INFORMATION TO USERS
This manuscript has been reproduced from the microfilm master. UMI films the text directly from the original or copy submitted. Thus, some thesis and dissertation copies are in typewriter face, while others may be from any type of computer printer.
The quality of this reproduction is dependent upon the quality of the copy submitted. Broken or indistinct print, colored or poor quality illustrations and photographs, print bleed through, substandard margins, and improper alignment can adversely affect reproduction.
In the unlikely event that the author did not send UMI a complete manuscript and there are missing pages, these will be noted. Also, if unauthorized copyright material had to be removed, a note will indicate the deletion.
Oversize materials (e.g., maps, drawings, charts) are reproduced by sectioning the original, beginning at the upper left-hand comer and continuing from left to right in equal sections with small overlaps. Each original is also photographed in one exposure and is included in reduced form at the back of the book.
Photographs included in the original manuscript have been reproduced xerographically in this copy. Higher quality 6" x 9" black and white photographic prints are available for any photographs or illustrations appearing in this copy for an additional charge. Contact UMI directly to order.
A Bell & Howell Information Company 300 North Z eeb Road. Ann Arbor. Ml 48106-1346 USA 313.'761-4700 800/521-0600
INFRARED HEATING AND WELDING
OF
THERMOPLASTICS AND COMPOSITES
DISSERTATION
Presented in Partial Fulfillment of the Requirements for
the Degree Doctor of Philosophy in the Graduate
School of the Ohio State University
By
Yang Shiau Chen, B.S., M.S.
The Ohio State University
1995
Dissertation Committee Approved by
Avraham Benatar
Chon L. Tsai tjun\ Ly James Lee
Adviser
Welding Engineering Graduate Program DMI Number: 9544534
UMI Microform 9544534 Copyright 1995, by UMI Company. All rights reserved.
This microform edition is protected against unauthorized copying under Title 17, United States Code.
UMI 300 North Zeeb Road Ann Arbor, MI 48103 To God and My Parents
I ACKNOWLEDGMENTS
I wish to express my appreciation to my adviser Dr. Avraham Benatar, for helping me to develop this research. His guidance has been invaluable in my years at The Ohio State University. It was a great experience and an honor to work with someone who gave me the opportunity to get involved in the plastics joining field. I would also like to thank the members of my committee, Professor Chon L. Tsai, for his constant encouragement and guidance during the course of this study, and Professor James Lee, who gave me lots of advice on polymers and helped me use the equipment in the polymer laboratory in chemical engineering. I would like to thank
Professor David A. Rigney for his suggestions and comments. Many thanks go to the National Science Foundation and Edison Welding Institute for financial assistance.
Mr. Stefan N. Fodor deserves thanks for being a good friend during these difficult months of writing and rewriting. I also wish to thank Dr. K.
Bolland for his friendship and for reviewing my dissertation. Special thanks go to my parents for their love and their faith in me
during my education. Love and thanks also go to my wonderful wife, Pi
Hsiang Kao for her unwavering love and support. Love my sweet sons,
Anthony and Andrew. VITA
October 13, 1960...... Born in Taipei, Taiwan, R.O.C.
1985 ...... B.S., Mechanical Engineering, R.O.C.
1987 ...... M.S., Mechanical Engineering, R.O.C.
1987-1989 ...... Second Lieutenant, Chinese Army Corp, R.O.C.
1990...... Graduate Research Associate, The Ohio State
University
FIELDS OF STUDY
Major Fields: Welding Engineering
Research in designing, manufacturing, and processing of polymer and
polymeric composites, quantitative process modeling, simulation and
control.
IV TABLE OF CONTENTS
DEDICATION...... I ACKNOWLEDGMENT...... II VITA...... Ill LIST OF FIGURES...... VIII LIST OF TABLES...... XVI CHAPTER PAGE I. INTRODUCTION...... 1 1.1 Introduction ...... 1 1.2 Description of IR Welding ...... 3 1.3 Mechanism and Application ofIR ...... 4 1.4 Literature Review ...... 8 1.5 Objectives of This Work ...... 10 II. IR WELDING SETUP...... 13 2.1 Introduction ...... 13 2.2 High Intensity IR Modules with Power Controller ...... 16 2.2.1 High intensity IR heater ...... 16 2.2.2 Power controller ...... 18 2.3 Two Independent Air Cylinder Setups ...... 18 2.4 A Process Controller System ...... 19 2.5 Lamp Cooling Setup...... 20 2.6 Foundation and Fixture ...... 20 III MODELING AND MEASUREMENT OF RADIATION FIELDS...... 23 3.1 Introduction ...... 23 3.2 Lamp-U-Shape Reflector System...... 24 3.3 Mathematical Model of Radiation Field ...... 27 3.3.1 Direction radiation ...... 27 3.3.2 Indirect radiation ...... 31 3.3.3 Total Radiation ...... 37 v 3.4 Experimental Setup ...... 38 3.5 Results and Discussion ...... 39 3.6 Summary...... 42 IV. MODELING AND MEASUREMENT OF RADIANT ENERGY DISTRIBUTION...... 45 4.1 Introduction ...... 45 4.2 Parameter Estimates...... 47 4.2.1 Effects of wavelength, output power, and applied voltage ...... 47 4.2.2 Effect of heating distance ...... 49 4.3 Normalization of Parameters ...... 53 4.3.1 Statistical independence of the applied voltage or dial scale ...... 55 4.3.2 Statistical independence of the heating position (X, Y, Z) ...... 58 4.4 Discussion ...... 68 4.4.1 Effect of applied voltage ...... 68 4.4.2 Effect of the heating position (X, Y, Z) ...... 69 4.5 Summary...... 70 V. HEAT TRANSFER MODELING OF IR WELDING...... 78 5.1 Introduction ...... ;...... 78 5.2 Background ...... 79 5.3 Penetration Depth ...... 82 5.4 Evaluation of Reflection Energy ...... 84 5.5 Theory Considerations ...... 87 5.6 Measurement of Melting Zone ...... 97 5.7 Temperature Measurement ...... 97 5.8 Results and Discussion ...... 100 5.9 Summary...... 103 VI. IR WELDING OF POLYPROPYLENE...... 110 6.1 Introduction ...... 110 6.2 Welding Procedures ...... 112 6.3 Mechanical Testing ...... 113 6.4 Microstructural Examination ...... 113 6.5 Results and Discussion ...... 116 6.5.1 Mechanical Testing ...... 116 6.5.2 Assuring High Joining Quality...... 122
VI 6.5.3 Microstructural Testing...... 130 6.4.3.1 Quality evaluation ...... 130 6.5.3.2 Quantitative evaluation ...... 143 6.6 Summary...... 149 VII. IR WELDING OF HIGH TEMPERATURE THERMOPLASTICS 145 7.1 Introduction ...... 152 7.2 Experimental Procedures ...... 154 7.3 Microscopic Evaluation ...... 156 7.4 Mechanical Testing ...... 157 7.5 Results and Discussions ...... 158 7.5.1 Effects of welding conditions on tensile strength ...... 158 7.5.2 Evaluation of PBT welds ...... 168 7.6 Summary ...... 175 VIII. IR WELDING OF GLASS FILLED POLYETHER SULFONE COMPOSITE...... 145 8.1 Introduction ...... 182 8.2 Welding Procedures ...... 184 8.3 Mechanical Testing ...... 185 8.4 Microscopic Evaluation ...... 186 8.5 Results and Discussions ...... 187 8.5.1 Tensile Testing ...... 187 8.5.2 Microstructural Testing ...... 192 8.6 Summary ...... 193 VIIII CONCLUSIONS AND RECOMMENDATIONS...... 196 APPENDICES A. Coefficient for Indirect Radiation Model ...... 201 B. Radiation Heat Transfer Program ...... 209 C. Uniform Energy Distribution Calculation ...... 211 REFERENCES...... 217
VII LIST OF FIGURES
1.1 Schematic pressure-time curve showing the three phases of IR welding 5
2.1 The installation of the overall infrared welding system ...... 14
2.2 Flow chart of IR preliminary adjustment and welding process ...... 15
2.3 The installation of the system for the irradiation field of the IR source 17
2.4 IR welding system control flow chart ...... 18
2.5 The schematic of sample fixturing of IR welding system ...... 21
3.1 Variable involved in theoretical analysis for direct radiation ...... 24
3.2 Variables involved in the upper and lower bonds of the variables ...... 27
3.3 Variables involved for projection of a ray on a U-shaped reflector ...... 31
3.4 A power/energy meter connecting with a detector head ...... 39
3.5 Attached aluminum plate on the detector head ...... 40
3.6 Comparison of experimental and calculated radiation field ...... 42
3.7 The calculated values for both direct and indirect radiation fields ...... 43
4.1 The effect of applied voltage on T3 beak wavelength ...... 48
4.2 The effect of applied voltage and power output on the dial scale ...... 50
4.3 A plot of voltage vs. power output ...... 51
4.4 Schematic assembly diagram...... 52
VIII 4.5 The effect of the heating distance and the dial scale on power output 54
4.6 Comparison of the normalized heating distance 0.25 cm, 0.89 cm and
1.5 cm...... 56
4.7 The relative power distribution in the plane x= 0.89 cm ...... 59
4.8 The relative power distribution in the plane y =2.5 cm ...... 60
4.9 The relative power distribution in the plane z =0 ...... 62
4.10 The comparison of the normalized curves in different planes z = 0, 0.25
and 0.83 cm ...... 64
4.11 The comparison of the normalized curves in different planes x= 0.25cm,
0.89 cm, 1.5 cm, and 2.5 c m ...... 65
4.12 The comparison of the normalized curves in different planes x = 0, 2.5 cm
,5 cm,and 7.5 cm ...... 67
4.13 Comparison of the normalized from result, mathematical model result, and
measured result in the x direction at y = 0 and z = 0 ...... 70
4.14 Comparison of the normalized from result, mathematical model result, and
measured result in the x direction at x = 0.89 cm and z = 0 ...... 71
4.15 Comparison of the normalized from result, mathematical model result, and
measured result in the x direction at x = 1.5 cm and y = 0 ...... 73
4.16 Comparison of the normalized from result, mathematical model result, and
measured result in the x direction at x = 2.5 cm and y = 0 ...... 74
4.17 Power distribution of the two modules combined to form a heat panel to
direct heat at a certain heating distance...... 76
IX 4.18 Comparison of theoretical and experimental results for uniform energy
distribution within two parallel reflectors in the z direct at (a) x = 5 cm
and y =0, (b) x = 2.5 cm and y = 0 ...... 77
5.1 A plastic specimen fill the rectangle shape of aluminum plate ...... 84
5.2 A schematical setup for the measurement of the reflected energy from the
surface of polypropylene ...... 87
5.3 Total radiation Energy ...... 87
5.4 A physical boundary including energy inputs and outputs ...... 89
5.5 Normalizable curves of polypropylene penetration as function of
thickness at two different heating distance 0.89 cm and 1.5 cm ...... 91
5.6 Normalizable curves of PBT penetration as function of thickness at two
different heating distance 0.89 cm and 1.5 cm ...... 92
5.7 Normalizable curves of black PBT penetration as function of thickness at
tw o different heating distance 1.5 cm ...... 93
5.8 One-dimension radioactive heat transfer in plastics ...... 94
5.9 Effect of infrared energy on the imbedded thermocouples ...... 98
5.10 Temperature history at two different locations (1.25 cm and 2.5 cm) for
different heating times : (a) 7 seconds, (b) 10 seconds, (b) 14 seconds,
and (d) 25 seconds ...... 101
5.11 Experimental results for penetration depth as function of thickness ...... 102
5.12 Comparison of theoretical and experimental melting zones at heating
distance 0.89 cm ...... 104
x 5.13 Comparison of theoretical and experimental melting zones at heating
distance 1.5 cm ...... 105
5.14 The increase of melting zone under the conditions of heating distance
0.89 cm, heating times (a) 15 sec. (b) 20 sec. (c) 24 sec ...... 107
5.15 Comparison of theoretical and experimental temperature histories at
different location (1.25 cm and 2.5 cm) at different heating distances
(a) 0.89 cm and (b)2.5 cm ...... -...... 108
6.1 (a) sample dimension and (b) the dogbone shape of the welded specimen.. 115
6.2 Effect of heating distance on joint strength for PP: welding pressure
constant at 0.44 MPa, change-over time 1 second ...... 118
6.3 The melting zone at different heating distances: (a)0.25 cm, (b)0.89 cm,
and (c) 1.5 cm ...... 120
6.4 Effect of forging pressure on joint strength for PP ...... 121
6.5 Effect of change-over time on joint for PP ...... 123
6.6 Effect of forging pressure and change-over time on joint strength for PP.. 124
6.7 Load-deformation curve for unwelded PP specimens ...... 125
6.8 Brittle behavior under the condition of heating distance 1.5 cm,
heating time 41 seconds, change-over time 1 second, forging pressure 0.44
M Pa...... 127
6.9 The deformational behavior of the high quality joining in polypropylene... 129
XI 6.10 Polarized optical micrography on polypropylene surface under the
conditions; heating time 7 seconds, change-over time 1 second, and
forging pressure 0.44 MPa. (x50)...... 131
6.11 Six regions weld zone in the polypropylene welds ...... 132
6.12 Polarized optical micrography on polypropylene surface under the
conditions; heating time 7 seconds, change-over time 1 second, and
forging pressure 0.44 MPa. (xlOOO)...... 134
6.13 Polarized optical micrography on polypropylene surface under the
conditions; heating time 11 seconds, change-over time 1 second, and
forging pressure 0.44 MPa. (x50)...... 135
6.14 The complete weld bead in the longitudinal direction under the
heating time 11 seconds, change-over time 1 second, and pressure 0.44
M Pa...... 137
6.15 The complete weld bead in the transverse direction under the heating
time 11 seconds, change-over time 1 second, and pressure 0.44MPa ...... 138
6.16 Polarized optical micrography on polypropylene surface under the
conditions; heating time 15 seconds, change-over time 1 second, and
forging pressure 0.44 MPa. (x50)...... 139
6.17 Polarized optical micrography on polypropylene surface under the
conditions; heating time 11 seconds, change-over time 0.8 second, and
forging pressure 0.44 MPa. (x50)...... 140
XII 6.18 Polarized optical micrography on polypropylene surface under the
conditions; heating time 11 seconds, change-over time 0.8 second, and
forging pressure 0.44 MPa. (x200)...... 141
6.19 Polarized optical micrography on polypropylene surface under the
conditions; heating time 11 seconds, change-over time 0.8 second, and
forging pressure 0.44 MPa. (x200)...... 142
6.20 Polarized optical micrography on polypropylene surface under the
conditions; heating time 11 seconds without forging pressure in IR
welding (x200) ...... 145
6.21 Polarized optical micrography on polypropylene surface without forging
pressure in hot plate welding (x200) ...... 146
6.22 Polarized optical micrography on polypropylene surface under the
conditions; heating time 7 seconds, change-over time 1 second, and
forging pressure 0.44 MPa. (xlOOO)...... 147
6.23 Polarized optical micrography on polypropylene surface under the
conditions; heating time 7 seconds, change-over time 1 second, and
forging pressure 0.44 MPa. (xlOOO)...... 148
7.1 The dimension of the specimen ...... 156
7.2 ...... The dogbone sample dimensions ...... 158
7.3 Joint strength for high temperature resistance thermoplastic white PBT... 159
7.4 Joint strength for high temperature resistance thermoplastic black PBT... 160
XIII 7.5 Effect of the heating distance on joint strength for black PBT; welding
pressure 0.44 MPa, change-over time 0.8 second ...... 162
7.6 Effect of the power level on joint strength for black PBT; welding
pressure 0.44 MPa, change-over time 0.8 second ...... 163
7.7 Effect of the pressure on joint strength for cream white PBT ...... 165
7.8 Effect of the pressure on joint strength for black PBT ...... 166
7.9 Effect of the change-over time on joint strength for black PBT ...... 167
7.10 Optical micrograph of PBT after 2 hr etching treatment (x50) ...... 170
7.11 Optical micrograph of microtome sectioning PBT (x50) ...... 171
7.12 Optical micrograph of microtome sectioning PBT (x200) ...... 172
7.13 Optical micrograph of PBT after 4 hr etching treatment (x200) ...... 173
7.14 Optical micrograph of PBT after 6 hr etching treatment (x400) ...... 174
7.15 Optical micrograph of PBT after 6 hr etching treatment (x400) ...... 176
7.16 Optical micrograph of microtome sectioning PBT (x400) ...... 177
7.17 Optical micrograph of microtome sectioning PBT (x400) ...... 178
7.18 Optical micrograph of microtome sectioning bulk PBT(xl500) ...... 179
7.19 Scanning electron microscopy evaluation of PBT after 6 hr etching
treatment (x2000) ...... 180
8.1 The dimension o f the specimen ...... 185
8.2 The dogbone sample dimensions ...... 186
8.3 A 20% glass-filled reinforced PES weld in longitudinal section ...... 187
8.4 A 20% glass-filled reinforced PES weld in transverse section ...... 187
XIV 8.5 Effect of heating time on joint strength for PES ...... 189
8.6 Effect of change-over time on joint strength for PES ...... 190
8.7 Effect of forging pressure on joint strength for PES ...... 191
8.8 Polarized optical micrography of a 20% glass-filled reinforced PES weld
in longitudinal section under the conditions; heating time 15 sec., change
over time 1 second, and forging pressure 0.744 MPa(x50) ...... 194
8.9 Polarized optical micrography of a 20% glass-filled reinforced PES weld
in transverse section under the conditions; heating time 15 sec., change
over time 1 second, and forging pressure 0.744 MPa (x50) ...... 195
9.1 Multiple modules combined to form a heat panel to direct at a moving
plate...... 200
C.l Calculated result for power distribution at heating distance 2.5 cm 211
C.2 Calculated result for power distribution ...... 212
C.3 Calculated result for translation of power distribution ......
213
C.4 Uniform power distribution ...... 213
C.5 Calculated result for power distribution ...... 214
C.6 Calculated result for power distribution ...... 215
C.7 Calculated result for translation of power distribution ...... 216
C.8 Uniform power distribution ...... 216
xv LIST OF TABLE
5.1 The evaluation of fraction of reflection energy ...... 86
6.1 Melting zones with variation of welding parameters ...... 141
XVI CHAPTER I
INTRODUCTION
1.1 Introduction
To join plastic products such as sheets, films, and pipes welding is
usually used. Depending on the welding process, manual or automated
equipment may be used. Generally speaking, welding processes can be
divided into non-contact and contact methods. Non-contact welding
methods, such as hot gas, extrusion, infrared (IR) and laser welding, are of
considerable importance for future industrial applications. The main
advantages of these welding methods are their ability to make rapid heating and capacity to minimize contamination risk. From the standpoint
of productivity and weld joint quality, these welding processes not only provide continuous, rather than step-wise operation, but also generate higher joint strengths. Nonetheless, contact welding of plastics, especially
l hot plate welding, can also generate high quality joints. However, the
main disadvantages of hot plate welding are relatively long processing
times and sticking to the hot tool. These are precisely the reasons for the
emergence of IR welding. It is intended to combine the advantages of both
contact and non-contact welding methods. Since IR welding is a non-
contact process, no sticking of the part to the heat source occurs. Another
important difference between the two welding processes, IR and hot plate,
is cycle time. For IR welding the total cycle time can be much shorter than
for hot plate welding. Therefore this study is aimed at developing a better
understanding of the IR welding process.
IR welding, which utilizes non-contact heating, can be used to join a
wide variety of plastics while other non-contact methods such as hot gas or
hot air welding can only be used for specific plastics. Furthermore, hot gas
or hot air welding can carry moisture or oxidants which may alter the
chemical structure of the molten polymer at the joint interface. For plastic joining IR welding is more cost-effective than laser welding. Briefly, the
advantages of IR welding over other non-contact welding processes include less transmission losses in air, direct heat transmission into the material, fast response time (i.e., 1.5 seconds) to achieve full power, and low weight
equipment. However, the heating during IR welding in plastics is not well 3 understood. Therefore it is important to determine how IR welding works for polymers, thus understanding heat transfer phenomena, radiation field, and ascertaining the optimum welding parameters.
1.2 Description of IR Welding
In IR welding, radiation energy from two IR sources, interacting with the two parts under zero pressure, is used to create a weld. IR welding can be divided into three phases, which are schematically shown in Figure 1.1 : (1) heating, (2) change-over, and (3) joining and cooling under pressure. The radiation energy causes the parts to heat and to form a molten layer. The parts to be joined are then brought together under pressure. By understanding the different phases in IR welding the process may be modeled. Rectangular thermoplastic specimens can be accurately butt-welded in a controlled manner.
The heating phase is carried out by moving the IR heating modules so that they are exposed to the parts’ surfaces until the molten layer thickness is large enough to ensure excellent joining. No pressure is needed during heating due to the non-contact nature of the heating process. Therefore, no polymer orientation occurs during the heating phase. 4
During the change-over phase, the IR heating modules are retracted
and the parts are brought into contact. Since the parts being joined are
cooled by conduction and convection during this phase, the change-over
time should be as short as possible.
Finally, constant pressure is applied to the parts. This forces molten
material out from the welded interface resulting in the formation of a weld bead on the outside surfaces. When the welded parts have cooled
sufficiently, the pressure is released and the parts are removed from the
IR welder.
1.3 Mechanism and Application of IR
Since a large proportion of the radiant energy falls in the near infrared region of the electromagnetic spectrum (0.75|jm-3pm), IR welding provides a practical solution to joining of plastics, especially for high
temperature polymers. At these wavelengths the radiation is not affected by the surrounding air, which makes the process energy efficient and hence practical. The tungsten filament in the IR welding assembly reaches
2200°C ( 4000°F ) within 1.5 seconds, thus generating radiation with a peak power at 1.15 pm [1], This wavelength can be obtained by the welding of IR phases three the curveshowing pressure-time Schematic 1.1 Figure Pressure (kPa) a 0X0 0X0 a etn ieCag-vrTm Joining & Cooling Change-Over Time Heating Time Heating Heating Time q - w □-* Time (sec) Time Change Over Joining & Cooling Joining
distribution of radiation at 2200°C in accordance with Wien's
displacement law [2]:
2898 \m °K max y
where tanax is the peak wavelength. This near infrared radiation is not absorbed by the air [1], so that most of the generated radiation transmits in polymers. In as much as infrared energy will penetrate within the polymer, an increase in molecular action as well as a temperature increase occurs. As a result, a temperature gradient is developed in the part.
Whenever a temperature gradient exists heat conduction will occur, causing the low temperature areas to heat. In other words, this penetrating heat produces both conductive and radiative heat transfer, thus causing rapid heating throughout the polymeric samples.
Applications of IR processing include: coil coating, industrial drying, continuous production of polymeric composites, curing, ink-jet printing, and sealing [3]. For example, IR heating modules are used successfully today for combining multiple modules to form heat panel for industrial drying or space heating, and even for arranging individual modules to completely surround a pipe for heat treatment. Accordingly, IR assures industry of dust-free processing conditions. Nevertheless, no IR processes are widely used in industry for plastics joining.
To make IR welding more acceptable to industry, the main purpose of this study is not only to continue the previous work of other researchers,
[4-7], but also to analyze the heating mechanism. The parameters of the IR welding process are based on the welding parameters of hot plate welding and have been adequately described for an IR radiation welding system in the published literature [4], However, heating distance becomes a critical welding parameter for certain material, design or technological considerations. This is also important in fixture design and energy efficiency when minimizing the heating distance, which factors have not been considered before. Consequently, developing an integral technique which can give the greatest joint quality is difficult. Therefore, it is important to develop an approximate model which can simulate the heating process. For the development of the heating analysis, it is also critical to determine the required melting zone size which can guarantee excellent joint strength. The goal is to investigate the previous known parameters (heating time and change-over time) to develop new ones such as heating distance, forging pressure, power level, and pigment and fiber 8 orientation effects, and to evaluate joint quality.
The next stage is focused on modeling the radiation fields.
Analyzing the radiation field is important for determining the efficiency of the IR welding process and it includes studying variables such as type or size of lamp, heating distance, and reflector shape. Conductive and radiative heating modes as well as a large heating rate make IR welding very useful in industrial and laboratory heating applications.
1.4 Literature review
Recent studies [4-7] have shown that infrared techniques can be used to join a number of thermoplastic materials for various applications.
IR joining of three types of thermoplastics (i.e. HDPE, ABS, and glass fiber reinforced PPS) has been successfully demonstrated [4], This study shows some of the more important parameters influencing operation, including irradiation time, change-over time, absorption index, transmission index, and reflection index. Unfortunately, this study [4] did not describe in detail how these parameters influence the joint strength, or consider the effect of heating distance, forging force during the cooling stage, or thermoplastic pigments. Also, no analytical model simulating the heating phenomena is currently available. This is an important
consideration when process control becomes of interest.
A single-lamp infrared reflector has been utilized in the
manufacturing processing of thermoplastic composites [5]. In this method,
bonding can be accomplished for thermoplastic matrix composites such as
PEEK and Ultem polyetherimide with a melt-flow temperature greater
than 316°C. Likewise, in the comparison of three welding methods, focused infrared heating, resistance welding, and ultrasonic welding, the
best results for joining graphite/PEEK thermoplastic composites were
obtained with the focused infrared heating method. This indicates that the infrared heating process performs better than the others. Moreover, focused infrared heating utilizes a robotic system for applying the required high temperature throughout the material, thus minimizing fiber-matrix
disturbance. In conclusion, focused infrared heating can be concentrated at the interface and is capable of being automated.
Joining of plastic pipe by the convection and radiation method was
also shown to he possible [6]. Furthermore, uniform heating, small weld bead size, good weld strength, and good repeatability have been reported.
A longer heating time to achieve a qualified joint is a disadvantage 10 because the convection loss in the air and the conduction loss in the
radiator ultimately generate a slower heating rate when the radiation is in
the medium infrared (3pm-30|jm)range.
When welding polymeric matrix composites, the best results were
achieved with radiant welding in comparison with other welding processes
such as heated tool or vibration welding. For a material whose bulk tensile
strength is 110N/mm2 a maximum welding factor of 0.83 was achieved
using radiant welding [7].
1.5 Objectives of this work
Since IR heating does not require air movement to transfer heat, the
possibility of dust contamination is reduced, and the total cycle time is
short. Infrared welding will provide industry in the near future with a fast, efficient, controllable method for joining thermoplastics. To date, based on the large number of joints that have been and continue to be
made by different welding methods, high-quality joints are obtained if
correct joining procedures are followed. However, the IR welding method has not been well understood, [4-7]. In particular, it is not clear which variables are most critical to proper joining, and how much these parameters may deviate from recommended values in order to have an acceptable joint. To fully understand the IR welding process, the complete
results from this study become critical keys to solve practical problems.
Therefore, the primary objectives of this study are to develop a better
understanding of IR welding and to study the feasibility for IR welding for
commercial use. This will be done through studying the physical phenomena of IR joining and the influence of joining parameters on
strength, studying the resulting microstructure, and modeling the heating phase including models for field distribution and heat transfer. In this
dissertation, principal activities consist of:
(1) Develop an IR welding system
(2 ) Investigate the IR welding of polypropylene (PP).
(3) Investigate the IR welding of a high temperature thermoplastic (PBT)
(3) Investigate the IR welding of glass-filled reinforced polyether sulfone
(PES)
(5) Evaluate the radiant field by establishing mathematical and
experimental models for the 3-D IR intensity heating distributions.
(6 ) Develop the heating models by evaluating the depth of penetration,
the reflected flux (R), the absorbed flux (A) and the transmitted flux
(T).
(7) Correlate material behavior to the welding power level. At this point pigmentation effect of thermoplastics will also be evaluated.
(8 ) Optimizing IR welding parameters in order to achieve a high joint
efficiency (approximately 1 0 0 % joint strength) and low welding times.
(9) Examine the microstructure of the welded material.
(10) Evaluate the Heat Affected Zone.
(11) Develop a more effective IR welding method, by evaluating welding
parameters, IR source modeling, thermal radiation heat transfer, and
microstructural evaluation. CHAPTER II
DESIGN AND CONSTRUCTION OF THE IR WELDING SYSTEM
2.1 Introduction
An infrared welding system was designed and built. It contains two
15 cm long T3 quartz tungsten filament infrared lamps with reflectors.
Figure 2.1 shows an overall view of the research IR welding machine. The entire IR welding system, as shown in Figure 2.1, includes high intensity
IR modules with power controller, two independent pneumatic cylinder systems, timer control system, cooling device, foundation, and fixture.
The operation of this IR welding machine consists of preliminary alignment and welding process ( as shown in Figure 2 .2 ). The fixturing is capable of positioning specimens accurately for butt welds. The heating distance must be taken into consideration when the specimens are
13 Figure 2.1 The overall infrared welding system 15 Power On
Set ler | Power Controller
Heating Distance Alignment
Check
Change-Over Adjustment
Preliminary Adjustment
Infrared Welding Process
Welding Porcess Heating Time Monitoring
Joining & Cooling
Stop
Fan Cooling
Power Off
Figure 2.2 Flow chart of IR preliminary adjustment and welding process 16 positioned. The alignment of the machine can be checked and adjusted by simply activating the IR welding machine through a manual cycle. After alignment, the specimens retract to their original positions precisely. The welding process can be automated under the required joining conditions.
Therefore, the specimens can be welded in either automatic or manual operation. The IR welding parameters, such as the heating distance, the welding time, the change-over time and the pressure, can be controlled independently and accurately.
2.2 High Intensity IR Modules With Power Controller
2.2.1 High intensity IR heater
The L12HVC high intensity IR heater, manufactured by ERASER
CO., features two modules in parallel with two reflectors including two IK-
0011 quartz halogen tubular infrared lamps illustrated in Figure 2.3. The
IK-0011 infrared lamp has a power rating of 1200 watts per unit length and 115-144 volts.
The high intensity IR heater is fixed and mounted on the air cylinder.
This IR heater, 15 cm x 10 cm x 375 cm, can be adjusted in the
horizontal and vertical direction with rotation and in-plane movement. Figure 2.3 The IR source showing the T3 lamp and reflector 18 modules. Therefore samples with unequal size can be welded successfully with excellent joining quality.
2.2.2 Power controller
The Lux-Therm Model LX25 power controller, manufactured by
Eraser Co., operates in 115v, 60Hz. It uses a phase angle fired triac power control to allow variation of voltage to the lamps. The triac can reduce power consumption to prevent lamp flickering and to prolong lamp life.
The maximum current is 25 amps. This power controller connects these IR modules and a digital timer in series, as shown in Figure 2.4.
Power Controller Time Controller Pressure Controller
Pressure a Controller Valve Switches
Fan Power On/off Cooling
Figure 2.4 IR welding system control flow chart 19 2.3 Pneumatic System
Two independent air cylinders are used to control two different
movements. One 6x399, manufactured by Speedaire Inc., is designed for
extending and retracting the IR modules and the other AH4,
manufactured by Trico Inc., is used for pushing the parts together. In
addition, two W64 pilot controller solenoid valves, manufactured by Ross
Co., are used to control the two air cylinders. These solenoid valves are
activated by the timer. The gage pressure in each air cylinder can be controlled by a pressure regulator from 0.083 to 0.69 MPa.
2.4 Process Controller System
The timer 365, manufactured by Automatic Timing & Controls Co., can be set from 0.1 second to 999 hours. When the timer is activated by pressing the start timer button, the IR lamps receive full voltage. At the end of the timing period, the voltage to the lamps is turned off. This timer controls the heating time and change-over time.
Two snap action (limit) switches 6x293, manufactured by Cherry
Co, in series, are used for the change-over time control. Upon retraction of the lamp air cylinder the switches are turned on. As these snap action switches are activated a single solenoid valve activates the part cylinder 20 which brings the parts together. Using two switches in series prevents unwanted movement of the system during the heating stage. Thus, if one of the switches is accidentally turned on, the system will not be activated since they are coupled in series. Therefore, putting the switches in series the two parts will not be accidentally pressed together during the heating stage, thus ensuring that no damage occurs to the IR lamp, as shown in
Figure 2.4.
The change-over time can be varied by using flow control valves.
These accurate and efficient control valves provide infinite control settings from full-close to full-open under the same pressure in the process.
2.5 Lamp Cooling Setup
A fan is used for cooling. Cooling must be provided to each module to maintain the lamp end seals under their rated temperature of 343°C
(650°F). Therefore after heating cooling air is blown on to the lamp and reflector to cool them as quickly as possible.
2.6 Foundation and Fixture
The entire foundation has dimensions is 75 cm x 50 cm x 40 cm as shown in Figure 2.4. The distance from one surface of the part to the other should be kept as short as possible, so the change-over time will be short.
One end is fixed rigidly and the other is moveable by approximately 10 cm in this foundation, as shown in Figure 2.5. Two fixtures are applied to confine the parts at an appropriate position. They also offer strong support and constrain the parts for proper alignment so that the parts to be joined will perfectly match.
Moving
Fixture
Figure 2.5 The schematic of sample fixturing of IR welding system CHAPTER III
MODELING AND MEASUREMENT
OF
RADIATION FIELDS
3.1 Introduction
At present, very limited information has been given on radiation fields in IR welding. Many studies made on the radiation field, [8-17], were devoted to photochemical processes in ultraviolet (UV) radiation. The shape of the reflectors can be divided into three categories: circular, parabolic, and elliptical. The reflection of radiation from an elliptical reflector is much higher than for a circular or parabolic reflectors.
Rigorous modeling and experimental validation indicate that systems incorporating the total power of circular or parabolic reflectors can be precisely controlled using this model [8-17].
22 23
No incident source system can be used without a theoretical and experimental analysis. The radiation field, can be divided into two parts, direct and indirect, or so called reflected radiation. This model is not only the basis for the present study, but also constitutes the foundation for future work.
Since the generated radiation fields are not uniform, problems arise when heating large samples. Thus, by using this model the radiation field can be determined analytically and an appropriate reflector design can be employed to unify the field.
3.2 U Shape Reflector System
As was discussed earlier, it is preferable to employ an elliptical shape for the reflector in order to obtain a good radiation pattern from the lamp. Nevertheless additional analysis is necessary to find ways to improve heating efficiency. The purpose of this study is to establish a comparison of efficiency between elliptical and U shaped reflectors. The comparison is as follows: 24
i i i
p=p>
i
Figure 3.1 Variables involved in theoretical analysis of direct radiation
[10] 25
ADVANTAGES (U SHAPE) ADVANTAGES (ELLIPTICAL SHAPE)
1. Light weight & small modules 1. Small heating area
2. Inexpensive 2. Focused energy
3. Short heating distance 3. Fast cooling system(water cooling)
4. Easy to replace lamp
DISADVANTAGES (U SHAPE) DISADVANTAGES (ELLIPTICAL)
1.Easily broken lamp connector 1. Heavy
2 . Divergent radiation energy (loss) 2. Expensive
3. Long focal distance
The advantage of a U shaped reflector is that the light weight heat source can be easily, quickly, and precisely moved. Other major advantages are low cost and small size. From the comparsion of U shape and elliptical shape, the U shape reflecting surface seems preferable for joining parts in IR welding. 26 3.3 Mathematical Model of Radiation Field
3.3.1 Direct Radiation
The incremental rate of energy emission at a frequency u (with o between u and u+du) about the direction of incidence given by 0 and § [1 0 ] is
d Ee 4, p u = — sin2 0 cos <()d0 d<|)dpd Adu (3.1)
where Ne describes the number of emitters per unit volume of source andp 0 di) represents the probability of emission per unit time at a given frequency interval dl). In Equation 3.1, h is Planck’s constant and p is the radius in spherical coordinates as shown in Figure 3.1. At any point the incremental radiation flux density emitted from the elementary volume dV with direction 0 , <|) and for the whole range of frequencies, as shown in
Figure 3.1, is given by
d qe.*.p sin2 0 cos where 27 > X p sm Figure 3.2 Variables Involved in the Upper and Lower Bounds of the Variable p and represents a property of the radiation source. Therefore, the total energy per unit area and unit time from the whole volume of the lampshown in Figure 3.1 can be obtained from Q(z0,r)=Kf f fsin2 0 cos<|)ded<|)dp (3.4) As shown in Figure 3.2, the upper and lower bounds of the variable p are Pupper = P2 sin6 “ rcos Plower = Pi sinG - rcoscj) = - ( r L2 - (r2 - r2 cos2 Finally, the upper and lower bounds are „ _ rcos<|) + (r2 cos2<|>-r2 + rL2)05 Pi,2 ~ ------r-r------(3.5) sin0 Substituting (3.5) into (3.4), and performing the integration 29 P2 2 2 2 2 0.5 Q(z0,r) =K jjJsin2 0cos<|)dpd0d (3.6) From Figure 3.1, the limits of the variable 0 are ®upper => Pi COS0J = (L - z0) ®lower P i COS02 = — Zq Combining Equations (3.5) and (3.7), the upper bound of the variable 0, is (L -z 0) = cot0,(rcos(l>-(r 2 cos 2 (J>-r2 + rL2)05) (3.8) 9 = tan"1 ((r cos ^ ~ (rZ cosZ 4>~ + rL2)°5) L -z n Similarly, the lower bound of the variable 0 2 is —z„ = cot02(r cos(J) - (r2 cos2 02(4>) Q( =2K{ (r2 cos 2 <[> - r2 + il 2 )O‘5 (cos 0 2 (<]>) - cos0i((|))cos(j)d(l) ♦ (3.10) 3.3.2 Indirect Radiation Indirect radiation expresses the radiation that comes from a differential volume of the lamp with direction 0 and d qe+p = KTrfsin 2 0 cos deddp 1 where k was defined in Equation 3.3 and represents a property of the radiation source. Trf is the reflection coeficient of the reflector shown in Figure 3.3. Therefore, the total energy per unit area and unit time which strikes the differential area at any given point, from all directions in space and from the whole volume of the lamp, can be obtained from 31 Reflector .3325 0.2325 0.25 Lamp P(X p,Y p) Figure 3.3 Variables involved for projection of a ray on a u-shapeed reflector 32 Q in (zo, r) = KTrf 1 1 | sin2 0 cos4)d0dcj)dp (3.12) The upper and lower bounds of the variable p are D cose - pE[2 sin 0 = (rL2 - D2 sin2 s ) ) 05 (3.13) Finally, the upper and lower bounds are _ rcos Substituting (3.13) into (3.12), and performing the integration Pe2 Qin (z0, r) = KTf | | J| sin2 9 cos(j)dpd 0 d 2 2 2 .0 .5 t r r ■ 2r.2(rr - D COS s) . . = KTc|fsm 0 . cos(j)d0 d(}) (3.15) Dsins = t = -[(X P +P + 0.25)sin Dcoss = - [(Xp + P + 0.25 cos<))e +(YP +Q)sin(J>E] (3.16) 33 where Dsin s and Dcos e have been derived in Appendix A. From the law of reflection, the angle between Ei, N and Ee, N will be equal: Z^p,^i = ZN p,E{ (3.17) Thus, the dot product of these vectors for these angles will be t p - f ' = N p-E{ (3.18) The Ei, Ee, and NP have been derived in Appendix A. Therefore, 4>E can be found. 2 , (m -l)sind) + 2mcosd) -0.815 - P < Xp < -0.4825-P , (3.19) -0.3325 - Q < YP < 0.3325 - Q m = 4 Y p-*Q ) where XP+P + 0.4825 -0.4825 - P < Xp < P, (3 20) E -0.3325-Q Integration along 0 must be computed numerically. The limits of integration are given by L - k = (pi +pe) cos 0 i —K = (p i + p E)CO502 (3 21) Substituting Eq.(3.14) into Eq.(3.21), it is shown that this dependence is 'p{+Pcoss-(rL 2 -D sin . 2 s) °'5 j 0 i = tan 1 L - k rpf +Dcoss-(rL 2 -D sin . 2 s) 05 0 2 i = tan I _K (3.22) where 0.5 Pi(4>) = Pi sin 0 = [(P + 0.4825)cos(|) + Qsin ()>] + |r 2 -[ (P + 0.4825)sin (j) - Q cosij )]2 J (3.23) The coefficients of pi have been derived in Appendix A The limiting variable <|) for any angle must be tangent to the lamp boundaries, which translates into : Pe, = P e, (3-24) By substituting the equality in Equation (3.24) into Equation (3.14) the following result is obtained: D2sin2e = rL2 (3.25) Besides, two regions in the coordinates referred to the point I, as shown in Equation (3.26), can also be divided into three regions in the (j)-direction for any angle. The following expression is obtained: 0.4825-P < XP <-P, , -0.815-P By transferring Equation (3.26) from the global Cartesian coordinate system (X, Y) to the spherical coordinate (r, 0, <|>), these limiting , + -i /D.3J25_-r._Q,\ A + -i /-0-3325 - Q \ d)A| = tan I ) rf>A2 = tan ^ ) T v -P v - P - 0.4825 t -i ; ..0.3325- Q \ , „ -i/ -0.3325-Q ) tan ) d>B2 = tan ) - P - 0.4825 v ' - P - 0.4825 . + -i / -0.3325- Q ' . + -i / -0.3325 - Q \ d»c, = tan chA2 = tan V V -P 1 ^ - P - 0.4825 (3.27) N ote that Finally, the complete expression becomes : Q* = KT- { C C( + J^ 2 A(<)))|cos 0 b2 (<|>) - cos 0 B, (<|>)] cos + J^ C2 E( The coefficients of C((f>), A((j)), and E(<|)) have been derived in Appendix A. Also, 0 ^ 2 ,0B2 > ®B2 ’ 0 C2 anc* have been derived in Appendix A. 37 3.3.3 Total Radiation The total energy is obtained by combining the direct and indirect radiation. Therefore, substituting Eq.(3.10) and Eq.(3.27), the complete form is QT = Q + Qit Qt = 2KJ(r2 cos2 (j) - r2 + q,2 )°'5 (cos 0 2 ( < f » B2 + J A( 3.4 Experimental Setup A Model FM Coherent FieldMaster power/energy meter connected to a LM-1000 detector head was used to measure the irradiation of the IR source, shown in Figure 3.4. The LM-1000 detector head has a water cooling system with spectral response in the 0.3 to 10.6 pm range. 38 A rectangular shaped aluminum plate, as illustrated in Figure 3.5, was attached to the outward surface of the head in order to measure the actual power absorbed by the joining parts at a specific heating distance. Throughout the measurements the IR source is placed in front of this assembly at different heating distances. It takes 10-13 seconds for the device to stabilize, displayed at room temperature (20°C), sifter turning on the IR source. By varying the heating distance, the power absorbed by the detector head also varies. For each heating distance the value on the power/energy meter is read. Finally, the complete experiment results will be determined by sequentially changing the distance between the detector head and the IR source. 3.5 Results and discussion A normalized energy comparison of experimental results and calculated radiation field based on the Equation 3.29 is shown in Figure 3.6. In Figure3.6 a value of 100% relative energy corresponds to the energy obtained at a heating distance of 0.254 cm, with all other relative energy values are normalized to this maximum. The calculated results and the experimental data are in good agreement. The predicted values are calculated by using Mathcad considering both direct and indirect incident Figure 3.4 A power / energy meter connecting with a detect or head Figure 3.5 Attatched aluminum plate on the detect or head radiation as illustrated in Figure 3.7. It is very clear that the direct and indirect radiation make different contributions to the total energy reaching the parts. The indirect radiation gives more energy than the direct radiation. According to the literature [1], the reflection coefficient of the reflector, Trf , is approximately 0.95 because the pure gold-fired ceramic reflector will withstand high temperatures and not oxidize. Moreover, the radiation energy reaching the parts rapidly decreases when the heating distance increases from 0.254 cm to 1.32 cm, followed by a much slower decrease in the 1.32 to 5.08 cm range. 3.6 Summary 1. Since the proposed model is suitable for modeling experimental results, it can also be concluded th at: a.The effective absorption energy by the joining parts is principally provided by the reflected radiation. b.The contribution of reflected energy to the total energy reaching the joining parts ranges from 33% to 98%. This also shows that decrease in heating distance increases incidence efficiency. 2. A reflector coupled with an IR source provides another welding method for plastics that generates excellent joining efficiency. 42 0.9« O — Calculated Results X— Experim ental Results O . 8 1 0.7< O . 6 1 0.5i 0.4. 0.3. 0 . 2< 1.25 2.5 3.75 H eating D istance(cm ) Figure 3.6 Comparison of experimental and calculated radiation fields 0.6 0.5 0.4 >» Indirect R adiation 0.3 >» 15JD 0.2 0 .1 -. D irect R adiation 1.25 2.5 3.75 D istance(cm ) Figure 3.7 The calculated values for both direct and indirect radiation fields The available theoretical model can be widely applied to large surface joining areas. When multiple reflectors are combined to form a heat panel the only tool which can predict the resulting radiation field is this proposed model. In most cases, in order to make a first estimate of the radiation energy involved in the process, it may be a good approximation to consider the total energy absorbed by the parts. In some cases, coupling a reflector to the IR set-up can be useful for photochemical reactions or thermoplastics-thermoset bonding because of the contact-free, light weight, and cost-effective advantages. CHAPTER IV MODELING AND MEASUREMENT OF RADIANT ENERGY DISTRIBUTION 4.1 Introduction Infrared radiation is usually most effective and easily applied when the lamps are located in a plane or planes parallel to that of the plastics. When the use of radiant heat is being considered, calculations employing known power requirements are of greatest value in determining the estimated heat. Knowing these requirements, it is possible to arrive at a decision as to the approximate power of infrared radiation for the purpose of infrared welding. The approximate number and arrangement of the infrared units and the adequate heating time required for a specific polymer are most readily determined from data obtained from previous installations of the same character. 45 46 The modeling is based on the prototype IR welding system. The relationship between prototype and model is broadly referred to as similarity. The condition of similarity between the IR welding system and the model can be established by a normalization analysis. This normalized analysis includes the data from calculations of mathematical modeling and the results of measurement. Therefore, by combining the results of mathematical modeling with experimental measurements, the IR radiant model becomes a semi-empirical model. Although the theoretical model, described in Chapter 3, can be applied to large surface joining areas, it may be too complicated for industrial applicability. However, this simple semi-empirical model is much more convenient for industrial use due to its simplicity and can also be applied for applications involving both large and small joining surfaces. IR power modeling using normalization analysis is a very useful tool for discovering fundamental physical phenomena. Particularly in cases involving non-linear equations this study emphasizes the established relationships between prototype and suitable model, and applies certain properties of the variables appearing in IR welding applications. In as much as IR welding applications involve small and large area joining, the foundations of the approach to modeling and testing may be traced back 47 to industrial inquiry and the study of natural phenomena. Thus, the purpose of this study is to generate a model of the radiation energy field emitted by the IR source. 4.2 Parameter Estimation 4.2.1 Effects of Wavelength, Output Power, and Applied Voltage Thermal radiation is the radiant energy emitted by particles of matter. Photons or quanta are the basic units of radiant energy. When a photon is emitted, the emitting particle loses energy. Therefore, it is convenient to discuss the radiation process by utilizing a photon point of view. In absorption processes, photons transfer part of their energy to the material. The photon energy is given by its frequency or wavelength (E = h u), where h is Planck's constant and u is the photon frequency. A tungsten- halogen lamp T3 radiates principally at a peak wavelength of 1.2 microns. As the frequency or wavelength of the photon changes, the emitted energy changes accordingly. Therefore, the total output energy of the tungsten- halogen lamp changes. The more important point is that the applied voltage will affect the wavelength of a photon as shown in Figure 4.1, [2], When the voltage changes the temperature of the free electrons of the tungsten filament changes, thus changing the occupation of energy levels. When the electrons return to their original energy level a photon is Figure 4.1 The effect of applied voltage on T3 peak wavelength [2] wavelength peak T3on voltage of applied effect The 4.1 Figure Peak W avelength (m icrons) 0.5 1.5 2.5 0 2 1 3 0 0.5 V-Applied/V-Rated 1 1.5 2 emitted at a frequency equal to the difference in the attained energy level and the ground state of the electron. As a result, the applied voltage dominates the total output energy of the tungsten-halogen lamp. The technology of controlling tungsten-halogen radiant sources is well understood and tested. Figure 4.2 shows the relative values of the applied voltage and output power on the dial scale. Increasing the scale value increases the applied voltage and also increases the output power of the tungsten-halogen lamp. Figure 4.3 indicates that the relative applied voltage is linearly proportional to the relative power output. The applied voltage is measured by a multimeter in series with two lamps as shown in Figure 4.4. The dial scale (%) is the scale on the power controller which controls the applied voltage of the tungsten-halogen lamp. 4.2.2 Effect of Heating Distance As discussed in the previous section, when a photon is emitted the energy of the emitting particle is accordingly reduced. A point source of radiation is viewed from infinite distance (R). The flux coming forth from a point source falls off as 1/R2. This means that in spherical coordinates the flux is only dependent on the radial direction. However, as shown in Chapter 3, for a tungsten-halogen lamp the radiation field is more Figure 4.2 The applied voltage and power output at different dial scales dial different at output power and voltage applied The 4.2 Figure Relative Values (%) 100 20 40 30 60 70 50 80 90 . 20 t l o V ) % ( e l a c S l a i D 0 60 40 r e w o 80 100 51 S!W5!S8?iSSSSSS5SSS) go 80 70 60 I £ vcas n 40 30 20 10 0 20 40 60 80 100 Relative Volt % Figure 4.3 A plot of voltage versus, power output 52 M ultimeter IR modulus Power Controller Figure 4.4 Schematic Assembly Diagram complicated. The flux in a tungsten-halogen lamp depends on the three directions (x, y, and z). This dependency, when transformed to spherical coordinates, can not be expressed as dependency on the radius. Therefore, the radiant field of the lamp can not be simplified to that of a point source. Does the applied voltage influence the heating distance? Figure 4.5 shows that for different heating distances (0.25, 0.89, and 1.5 cm), by increasing the dial scale the output power of the tungsten-halogen lamp also increases. However, the applied voltage also increases as the dial scale increases. Thus, it appears that the applied voltage influences the heating distance. Further analysis shows a comparable trend in the power output of the tungsten-halogen lamp. Actually, this is far too complicated for simple presentation by a family of curves. In many cases it is convenient to go through the analysis in normalized form. A simplification can be obtained if the data in Figure 4.5 are normalized and rearranged. Thus, in this analysis, a general plot can easily be generated as shown in Figure 4.6 where all heating distances exhibited in Figure 4.5 have been included. Thus, the influence of heating distance may be presented as a single curve, as shown in Figure 4.6, rather than as a family of curves, as shown in Figure 4.5 4.3 Normalization of Parameters Three uses of normalization are discussed: parameter simplification, engineering approximation, and combination of variables. The way to normalize is as follows: 1. Summarize sets of data in order to obtain curves. Figure 4.5 The effect of the heating distance and the dial scale on power on scale dial the and distance heating the of effect The 4.5 Figure power (w atts) 20 25 35: output : 0 20 40 Dial Scale(%) Dial 60 80 .5 cm 0.25 .9 cm 0.89 . cm 1.5 100 54 55 2. Compare several sets of data in terms of constant curves. 3. Normalize these parameters. 4. Obtain generally useful formulas. 4.3.1 Statistical Independence of the Applied Voltage or Dial Scale Figure 4.5 shows the actual radiant power output emitted in an area of 2.5 cm by 0.95 cm. In Figure 4.5 , W0, Wj, and W 2 indicate the maximum power output of the lamp when a heating distance of 0.25 cm, 0.89 cm and 1.5 cm has been set up. The maximum power at each heating distance represents the value of 100% Dial scale in Figure 4.5. W0 = KL0 , Wj = KLj, W2 = KL2 => Wn = KLn (4.1) where Ln is a function which describes the influence of the dial scale on power output, and K is a scaling factor which shows that all curves have a similar trend, as seen in Figure 4.6. The relative energy (W/Wmax)%, as shown in Figure 4.6, is defined as a power (Dial Scale) (4.2) Power (100% Dial Scale) 56 100' 90 80 70 60 50 40 30 20 20 40 60 80 100 Dial Scale(%) Figure 4.6 The comparison of the normalized heating distance 0.25 cm, 0.89 cm, and 1.5 cm By taking each point from each curve shown in Figure 4.5 and dividing by the corresponding maximum power the normalized power term is obtained as shown in Figure 4.6 Wg^=W/^=W^ (43) w„ w , w2 Assume th a t: W„(n) = K'Ln In a manner similar to the derivation of (1), the normalized applied voltage term is derived as shown in Figure 4.2. v ( n) Vvo =f D is the scale reading which ranges from 0 to 1. The plot of the normalized power output for each heating distance has the same trend as the plot which exhibits the dial scale’s influence on applied voltage as shown in Figure 4.2. Therefore it can be written: 58 Using these normalized forms the power output can be obtained in a different way. W„(n) = KLn = Kf(D)Ln = K ^ - L n (4.6) v o where f(D) = 14D5-48D4+ 60D3-33D2+9D- 0.857 was obtained by employing a polynomial function in the least square fit method to fit the curve(s) in Figure 4.6. 4.3.2 Statistical Independence of the Heating Position (X, Y, Z) Plotting is instructive when only one independent variable is thought to be influential. When the data can be directly represented by linear spatial dependence on the X and Y coordinates, it is fortunate. However, the radiant power spatial distribution of the tungsten-halogen lamp is unfortunately not as simple. The relative energy distribution is calculated relative to the source characteristic K (see Equation (3) from Chapter 3), using the model presented in Chapter 3, at various positions in the y-z, x-z and x-y planes. For example, Figure 4.7 shows the power distribution within the y-z plane when x = 0.89 cm, while Figure 4.8 shows 59 0.6 0.5 B o-4- c UJ u 0.3 > 1b u 0.2 CC 0.1 Hi. *< *{'*+ j-** t -1.52 -0.84 -0.254 Y - a x i s 7.6 Z - a x i s 0.254 0.84 1.52 Figure 4.7 The relative power distribution in the plane x = 0.89 cm 60 X - a x i s Figure 4.8 The relative power distribution in the plane y = 2.5 cm the power distribution within the x-z plane when y = 2.5 cm. Similarly, Figure 4.9 shows the power distribution within the x-y plane when z = 0 cm. From this analysis, as the distance from the source increases the power distribution gradually diverges, thus decreasing the power. Therefore, it is clear that a U-shaped reflector does not possess a focal point. Determing the applicability of this welding system to a higher output power is also important. Normalization may also be used to investigate the influence of each independent variable (x, y, z) on the output energy distribution. Therefore, a single equation or a set of equivalent equations is desired for each independent value. A recommended procedure for this type of analysis is: 1. Use the model in Chapter 3 to estimate the relative energy radiated within each plane, x-y, y-z and z-x. 2. Normalize the relative energy calculated at each point within a given plane to the maximum value, which can be found nearest to the heat source. 3. Fit the curves of normalized relative energy vs. position, thus generating a set of equations. 4. Compare experimental measurements to the mathematical model obtained from normalization. Figure 4.9 The relative power distribution in the plane z = 0 cm 0 = z plane the in distribution power relative The 4.9 Figure Relative Energy s i x a - Y s i x a - X 62 63 In short, to determine the applicability of this IR welding system all possible information from existing data must be analyzed. In the power distribution calculation, the data are analyzed and normalized using equation (4.3) for various x-positions in the y-z plan. Figure 4.10 shows that the relative energy distribution in the x- directionhas the same trend when y varies from 0 to 7.6 cm at z = 0, 0.25 cm, and 0.83 cm. Therefore, the normalized energy distribution for different y-positions in the x-direction overlap. Because of the same trend, one may calculate the average energy distribution in a normalized form at various x-positions in the y-z plane. Thus, the normalized energy distribution equation for the x-direction is: Wx = -0.285556^n(X) + 0.27 (4.7) Following a similar procedure, Figure 4.11 shows the normalized data at various y-positions in the x-z plane. The normalized energy distribution equation for the y-direction is: WY = 1 + 0.0276Y - 0.078Y2 + 0.05Y3 - 0.015Y4 (4.8) 64 0.5 10 15 2.0 25 30 35 4.0 4.5 5.0 Distance (X-axis) 0.25 Figure 4.10 Comparison of the normalized curves in different planes z = 0 cm , 0.25 cm, and 0.89 cm 0 0.5 1.0 1.5 2.0 2.5 3j0 3.5 4 0 4,5 5 Distance (X-axis) 0.89 65 1 1 ■ ■ ■ ■ ■ 0.9 i 0.9 ■ 0.8 0.8 8 0.7 «P 3? §0.6 § 0 .6 o B g ■ W 0.5 W 0-5 J 0.4 Jj 04 .3 (8 0.3 « 0.3 0.2 0.2 0.1 0.1 0 0 1.25 2.5 3.75 5.0 6.25 7.5 3 1.25 2.5 3.75 5.0 6.25 7.5 Distance (Y-axis) Dislance (Y-axis) 0.25 0.89 1 " i 0.9 1 0.8 B 0.7 Sp feo.e 1 B i£ °-5 £ •5 0.4 3& £ 0.3 0.2 0.1 0 1.25 2.5 3.75 5.0 6.25 7.5 1.25 2.5 3.75 5.0 6.25 7.5 Dl^ance (Y-axis) Distance (Y-axis) 1.53 2.54 Figure 4.11 Comparison of the normalized curves in different planes x=0.25 cm, 0.89 cm, 1.53cm, and 2.54cm 66 Similarly, Figure 4.12 shows the normalized data at various z-positions in the x-y plane. However, for this case the normalized energy distribution curves for different x-positions in the z-direction do not overlap when y = 0, 2.5 cm, 5 cm and 7.5 cm but a similar trend exists for all y planes. Therefore a set of the normalized equations is desired for each x-position of 0.89 cm, 1.5 cm, 2.5 cm, and 5 cm. These normalized equations were obtained by using polynomial equations to fit the curve in Figure 4.12 and are suited for any y-direction. Wz = 1 - 0.823Z + 0.046Z2 at X=0.89 cm Wz= l - 0.879Z- 1.327Z2 + 0.918Z3 - 0.1898Z4 at X=1.5 cm Wz = 0.9989 - 0.00928Z- 0.897Z2 +Q7133Z3 -0.257Z4 +0.046Z5 -0.0033Z6 at X=2.5 cm Wz = 0.9992 + 0.041Z - 0.764Z2 + 0.5099Z3 - 0.1418Z4 + 0.0149Z5 at X=3.125 cm Wz= 1 + 0.0245Z - 0.35185Z2 + 0.1562Z3 - 0.0295Z4 + 0.002243Z5 at X=5 cm (4.9) 67 1 B a D ■ a 0.9 a 0.8 a a a 0.7 r a £ ° . 5 © • J o .4 § o . 0.2 0.1 0 3 0.25 0.5 0.75 1.0 1.25 1.5 0.5 0.75 1.0 Distance (Z-axis) Distance (Z-axIs) 0 2.5 1 ) ■B a ■ 5a “■ a 0.9 o 0.6 a a o 0.7 a J 0 . 5 © ’5 0.4 £ ^ 0.3 0.2 0.1 0 ) 0.25 0.5 0.75 1.0 1.25 1.5 0.5 0.75 1.0 Distance (Z-axis) Distance (Z-axis) 5 7.6 Figure 4.12 Comparison of the normalized curves in different planes x=0 cm, 2.5 cm, 5 cm, and 7.6 cm 68 Using the normalized equations the spatial power distribution can be calculated much more easily if the power output is known at one specific location in space, which is chosen arbitrarily by the designer. 4.4 Discussion 4.4.1 Effect of Applied Voltage The power level is one of the joining parameters which influences the joint strength. The applied voltage controls the power level by simply adjusting the dial scale. The results obtained indicate that the applied voltage is an independent variable. The normalized voltage equation provides a good tool for predicting the required power level and is useful for better control of the IR welding process. For example, for a full scale reading of 100%, the emitted power to a plastic sample with a cross- sectional area of 2.5 cm by 0.95 cm is 36 Watts, for a heating distance of 0.25 cm, and 21 Watts for a heating distance of 0.89 cm. If for some reason the heating distance is fixed at 0.25 cm and the power required is 21 Watts, then by implementing the above described model the required scale setting can be determined. By first finding f(D), then D, the required power level can be determined as follows: Given: 0.25 cm — ► 36 watts in full power 100% Find: 0.25 cm — ► 21 watts in what level power ? % 69 Solve equation (4.5): w 21 = f(D) => — = 0.583 = f(D) Wn 36 from Equation (4.6): f(D) = 14D5 - 48D4 + 60D3 - 33D2 + 9D - 0.857 If D = 0.65 = 65% then,f(D) = 0.583 Thus, based on the calculation from Equations (4.5) and (4.6) the predicted scale setting (D) is 65%. Experimental measurements show that a setting of 65% provides a total of 21 W in the joining area. Therefore, it seem that the measured power output tends to agree with the predicted level. 4.4.2 Effect of the Heating Position (X, Y, Z) The effect of the heating position (x, y, z) was investigated. The normalized equations of the energy distribution were evaluated for two cases: (y = 0, z = 0) and (x = 0.89, z = 0). Figures 4.13 and 4.14 show a comparison of the predicted normalized form, the mathematical model, and experimental measurements for the power distribution of an IR welding system. The results show that this normalized model is also in excellent agreement with the mathematical model and experimental 70 Mathematical Model 0 . 9 , Q Measured Results • Normalized Results 0 . 8 , 0 . 7, 0 . 6, 0 . 5, 0 . 4, 0 . 3, 0 . 2, 0.1 1.25 2.5 3.75 D istance (X -axis) Figure 4.13 The comparison of the normalized form result, mathematical model result, and measured result in the x direction at y=0 cm and z=0 cm 71 ■) » r i r i)in f W 'F i ^ > W y 0.9 0.8 0.7 0.6 4* 0.5 iiitm Mathematical Model £ 0.4 Ob Normalized Results $ Measured Result 0.3 0.2 0.1 2.5 3.75 6.25 7.51.25 Distance (Y-axis) Figure 4.14 The comparison of the normalized form result, mathematical model result, and measured result in the y direction at x=0.89 cm and z=0 cm measurements. As shown in Figures 4.13 and 4.14, the normalized form typically predicts the power distribution to within a small error. For a complete comparison between the predicted normalized form, mathematical model, and experimental measurements, the effect of changing heating position along the z-direction needs to be analyzed. Figure 4.15 and 4.16 show comparison for two cases: x = 1.5 , y = 0 cm and x =2.5, y = 0 cm. Thus, from these cases it can be observed that if only the power distribution is required, by employing the normalized equations the analysis is considerably simplified. 4.4.3 Summary The spatial distribution of the radiant power has been predicted by employing a normalization technique. When the power distribution is required the normalized form yields reasonable results. The mathematical model is much more complex but provides a more accurate view. Another important Concept in this modeling of radiation energy for future work is to generate uniform energy distribution in the application of large area heating. For examples, based on Equation (4.9) the predicted energy for one reflector with lamp is illustrated in Figure 4.17. This energy distribution shows that high energy is located in the center of the 73 0.9 0.8 0.7 0.6 ^ Measured Results 0.5 Mathematical Model »Mm Normalized Results tt. 0.4 0.3 0.2 0.1 0.25 0.5 0.75 1.25 Distance (z-axis) Figure 4.15 The comparison of the normalized form result, mathematical model result, and measured result in the z direction at x=1.5 cm and y=0 cm 74 0.9 0.8 0.7 0.6 i Mathematical Model Measured Result Normalized Results “ 0.4 0.3 0.2 0.1 0 0.25 0.5 0.75 1 1.25 1.5 Distance (Z-axis) Figure 4.16 The comparison of the normalized form result, mathematical model result, and measured result in the z direction at x = 2.5 cm and y = 0 cm lamp and gradually decreases away form the center. From this predicted value it can be further evaluated that uniform energy distribution can be generated by putting two reflectors adjacent to each other. Figure 4.17 shows uniform energy distribution within two reflectors. The calculation results for two cases: (x=2.5, y=0) and (x=5, y=0) have been derived in Appendix C. Figure 4.18 shows that the calculated and experimental results are in good agreement in two cases. For generating uniform distribution the distance between two parallel reflector for these cases is different. In the case (x=2.5, y=0) the distance is 5 cm. In the other case (x=5, y=0) the distance is 7.5 cm. 76 Figure 4.17 Power distribution of the two modules combined to form a heat panel to direct heat at a certain heating distance 77 2* 1 .8 ' 1.6i ^ 1.4i >*g> « 1.2* ~ mr- » o fio 1 1. .a 5DC 0.8 • 0.61 Experimental Results ■Calculated Results 0.4' 0.2 ' O' 1.25 2.5 3.75 Z-axis (a) 2 ' 1 P 1.8' 1 1.6' | $. || 1.4' p >* ♦ 1.2' ♦ ♦ ♦ 1 & i & o 1 ' .a 1 JS 0.8' I 1 0.6< + Experimental Results 1 <■■■»Calculated Results 0.4' 1 0.2' 1 p (3 1.25 2.5 3.75 5 6.25 7 Z-axis (b) Figure 4.18 Comparison of theoretical and experimental results for uniform energy distribution within two parallel reflectors in the z direction at (a) x = 5 cm and y = 0 cm, (b) x = 2.5 cm and y = cm CHAPTER V HEAT TRANSFER IN IR WELDING 5.1 Introduction The interaction of radiation with other modes of heat transfer (i.e., conduction and convection) has recently aroused considerable attention. Such interaction effects may be broken down into two main categories. The first, category involves radiation interaction at the physical boundaries where conduction or convection processes occur. One example, in IR welding, is transient conduction in a solid including the penetration layer due to radiation interaction. The second category involves radiation interaction through an absorbing-emitting medium such as water in a vapor or liquid state, for which the net radiant energy is transferred to medium. Consequently, the 78 79 conduction or convection process may be thought of as one involving a heat source or internal heat generation. For this process the heating mechanism is the major concern, thus the phenomenology of IR welding of plastics is very important for this study. Therefore, the purpose of this section is to model the temperature rise rate during heating, caused by the IR irradiation of the sample surface. The generated heat will diffuse into the samples by conduction, while penetration is achieved by the radiation interaction. The diffusion and penetration rates will in turn control the temperature rise rate at the interface and within the sample. 5.2 Background As mentioned before, previous work has been done in comparing the IR joining process to other joining methods [3-6], as well as describing the effective parameters of the process. The heating mechanism which occurs during IR welding of polymers is specifically different from that achieved during hot plate welding. Nonetheless, IR welding could be a further development of the heated-tool butt welding by placing an IR source instead of a hot plate. Therefore, a comparison of the heating mechanism for hot plate and IR welding of plastics should be evaluated before modeling the heat transfer phenomena occurring during IR welding of polymers. It is of interest to investigate the limiting cases of the heating mechanism occurring during hot plate and IR joining of polymers. First, consider the case of hot plate welding of polymers. Neglecting contact resistance the boundary condition at the hot plate - sample interface is given by: T(0,t) = Tpiate t>0 (5.1) The plastic material is maintained at the hot plate temperature at their common interface, x=0. The heating mechanism of IR welding is considered to be radiative as well as conductive transport. Thus, both types of transfer, radiative and conductive, contribute to the energy transfer through a plane layer of the polymer. Consequently, the contribution of radiative energy in heating the polymer in IR welding will 81 where E is the differential rate of energy, and Q is rate of energy per unit area, a is the fraction of absorbed energy and d is the penetration depth. From equations (5.1) and (5.2) a comparison of the two heating mechanisms can be executed. In hot plate welding the sample is heated only on the surface which is in direct contact with the heated plate. On the other hand, in IR welding the radiant energy may be visualized as being transported by photons traveling within the sample through the so-called penetration length. From these it can be qualitatively shown, that the radiant energy transfer during IR welding provides deeper penetration than the conductive energy transfer which appears during hot plate welding. An additional quantity of importance is the net heat transfer across the length of the sample. The radiative energy transfer contribution to the sample is always maintained at a roughly constant rate. However, the conduction energy transfer contribution for hot plate welding varies as a function of the interface temperature. Once the temperature of the surface increases, the heat-transfer flow rate into the sample decreases. In other words, for hot plate welding not only does the low thermal conductivity reduce the heating process, but also the decaying heat flow from the hot 82 plate to the part help reduce the rate of heating. Therefore, IR does provide a better heating mechanism for plastic joining. 5.3 Penetration Depth The maximum possible penetration depth or the penetration limit of IR welding has not been studied. It is important to understand the penetration capability of the IR welding process compared to purely heat conduction penetration to determine if additional information on penetration is required. Therefore, the heating mechanism of IR welding including heat conduction and radiation must be analyzed by FEM analysis. The penetration energy is measured by the experimental setup of the radiation field. To measure the penetration energy a series of samples whose thickness varied from 0.025 cm to 3.25 cm were prepared. By referring to Figure 3.5, the shape and dimensions of the samples are determined by the rectangular opening (2.75 cm x 0.94 cm) in the middle of the aluminum plate attached to the detector head. Therefore, the cross-sectional area of the sample must coincide exactly with the rectangular opening in the aluminum plate, thus filling it completely and ensuring a tight fit, as can be seen in Figure 5.1. When the radiation energy hits the surface of the 83 Figure 5.1 Aplastic specimen fills the rectangle shape of aluminum plate 84 polymer, the transmission energy is detected and measured by the detector. The actual transmission power is then read on a power/energy meter. The rectangle shape required is the exact size of the cross-sectional area of the welded specimen. The polymer specimen is fixed on the aluminum plate and fills the whole rectangle shape as shown in Figure 5.1. 5.4 Estimating the IR Reflection of Polypropylene To evaluate the reflection of energy from the surface of the sample, thick (1.1 cm) section was prepared. The shape and dimensions of the sample are determined by the rectangular opening (2.75 cm x 0.94 cm) in the middle of the aluminum plate attached to the detector head as shown in Figure 5.1. When the radiation energy hits the surface of the polymer, the actual transmission power is then read on a power/energy meter after approximate 12.5 seconds. Once the power was turned off, the hot polymer sample was put into water immediately. A magnetic stirrer excited a magnetic mixing bar to keep the temperature uniform in the water. The effect of increasing the water temperature through stirring was measured and it was found to be less than 0.1°C over 60 seconds. The temperature of the water was measured by a thermocouple until the temperature stopped rising. Finally, The absorbed energy and the fraction of reflected energy can be obtained: Qpp— PwCw(Twater - Troom) Vvv + p«rCpp(Tpp - Troom)Vpp Qin= t(Wtotal - Wtrails) Oin - Opp 1 0 0 t Wtotal (5.4) where Qin and QPP are the theoretically absorbed energy and actual absorbed energy relatively, t is the radiant heating time, p is density, C is heat capacity, Wtotai is emitted power, Wtrans is transmitted power, Twater and TPP are equal to the final temperature in water, r is the fraction of reflected energy. In this measurement, the heating distance is 0.352 inch. The heating times are 12.56 and 12.45 seconds for different volumes of water, 65 and 34 cm3, relatively. The compared results for different volumes of water are shown in Table 5.1. 86 Table 5.1 The evaluation of fraction of reflected energy Vw Wtotai (watts) Wtram(watts) t (seconds) T* room TV r ) / 7 water,!' PP ( C) QPP(joules) Qin(joules) r (%) 65 20 2 12.56 23.1 23.9 218.64 226.08 3 34 20 2 12.45 22.9 24.4 214.65 224.1 3.7 During this test and analysis, the fractions of reflected energy r were 3% and 3.7% with respect to the water volumes 65 and 34 cm3. It is reasonable to assume that in plastic the reflected energy is negligible. This method was conducted to better understand for very limited effect on the reflected energy. Based on this investigation the reflected energy from the surface of the polymer is neglected for the next theoretical section. Moreover, as compared to the theoretical part the temperature and melting layer measurements will confirm this assumption. 87 Sample Thermocouple Magnetic Stirring Bar Water O O Magnetic Stirrer Thermometer Figure 5.2 A schematical setup for the measurement of the reflected energy from the surface of polypropylene 5.5 Theoretical Considerations A >1f Figure 5.3 Total radiant energy When a body is irradiated, part of the incident radiant energy is reflected, part absorbed, and the rest transmitted. The incident energy I is 88 equal to the sum of the reflected (R), absorbed (A) and transmitted energies (T): I = R + A + T (5.5) In the case of an infinite-body approximation, the penetration depth of the radiation into the material is much less than the length of the body [18]. Thus, the transmitted energy can be neglected. Assuming that in plastic the reflected energy is negligible, it can then be concluded that the absorbed energy is approximately equal to the incident energy. Consequently, I = A (5.6) It is clear that almost all the radiative energy is transferred to the polymer. In generating the differential equations which describe the radiative (infrared radiation) heat transfer mechanism the energy balance concept as well as the rate equation is employed. In order to conceptualize an energy balance equation one must first look at the system with its energy inputs and outputs and generate a mathematical statement of energy conservation. In this case the mathematical statement is that 89 Ec E in Figure 5.4 A physical boundary including energy inputs and outputs. Ejn + Eg Ec + Eout (5.7) Ein is energy coming in. Eg is internally generated. Ec is change in internal energy. Eout is energy coming out. Energy conducted in + heat generated within element = change in internal energy + energy conducted out 90 The energy terms are as follows: Energy in = Ein = -kA — ■ + (1 - a)q (5.8) dX Energy generated within element = Eg = 0 (5.9) Change in internal energy = Ec = pCpA— dX (5.10) Energy out = Eout = -kA-^J + (-kA-^)dX+(1 - a)q + — 1 ^ dX (5.11) 8X 8X dX dX where k is thermal conductivity, A is the cross-sectional area of the sample, q is radiative flux and a is fraction of absorbed energy. In the cases of polymers (i.e. PP and PBT) these the fraction of absorbed energy have non-linear dependence on the thickness as shown in Figure 5.5, 5.6, and 5.7. In the cases where the heating distance varies the fraction of absorbed energy maintains the same trend. This is an indication that the fraction of absorbed energy is independent of the heating distance. Therefore, radiative flux can be separated from the differential term as a constant value. 91 0.89 0.9 0.8 0.7 0.6 0.5 0 .4 0.3 0.2 0.25 0.5 0.75 Thickness(cm ) Figure 5.6 Nonlinear curves of natural color PBT penetration as a function of thickness at two different heating distances 0.89 cm and 1.5 cm 92 1 .2 v 05 Jsj 0.3 0.2 -• 0.1 ■ ■ 0 4 h — 1 h 1------1------1------4 ------0 0.0125 0.025 0.0375 0.05 0.0625 0.075 0.0875 0.1 Thickness (cm) Figure 5.7 Nonlinear curves of black PBT penetration as a function of thickness at two different heating distance 0.89 cm and 1.5 cm 93 0.9- • / 0.8 “ 0.7“ Sd I 0.6. W ■O 0.2 - 0.1 -H ------4------1------1------1— ------1------1------0.0125 0.025 0.0375 0.05 0.0625 0.075 0.0875 0.1 Thickness (cm) Figure 5.7 Nonlinear curves of black PBT penetration as a function of thickness at two different heating distance 0.89 cm and 1.5 cm Figure 5.8 One-dimensional radiative heat transfer in plastics The energy balance equation can now be written directly by combining the above mentioned relations, (5.8), (5.9), (5.10), (5.11). x=d . 52T q da k— — +—— = pCp-^ X>0,t>0 (5.12) dX2 A ax X =0 where p and k are assumed constant and CP is changed with temperature. This radiative and conductive heat transfer equation is suited for PP and PBT, but may be extended to other polymers as well. This is similar to the radiative and conductive heat transfer model for glass [19-22]. The heat transfer equation in glass is given by: k A aQr ot (B13) dX2 ox P at v ' where Qr is the radiative flux 95 As an example in the use of the boundary conditions, a simple heat conduction problem, involving a change of phase, can be solved. Assume a semi-infinite plastic sample extending over positive x. Initially, the temperature T0 is room temperature, and the heat flux q(x,t) is given for both penetration depth d>x>0 and time t>0. If h is the convection heat- transfer coefficient, the boundary conditions are - k |£ = h(T-TJ X=0, t>0 (5.14) dX T(oo,t)=T0 X=oo t>0 (5.15) Define x=s(t) to be the location of the melt line . The melting temperature will be chosen to be Tm for this problem. Hence T[s(t),t]=Tn (5.16) Furthermore, from heat balance across the melt line, H = /p C p(T)dT (5.17) where p is the density and L the latent heat of melting. It should be noted that the specific heat can be temperature dependent. Therefore, the latent heat, which is related to the specific heat, is obtained by defining the enthalpy H, which has units of heat/volume, of the material as a function of temperature. The enthalpy H is defined as the integral of density times specific heat with respect to temperature shown as Equation (5.16). The initial condition is T(X,0) = T0 (5.18) In the case of polypropylene, the material properties are [23] : _ 2 watts Thermal Conductivity: 0.1176x10 7cm 3 ^ C Density: 0.9—^ cm _ 4 watts Heat Convection Coefficient: 4.5x10 cm 20 f C Latent Heat: 100 ^°U cm Heat Capacity: 4.187(0.3669 + 0.00242T) —^ Melting Point: 165°C 97 This mathematical model can be calculated using a finite-element package, ANSYS. This package can be used to analyze transient heat conduction problems with phase change effects. The computer program for calculation of radiation heat transfer in the case of polypropylene is shown in Appendix B 5.6 Measurement of Melting Zone During heating, the melting zone is clearly visible on the specimen. Consequently, it is possible to determine accurately the molten layer. For this purpose, the velocity of the melt front can be tracked by using a Sony video camera. The molten layer profile is thus recorded on video tape. 5.7 Temperature Measurement A preliminary study of the temperature measurement has been done. Because the radiation energy penetrates into the polymers, the imbedded thermocouple wire absorbs radiation and heats up. Therefore, the resulting temperature readings are higher than the actual sample temperature. In order to avoid this problem a manual method was employed using two thermocouples as shown in Figure 5.9. Thus, the actual temperature of PP can be measured during IR heating. Thermocouple #1 is fixed to the underside surface of the PP sample. As the IR energy hits the sample the temperature exhibited by thermocouple #1 98 Infrared Energy power off 1 h V y Moving Thermocouple 200 200 180 y 160 140 120 A / 1 ^ 100 2 / j- s' § 8 0 1 a / .r~' V-/ v/ S 60 40' a 1 20 0 5 10 15 20 Time (seconds) Time (seconds) Figure 5.9 Effect of infrared energy at heating distance 0.89 cm on the imbedded thermocouples 99 rises. At this time thermocouple #2 is manually held near the underside surface of the sample, thus explaining the oscillation of the temperature readings as shown in Figure 5.9a. After a certain time the incoming IR radiation is stopped and thermocouple #2 is embedded into the sample as close as possible to thermocouple #1. Once the power was turned off, as seen in Figure 5.9b, the temperature reading from thermocouple #1 exhibits a drastic decrease within 1 second, to finally reach the value shown by thermocouple #2. Therefore, at this point both thermocouples show the same temperature reading, which is actually the true temperature of the sample at this point in time. As time increases and cooling ensures, both thermocouples show identical readings. From the above described measurements it can be concluded that the thermocouple wire requires 1 second to lose the absorbed energy. Therefore, after 1 second the thermocouple reads the actual temperature of the polymeric sample. Keeping this in mind, the actual heating history can be determined as follows. Two thermocouples were embedded within the sample at a distance of 1.25 cm and 2.5 cm from the surface exposed to the thermocouple. The temperature histories were then determined for different heating times: 7, 10, 14 and 25 seconds. Taking into account the 1 second thermocouple reaction time the temperature of the sample for 100 each specific heating time was taken at that heating time plus 1 second. For example, if the heating time was 7 seconds the temperature of the sample at 7 seconds was taken as that corresponding to 8 seconds. Another interesting fact shown in Figure 5.10 is that once the power has been turned off the temperature readings drop off drastically. This dramatic cooling rate occurs for only 1 second, followed by a more gradual rate. This 1 second is actually the thermocouple response time, thus proving that 1 second is truly the thermocouple reaction time and that the manual method described in the previous paragraph is valid. Using the determined temperature values the heating history can be generated as shown in Figure 5.15. 5.7 Results and Discussion In the infinite-body approximation, the incident IR radiation becomes progressively absorbed as it travels through the specimen. As shown in Figure 5.11 the depth of penetration depends on the heating distance. The contribution of the total incident flux which originates from a surface layer to the depth of penetration is the radiative energy term in 101 120 1 GOwtfflmmmmmmmM 1.25 1.25 140 100 120 2.5 100 2.5 40' 10 Tim e (sec) Time (sec) (a) 7 seconds (b) 10 seconds 180* iiSSSKSBBSSSSKSSKS} (c) 14 seconds (d) 25 seconds Figure 5.10 Temperature history at two different locations (1.25 cm and 2.5 cm) for different heating times : (a) 7 seconds, (b) 10 seconds, (c) 14 seconds, and 25 seconds. 102 0.5 0.75 T hickness (cm ) Figure 5.11 Experimental results of penetration depth as function of thickness the radiative and conductive heat transfer equation. Thus, the results from the computer simulation, Figures 5.12 and 5.13, show surprisingly good agreement between the computer simulation and experimental results. It should be noted that the experimental results are recorded as shown in Figure 5.14. Also, in order to achieve acceptable weld quality it is essential to have a sufficiently thick film of molten material in the weld zone. Experiments show that when the joining efficiency reaches 90% the molten zone of the parts is approximately 1.5mm (1/16 inch). As expected, both figures show an increase in the thickness of molten layer with increasing heating time. Note that this relationship is consistent for the two different heating distances (0.89 cm and 1.5 cm) investigated using PP. The IR heating mechanism for polymers is different from many other materials. Radiation penetrates into the polymer and increases the temperature inside the polymer. Figure 5.15 shows good agreement between computer simulation and measured heating histories of PP samples for two different heating distances, 0.89 cm and 1.5 cm. 5.8 Summary 1. Radiative energy affects not only the surface of the sample, but also its interior. 2. Penetration of radiative energy transfer is the important mechanism for heat transfer in IR welding. 104 3 5 3 0 2 5 E xperim ental 20 R e s u l t s 0.5 Melting Zone (mm) Figure 5.12 Comparison of theoretical and experimental melting zones at heating distance 0.89 cm 105 50 45 40 35 p e r i m e n t a l R e s u l t s55,S 30 i E H 25 M) / C o m p u t e r 20 S i m u l a t i o n 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 Melting Zone (mm) Figure 5.13 Comparison of theoretical and experimental melting zones at heating distance 1.5 cm 106 Figure 5.14a Melting zone 0.5 mm, under the heating time 15 seconds 5.14b Melting zone 1 mm, under the heating time 20 seconds 4 (c) Melting zone 1.5 mm, under the heating time 24 seconds Figure 5.14 The increase of melting zone under the conditions of heating distance 0.89 cm, heating times (a) 15 sec., (b) 20 sec., (c) 24 sec. 250 1.25 200 i theory experim ent 1 5 0 1 2.5 o 100 50 10 15 20 25 Tim e (sec) (a) Heating Distance 0.89 cm 300 theory 250 experim ent 1.25 200 2.5 50 100 50 0 10 20 30 40 50 Tim e (sec) (b) Heating Distance 1.5 cm Figure 5.15 Comparison of theoretical and experimental temperature histories at different locations (1.25 cm and 2.5 cm) at different heating distances (a) 0.89 cm, (b) 2.5 cm 109 3. Deep-penetration-type IR welding indicates that sufficient energy creates fast and uniform heating through the thickness direction, low temperature gradient, and low residual stress within the welds. 4. All the energy is absorbed within the penetration depth with no transmitted energy for PP and PBT because the penetration depth is much smaller than the length the samples. 5. Based on the negligible reflection of polypropylene, the electromagnetic radiation emitted by an IR source transfers its radiation energy totally into the plastic part. The agreement obtained when comparing experimental and theoretical results (Figures 5.11 and 5.12) also shows that neglecting the reflected flux is a valid assumption. 6. An important point for energy reflection is radiative penetration. For example, for metals the reflective component can not be neglected because only a small amount of energy penetrates the metal surface. This means that less energy can be absorbed by the metallic medium. Plastics absorb radiation better. The radiative energy can even penetrate through the polymeric material if the sample is thin and/or the medium is transparent. 7.The effect of IR radiation on heating of the thermocouple wire has been determined. For PP one should use a one second after the power is turned off to read the actual sample temperature in the sample. CHAPTER VI IR WELDING OF POLYPROPYLENE 6.1 Introduction Plastics play an important role in daily life. Moreover, in many cases, complex structure of plastics or multimaterial parts can be made more easily through joining than by complex moldings. Therefore, new joining technology is very important for making complex parts and assemblies quickly and at low cost. For some thermoplastics butt joining is extensively used to fabricate load supporting structures. However, the problem is how to choose which welding method generates the best welds. Typically, for butt joining, the surfaces to be joined are heated, melted, and then pressed together. For instance, gas pipe or thick plate can be joined by traditional hot plate welding. On the other hand, IR welding can 110 Ill which are so common to hot plate welding [24,25]. Both IR and hot plate joining methods are butt joining methods. One major difference between IR and hot plate joining methods is that IR can join high temperature plastics in a very short cycle time, while the cycle time for hot plate welding is much longer. Therefore, it is suggested that IR welding could become in the near future an extension of butt joining methods. This chapter describes how joining parameters interrelate with each other in order to generate a high welding factor, ( joint strength/bulk strength) reaching approximately 1.0 for Polypropylene (PP). The important parameters which have been identified are heating time, change-over time, and applied pressure. Each of these parameters may affect the joint strength as well as the microstructure of the joint. Therefore, the morphology of PP welds needs to be studied using optical microscopy. This technique can be used to characterize the weld line, flow layer and spherulites which grow within the welded samples. As stated in the previous section, the focus is on the joint strength with emphasis on how one can obtain high quality butt joints. At this point it is pertinent to ask, when during welding are unsatisfactory joints generated, during the heating, change-over, or forging phase? How do these phases affect the joints? Is there any way to improve the joint 112 strength? This study seeks to answer some of the above questions for IR welded PP joints. The results are part of a larger study of IR welding, which includes the influence of welding conditions. The welding parameters were varied as follows: heating times: 7 seconds - 46 seconds, heating distance: 0.25 cm - 1.5 cm, change-over time: 0.8 seconds -1.5 seconds, forging pressure: 0.44 MPa - 1.18MPa. 6.2 Welding Procedures Polypropylene specimen were prepared as coupons with dimensions of 15 cm x2.75 cm x 0.94 cm as shown in Figure 6.1a. Sample cleaning and polishing were not required in IR welding. The sample length was cut precisely to insure a constant heating distance for the joining samples. The heating distance is defined as the distance from the surface of the specimen to the edge of the reflector. During welding the specimen was fixed using a finger clamp. During heating, the IR modules were moved in position and radiated the surface of the PP samples. After heating, the IR modules were retracted and the specimen were brought into contact and 113 allowed to cool under pressure. After sufficient cooling, for approximately 30 seconds, the welded sample was removed from the IR welder. 6.3 Mechanical Testing Destructive testing enables one to obtain an estimate of the strength of the welded joint. To obtain statistical data on the joint strength, at least five samples should be used for the preparation of specimens. These specimens were machined to produce tensile bars which conform with ASTM D638-91. It is possible to use the router to machine the samples into a dogbone shape as shown in Figure 6.1b. After machining, the surfaces were polished with fine emery paper. After tensile testing the welding factor, A was calculated as follows [26]: ^ tensile strength of welded material tensile strength of bulk material Tensile tests were carried out on an Instron tensile testing machine at room temperature and at constant crosshead speed of 0.5 cm min'l. The butt welded specimens had a cross-sectional area of approximately 1.875 cm by 0.94 cm with a gage length of 2 inches and with the weld bead left on the top and bottom surfaces. 114 6.4 Microstructural examination The main features of the microstructural examination were as follows. For reflected light microscopy, thick (0.3125 cm) sections were cut with a saw. The specimens were cut to enable investigation of transverse and longitudinal sections of the weld. To be consistent, for all specimens transverse and longitudinal sections were removed from the central area. The resulting polymer specimens were then mounted in room temperature cure epoxy resin. The samples were then ground and polished for micrographic examination. In the as-polished condition, all specimens were etched using a chromic acid etchant. The 6M chromic acid was prepared by dissolving chromic oxide (CrzOs) flake in distilled water[27]. The welded polypropylene specimens were then submerged in the. chromic acid solution and maintained in an oven at 90°C in a tank for times ranging up to 2 hrs [28]. After treatment with chromic acid at 90°C the specimens were washed with water and then immersed in acetone for at least 1/2 hr. Micrographic (magnification x50 to xlOO) examination of cross sections was performed to adequately characterize the welded as well as the unwelded areas. This microscopic technique was used to investigate 115 2.75 cm 0.94 cm 15 cm (A) v_ y Weld bead (B) Figure 6.1 (a) sample dimensions and (b) the dogbone shape of the welded specimen 116 and analyze the causes of unsatisfactory welds that have contributed to brittle fractures during tensile testing. The required microscopic analysis was performed using reflected light microscopy. 6.5 Results and Discussion 6.5.1 Mechanical Testing Tensile tests were used to examine the mechanical properties of the welded joint. These tests were performed in accordance with the ASTM D- 638 standard test method for tensile properties of plastic. A series of welds were made over a range of heating distances and times, while maintaining a constant joining pressure (0.44 MPa). Increasing the distance between the lamps and the samples increases the spread of the IR radiation and reduces the flux density. As shown in Figure 6.2, as the heating distance increases, the heating time increases substantially for the same joint strength. Therefore, the heating time required to create the same molten layer thickness increases relative to heating distance. However, as the heating time increases, not only the thickness of the molten layer, but also the surface temperature, increases. Therefore, Figure 6.2 shows that the joint strength initially increases with increasing heating time for the three different heating distances until a maximum is reached. It also shows that the joint strength decreases when the heating 117 time increases beyond the maximum strength point. Prolonged heating may possibly cause polymer degradation and/or vapor oxidation entrapment within the polymer. At the same maximum joint strengths for three heating distances, Figure 6.3 shows the molten layer thickness to be the same in all these cases. During the joining phase, finding an optimum joining pressure is mandatory for obtaining high quality joints. Figure 6.4 shows the effect of pressure on joint strength for a heating distance of 0.25 cm and change over time of one second for different heating times. In accordance with Figure 6.4, a heating time of 7 seconds is too short and by increasing the welding pressure more of the molten polymer is squeezed out, making the joint weaker. On the other hand, a heating time of 15 seconds is too long. However, increasing the pressure increases the joint strength for both heating times of 11 and 15 seconds until a maximum is reached. This is probably due to squeeze out of degraded polymer, while a certain amount of molten polymer still remains available to form a good joint. By further increasing the pressure, the joint strength is reduced. As in hot plate welding [ ] this is because less molten polymer remains at the joint and adverse molecular orientation due to flow increases. In addition, a heating time of 11 seconds gives the optimum strength, equal to the bulk strength. 118 41.34 34.45 ■ 0.25 0.89 27.56. -a 20.67. oo 13.78. 6.89 0 10 20 30 40 50 H eating T im e(sec) Figure 6.2 Effect of heating distance on joint strength for PP; welding pressure constant at 0.44 MPa, change-over time 1 second. 119 •■j rrstttpsw?; m m m (a) Melting zone 1.5 mm at heating distance 0.25 cm and heating time 11 sec. (b) Melting zone 1.5 mm at heating distance 0.89 cm and heating time 24 sec. 120 (c) Melting zone 1.5 mm at heating distance 1.5 cm and heating time 41 sec. Figure 6.3 The melting zone at different heating distances: (a) 0.25 cm, (b) 0.89 cm, (3) 1.5 cm 121 bulk 11 sec. I ■15 sec. 27.56- • 7 sec. f , 20.67- 13.78- • 6.89 0.34 0.589 1.378 Pressure (MPa) Figure 6.4 Effect of forging pressure on joint strength for PP; heating distance 0.25 cm and change-over time 1 second Another dominant parameter is change-over time. The shorter the change-over time, the higher the joining quality. Figure 6.5 shows that reducing the change-over time increases the joint strength for a heating distance of 0.25 cm, heating time of 11 seconds, and welding pressure 65 psi. It shows that a change-over time of 0.8 seconds generates an optimum strength equal to or slightly exceeding the bulk strength for PP. In contrast to Figure 6.5 which shows the effect of change-over time for a constant pressure, Figure 6.6 shows the effect of welding pressure and change-over time on joint strength for a heating distance of 0.25 cm and a heating time of 11 seconds. Under these welding conditions, Figure 6.6 also shows that for different change-over times the influence of welding pressure on joint strength does not increase or decrease monotonically. In the change-over phase, convective and conductive cooling of the molten polymer occurs. If the optimum molten layer thickness required to achieve the maximum strength is reached during heating, then by increasing the pressure the joint strength is reduced. 6.5.2 Assuring High Joining Quality The effects which generate variations of joining quality during IR welding should be recalled once more. Prediction of the joining efficiency is very important. In this study, a limited investigation was performed to describe how high joining quality can be generated. 123 37.9 34.4 A l 31.0 — [ 27.6 24.1 20.7 17.2 6 55 1 3.8 1 0.3 6.89 3.4 + 1— 0.5 1 1.5 Change-over Time(sec) Figure 6.5 Effect of change-over time on joint strength for PP; heating distance 0.25 cm, heating time 11 second, and pressure 0.44 MPa. 124 41.3 34.4 *s 0.8 sec* 1 sec! 27.6 I) 20.7 13.8 6.89 0.34 1.03 1.38 Pressure (MPa) Figure 6.6 Effect of forging pressure and change-over time on joint strength for PP; heating distance 0.25 cm and heating time 11 second. 125 6230 5340 4450 8 1780 890 0 0.25 0.5 0.75 1.25 1.5 1.75 2 2.25 Displacement (cm) Figure 6.7 Load-deformation curve for unwelded polypropylene specimens 126 During tensile loading of the specimens, the deformation characteristics up to the onset of necking were studied visually. To understand joining quality, tensile testing machines are equipped to plot, the load carried by the specimen versus its deformation. Figure 6.6 shows a typical load-deformation curve for an unwelded PP specimen. After M, the maximum load, the specimens may undergo what is called necking where highly localized strain takes place [30]. Necking invariably continues until all of the narrow central section has become oriented or until some serious flaw is encountered. If allowed to continue, the necking process encounters the greatly increased cross section of the specimen ends. The strength of the highly oriented necking section is usually not large enough to overcome the resistance of the enlarged section at the ends of the specimen, thus at this point the necking process ceases and rupture occurs at R. However, in principle, if insufficient heating or overheating takes place, necking would not occur for all IR welds. Examination of the strength of an IR welded joint over its section shows that the maximum load would never exceed the maximum tensile load M as shown in Figure 6.8. For instance, analysis of the joint strength for PP make it possible to understand the relationship between heating time and joining strength. Figure Brittle 6.8 behavior under the condition of 1.5 cm,distanceheating forgingpressure 0.44 MPa. heating 41time seconds, change-over 1 second, time and Tensile Force (Newton) -445 445 890 2229 3119 2670 1335 1780 356G .5 . .5 0.2 0.15 0.1 0.05 Displacement(cm) .5 . .5 0.4 0.35 0.3 0.25 The data shown in Figure 6.8 were generated under the following conditions: heating distance of 1.5 cm, heating time 41 seconds, change over time one second, and a forging pressure 0.44 Mpa and it indicates that this response is similar to a typical curve for brittle plastics. The joint shown in Figure 6.8 failed prior to necking. The reason for failure of welded joints is that some insufficient joining areas or defects exist across the cross section of the joint, as well as weld line and notch effects. This was also confirmed later on in the study of the microstructure of the PP welds. Some parameters cause unqualified failure according to the investigation of the microstructure of the PP joints. The heating time, change-over time, and pressure during joining affect joining quality and probably create joint defects. As a result of studying the effects of failure for weak joints, it is possible to optimize the welding parameters and reach 100% of the bulk strength as a final goal for joint strength. A good joint can be developed by improving the parameters to achieve a welding factor of approximately 1.0. The deformation behavior of a PP welded specimen is illustrated in Figure 6.9. As can be seen, the load at failure is close to the maximum value obtained in the unwelded specimen. Once it reaches the maximum value, a neck begins to form which propagates along the tensile specimen. A significant difference in the welded case is that necking appears only in 6230 5340 44501 © 3560 2670i 1780« 890 0.25 0.5 0.75 Displacement (cm) Figure 6.9 The deformational behavior of the high quality join in g polypropylene 130 a localized area near the joint. From the data presented above it is apparent that satisfactory welds can be produced for PP. At this point this result confirms that a high welding quality, which would reveal necking under a tensile test, is assessed in terms of a high welding factor, approximately 1.0. 6.5.3 Microstructural Testing 6.5.3.1 Qualitative Evaluation It is doubtful that the weld microstructure will be the same as that of the bulk material in view of the existing large thermal gradients caused by the infrared welding process. Certainly, the microstructural features of polypropylene welds have been clarified in terms of different zones due to flow phenomena in PP and the resulting small spherulites. It can be seen that in the coarse spherulite zone, the spherulitic morphology of the parent material changes as is illustrated in Figure 6.10. Figure 6.10 shows the cross section micrograph of a welded PP sample with only 86% joining strength. For low strength samples there is a visible weld fine in the middle of the weld zone. Therefore, as shown schematically in Figure 6.11, the weld may be idealized as consisting of six zones: the weld line, the transcrystallization zone, the small spherulites zone, the temperature gradient fine-grained spherulitic and nuclei structure zone, the deformed material zone, and the coarse spherulites zone. In the middle of the weld weld line Figuure 6.10 Polarized optical micrography on polypropylene surface u n d e r the condition; heating time 7 sec., change-over time 1 sec., and forging pressure 0.44 MPa ( x 50 ) 1. weld line 2. transverse recrystallization zone 3. small spherulites 4. temperature gradient fine-grained spherulites and nuclei structure 5. deformed material 6. coarse spherulites Figure 6.11 Six regions weld zone in the polypropylene welds zone is the weld line. A transcrystallization zone exists right next to the weld line. A small spherulite layer is adjacent to the transcrystallization zone, followed by the small spherulite layer which is referred to as the recrystallized layer with fine-grain spherulites and nuclei structure. In this layer, the size of fine-grained spherulites gradually reduces and forms the nuclei structure toward the deformed material, which represents the fifth region. Finally, the sixth region is the bulk material with coarse spherulites. Generally, it can clearly be seen that the optical characteristics of the joint zones are significantly different from those of the adjacent bulk material. Examination of the infrared welds made using different welding conditions indicates that the joining strength is related to heating time, change-over time, and forging pressure. Changing weld conditions modifies the heat affected zone (HAZ). A good example of the weld fine and transcrystallization zone which can be obtained in high magnification (xlOOO) is shown in Figure 6.12. Under some conditions, the weld line also disappears in some areas throughout the whole weld zone as shown in Figure 6.13. The only difference in welding conditions between the samples shown in Figures 6.10 and 6.13 is the heating time. The sample displayed in Figure 6.13 exhibiting a joint strength of 94% of the bulk strength at an optimum heating time of 11 seconds. Furthermore, Figures PLEASE NOTE Page(s) not included with original material and unavailable from author or university. Rimed as received. UMI 135 Figure 6.13 Polarized optical micrography on polypropylene surface under the conditions; heating time 11 sec., change-over time 1 sec., and forging pressure 0.44 MPa (x50) 6.14 and 6.15 show the complete weld zone in the transverse and longitudinal sections. Since the weld line is not visible the joining strength increases to 94% of the bulk strength. If the sample is overheated, notch effects at either end of the weld, as shown in Figure 6.16, may reduce the joint strength. This is probably due to squeeze out of the molten polymer and oxidation vapor entrapment. Some vapor probably remains at the edges of the samples as a result of the notch effects. It should be noted that the notch remains within the part width, so removal of the flash would not remove the notch. Infrared welding provides an excellent opportunity to generate high quality welding of PP. By changing the change-over time, the joining strength can reach maximum values. For low change-over times, Figure 6.17 clearly shows the complete disappearance of the weld line. This factor probably generated the high joint strength. Figures 6.18 and 6.19 show the temperature gradient recrystallized layer with fine-grained spherulites. This layer is due to fast recrystallization as result of conduction cooling into the bulk material. Further testing was done to evaluate this temperature gradient fine-grained spherulite region. The ideal procedure for this study consists of three stages. First, the sample is heated by IR source until a melt layer of a desired thickness is formed. Second, the sample is freely cooled down to room temperature without applying > Figure 6.14 The complete weld bead in the longitudinal direction under the heating time 11 sec., change-over time 1 sec., and pressure 0.44 MPa. LO Figure 6.15 The complete weld bead in the transverse direction under the heating time 11 sec., change-over time 1 sec., and pressure 0.44 MPa. u> 00 139 Figure 6.16 Polarized optical micrography on polypropylene surface under the conditions; heating time 15 sec., change-over time 1 sec., and forging pressure 0.44 MPa (x50) Figure 6.17 Polarized optical micrography on polypropylene surface under the conditions; heating time 11 sec., change-over time 0.8 sec., and forging pressure 0.44 MPa (x50) 141 Figure 6.18 Polarized optical micrography on polypropylene surface under the conditions; heating time 11 sec., change-over time 0.8 sec., and forging pressure 0.44 MPa (x200) 142 Figure 6.19 Polarized optical micrography on polypropylene surface under the conditions; heating time 11 sec., change-over time 0.8 sec., and forging pressure 0.44 MPa (x200) 143 forging pressure and finally in the third and last stage the sample is optically examined. On the basis of the results presented for IR welding so far, it is expected that the temperature gradient recrystallized layer with fine-grained spherulites will also appear in hot plate welding. Therefore, the same procedure for microstructural examination in the case of hot plate welding was done. The resultant microstructures are shown in Figures 6.20 and 6.21. In both cases, hot plate and infrared welding, the temperature gradient recrystallized layer with fine-grained spherulites structure exists near the bulk material. In accordance with Figures 6.20 and 6.21, it can be predicted that spherulites near the bulk material do not have enough time to recrystallize due to the fact that heat rapidly diffuses into the bulk material. This is similar to what happens on the surface of the plastic parts [30,31], which have small spherulites whose growth is determined by a quench rate declining with depth, since rapid quenching of the surface is usually involved in the formation of a transcrystalline zone [31-34], This transcrystalline zone also occurs in welds with relatively low joint strength as illustrated in Figure 6.12. Figures 6.22 and 6.23 show fine-grained spherulites, nuclei structure, and a deformed structure. 144 6.5.3.2 Quantitative Evaluation For quantitative analysis the weld zone can be measured. As mentioned, micrographic examination of the weld cross section can be made through the complete weld zone. Then one finds the center of the entire cross section and measures the width of the Heat Affected Zone (HAZ). The Heat Affected Zone (HAZ) is defined as the region from one side of the deformed structure to the other side of it. Table 6.1 shows the range of conditions used to produce nominally 86%, 94%, 100% joining quality infrared welds of PP. By simply reducing the change-over time to 0.8 seconds, under the optimum heating time of 11 seconds, a HAZ of 20 pm and joint strengths exceeding the bulk strength were formed. Increasing the forging pressure to 150 psi, resulted in a HAZ of 16.4 pm and a joint strength reaching the bulk strength. However, in the case of the 19.2 pm HAZ the joint strength obtained was found to be 84% of the bulk strength due to notch effects. Tensile testing shows that 11 seconds is the critical value for obtaining the highest quality. Therefore, measuring the HAZ width represents another way to analyze quantitatively the effect of heating time on joint strength. Table 6.1 shows that insufficient heating time creates a larger HAZ. Overheating results in the smallest weld zone. For long heating times, the combination of low viscosity with relatively high forging pressure, causes the size of the HAZ to diminish due to excessive squeeze out. 145 Figure 6.20 Polarized optical micrography on polypropylene surface under the conditions; heating time 11 sec., without forging pressure in IR welding (x200) 146 Figure 6.21 Polarized optical micrography on polypropylene surface without forging pressure in hot plate welding (x200) Figure 6.22 Polarized optical micrography on polypropylene surface under the conditions; heating time 17 sec., change-over time 1 sec., and forging pressure 0.44 MPa (xlOOO) Figuure 6.23 Polarized optical micrography on polypropylene surface under the conditions; heating time 7 sec., change-over time 1 sec., and forging pressure 0.44 MPa ( x 1000 ) -t— 00 6 Summary The validity of the proposed IR joining method has been established by analysis of the parameters for PP welding. A systematic study of tensile joint strength and microstructure of IR welded PP samples has shown that a broad range of welding parameters may be used successfully for analysis. The results obtained indicated that PP is exceptionally well suited for IR welding, and may provide an appropriate base for developing a suitable welding procedure for high or low temperature polymers with good welding characteristics. The IR joining method enables the development of a simple and stable joining process which can generate a high welding factor exceeding 0.9. By controlling the joining parameters, such as heating distance, heating time, change-over time, and joining pressure, a welding factor of approximately 1.0 can be obtained. The best way to choose the joining parameters is to start with the heating distance. 150 Table 6.1 Melting zones with variation of welding parameters Heating Time Change over porgmg Pressure Joining Strength HAZ(pm) (seconds) (seconds) (psi) (100 %) 7 1 65 50 86 11 1 65 26 94 11 0.8 65 20 100 11 1 150 16.4 100 7. After choosing the proper heating distance, the heating time which can provide a high welding factor, greater than 0.9, is chosen. The gage pressure employed is 0.44 MPa, which practically speaking is the lowest applicable pressure. 8. Under this heating time by controlling the pressure or change-over time a welding factor of 1.0 can be obtained. Generally speaking, for different heating times, the required pressure to generate a welding factor of 1.0 is also different. Higher forging pressure will squeeze out too much molten material from the weld region. However, if the heating time is too long a thicker molten layer is generated and thus higher pressure is needed in order to maintain a high quality joint. 9. If the change-over is as short as possible, the required pressure is lower. 10. Microscopic evaluation of PP welds has been conducted to demonstrate that improvements in the welding process are needed. 11. The weld zone can be divided into six regions. These are the weld line, the transverse recrystallization zone, the small spherulites zone, the temperature gradient recrystallized layer with fine-grained spherulites and nuclei structure, the deformed material zone , and the coarse spherulites zone. 12. Tensile testing is used to evaluate PP weld strength. Although this test is used as an acceptance criterion to evaluate joining strength, it may not describe detailed microstructural mechanisms. Nevertheless, morphology evaluation can be obtained from reflected light microscopy with additional information from complementary examination techniques, such as transmitted light microscopy, scanning electron microscopy, and transmission electron microscopy. 13. The technique developed throughout this study for PP can easily be adapted as a standard test procedure for quality evaluation to improve proper welding guidelines. CHAPTER VII IR WELDING OF POLYBUTYLENE TEREPHTHALATE (PBT) 7.1 Introduction Sticking of the polymer to the hot plate is a frequent problem in hot tool welding of high temperature thermoplastics. Therefore, for welding high temperature polymers IR replaces hot plate welding by switching to a non-contact heating method. In order to prove this concept, a thermoplastic polymer PBT, which can withstand the high temperatures required for the application of IR welding, is selected. Polybutylene Terephthalate (PBT), a thermoplastic polyester of high temperature resistance has been in use for over 20 years. This high- performance thermoplastic polymer has a desirable combination of 152 153 mechanical properties, and high quality surface finish [35]. Moreover, PBT outperforms metals such as steel, brass and aluminum because of lower friction coefficient and higher strength per weight ratio [36]. Therefore, to date PBT is a commonly used plastic material for many applications such as electrical, telecommunications, automotive, household appliances, food processing industry (i.e., piston pump and valve body), food and medicinal packaging and precision engineering [37]. Most applications for plastics require some assembly. Welding methods are well suited for parts that need to be assembled. By virtue of the well-balanced properties of PBT and its overall application in almost all branches of industry, it is useful to study IR welding of this high temperature thermoplastic material. IR welding is especially suited for joining of high temperature thermoplastics. Recent advances in aerospace and other technologies have produced a continuing and growing need for materials which will withstand prolonged exposure at high temperature and/or relatively short exposure at even higher temperatures. The needs have become particularly acute for high temperature materials used in electrical and thermal insulation, gaskets, flexible tubing, and many other items manufactured from fibers, films, and plastics. Since these materials are 154 increasingly being used in complex structural assemblies, the importance of joint integrity and cost-effectiveness should be addressed. Polybutylene Terephthalate (PBT) is a commonly used high temperature polymer which is difficult to join using hot plate welding. Plastic automobile bumpers manufactured from Polybutylene Terephthalate can be joined by vibration welding [38,39], and are reparable using hot gas welding [40]. The repaired bumpers provide satisfactory results in service and exhibit comparable strength to the undamaged bumper. Due to the fact that the PBT sample had not sustained serious damage and was reparable for many applications, it is possible to use IR welding to join PBT. IR welding provides joining quality for high temperature thermoplastics without any sticking problem. 7.2 Experimental Procedures In this study, two extreme from the colorant point of view cases, were investigated, natural (cream white) and black PBT (KR 4036) which are thermoplastic polymers of an unfilled or unreinforced grade. Ultradur is the tradename for semicrystalline polyester (PBT) manufactured by BASF. It should be noted that the black PBT is supplied with 0.5% carbon additive, which does not affect the mechanical properties of PBT. The goal of this study is to determine how pigmentation affects the heating mechanism. This allows the possibility of achieving excellent and rapid joining when pigmentation is present. In addition, this information can be used to increase productivity and reduce costs. In particular, when pigmentation is present in thermoplastics, the power level of the radiation energy can be reduced. For example, less heating time is required for black material than for non-colored material. This example indicates that natural color thermoplastics can be replaced by colored thermoplastics. Also, the black material can be welded using a relatively low power level of the radiation. Both natural and black color Polybutylene Terephthalate (PBT) specimens were prepared as coupons with dimensions of 15 cm x 2.8 cm x 1.25 cm inches as shown in Figure 7.1. Sample cleaning and polishing were not required in IR welding. The sample length was cut precisely to ensure a constant heating distance for all samples. For black color PBT, the power level varied from 100% to 62.5 %, while for natural color PBT the power level was maintained at 100%. During welding the specimen was fixed using a finger clamp. During heating, the IR modules were moved in position and radiated the surface of the PBT samples. After heating, the IR modules were retracted and the specimens were brought into contact and allowed to cool under pressure. After cooling for 156 approximately 30 seconds, the welded sample was removed from the IR welder. Five samples were welded for each set of parameters. 2.8cm 1.25cm 15 cm Figure 7.1 The dimensions of the specimen 7.3 Microscopic Evaluation The morphology of PBT welds can be studied either by microtoming or an etching technique. Due to difficulties in observing the microstructures of bulk samples of PBT, relatively little effort has been devoted to relating the properties of semicrystalline plastics to their microstructures. Microstructures in polybutylene terephthalate (PBT) can be revealed by (i) sectioning polymeric samples and observing the sectioned microstructures by small-angle light scattering or transmission microscopy [41-42]; (ii) etching polymers in a manner analogous to the acid etching of metals, (iii) other methods including scanning electron micrography [42-43], X-ray diffraction [44-46], and DSC analysis [44-47]. The etching method is simple, economical and desirable. Because of the 157 difficulty in sectioning on a micro scale, the fine spherulites in the weld bead of a PBT sample may not be distinguishable. In addition, the microstructures of welded black PBT specimens may not be easily revealed by simply sectioning and polishing the sample. Therefore, the etching technique may prove to be a better way for examining the treated surface. The polymeric etchants slowly dissolve the amorphous regions from the polymer surface. Chromic acid (teOa) was used in this study to reveal the crystalline microstructure of the welded PBT. The amorphous areas were attacked and removed at a very slow rate to uniformly reveal the microstructure. Selective attack is limited to the surface. Microtomy in transmitted light provided morphological information concerning PBT welds. 7.4 Mechanical Testing Butt welded samples were prepared for tensile testing following ASTM D-638, assuring consistency in testing. Dimensions of testing samples (dogbone) were constrained by the thickness and curvature of the parts. Final specimen geometry used in the tests yielded a 5 cm gage length with a cross-sectional gage area of approximately 1.875 cm by 1.25 cm as shown in Figure 7.2. 158 Tensile testing specimens were prepared by machining the butt welded samples to the same dimensions as indicated by the standard. The tensile testing results were averaged over five samples and a standard deviation was calculated. The tensile testing results were done on an Instron tensile testing machine TT-D was used to determine tensile properties of the joints at room temperature. A crosshead speed of 0.5 cm/min was employed. 5 cm- T ------1.875 cm I V Weld bead Figure 7.2 The dogbone sample dimensions 7.3 Results and Discussions 7.3.1 Effect of welding conditions on tensile strength Tensile strengths of IR welded high temperature natural and black colored PBT are shown in Figures 7.3, and 7.4 respectively. The joining conditions in both cases were: heating distance 0.89 cm, change-over time 159 6 2 . 0 . 5 5 . 1 . IL0 . 8 9 c m 4 8 . 2 . V 4 1 . 3 . • white (distance 0.89 cm, change-over 0.8sec, 0.44 M Pa) 3 4 . 4 . 2 7 . 6 . 2 0 . 7 . » 1 3 . 8 . 6 . 8 9 . 1 0 . - — i ------! ------S ------1 ------1 ------8 ------1 ------0 5 10 15 20 25 30 35 40 Heating Tim e(sec) Figure 7.3 Joint strength for high temperature thermoplastics cream white PBT 160 CO OZ.Uft I 55.1 . 48.2 . IUr • * f t 4 1 3 • m r black (distance 0.89 cm 34 4 ■ J 1 ’ | change-over 0.8 sec.,0.44 MRa) 0) f I GO£ 27 6 • 20.7 . 13.8 • 6.89 i 0 • 0 5 10 15 20 Heating Time(sec) Figure 7.4 Joint strength for high temperature thermoplastics black PBT 161 0.8 second, and a pressure0.44 MPa. Figures 7.3 and 7.4 show that by increasing the heating time the weld strength increased until a maximum was reached. Once the weld strength has reached maximum, overheating produces a decrease in joint strength for natural color PBT, while only a slight decrease is observed for black PBT. Although the natural color required longer heating times, it provides better joints than the black PBT. Both heating distance and power level showed significant effect on the joining strength of black PBT. As shown in Figure 7.5, upon increasing the heating distance for black PBT, the joining strength improves slightly. On the other hand, for the same heating distance a lower power level significantly increased the weld strength as shown in Figure 7.6. As the heating distance decreased or the power level increased, more surface heating occurred for the black PBT, thus not generating a sufficiently thick molten layer. Therefore, reducing the power level and increasing the heating distance is required. As expected, the natural color PBT provided better subsurface heating, thus achieving strong weld joints with short cycle times. Comparing Figures 7.6 and 7.3, it is apparent that at a large heating distance and a low power level the heating time for the natural color PBT is greater than the heating time for the black PBT, which also exhibits a relatively lower joint strength. 162 Heating Time (sec.) Figure 7.5 Effect of the heating distance on joint strength for black PBT welding pressure 0.44 MPa, change-over time 0.8 sec 163 68.9 » » 62.0 55.1 75% power level 62.5% p o w er le v e l 100% p o w er le v e l 482 1.5 cm 1.5 cm 1? 41.3 ECL B 34.4 & 1.5 cm 8 27.6 20.7 13.8 6.89 i ” 1 " I - 3 10 20 30 40 50 60 70 Heating Time (sec) Figure 7.6 Effect of the power level on joint strength for black PBT welding pressure 0.44 MPa, change-over time 0.8 sec Likewise, the heating distance and power level affect the joining quality of natural color and black PBT. Consider other parameters that affect the joining quality for both cases. A typical example of a major influential parameter is the forging pressure during the cooling phase, as shown in Figures 7.7 and 7.8. As was observed with IR welds of PP, there is also an optimum pressure for natural and black colored PBT. At low pressures, it is possible that not all of the degraded polymer has been squeezed. A further increase in pressure resulted in the reduction of the molten layer thickness and an increase in adverse molecular orientation, therefore decreasing the strength. One important observation concerning the effect of joining pressure is that 0.88 MPa is the optimum pressure required to achieve intimate contact for both natural and black colored PBT. Even if the heating distance, heating time, or color changes, the optimum pressure does not change. It seem that this optimum pressure may be due to the viscosity or flow of the molten PBT. Another important parameter is change-over time. Higher joining quality is obtained for shorter change-over time. Figure 7.9 shows that for a change-over time of 0.8 second, over 90% of the joint strength for PBT was reached. During the change-over time, convective and conduction cooling of the molten polymer occurred. 165 Figure 7.7 Effect of pressure on joint strength for cream white PBT, heating distance 0.25 cm, heating time 29 sec., change-over time 0.8 sec 166 Forging P ressure (MPa) Figure 7.8 Effect of pressure on joint strength for black PBT, heating distance 0.25 cm, change-over time 0.8 sec 167 Change-Over Time (sec) Figure 7.9 Effect of change-over time on joint strength for black PBT, heating distance 0.25 cm, heating time 57.6 sec., power level 62.5%, and forging pressure 0.88 Mpa 168 Consequently, by reducing the change-over time, conduction and convection losses decrease, thereby conserving a sufficiently thick molten layer which in turn results in a higher joint strength. 7.3.2 Evaluation of PBT Welds The maximum joint strength for PBT was approximately 90% of the bulk tensile strength and the related microstructure was revealed using reflected and transmission microscopy as shown in Figures 7.10 and 7.11 respectively. The weld was etched using chromic acid at 90°C for 2 hrs and only four zones were observed. These regions are the weld line, small spherulites, deformed material, and coarse spherulites. By sectioning a PBT weld, as shown in Figure 7.12, the weld line and deformed material were observed using transmission microscopy. However, when using reflected microscopy it was more difficult to reveal the spherulites when the magnification was less than 200X. This implies that not all of the amorphous material on the sample surface of PBT was removed after the 2-hour chromic acid etching. Therefore longer etching times are needed. PLEASE NOTE Page(s) not included with original material and unavailable from author or university. Rimed as received. UM1 170 Weld Line Figure 7.11 Optical Micrograph of the microtome sectioning PBT (x50) 171 Figure 7.12 Optical Micrograph of the microtome sectioning PBT (x200) As the etching time was prolonged to four hours at 90°C, both the weld line and small spherulites were observed at 200X as shown in Figure 7.13. To better reveal the weld line and small spherulites at 400X, as shown in Figure 7.14, the etching time was increased to 6 hours at 90°C. Based on the observations of the welds prepared by the aforementioned progressive-etching technique, it is possible to use conventional reflection microscopy, in addition to transmission microscopy, in order to study PBT weld microstructures with various degrees of etching. As shown in Figure 7.15, deformed material was observed adjacent to the bulk material. Elongated and deformed spherulites were found in the deformed material as can be seen in Figures 7.16 and 7.17. Next to the deformed material, there exits a region containing fine-grained spherulites which were believed to be influenced by the temperature gradient. As for the microstructure of the bulk material, the morphology of the spherulities is shown in Figure 7.18, which is in accordance with similar observations previously illustrated in published literature [47-48]. Another useful technique, which plays an important role in the study of polymers is scanning electron microscopy (SEM). Figure 7.19 shows an SEM evaluation of the bulk material, which further ascertains that very small spherulites of approximately 2 pm are uniformly distributed. These small spherulites of 2 pm are probably the so called nuclei structure. 173 Weld Line Small Spherulites Figure 7.13 Optical Micrograph of PBT after 4 hr Etching Treatment (x200) 174 Weld Line Small Spherulites Figure 7.14 Optical Micrograph of PBT after 6 hr Etching Treatment (x400) 175 Deformed Material Figure 7.15 Optical Micrograph of PBT after 6 hr Etching Treatment (x400) 176 Temperature gradient coarse spherulites Fine-grained spherulites Deformed Material Small Spherulites I I Figure 7.16 Optical Micrograph of the sectioned PBT (x400) 177 temperature gradient fine-grained spherulites deformed coarse weld line material small spherulites spherulites 4, ^ I Figure 7.17 Optical Micrograph of the microtome sectioned PBT (x400) 178 Figure 7.18 Optical Micrograph of the microtome sectioned bulk PBT (xl500) 179 Figure 7.19 Scanning electon microscopy evaluaiton of PBT after 6 hr etching treatment (x2000) 180 It can be seen that five different zones were distinguished within the observed cross sections. These regions are the weld line at the center of the weld bead, fine spherulites adjacent to the weld line, temperature gradient fine-grained spherulites and nuclei structure, deformed material, and larger spherulites within the bulk material. 7.4 Conclusions 1. General-purpose PBT is one of the most widely used temperature- resistant thermoplastics. IR welding is suitable for joining PBT. 2. There are three key considerations to keep in mind when IR welding high temperature thermoplastic PBT, whether natural or black colored: (a) tensile strength of welds is almost the same as that of the base materials, (b) high temperature thermoplastics can be joined by IR welding, (c) joint strength is influenced by the following welding parameters: heating distance, heating time, change-over time, and welding pressure. 3. When comparing the IR welding process for natural and black colored samples, the colored samples exhibited significant differences during the heating stage resulting in a lower joint strength, around 70 % of the bulk strength. 4. For black PBT the heating distance and power level dominate the required cycle times and thus the resulting joint strength. Both transmitted (X1000) and reflected (X400) light microscopy can be used to examine the welds in PBT to reveal the welded region. Scanning electron microscopy provides an image of a layer, which in turn contain very small spherulites. The very small spherulites of approximately 2 jim are probably the so-called nuclei structure. Microstructurally the weld zone consists of five distinct regions: the weld line at the center of the weld bead, fine spherulites adjacent to the weld line, temperature gradient fine-grained spherulites and nuclei structure, deformed material, and larger spherulites within the bulk material. CHAPTER VIII IR WELDING OF GLASS FILLED POLYETHER SULFONE COMPOSITE 8.1 Introduction As previously shown the IR welding process has been employed to explore a new field for low and high temperature thermoplastics such as PP, and natural and black color PBT. Large area joining provides one special opportunity for IR welding in the future. Along with these considerations, industry may require knowledge of IR welding of polymeric matrix composites. Thus, there is a need for an established IR welding method which adresses the joining capabilities of composite materials. Composite materials are now a major field of research and development activity, and they are rapidly becoming important as structural materials. Therefore, the objective of this study is to present an analysis of the effects 182 of IR welding parameters on the joining properties and to develop a better understanding of the physical processes involved. At present, no comparable data exits which covers all the welding parameters, including heating time, change-over time, and forging pressure. Several publications [49-50] dealing with the IR welding process involve the comparison of three welding methods, hot plate, vibration and IR welding processes, for joining 30% glass fiber reinforced Polyether Sulfone (PES). In these comparative investigations, the IR welding method generated better joining results than either hot plate or vibration welding of 30% glass filled PES composite. Throughout these studies the effect of the IR welding parameters on joint strength was not performed. Therefore a complete study will be performed on the effect of IR welding parameters, while also performing a morphological evaluation of welds in glass reinforced PES composites. Polyether Sulfone (PES) is a high performance amorphous thermoplastic with very high resistance to heat. It is utilized for high- quality engineering parts exposed to high temperatures in applications where the thermal properties of other engineering plastics, e.g. nylon, and polycarbonate, would no longer suffice. Glass-filled grades of PES are used 184 for precise, electrical, and medical applications. It is also suited for aircraft components and pumps that operate in hot water. 8.2 Welding Procedures Polyether Sulfone composites with 20% by weight glass fibers JF- 1004, manufactured by LNP Engineering Plastics, were prepared as coupons with dimensions of 10 cm xl.875 cm x 0.0625 cm as shown in Figure 8.1. The samples were thoroughly dried prior to welding by placing them in an oven for a minimum of 6 hours at 200 °C. During welding the heating distance was kept at 0.12 cm. The sample length was cut precisely to insure a constant heating distance for all samples. The heating time was varied from 11 seconds to 16 seconds and the change-over time from 0.8 seconds to 2 seconds. The power level was maintained at 100%. During welding the specimen was fixed using a finger clamp. During heating, the IR modules were moved in position and radiated the surface of the PES samples. After heating, the IR modules were retracted and the specimens were brought into contact. The pressure with which the PES specimen were joined together was controlled using a pressure regulator connected to the pneumatic cylinder. The forging pressure was varied from 0.56 MPa to 1 MPa After cooling the samples for 30 seconds, the welded sample was 185 removed from the IR welder. Four samples were welded for each set of parameters. ,9cm 0.625 cm 10 cm Figure 8.1 The dimensions of the specimen 8.3 Mechanical Testing Welded and bulk material samples were tested in tension following ASTM D-638. All samples were machined to the dogbone shape, as determined by the thickness of the samples. This dogbone shape was machined by a router and the surface was polished with fine emery paper #240. The dogbone shape used in the test yields a 5 cm gage length with a cross-sectional gage area of about 1.25 cm by 0.625 cm as shown in Figure 8.2. For the welded samples the weld bead on the top and bottom of the sample were not removed and the weld was placed in the center of the gage area. 186 An Instron tensile test machine model #4468 with 44500 newton load cell was used to determine the tensile strength of the samples at room temperature. A constant crosshead speed of 0.5 cm/min and load range of 13350 newton were employed. The average tensile strength and standard deviation of the tested specimens were calculated. Weld bead Figure 8.2 The dogbone sample dimensions 8.4 Microscopic Evaluation Microscopic analysis was performed using an optical reflected microscope. It was used to characterize glass fiber orientation at the welded area. The observation area in either transverse and longitudinal direction as shown in Figures 8.3 and 8.4 was sectioned, ground, and polished for micrographic examination. 187 A Figure 8.3 A 20% glass-filled reinforced PES weld in longitudinal section Figure 8.4 A 20% glass-filled reinforced PES weld in transverse section 8.5 Results and Discussion 8.5.1 Tensile Testing Figure 8.5 shows the effect of heating time on joint strength for 20% glass-filled reinforced PES. The joint conditions were a heating distance of 0.12 cm, and a change-over time of 1 second, a forging pressure of 0.744 MPa. It can be seen in Figure 8.5 that increasing the heating time 188 110.24 s £ $> 96.46 • «9 t 9 82.68 ■ 68.9 ■ I l / f t £ I “ £ ■ f 55.1 • s \ m $I p 41.3 ■* ! Heating Distance: 0.12 cm Forging Pressure: 0.744 MPa I 27.6 ■ Change-over: 1 sec j j 5. r £ f 13.8 ■B ] I 0 ■ 5 7 9 11 13 15 17 19 21 Tim e (sec) Figure 8.5 Effect of heating time on joint strength for PES 189 increased the weld strength until an optimum was reached. For 20% glass- filled reinforced PES joint strengths of 84% of bulk strength were achieved for a very short heating time of 15 seconds. Heating beyond the optimum resulted in a slight decrease in strength for 20% glass-filled reinforced PES. Figure 8.6 shows that for change-over times of 1 and 0.8 seconds weld strengths exceeding 84% of the bulk strength were achieved for a heating distance of 0.12 cm, a heating time of 15 seconds, and forging pressure of 0.744 MPa. Therefore the optimum change-over time for obtaining a high joining quality is 1 second or less. Due to convection and conduction cooling of the molten polymer, increasing the change-over time beyond 1 second results in a decrease in joint strength. As was observed with IR Polypropylene and polybutylene terephthalate welds in previous chapters, increasing the pressure improves the joint strength until an optimum is reached. Figure 8.7 shows the effect of forging pressure on 20% glass-filled reinforced PES. These welds were prepared using a heating time of 15 seconds at a heating distance 0.12 cm and with a change-over time 0.8 second. In this case the optimum was achieved at a forging pressure of 0.744 MPa producing a weld strength of 75.79 MPa which is 84% of the bulk strength. Further increases in pressure beyond the 190 ■\W/AW#/MVM&MX*M&iW*fMWAV/&AWAWA 96.46 82.68 ■ • 68.9 1 E I 55.1 i 25 41.3 Heating Distance: 0.12 cm Heating Time: 15 se c 27.6 - • Pressure: 0.744 MPa 13.8 - 4 2 .50.5 1 1.5 2.50.5 Change-Over Time (sec) Figure 8.6 Effect of change-over time on joint strength for PES Figure 8.7 Effect of forging pressure on joint strength for PES strength on of joint Effect forging pressure 8.7 Figure Strength (MPa) 26 ■ • 82.68 64 ■■ 96.46 27.6 38 ■13.8 • 41.3 55.1 68.9 .1 .8 .5 .2 .9 .6 .3 .9 .6 1.03 0.96 0.89 0.83 0.76 0.69 0.62 0.55 0.48 0.41 -+■ -9- hneOe: . ec se 0.8 Change-Over: etn ie 1 ec se 15 Time: Heating cm 0.12 Distance: Heating Pressure (MPa) Pressure " 4 - -I 1.10 191 192 maximum result in reduction of the molten layer thickness and adverse molecular and fiber orientation, thereby decreasing the joint strength. 8.5.2 Microstructural Testing As shown in Figure 8.8, microstructural examination of the weld in the longitudinal section shows fiber orientation obtained for welds whose strength is 84% of the bulk strength. The optimum joining parameters employed were a heating time of 15 seconds, a heating distance of 0.12 cm, a change-over time 0.8 second, and forging pressure 0.744 MPa. Since glass fibers align themselves in the flow direction, then orient in thickness direction (the direction of maximum flow). Fiber orientation of 20% glass- filled reinforced PES were compared for the longitudinal and transverse sections in order to further understand the effect of IR welding on joint strength. Figure 8.9 shows the microstructural examination of the weld in the transversal section. As can be seen, few fibers orient in the transverse section as compared to the longitudinal section. Due to more flow in the thickness direction the glass fibers align themselves in this direction and thus are much more visible when a longitudinal section is performed. Due to glass fiber orientation in the thickness direction the strength properties of reinforced PES are lower in this direction, therefore welds should not be loaded at full capacity in the longitudinal direction. 193 8.6 Summary Based on this study the joint strength of 20% glass-filled reinforced PES is 84% of bulk strength, while for the joint strength of 30% glass-filled reinforced PES [49-50] similar results of 83% of the bulk strength was obtained. The microstructural examination of the weld indicate that the fiber orientation affects the joint strength of PES composite materials. The fiber orientation at the weld tends to be greater in the thickness direction as compared to the other directions thereby reducing the weld strength as compared with the bulk strength of the composite. 194 > Thickness direction Fiber _> Orientation Figure 8.8 Polarized optical micrography of 20% glass-filled reinforced PES weld in longitudinal section under the condition; heating time 15 sec., change-over time 1 second, and forging pressure 108 psi (xlOO) 195 Fiber - > Orientation Figure 8.9 Polarized optical micrography of 20% glass-filled reinforced PES weld in transverse section under the condition; heating time 15 sec., change-over time 1 second, and forging pressure 108 psi (x50) 196 CHAPTER VIIII CONCLUSIONS AND RECOMMENDATIONS 1. A prototype IR welding system was designed and constructed. 2. Evaluation of the IR welding process was performed using theoretical and experimental analysis, including mathematical model of the radiation field, simplified semi-empirical model of radiation field, heat transfer model, effects of process parameters on joint strength, and microstructural analysis. 3. The Combination of the modeling and measurement of the radiation field with the heat transfer model permits a thorough understanding of the heating during IR welding process. 4. This technique is very promising for welding of high temperature thermoplastics and their composites. 5. Although the black PBT performed well in IR welding, the effect of colorant additives on the IR welding process must be clearly understood. One must understand the concepts behind the process 197 and how the materials are affected by IR radiation before reaching a "press of a button" type conclusion. Heating time for black-colored PBT is much shorter than for cream white PBT. According to the present analysis, it seems that radiation absorption could be one reason why the heating time for cream white and black colored PBT is dramatically different. 6. The radiation absorption capability of cream white PBT decreases along the sample length and its penetration depth is relatively larger than in the case of black colored PBT. Conversely, this also implies that a large part of the energy is absorbed by the black PBT within a very small thickness layer. This drastic absorption generates the fast heating behavior and short welding cycle characteristic of black PBT. 7. IR heating may create new fields of investigation with potential applications such as thin film sealing and forming, repairing, large and/or small area joints. 8. A limitation of the IR joining process is the geometrical complexity of the parts. 9. A camcorder method was devised for judging weld quality by evaluating the size of the melting zone. As in any welding/bonding method which requires melting of the material at the interface, some of the melt may be displaced beyond the immediate joint area, thus 198 creating flash. A flash trap, volumetrically proportional to the amount of molten material being displaced during welding, can be incorporated into the joint design if the amount of molten material is known. 10. The most important joining parameter is the heating distance from the IR modules to the parts. When making a joint between dissimilar materials the distances from the IR source to each part are different. 11. All dissimilar thermoplastics that can be joined by other welding methods can also possibly be joined by IR welding. For example, using vibration welding, ABS can be welded to ABS, acrylic, and an ABS- polycarbonate alloy; acrylic can be welded to acrylic, ABS and polycarbonate. IR welding can possibly generate these dissimilar welds by adjusting the heating distance correspondingly for each type of polymer in order to provide sufficient energy to the dissimilar thermoplastics. 12. In the semi-empirical model for IR heating, the radiation fields inside multiple reflectors could be combined and evaluated as only one reflector. Theoretically, there is no limit to the number of reflectors which can be overlapped as is illustrated in Figure 9.1. Hence, IR heat panels to join a large workpiece could be successfully built and 199 evaluated, and would have significant economical and technological advantages. 13.IR is suitable for a wide range of applications from welding high heat resistant thermoplastics without a sophisticated heated-tool butt welding system, which may generate a sticking problem to adhesive bonding of thermoset plastics which could be photochemicaly or non- photochemicaly reactive. 7,0° Cot**5 APPENDIX A COEFFICIENT FOR INDIRECT RADIATION MODEL (1) Derivation of Dsins and Dcoss: D 2 = (X - 0.2325)2 + Y2 (1) Assume a point P on the reflector whose position is given by the coordinates Xp, Yp in regard to the global Cartesian coordinate sytem (X,Y) located at the chosen heating point I. as shown in Figure 3.3 D2 = (Xp + P + 0.4825 - 0.2325)2 + (Y„ + Q)2 -0.815-P D2 = (X P + P + 0.4825 - 0.2325)2 + (±0.3325 + Q)2 201 202 -0.4825- P Dsins = t = -[(XP + P + 0.25)sin<|)E -(Y P +Q)cos(j)E] Dcose = -[(Xp +P + 0.25 cos(|)E +(YP + Q)sincJ)E] (2) Derivation of tJ)E lij = sin0cos<|^ + sin0sin(|jf+cos£fic IiE = sin ©cos ^e^ h- sin 0 sin Unit normal vectors to the reflector at point P can be written as follows : [(Xp + P + 0.4825^ + (YP + Q)f] N P = (X p+P + 0.4825) 2 + (Y p+ Q )2 R R -0.815-P < Xp < -0.4825-P, where (6) -0.3325 - Q < YP < 0.3325 - Q w -0.4825-P 202 203 from the law of reflection, the angle between Ei, Np and Ee, Np will be equal: Thus, the dot product of these vectors for these angles will be Therefore, (m — 1 )sin -0.815 - P < Xp <-0.4825 - P, where (10) -0.3325-Q (YP+Q) m = c------XP+P + 0.4825 -0.4825 - P < Xp < P, -0.3325- Q 203 204 (3) Derivation of pi The coordinates referred to point I are deduced from the coordinates of point P where the energy from the IR source reflects off the reflector Xp = PjSinGcostj) Yp = pjSinGsinc)) (12) Zp = pj cos0 where 21°- {R2 - [(P + 0.4825)sin <}> - Q cos -0.815 - P < Xp < -0.4825 - P, (13) -0.3325 - Q < Yp < 0.3325 - Q and v . _ , ±0.3325-Q Xp = pj sin 0 cos (14) Zp = Pj cos0 where 204 205 +0.3325 -Q sin (|) -0.4825-P Note that rL2 > D2 sin2s >[(XP + P + 0.25)sin(|) - (YP + Q)cos The integration limits of a function of <1> result: P{(4>)sin4> + Q m(4>) = (())) cos (4) Derivation of C(4>), A(4>), E(4>),0 b2 , 0bi ,0A2, 0m , 0C2 ,and 0ci (r^-D W e)" = |■ rL22 - [(Pi(4>) cos 4> + P + 0.25)sin 4>e (4>) - (P{ 4> sin 4> + Q )cos4>E(4>) j j v 0.5 ( rL2 - D 2 sin2 e ) = A(4>) (17) 205 206 D cos s = -[(pj( eE,#) ] L K (19) 6^ )=ta„_,[Pi*+B#)-AA] K *1)^ ^ 4» < 4»b, (20) 2 0 .5 (rL2-D 2 in2 s )°5 = | rL2 - [ ( ^ + P + 0.25) sin 4>E(4>) - 0.3325 cos (21) D cos s = -r ( -• - - ^ + P + 0.25) sin <1>e(4>) + 0.3325 cos <|»B (<|>) 1 = D(4*) L tan.]) (22) 206 207 0.3325- Q _,.v sint + D^) - C () Qa@) = tan ~' L- k (23) 0.3325 + Q + p ^ _ c((j)) sin 4> 0A.C+) = tan ’ tt,A2 and 03 ■—D ^^95 _O ^ ^ ( rL2- D 2sin2s ) = {rL2- [ ( — _ _ ^ _ _ 1 + P + 0.25)sin<|>E(-<|>) — (-0.3325)cos<))E((jj)j J = E( (25) _0 3325 — o D cos s -[ ( tan, + P + 0.25 ) cos tJ)E ({]))+( 0.3325 ) sin -0 .3 3 2 5 - Q ,f— 5ta+ '+F® - E») 00,# )= tan ------(27) 207 208 Gc,#) = tan 1 ------K 208 A P P E N D IX B RADIATION HEAT TRANSFER PROGRAM /PREP 7 ETYPE STAT ET, 1,32 R, 1 RMOD, 1, 1,2.66e—4 MP , DENS,1,.9e3 MP,C,1,1.8e3 MP,KXX,1 , .1176 MPTEMP,1,25,164,165,166,167,800 MPDATA,enth,l,l,0.0,312e6,314.9e6,404.9e6,407.7e6,4076.8e6 N, . 0 N, .0005 N, .001 N, .0015 N, .002 N, .0025 N, .003 N, .0035 N, .004 N, .0045 N, .005 N, .0055 N, .006 N, .0065 N, .007 N, . 0075 N, .008 N, .0085 N, .009 N, . 0095 N, .01 N, .02 N, .04 E.1,2 EGEN,22,1,1 FINISH /SOLU ANTYP,4 TRNOPT,FULL,-999 STAT 1-999 above = only method changed, timint,off 209 CNVTOL , TEMP AUTOTS , ON TUN IF,25 time,1e-0 6 lswrite timint,on T IN T P , , , , 1 KBC , 1 TIME , 41 DE LTIM,0.01, , ,ON OUTPR,ALL OUTRES,AL L,10 F ,1,HEAT,0.65782 F ,2,HEAT,0.62493 F,3,HEAT,0.59204 F ,4,HEAT,0.55914 F ,5,HEAT,0.52625 F,6,HEAT,0.493364 F ,7,HEAT,0.46047 F,8,HEAT,0.42758 F ,9,HEAT,0.39469 F,10,HEAT,0.3618 F,11,HEAT,0.32891 F , 12,HEAT,0.296 F,13,HEAT,0.263127 F,14,HEAT,0.23023 F,15,HE AT,0.197345 F ,16,HEAT,0.16445 F ,17,HEAT,0.13156 F ,18,HEAT,0.09867 F ,19,HEAT,0.065782 F,20,HEAT,0.032891 D,23,TEMP,25 lswrite lssolve,1,2 finish /postl A P P E N D IX C UNIFORM ENERGY DISTRIBUTION CALCULATION By using Mathcad the result of uniform energy distribution can be calculated: Heating Distance : 2 inch (5 cm) z := 0,0.1..2 f(z) := 0.9989- 0.00928-z- 0.897 -z +■ 0.7133-z3 - 0.257-z + 0.046-z - 0.0033-z 1 f(z) 0.5 0 0 0.75 1.5 2.25 3 z Figure C.l Calculated result for power distribution at heating distance 2.5 cm 211 212 Rotation: z := -2,- 1.9.. 0 g(z) := (0.9989 + 0.00928 z) - 0.897-z2 - 0.7133-z3 - 0.257-z4 - 0.046-z5 - 0.0033-z g(z) 0.5 -2 -1.25 -0.5 0.25 1 z Figure C.2 Calculated result for power distribution Translation: z := 0,0.1..2 a := 2 h(z) := 0.9989 +- 0.00928-(z_ a) - 0.897-(z- a)2 - 0.7133-(z_ a)3 - 0.257-(z_ a)4 - 0.046-(z- a / - 0.0033-(z- af 1 0 1 01 2 3 z Figure C.3 Calculated result for translation of power distribution Finally, J(z) = f(z) + h(z) 2 J(z)i 0 0 1 2 3 z Figure C.4 Uniform power distribution Heating Distance: 3 inch (7.5 cm) z := 0,0.1.. 3 f(z) := (l + 0.0245-z- 0.3 51854-z2 + 0.1562-z3 - 0.0295-z4) -t- 0.002243-z 1 f(z) 0 1 2 3 40 Figure C.5 Calculated result for power distribution Rotation: z = -3,-2.9.. 0 g(z) (1 - 0.0245 z) - 0.351854-z - 0.1562 z - 0.0295-z - 0.002243- -4 -2.5 1 0.5 2 z Figure C.6 Calculated result for power distribution Translation: z := 0,0.1..3 a := 3 h(z) := (1 - 0.0245-(z- a)) - 0.351854-(z- a)2 - 0.1562-(z_ a)3 -0.0295 (z- a)4 - 0.002243-(z- a)5 1 0 -2 0 2 4 z Figure C.7 Calculated result for translation of power distribution Finally, J(z) := f(z) + h(z) J 1 1 0 l.f 1 1 1 1 n 0 1 2 3 4 z Figure C.8 Uniform power distribution REFERENCES 1. IR Lamp User Manual, The Eraser Co. Inc. 2. Robert Siegel, John R. Howell, Thermal Radiation Heat Transfer, Hemisphere Publishing Corporation, 1992. 3. The Use of High Intensity Infrared, Research Inc. 4. H. Potent, P. Michael, and M. Heil, Infrared Radiation Welding : A Method for Welding High Temperature Resistant Thermoplastics, ANTEC, 1991. 5. E. M. Silverman and R. A. Griese, Joining Methods for Graphite/PEEK Thermoplastic Composites, SAMPE Journal, 25, 5, 1989. 6. N. S. Taylor, F Klaiber, and J Wermelinger, Assessment of PP and PVDF Joints Made by a New Infrared Joining System, ANTEC, 2647- 56, 1993. 7. H. Potent, J. Natrop, T. Klit Pedersen, M. Uebbing, Comparison of Three Methods for Welding Glass-fibre Reinforced PES, ANTEC, 1274- 1279, 1994. 8. S. M. Jacob and J. S. Dranoff, Light Intensity Profiles in an Elliptical Photoreactor, AICHE journal, 15, 1, 1969.10.H. A. Irazoqui, J. Cerda, and A. E. Cassano, Radiation Profiles in an Empty Annular Photoreactor with a Source of Finite Spatial Dimensions, AICHE 217 218 Photoreactor with a Source of Finite Spatial Dimensions, AICHE journal, 19, 3, 1973. 9. J. Cerda, H. A. Irazoqui, and A. E. Cassano, Radiation Fields Inside an Elliptical Photoreflector With a Source of Finite Spatial Dimensions, AICHE journal, 19, 5, 1973. 10. H. A. Irazoqui, J. Cerda, and A. E. Cassano, Radiation Profiles in an Empty Annular Photoreactor with a Source of Finite Spatial Dimensions, AICHE journal, 19, 3, 1973. 11. O. M. Alfano, R. L. Romero, and A. E. Cassano, A Cylindrical Photoreactor Irradiation From the Bottom-I. Radiation Flux Density Generated by a Tubular Source and a Parabolic Reflector, Chemical Engineering Science, 40, 11, P 2119-2127, 1985. 12. O. M. Alfano, R. L. Romero, and A. E. Cassano, A Cylindrical Photoreactor Irradiation From the Bottom-II. Models for the Local Volumetric Rate of Energy Absorption With Polychromatic Radiation and Their Evaluation, Chemical Engineering Science, 41, 5, P 1155- 1161, 1986. 13. O. M. Alfano, R. L. Romero, and A. E. Cassano, A Cylindrical Photoreactor Irradiation From the Bottom-Ill. Measurement of Absolute Values of the Local Volumetric Rate of Energy Absorption Experiments With Polychromatic Radiation, Chemical Engineering Science, 41, 5, P 1163-1169, 1986. 14. M. A. Claria, H. A. Irazoqui, and A. E. Cassano, Modeling and Experimental Validation of the Radiation Field inside an Elliptical Photoreactor, Chemical Engineering Journal, 33, P 119-131, 1986. 15. O. M. Alfano, M. Vicente, S. Esplugas, and A. E. Cassano, Radiation 219 Field Inside a Tubular Multilamp Reactor for Water and Waste water Treatment, Ind. Eng. Chem. Res. 29, P1270-1278, 1990. 16. M. Vicente, O. M. Alfano, S. Esplugas, and A. E. Cassano, Design and Experimental Verification of a Tubular Multilamp Reactor for Water and Wastewater Treatment, Ind. Eng. Chem. Res. 29, P1278-1283, 1990. 17. S. Esplugas, 0. M. Alfano, M. Vicente, and A. E. Cassano, Effect of the Reflector Shape on the Performance of Multilamp Photoreactors Applied to Pollution Abatement, Ind. Eng. Chem. Res. 29, P 1283-1289, 1990. 18. A.Leupin, H. Vetsch and F. K. Kneubuhl, Investigation Comparison and Improvement of Technical Infrared Radiators, Infrared Phys., 30, 3, 199-258, 1990. 19.M.Quinn Brewster, Thermal Radiative Transfer & Properties, John Wiely & Sons, Inc., 1992. 20. N. D. Eryou, L. R. Glicksman, An Experimental and Analytical Study of Radiative and Conductive Heat Transfer in Molten Glass, Transactions of the ASME, May, 224-230, 1972. 21. E. E. Anderson, R. Viskanta, W. H. Stevenson, Heat Transfer Through Semitransparent Solids, Journal of Heat Transfer, May, 179-186, 1973. 22. S. H. Chan, D. H. Cho, G. Kocamustafaogullari, Melting and Solidification with Internal Radiative Transfer-A Generalized Phase Change Model, International Journal Heat Transfer Mass Transfer, 26, 4, 621-633, 1983. 220 23. J. Brandrup and E. H. Immergut, Polymer Handbook, third edition, A Wihey-Interscience Publication, New York, 1989. 24. E. Pecha, Welding of Plastics parts by the Hot Plate Method, IPE Internation, Industrial & Production Engineering, 10, 120-124, 1986. 25. V. K. Storkes, Joining Methods for Plastics and Plastic Composites an Overview, 29, 1310-1324, 1989. 26. G. Menges and J. Zohren, Plastverarbeiter, 18,165, 1967. 27. V. J. Armond and J. R. Atkinson, The Effect of Chromic Acid on Polypropylene, J. Material Science, 332-335, 1968, Letters. 28. M. J. Day and M. F. Gittos, The Application of Microscopy to Welded Joints in Thermoplastics, 390, 1-29, 1989. 29.C. B. Bucknall, I. C. Drinkwater, and G. R. Smith, Hot Plate Welding of Plastics: Factors Affecting Weld Strength, Polymer Engineering and Science, 20, 432-440, 1980. 30. P. J. Flory, Principles of Polymer Chemistry, Cornell University Press, Ithaca, New York, 482, 1953. 31.R. J. Barriault and L. F. Gronholz, Formation of Spherulitic Structure in Polyhexamethylene Adipamide ( Nylon). I. Structure and Optical Properties of Spherulites at Room Temperature, Journal of Polymer Science, XVIII, 393-403, 1955. 32.Linda A. Sawyer and David T. Gubb, polymer Microscopy, Chapman and Hall Ltd, New Yark, 1987. 33.D. Fitchmum, and S. Newman, Surface Morphology in semicryatalline Polymers, 7, 301-305, 1969. 221 34.T. K. Kwei, H. Schonhorn, and H. L. Frisch, Dynamic Mechanical Porperties of the Transcrystalline Regions in Two Polyolefins, 38, , 2512-16, 1967. 35. Booklet" Product Line Performance Guide PL-1" Hoechst Celanese. 36. Booklet" HYDEX 4101", A.L. Hyde Co. 37.Booklet" Ultradur ( Polybutylene terephthalate)", BASF Co. 38. First Issue of Joining & Materials, Welding Plastics Broadens Materials Application, Joining & Materials, July, 1988. 39. V. K. Stokes, Vibration Welding of Thermoplastics Part IV: Strengths of Polybutylene Terephthalate, Polyetherimide, and Modified Polyphenylene Oxide Butt Welds, Polymer Engineering and Science, 28, 998-1008, 1988. 40. J. R. Atkinson, and B. E. Turner, Rep air ability of Plastic Automobile Bumpers Hot Gas Welding, 29, 19, 1368-1374, 1989. 41. R. S. Stein and A. Misra, Morphological Studies on Polybutylene Terephthalate, Journal of Polymer Science, Polymer Physics Edition, 18, 327-342,1980. 42. S. Y. Hobbs and C. F. Pratt, The Effect of Skin-Core Morphology on the Impact Fracture of Polybutylene Terephthalate, Journal of Applied Polymer Science, 19, 1701-1722, 1975. 43. A. Ajji, J. Brisson, and Y. Qu, Orientation and Mechanical Properties of PBT and its Blends with a Liquid-Crystalline Copolyester, Journal of Polymer Science, Polymer Physics Edition, 30, 505-516, 1992. A 44. M. Yokouchi, Y. Sakaibara, Y. Chatani, H. Tadokora, T. Tanaka, and K. Yoda, Structurees of Two Crystalline Forms of Polybutylene 222 Terephthalate and Reversible Transition between Them by Mechanical Deformation, Macromolecules, 9, 2, 1976. 45. S. Y. Hobbs and C. F. Pratt, Multiple melting in Polybutylene Terephthalate, Polymer, 16, 462-464, June, 1975. 46. C. F. Pratt, and S. Y. Hobbs, Comparative study of Crystallization rates By D.S.C and Depolarization Microscopy, 17, 12-16, January, 1976. 47. B. Moginger, C. Lutz, A. Polsak, And F. Stuttgart, Influence of Processing on Mechanical Properties and Morphology of PBT. Kunststoffe German Plastics, 81, 3, 43-44, 1991. 48. B. Moginger, C. Lutz, A. Polsak, and Fritz Stuttgart, Einflub der Verarbeitung auf mechanische Eigenschaften und Morphologie von PBT, Kunststoffe, 81, 3, 1991. 49. H. Potent, J. Natrop, T. Klit Pedersen, M. Uebbing, Comparative Investigations into the Welding of Glass-Fiber-Reinforced PES, Jouranl of thermoplastics Composite Materials, 6, April, 147-159, 1993 50. H. Potent, J. Natrop, T. Klit Pedersen, M. Uebbing, Comparison of Three Methods for Welding Glass fiber Reinforced PES, ANTEC, 1274- 79, 1994