
50.0
0.5 0.4
Thbmm [mm )
0.3
0.2
0.1 0.0 vat%9(Mf to
Fig 2.6 The calibrationcurve of tester
49
2.3.2Yarn Tension Measurement Yarn tension is of essentialimportancein yarn winding. A single yarn tension probe was usedto measureyarn tensionasyarn was wound on the package.The device used employsa strain gaugebasedcantilever arrangement,and hasa naturalfrequencyof 2.5 kHz. It is thereforecapableof measuringrapid variations of yarn tensionthat occur in the yarn during winding. 2.3.2.1Yarn tension probe The single yarn tensionprobe usedin this work is a strain gaugebaseddevice. It usesa short cantilevermadeof duralumin.The cantileveris rigidly clampedto the body of the instrument at one end, and carries a 1.5 mm dia. ceramic pin. The cantilever tapers towards its outer end in order to minimise its massand to retain the high degree of stiffness.Four strain gaugesare bondedto the cantileverclose to its baseso that two of them undergo extension in responseto yarn tension, while the other two undergo compression.The ceramicpin works in conjuction with two fixed ceramicguides so as to load the cantilever under yarn tensionaccording to the well known 3-point loading principle. The gauges are connectedin a full Wheatstonebridge arrangementso as to achieve maximum sensitivity to yarn tensionvariations. This is important as the deflection of the cantilever underthe low valuesof yam tensioninvolved (2-50 g) will be quite small. The signal produced by the bridge is only a few millivolts, and is required to be amplified with a gain of the order of 1 k, so that a large enoughsignal is producedfor measurementusing a standardmultimeter or oscilloscope.
Sincethe gaugesare in thermalbalanceby virtue of therebeingbondedto the same metal body (cantilever), the main problem in amplification is keeping electrical noise
pickup to a minimum.This is achievedto a satisfactorydegreeby theuseof screened cabling,wherebythe caseof theprobeis electricallyearthed.A standardDC amplifier speciallymadefor usewith straingaugebridgesis usedfor this purpose(RS435-692]. The amplifieralsopowersthe bridge.In fact theelectricalgain of thebridgecanbe set by adjustingthe bridge supplyvoltage.Oncethis is set at a suitablevalue,the load
50
calibration of the probe is done using a variable resistor which directly sets the gain of the amplifier.
Using the storage oscilloscope the natural frequency of the probe was found to be about
2.5 kHz. Sincethe tensionfluctuationsof yarn involved are generallyof a much lower frequency, this ensures that the output signal of the probe are free of gauge resonance effects. + Bridge Supply
+Vs y,
R1 ý(
R4
Strain Gauge Amplifier
+ Input
P. C. B
- Input R3
'\
/-
R2 - Bridge Supply
-Vs Resistors are 120 ohm
Fig 2.7 Wheatstone bridge arrangement of the tension probe 2.3.2.2 Calibration
of the yarn tension probe
The probe, as shown by Fig. 2.8, has a central ceramic pin, which sensesthe tension of the yarn which is guided by the two ceramic guides on each side of the pin. The probe was mounted on a stand so as to place it in the yarn path. The probe was calibrated as shown by hanging weights from a yarn placed in the probe as shown by the figure. The output voltage of the device was measured on the oscilloscope as shown by Fig. 2.9.
- yarn r-
yarn tension measurement
oscilloscope
weight Stand
>
ý`T
Inverting _ amplifier
1__ Amplifier
Fig 2.8 Set-up to calibrate the single yarn tension measurement device
51
Weight (g)
Output (mV)
25
110
50
190
75
255
100
335
350
300
5 250 E
5
200
0 150
100
output=2.9xweight+41.8
I'll 20
30
40
50
60 70 (gj weight
80
90
100 110
Fig 2.9. Calibration of yarn tensionmeasurement When the device is used for measuring winding tension it was found that the yarn tensionwas in fact of a fluctuating nature. Since only the mean tensionof the yarn is mostly of interest, the small additional circuit was added to the output of the main amplifier, so that a smootheroutput proportional to the meantensionwas obtained.
bVtA Fromampl$fv
OuOkA +
toacNloscOpe
device Fig 2.10the Op-Ampfor smoothsignalfor yarntensionmeasurement
52
2.3.3Shaft Encoder In order to follow the progressof the winding of a yarn package,it is very useful to be able to track the rotational positions of the drum and the packagecontinuouslywhen required. It is thereforenecessaryto introduceshaft encoderson the drum and package axes.
Optical shaft encodersare the most convenient for this purpose, and they produce information about angular displacementin digital form. There are two main types of encoderdevice;the absoluteencoderandthe incrementalencoder.In this project, it was considerednecessaryto place an absolutetype encoderon the drum shaft. For reasons of simplicity and also greaterfreedomfrom false readings,it seemedbetter to place an incrementalencoderon thepackageshaft. The principle of the optical shaft encoderis asfollows. A glassdisk is rigidly connected to the input shaft. A precisepattern of concentricrings is printed on the disk. Each ring is divided into alternatedark and clear segmentsequal to 2", where n is the numberof the ring countedfrom the innermost.A light sourceis collimated, focusedand projected through the disk from one side and is detectedby a photo sensorplacebehindeachring. The patternof illumination of the sensorsdependson the angularorientationof the disk. As the shaft rotates the light source faces different patterns and the photo sensors produce a binary output dependingon this pattern. A shaft encoderwith 8 concentric rings will provide an 8-bit output of its position and henceresolve a revolution of its shaft into 256 parts[Sternheim89].
2.3.3.1AbsoluteShaft Encoder An absolute-shaft to the angular encoderprovidesa uniquebinaryword corresponding positionof its shaft.The outputsignalproducedis in straightbinary or in grey code form. The absoluteencoderchosento determinethe positionof the groovedrumis an 8-bit device.A full cycle of shaft rotationgives readingsin the range0-255 (i.e. divide a revolutioninto 256parts).On thewindertheencoderwassofitted asto givea0 reading
53
when the yarn is on theextremeleft side of the grooved drum, asthis gavea convenient referenceasthe startof a doubletraverse. A difficulty associatedwith an absoluteencoderis that, when a microprocessorreadsits output, the readingmay contain some inaccuracydue to the simultaneouschangingof one or more bits. There are variousencodingpatternsthat are normally incorporatedon the disc to minimise errors. The specificationsof the particular deviceusedaregiven in Appendix A. 1. 2.3.3.2Incremental Shaft Encoder An incrementalshaft encoderis basically similar in construction to the absolute type devicediscussedabove,but is simpler in that it provides a single bit output asits shaft is rotated.The encoderdisk is again divided into a numberof segmentswhich is a power of two. Thus an incrementalencodermay divide a completerotation of its shaft to 256 or 512 divisions. However an externalcounter is necessaryto be incorporatedwith an incrementalencoder,so that the amountof rotation undergoneby it can be determined. The total count of this counter can be zeroed at any arbitrary position of the shaft encoder,hencethe incrementalnatureof thereadingobtainedwith this device.However an incremental shaft encoder normally has two other outputs; one to indicate the completion of each rotation and the other which is similar to the normal output as indicatedabout,but in quadraturewith it. A 512 count incremental shaft encoder was attached to the package shaft on the experimentalwinder. The device was however usedto produce an 8 bit reading of its position by feeding its output into a divide by two counter, and reading only the Q2 to Q9 outputs of it, as shownby Fig. 2.23.The particulars of this deviceare also given in Appendix A. 2.
54
2.3.4Measurement of packagediameter and compression
2.3.4.1Linear variable differential transformer A Linear Variable Differential Transformer (LVDT) was usedto measurethe package diameterwhile the packagewas on the drum. LVDT consistsof one primary winding and two identical secondarywindings connected in series opposition. A movable magneticarmatureattachedto a brass push road takesplace betweenthe primary and secondarywindings. When the primary winding is excited by AC supply voltage the combinedoutput of the two secondariesrepresentthe difference betweenthe voltage induced in them. The relative position of the magneticarmaturein the coils shows the radial displacementof the wound package.Fig. 2.11 showsthe electrical diagramof the LVDT usedin this work. Fig. 2.14 and Fig. 2.15 also show the circuit diagramof LVDT.
Red
Green
D.C. Input
RL
1A 47KQ 0.022 NF
Yellow Brown
Blue
0
Black Armature
Fig 2.11Schematic diagramof theLVDT The LVDT was mountedon a plate on the machineframe as shownFig 2.12,so that its probe sensedthe height of the packagecentre.As a result, what it actually measuresis the compressesradius of the packageon the grooved drum. As the packageundergoes some compressionagainstthe grooved drum due to the pressureimposed on it during winding, knowing thecompressedradius of the packageis of importance.
55
Package
- Drum
Fig 2.12 The method of measuring deformed wound package radius 2.3.4.2. Potentiometer A potentiometer (POT) was adopted for measuring the un-deformed radius of the yarn package by sensing the outside surface of the package using a wide aluminium roller. The roller is attached to a light radius arm mounted on the potentiometer shaft. The potentiometer is mounted as shown in Fig. 2.13, so that it rides on the package surface without deforming it. The potentiometer chosen requires very little torque to turn it, and hence does not cause the roller to compress the package. POT light wheel
0 wound package surface
fix point on package holder Fig. 2.13 The method of measuring un-deformed wound package radius
56
c ýT
i
ýij
i
sO w
r
M
I1NI
O
P
1 ýý
2i
H, "
r
IN -
------
I
"
M
N
ý'
-IFs
ý,,i6
Fig 2.14L.V.D.T excitationandsumming junctioncircuitdiagram(1)
IIS"
ist
57
U s
J
13
-0
3
let iýý. ýIp ýý
F-*
ýý J
0
naa
w
Fig 2.15L. V. D.T. excitationandsummingjunction circuit diagram(II)
58
There are three terminals on the POT. A constant voltage of 5.0 V is applied to the main terminals of the POT. The central terminal, connected to the slider, provides a signal proportional to the position of the roller, and hence a measure of the undeformed radius of the package.
2.3.5Interface Card The interface card enablesthe connectionof the signals producedfrom the different sensorsand amplifiers provided on the winder. It fits into one of the expansionslots in the mother board of the PC. There are two common types of signal, analogue and digital. Analogue signalsvary continuously in amplitude and time, and can be carried by a single conductor. Digital signals are usually in binary, which meanseach such signal will havea numberof bits (binary digits), and requiresa correspondingnumber of conductorsto carry them. The interfacecircuit is basedon the IBM prototypeboard [ RS 435-686].The circuitry on the boardis provided to buffer anddecoderelevantsignalsto simplify the interfacing of the board circuitry to the PC systemand a grid matrix area is provided to facilitate the developmentand constructionof electronicscircuits. The signal from shaft encoders are availablein binary form. Their 8 output lines areconnectedto port A of the interface card. The computer can read the shaft encodersdirectly by addressingport A. The L.V. D.T and the POT signals are sent to the Analogue Digital Converter (A/D). The A/D circuit is basedon chip ZN448 in a bi-polar configurationwith an input rangeof ± 5V. The outputsfrom Brushlessmotor and the steppermotor are takenfrom a Digital to Analogueconverter(D/A); basedon chip AD558N and latch 74LS373.The D/C gives an output of 0-5V for convertednumberof a 8-bit number0-255.
The computerreceivesinformation from the systemthrough the channelsof the interfacecircuit, for controlling the operationof the system;mainly the absoluteand the incrementalshaft encoder,LVDT and POT, steppermotor and brushlessservo motor as
shownin Fig 2.16aandFig 2.16b. In this work, theoutputof both shaftencoderswasconnected to theinterfacecard.The The datawere bufferedand can be readby the computeron the specificaddresses.
59
outputs of the LVDT and the POT were also connected to the interface card. Both
output signals were converted from analogue into digital by two eight-bit A/D converterson the board. Moreover, when necessarythe output from the computer can be convertedinto analogueby the D/A converterchip. Each device connectedto the card has a specific addressusedby the microprocessorin addressingit. The interface circuit diagramis given in Fig 2.17 andthe actualaddressesassignedto devicesis given in Table 2.2.
HEXADECIMAL
DECIMAL
300
768
Start conversionof A/C
301
769
ReadPOT A/C
302
770
Write brushlessmotor D/A
303
771
Write steppermotor D/A
304
772
ReadabsoluteSE A/C
305
773
ReadincrementalSE A/C
306
774
ReadL.V. D.T A/C
CHIP OPERATION
Table 2.2 Addressesusedby the interfacecard
2.3.6Calibration of the LVDT and the POT beforeadopting Severalmethodsof calibratingtheabovetwo deviceswereconsidered the method describedbelow. Severalwound packageswere preparedto a number of different outside diameters,which fitted snugly inside a rigid card board tube of a
diameter,cut to the samelength as the package.A metalsleevewas corresponding attachedto a normalwinding tube as shownby Fig. 2.18.This togetherwith the 4 larger
tubesprovided5 'gauges',usingwhich the LVDT andthePOTcouldbe calibrated.The hard surfacesprovided on each gauge minimised the oscillation of the POT as the grooveddrum was rotatedduring calibration.
60
MICROPROCESSOR II
PROGRAM
INTERFACECARD
ABSOLUTESHAFT ENCODER
EMENTALSHA ENCODER
L.V. D.T
ýý'ý POTENTIOMETER
BRUSHLESSSERVO " MOTOR
Ii: ----------
Mechanical connection Electronicl connection
STEPPER MOTOR
s
RANDOMWINDER
Fig 2.16aElectronic and Mechanicalconnectionsof devices
61
Fig 2.16bA view of winding machine
Cl)
CO fI)
m a a Cm U) i
0 ýp
O
N
v
m
O
,n
N
Cd
U,
A
O
O
f
01
N
I go b .1 I
2p
63
The calibration of the POT and the LVDT were done together as the grooved drum was allowed to run slowly. A program was written in 'C' to read the position of LVDT and POT and display the results on the monitor of the PC. The program is given in Appendix C. 1. All readings taken were saved and plotted on the graphs.
ýý Fig 2.18 The original tube with a metal sleeve The results of the calibration procedure are given below
in Table 2.3. The
corresponding graphs are shown in Fig. 2.19a and 2.19b.
LVDT reading (binary value)
POT reading (binary value)
2 kg
1.25 kg
0.7 kg
2 kg
1.25 kg
0.7 kg
30
24.3
24.7
25
16.3
16
15.9
41.8
65.7
66.1
66.6
59
58.5
59.2
53.8
109.7
110.2
111.2
97.8
97.1
97
62.6
142.6
142.9
143.9
127.9
127.5
127.5
77.9
200.9
201.9
202
175.4
175.6
175.9
Radius
Table 2.3 Calibration data of LVDT and POT read from computer
64
220 pressure 2 kg pressure 1.25 kg pressure 0.7 kg
200
Xý
180 160
x
Output Reading 140 [byte 0-2551 120
a'
100 80.
9
60. 40
Output=(-81.19+3.32*radius) 20.
30
40
50 60 Radiusof Tubes[nTj
70
80
Fig. 2.19aCalibration graphof LVDT
180160
pressure2 kg " pressure1.25kg +* pressure0.7kg
4
140 Output Reading [byte 0-2551
r
120 100
a
80 60 40
Output=$1.198+3.31"radius 20
0 30
40
50 60 Radiusof Tubes[ml
70
80
Fig. 2.19bCalibration graphof POT 2.3.7 Frequency Inverter The winding machine was originally fitted with a3
phase fixed speed motor. By
supplying the motor from a frequency inverter drive, it could be run as a variable speed motor. This enabled the grooved drum to be run at different speeds as required. The inverter allows the speed of the motor to be set from its front panel. The motor could be
65
enabledor disabledfrom a simple switch. Fig. 2.20 shows the connectiondiagram of inverter,and its technicalspecificationsare given in Appendix A. 3.
LED display
Switch
OO Function keys --Do-
OOI
Control Terminals
) -4-]
u
Power Terminals
M2/v
Potentiometer To Motor
ýL
Power Supply
Fig.2.20 Control connectionsfor operationasa speedcontroller The speedof the motor was set usinga potentiometer(10 ku).
2.3.8Stroboscope As in the study of rotating machine parts, it as realised that a stroboscopewill be particularly suitable for the study of packagewinding, as it seemedto be useful for
indicatingwhenribboningoccurredon thepackage. A Genrad Strobolume Type 1540 stroboscopewas available, and it was controlled through the interface card using one bit from the digital output port. The bit was activated when ribboning was detected by the control programme, and the flash occurred each time the package arrived at a chosen orientation. This enabled the package to be visually 'frozen', to facilitate the observation of the ribboning or the ribbon breaking actions.
66
2.3.9Oscilloscope Extensiveuse of a digital storageoscilloscopewas made in this work to observeand measurethe various signals of electronic circuits. For example the calibration of the yarn tensioner,checkingthe quality of signalsfrom shaft encoders,the determinationof the speed of package and drum were greatly facilitated by the use of this instrument.
The oscilloscope which was used in the research is a 60 MHz digital storage device
madeby PHILIPS (model 3331). It has two input signal channels,thereforetwo signals can be simultaneouslyshown on its screen.The signal displayedon the screencan be magnified several times on the time scale for ease of measurement,as well as waveformssavedfor comparisonor printout purposes. 2.3.10Tachometer A tachometer is another essential basic instrument in a study such as this, to conveniently measurethe speedof the motor, the drum etc. A digital photo / contact tachometerwas usedin this work. This tachometerhas an integratedoptical transmitter and receiverat the top of the unit and a mechanicalcontactassemblyat the baseof the unit. The 5-digit LCD automatically inverts when switchedfrom optical to contact so that the display is always the right way up when viewed. The specifications of the tachometerare given below. "
Photo tachometer- direct readingrpm
9 Contact tachometer- direct reading,rangesrpm, m/min, ft. /min. 9 Memory recall of last reading
Measurement range
5 to 19999rpm(optical) 0.5 to 19999rpm(contact)
Resolution
0.1rpm
Accuracy Sensingrange
±0.05%+1 digit (dependon ambientlight conditions) 50 to 150mm
Overall dimensions
H. 215xW. 65xD. 38
Battery
4x AA Cells
Table 2.4. Technical specifications of the tachometer
67
2.3.11Electronic circuits A number of basic electronic circuits were built to perform a number of different functions of measurementand to enablethe interfacing of the different sensorsusedon the winder. 2.3.11.1Frequency to Voltage converter for checking yarn package speed Much of the measurementscarried out on the winding operation were basedon the assumptionof constant speedof drum rotation. Initial measurementsof drum and packagespeedindicated significant fluctuations,which were difficult to explain. Since the output of the shaft encoderswere in the form of pulses,it was not too easyto use them to determineany fluctuation of drum speedtaking placewithin eachrevolution of its rotation. Therefore it was decidedto build a frequencyto voltage converter,which would be fed from the leastsignificant bit from a shaft encoder,which could be usedto display any fluctuation of the drum or thepackageon the oscilloscope. The circuit constructedwas basedon the LM2917 chip which is a monolithic frequency to voltage converter with a high gain op amp/comparatordesignedto operatea relay, lamp, or other load when the input frequency reachesor exceeds a selected rate [National Semi.]. The circuit diagram employedis given in Fig. 2.21. A single bit (bit 0) from either of shaft encoderconnectedto pin no. 1 as input while the oscilloscope was connected to the V0t. As the machine runs, the analogue voltage, which is proportional to the speed of the shaft encoder concerned is displayed on the oscilloscope.
It should be mentionedthat the shaft encodersare connectedusing bellows type couplings which are rotationally rigid. Hence they follow the speedof the drum or the
packagewithouterror.
68
Freauency to Voltage Converter (LM 2917) VcC-15 V
V out OK
(Fre
Vout=Vcc*Finput*C1*R1*K (K=gainconstant typically 1) around
Fig. 2.21 The connectionof FN converter
2.3.11.2Amplifier card 2.3.11.2.1Zero adjustment and level shifting for POT and LVDT As seenin Fig. 2.22,two op-amps(operationalamplifier) are usedto adjust the amount of gain and level shifting for signalsprovided by the LVDT and the POT. The op-amps are basically differential amplifiers, and are hencecapableof amplifying the difference betweenthe signal levels impressedon their inverting and non-inverting terminals.The negative feed back applied between the output terminal and the input determinesthe gain of an op-amp. The variable resistor is to adjust the Voutl to a required. LM348 analogueamplifier chip employedis a quad741 type op-amp.Two.of theseare usedfor the POT andthe other two for the LVDT.
69
zero adjustmentand level shifting for LVDT and POT voltaae
Vin (from LVD' or from PO' Vout2
5V
Fig. 2.22 The schematicdiagramof amplifier for POT and LVDT With reference to Fig 2.22 the gain of amplifiers is obtained from the following relations: Rf1 Gain = Rin
and
Rfl Vout, _ -Vin Rin
Rf2 Gain 2= Rin 2
and
Vout
Rf 2 2= -Vout, Rin2
2.3.11.2.2Counter for incremental shaft encoder A 12- stagebinary ripple counter HCF4040BE was usedto count the pulsesfrom the , , incremental shaft encoder.The output signal (A signal) was fed to the counter as shown. As the incrementalencoderof the 512 count/revolutiontype, to obtain an 8 bit output which moved throughvalues0 to 255 eachrevolution, the bits 02 to Q9 from the output of thecounterwere employedasshownby Fig. 2.23.
70
Counter for incremental shaft encoder
3.3 K
a
11
=
7
RESET
e
1.5K
5
10
INPUTHCF4040BEY 12 stage Ripple counter
4
Signal from incremental
3 2
shaft encoder
13
3.3K
(green)
+5V
VDD SUPPLY
12
16
V$8 2"
Q2
purple
Q3
orange -
Q4
Grey
Q7
blue
Q5
yellow
QB
brown -
Q8
red
Qe
green -
27
navy
,
2' 22 V 2'
-
2s 2'
bIadcJl-IU
Fig.2.23.Conversionof signal from the incrementalshaft encoderto 8 bit binary code 2.3.11.2.3Voltage regulator for POT The switching action of the PC power supply and the frequencyinverter for the motor, and the mains 50 Hz power supply being all presenttogether,the winder was found to havea noisy electricalenvironmentwhich conflicted with having cleananaloguesignals to be available for feeding to the amplifier circuits. So it was necessaryto provide the voltage for supplying the POT by deriving it from +15 V from the main power supply. A 5V positive regulator was available and was usedfor this purposeas shown by Fig 2.24. The high level of regulation provided by the chip and its very low output impedancepermittedthenoise level in its output to be very low.
Input
+15V
+5V
7805 100mF
Ground
Fig.2.24Thevoltageregulator
TT
100n
Outputto POT
71
2.3.11.2.4 Stepper motor drive card for lifting the package Action of the lifting the package discussed in Chapter 5 required a special card to drive the stepper motor used. A proprietary stepper motor control card type supplied by RS Components was used for the purpose. The configuration of the card is given in Fig 2.25. The card basically requires 3 signals from the PC through the interface card described above. These are a single bit pulse train which decides the speedof the motor, a single bit for enabling the drive card, and a single bit for deciding the direction of rotation. Of these the pulse train is provided by a frequency generator on the motor control card itself. The other two signals are generated under programme control by the PC. The circuit diagram of computer card and stepper motor and all connection for the action of lifting package are given in Appendix A. 5.
stepper motor card
computer card r--
Box holding cards
to emergency switch
Fig 2.25 Basic configuration for the lifting the package
The more information about the action is given in Chapter 5. 2.3.12 Brushless Servomotor and controller card The initial operation of the winder indicated the inadequacy of the power of the standard motor fitted on it, to carry out all the experiments planned in this research. A DC brushless motor was available, and was considered greatly superior in its power rating and versatility of control compared with the standard motor powered from the frequency inverter. The motor has its own microprocessor controlled power supply, and can be run so as to provide a variable speed profile under program control. The latter stages of experiments were carried out by using this motor to replace the standard motor.
72
This motor is of the BRU-200 series supplied by Robbin & Myers. The motor directly
replacesthe old motor by direct connectionto the grooveddrum shaftof the winder. 2.3.13 Low Voltage Power Supply On the systemseveraldifferent power supplieswere usedto provide the necessaryDC power for the electronic circuits. The separatepower supplies are important, as they greatly reducethe chanceof electrical noisebeing createdin the signals,if all circuits were supplied with power derived from one source.The circuit diagram concernedis shownby Fig 2.26. One of thesewas built to provide a highly regulated dual rail 15V and a +5V supply for the amplifier circuitry for the LVDT, the POT and the divide by two counterfor the incremental shaft encoder. A separatesupply powers the lamp side of the shaft encoders.A general purpose power supply (Thurlby Thandar PL 310) was used to obtain +5V to be appliedto the PC interfacecard for the steppermotor drive. A further generalpurposepower supply (Farnell PDD 3010) was usedto power the steppermotor drive. 2.3.14Actuators
2.3.14.1Steppermotor drives In order to implementthe anti-ribboningmethodby lifting the package,a gearedstepper motor was usedto operatethe arm as shown by Fig 5.8 in Chapter5. Technical details of the steppermotor are given in Appendix A 5.1. The motor is controlled through the computer.
2.3.13.2SafetySwitches The safetyswitcheswereusedto disconnectthe powersupplyof the steppermotorsif the packagelifting arm were to lose control when the anti-ribboning was in progress. This could be reset again by positioning the arm under the package and then the operationwould start again.Safetyswitchesconnectionare shownin Appendix A 5.2.
U, U, 0 M Z
N 0
> N
> as
Ö
74
2.4 PC Hardware A standardPC typecomputerwas usedasthecontroller of the experimentalwinder. The specificationsof PC are "
14" monitor VDU, resolution 480 x 640 ( Elonex SuperVGA)
"
Motherboardwith 200 Mhz. processor, Motherboard in PC has 3 ISA (Industry StandardArchitecture) and 2 PCI (PeripheralComponentInterconnect)slots.
"
32 Mbytes RAM
"
518 Mbytes Hard disk
"3 "
1/2" floppy disk drive Keyboard,Mouse,Printer (PanasonicKX-P1180 Multi-mode printer)
2.5 Software The PC has installed in it MS-DOS version6.20, Windows 3.11, and the programming languageBorland C++ version3.1( Borland Internationalinc. 1992). All programs in this project were written in the C language.For analysingdata, the MicrocalT" OriginTm version 4.10 16 Bit and Microsoft Excel were used on a separatecomputer. 2.6 Noise on the signals and other problems The modified winder has a numberof electrical and electronic circuits associatedwith it. Someof thesehandleanaloguesignalsand othersdigital signals.Someinvolve small
currentswhile otherssuchasthesteppermotorcontrolcardhandlehigh currents. One of the problemsexperiencedin the operationof all thesetogetheris the possibility of one circuit affecting another,as seenby the noise contentof the various signals.The main supply at 230 v and 50 Hz, causesconsiderablenoise in the signals as a 50 Hz componentriding on the DC level of the signal. An evengreaterproblem was found to be causedby the electrical switching (EMI) within the PC power supply and the frequency inverter used to drive the winder motor. EMI causesnoise by inducing voltagesto appearon the high impedanceinput segmentsof circuitry.
75
Theseproblemswere experiencedat the beginningof the experimentsand a numberof measureswere taken to reducethe problem to a tolerable level. The oscilloscopewas very useful for observingnoiseon the signalsas well as for checking the effectiveness of the variousnoisereductionmeasuresthat were implemented. 9 The frequencyinverter was separatedfrom thejunction box where many of the signal carrying wires meet, by as much as 1 metre,and ferrite sleeveswere put on the mains cable that suppliesthe inverter. This producedsomeeasingof the noise. "
The 3 core power cablebetweenthe motor and the inverter was replacedwith a screenedtype anda ferrite sleevewas added.This provided somefurther relief.
"
It was recognisedthat the motor had not beendirectly earthed.The inverter and the motor were separately earthed and this contributed to a considerable reductionof noise.The lids of die-casthousingusedto containthe circuit boards boxeswereproperly closed.
"
Shaft Encoderwas earlier poweredby the sametransformer that powered the LVDT circuits. As the shaft encoderseemedto be was drawing a considerable current the supply to the shaft encoderwas separated.The board which has amplifier circuit for POT and LVDT was powered with that separated power(+15V and +5 V).
"
All mainsplugswere checkedto ensuretheywere properly grounded.
9 Thereis a small motor which runs a pressureshoesfor yarn tension,which operatedfrom 415 V, 3 phasesupply. It was thoughtthat that threephasesupply causedsomenoise,but careful checkingindicatedthat it was not so. These measuresfinally achieved an acceptablenoise level in the various signals. As
indicatedabove,the separation of the power suppliesenabledsomemeasureof noise reductionby reducing'crosstalk' due to feedbackbetweencircuits acrosspower supply lines.
76
CHAPTER THREE
MEASUREMENT OF PACKAGE DENSITY, SLIPPAGE AND DEFORMATION
3.1 Introduction The Study of the random winding processshould lead to the determinationof such aspectsas the distribution of yarn in the package,deformation of the packageat the contact point between the drum and the package,and the nature of slip between the packageand the drum. The drum surface is hard and non-deformable,but the wound packagesurface is relatively soft, deformableand non-homogenousin its construction dependingon the yarn tensionandthe pressuremaintainedbetweenthe packageand the drum. Density distribution of a cheesepackagesshows differencesdependingon the type of winding used. Generally it is known that the density of the cheesedependson the cradle pressure,the winding tension and the wind angle. In random winding, the wind angle is constant throughout the package.In precision winding the wind angle diminisheswith increasingdiameterof the package. According to Wegenerand Schubert[Wegener68], the density distribution in a cheese wound packageproducedby precisionwinding shows a sharp reduction near the core, accompaniedby a gradual increaseof the density with increasingradius. Finally the densityfalls againsharplynear thefinal outer radiusof the cheese,asshownby Fig.3.1.
In this chapter,preparationof yarn packagesfor each of the experimentwill be explainedand the resultswill be given. 3.2 Physical properties of Yarn The parametersof the yarn usedfor winding were testedin the testing laboratory under standard testing conditions. Tensile properties were tested on the Statimat at 100 mm/min on a samplelength of 200 mm. Yarn twist was measuredon a twist tester and its count was measuredby winding 100 m of it on the yam count reel andweighting on a sensitivebalance.Yarn specificationsare given in Table 3.1.
77
0.90
0.85 Density of cheese
0.80
[g/cm3] 0.75
0.70 0.65
0.60 60
80
100
120
140
Cheeseradius[mm]
Fig 3.1 Variation of the densityof cheesewith radiusin precisionwinding Parameters of Yarn Type of yarn
Worsted
Yarn colour
Ecru
Yarn count
R42/2 tex
Yarn twist
544 t/m
Twist directionof yarn
S
Average Elongation
18.44%
Averagetenacity
8.38cN/tex
Average work to rupture
884.95cN/tex
BreakingForce Modulus(0-3%)
309.21cN 115.29cN/tex
Modulus(0-5%)
117.87cN/tex
Table 3.1 Yarn specifications
160
78
3.2.1 Elastic Modulus of Yarn Elastic Modulus of Yarn, EY, is defined as the force or tensionrequired in the yarn to producea unit strain in the length of the yarn. It is expressedin grammes.It was found by usingthe InstronTestile Tester,Model 1122. The Instron Tensile tester consists of a recording unit, acrosshead unit and a control panel. There are two jaws in the crossheadunit that allow a specimen to be clamped in the vertical direction. The upper jaw is suspended from a load cell which is fixed centrally in the stationary crosshead and driven upwards and downwards by screwed rods on each side, whereas the lower jaw was fixed at the surface of the tester. The loadextension curve was obtained on the chart in the recording unit. The movement of the chart was proportional to the extension of the sample. A pen on the chart unit moved across the chart when the tensile force applied to the sample of yarn. The speed of the chart and crosshead could be adjustable due to the best-proportioned curve. The adjustments necessarycan be made on the control panel.
The specificationsbelow were usedin the test: "
The length of the yarn was 150 mm.
"
The speedof thecrossheadwas 5 mm/min.
"
The speedof the chartwas 200 mm/min.
"
Maximum load appliedwas 100 g.
Therefore,4 an of chartpaperrepresented1 mm extensionof the yarn sample. The test was repeated3 times on a sample.The tester was set to automaticreversing movement. When the load reached the maximum load, 100 g, the upper jaw was stoppedanddriven in oppositedirection until the original point of the sample.Therefore the extension and recovery curves were both obtained on the chart. The modulus of elasticity could be calculatedfrom the curveson the chart.
79
3.2.1.1Results The curve in Fig. 3.2 was obtainedfor yarn used in the experiment.The extensionof yarn is significantly different for low loads, as comparedwith that for higher loads. However, the tension of the yarn during the winding test was between 10-50 g. The modulus of the yarn could be obtained in two regions, 10-20 g and 20-50 g, by constructingtangentsto the curvesas shown.
4
10 0.763 30 1
0
10
20
30
40
50 60 Tension(d
70
80
90
Fig. 3.2 Instron Load-Extensioncurve
ElasticModulusat theregionof 10-20g.:
EY _
10 tensiongradient correspondingstrain _ 0.38%50
3866 g
Elastic Modulus at the region of 20-50g.:
EY _
tensiongradient 20 _ 0.7%0 correspondingstrain
5898 g
100
110
80
3.2.2Transverse Modulus of Yarn
3.2.2.1Measurement of Yarn Thickness In this test, it was necessaryto measurethe compressionof yarn undervarying loads andwith different numbersof intersectingyarn layers.To ensurea goodaccuracyin the measurement,it was important to placethe yarns sufficiently close to eachother with equal spacingbetween them. Actually the angle betweenthe direction of yarn layers should be equal to that betweenintersectingyarns in the packagewhich is about 60°. However,90° was considerednot an unrealisticvaluein the first instance. The yarns in eachlayer were placed in the coils of four close coiled springs,placedas shown by Fig. 2.3 in Chapter2 using suitable wire hooks. This ensuredan equal and close spacingof yarns in eachlayer. When pressureis applied using the load plate, the outermostlayers of yarn will have contactalong their full length on the metal surfaces, whereas those that are held within the layers have yarn that bear on each other at intersections.
Tests were carried out with different numbersof yarn layers placed under the loading plate. It was ensuredthat therewere equalnumbersof yarn intersectionson eachside of the plate. Different numbersof yarn intersectionswere formed by laying the yarn under a low tension , of which the overall thicknesswas measuredwith a digital voltmeter. The measurementwas repeatedunderdifferent loads. Table 3.2 gives the resultsof the testing carried out on a Tex R42/2 worsted yarn. The results were used to obtain the relationship between the thickness of yarn under different pressurelevels.
81
LOAD
LOAD/POINT
(g)
(g)
THICKNESS (mm) 2 LAYER(t2)
3 LAYER(t3)
6 LAYER(t6)
130.5*
16.32
0.31
0.45
155.5
19.44
0.30
0.42
-0.86
00
180.5
22.57
0.29
0.40
0.82
ö
230.5
28.82
0.28
0.37
0.72
280.5
35.07
0.27
0.34
0.66
330.5
41.32
0.26
0.33
0.60
380.5
47.57
0.26
0.31
0.57
430.5
53.82
0.25
0.30
0.54
480.5
60.07
0.25
0.30
0.52
530.5
66.32
0.24
0.29
0.50
580.5
72.57
0.24
0.28
0.49
130.5
7.25
0.34
0.50
--
155.5
8.64
0.33
0.48
--
180.5
10.03
0.32
0.45
230.5
12.81
0.31
0.42
-0.92
280.5
15.59
0.30
0.39
0.92
330.5
18.36
0.29
0.37
0.85
380.5
21.14
0.28
0.35
0.80
430.5
23.92
0.28
0.33
0.76
480.5
26.70
0.27
0.33
0.73
530.5
29.47
0.27
0.32
0.70
580.5
32.25
0.26
0.32
0.68
130.5
4.08
0.39
0.52
--
155.5
4.86
0.38
0.50
--
180.5
5.64
0.37
0.47
230.5
7.20
0.35
0.44
-0.96
280.5
8.77
0.34
0.41
0.93
330.5
10.33
0.33
0.38
0.85
380.5
11.89
0.32
0.37
0.81
E z
22
c
c
0
430.5
13.45
0.31
0.36
0.77
.8 1
480.5
15.02
0.35
z
0.30
0.74
530.5
16.58
0.29
0.34
0.71
580.5
18.14
0.29
0.33
0.6
* Note: Weightof loadplate=130.5g.
Table 3.2 Thicknessmeasurements
82
3.2.2.2Results On the tester, the outermost yarns of two or more layers placed on the tester contact with the flat metal surfaceson the one side, and makeyarn to yarn contacton the other. All yarn in betweenbearon other yarns lying in a transversedirection. What is basically required is to determine is the thickness of a yarn, when compressedtransversely betweentwo similar yarns. x= thicknessof an intermediateyarn, and
Let
y= thicknessof an outer yarn, Thesewill be ideally constantfor a given level of load
Thusthemeasured thicknessof n layers T=(n-2). x+2y If t is themeasuredthicknessof n layers then (t-T) is a measureof the error in any one measurement.As the measurementis repeatedfor a numberof valuesof n, it is possible to calculate z and y, the best estimatesof x and y, basedon the method of least squares.
Theerrorsof measurement are (t; - Ti) (t; - T;) = t; -2y-(n1
-2)x
defining, With themethodof leastsquares, N
S=2: (t; -2y-(n1
-2)x)
2
S= Sumof squaresof errors
as os o ay ax
83
N
as
2(t; -2y - (n; -2)x)(-2)=0 -=1: 0Y ._ý ý (t -2 y- (n; -2)x-)=O NN
ýt1
-2Ny-ý(n1 N
as
Also
äx
(3.1)
-2)=0 2(t; - 2y - (n; - 2)x)(-(n;
=
- 2)) =0
;.,
NNN
(t(n; -2)-2yý(n;
-2)-xý(n;
(3.2)
-2)2=0
Solving equation1 and 2. Firstly, y is taken from equation1.
NN
t1 -x1:
(ni_2-2)
2N
Substitutey into Equation2
[±t1-i(n1-2) NNN
Nýt; (n; -2)-
ý(ni
-I
-2)-21i
ýt;
Nýt; (n; -2)-ýt;
-xj(n;
-2)-
-2)
ý(n;
ý(n; -2)+xý(n; i. 1 W
_ XT z
(11i
-2)-xNý(n,
-2)ý(n;
-2)Z=0
.,
-2)2=0
-2)-xNý(n;
-2)2=0
84
NNNNN
N5"t; (n; -2)-I: i=1
i=1
t; Y(n; -2)=x i=1
NI(n; i=1
-2)Z-., i=1 (n; -2),
W
(n; -2)
NNN
NEt; (n; -2)-Yt; XNNN
E(n;
-2)
(3.3)
NJ] (n; - 2)2-ý (n; - 2)ý (n; -2) i=t i=, i=,
NN
t1 y
(ný -2) X
2N
2N
(3.4)
whereN= Numberof 'stacks'testedeachwith a different value of n
Oncez is obtained,thetransverse elasticmodulusE maybe definedasfollows.
E_
AID d
(3.5)
X
For this case z and y were calculatedfor a given load at eachintersection.Resultsare given in Table 3.3.
85
The results of the abovecalculations,for a rangeof load values,are given in Table 3.3 below.
load/point
y
Oload
(mm)
OX x
(g)
E (g)
0.16
0.16
0.02
1.56
63.90
8.77
0.15
0.15
0.10
1.56
15.27
10.33
0.14
0.15
0.06
1.56
27.73
11.89
0.13
0.14
0.92
12.81
0.16
0.14
-0.21 0.23
0.65
-4.37 2.77
13.45
0.12
0.14
0.05
1.56
28.63
15.02
0.11
0.14
0.57
15.59
0.16
0.13
-0.41 0.32
0.99
-1.39 3.08
16.58
0.11
0.13
0.04
1.56
39.46
18.14
0.10
0.13
0.22
18.36
0.15
0.13
-0.40 0.03
1.08
-0.55 42.52
19.44
0.14
0.15
0.07
1.70
25.91
22.57
0.13
0.14
0.07
1.35
20.74
23.92
0.12
0.12
0.05
2.78
59.29
26.70
0.12
0.12
0.05
2.12
42.49
29.47
0.11
0.12
0.04
2.78
71.02
32.25
0.11
0.12
0.08
2.82
34.27
35.07
0.10
0.13
0.15
6.25
42.76
41.32
0.08
0.13
0.06
6.25
99.89
47.57
0.08
0.12
0.06
6.25
104.49
53.82
0.07
0.12
0.06
6.25
103.46
60.07
0.07
0.12
0.04
6.25
146.15
66.32
0.07
0.12
0.04
6.25
168.55
(g)
X (mm)
7.20
Table3.3 Resultsof Analysis
86
0.20
0.15
X [mm]
0.10
0.05
0.00
0
20
40
60
80
Load / Crossing point [g/point)
(-(load/point-72))
-0.00287 + 0.15383e
72.78
0.20
0.15
Y
[mm]
0.10
0.05
0.00 0
20
60 40 [g/point) Load / Crossing points
(-(load omt-72))
y=0.11966 + 0.0362.e
Fig 3.3 Variation of z and y with load
1184
50
87
200
150 E
[9]
100
50
0
20
40 Load / point [g]
60
80
Fig 3.4 Variation of E with load
E= 14.8.e
load/point 26.65
3.3 Preparing yarn packages Random wound packagewas preparedto different diametersby varying the winding tensionandpressurevaluesappliedin eachcase. The pressuresettings chosenwere 2 kg, 1.25 kg and 0.7 kg. The winding tension settings applied to the yarn were maximum, medium and minimum, which correspond to 30 g, 19 g and 8g respectively.The diametersof packagechosenwere 80 mm, 120mm, and 160 mm. The experiment are basedon all the combinationsof tension, pressureand diametervaluesgiven above.
88
3.3.1Pressure On the winder, the pressureto be applied on the packagecan be set to a desiredvalue. This pressureis provided by a spiral spring in a way that it can also compensatefor the growing weight of the package[Schlafhorst67]. The initial tensionof the clock spring is adjustable.The rangeof setting available for the spring are from 1 to 6. The scale position 1 gives the maximum pressure,whereas the position 6 gives the minimum pressure.If necessaryan additional weight can be put on the packagecradleby means of the moveableweight. The effect of this additional weight of 1 kg can be changedby moving it upwardsor downwards. The pressure can be checked by means of a spring balance before starting any experiment.The pressuresettingschosenfor this experimentare thefollowing: 2 kg (maximum) - the scalewas 1 and additional weight on the packagecradle is on and it was placedat full length.
1.25 kg (medium)- the scalewas 3.5 andadditionalweight on the package cradleand it was placefull length. 0.7 kg (minimum) - the scalewas 4.5 and no additional weight on the package cradle. It was observedthat the yarn could not be wound to the packagewith under 0.7 kg pressure.Below this setting, the tube will not remain on the drum with sufficient
Hencetheminimumpressure pressure chosenwas0.7 kg. 3.3.2 Yarn winding tension The winder has an adjustabletensionerwhich was identified in Chapter2. As checked by the electronictensionprobe(describedsection2.3.2 in Chapter2), the maximum and minimum tensionsthat could be obtainedby adjusting the tensioneramountedto 30 g and 8g respectively.Therefore the medium tension setting was chosenas 19 g. The setting was facilitated by the colour coding of tensionvalues marked on the tensioner adjustment.
89
As a rule of thumb,on a winder the winding tensionusedshould be about 10-15 % of the mean yarn-breakingload [Oxtoby 87]. Weak places on the yarn can be removed underthis tensionin the caseof naturalfibres, while strong syntheticsand blendsdo not necessarilyfollow this. Too high a winding tensionmay permanentlyimpair the elastic properties of a yarn and cause fabric faults. According to this rule, the maximum tensionappliedto chosenworsted yarn should be 35 g due to the meanbreakingload of the yarn.
3.3.3Oscillationof drum On the winder, the grooveddrum canbe given an axial oscillatory motion that displaces the yarn laterally on the edgesof the package.This enablessoft edgeson the package which are important for certain operations such as pressure dyeing. The lateral displacementmotion moves the yarn guide cylinder sideways 30 times per minute. 3 different displacementsare provided; 2 mm, 3 mm and 4 mm. If necessary,it can be cancelled completely. For producing the packagesfor experiments, the maximum displacementof 4 mm, was used. Use of the lateral displacementof the package preventedthe formation of hard edges,and allowed the packageto contact the drum properly along its full length. 3.3.4 Packagediameter The wound packagediameterschosenfor experimentwere 80 mm, 120 mm, and 160 mm. A computerprogram was written in the C-languageto check the diameterof the packageby meansof the LVDT during winding of yarn to the package[Appendix C. 1]. Only one variable had to be changedon the programme,namely the packagediameter. For example,when 120mm packagediameterwas requiredto be produced,the variable was assignedthevalue 120.
3.3.5Speedof winding drum The speedof thedrumwasset 1800rpm,corresponding to a yarnspeedof 510in/min This corresponded to a speedsettingof Decimal215on the motorchive approximately. controller.
90
There are several ways to check the accuracy of the drum speed. One way is to count the number of revolution of the drum in a time using the absolute shaft encoder connected to the drum shaft. A program was written for this purpose [Appendix C.2]. It was found that the maximum value 255 of the byte gives 2133 rpm. But this speed was not chosen because in any case the maximum speed would be needed for another purpose. The particular speed of 1800 rpm was also the speed used with the original motor fitted on the winder using the frequency inverter.
Table A. 4 in the Appendix A gives the speedsreachedfor different values of speed commandsignal appliedto theBRU 200 controller. The other way of the checkingthe speedis to use the optical tachometer.A small piece of retro reflective tapewas stuck on the drum, to enablethe tachometercount the each revolution of the drum.There was additionally thepossibility of using theV/F converter circuit given in section 2.3.11.1 in Chapter2 using any of the bits output from the absoluteshaft encoder.All these methodsapplied to check the drum speedgave the sameresult. As the drum diameter is 0.09m., the surface speedof the drum is closely 510 m/min. 3.4 Density of random wound packages
3.4.1The method of finding the density of random wound packages The investigation of the density distribution of the packageis one of the important aspectsof the study of the random wound package.It was intended to measure a numberof packagespreparedunder different pressure,winding tensionand diameterfor this purpose.In order the obtain information on the distribution of yarn weight with packagediameter,it was necessaryto find the volume and the weight of the packages, as theywere wound from the minimum size to the final size. The weight readingswere obtained on an accurateweighing scale. The gross weight of the package minus the weight of the tube gave the total weight of yarn on the package at each stage of completionof it.
91
Packagesshapeis cylindrical exceptthe side facesof it. During winding eachnew layer producedpressureon the previous one. At increasingpackagediameter,more pressure was being appliedto the inner layersof the package.One of the resultsof this is the two outer facesof the packagebecomingconvex.
i i i
ý
ý
'ý i
Fig 3.5 sideview of a package To calculate the packagevolume, the diameter and length of the packageshould be measured.The diameteris relatively easy to measure.The length of the packagewas taken as the averageof the sum of the maximum and minimum lengthsof the package. The equationfor density of the packagewas takenas the following:
(3.6)
Density(g/ cm3) =nw 4.
(D2 -d2)a
Where w- the net weight of the yarn on the wound packagein gram. (w= the weight of the package- the weight of the tube*) * the meanweight of thetube is 54.7 g.
D- thediameterof thepackagein cm includedthetubediameter d- the diameterof the packagein cm subtractedtube diameter
1- the averaged lengthof thepackagein cm
92
3.4.2.The results of the density of the random wound packages Randomwound packagesproducedunder different winding tensionand pressurevalues were measuredfor weight and volume to determinetheir density. All results obtained are shownin the Appendix B. 1. The distribution abovenaturally dependson the type of yarn involved. The yarn usedin this experimentR42/2 worsted. However with a different yarn, the results are unlikely to deviateto any large degree. The density of packageswas averagedover 3 packagesfor each of pressureand yarn tension.They are tabulatedin Table 3.3.
PRESSURE (kg)
5 Z
2kg
1.25kg
0.7 kg
30 g
0.478
0.457
0.444
19 g
0.435
0.414
0.4
8g
0.381
0.36
0.351
Table 3.4 Density of randomwound packages
The resultsof Table 3.4 were plottedyarn tensionagainstdensityof packagein Fig 3.6. According to the Fig.3.6, when more tension applied to the yarn, the packagedensity increasesin the manner shown, under a constant pressure.It is also clear that the pressureof packagecradleinfluencesthe density. An increaseof yarn tensionfrom 8 to 30 g has been observedto increasepackagedensity from 0.381 to 0.478 g/cm3at the maximum pressureof 2 kg. The difference in density is 0.097 g/cm3. Although an increaseof pressurefrom 0.7 kg to 2 kg has beenobservedto increasepackagedensity from 0.444 to 0.478 g/cm3 at the maximum yarn tension applied, the difference in
93
density brought about is 0.034 g/cm3. It is clear that changing yarn tension has a greater effect on package density than pressure applied to the package holder.
0.48
.LKý 9 ý 2`'v.
0.46 0.44 0.42
0.40 0.38 0.36 0.34' 5
I'IIa 10
15
20 Yarn Yension [g]
25
30
Fig 3.6 The densitydistribution in termsof yarn tensionand pressure
For a betterunderstandingof the variation of densityof wound yarn packages,they can be plotted in a 3D representation.Using the graphic software Origin, the graphswere preparedas given below. The X, Y, and Z-axis show the radius of the package,yarn tension and density of the package respectively. Each graph has been drawn for a constant pressuresetting, and they show a similar density distribution. If pressureis
kept constant,a lower winding tensionmakessoft packages;and a higher winding tensionmakeshardpackages.
94
PRESSURE2 ka
0.48
0.46
0.44
042
0 4C
30
03
s
-eý+aeeýmmý
...
80
Fig 3.7 the density distribution at the pressure2 kg PRESSIFE
1 25 k.
046
044
042 21 0.40
0.3E
0.3
5
30
sýý --'peck. /mmý
80
."
Fig 3.8 the density distribution at the pressure 1.25 kg PSS
IRF n7
kn
0.44 042
0.40 r 0.38
o.3 5
oq
. igac4
aCs/rnm/
80
Fig 3.9 the density distribution at the pressure 0.7 kg
30
95
3.4.3 Discussionof Density The results of density distribution in packagesare very much as expected. The measurementof package dimensions proved to be rather complicated. The wound packageis not isotropic and is deformable.The side facesof the edgesare not flat and that is why the axial length of packageswas averaged.Due to the difficulty of accurate measurement,somesmall fluctuationof the resultscan be expected. 3.5 Slippage
3.5.1 Introduction As a round body drives anothersuch body by frictional contact, there will be some opportunity for slippage between them to occur. In random winding, while yarn is wound asa cheese,slippageis possiblebetweenthe drum as driver andthe packageas a driven body. This combination is rather different from the combination of two hard bodies becausethe wound packageis deformableat the contactpoint. Due to a number of reasons,slippagemay vary as the packageradius increases.As packagediameter increases,the deformationis likely to increase. The amountof slippagemay dependon : "
type of yarn being wound i.e. cotton, fibro, wool
"
yarn count
"
yarn tension
"
surfaceof the driving drum (steel,bakelite,chromium plate etc.)
"
pressurebetweenthe drum and the package
3.5.2The methodof measurementof slippage While some mechanicalslippageis possible between the two bodies concerned,it is
it asthepackagerotates. extremelydifficult to devisea methodof measuring The method of the slip observation adopted is as follows. First the packageswere preparedin different sizes under different settings of yam tension and pressure. The preparedpackageswere mounted on the winder. The winder was run without any addition of yarn. The effect of the yarn drag was neglected. The packagepressure
96
applied during experiment was the same as that used for preparing the package i. e. If a package was produced with 2 kg pressure, the pressure used during the measurement of slippage was 2 kg for that package. The number of package revolutions produced was counted for a fixed number of drum revolutions. 2000 drum revolutions were considered adequate for the measurement. A computer program was written in C language to count the package and drum revolution, at the same time the drum revolution is comparing the fixed number indicated [Appendix C.3]. A loop in the program ran 3 times while counting the package revolutions for the fixed number of drum revolutions and the results of package revolutions were averaged.
One of the points to be noted is that during the preparation of the packages the maximum oscillation of winder, 4 mm, was employed in order to obtain soft edges to the package. This would allow package contact with the drum over the full axial length of the package. 3.5.3 Provision of the drag effect
3.5.3.1The pulley To give a same effect of yarn drag to the winding package when a package was rotated with no yarn winding, an aluminium pulley was mounted on the package shaft, on which a weighted cord was added to provide a controlled amount of frictional drag. The
detailsof thepulley are shownby Fig 3.10.
R 35.5 mm
3.5 mm a- 1.41-
Package
E4E
--------------
E--
Pulley
1.5 mm
Side view
*-T 5.5 mm
Fig 3.10 a) the dimension of the pulley
Grooved drum
b) the pulley on the machine
Motor
97
3.5.3.2The calculation of weight hung
T2
Fig. 3.11 The drag due to cord with weight
With referenceto Fig 3.11 , the following equationcan be written; TZ = eN.
a
T, T2 = T,.eµ,°`
(3.7)
Where
T2 - tensionon the cord at the left handsideof the pulley T1- tensiondue to theweight on the cord at the right handsideof the pulley µ- Coefficient of thefriction betweenthe cord and the pulley a- the anglewhich cord coversthe pulley Torque due to yarn = T.d/2
(3.8)
Where T- yarn tension d- packagediameter The force shouldbe samewith the torque due to yarn (T2-T1).r = T.d/2 where
r is theradiusof thepulley
(3.9)
98
Substituting for T2 from equation (1)
(T,
).r = T.d/2 .e"-T, T.d/2 Tý(el*' - 1).r
(3.10)
T, is the weight that is hung from the cord to suit the packagediameterand tension applied to the yarn. The coefficient of friction (µ) betweenthe cord and the pulley is approximately 0.25 and the angle of the wrap cord on the pulley ((x) is 4/3n. The chosenpackagediameteris 120 mm on this calculation.The resultsare approximately samewhen the diameteris chosen80 and 160 mm. The 120mm packagediameterwas chosenbecauseit is close to the mid value of packagediametersthat can be handleon the winder. The inner radiusof the pulley is 35.5 mm. Substitutingall valuesto the equation(4), Weight = T1 = 0.9.T The weight dependson what winding tensionis appliedto the yarn. Maximum tensionappliedis 30 g, theweight will be 27 g- in practice30 g. Medium tensionapplied is 19 g, the weight will be 17.1 g- in practice20 g.
Minimumtensionappliedis 8 g, theweightwill be 7.2g- in practice10 g. 3.5.4Resultsof Slippage The results of slippageare astabulatedin Table 3.5. The radiusindicatedon the table is the actual radiusreadingi. e. the deformedpackageradius,from the LVDT. All individual resultsare given Appendix B.2. The methodof calculationof slippageis given below. Vdrum=2.7t.N. R
(3.11)
Vpack. ge=2.n. n,r
(3.1)
(N. R-n.r) OV=Vdrum-Vpackage=2.7t.
(3.13)
99
AV 2. (N.R - n.r) 100= n. 100=[1- n.r ]. 100 slippage(%)=. V 2.7c. N.R N.R slippage(%)= [1-
n.r 1.100 100 n.r = - 900 2000.45
(3.14)
To find theloss of thepackagerevolution: Assumingno slip, the velocity of the packageshould be the sameas the velocity of the drum at thecontactpoint. Vdrum
=
Vpackage
2.ic.N.R = 2.7t.n.r R n
teo.rev
= N" ract.
radf
Lost revolution = n1-nm
(3.15)
where nt - theoreticalrevolutionsof package nm- measuredrevolutionsof package DIA. (mm)
PRESSURE (kg)
2
2
2
1.25
1.25
1.25
0.7
Slip(%) LostRev.
1.17 26.34
1.7 38.41
2.95 66.81
1.21 27.31
1.62 36.61
1 50.44
1.58 35.64
0.7 0.7 Yarn Med Max Min Max Med Min Max Med Min (8 g) (30 g) (19 g) (8 g) (30 g) tension(g) (30 g) (19 g) (19 g) (8 g) Rev 2229.3 2222.9 2194.5 2234 2224.7 2205.2 2220 2216.1 2213.3 Radius 39.9 39.8 39.8 39.8 39.9 39.9 39.9 39.9 39.9 Density 0.4564 0.4033 0.3632 0.4323 0.3954 0.3558 0.4201 0.3847 0.325
80
Rev
120
Radius Density Slip(o) LostRev
1.75 39.54
42.34
1486 1484.3 1463.6 1489.8 1500.3 71-40- 1495.6 1492.5 1487.4 59.9 60.0 59.9 60.3 59.7 59.8 60. 60.1 60.1 0.4599 0.4207 0.3700 0.4547 0.4233 0.3451 0.4336 0.3847 0.3415 1.1 1. 2.59 T879 16.5 15.7 2.74 7.24 22.2 4.4 5.0 10.1
Rev
160
1120.1 1118.9 1102.3 112 1117.3 1109.8 1123.5 1119.4 111511 .4 79.7 79.6 79.7 80.0 80.1 80.1 79.8 79.8 .0 TW Density - 6 432 0. 0.3432 0.4-4W57 67 0.4028 0.341 .43 Slip(%) 1 1.04 2.39--- 0.66 1 0.311 Vr0.76 Radius
F , 0
Lost Rev
9.13
11.75
26.93
7.42
7.7
15.2
uensiry gtcm-, xaaius mm, revolution rev as drum count 20(X) rev
Table 3.5. the results of the slippagetest
4.32
4.2
8.
100
Slippageis usually calculatedasa percentage.The resultswere plotted as3D graphsof packagediameterand density againstslippage.There are three different graphsdrawn in termsof packagecradlepressure.In Figure 3.12, thepressureis maximum,2 kg. The maximum slippagewas found to occur whenthe packagesare soft andradiusis small. As the radius of packageincreases,the amountof slippageslightly decreases,whereas when a packageis harder,the slippagedecreasesvery much. In general,as the density of a packagerangesfrom soft to hard, the slippageis slightly reduced.For the pressure of 0.7 kg, as the radius of packagesincreases,the slippagedecreasesrapidly until the radius of 60 mm, after which there is no significant reduction. In Fig 3.14 the graph plotted with medium pressureof packagecradle.It can be noticedthat the pattern is the middle of two graphs, one is with minimum pressure,and the other is maximum. Increasingeither packagedensityor packageradiusmakea small drop on the slip. It can be said that the deformation on the contact line in axial direction betweenthe drum and the packagedependson the packageholder pressureapplied and the density of packageproduced.The radius of the packageat the contactline is smaller than the normal package radius. The small radius depends on the deformation. When the slippageis calculated,the deformedradiushasbeenconsidered.It might be thought that the explanation of the slippagedependson what happenson the contact line between packageand drum. The pressureapplied affects the area of contact of the packageon the drum. A higher pressureproduces more contact area depending on the package density. When the packageturns, the material in the areaof contactis likely to stretch. As material of the package deforms as it meets the drum during rotation, sufficient torque shouldbe appliedto the drum to supply the work of this deformation. As material so compressedleavescontactwith the drum, it recoversits normal shape and releasesthe energy earlier stored in it due to compression.When packagesjust leavethe drum, it startscovering original radius.As someof this energywill be simply dissipatedin the material againstinternal friction, the effect may contribute to somelost revolutionsof the drum.
101
PRESSURE 2 kg 30
2.5
2.0
1.5
10
0.5
D.46 4
0.0
1'o"/mTj
9D 0.34
Fig 3.12 theslippageat pressure2 kg PRESSURE 125 kg
-ý_
3.0
25
20
1.5
1.0
0.5
0.0
7o 46 o 44 042
40
om 038
50 6-0
"O'ul of
80
038 034
Fig 3.13 the slippageat pressure1.25kg PRESSURE 0.7 kg 3.0fr
2.5
20 f
15ý. ,0
05 044 0 42 040 ý. 038
r' sue' U
40
0.06
0.34
Fig 3.14 the slippage at pressure 0.7 kg
102
3.5.4.1Results of Slippage with Yarn Drag Effect With suitable weights hungon the cord on the pulley the sametests that were donefor slippage were repeated.The counts of revolutions of the package were saved and tabulatedin Table 3.6. As mentionedbefore the weight hung was so as to producethe sameeffect asthe yarn tensionappliedduring the winding of a particular package.
DIA. (mm)
80
120
160
PRESSURE (kg)
2
Yarn tension(g) Rev Radius
Max (30 g) 2230 39.9
2
2
Med Min (19 g) (8 g) 2224.2 2196.1 39.8 39.8
1.25 Max (30 g) 2233 39.8
1.25
1.25
Med Min (19 g) (8 g) 2224.3 2206.1 39.9 39.9
0.7 Max (30 g) 2214 39.9
0.7
0.7
Med Min (19 g) (8 g) 2211.1 2212.2 39.9 39.9
Slip(%)
1.14
1.64
2.88
1.25
1.64
2.2
1.85
1.97
1.93
Lost Rev.
25.64
37.11
65.21
28.31
37.01
49.54
41.64
44.54
43.44
Rev Radius
1486.5 1484.7 1464.4 1490.3 1500.9 1483.6 1494.9 1492.3 1487.6 59.9 60.0 59.9 60.3 59.7 59.8 60.0 60.1 60.1
Slip(%)
1.07
1.02
2.54
0.15
0.44
1.42
0.34
0.35
0.66
Lost Rev
16
15.3
38.1
2.24
6.64
21.42
5.1
5.2
9.9
Rev Radius
1120.4 1119.3 1102.9 1120.6 1117,5 1110.2 1123.6 1119.6 1115.4 79.6 79.7 79.8 80.0 79.7 80.0 79.8 80.1 80.1
Slip(%)
0.78
1.0
2.33
0.64
0.67
1.32
0.37
Lost Rev
8.83
11.35
26.33
7.22
7.5
14.8
4.22
.36
0.7
4
8.2
Table 3.6 The resultsof slippagetestwith hungweight 3.5.5 Comparison of the two experiments
DIA. (mm)
80
120
PRESSURE (kg)
2
2
2
1.25
1.25
1.25
0.7
0.7
0.7
Yarn tension(g)
Max (30 g)
Med (19 g)
Min (8 g)
Max (30 g)
Med (19 g)
in (8 g)
Max (30 )
Med (19 g)
Min (8 )
Rev
2229.3 2222.9 2194.5 2234
160
2216.1 2213.3 2211.1 2212.2
Rev Drag Rev
1486 1484.3 1463.6 1489.8 1500.3 1483 1495.6 1492.5 14 7.4 1486.5 1484.7 1464.4 1490.3 1500.9 1483.6 1494.9 1492.3 14 7.6
0.4
0.
2233
2224.3 2206.1
2230 0.7
Subtracted 0.5
2224.2 2196.1 1.3 1.6
2224.7 2205.2 2220
Drag Rev Subtracted
.1
0.5
-0.4
0.9
2214
.6
.5
.6
1120.1 1118.9 1102.3 1120.4 1117.3 1109.8 1123.5 1119.4 1115.1 1120.4 1119.3 1102.9 1120.6 1117.5 1110.2 11 .6 1119.6 111 .4 1 0.1 1 0.2 Subtracted 0.3 0.4 0.6 0.2 1 0.2 1 0.4 Rev Drag Rev
Table3.7The resultsof comparingtwo testsof slippage
103
xz -
(ObservedFreq.- ExpectedFreq.)' ExpectedFreq.
Effect of drag on revolutionsof packageat 120 mm Observed
Expected
frequencies
frequencies
Fo
F,,
Press.2 /ten. 30 g
1486.5
1486.94
Press.2 /ten. 19 g
1484.7
1486.94
Press.2 /ten. 8g
1464.4
1486.94
Press.1.25/ten. 30 g
1490.3
Press.1.25 /ten. 19 g
Deviations
Fo-f
(Fö fe)2/fe
-0.444
1.32578E-4
-2.244
0.00339 0.3418
1486.94
-22.544 3.356
0.00757
1500.9
1486.94
13.956
0.13099
Press.1.25 /ten. 8g
1483.6
1486.94
0.00752
Press.0.7 /ten. 30 g
1494.9
1486.94
-3.344 7.956
Press.0.7 /ten. 19 g
1492.3
1486.94
5.356
0.01929
Press.0.7 /ten. 8g
1487.6
1486.94
0.656
2.8941E-4
2.704
0.5535
13385.2
Total
13382.496
0.04257
Observedfrequenciesare the revolutionsof packagewith drag effect (from Table 3.6) "
Expectedfrequenciesare the mean of the revolution of packagewithout drag effect (from Table 3.5)
X9,0.05 =16.92
0.5535«
x9,o.os
So the drag effect doesnot causeany significant changeof slippage.
104
Effect of dragon revolutionsof packageat 80 mm Observed
Expected
frequencies
frequencies
Fo
Fe
Press.2 /ten. 30 g
2230
2217.77
12.23
0.06744
Press.2 /ten. 19 g
2224.2
2217.77
6.43
0.01864
Press.2 /ten. 8g
2196.1
2217.77
0.21174
Press.1.25 /ten. 30 g
2233
2217.77
-21.67 15.23
Press.1.25 /ten. 19 g
2224.3
2217.77
6.53
0.01923
Press.1.25 /ten. 8g
2206.1
2217.77
Press.0.7 /ten. 30 g
2214
2217.77
Press.0.7 /ten. 19 g
2211.1
2217.77
Press.0.7/ten.8g
2212.2
2217.77
19951
19959.93
Total
Deviations
(Fý fe)2/fe
Fo-fe
-11.67
0.10459
0.06141
-3.77
0.00641
-6.67
0.02006
-5.57 -8.93
0.01399 0.52351
Effect of drag on revolutions of package at 160 mm Observed
Expected
frequencies
frequencies
Fo
F,
Press.2 /ten. 30 g
1120.4
1116.31
4.09
0.01499
Press.2 /ten. 19 g
1119.3
1116.31
2.99
0.00801
Press.2 /ten. 8g
1102.9
1116.31
Press.1.25/ten. 30 g
1120.6
Press.1.25 /ten. 19 g
Deviations Fö fý
(Fý fß)2/fý
0.16109
1116.31
-13.41 4.29
0.01649
1117.5
1116.31
1.19
0.00127
Press.1.25/ten. 8g
1110.2
1116.31
-6.11
0.03344
Press.0.7/ten.30 g
1123.6
1116.31
7.29
0.04761
Press.0.7 /ten. 19 g
1119.6
1116.31
3.29
0.0097
Press.0.7 /ten. 8g
1115.4
1116.31
-0.91
7.41819E-4
10049.5
10046.79
2.71
0.29334
Total
105
3.5.6 Discussionof Slippage There are a few references about the slippage of cheese packages in winding. One of these was published by Osawa and Koyama [Osawa 72]. Their experiment is somewhat different from the experiment performed in this project as some of the settings used were unrealistically high. Their result was that slippage increases when increasing pressure or yarn drag is applied to a cheese package. In this project, the conditions of winding were the same as those normally used in the textile industry. Packages were produced under limited values of tension and pressure and slippage determined.
3.6 Deformation of a wound packageat the contact point
3.6.1 Introduction The deformation occurs at the contactpoint between the drum and the packagedue to packagecradle pressureand the own weight of the package.As mentionedbefore, the drum is hard and is not deformable.The packageis relatively soft and is non-isotropic due to the yarn. Besides,the surface of the packageconsistsof yarn laid at a constant wind angle.The winding createsspacesbetweensuccessiveturns of yarn wound. 3.6.2The Method of Measuring PackageDeformation Experimentswere conductedat different diameters,pressureand yarn tensionvaluesin order to determine package deformation at the point of contact with the drum. The LVDT, located on the packagecradle gives the deformedradius of the package;the POT which is located on the top of the packagegives the non-deformedradius of the package.The differenceof the two readingsis thereforethe amountof deformation. However, on account of the fact that the wound packageis not quite cylindrical, the deformation showsvariation at different points on the package. Therefore it was decided to measuredeformation while the packageis rotated on the winder, and the average of a sufficient number of such readings will give a better estimateof deformation.
106
However,it was found that the readingfrom the POT had considerablejitter due to the light cylinder connectedto its armjumped excessivelyon the packagewhile the winder was operated.Another solution was neededto measurethe undeformedradius of the package. The solution adopted was to use a screw and nut arrangementto gently raise the packagearm till the packagejust cleared the drum surface.As this point the LVDT should give the un-deformedradius of the package. First, the packagewas allowed to rotatenormally, and the LVDT was read to obtain its deformedradiusas describedin the previousparagraph.After that, the screw was locatedbetweenthe flat surfaceunder packagecradleand the packagecradlearm. The nut on the screwwas turned slowly by handto lift the packagearm while running the winder. The point of stoppingwas when the packageonly just touchedthe drum. At this point, the packagespeedwas almost zero. The reading at that point from the LVDT gives the un-deformed radius of the package.The difference of the two reading gave the deformation of the package.A programin C languagewas written to follow the LVDT readingon the monitor for both deformedradius and un-deformedradius.The programgives the averagereadingof 10 revolution of thepackage[Appendix C.4]. 3.6.3 Results of deformation measurement The reading of the LVDT was obtained for different values of package diameter, pressure between drum and package, and winding tension. The calculation of deformation was done and tabulatedbelow in Table 3.8. More information about the resultsof deformationis given in Appendix B. The results are shownby Fig 3.15,3.16, 3.17.
The amount of deformation dependson the package cradle pressureas mentioned before. At thesamepressure,the deformationdependson the packageradius.Generally, increasingpackageradius causedthe greater deformation contact point as shown by
eachgraph. At the pressureof 1.25 kg as in Fig 3.16, the deformation generally shows a decrease with increasingdensity, and an increasewith increasingpackageradius.The behaviour
107
is similar at the lower pressure of 0.7 kg, but is more complex at the higher pressure of 2 kg.
DIA. (mm)
PRESSURE (kg) Yarn tension (g)
2
2
2
1.25
1.25
1.25
0.7
0.7
0.7
Max (30 g)
Med (19 g)
Min (8 g)
Max (30 g)
Med (19 g)
Min (8 g)
Max (30 g)
Med (19 g)
Min (8 g)
80
Radius Density Def
39.9 39.8 39.8 39.8 39.9 39.9 39.9 39.9 39.9 0.4564 0.4033 0.3632 0.4323 0,3954 0.3558 0.4201 0.3847 0.3325 0.32 0.44 1.35 0.34 0.43 0.75 0.18 0.18 0.4
120
Radius Density
59.9 60.0 59.9 60.3 59.7 59.8 60.0 60.1 60.1 0.4599 0.4207 0.3700 0.4547 0.4233 0.3451 0.4336 0.3847 0.3415
160
0
Def
0.9
Radius Density
79.7 0.452
Def
0.8
1.02
1.67
0.45
0.57
1.14
0.28
0.37
0.55
79.6 79.7 79.8 80.0 80.0 79.8 80.1 80.1 0.4060 0.3432 0.4405 0.4028 0.3448 0.4367 0.4009 0.3461
1.27
2.2
0.41
0.76
1.11
0.53
0.53
0.79
Density [g/cm3],Radius[mm], deformation[mm]
Table 3.8. Deformationof packages 3.7 Discussion The deformationof the packageon the winding drum is of interest in randomwinding. The amountof compressionwhich dependson the basicwinding parametersasgiven in the previous sectionwill havea direct influenceon the speedof packagerotation. When
with increaseof packagecradlepressure.It otherthingsareconstant,slippageincreases can be seen for example by comparing Fig.3.14 and 3.17 that while the amount of
increases compression with increasingpackageradius,otherthingsbeingconstant,the slippageshows a decrease.This could be partly due to the increaseof surface area of contact for the same amount of compression(depth of penetration) as the package
radiusincreases.
108
Fig 3.15 Deformation of package at pressure 2 kg
2.5
2.0 1.5 tic
0.'.
ons 0.44
0 42 10
0
_., 490 ["j
80
034
Fig 3.16 Deformationof packageat pressure1.25kg PRESSURE 0.7 kg 2.5 'T-
20
I
1.5
11.0 0
o44 042 040 o.3e
0
0.36 .. ,y-ke9O
Imml
80
Fig 3.17 Deformation of package at pressure 0.7 kg
0.34
.ate
109
CHAPTER FOUR
RIBBONING ON THE RANDOM WINDER
4.1 Introduction A randomwinder tendsto produce ribboning or patterning at certain diametersof the packagebeing wound. The ribboning causesthe packageto deviatefrom the desired shapeas well causeunwinding difficulties during yarn withdrawal from the package. Where packagesare dyed it causesthe packageto dye non-uniformly. In this chapter, ribboning will be defined and controlling the random winding operationby the use of electronicswill be explained
4.2 Definition of Ribboning Ribboninghasbeendefined asthe formation of visible bandsof yarn by the overlapping or very close winding of successiveturns of yarn that tends to occur as the package diameterattains certainvalues.So if a double traverseof yarn begins at a certain point on the packagebeing wound; and if a number of successivedouble traversesof yarn begin at the samepoint on the package,this leadsto the formation of a visible bandor ribbon of yarn. As explained before, ribboning is most intense( major ribboning)at points where the packagereachesinteger values of wind per double traverse.It persists longest at the lower integer values of it as permitted by the maximum packagesize wound. This is becausethe rate of changeof packagediameteris less at theselower valuesof wind per doubletraverse.
Wherethe traverseratio differs from an integerby suchvaluesas0.5 and0.333 etc, ribboningstill occurs,but in this caseit is lessintensecomparedto majorribboning.
110
4.3 Analysis of ribboning on the random winder
4.3.1Configuration of experimental winder Fig 4.1 shows the configuration of the experimental apparatus.There are two shaft encoderslocatedon the winder. Oneof theseis attachedto the drum shaftof the winder. The other is mountedon the packageshaft at the sameside of the winder. Both were connectedto interfacecard via junction box so that their angularposition can be read as an 8-bit digital numberby the computer.The cablesfrom the shaft encodersare leadto a pair of D-type connectors,to facilitate checkingof all the signals provided by them. The 8 bit angularposition readingis fed to the PC interfacecard usingscreenedwire. A programwas written in the C languageto readboth shaft encoders. The constant speedmotor provided on the winder was replacedwith a DC brushless motor [Electro-craft] whosespeedcan be directly controlled by an externalDC voltage. The motor is capable of any speed from 0-3000 rpm, as well as very rapid speed fluctuation under external control. More information about the shaft encodersand the motor was given in Chapter2. The function of the pulley shownis coveredon page96.
lunctionbox II 14,
ShaftEncoderI
f'-fýOT fPackage ~
ShaftEncoderII
functionboxI
GroovedDrum
LVDT fey
Motes seryodriye
Fig 4.1Configurationof thewinder
111
4.3.2The method of following yarn on the drum and the package On the experimental winder the shaft encoder on the drum was set to read to zero when the yarn is at the left hand end of the groove. As the grooved drum has a drum constant of 3, by reading the package shaft encoder reading at every third zero crossing of the drum, a convenient reference is obtained to identify the beginning of a double traverse.
4.3.2.1The program for reading shaft encoderon the drum The programwritten for this purposeenablesthe PC to read the shaft encoderwhen the drum is madeto rotateat the requiredspeed.It writes the 0-255 value of shaft position on the VDU. At any constantspeedof the drum, eachvalue of shaft position should be written very closely the same number of times, for one rotation of the drum. For instance,the numberof readingbetween 1 and 2 should be as same as the numberof readingbetween244 and 245. The operation of both shaft encoderswas painstakingly checkedfor correct operation, before experimentswere started.When testedas above,the incrementalshaft encoder was quite well behaved,but the absoluteshaft encoderoccasionally provided solitary rogue values. There were two possible causesof this. One was electrical noise, as described in Chapter 2. The other was the intrinsic problem with absolute shaft encoders,in that if a reading was taken at a time one or more bits were changing,a faulty readingcould be obtained.This problem was far less severewith the incremental encoder,as the error was no more than one least count. As confirmed by observations over a number of hours, the spurious values provided by the absoluteencoder were always single, and these were comparatively rare in the readings obtained in each revolution.
4.3.2.2Experimentsto observeribbon formation in winding The observationor detectionof ribboningis basedon the following procedure.Each time a new double traversebegins,the readingof the packageshaft encoderis taken.As this procedureis repeated,thereadingof the packageshaft encoderis comparedwith its previous reading. If the difference is zero or very close to zero, this indicates major
112
ribboning. If the difference is 180°, 90°(or 270°), 45°(or 315°) etc. diminishing magnitudesof minor ribboning is indicated. Since the beginning of a double traversewas chosenas when the drum shaft encoder readszero, it is first necessaryto detectwhen the grooved drum is at this position, in every 3 revolutions (as the drum constantis 3). Since the readingof 0 from this shaft encoderwill last in the duration betweenangle readingof 255 and 1, it was arbitrarily chosento detectthe transition from 0 to 1 as the reference,so that the packageshaft encoderreadingcan be obtainedwith the bestaccuracy.However since the drum shaft encoderinvolves the occasionalroguenumbers,a procedureis necessaryto avoid any difficulties that canbe causedby suchnumbers. The programwas modified to be able to eliminate such false values.Two methodsof achievingthis were tried. The first methodis to usetwo stagesof nestedloops asshown by Fig 4.2. The routine is not disturbedby false numbers,and exits only when a reliable transitionof the readingfrom 0 to 1 occurs. The secondmethod consistsof taking 4 successivereadings,and comparing them to check that they are all the same.This is repeatedtill the readingof 0 is achieved.The routine then waits till the reading of 1 has reliably arrived. Both these methodswere tried andwere found to work reliably. The programis given in Appendix C.5. 4.3.3 Graphic representation of progressof winding The yarn starts winding at the left hand side of the drum, and goesback to the same point, after the drum has made 3 revolutions. Ribboning is expectedeach time the packageshaft encoderrecordedthe samereading over a number of successivearrivals of thedrum at the beginningof a doubletraverse.Thereforeit appeareduseful to plot on the VDU the value of the packageshaft encoder,each time a new double traverse began.
113
main program Read address 772 a=inportb(772)
O
a=0 yes Read address 772 a=inportb(772) no
a=0
dyes return to main program Note:a-a variable
(a)
main program x=0 Read address 772 a[x]=inportb(772)
x=x+1 yes
x5 yes
nb. O
no read diameter
of package
Rib=0
Check nbbonmp new value for nb
yee
no
,..
nD-S
Duration of lowering am, oll switch is off
yes
no
Y.. (ADt k..
k=3
y8s
knO
no
-
yes
--
m stars moving Gaü the army
no
oreeklnp by Mang package
5
Return to main programme
stop the arm Dark moving stop_stepl°y no-e
yet
nW4
yea
'01
no
roubne for nbbon
RID-0
is switch ofti different than
-'
252
References [1] Rebsamen, A. Modern PackageBuilding. Textile Asia 1988; September, 130-136. [2] Durur, G., PhD thesis( In preparation), University of Leeds, U.K., 2000
[3] DC BrushlessMotor DM-20 andBRU-200 Servodrive,Electro-Craft Ltd., Fourth Avenue,Crewe,U.K. [4] Osawa,G. and Koyama,T. On the SlippagebetweenYarn-Packageand Grooved Drum During Winding. Journal of the Textile Machinery Society of Japan, 1972;25(6) T.113-123. [5] Nobauer, H. and Baumeler,M. Measurementof unwinding resistanceof yarn packagesusing the PackagePerformanceAnalyser. Textil Praxis International 1992;November,1034-1040.
253
AppendixE3
"The useof the Angle of DoubleTraversediagramto study packageformation in randomwinding andthe effectivenessof different techniquesof ribbon breaking" P. Bandara,G. Durur
PaperID: M2000-156
InternationalConference`Mechatronics2000', 6-8- September2000,Atlanta, Georgia, USA ( Acceptedfor presentationand publication in proceedings)
254
ACTIVE RIBBON BREAKING IN RANDOM WINDING G. Durur, P. Bandara Schoolof TextileIndustries,University of Leeds,LeedsLS2 9JT, U.K
Abstract: Ribboning is a basic problem in randomwinding of yarn packagesby the grooved drum method. Severalmethodsof 'ribbon breaking' are used on commercial winders, but their effectivenessis rather limited. 'Active' ribbon breaking can be achieved,under programmedcontrol, by initiating corrective action at times when ribboning actually occurs, thereby providing better unwinding characteristicsto the wound package. Unwinding testscarried out on packagespreparedby this method confirmed thegreatereffectivenessof active ribbon breaking.
1. INTRODUCTION Random winding is widely used for preparing yarn packagesrequired in a variety of textile processes,particularly in weaving and knitting. Such a yarn packagenormally has the form of a cheese(cylinder) or a cone,and has a single yarn that is cross-woundso as to producea stable and compact body. In use, the package is held stationary,and the yarn is withdrawn by pulling it over one end of the package.In a well prepared package,the yarn should unwind with a uniform unwinding tension. On a random winder, shown schematically by Fig. 1, a cylindrical drum with an endlesshelical groove is usedto traversethe yarn over the width of the package, and the package is rotated by contactwith thesurfaceof the drum.
POT
LDT
om. ma«
Yom (End View)
Fig. 1 Configuration of the Apparatus As the drum is rotated at a constantspeed,this ensures a constant yarn winding speed, up to several hundred meters/sec. As the package
gradually builds up during winding, the wind ratio, which is the number of packagerotations per double traverse of yam, gradually decreases and passes through several integer values (Rebsamen,1988). Eachtime theratio reachesan integer, successiveturns of yarn are laid on one another,and the effect persiststill the ratio moves away from the integer value as the package diameter gradually increases. This bunching of yarn turns is known as ribboning, and is undesirabledue to the distortion of packageshape it causes,as well as the greatertension that needs to be appliedto the yarn in over-endunwinding at thesepoints. There are several 'ribbon breaking' techniques used in random winding. Each of these causes someabrupt slippageof the packageon the drum, once every few seconds during winding. This helps to disperse the yarn turns somewhat, and thereby minimise ribboning. However as the action is 'passive', in that the moments of its application are not coincident with those when ribboning actually occurs. The effectiveness of ribbon breaking is thereforenot optimum. Winders are available now which use speed controlled motors under microprocessorcontrol, which prevent ribboning by switching traverse rates so as to avoid the integer values of wind ratio. However, the grooveddrum randomwinder is still widely used by the industry due to its simplicity, and standard methods of ribbon avoidanceare still applied on them.
255 2. OBJECTIVES The research reported here was intended to devise a means of improving the effectiveness of ribbon breaking on a standard random winder by applying such action precisely at the times when ribboning actually occurred during winding. For this reason, it was necessary to provide the winder with the capability to detect the onset of ribboning whenever it occurred.
3. EXPERIMENTAL The project (Durur, 2000) used a commercial winding head modified by adding a shaft encoder (8 hit) on the winding drum as well as on the package spindle. The basic winder incorporates a mechanical ribbon breaking mechanism, which briefly decouples the drum from the drive several times a minute. At each re-engagement of the drive some slippage is caused between the drum and the package as required. Two additional methods of ribbon breaking were introduced on the winder, One of these was implemented by replacing the standard drive motor by a DC hrushless motor (Electro-Craft), which can be driven so as to produce sharp changes of drum speed. The other was to introduce a stepper motor controlled package lifting arm, which could be used to lift the package minutely off the drum and back at appropriate moments. Potentiometric sensors were added to measure the package diameter and the amount of package compression against the winding drum as winding progressed. All the above devices were interfaced to a PC for control and data acquisition purposes. These modifications resulted in a random winder that offered considerable scope for its mechatronic control. The use of the shaft encoders proved somewhat difficult due to the pick up of electrical noise by electronic circuitry and measures were taken to minimise the problem. It was found that errors due to this problem could be eliminated by comparing four successive readings from the shaft encoders. This eliminated 'rogue' reading due to noise, as well as any erroneous readings that could occur as the shaft encoder stepped from one reading to the next. At a yarn winding speed of 450 m/min, each double traverse of the yarn took about 140 mS. The reading of the shaft encoders by the PC, even with error checking, was comparatively much faster at 1.7µS per reading and allowed the above procedure to be satisfactory. The PC was programmed using the C language. The apparatus was first used to study the random winding process. It was quickly realised that by
plotting the angular position of the package spindle at every start of a double traverse, an Angle of Double Traverse (ADT) diagram could be constructed on the VDU of the PC which was found to be very useful in following the progress of winding. It was seen that by rotating a yarn package of constant diameter (i. e. a package on which no additional yarn is wound) at the normal running speed, a very regular dot pattern is shown by the ADT (Fig. 2). The inclination of the 'lines' depends on the particular wind ratio that corresponds to the package diameter. It was clear that despite the friction drive to the package, the package rotation was quite uniform, with no suggestion of random slipping. However as indicated by the measured diameter of the package and its deformation on the winding drum, a minute slippage of constant magnitude is likely to occur.
.............
Fig. 2
Rotation of constant diameter package with no drag ( diameter :80 mm, wind ratio: 3.4)
This may be mechanical slippage, or could be the result of the package surface at the point of drive being slightly stretched due to its elasticity (Osawa and Koyama, 1972). Even with some mechanical drag introduced on the package spindle, corresponding to the tension of yarn during winding, the figure remained unchanged. When winding a package normally i. e. with its diameter gradually increasing, but with ribbon breaking measures disabled, the slope of the lines gradually changes, as shown by Fig. 3.
"S
."
"..
".....
.
..
"" ..
.""
"". ""
.. "
""""
J" ":
"""...:
"
"
"". "". "
"
:" "
" .
."
"
:
" ::.
'.
..
".. ".
"
'. .
... ... """
.:.
.. """..
.;
"
.
:.
"S,
.
"
"""...
, ".
"ý ...
"ý
55
Fig. 3 Winding of package at wind ratio 3.3
256
Depending on the actual value of wind ratio reached, the lines form a single peak or several of them. Where a single peak is seen, the wind ratio has an integer value at that point, where 3 peaks are seen the wind ratio is an integer plus 0.33 etc. When the ratio is an integer, the ribboning that occurs is most undesirable (major ribboning). Ribbon breaking is mostly needed at these intervals. Where the ratio is high, and not an integer, any ribboning (minor ribboning) that occurs quickly passes away. When the wind ratio descends to lower values such as 3 and 2, even minor ribboning persists longer.
The main purpose of the modifications made to the winder was to be able to apply the ribbon breaking action at moments of ribboning, so as to maximise the beneficial effect. As described above this could be done by programming the PC to detect the onset of ribboning, and then apply the ribbon breaking action continuously, till no more ribboning is detected. Basically this action can be implemented by detecting the beginning of each double traverse, for example at the left hand end of the grooved drum, and checking the corresponding package angular position for coincidence within a small tolerance. This procedure can be extended to detect not only major ribboning as above, but also any minor ribboning as necessary. Each time such ribboning is detected, the programme will initiate the ribbon breaking procedure.
lt
is really the major ribboning that cause significant difficulties with unwinding, and hence those that demand dispersal. The moments of major ribboning can be detected by software, and ribbon breaking can be effected 'actively' as compared with the passive ribbon breaking normally employed.
.". ..
Fig 5 shows the effect of package slippage caused by introducing speed fluctuation of the drive motor itself, which is seen to be superior to the effect produced by the standard procedure provided on the winder.
"...
It was realised however that the most effective method of ribbon breaking is to lift the package at brief intervals during ribboning. This achieved the best ribbon dispersal as shown by Fig. 6. The package lifting arm acted on the hub of the package arm, so as to minutely lift the package off the drum. This action slows the package down, thereby causing the wind ratio to drop below the value causing ribboning. The arm uses a microswitch to sense the lifting action, and the shaft encoders help to control the lift and the duration of lift to the best effect. The microswitch also helps to keep the arm following the increasing diameter of the package. The flowchart of actions that were programmed, for ribbon breaking by this method is given in the Appendix. The routine shown additionally also plots the ADT diagram.
j
"..;
Fig. 4 Effect of normal ribbon breaking at wind ratio 3. Fig. 4 shows the effectiveness of ribbon breaking by using the standard method of disconnecting drive to the drum approximately every 2 seconds, provided on the basic winder. The method diminishes ribboning to a noticeable extent, but some remnant ribboning could be seen on the package.
:ý
i"
"
ýi .)v .. .., "
.; il
;.....
r ro 'O
Fig. 5 Effect of active ribbon breaking at wind ratio 3
.4.0
."'
W
Fig. 6 The results of lifting the package at wind ratio: 2.
257 However the best proof of the effectiveness of ribbon breaking is obtained by comparing unwinding tension relating to each method (Nohauer and Baumeler, 1992). A number of packages were wound under the same winding conditions, but using each of the above 3 methods of ribbon breaking, and were evaluated as for evenness of unwinding tension. Tension was monitored during the unwinding of each package, by withdrawing yarn at 480 m/min. A tension probe having a natural frequency of 2.5 kHz was used and tension readings were recorded as 8 hit values, to provide some 1M readings per package. The results of each test were used to construct a histogram of withdrawal tension. These are compared in Fig. 7. The results show that active ribbon breaking using package lifting gave the lowest unwinding tensions and hence the best unwinding characteristics. Yrnn
n V.!
mit32
un.. n U'
The method of lifting the package minutely off the drum was found to be the most effective for ribbon breaking, when used 'actively' as described. However this method is more difficult to implement due to the increasing package diameter as winding progresses. In this work an arm attached to a geared stepper motor was used to implement the procedure, but a simpler mechatronic solution is desirable.
REFERENCES
Durur, G., (2000), PhD thesis (In Preparation), University of Leeds, U. K.
ýýen20
zýYln
ý+nyirWnMnhrtak -g-I nlhnak rTu)1edmror
"n 1-mim-
relatively low in cost, methods such as the one reported here will prove to be useful in producing better random wound yarn packages.
$nm
,
Electro-Craft Ltd., DC Brushless Motor DM-20 and BRU-200 Servodrive, Fourth Avenue, Crewe, U. K.
rnn mrn vrrn e Ja. 2
Al
4$
Fig. 7 The results of unwinding yarn tension
Nobauer, H Baumeler, M., (1992), and Measurement of unwinding resistance of Package the yarn packages using Performance Analyser, Textil Praxis International, November, pages 1034-1040 Osawa, G. and T. Koyama, (1972), On the Slippage between Yarn-Package and Grooved Drum During Winding, Journal of the Textile Machinery Society of Japan, 25, No: 6, T. 113-123.
4. CONCLUSIONS The ADT diagram is a useful means for studying the random winding process. It showed that no irregular slipping takes place in random winding under normal yarn winding tension settings, despite the indirect (surface) drive to the package employed. It was particularly useful in studying the effectiveness of the 3 ribbon breaking methods studied in this work. As the standard ribbon breaking measures provided on random winders are applied at constant time intervals, it was felt that 'active' ribbon breaking as described above would prove to he more effective. This was in fact found to be the case by comparing the unwinding tension measured on yarn packages prepared by the different methods. The standard random winder will continue to he used by the textile industry due to its simplicity and its suitability for use in automatic winders. As the use of microprocessor control has become
Rebsamen, A., (1988), Modern Package Building, Textile Asia, September, pages 130-136,
APPENDIX Key to Symbols used in flowchart: drum rotations within double traverse (0-2) y: number of double traverses produced l no. of double traverses for diagram (0-500) rib : variable to indicate stage of ribbon breaking (0= no action required) ( 1-6 = stages of arm movement) lift_step, stop_step: time duration of arm movements expressed in double traverses k
258 MAIN PROGRAM FOR DETECTING OF RIBBONING AND DISPLAY ON VDU
ROUTINE FOR RIBBON BREAKING BY LIFTING PACKAGE From main programme
Start Duration of arm top point
set drum byte (man-255)
rib-3 no
yes
preparethe graphs on the monitor nb"4
Y-stop_stepvS yes
k=0 (grooves turn) y=O(no of DT)
no
END
Y-17000 no yea
nb=2
t-0
Duration of lifting arm yes
no 0500
clear view on he monitor
no
y-lill_step>3
yes
yes
slop lifting the arm
no read package diameterfrom
LVOT
stop_5tep"y
rib"3 Rib=O
yes
-
check arm position
no find the beginning of DT (involves
reading 4 SEI
nb-I
values)
yes
- ---y
no
start lifting the arm ý-
-
tth
no
stepay
k=0
Yes Duration of package on the drum
-
read package SE2
-
nb-S
yes
-
--
--
--
-
no plot dot on the monitor for yam
position y-stop_steplo5 yes
rib-0
no read diameter
of package
Rib=0
yes
Check ribboning, new value for nb
na
rib-5
yes
Duration -
of lowering
to, y.
stop the arm moving back
no
stop-step I -Y nb. 6
yes
k_0
no nb-4
routine for ribbon Rib=O
yes
is off
is switch off? (AD1 different than 'Or) yes
k""
k=3
amt till switch
no
no
breaking
yes
stars moving
back the arm
no
by IINng
rib-5
package
_ Return to main programme
