EJECTION FORCES AND STATIC FRICTION COEFFICIENTS
FOR RAPID TOOLED INJECTION MOLD INSERTS
DISSERTATION
Presented in Partial Fulfillment of the Requirements for
the Degree Doctor of Philosophy in the Graduate School of
The Ohio State University
By
Mary E. Kinsella, M.S.
*****
The Ohio State University
2004
Dissertation Committee:
Approved by
Professor Blaine Lilly, Adviser
Professor Jose Castro
Professor Jerald Brevick
______________________________
Adviser
Industrial and Systems Engineering
Graduate Program
ABSTRACT
While manufacturing is typically considered a high-volume industry, the necessity
for small quantities of products and components exists for aerospace customers and those
producers wishing to mass customize their products. Because of the high cost of tooling,
injection molding processes are seldom used to produce only small quantities of parts.
This, however, can be remedied if cost effective tooling methods are implemented.
Rapid prototyping processes show great potential for such tooling applications because
they generally require shorter lead times, produce less waste, and, in some cases, use less
expensive materials.
The research presented herein studies the feasibility of using injection mold
inserts produced with additive methods by investigating ejection and friction. Through
experimentation, the application of P-20 steel, laser sintered LaserForm ST-100, and
stereolithography SL 5170 tools to produce limited quantities of a thin-walled cylindrical
part are explored. A substantial amount of data and statistical analysis are provided that
reveal conditions during the actual injection molding process, and comparisons are made
among the three insert types. Experimental ejection forces from each tool type are
compared with model-based calculations, and apparent coefficients of static friction are
calculated and compared to standard test results. Based on the data and analyses, the
benefits and limitations of using rapid tooled injection mold inserts are presented.
ii
For Michael, Amelia, and Nathaniel
iii
ACKNOWLEDGMENTS
“Gratitude is not only the greatest of virtues, but the parent of all the others.”
--Cicero
At the Materials and Manufacturing Directorate in the Air Force Research
Laboratory, I am grateful to Charlie Browning, Bill Russell, and Chuck Wagner for
providing the time and funding to complete this research; to John Jones for assembling
and programming the data acquisition system; and to Neal Ontko, Nick Jacobs, and Ben
Gardner for performing friction tests.
At the NASA Marshall Space Flight Center, thanks to Ken cooper, who provided
the laser sintered and stereolithography inserts for the experimental work.
At The Ohio State University, many thanks to the following people: Brian
Carpenter for finite element modeling of the inserts for thermal and deformation
simulations, and for helping with experiments; Bob Miller for his machining and
injection molding expertise; Mary Hartzler for providing machining services and design
consultation; Barney Barnhart for providing equipment and expertise for the
thermoplastic tensile tests; Mauricio Cabrera-Rios for design of experiments and
statistical analysis consultation; Narayan Bhagavatula for helping with tensile tests and
injection molding simulations; and especially my adviser, Dr. Blaine Lilly, and Drs. Jose
Castro and Jerry Brevick for serving on my dissertation committee.
Finally, I extend my gratitude to my parents, Robert and Carolyn Corbin, who
deserve more than I can ever express.
iv
VITA
June 5, 1961
Born – Longview, WA, USA
1983
1983 - 1986
1987 - Present
Bachelor of Science, Applied Science
Miami University, Oxford, OH, USA
Production Supervisor, NCR Microelectronics
Fort Collins, CO, USA
Project Engineer, Materials and Manufacturing Directorate
US Air Force Research Laboratory
Wright-Patterson AFB, OH, USA
1991
Master of Science, Materials Engineering
University of Dayton, Dayton, OH, USA
PUBLICATIONS
Kinsella, M. E., Heberling, M. E. 1997, “Applying Commercial Processes to Defense
Acquisition,” National Contract Management Journal, vol. 28, issue 1, p. 11.
Kinsella, M. E., Lilly, B. L., Bhagavatula, N., Cooper, K. G. 2002, “Application of Solid
Freeform Fabrication Processes for Injection Molding Low Production Quantities:
Process Parameters and Ejection Force Requirements for SLS Inserts,” Proceedings of
the 13th Annual Solid Freeform Fabrication Symposium, Austin, TX, pp. 92-100.
FIELDS OF STUDY
Major Field: Industrial and Systems Engineering
v
TABLE OF CONTENTS
Abstract ........................................................................................................... ii
Acknowledgments.......................................................................................... iv
Vita .................................................................................................................. v
Table of Contents ........................................................................................... vi
List of Figures ................................................................................................ xi
List of Tables................................................................................................. xv
Chapter 1 Introduction .................................................................................... 1
1.1 Background ........................................................................................................... 1
1.2 Problem Statement ................................................................................................ 6
1.3 Research Objective.............................................................................................. 10
1.4 Research Description........................................................................................... 11
1.5 Organization........................................................................................................ 12
Chapter 2 Literature Search .......................................................................... 14
2.1 Ejection Force ..................................................................................................... 14
2.1.1. Ejection Force Models............................................................................... 14
2.1.2 Shrinkage ................................................................................................... 22
vi
2.1.3 Friction and Adhesion ................................................................................ 25
2.2 Rapid Tooling ..................................................................................................... 34
2.2.1 Background ................................................................................................ 34
2.2.2 Stereolithography and Laser Sintering for Injection Molding Tools........... 44
2.2.3 Summary .................................................................................................... 48
Chapter 3 Theory........................................................................................... 50
3.1 Thermoplastic Materials...................................................................................... 50
3.1.1 High Impact Polystyrene ............................................................................ 51
3.1.2 High Density Polyethylene ......................................................................... 54
3.2 The Adhesion Component of Friction ................................................................. 55
3.3 Ejection Force Model Derivation ........................................................................ 60
3.3.1 Model derivation ........................................................................................ 60
3.3.2 Additional Consideration for Strain............................................................ 65
Chapter 4 Experimentation ........................................................................... 66
4.1 Friction Testing ................................................................................................... 66
4.1.1 Friction Test Apparatus .............................................................................. 68
4.1.2 Test Matrix and Procedure ......................................................................... 70
4.2 Measurement of Elastic Modulus ........................................................................ 72
4.3 Injection Molding................................................................................................ 75
4.3.1 Mold Design and Materials ........................................................................ 75
4.3.2 The Injection Molding Process................................................................... 79
4.3.3 Design of Experiments ............................................................................... 84
vii
4.3.4 Experimental Procedure ............................................................................. 87
4.4 Set-up and Data Acquisition................................................................................ 88
4.4.1 Temperature Measurement and Thermal Model ......................................... 89
4.4.2 Ejection Force Measurement ...................................................................... 98
4.4.3 Diameter and Thickness Measurement ..................................................... 100
4.4.4 Calculation of Static Friction Coefficient ................................................. 101
Chapter 5 Results and Analysis .................................................................. 102
5.1 Injection Molding Experiments ......................................................................... 103
5.1.1 Experimental Results and Discussion....................................................... 103
5.1.2 HDPE Experimental Ejection Force Results............................................. 105
5.1.3 HIPS Experimental Ejection Force Results .............................................. 107
5.1.4 Experimental Ejection Force Results from the P-20 and ST-100 Inserts .. 109
5.1.5 Experimental Ejection Force Results from the SL 5170 and SL 5170/P-20
Inserts................................................................................................................ 109
5.2 Statistical Analysis ............................................................................................ 112
5.2.1 DOE Results............................................................................................. 112
5.2.2 Main Effects and Interactions................................................................... 112
5.3 Standard Friction Testing Results...................................................................... 120
5.3.1 HDPE Standard Friction Results .............................................................. 120
5.3.2 HIPS Standard Friction Results ................................................................ 124
5.4 Reliability of the Data ....................................................................................... 128
5.5 Calculation of Ejection Force Using the Model................................................. 130
viii
5.5.1 Calculated Ejection Force for HDPE........................................................ 133
5.5.2 Calculated Ejection Force for HIPS.......................................................... 134
5.5.3 Possible Sources of Error ......................................................................... 135
5.6 Calculation of Apparent Friction Coefficients using the Menges Model ........... 138
5.6.1 HDPE Apparent Coefficient of Friction Results....................................... 140
5.6.2 HIPS Apparent Coefficient of Friction Results......................................... 141
5.6.3 Apparent Friction Coefficient Results from the P-20 and ST-100 Inserts. 143
5.6.4 Apparent Friction Coefficient Results from the SL 5170 and SL 5170/P-20
Inserts................................................................................................................ 144
5.6.5 Comparing Calculated Friction Results to Standard Friction Test Results 146
5.7 Other Observations of Rapid Tooled Inserts...................................................... 150
Chapter 6 Conclusions ................................................................................ 156
6.1 Molding HDPE and HIPS with ST-100 and SL 5170 Inserts ............................ 156
6.1.1 Benefits and Limitations of Using Rapid Tooled Injection Mold Inserts .. 156
6.1.2 Friction and Ejection Force Considerations.............................................. 158
6.2 Using a Model to Determine Ejection Force and the Coefficient of Friction..... 160
6.3 Implications and Future Work........................................................................... 162
6.4 Summary........................................................................................................... 165
LIST OF REFERENCES ............................................................................ 167
Appendix A Data Tables............................................................................. 174
A.1 Tensile Test Data Table.............................................................................. 176
A.2 Modulus Look-up Table ............................................................................. 177
ix
A.3 Thermal Analysis Convergence Table........................................................ 178
A.4 Sample Experimental Data Set, All Runs ................................................... 182
A.5 Sample Experimental Part Dimensions (2 Runs Shown) ............................ 184
A.6 Experimental Data and Calculated Coefficient of Friction (Menges), Run
Average............................................................................................................. 185
A.7 Analysis of Variance Tables by Set ............................................................ 187
Appendix B Mold and Canister Drawings.................................................. 190
B.1 Part Drawing............................................................................................... 191
B.2 Mold Insert Drawings ................................................................................. 192
B.3 Mold Assembly Drawings .......................................................................... 205
x
LIST OF FIGURES
Figure 1.1: Importance characteristics for various tool types……………………….. . 4
Figure 2.1: A schematic of the laser engineered net shaping (LENS™) process…... 36
Figure 2.2: A schematic of the selective laser sintering process……………………. 38
Figure 2.3: A schematic of the stereolithography process…………………………...42
Figure 3.1: Polystyrene monomer……………………………………………………52
Figure 3.2: High impact polystyrene………...…………………………… ............... 53
Figure 3.3: Polyethylene monomer. .......................................................................... 55
Figure 3.4: High density polyethylene linear molecule. ............................................ 55
Figure 3.5: Thin-walled cylindrical pressure vessel. ................................................. 61
Figure 3.6: Section of the part and the core with associated stresses. ........................ 62
Figure 4.1: Schematic of friction apparatus............................................................... 68
Figure 4.2: Friction test apparatus: sled on plate specimen inside furnace and tester…
.................................................................................................................................. 69
Figure 4.3: Tensile testing apparatus with tube furnace............................................ 73
Figure 4.4: HIPS specimens after tensile tests.......................................................... 74
Figure 4.5: Elastic modulus at various temperatures for HDPE and HIPS. .............. 74
Figure 4.6: Sprue side of the MUD base mounted in the injection molding machine
with SL 5170 cavity insert. ....................................................................................... 76
xi
Figure 4.7: Core and cavity inserts, before final machining, made of SL 5170, P-20
steel, and LaserForm ST-100. ................................................................................... 76
Figure 4.8: Canister part with vent holes and no taper.............................................. 78
Figure 4.9: Sumitomo SH50M injection molding machine. ..................................... 79
Figure 4.10: Signal conditioner and computer with front panel for data acquisition,
and core side of mold with thermocouple and load cell sensor wires. ....................... 89
Figure 4.11: Thermocouple placement within core insert.......................................... 90
Figure 4.12: Representative thermal traces of the injection molding cycle............... 91
Figure 4.13: Graphs of the thermal analysis results for each material combination... 94
Figure 4.14: Representative ejection force traces ...................................................... 99
Figure 4.15: Digital pictures of HDPE canisters for measuring inside and outside
diameter. ................................................................................................................. 101
Figure 5.1: Experimental ejection force results for HDPE, all runs........................ 106
Figure 5.2: Experimental ejection force values for HIPS, all runs.......................... 108
Figure 5.3: Experimental ejection force results from the P-20 and ST-100 inserts. 110
Figure 5.4: Experimental ejection force results from the SL 5170 insert and the
combination SL 5170/P-20 insert. ........................................................................... 111
Figure 5.5: Main effects and interactions for HDPE with the P-20 insert............... 114
Figure 5.6: Main effects and interactions for HIPS with the P-20 insert................. 115
Figure 5.7: Main effects and interactions for HDPE with the ST-100 insert. ......... 116
Figure 5.8: Main effects and interactions for HIPS with the ST-100 insert. ........... 117
Figure 5.9: Main effects and interactions for HDPE with the SL 5170/P-20 insert.118
xii
Figure 5.10: Main effects and interactions for HIPS with the SL 5170/P-20 insert.119
Figure 5.11: Standard friction test results for HDPE; means and ranges shown in the
table. ....................................................................................................................... 122
Figure 5.12: Standard friction test results for HIPS; means and ranges shown in the
table. ....................................................................................................................... 123
Figure 5.13: Sample plot of load vs. time for HIPS on SL 5170 from elevated
temperature tests. .................................................................................................... 126
Figure 5.14: Sample plot of load vs. time for HIPS on P-20 from elevated
temperature tests. .................................................................................................... 127
Figure 5.15: Calculated values for ejection force for HPDE compared with
experimental values, averaged across all runs. ........................................................ 134
Figure 5.16: Calculated values for ejection force for HIPS parts from the P-20 and
ST-100 cores compared with experimental values, averaged across all runs........... 137
Figure 5.17: Calculated values for ejection force for HIPS parts from the SL 5170
core compared with experimental values, averaged across all runs. ........................ 137
Figure 5.18: Calculated values of the apparent coefficient of static friction for HDPE,
all runs. ................................................................................................................... 140
Figure 5.19: Apparent coefficients of friction calculated from experimental results
for HIPS, P-20 and ST-100 results only, and results from all runs. ......................... 142
Figure 5.20: Apparent coefficient of static friction for parts from the P-20 insert. . 143
Figure 5.21: Apparent coefficient of static friction for parts from the ST-100 insert.
................................................................................................................................ 144
xiii
Figure 5.22: Apparent coefficient of static friction for parts from the SL 5170 insert.
................................................................................................................................ 145
Figure 5.23: Apparent coefficient of static friction for parts from the SL 5170 core
with the P-20 cavity. ............................................................................................... 146
Figure 5.24: Average apparent coefficient of static friction for HDPE compared to
standard test results. ................................................................................................ 148
Figure 5.25: Average apparent coefficient of static friction for HIPS compared to
standard test results. ................................................................................................ 149
Figure 5.26: Defects in the SL 5170 core. .............................................................. 152
Figure 5.27: Simulation results of HDPE injection into SL 5170 insert, no packing.
................................................................................................................................ 153
Figure 5.28: Simulation results of HIPS injection into SL 5170 insert, no packing.154
Figure 5.29: Simulation results of HIPS injection into SL 5170 insert, with packing.
................................................................................................................................ 155
Figure A.1: Sample plots from tensile test data....................................................... 175
xiv
LIST OF TABLES
Table 1.1: Properties of tooling materials.................................................................... 9
Table 2.1: Vanguard System Specifications from 3D Systems.................................. 40
Table 2.2: LaserForm ST-100 (Sintered and Infiltrated) Material Properties from 3D
Systems. .................................................................................................................... 40
Table 2.3: SLA 250 System Specifications from 3D Systems. .................................. 43
Table 2.4: Vantico SL5170 Typical Properties (90-minute UV post cure). ............... 43
Table 4.1: Friction data for polymers on steel. .......................................................... 67
Table 4.2: Friction test matrix. ................................................................................. 71
Table 4.3: Injection molding machine specifications................................................ 80
Table 4.4: Injection Molding Parameters .................................................................. 82
Table 4.5: Typical data for Lutene-H ME9180. ........................................................ 83
Table 4.6: Typical data for BASF PS 495F. .............................................................. 83
Table 4.7: Experimental design for six sets, including process parameters. ............. 86
Table 4.8: Resulting convergence times from the thermal simulation. ..................... 93
Table 4.9: Input conditions for the thermal analysis.................................................. 93
Table 5.1: Experimental ejection force results for HDPE and HIPS according to
packing time, cooling time, and packing pressure parameters................................. 104
xv
Table 5.2: Results from the designed experiment indicating which factors had a
significant effect on ejection force. ......................................................................... 113
Table 5.3: Surface roughnesses of all plates (friction tests) and cores (injection
molding experiments).............................................................................................. 129
Table 5.4: Calculated values of ejection force for HDPE from the Menges equation
and experimental data. ............................................................................................ 131
Table 5.5: Calculated values of ejection force for HIPS from the Menges equation
and experimental data. ............................................................................................ 132
Table 5.6: Calculated apparent coefficient of friction results according to packing
time, cooling time, and packing pressure parameters. ............................................. 139
xvi
CHAPTER 1
INTRODUCTION
1.1 Background
Manufacturing is predominantly a high volume industry and is constantly striving
toward greater efficiencies at lower cost. A growing sector of manufacturers, both in the
aerospace and consumer markets, however, is targeting small quantity production to meet
customer needs. The demand for low volume production is strong in the aerospace
industry, where customer organizations such as the military services and NASA need
relatively small numbers of end products to accomplish their missions. Product variety is
generally higher, and lot sizes smaller, in the defense industry compared to other
industries. The U.S. Defense Department and other government organizations have been
focused on finding ways to build low volume products more cost effectively (e.g.,
Kinsella 2000). These organizations and their contractors must implement new methods
of producing extremely robust equipment in reduced time at reduced cost (Kaminsky
1996).
1
Small production quantities in the consumer market have historically applied to
prototypes and market testing. In the past decade, however, market forces have altered
the way industry looks at low volume requirements. Increasing product variety and
shorter product lifetimes have led to mass customization, in which products are designed
and made to order for individual customers, but produced by methods that still allow for
economies of scale. Mass customization is already evident in a number of industries,
including pagers, fiber optics, and blue jeans (Victor & Boynton 1998). The concept
continues to spread as customers become more particular and manufacturers become
more flexible.
Thermoplastic injection molding is inherently well suited to high volume
production requirements. A quality mold, running with material and process variables
under tightly controlled conditions, is capable of producing very large quantities of parts
with little or no manual intervention. When coupled with automated material feeding and
robotic part removal systems, injection molding operations can be extremely cost
effective over large production runs, turning out millions of components per year at a cost
of a few pennies per part. At these production scales, the cost of the tooling essentially
disappears, and the part cost results almost entirely from material, handling, and
overhead.
Minimum economic production quantities for injection molded parts are typically
large due to tooling costs, which are incurred at the beginning of the product life. Molds
are expensive, regardless of part size, and typically require production volumes of tens of
thousands of parts in order to amortize their costs. For this reason, injection molding is
2
generally feasible only when the total production run is large enough to recoup the cost of
the tool. Many design situations exist in which the complexity and versatility of injection
molded thermoplastics would be an ideal solution were it not for the high initial cost of
the tooling. For many low volume applications, the ability to use the best engineering
solution is inhibited by the inability to cost effectively produce the necessary tooling.
Injection molds expressly designed for low volumes have successfully been
fabricated for prototyping use and as bridge tools. Small numbers of prototypes are
typically built to test out a design for fit and function, and to allow changes to be made
before tooling designs are finalized. Since so many consumer products use injection
moldings, it is often necessary to build prototype tools to produce the design prototypes,
especially in cases where the prototype must be fully functional to answer questions of
strength, rigidity, etc. Bridge molds, on the other hand, are built and put into production
very quickly. That is, they produce a small volume of parts prior to the completion of
the final tooling, thus “bridging” the gap between prototype tools and final production.
While prototype tooling can reduce time to market by accelerating the product
development cycle, bridge tooling is designed to get a new product to market quickly
while the high volume production tooling is still under construction. In both cases,
reduced time to market is the prime consideration, and the cost of the tooling will
eventually be folded into the total tooling cost and amortized over the lifetime of the
product.
As opposed to prototype and bridge tooling, molds intended specifically for low
volume production have many fewer products over which to amortize costs. In low
3
volume production environments, tooling must be low cost, unless the product is very
expensive. Time to market is a secondary consideration in this scenario. For these
reasons, the strategies for determining tooling methods will differ between prototype or
bridge tooling and low volume production tooling.
To distinguish low volume production tools from prototype, bridge, and high
volume tools, four characteristics typically dominate: cost, durability, cycle time, and part
quality. Part quality means, for example, surface roughness, residual stress, and
dimensional accuracy. The importance of these characteristics varies, depending on the
tool type, as shown in Figure 1.1. For example, cycle time is very important for bridge
and high volume production tools, but not for prototype or low volume production tools.
Wear resistance is more important for high volume production tools than for any of the
others. The most important characteristics for low volume production tools are cost and
part quality.
IMPORTANCE CHARACTERISTIC
Cost
Durability
Cycle Time Part Quality
High Volume
Production
TOOL
TYPE
Prototype
Bridge
Low Volume
Production
High Importance
Low Importance
Figure 1.1: Importance characteristics for various tool types.
4
There are a few ideas that have been implemented for reducing tooling costs for
low volume production. For example, a less expensive tooling material can be used that
is easier to machine, such as aluminum instead of steel. Universal mold bases with
interchangeable inserts are another possibility, though this can be problematic if the piece
parts differ widely in design. “Family molds,” in which several components of an
assembly are molded together in the same tool, have been used with limited success due
to constraints imposed by filling and cooling. Any or all of these approaches can be
implemented to minimize tooling costs.
The application of rapid prototyping processes for the purpose of making tools,
known as rapid tooling, has been the object of much interest for prototype and low
volume production. Rapid tooling encompasses many processes based on the rapid
prototyping concepts of additive, layer-by-layer manufacturing. While these processes
are still finding their way into the injection mold market, they hold significant potential
for tools intended to build small quantities of parts.
Depending on the process used, rapid tooled molds are made from various
materials, which typically have much lower strength and thermal conductivity than the
tool steel used in conventionally machined molds. For these reasons, it is generally
believed that rapid tooled molds are inadequate for quality production injection molding.
If, however, the mold is required only to make a small quantity of products, and if
molding conditions are allowed to vary from those used with machined steel molds, rapid
tooling may be an economical alternative. These variations may occur at the expense of
5
cycle time, but cycle time is generally not considered to be as critical for low volume
production.
1.2 Problem Statement
Manufacturers who currently build products in small quantities, such as aerospace
systems, can benefit from injection molding tools that will cost effectively produce low
volumes of production parts. A growing need for such tools is evidenced by the
increasing applications for mass customization. Furthermore, if injection molds for low
volume production become economically feasible, then manufacturers will most likely
discover their overwhelming potential.
Aerospace applications that require small quantities of molded parts, especially
for the military, include composites and electronics packaging. In the composites area,
injection molding and related processes can be used to mold filled thermoplastics for
structural components or for resin transfer, such as for aircraft skins. Future aerospace
applications will also include micro molding for microelecromechanical systems.
Mass customization refers to the mass production of customized products
(Anderson & Pine 1997). The goal of mass customization is to develop, produce, market,
and deliver affordable goods and services with enough variety and customization that
nearly everyone finds exactly what they want (Pine 1993, p. 44). A modular approach to
injection molding using rapid tooled inserts facilitates mass customization at the
fabrication level, allowing smaller quantities to be customized economically.
6
Thermoplastic injection molds, in general, must perform several functions,
including distribution of the melt, formation of the melt into its final shape, cooling of the
melt, and ejection of the part. To meet these requirements, high volume molds are
traditionally machined of steel, are very strong, and have good thermal properties. Rapid
tooling processes, on the other hand, may be very well suited to building injection molds
for small quantity production. These processes have certain advantages for tooling
applications. For example, because rapid tooling processes can generate complex
geometries as easily as simple ones, they can build mold shapes and cooling lines that are
impossible to machine. Also, the capability for local composition control further
enhances the appeal of rapid tooled injection molds.
The material properties of rapid tools, however, vary from conventional molds,
i.e., strength and thermal conductivity can be much lower than for machined steel (Table
1.1). But for small quantity production, the robust material properties exhibited by
conventional machined steel molds may not be necessary. With less rigorous molding
parameters, such as injection pressures and temperatures, rapid tools might be used
successfully for injection molding. The properties of rapid tools may be adequate to meet
many small quantity injection molded part requirements, such as those for the aerospace
and mass customization industries. This research studied aspects of rapid injection mold
tooling in an attempt to find out if this is true.
The issue addressed in this research was whether or not rapid tooled injection
mold inserts are suitable for small quantity injection molding. Many aspects must be
researched in order to confirm any suitability, too many to address in a single project.
7
Therefore, this work has focused on aspects related to ejection force. Rapid tooling
materials must be able to withstand the forces inherent in the injection molding process,
including forces resulting from ejection of the molded part. Ejection force requirements
and the effects of process parameters on ejection force were investigated in this work.
Also included was the determination and analysis of friction coefficients from standard
test results, injection molding experiments, and an ejection force model. With respect to
these areas, this research provides a comparison of rapid tooled inserts to conventional
steel inserts, and further provides an assessment of the benefits and limitations of rapid
tooled inserts for injection molding small quantities of parts.
8
Process
Mold Material
Density Tensile Strength
3
kg/m
MPa
Hardness
Conductivity
o
W/m C
Baseline
Machining
o
P-20 Mold Steel [1]
7870
1080
30-35 HRC
47.6 @ 204 C
H-13 Tool Steel [1]
7800
1550
52 HRC
(R)
Moldmax XL [2]
(copper-nickel-tin)
8900
760
28-32 HRC
25.1 @ 199oC
63-70
3D Printing - Prometal
Bronze/infiltrant
8100
406
60 HRB
7.35
Laser Sintering - 3D Systems
Copper polyamide
3450
33.6
75 ShoreD
1.28 @ 40oC
S Steel w/bronze
7700
510
79 HRB
as machined
Rapid Tooling Materials [2]
Laser Generating - LENS
S Steel 316
Plastic Casting -CIBA
Ceramic-filled Epoxy
Stereolithography - 3D Systems
SL 5170 cured resin
8000 [3]
1220
[1] ed. Rubin 1990
[2] From company literature
[3] From www.matweb.com, various ss 316 properties
Table 1.1: Properties of tooling materials.
9
800
80 HRB [3]
64 (UFS)
91 ShoreD
59
85 ShoreD
0.92 @ 150oC
49 @ 100oC
56 @ 200oC
15 [3]
0.200
1.3 Research Objective
The purpose of this research was to determine the feasibility of using rapid tooled
inserts for injection molding small quantities of products. The objective was to
quantitatively determine the benefits and limitations of laser sintered and
stereolithography tools by: 1) comparing ejection force requirements among materials, 2)
learning which process parameters affect them, and 3) determining the friction
coefficients between the injection mold insert core and the thermoplastic part. The data
generated help to answer the following questions for two thermoplastic molding materials
(one amorphous and one crystalline) and two types of rapid tooled mold inserts:
•
How do ejection forces compare among conventional and rapid tooled
injection mold inserts?
•
How do model-based values for ejection force compare to experimentally
measured values?
•
Do cooling time, packing pressure and packing time affect ejection forces for
conventional and rapid tooled injection mold inserts in a similar manner?
•
What are the coefficients of friction between the thermoplastic materials and
the core materials during ejection?
•
How do standard friction coefficient test results compare to model-based
calculations?
•
Based on these data, what are the potential benefits and limitations for using
rapid tooled inserts for small quantity injection molding?
10
1.4 Research Description
The present research investigated the ejection portion of the thermoplastic
injection molding process. First, ejection forces were measured experimentally for parts
produced from steel, laser sintered steel (infiltrated with bronze), and stereolithography
resin mold inserts. A full factorial statistical experiment was designed to determine the
effects of three process parameters on the ejection force. Second, the experimental
ejection force values were compared to calculated values from an ejection force model.
Standard friction testing was conducted to determine static friction coefficients to use in
the model. Model-based values for the static friction coefficients were also determined
and compared with the standard test results. These results, along with observations of
tool performance, provide some indication of how successfully the rapid tooled inserts
might apply to injection molding.
The chosen part for the experiments is a vented, closed-end cylinder, similar to
the plastic canisters used to store photographic film. The thermoplastic materials were
chosen according to their moldability for the given application and their range of
applications for manufacturing and consumer products. High density polyethylene
(HDPE), a semicrystalline thermoplastic, and high impact polystyrene (HIPS), an
amorphous thermoplastic, are widely used consumer resins, and are known to be well
suited to injection molding.
The experimental core and cavity pairs were built as inserts that were fitted to a
standard mold base. The rapid tooling processes and materials have been chosen
11
according to their potential application for injection molding and to their availability for
experimentation. These processes are laser sintering and stereolithography. While the
experiments with the laser sintered insert were similar to those with the baseline steel
insert, the stereolithography insert posed more of a challenge due to the softness of the
material and its insulating qualities.
The number of experimental runs was determined by the designed experiment,
which varied three input parameters: packing pressure, packing time, and cooling time.
These input parameters are key in defining an optimal injection molding process and
producing a quality part. For each experimental part, ejection force, temperature at
ejection, and part diameter data were collected. The ejection force data were compared
with values calculated using a model for estimating ejection force developed by Menges.
Apparent coefficients of friction for all material pairs were calculated using the Menges
model and data from the experiments. These values were compared with results from
standard friction tests. Statistical analysis was performed to determine the effects of the
three input parameters on ejection force.
1.5 Organization
The remainder of this document is organized as follows. Chapter 2 is the result of
a literature search of relevant previous work. It first presents topics related to ejection
forces in injection molding, such as ejection force models, shrinkage, friction and
12
adhesion. A section on rapid tooling is also presented that includes a background of rapid
prototyping processes and details of the stereolithography and laser sintering processes.
Chapter 3 presents the theoretical basis for this work. This includes the materials
aspects of the two thermoplastics used in the experiments, HDPE and HIPS, and further
definition of the coefficient of friction. The last section in Chapter 3 derives the equation
for ejection force.
Chapter 4 provides details on how the experimental work was accomplished. It
describes the standard friction test, the injection molding experimental design, the part
and tool designs and data acquisition. Chapter 5 presents all test and experimental results
and statistical analysis, and Chapter 6 presents conclusions, implications and future work.
References are listed following Chapter 6. There are two appendixes: Appendix A
includes data tables, and Appendix B includes part and tool drawings.
13
CHAPTER 2
LITERATURE SEARCH
This chapter presents comprehensive results of a literature search on the topics of
ejection force and rapid tooling. Extensive work has been published on topics related to
ejection forces, including shrinkage, friction, adhesion, and modeling. Topics cited in
rapid tooling include various rapid prototyping processes and works specifically
pertaining to stereolithography and laser sintering.
2.1 Ejection Force
2.1.1. Ejection Force Models
Several researchers have developed force equations for the ejection of parts from
injection mold cores based on mechanical or thermo-mechanical models. Most of these
equations derive from the friction-based concept FR = f × p A × A (see derivation in
Chapter 3), where FR is the ejection (or release) force, f is the coefficient of friction
between the mold and the part, pA is the contact pressure of the part against the mold
14
core, and A is the area of contact. While area is a straightforward measure, friction
coefficient and contact pressure have various interpretations or methods of estimation. A
number of models and variations are summarized below.
The version of the ejection force equation developed by Menges et al for a vented
cylinder defines contact pressure as
p A = E (T ) × ∆d r × s m
2.1
and therefore ejection force is:
FR = f × E (T ) × ∆d r × s m × 2πL
2.2
where E(T) is the elastic modulus of the thermoplastic part material at ejection
temperature, ∆dr is the relative change in diameter of the part immediately after ejection,
sm is the thickness of the part, and L is the length of the part in contact with the mold core
(Menges, Michaeli, & Mohren 2001). The rationale for this formulation is that shrinkage
of the part is constrained by the core, thus causing stresses to build up in the cross
sections of the part and resulting in forces normal to the surfaces restrained from
shrinking. When the part is ejected from the mold, the stored energy-elastic forces can
recover spontaneously. The relative change in circumference, measured immediately
after ejection, is used as a measure of tensile strain in the cross section of the part while it
is still on the core. The strain multiplied by the elastic modulus, the surface area in
15
contact, and an assumed friction coefficient then gives an estimate of the force required
to remove the part from the core.
Malloy and Majeski (1989) referenced the ejection force equation as used by
Menges et al and a more detailed version by Glanvill, as shown below. Their paper
examined ejection variables with respect to designing ejector pins.
Burke and Malloy (1991) further discuss aspects of contact pressure and
coefficient of friction. They showed that ejection force is affected by cooling time,
surface finish, direction of polish, and draft angle. Their version of the ejection force
equation for a box-shaped part (not vented) is as follows:
FR = f × E (T ) × α × (TS − TE ) ×
8s m L
+ (W1W2 PA )
1 −ν
2.3
where α is the coefficient of thermal expansion (contraction), TS is the temperature at the
onset of shrinkage (determined using a secondary empirical calculation), TE is the
temperature at ejection, ν is Poisson’s ratio, W1 and W2 are the widths of two sides of a
rectangular core, and PA is atmospheric pressure. The authors applied this equation to
determine the apparent coefficient of friction at various surface finishes.
Michalski (2000) used a version of the equation for closed cylindrical sleeves to
measure ejection force for film canisters. This version of the ejection force equation took
into account vacuum forces and an adjusted value for f due to the taper of the part
(Menges, Michaeli, & Mohren 2001).
16
Glanvill (1971) is another oft cited reference for ejection force. His equation
defines contact pressure as
pA =
α (Tm − Te ) × E
1 ν
−
2t 4t
2.4
where Tm is the softening point of the thermoplastic and t is thickness. Thus,
FR =
α (Tm − Te ) × E × πL × f
1 ν
−
2t 4t
2.5
Hopkinson and Dickens (1999, 2000a, 2000b) used Glanvill’s equation to predict
ejection force for parts molded with stereolithography tools. The authors have done
extensive work with stereolithography tools as described later in this chapter.
A model by Pham and Colton (2002) was developed from Glanvill for
stereolithography molds, taking into account friction and shrinkage, as well as the stairstep (roughness) aspect of stereolithography molds with draft angles θ. They defined two
components of force, one due to friction and another due to the stair-step surface as
follows:
FR = F fric.therm + Fdef . stair
2.6
17
The ejection force is
FR = A × ( f eq cosθ − sin θ ) × Ptherm
2.7
where contact pressure is
Ptherm =
α p ∆T p rp − α m ∆Tm rm
 1
rm 
 E p
 rp2 + rm2
 1 −ν m 

+
+
ν

p
 r2 − r2

E m 
p
m


2.8
where r is the hydraulic radius, the p subscript refers to the part, and the m subscript
refers to the mold. This model derives from the ejection force equation for a general
mold with a core feature and uses an approximation for thick-walled cylinders. The
model was applied, along with finite element analysis and experimentation. Results
showed the Pham and Colton model to be more accurate than Glanvill in this case.
Colton, Crawford, Pham, and Rodet (2001) showed that the ejection force model
for stereolithography molds gives reasonable results when compared to experimental
results. In this work, the build orientation of the stereolithography tool had no effect on
mechanical properties. Mechanical properties of the mold were shown to degrade with
higher temperatures. Brittle fracture of the molds occurred below the glass transition
temperature, while yielding occurred above the glass transition temperature.
Palmer and Colton (2000) used this model to predict ejection failures of
stereolithography mold features based on height ratio, aspect ratio, and draft angle.
18
Height ratio was the most critical factor in determining feature life, while aspect ratio had
no conclusive effect. As expected, larger draft angles increased feature life. Fatiguebased chipping failures also occurred.
Cedorge and Colton (2000) studied the stair step effect of stereolithography tools.
Surface roughness resulting from the stereolithography build process depended on layer
thickness and draft angle. The authors showed a trade-off between these two parameters
in terms of ejection force. For tools built with thin layers, ejection force decreased with
draft angle, while for thick layers, ejection force increased with draft angle.
Colton and LeBaut (2000) showed that ejection force decreases with number of
shots in a stereolithography mold. This was because the mold gradually heated up, and
shrinkage was less because the mold and part temperatures were closer together. The
authors also found that the stereolithography material continued to cure and become
harder.
Another version of the ejection force equation was presented by Shen et al (1999)
for hollow, thin-walled cones, taking into account draft angle θ and vacuum forces:
FR =
2πEεs m L cosθ ( f − tan θ )
×
+ 10 B
1 −ν
1 + f sin θ cos θ
2.9
where ε is elastic strain in the thermoplastic and B is the projected area of the core
surface in the core axis direction. The first term in this equation refers to contact
pressure, which was determined from a force and stress analysis for a hollow, thin-walled
cone. The second term refers to the friction force, and the third term refers to the vacuum
19
force. Experiments by Shen, et al showed agreement with model results, though molding
parameters were not discussed.
Pontes, et al derived a thermo-mechanical model for amorphous materials based
on average internal stress (Pontes et al 2001, Pontes, Brito and Pouzada 2002, Pontes, et
al 2002; see also Jansen and Titomanlio 1996). The model assumed that stresses in each
layer of the part start to develop when the layer solidifies, and relaxation in the solid
polymer is negligible because of the high cooling rate. For a cylindrical part
σ θθ =
E (Te )
E (Te ) t
× (− βPs + α (Ts − Te )) −
×
×δ r
1 −ν
1 − ν 2 Dm
te
t r*
2.10
where σ θθ is the average circumferential stress before ejection, β is compressibility, PS
is the pressure as each layer of polymer solidifies, Dm is the center thickness coordinate,
δr is thickness shrinkage, te is the time of ejection and t r* is the time of solidification. The
first term of this equation represents pressure induced effects, the second term represents
thermal contraction, and the third term represents thickness shrinkage, which reduces
average internal stresses. The authors found that ejection force decreased with increasing
surface temp at ejection (for polystyrene), increased slightly, then decreased with
increasing holding pressure (for polystyrene and polypropylene), and decreased with
increasing holding pressure (for polycarbonate). Experimental results agreed with the
model.
20
Kabanemi, et al (1998) derived a numerical model for prediction of residual
stresses, shrinkage and warpage for thin, complex injection molded products. Wang,
Kabanemi, and Salloum (1997, 2000) presented the numerical approach to predict
ejection force from mold-part constraining forces and friction forces. It included finite
element thermoviscoelastic solidification analysis to account for stress and volume
relaxation of polymers under cavity-constrained conditions, and predicted distribution of
ejection force among ejector pins. The model worked well for a rigid polymer
(polycarbonate), but HDPE had significant post-molding shrinkage and warpage that was
not taken into account.
Several examples of research using models for injection force have been
described in this section. Many researchers have used the Menges or Glanvill models,
while others have derived their own models. Much of this work has shown the effects of
various parameters on ejection force and has illustrated the many different variables that
need to be taken into account. The present work follows up on these ideas by
determining the effects of three parameters on ejection forces for three mold insert
materials and two thermoplastic materials, and by applying an existing ejection force
models (from Menges) and comparing the results to experimental values. The present
work is unique in that it includes three different injection mold inserts in the same
experiment, and two of these are made by rapid prototyping processes. It is also unique
in that it includes values for modulus at temperature, standard measurements of
coefficients of friction at elevated temperatures, and near real time measurements of part
diameters (to determine shrinkage and part thickness).
21
2.1.2 Shrinkage
An important aspect of the above ejection force models is shrinkage of the
thermoplastic part. Shrinkage influences the contact pressure of the part against the core
and can affect strain and friction. The extent of shrinkage that occurs depends on
material properties and process conditions. The following works, most by researchers
previously mentioned, address shrinkage in the context of thermoplastic injection
molding and ejection forces.
Malloy and Majeski (1989) explained aspects of shrinkage that relate to the
injection molding process. They stated that shrinkage values for thermoplastics are often
given in ranges because they vary both parallel and perpendicular to flow and with
process conditions. Standard shrinkage values, however, have limited value in
determining ejection forces since ejection is normally at elevated temperatures. Deep
gates, long holding times, and high holding pressures in the injection molding process
can compensate for shrinkage of the part. In calculating the ejection force, accurate
values of the coefficient of thermal expansion may not be available since it is a function
of temperature and pressure in the process. Therefore, shrinkage (strain) values can be
used instead.
Burke and Malloy (1991) stated that shrinkage results from thermal contraction
and directional distortion. Thermal contraction is due to atomic vibration in which atoms
move closer together at lower energy levels, and directional distortion results from
orientation of polymer molecules during flow, and their subsequent relaxation back to a
22
coiled state after flow ends. Shrinkage is greatly influenced by ejection temperature, is
material dependent, and varies for amorphous and semicrystalline polymers.
Semicrystalline materials exhibit greater shrinkage due to phase transformation of the
crystalline portion: random amorphous coils and high free volume in the melt reduce to
orderly packed chains in the crystal lattice. Amorphous polymers, on the other hand,
contract much more gradually.
Michaeli et al (1999) modeled the development of material properties and
crystallization due to processing. They found that the temperature at which the
crystallization peak occurs decreases, and the crystallization interval widens, with
increasing cooling rate. That is, crystallization starts earlier at lower cooling rates.
Menges, et al (2001) stated that, for both amorphous and crystalline
thermoplastics, holding pressure exerts the greatest effect on shrinkage (degressive
effect) in the injection molding process. The temperature of the material is the second
major factor influencing shrinkage. Higher temperature results in higher thermal
contraction potential, but also lowers viscosity for better pressure transfer. With a longer
holding time, the effect of improved cavity pressure predominates for crystalline
materials. Menges et al provided shrinkage values for some thermoplastics, but stated
that the best data are found through experience.
Pantani and Titomanlio (1999) found that higher pressure histories inside the
injection mold cavity – obtained by increasing either holding time or holding pressure –
result in a lower final shrinkage and in a delayed start of shrinkage inside the mold for a
polystyrene plate.
23
In mold experiments with polycarbonate, Pontes et al (2001) found that increasing
holding pressure reduces contact pressure by decreasing diametrical shrinkage. Holding
time, however, had no effect on the shrinkage due to fast solidification of this material.
In their ejection force modeling work, the authors described average circumferential
stress to include volumetric shrinkage due to thermal contraction (and crystallization) and
thickness shrinkage, which reduces stress. Ejection force, then, depended on elastic
modulus, friction coefficient, part thickness, and variation of the volumetric shrinkage.
In their model, initially, ejection force increased (or plateaued) with increasing holding
pressure because of thickness shrinkage, while at higher holding pressures a reduction in
volumetric shrinkage reduced ejection force.
As indicated by the research described in this section, shrinkage varies with
parallel and perpendicular flow, injection and ejection temperature, holding pressure and
time, material structure and properties, and pressure histories. In the present work, while
shrinkage is not analyzed directly, it is measure and used in the Menges model to
calculate ejection force. Shrinkage influences the contact pressure of the part on the core
(see Chapter 3). Furthermore the effects of cooling time, packing pressure, and packing
time on ejection force are determined in part by the shrinkage characteristics of the
thermoplastic material.
24
2.1.3 Friction and Adhesion
Friction is another important aspect in determining ejection forces. Friction
between the thermoplastic part and the injection mold core not only depends on the
mechanical relationship between the two surfaces, but also on an adhesive component
inherent in the properties of the two materials at processing conditions. The following
works address friction and adhesion, some in general terms, others specifically as they
apply to polymers and injection molding.
Contact between two solids occurs only at asperities (ed. Eley 1961, Ch. V).
Extremely high pressures are produced at these contact points and, in metals, plastic flow
occurs. Under plastic conditions, the area of real contact is directly proportional to the
load and is independent of the apparent area of contact. During sliding at slow speeds,
with no temp increase, fragments of one metal can strongly adhere to the other (cold
welding). The frictional force is then the force required to shear the junctions formed in
this way.
With softer metals junctions are more ductile and easily deformed, and
appreciable adhesion may occur. Relatively smaller adhesions occur in plastics due to
higher elastic recovery. Adhesion may thus occur by reducing elastic stress or by
increasing ductility.
A number of concepts relating polymer friction and adhesion to thermoplastic
injection molding (with steel molds) were presented by Burke and Malloy (1991) and are
summarized below. More on friction and adhesion theory is presented in Chapter 3.
25
•
Plastics have relatively low modulus values, which lead to frictional values that
are not always directly proportional to load. This is attributable to adhesion and
deformation.
•
Theoretical calculations show that van der Waals forces, which attract molecules
with permanent dipoles, and London dispersion forces, which cause dipoles
created by motion of electrons in the molecule, are great enough to produce bonds
exceeding the cohesive strength of most adhesives.
•
In the surface energy theory, a liquid may wet and spread over a solid surface if
the critical surface tension of the solid is greater than that of the liquid. Heat
decreases viscosity and improves wettability; heat and pressure promote wetting
and spreading. Molten polymers on steel under injection molding conditions are a
good environment for wetting and spreading.
•
Wetting and spreading does not necessarily imply adhesion. Apparently both
physical adsorption and surface energy criteria must be met for adhesion to occur.
An increase in adhesion will increase the apparent coefficient of friction, which
depends on the specific polymer-steel combination.
•
Surface roughness causes mechanical coupling and increases surface area over
which van der Waals forces can act. Imperfect surfaces lead to inherent voids or
trapped gas bubbles and imperfect molecular fit, limiting the bond strength.
•
The coefficient of static friction increases with increasing surface roughness,
depending on viscosity and pressure applied. A highly viscous material under
low pressure may not wet the steel. The direction of polishing affects part
26
ejection. The coefficient of friction decreases with increasing cooling time
because shrinkage decreases the mechanical anchorage of the polymer, i.e., it no
longer completely penetrates irregularities in the mold.
Looking at polymer adhesion from the standpoint of wear, Briscoe (1981)
summarized several fundamental aspects, including cohesive wear and interfacial wear.
Cohesive wear mechanisms occur adjacent to the interface, e.g., abrasion and fatigue
wear induced by tractive stresses. Interfacial wear processes dissipate frictional work in
much thinner regions and at greater energy densities, e.g., transfer wear and chemical or
corrosive wear. In interfacial wear, frictional work originates from adhesive forces
emanating from the contacting solids. These forces generate localized plastic surface
deformation and transfer of relatively undegraded polymers to the counterface in certain
systems.
Also discussed in this paper is natural adhesive or transfer wear, specifically,
initial adhesion. Briscoe stated that initial junction strength is a function of the
interaction of surface forces and mechanical properties of the contact. For polymers, the
surface forces consist of van der Waals, coulombic and possibly hydrogen bonding
forces. The higher the surface free energy of the polymer, the greater the adhesive force.
Polymers above the glass transition temperature will adhere more strongly because they
conform to surface imperfections and have a relatively low level of stored elastic strain.
Very clean metal surfaces may promote chemical bonding. Essentially brittle and highly
27
elastic crosslinked polymers tend to fail at the interface. This includes polymers below
the glass transition temperature and crosslinked systems.
Czichos (1983, 1985) investigated contact deformation, static friction, and
tribological behavior of polymers. In his work contact deformation was measured for
four crystalline thermoplastics in loading and unloading conditions. A model was
proposed that takes into account elastic, viscoelastic, and viscoplastic components.
Polymer to polymer coefficients of friction were measured, using a pin-on-disc
configuration, and plotted against sliding distance. Experimental frictional work was
plotted against the work of adhesion using the Dupre equation (see Chapter 3), showing
that a reasonable correlation exists. Also, coefficient of friction and wear rate were
plotted against surface roughness for four polymers against steel. The author found that
adhesion was the primary influence for very low surface roughness, while abrasion was
the primary influence for higher roughness.
Benabdallah and Fisa (1989) measured the static friction coefficient between a
steel surface and three thermoplastics with surface roughnesses varying between 0.4 µm
RMS to 40.5 µm RMS. Parameters measured were normal load, relative displacement
and tangential force. In these tests, static coefficient of friction decreased with increasing
normal loads ranging up to 160 N. This is explained by an increasing influence of the
adhesion component of friction. Also, the friction coefficients decreased with increasing
surface roughness since, with smoother surfaces, there is more adhesion. The authors
present a model for friction coefficient µs based on this work:
28
µs = αFNn
2.11
where FN is normal load, α is a proportionality constant that depends on polymer surface
roughness and n is experimentally determined for each polymer.
Benabdallah (1993) investigated static shear strength during contact between a
bulk plastic and a metallic plate, both with smooth surfaces. The experimental equipment
included one apparatus to measure static friction force and another to measure the real
area of contact. The adhesion component of friction was approximated by the measured
static friction force. The author determined bulk shear strength of plastics experimentally
following ASTM D732-78 and found surface energies from the Young equation (see
Chapter 3). In this work the friction force was assumed to consist only of the adhesion
component due to the smoothness of the contacting surfaces. That is, the deformation (or
ploughing) component was not considered. Friction force equaled the maximum
tangential load, which corresponded to the minimum force required to initiate motion.
The paper includes plots of the adhesion component of friction against the calculated
work of adhesion according to the Dupre equation, Wa = 2φ (γ 1γ 2 ) 2 , where the
1
interaction parameter, φ, equals 1, and the static shear strength against the real contact
pressure (the ratio of applied load and real area of contact).
Benabdallah then combined the Young equation with the geometric mean
equation to obtain:
29
γ L (1 + cosθ )
( )
2γ
d
L
1
2
( )
= γ
p
S
1
γ p
2 L
γ d
 L



1
2
( )
+ γ Sd
1
2
2.12
This equation is in the form y = mx+b. Given known surface energies of six liquids,
contact angles were measured and x and y plotted. Then the square of the intercept
determined the dispersion component, the square of the slope determined the polar
component, and the addition of the two gave the total surface energy of the solid γs.
The author concluded that the adhesion component of friction increases with the
real area of contact and is large when the surface energy of the plastic material is high. It
was also found that a correlation may exist between the adhesion component of friction
and the work of adhesion when evaluated as a function of the real area of contact.
Menges and Bangert (1981) measured static coefficients of friction for
determining opening and ejection forces in injection molding. This work looks at the
effects of various parameters, including surface (contact) pressure, cooling time, mold
temperature, holding pressure, and surface roughness. Various thermoplastics were
studied, but results were reported only for polypropylene. In all cases, the friction
coefficient decreased with surface roughness in the range 1 to 35 microns. In general the
friction coefficient also decreased with increasing cooling time. The effects of other
parameters were varied. The friction coefficient results varied from standard
measurements for polypropylene.
Balsamo, Hayward and Malloy (1993) conducted studies on ejection forces and
coefficients of friction. The authors found, first, that external lubricants have a large
30
effect on ejection force. They also measured static friction coefficients for polystyrene,
polypropylene, a polycarbonate/polyethylene alloy, and filled polycarbonate parts on
nickel, steel, and polytetrafluoroethylene (PTFE)/nickel plated mold cores. While noting
that the friction test does not exactly duplicate the injection molding environment, they
found that friction coefficients are generally lowest on PTFE/nickel surfaces and highest
on steel surfaces. Some coefficients changed significantly with temperature.
In Dearnley’s work (1999) to study low friction surfaces for injection molds, steel
rings were coated with TiN (polished), CrN (polished and spark eroded), and MoS2
(polished and spark eroded) and used as a core around which an acetal ring was molded.
Coating thickness, surface hardness, and surface roughness were measured, and friction
force was determined experimentally. Spark eroded surfaces were found to have higher
roughness values and higher friction forces compared to polished surfaces. The author
attributed this to mechanical interlocking. Polished CrN had the lowest friction forces
even though polished TiN and MoS2 had lower roughness values. This was attributed to
possible differences in chemical behavior at the interface, e.g., lower surface energy (or
wettability) of CrN coatings.
Pontes et al (1997) studied the effects of processing conditions on ejection forces
for tubular moldings. Parameters included surface roughness, injection temperature and
holding pressure. Two thermoplastics were used, one amorphous and one crystalline.
For polyphenylene ether (PPE), ejection force increased with injection temperature and
decreased with holding pressure, as would be expected. For polypropylene, ejection
force decreased initially with surface roughness (less than 0.75 microns), then increased.
31
For polypropylene, ejection force decreased as injection temperature increased, indicating
that the core surface temperature was different than ejection temperature, and that
deformation ability increased with higher temperature. For polypropylene, ejection force
decreased with increasing holding pressure, as was expected. The diametrical shrinkage
was determined from shrinkage at room temperature and the coefficient of thermal
expansion. Calculated values for the equivalent coefficient of friction of PPE
(amorphous) were near the lower range of published values. The authors concluded that
adhesion appears to be an important factor for semicrystalline materials molded on
surfaces with low surface roughness.
Sasaki et al (2000) molded cylindrical parts with polypropylene, polymethyl
methacrylate (PMMA), and polyethylene terephthalate (PET) at a range of surface
roughnesses from 0.016 to 0.689 microns Ra. In all cases, ejection force increased
significantly when surface roughness approached zero. Optimum surface roughness (in
terms of ejection force) for polypropylene and PET was approx. 0.2 microns and for
PMMA was 0.009 microns. For lower values of roughness, the meniscus force or van der
Waals force was thought to be the greatest factor, whereas for higher values of roughness,
the “engraving or scratching” of the surface came into play. Ejection force was also
measured on polypropylene and PET parts from cores with various coatings. Here,
tungsten carbide/carbon coating was found to be most effective for reducing ejection
force. TiN (HCD), TiN (Arc), DLC, and CrN coatings also showed ejection force
reduction effects.
32
Ferreira et al (2001) friction tested polycarbonate and polypropylene using a
special prototype apparatus. The testing procedure included heating the specimens to
processing temperatures, applying a normal load (so that the specimen replicated the
mold surface), cooling to ejection temperature, then pulling the specimen. At room
temperature, the coefficient of friction of polycarbonate at 0.32 was similar to published
values at 0.31. At high temperature, the coefficient was much higher at 0.47. For
polypropylene at high temperature, the coefficient of friction was 0.19, much lower than
published values, 0.36. A similar approach to imprinting the mold surface onto the
specimens was used in the standard friction tests in the present work (see Chapter 4).
In related work, design of experiments was used to determine the effect of polish
direction, surface roughness, and temperature on the coefficient of friction (Ferreira et al
2002). Results showed that testing temperature and surface roughness had a significant
effect on the coefficient of friction for polycarbonate. For polypropylene, none of the
parameters had a significant effect on the coefficient, except possibly the interaction of
polish direction and roughness. Friction values for both polymers were higher than
published values.
Muschalle (2001) measured the coefficient of friction for polycarbonate
(amorphous) and polypropylene (semi-crystalline) materials against steel with two
different surface roughnesses, machining directions, and temperatures. Results showed
that the friction coefficient for polycarbonate was higher at higher temperature, while that
for polypropylene was lower at higher temp. Also, the coefficient for polypropylene was
33
higher when temperature and pressure caused surface reproduction of the metal on the
plastic.
The works described in this section indicate the many aspects of friction, which
includes both a mechanical and an adhesive component. No unifying theory seems to
exist for friction, but, rather, it can be explained by one or another or a combination of
concepts. This can be seen in the friction testing in the present work, where coefficients
of static friction are influenced by adhesion and/or mechanical components of friction to
varying degrees (reference Chapters 5 and 6).
2.2 Rapid Tooling
2.2.1 Background
One of the most promising techniques for low volume, net shape manufacturing
tools is rapid tooling, i.e., the application of rapid prototyping processes for the purpose
of making tools. Rapid tooling processes are additive and produce a tool or pattern from
a CAD model. Direct rapid tooling processes generate the tool itself from the CAD file,
while indirect rapid tooling processes require intermediate steps and usually generate a
pattern from which a tool is made. A wide range of materials can be used in rapid tooling
processes, from waxes and resins to ceramics and metals. Those processes that use metal
materials and build tools directly tend to be more suitable for production tools. These
34
include laser sintering, 3D printing, and laser generating (Karapatis, van Griethuysen &
Glardon 1998).
To date most rapid tooling technology for production parts has been aimed at
meeting the same process requirements as conventional tooling. Those direct and
indirect rapid tooling processes that use metal materials have been most successful with
this approach. Tools from non-metal processes, however, can possibly be used under
non-conventional molding conditions. These include, for example, stereolithography and
cast epoxies. Several rapid tooling processes are described in the following paragraphs.
Laser generating processes deposit highly dense metal materials and come very
close to meeting the material properties of conventional molds. The laser engineered net
shaping (LENSTM) process, for example, focuses a high power Nd:YAG laser and creates
a molten puddle on a substrate, into which metal powder is injected (Keicher, Gorman &
Taute 2001). A schematic of the process is shown in Figure 2.1. An injection mold
insert with conformal cooling channels was successfully built for a high volume
automotive part using the LENSTM process (Optomec 2001).
35
Figure 2.1: A schematic of the laser engineered net shaping (LENS™) process
(Castle Island 2003).
Three other rapid tooling processes that use metal materials are laser sintering,
direct metal laser sintering, and 3D printing. In the laser sintering process, a laser is
scanned over powdered material with a binder coating, and the part is built layer by layer.
Laser sintering is described in detail later in this section. Direct metal laser sintering
(DMLS) is a similar process in which the metal itself is sintered without any polymer
binder. 3D printing processes spray a binder material in an ink-jet-printing fashion onto
successive layers of metal powder. All of these processes include steps for debinding,
sintering and infiltration.
36
Stereolithography, like laser sintering, employs a scanning laser, but uses liquid
resin build materials. The stereolithography process is described in detail later in this
section.
Indirect processes generate a pattern using rapid prototyping techniques, then
build a tool from that pattern. Although indirect processes have more steps, they benefit
from a wider range of material choices. At the Pennsylvania State University, a powder
metallurgy process was used to make a mold insert (Weaver et al 2000). From a three
dimensional model, a pattern was generated using a three dimensional plotting process, a
negative of the tool was cast in silicone rubber, and a slurry of steel and ceramic was cast
into the negative to make the final tool insert. Such a tool has excellent mechanical
properties and is capable of high volume production.
Another option for indirect tooling is cast epoxy, in which a blend of resin and
aluminum filler is cast over a rapid prototyped pattern. The resulting tool insert is
machineable, capable of withstanding typical molding pressures and temperatures, and
can produce low volumes of prototype or production parts.
The two rapid tooled injection mold inserts for this research were built using laser
sintering and stereolithography processes. These represent two very different processes
among the spectrum of rapid tooling techniques. While the laser sintering material is
more like conventional tool steel, the stereolithography material is unlike what you would
expect in a production tool. The two processes were chosen first for their availability,
and second based on their economic potential for producing small quantities of parts. In
the remaining paragraphs of this section, these processes will be described in detail.
37
Laser sintering is an additive layer process in which a laser melts powdered
material by cross sections to build a part. The process is versatile in that it can use any of
several powdered materials including polymers, ceramics, and metals. For the Selective
Laser Sintering (SLS®) process, developed by the University of Texas at Austin, early
processing materials included wax, polycarbonate, unreinforced nylon, and glass
reinforced nylon (McAlea et al 1995). Today metal materials can be sintered and
infiltrated for higher densities. The laser scans across each layer of metal powder coated
with a polymer binder, and fuses the binder to create the tool (Figure 2.2). Later the
“green” part is sintered and infiltrated with copper or bronze (Beaman et al 1997, Kai &
Fai 1997).
Figure 2.2: A schematic of the selective laser sintering process (Castle Island 2003).
38
The selective laser sintering process was used to make one of the injection mold
inserts for this research. This process involves a polymer-coated 420 stainless steelbased powder, known as LaserForm ST-100, and a 3D Systems Vanguard machine. The
sintered ST-100 material was subsequently infiltrated with bronze. Specifications of the
Vanguard are shown in Table 2.1 and material properties of ST-100 are shown in Table
2.2.
39
Model Number
Laser
Wavelength
Power
Beam Diameter
Max. Scan Speed
Min. Layer Thickness
Build Chamber
LC-100
DEOS CO2 Laser
10.6 microns
100 W max at part bed
450 microns
10,000 mm/s (394 in/s)
0.10 mm (0.004 in)
381w x 330d x 457h mm
(15w x 13d x 18h in)
Table 2.1: Vanguard System Specifications from 3D Systems.
3
Density
Thermal Conductivity
7.7 g/cm
o
o
49 W/m K @100 C
CTE
Tensile Yield Str. (0.2%)
Tensile Strength
Young's Modulus
Elongation
Compression Yld Str (0.2%)
Hardness, HRB
56 W/m K @200 C
o
12.4 ppm/ C
305 MPa
510 MPa
137 GPa
10%
317 MPa
87 As infiltrated
79 As machined
o
ASTM D792
ASTM E457
o
ASTM E831
ASTM E8
ASTM E8
ASTM E8
ASTM E8
ASTM E9
ASTM E18
Table 2.2: LaserForm ST-100 (Sintered and Infiltrated) Material Properties from
3D Systems.
40
When the 3-dimensional part is initially built on the Vanguard System, the laser
heats the metallic particles above the glass transition temperature of the polymer coating.
The polymer softens and deforms, then fuses with other particles at each contact surface.
The temperature is such that melting of the metal does not occur, only viscous flow of the
polymer coating. The metal powder is then bound together by the polymer to form the
“green” part. After the build is complete, the green part is removed from the machine
and excess powder is brushed away. A furnace cycle follows in a reducing atmosphere to
burn off the polymer, sinter the steel powder, and infiltrate the part with bronze.
Infiltration eliminates any voids within the steel, resulting in a fully dense part (Bourell et
al 1994, McAlea et al 1995).
Stereolithography, a non-metal process, uses a laser to scan a vat of liquid resin
and build a part layer by layer. Stereolithography resin, in direct comparison to
conventional mold steel, has vastly different mechanical and thermal properties. With
enhancement, such as a metal backing, a metal coating, or water cooling channels, a
stereolithography resin mold still underperforms an aluminum one under traditional
molding conditions (Li, Gargiulo & Keefe 2000). Nevertheless, there are several
examples of research in stereolithography tooling, as described in the next section.
Stereolithography was one of the first rapid prototyping processes to emerge, and
the 3D Systems stereolithography apparatus (SLA) was a pioneer rapid prototyping
system in the late 1980s (Kai & Fai 1997). The SLA system consists of a control
computer, a control panel, a laser, an optical system, and a process chamber. The SLA
250, appropriate for many applications, has been widely used across the globe and, in
41
fact, was used to make mold inserts for this research. Specifications for the SLA 250
machine are shown in Table 2.3. The SLA uses a photo-curable liquid resin as a build
material. Many resins are available depending on the type of laser in the machine and the
requirements of the part to be built. For this research, one of the rapid tooled injection
mold inserts was built using the stereolithography process and SL 5170 resin from
Vantico. Properties of this resin are shown in Table 2.4.
The SLA set-up includes a vat of the photo-curable liquid resin, inside which an
elevator table is set just below the resin surface (Figure 2.3). A solid model CAD file in
.STL format is loaded into the machine. The model is sliced by the control unit into cross
sections, which are solidified by the SLA laser one at a time. After each layer is
solidified, the elevator drops just enough to cover the solid layer with a new coat of liquid
resin. The part is built in this manner from the bottom up. When completed, the elevator
raises the part out of the vat, and the excess liquid resin is removed.
Figure 2.3: A schematic of the stereolithography process (Castle Island 2003).
42
Laser
Wavelength
Power
Beam Diameter
Max. Drawing Speed
HeCd
325 nm
24 mW
0.20-0.28 mm
762 mm/s
Min. Layer Thickness
Elevator Resolution
Max. Part Weight
Vat Capacity
Max. Build Envelope
0.1 mm
0.0025 mm
9.1 kg
32.2 L
250 x 250 x 250 mm
Table 2.3: SLA 250 System Specifications from 3D Systems.
Tensile Strength
Tensile Modulus
Elongation at Break
Glass Transition Temp
CTE
Thermal Conductivity
Hardness, Shore D
Density
59-60 MPa
3737-4158 MPa
8%
o
65-90 C
o
90 ppm/ C
ASTM D638
ASTM D638
ASTM D638
DMA
TMA (T<T g)
o
0.200 W/m K
85
3
1.22 g/cm
DIN 53505
Table 2.4: Vantico SL5170 Typical Properties (90-minute UV post cure).
43
The mechanisms that are the basis for the stereolithography process are free
radical and cationic photopolymerization. Polymerization is the process by which
monomers are linked into larger, chain-like molecules called polymers. Further linking
leads to the crosslinking of these chains. In free radical polymerization, heat or light
energy decomposes an initiator to generate free radicals that catalyze the polymerization
process. In cationic photopolymerization, cationic photoinitiators cause reactions that
open molecular ring structures to catalyze the polymerization process. Free radical
photopolymerization is associated with acrylate resins, and cationic photopolymerization
is associated with epoxy resins.
The resins used in the 3D Systems’ SLA machines are UV-curable photopolymers
made up of photoinitiators and reactive liquid monomers. In the SLA polymerization
process, sufficient crosslinking is required to prevent the polymer molecules from
dissolving back into monomers. Furthermore, since the cured resin must withstand forces
during recoating, the polymer molecules must be sufficiently strong. Increasing the laser
power results in a higher polymerization rate and thus a faster build rate, but brittleness
also results, due to lower molecular weight. Cure depth must be deep enough to prevent
delamination, but not so deep as to cause distortion and, therefore, inaccurate parts (Kai
& Fai 1997, Beaman et al 1997).
2.2.2 Stereolithography and Laser Sintering for Injection Molding Tools
Rapid tooling processes lend themselves well to injection molds because of their
ability to generate complex shapes as easily as simple ones. Complex shapes that are
44
difficult or impossible to machine, detailed internal structures, and thin walls can be
readily generated. This allows the integration of conformal cooling channels within the
mold, which lower residual thermal stresses and can reduce cycle times (Sachs et al
2000). Some rapid prototyping processes have the ability to vary material composition
during fabrication. This local composition control benefits rapid tooling because it
allows tailoring of various material properties, such as conductivity, corrosion resistance,
and hardness (Cho, Sachs, & Patrikalakis 2001). It is theoretically possible, for example,
to build an injection mold with a core of highly conductive material, such as copper, and
surround it with a wear resistant material, such as stainless steel. Rapid tooling processes
lend themselves well to low volume production because they reduce the requirements for
labor intensive machining, minimize material waste, and, in some cases, use less
expensive materials. Thus they have the potential to reduce tooling costs enough to make
low volume injection molding economically feasible.
Research in the area of rapid tooling for injection molding is varied. The work
described in section 2.1.1 includes some research with rapid prototyped tools. Additional
work specifically pertaining to stereolithography and laser sintering, is summarized
below.
Laser sintering uses powdered metals for tooling and is not as challenging as
stereolithography in terms of strength and thermal conductivity, so there are more
examples of its use for injection molding prototyping in industry (e.g., Campbell 2000).
Laser sintering with copper polyamide has been used to build small, low volume mold
inserts, such as those for a brake reservoir and a glass guide (Nelson et al 1998). In these
45
cases, the advantages of laser sintering included durability for up to hundreds of parts,
low cost and lead time, and cycle times that are comparable to those with conventional
tools.
Pham, Dimov and Lacan (2000) studied characteristics related to laser sintered
tool accuracy, including shrinkage of the tool material and finishing requirements. Mold
insert accuracy requires fine tuning of scaling and offset factors due to shrinkage and
careful planning of tool finishing processes. Two case studies indicate successful use of
laser sintering for injection molding and gravity die casting. In cases for two injection
molded parts, Dalgarno and Stewart (2001) studied cycle time effects of conformal
cooling and molding costs based on tool durability, and compared laser sintered tool
results with conventional tooling. Due to tool finishing requirements, they found no lead
time advantage for the laser sintered tooling process. The laser sintered tools, however,
did exhibit cycle time savings with conformal cooling channels and economic benefits at
low demand rates. Other work includes optimization of shapes for heating and cooling
lines in mold inserts made with the direct metal laser sintering process (commercialized
by EOS), which sinters bronze particles and infiltrates with epoxy resin (Hopkinson &
Dickens 2000a).
Hopkinson and Dickens have investigated stereolithography tools for injection
molding, including tool failure, tool strength and ejection force. In a comparison of
stereolithography with aluminum injection mold tooling, they found that the low thermal
conductivity can be advantageous since the tool surface stays above its glass transition
temperature for easier ejection. Also, tool degradation due to thermal cycling is reduced,
46
and the ability to mold long thin slots is enhanced. Ejection forces in this work were
calculated using the equations developed by Glanvill and Menges et al (Hopkinson &
Dickens 2000b).
In other work, models were developed to predict tool strength and ejection force
(Hopkinson & Dickens 2000c, 2000d). Heat transfer through the tool was measured and
modeled. Then the results of the heat transfer analysis were used in a finite element
analysis model to predict tool strength. The model showed a decrease in tool strength
with increased cooling time before ejection. Ejection forces were predicted based on a
modified equation by Glanville and Denton. Longer cooling times were found to lead to
higher ejection forces, as expected, due to part shrinkage onto the core. The predicted
values, however, were approximately 30 percent lower than actual values, and the
measured values contained some inherent variation.
Harris and Dickens (2001) explored two design variables for stereolithography
injection mold inserts, namely, layer thickness and draft angle. They found that ejection
forces increase with increasing stereolithography layer thickness and decreasing draft
angles, thus increasing the risk of mold breakage. Interestingly, the linear changes in
these two variables cause nonlinear changes in the ejection force, suggesting that
optimum values must be found that balance ejection force requirements with desired part
design and economy of stereolithography process. A later paper describes their study of
the morphology of thermoplastic materials injection molded from stereolithography and
aluminum tools (Harris & Dickens 2003). Parts from the stereolithography tool had
longer cooling times and higher crystallinity. Experimental work demonstrated that
47
crystallinity can be controlled by using a nucleating agent or by adjusting melt
temperature.
The work of Dickens and Rudgley (2001) demonstrates the successful use of
stereolithography resin inserts to mold an engineering polymer. With much lower
injection pressure, speed, and clamping force, poly ether ether Ketone (PEEK), an
engineering polymer, was injected into a room temperature stereolithography mold insert.
The low thermal conductivity of the insert allowed the mold to fill completely at the
lower pressure level. The part was ejected at a higher temperature so that the insert
flexed during ejection. The part was molded successfully, and the slower cooling
resulted in higher crystallinity as compared to parts molded in a conventional tool.
2.2.3 Summary
There are many rapid prototyping processes in use or under development today,
some of which have been described in this section. A subset of the rapid prototyping
processes can be applied to make tools, including injection mold inserts. The present
work investigates tools from two of these processes, laser sintering and stereolithography.
Laser sintering with powdered metal has been successfully used to build injection molds
for limited quantities of parts. Stereolithography with epoxy resin has been the subject of
research for injection mold inserts, but has not been used for production to any significant
extent. This work follows up much of the work described previously in this section by
taking a systematic look at inserts from these two processes for molding two different
thermoplastic materials. Ejection forces and friction coefficients are measured, compared
48
with model-based calculations, and baselined against a machined steel insert. The data
collected help to determine the applicability of these rapid tools to injection molding, at
least in terms of their ability to withstand the forces of ejection.
49
CHAPTER 3
THEORY
This chapter includes necessary theoretical background on polymeric materials,
the coefficient of friction, and an ejection force model. First amorphous and crystalline
aspects of the thermoplastic materials used in this work are presented. Some friction
theory follows, including discussions on the deformation and adhesion components of
friction. The final section derives the primary ejection force equation used in this
research.
3.1 Thermoplastic Materials
Chemical structures and some properties of the two thermoplastics used in this
work are presented in this section because they relate to the shrinkage, friction, and
strength characteristics of the materials. These characteristics explain much of the
behavior of these materials in the present work during testing and experimentation and
described in Chapters 5 and 6.
50
A thermoplastic material is a polymer that has a linear macromolecular structure
and will repeatedly soften when heated and harden when cooled. Examples of
thermoplastics include styrenes, acrylics, polyethylenes, vinyls, and nylons. A crystalline
thermoplastic has sections of crystallinity, i.e., periodic ordering of molecules, whereas
an amorphous thermoplastic lacks any long range molecular order. The characteristic
differences between amorphous and crystalline polymers determine processing
parameters and influence the properties of an injection molded part.
Amorphous polymers have a second order transition, or glass transition
temperature, above which the material flows, and below which the material is glassy
(Trantina & Nimmer 1994). In general, they have lower and more uniform shrinkage,
greater post-mold stability, and high melt viscosities. Amorphous polymers also tend to
be more susceptible to chemical attack.
Crystalline polymers have a well-defined melting point below which crystals are
formed, and above which the crystals dissolve and the material flows. In general they
shrink more, and shrink more anisotropically, have low melt viscosities (long flow
lengths), and have more temperature dependent mechanical properties. Crystalline
polymers also tend to be more resistant to solvents.
3.1.1 High Impact Polystyrene
Polystyrene is a vinyl polymer (i.e., formed from hydrocarbon monomers with
double carbon bonds) having a phenyl group attached to every other carbon atom in its
hydrocarbon chain (Figure 3.1) (University of Southern Mississippi 2002). In atactic
51
polystyrene, the phenyl groups are distributed on either side of the carbon atoms in a
random fashion. Thus, it is amorphous because its unwieldy and asymmetric structure is
not conducive to regular crystal formation.
Figure 3.1: Polystyrene monomer.
Polystyrene is formed using the free radical vinyl polymerization process. This
process depends on the use of initiators that, upon splitting, produce free radicals. The
unpaired electrons in the free radicals attack the double carbon bonds, pair with one
electron from that bond, and cause the other electron to become a free radical. The chain
reaction continues in this way to propagate the polymer.
High impact polystyrene (HIPS) is formed by adding polybutadiene rubber
monomers during the polymerization process. HIPS is a graft copolymer that has a
polystyrene backbone chain with polybutadiene grafted onto it (Figure 3.2). The
polystyrene provides strength to the material, while the polybutadiene renders it less
brittle.
52
Figure 3.2: High impact polystyrene.
53
The HIPS material used in this work is BASF PS 495F. Its glass transition
temperature is 100 oC (212 oF). This material is more brittle at room temperature
compared to high density polyethylene. The properties of PS 495F are given in Chapter
4.
3.1.2 High Density Polyethylene
Polyethylene is also a vinyl polymer with a very simple hydrocarbon chain
(Figure 3.3). High density polyethylene (HDPE) has linear molecules (Figure 3.4) that
can pack more tightly together, as opposed to low density polyethylene that has branched
molecules. Because of its regular symmetric structure, HDPE is conducive to crystal
formation and is considered a crystalline polymer.
HDPE cannot be produced using free radical vinyl polymerization because some
termination reactions result in branching of the molecules. Instead, the Ziegler-Natta
vinyl polymerization process is used. The Ziegler-Natta process involves transition metal
catalysts and co-catalysts based on the Group III metals, and it can produce polymers of a
specific tacticity (University of Southern Mississippi 2002).
54
Figure 3.3: Polyethylene monomer.
Figure 3.4: High density polyethylene linear molecule.
The HDPE used in this work is Lutene-H ME9180 from LG Chem. Its crystalline
melting point is 133 oC (271 oF). More on material selection is presented in Chapter 4.
3.2 The Adhesion Component of Friction
A few introductory concepts of adhesion are presented in this section because
adhesion plays an important part in the friction between the part and the injection mold
55
core. This can be seen in the present work, especially in the case of a HIPS part molded
in an epoxy insert, as described in Chapters 5 and 6.
In the basic friction equation, the friction force F between two sliding bodies is
equal to the normal force N pressing the bodies together, multiplied by a constant, i.e., the
coefficient of friction µ. The force required to initiate motion between the two bodies is
typically higher than the force required to maintain motion. Thus the coefficient of static
friction, µstatic, is defined as the ratio of the force necessary to initiate motion to the
normal force:
µstatic =
Fbreakaway
N
3.1
Friction is comprised of a deformation component and an adhesion component,
the latter of which is typically more prominent for polymer materials. While the
deformation (or mechanical) component of friction tends to be more easily defined, the
adhesion component is rather more complex. The following paragraphs include
theoretical background on adhesive bonding and adhesion theory.
In adhesive bonding, the surface tension of the adhesive should be less than the
free surface energy or critical surface tension of the adherend (ed. Cagle 1973). This
allows the adhesive to wet and spread. Wettability or tendency to adsorb can be
measured by the contact angle (between the adhesive and the surface to be bonded) or the
work of adhesion.
56
Forces in the wetting and spreading phenomena include chemical bonds,
mechanical entanglement, physical and chemical adsorption, electrostatic forces of
attraction, and combinations thereof. Physical adsorption involves secondary attractive
forces, i.e., van der Waals forces: molecules with permanent dipoles, dipoles induced by
permanent dipoles in neighboring molecules (Debye forces), and London dispersion
forces. Dispersion forces are dipoles produced by the motion of electrons and are
independent of molecular polarity. Dispersion forces are considered to be the major
attractive force even when polar groups and hydrogen bonding groups are present.
Hydrogen bonding is demonstrated by molecules with hydroxyl groups.
If the critical surface tension of the solid is greater than the surface tension of the
liquid, a good bond can occur. Surface free energies of metals range from 100 to 3000
ergs/cm2, while organic liquids (including molten thermoplastics) have surface free
energies of less than 100 ergs/cm2. Heat serves to increase the ability of the adhesive to
adsorb, dissolve, and disperse. Heat also decreases viscosity, thus increasing wetting and
adsorption. Pressure and heat together improve wetting and spreading of more viscous
materials.
Some of the equations of adhesion theory that derive from surface energy are
introduced below. Surface energy or surface tension is represented by γ, where subscripts
S and L represent solid and liquid, respectively (Wu 1982). Interfacial energy is
represented by γLS. The Young equation relates contact angle θ, formed between a drop
of liquid and a solid surface, to interfacial tensions as follows:
57
γ LV cos θ = γ SV − γ SL
3.2
where θ is the contact angle of a liquid on the plane surface of a solid, γLV is the surface
tension of the liquid in equilibrium with its saturated vapor, and γSV is the surface tension
of the solid in equilibrium with the saturated vapor of the liquid.
The Dupre equation for the work of adhesion Wa defines the work required to
reversibly separate the interface between two bulk phases and can be written as
Wa = γ L + γ S − γ LS
3.3
The Young-Dupre equation relates the work of adhesion, a thermodynamic
parameter, to two easily determined parameters, the contact angle and the liquid-vapor
surface tension:
Wa = γ LV (1 − cosθ )
3.4
When the surface energy of the liquid is smaller than that of the solid, θ will be small,
and adsorption will occur. Various molecular forces are linearly additive, and the work
of adhesion can be separated into two terms, a dispersion component and a polar
component:
Wa = Wad + Wap
3.5
58
The geometric mean relation is used when the interface is made up of a low-energy and a
high-energy material. Then the dispersion component of the work of adhesion is:
(
Wad = 2 γ Ld γ Sd
)
1
2
3.6
If dipole-dipole interaction is predominant, then the polar component of the work of
adhesion is:
(
Wap = 2 γ Lp γ Sp
)
1
2
3.7
It can be seen that if one of the materials is non-polar, then the polar component of
adhesion is zero. If the surface energies of the two materials due to polarity are similar,
the polar component of adhesion will be maximized (Wu 1982).
In the present work, adhesion is found to be high between HIPS and SL 5170
resin (see Chapter 5). The work of adhesion, or the work required to separate these two
material surfaces, is high compared to the other material pairs studied. The high work of
adhesion may be due to dispersive interactions, polar interactions, or both.
59
3.3 Ejection Force Model Derivation
The ejection force model derived in this section is a key component of the present
work. It is used for theoretical comparison to experimental measurement, both for the
ejection force and the coefficient of static friction, for all the injection mold insert and
thermoplastic material combinations used.
3.3.1 Model derivation
Ejection force equations are derived from the empirical law of the friction
phenomenon, presented above, in which the friction force between two surfaces is
proportional to the normal force pressing the two surfaces together:
F = µN
3.8
where N is the normal force and µ is the coefficient of friction, a characteristic constant
of the materials involved.
For deep parts with cores and cavities, the friction force is equal to the release
force FR, and the normal force results from the product of the contact pressure P and the
area of contact A (see Burke 1991):
60
FR = µPA
3.9
The stresses in an injection molded cylindrical part before ejection can be
modeled as stresses in a thin-walled cylindrical pressure vessel as shown in Figure 3.5
(Beer & Johnston 1981, p. 326). The radius of the core is r, and t is the wall thickness.
The stresses exerted on a small element of wall will be determined. The sides of the
element are respectively parallel and perpendicular to the axis of the cylinder. The vessel
and its contents are axisymmetric, so there are no shear stresses on the element, and σ1
and σ2 are principal stresses. The hoop stress is represented by σ1, and the longitudinal
stress is represented by σ2.
Figure 3.5: Thin-walled cylindrical pressure vessel.
For thin-walled pressure vessels, the term t/2r is considered sufficiently small
such that the stresses do not vary across the wall, and thus the core radius may be used in
the calculation in lieu of the mean radius of the wall section. Also, in this case the
61
longitudinal stress is assumed to be insignificant relative to the hoop stress, so only σ1
will be calculated here.
A detached portion of the part and the core, bounded by the xy plane and by two
planes parallel to the yz plane and separated by a distance ∆x, is used to determine the
hoop stress σ1 (Figure 3.6). The forces in the z direction acting on this free body are the
elementary internal forces on the wall sections σ1 dA and the elementary pressure forces
acting on the projected area of the core p dA.
Figure 3.6: Section of the part and the core with associated stresses.
The resultant of the internal forces σ1 dA equals the product of σ1 and the crosssectional area of the wall 2t ∆x. The resultant of the pressure forces p dA equals the
product of p and the area 2r ∆x. The sum of the forces in the z direction are:
62
ΣFz = 0 :
σ 1 (2t∆x ) − p (2r∆x ) = 0
3.10
Solving for hoop stress σ1:
σ1 =
pr
t
3.11
Rearranging equation 3.11 to solve for the pressure force, i.e., the contact pressure P in
this case:
P=
σt
rc
3.12
Next, Hooke’s Law is applied, assuming elasticity in the solidified part.
According to equation 3.12, contact pressure P is proportional to tensile (circumferential)
stress σ. Stress σ is directly proportional to the elastic modulus and strain Eε:
σ = E (T )ε
3.13
where E(T) is the elastic modulus at the ejection temperature, and ε represents
engineering strain. The injection molding case involves changing temperatures, thus
strain can be represented by thermal strain as follows:
63
ε = α (TM − TE )
3.14
where α is the coefficient of thermal expansion, TM is the melt temperature, and TE is the
temperature at ejection. Combining equations 3.13 and 3.14 gives:
σ = E (T )α (TM − TE )
3.15
As previously mentioned, the term t/2r must be sufficiently small in order to
apply the equation for thin-walled pressure vessels. For example, ensuring that t <
2rc
is
10
a good rule of thumb (Popov 1976, p 290). Combining equations 3.12 and 3.15 gives:
P=
E (T )α (TM − TE )t
rc
3.16
With the area of the cylinder A = πDc L and equations 3.9 and 3.16:
FR =
µE (T )α (TM − TE )tπDc L
rc
3.17
Menges et al (2001) approximate strain by the relative change in diameter ∆dr of
the cylinder immediately after ejection. With this change, then, the ejection force is:
64
FR =
µE (T )∆d r tπDc L
rc
3.18
3.3.2 Additional Consideration for Strain
The description of strain in equation 3.14 may be a simplification considering
other transformations within the material. For example, in addition to thermal strain,
there may also be crystallization strain for crystalline materials, reaction strain for
thermosets, and hydrostatic strain due to the compressibility of the material (Jansen &
Titomanlio 1996). Total strain would be the sum of thermal strain, hydrostatic strain, and
crystallization and reaction strain as applicable. In this work, reaction strain does not
apply, and crystallization strain would only apply to HDPE, since HIPS is amorphous.
By using a measure of relative change in diameter in place of thermal strain, as shown in
equation 3.18, all aspects of strain are taken into consideration. Thus ∆dr represents total
strain, and improves the accuracy of the ejection force model.
65
CHAPTER 4
EXPERIMENTATION
This chapter describes the details about how data were collected for this research.
For friction testing, this includes a description of the test apparatus and the test matrix
and procedure. The process for measurement of thermoplastic modulus at temperature is
then presented. For the injection molding experiments, the mold and part design, the
injection molding machine and process parameters, and the experimental design and
procedure are all explained. The last section on data acquisition summarizes core
temperature, ejection force, and part diameter measurements.
4.1 Friction Testing
Most published data on coefficients of friction for thermoplastics result from
room temperature tests against steel or against like materials. For example, the ASM
Handbook lists friction data for polymers on steel as shown in Table 4.1 (ASM
International 1992). Actual friction coefficients during the injection molding process are
difficult to determine because of the rapidly changing temperature and pressure
66
environment that exists. In order to have reasonable values against which to compare
friction values determined from the experimental injection molding data, standard friction
testing was conducted using the same mold insert materials and thermoplastics used in
the experiments while more closely simulating processing conditions.
Fixed
Specimen
Steel, 52100
Steel, carbon
Steel, mild
Moving
Specimen
HDPE
HDPE
Polystyrene
polystyrene
Test
Static
Geometry
CoF
pin-on-disc
--pin-on-flat
0.36
pin-on-flat
0.43
thrust washer 0.28
Kinetic
CoF
0.25
0.23
0.37
0.32
Table 4.1: Friction data for polymers on steel.
The coefficients of static friction of HDPE and HIPS were measured against P-20
mold steel, LaserForm ST-100, and SL 5170 stereolithography resin following a
modified ASTM D 1894, “Standard Test Method for Static and Kinetic Coefficients of
Friction of Plastic Film and Sheeting.” A schematic of the friction apparatus is shown in
Figure 4.1. Measurements were made first at room temperature; second, at ejection
temperature; and third, after the specimen was heated to a higher temperature, pressed
against the plate, and cooled to ejection temperature. The purpose of these tests was to
compare coefficients of friction among injection mold insert materials, and at elevated
67
temperatures that more closely resemble processing conditions. The polymers tested
were identical to the polymers used in the subsequent molding experiments.
A.
Sled
B.
Plane
C.
Supporting Base
D.
Gage
E.
Tensile Tester Crosshead
F.
Braided Wire
G.
Pulley
Ref. ASTM D 1894, figure 1.
Figure 4.1: Schematic of friction apparatus.
4.1.1 Friction Test Apparatus
The friction tests were conducted in accordance with the modified ASTM D 1894
procedure. The equipment used consists of a Coefficient of Friction Sled Fixture
(Material Testing Technology Co.) installed in an Instron model 4507 tester equipped
with a Sensotec 25-lb load cell and a Bemco furnace (Figure 4.2). Data were collected at
a rate of 100 samples per second. Thermocouples were mounted at two locations on each
plate specimen to measure temperature.
68
Figure 4.2: Friction test apparatus: sled on plate specimen (left) inside furnace and
tester (right).
Modifications to the ASTM D 1894 standard test include elevated temperatures
for some tests, a slower pull speed of 25 mm (1 inch) per minute instead of the
recommended 150 mm (6 inches) per minute, and a shorter pull distance, i.e.,
approximately 25 mm (1 inch) vs. the recommended 125 mm (5 inches). The slower pull
speed was more appropriate for measuring static coefficient of friction since the higher
speed caused a sudden jerk of the sled, and an unreliable measurement of force. The
69
shorter pull distance was used because, for determining static friction, only the force
necessary to set the sled in motion is required.
4.1.2 Test Matrix and Procedure
The friction test matrix is shown in Table 4.2. Two thermoplastics and three mold
insert materials were tested under three temperature conditions. The thermoplastic
specimens were 63.5 mm (2.5 inches) square and attached to the sled using double-sided,
high temperature fiberglass tape. The mold insert material specimens were
approximately 125 mm (5 inches) wide by 250 mm (10 inches) long and were positioned
on the base plate of the apparatus. The surface roughnesses of the plate specimens were
0.7 microns (28 microinches) for P-20, 0.2 microns (8 microinches) for ST-100, and 3.6
microns (142 microinches) for SL 5170.
The first temperature condition was room temperature, 22 oC (71 oF). The second
condition was ejection temperature, i.e., the temperature at which molded parts are
ejected from the injection molding machine. The second temperature condition simulated
the environment in which friction is encountered in the injection molding process, i.e., 50
o
C (120 oF) for P-20 and ST-100 plate materials and 55 oC (130 oF) for SL 5170 plate
material. The third condition consisted of first heating to an elevated temperature (120
o
C (250 oF) for HDPE specimens and 150 oC (300 oF) for HIPS specimens), then cooling
down to the ejection temperatures mentioned above before testing. While at the elevated
temperature, a 0.9 kg (2-lb) mass was placed on the sled. When the plate cooled to
ejection temperature, the weight was removed and the specimen tested. The purpose of
70
the third temperature condition was to imprint the surface of the plate specimen onto the
sled specimen. This simulated the environment as well as the surface condition that
occurs during ejection of a molded part.
Elevated Temp
Ejection Temp
Ambient
Plate Specimen
P-20 Steel
P-20 Steel
ST-100 (Sintered)
ST-100 (Sintered)
SL 5170 (Resin)
SL 5170 (Resin)
P-20 Steel
P-20 Steel
ST-100 (Sintered)
ST-100 (Sintered)
SL 5170 (Resin)
SL 5170 (Resin)
P-20 Steel
P-20 Steel
ST-100 (Sintered)
ST-100 (Sintered)
SL 5170 (Resin)
SL 5170 (Resin)
Sled Specimen
Temporary Load
HDPE
HIPS
HDPE
HIPS
HDPE
HIPS
HDPE
HIPS
HDPE
HIPS
HDPE
HIPS
HDPE
HIPS
HDPE
HIPS
HDPE
HIPS
None
None
None
None
None
None
None
None
None
None
None
None
0.9 kg (2 lbs)
0.9 kg (2 lbs)
0.9 kg (2 lbs)
0.9 kg (2 lbs)
0.9 kg (2 lbs)
0.9 kg (2 lbs)
Initial Plate Temp
o
o
RT 22 C (71 F)
o
o
RT 22 C (71 F)
o
o
RT 22 C (71 F)
o
o
RT 22 C (71 F)
o
o
RT 22 C (71 F)
o
o
RT 22 C (71 F)
o
o
50 C (120 F)
o
o
50 C (120 F)
o
o
50 C (120 F)
o
o
50 C (120 F)
o
o
55 C (130 F)
o
o
55 C (130 F)
o
o
120 C (250 F)
o
o
150 C (300 F)
o
o
120 C (250 F)
o
o
150 C (300 F)
o
o
120 C (250 F)
o
o
150 C (300 F)
Soak Time
N/A
N/A
N/A
N/A
N/A
N/A
1 min
1 min
1 min
1 min
2 min
2 min
1 min
1 min
1 min
1 min
2 min
2 min
Plate Temp at Pull
o
o
RT 22 C (71 F)
o
o
RT 22 C (71 F)
o
o
RT 22 C (71 F)
o
o
RT 22 C (71 F)
o
o
RT 22 C (71 F)
o
o
RT 22 C (71 F)
o
o
50 C (120 F)
o
o
50 C (120 F)
o
o
50 C (120 F)
o
o
50 C (120 F)
o
o
55 C (130 F)
o
o
55 C (130 F)
o
o
50 C (120 F)
o
o
50 C (120 F)
o
o
50 C (120 F)
o
o
50 C (120 F)
o
o
55 C (130 F)
o
o
55 C (130 F)
Table 4.2: Friction test matrix.
The test procedure began with insertion of the proper materials. The plate
specimen was cleaned with acetone prior to each set of tests. For room temperature tests,
the sled was pulled until it moved along the plate. In the ejection temperature tests, the
furnace was ramped until the plate reached the specified initial temperature. After the
apparatus was soaked for the specified amount of time, the sled was pulled until it moved
71
along the plate. In the elevated temperature tests, a weight was placed evenly on top of
the sled, and the furnace was ramped until the plate reached the specified initial
temperature. After the specified soak time, the plate was cooled to the specified
temperature at pull (ejection temperature). The weight was then removed, and the sled
was pulled until it moved along the plate. In all cases the pulling force and time were
recorded. Each set of test conditions was repeated five times, using a new sled specimen
for each test.
4.2 Measurement of Elastic Modulus
The equation for ejection force requires values for the elastic modulus of the
molding material at the ejection temperature. Elastic moduli for the HDPE and HIPS
materials used in this research were measured at various temperatures using ASTM D
638 “Standard Test Method for Tensile Properties of Plastics” as a guide. The testing
apparatus was an Instron model 1322 tensile tester with a tube furnace. An extensometer
with a 2-inch gauge and 50 percent strain was used to measure elongation (Figure 4.3).
72
Figure 4.3: Tensile testing apparatus with tube furnace.
ASTM Type I (dogbone) specimens were molded from each thermoplastic
material and were tested at room temperature, 30oC, and at ten degree increments until no
elastic region was detected (Figure 4.4). HDPE was tested through 70oC, and HIPS was
tested through 50oC. At least three samples of HIPS were tested at each temperature,
while at least 5 samples of HDPE were tested at each temperature. Data from the tensile
tests are included in Appendix A, and a graph of the results is shown in Figure 4.5.
Modulus values for HIPS at higher ejection temperatures than 50oC were determined by
extrapolating the graph. The lookup table generated from this graph and used in the
calculations of friction coefficient can also be found in Appendix A.
73
Figure 4.4: HIPS specimens after tensile tests.
Modulus (MPa)
4000
3500
HIPS PS 495F
3000
HDPE ME 9180
2500
2000
1500
1000
500
0
20
30
40
50
60
70
80
Temperature (°C)
Figure 4.5: Elastic modulus at various temperatures for HDPE and HIPS.
74
4.3 Injection Molding
4.3.1 Mold Design and Materials
Since this research involved the study of different rapid tooling materials, a
modular mold design was employed. The mold base was a steel Master Unit Die (MUD)
having a core and cavity that could be removed and replaced with those of other materials
(Figure 4.6). The baseline core and cavity were made of P-20 steel, a typical mold steel,
and the two rapid tooled core and cavity sets were made of SL 5170 epoxy resin and
LaserForm ST-100 material (Figure 4.7). The SL 5170 insert was built at NASA
Marshall Space Flight Center using the stereolithography process, and the ST-100 insert
was built at General Pattern Company (Blaine, MN) using the laser sintering process, as
described in Chapter 2. Machining allowances were included in the design of each rapid
tooled insert so that it could be machined to fit properly into the mold base. The baseline
steel insert core had a surface finish of Ra = 0.7 microns (28 microinches). The
stereolithography insert also had a surface finish of Ra = 0.7 microns (28 microinches),
and the laser sintered insert had a surface finish of Ra = 0.3 microns (12 microinches).
75
Figure 4.6: Sprue side of the MUD base mounted in the injection molding machine
with SL 5170 cavity insert.
Figure 4.7: Core and cavity inserts, before final machining, made of SL 5170 (left),
P-20 steel (center), and LaserForm ST-100 (right).
76
The experimental part was a closed-end, straight cylinder with a 32 mm (1.26 in)
outside diameter, 49.6 mm (1.95 in) height, and 1.2 mm (0.05 in) wall thickness (Figure
4.8). Design drawings for the canister and its injection mold are included in Appendix B.
The canister was designed with four vent holes in the base to prevent vacuum forces that
would result during ejection from the core. The part was similar in size and shape to the
canisters used to store 35 mm photographic film. This particular part was selected
because it required a simple core and cavity that eliminated the effect of corners, and its
non-tapered design allowed for a significant and measurable ejection force.
The dimensions of the SL 5170 inserts were modified to alleviate problems with
core swelling and parts sticking in the cavity. The core diameter was reduced by 0.1 mm,
(0.005 in) to prevent interference between the stripper plate and the core when the core
swelled with temperature. All parts from the SL 5170 core had this slight increase in
diameter. The cavity wall was tapered 0.42o, leaving the canister base dimension intact,
in an attempt to prevent parts from sticking in the cavity. Only the parts from the
experiments using both the SL 5170 core and cavity inserts were thicker at the rim, i.e.,
had an outside diameter of 32.8 mm (1.29 in), and were slightly tapered on the outside of
the wall.
77
Figure 4.8: Canister part with vent holes and no taper.
The mold was a single-cavity design with a heated sprue connecting directly to
the base of the canister. The hot sprue allowed more control over packing pressure, i.e.,
the packing material did not prematurely freeze at the gate. The core and cavity each had
a housing that fixed it to the MUD base. This ensured that there were no bolt holes
through the core or cavity inserts. The cavity insert had a square profile large enough to
provide for the possibility of adding cooling channels in the future. The core insert, on
the other hand, had a round profile so that material requirements could be reduced and the
machining process simplified. The core insert also had one flat surface for orientation. If
necessary, cooling channels could be added inside the core. Three thermocouples were
positioned at different depths inside each core insert.
The ejection system employed a stripper plate with a circular hole that fit around
the base of the core. The stripper plate was supported by four ejector pins that connected
to the mold ejector plate. Subminiature load cells for measuring ejection force were
78
positioned between each ejector pin and the mold ejector plate. Drawings of the mold
insert design are included in Appendix B.
4.3.2 The Injection Molding Process
A Sumitomo Injection Molding Machine, model SH50M, was used for the
experimental portion of this work (Figure 4.9). It was a horizontal press with a fully
hydraulic, 50-ton clamping system. Machine specifications are given in Table 4.3.
Figure 4.9: Sumitomo SH50M injection molding machine.
79
Model
Clamping System
Clamp Force
Distance Between Tie Bars
Overall Size of Platen
Opening Stroke
Ejector Type
Ejector Stroke
Ejector Force
Screw Diameter
Injection Capacity
Injection Rate
Nozzle Contact Force
Machine Weight
SH50M
Fully Hydraulic
50 metric tons (55.1 short tons)
325 x 325 mm (12.8 x 12.8 in)
470 x 467 mm (18.5 x 18.4 in)
440 mm (17.3 in)
Hydraulic, cross multipoint ejection (5pts)
70 mm (2.8 in)
2.2 metric tons (2.42 short tons)
28 mm (1.1 in)
3
3
70 cm (4.3 in )
3
3
99 cm /s (6.0 in /s)
4670 kgf (10297 lbf)
2.2 metric tons (2.42 short tons)
Table 4.3: Injection molding machine specifications.
The procedure for defining injection molding process parameters was intended
first to establish the volume of material required, and then to determine the velocity
required to completely fill the mold with no flashing. Barrel zone temperatures were set
based on commonly used temperatures for injection molding the given thermoplastic
material. For the steel and sintered inserts, with velocity at 50 percent of maximum and
packing pressure at zero, the screw position was initially set for a short shot, and then
gradually extended until the part filled. Velocity was increased if the part froze before
the entire shot could be injected, and decreased if flashing occurred. All experiments,
except those using the SL 5170 insert, were run with one stage packing pressure, 25%
maximum screw rpm, 5% maximum back pressure, and 15% maximum ejection velocity.
80
The velocity and temperature parameters for each set of experiments are shown in Table
4.4. Two levels of packing time were defined at 2 and 6 seconds. Three levels of cooling
time were defined at 5, 10, and 15 seconds. Packing pressure levels were defined at 0, 5,
and 10 percent of maximum (0, 10.93, and 21.87 MPa). Clamping force was 20 metric
tons.
Since the SL 5170 insert was expected to be less durable, temperature and
velocity settings were reduced as far as possible such that the mold would still fill. As
much as possible, the number of test runs on this insert were minimized. Screw rpm,
back pressure, clamp force, and ejection velocity were the same as above. Packing time
levels remained the same (2 and 6 seconds), while the number of cooling time and
packing pressure levels were reduced from three to two. Cooling times were greatly
increased to 120 and 150 seconds to allow for the low thermal conductivity of the
stereolithography resin. Packing pressure levels were 0 and 5 percent (0 and 10.93 MPa).
81
HDPE with P-20 Steel and LaserForm ST-100 Inserts
Velocity
35% or 56 mm/s (2.2 in/s)
Temperature Profile:
Sprue
o
210 C
Nozzle
Front
o
o
210 C
199 C
Middle
o
193 C
Rear
o
177 C
HIPS with P-20 Steel and LaserForm ST-100 Inserts
Velocity
40% or 64 mm/s (2.5 in/s) for P-20, 35% or 56 mm/s (2.2 in/s) for ST-100
Temperature Profile:
Sprue
o
221 C
Nozzle
Front
o
o
221 C
Middle
o
Rear
o
213 C
204 C
191 C
Front
Middle
Rear
HDPE with SL 5170 Insert, SL or P-20 Cavity
Velocity
25% or 40 mm/s (1.6 in/s)
Temperature Profile:
Sprue
o
177 C
Nozzle
o
o
177 C
o
o
171 C
166 C
160 C
Front
Middle
Rear
HIPS with SL 5170 Insert, SL or P-20 Cavity
Velocity
40% or 64 mm/s (2.5 in/s)
Temperature Profile:
Sprue
o
210 C
Nozzle
o
o
216 C
202 C
o
193 C
o
182 C
Table 4.4: Injection Molding Parameters
The thermoplastic materials used in the experiments, HDPE and HIPS, are
described in Chapter 3. Prior to molding, both materials were dried for two hours in a
desiccant dryer, with dew point at -40 oC (-40 oF) and air temperature at 71 oC (160 oF).
Materials data for HDPE (Lutene-H ME9180) are shown in Table 4.5, and for HIPS
(BASF PS 495F) are shown in Table 4.6.
82
Melt Flow Index
Density
Tensile Strength @ Yield
Tensile Strength @ Break
Flexural Modulus
Vicat Softening Temperature
18.0 g/10 min
3
0.958 g/cm
2
290 kg/cm (4125 psi)
<1000%
2
10,000 kg/cm (142 kpsi)
o
o
123 C (253 F)
ASTM D 1238
ASTM D 1505
ASTM D 638
ASTM D 638
ASTM D 790
ASTM D 1525
Table 4.5: Typical data for Lutene-H ME9180.
Melt Flow Index
Impact Strength, Izod
Tensile Strength @ Yield
Tensile Elongation @ Break
Flexural Modulus
Vicat Softening Temperature
7 g/10 min
112 J/m (2.1 ft-lb/in)
20 MPa (2900 psi)
55%
1655 MPa (240 kpsi)
o
o
101 C (214 F)
Table 4.6: Typical data for BASF PS 495F.
83
ASTM D 1238
ASTM D 256
ASTM D 638
ASTM D 638
ASTM D 790
ASTM D 1525
4.3.3 Design of Experiments
Statistical design of experiments has been used to design the six sets of
experiments run in this work. Each set was blocked by insert material and thermoplastic
material, and then randomized by packing time, cooling time and packing pressure. In
the first four experimental sets, there were two levels of packing time, three levels of
cooling time, and three levels of packing pressure as described above. Each combination
of factors was repeated eight times. The experimental design with process parameters is
shown in Table 4.7.
The last two experimental sets, using the SL 5170 insert, were designed to be
smaller than the other sets due to the expected low durability of the insert material. In
this case there were two levels each of packing time, cooling time, and packing pressure
as described above. Each combination of factors was repeated five times. A much
longer cooling time was required for this insert to accommodate its low thermal
conductivity. It was initially intended that cooling time levels be set at 150 and 180
seconds. However these levels were reduced to 120 and 150 seconds to reduce the
shrinkage on the core, and thus ejection force requirements. Also note that only a limited
number of parts could be processed using the SL 5170 cavities due to deformation that
caused sticking of parts (see Chapter 6). The designed experiment, therefore, was carried
out using the SL 5170 core with the P-20 cavity. While the use of steel cavity material
greatly changed the thermal performance of this insert and reduced the temperature at
ejection, it also allowed a complete experiment to be performed in which ejection force
from the SL 5170 core could be measured.
84
Design of experiments (DOE) analyses in Minitab® have been performed using
packing time, cooling time and packing pressure as factors and ejection force as the
response. Using analysis of variance (ANOVA), the effects of variables and their
interactions on each response were determined. ANOVA tables for each data set are
included in Appendix A. DOE results and graphs identifying main effects and
interactions are included in Chapter 5.
85
SET 1
P-20
HDPE
8 Reps
Run 1
Tp = 2 s
Tc = 15 s
Pp = 0%
Run 10
Tp = 6 s
Tc = 15 s
Pp = 10%
Run 2
Tp = 2 s
Tc = 5 s
Pp = 10%
Run 11
Tp = 6 s
Tc = 5 s
Pp = 0%
Run 3
Tp = 2 s
Tc = 15 s
Pp = 10%
Run 12
Tp = 6 s
Tc = 5 s
Pp = 5%
Run 4
Tp = 2 s
Tc = 15 s
Pp = 5%
Run 13
Tp = 6 s
Tc = 15 s
Pp = 5%
Run 5
Tp = 2 s
Tc = 10 s
Pp = 10%
Run 14
Tp = 6 s
Tc = 10 s
Pp = 0%
Run 6
Tp = 2 s
Tc = 5 s
Pp = 5%
Run 15
Tp = 6 s
Tc = 10 s
Pp = 5%
Run 7
Tp = 2 s
Tc = 5 s
Pp = 0%
Run 16
Tp = 6 s
Tc = 15 s
Pp = 0%
Run 8
Tp = 2 s
Tc = 10 s
Pp = 0%
Run 17
Tp = 6 s
Tc = 5 s
Pp = 10%
Run 9
Tp = 2 s
Tc = 10 s
Pp = 5%
Run 18
Tp = 6 s
Tc = 10 s
Pp = 10%
SET 2
P-20
HIPS
8 Reps
Run 1
Tp = 2 s
Tc = 15 s
Pp = 0%
Run 10
Tp = 6 s
Tc = 15 s
Pp = 10%
Run 2
Tp = 2 s
Tc = 5 s
Pp = 10%
Run 11
Tp = 6 s
Tc = 5 s
Pp = 0%
Run 3
Tp = 2 s
Tc = 15 s
Pp = 10%
Run 12
Tp = 6 s
Tc = 5 s
Pp = 5%
Run 4
Tp = 2 s
Tc = 15 s
Pp = 5%
Run 13
Tp = 6 s
Tc = 15 s
Pp = 5%
Run 5
Tp = 2 s
Tc = 10 s
Pp = 10%
Run 14
Tp = 6 s
Tc = 10 s
Pp = 0%
Run 6
Tp = 2 s
Tc = 5 s
Pp = 5%
Run 15
Tp = 6 s
Tc = 10 s
Pp = 5%
Run 7
Tp = 2 s
Tc = 5 s
Pp = 0%
Run 16
Tp = 6 s
Tc = 15 s
Pp = 0%
Run 8
Tp = 2 s
Tc = 10 s
Pp = 0%
Run 17
Tp = 6 s
Tc = 5 s
Pp = 10%
Run 9
Tp = 2 s
Tc = 10 s
Pp = 5%
Run 18
Tp = 6 s
Tc = 10 s
Pp = 10%
SET 3
ST-100
HDPE
8 Reps
Run 1
Tp = 2 s
Tc = 15 s
Pp = 0%
Run 10
Tp = 6 s
Tc = 15 s
Pp = 10%
Run 2
Tp = 2 s
Tc = 5 s
Pp = 10%
Run 11
Tp = 6 s
Tc = 5 s
Pp = 0%
Run 3
Tp = 2 s
Tc = 15 s
Pp = 10%
Run 12
Tp = 6 s
Tc = 5 s
Pp = 5%
Run 4
Tp = 2 s
Tc = 15 s
Pp = 5%
Run 13
Tp = 6 s
Tc = 15 s
Pp = 5%
Run 5
Tp = 2 s
Tc = 10 s
Pp = 10%
Run 14
Tp = 6 s
Tc = 10 s
Pp = 0%
Run 6
Tp = 2 s
Tc = 5 s
Pp = 5%
Run 15
Tp = 6 s
Tc = 10 s
Pp = 5%
Run 7
Tp = 2 s
Tc = 5 s
Pp = 0%
Run 16
Tp = 6 s
Tc = 15 s
Pp = 0%
Run 8
Tp = 2 s
Tc = 10 s
Pp = 0%
Run 17
Tp = 6 s
Tc = 5 s
Pp = 10%
Run 9
Tp = 2 s
Tc = 10 s
Pp = 5%
Run 18
Tp = 6 s
Tc = 10 s
Pp = 10%
SET 4
ST-100
HIPS
8 Reps
Run 1
Tp = 2 s
Tc = 15 s
Pp = 0%
Run 10
Tp = 6 s
Tc = 15 s
Pp = 10%
Run 2
Tp = 2 s
Tc = 5 s
Pp = 10%
Run 11
Tp = 6 s
Tc = 5 s
Pp = 0%
Run 3
Tp = 2 s
Tc = 15 s
Pp = 10%
Run 12
Tp = 6 s
Tc = 5 s
Pp = 5%
Run 4
Tp = 2 s
Tc = 15 s
Pp = 5%
Run 13
Tp = 6 s
Tc = 15 s
Pp = 5%
Run 5
Tp = 2 s
Tc = 10 s
Pp = 10%
Run 14
Tp = 6 s
Tc = 10 s
Pp = 0%
Run 6
Tp = 2 s
Tc = 5 s
Pp = 5%
Run 15
Tp = 6 s
Tc = 10 s
Pp = 5%
Run 7
Tp = 2 s
Tc = 5 s
Pp = 0%
Run 16
Tp = 6 s
Tc = 15 s
Pp = 0%
Run 8
Tp = 2 s
Tc = 10 s
Pp = 0%
Run 17
Tp = 6 s
Tc = 5 s
Pp = 10%
Run 9
Tp = 2 s
Tc = 10 s
Pp = 5%
Run 18
Tp = 6 s
Tc = 10 s
Pp = 10%
SET 5
Run 1
Run 2
Run 3
Run 4
Run 5
Run 6
Run 7
Run 8
SL 5170 Tp = 2 s Tp = 2s
Tp = 6 s Tp = 6 s Tp = 2s
Tp = 2s
Tp = 6 s Tp = 6 s
HDPE Tc = 150s Tc = 120s Tc = 150s Tc = 120s Tc = 150s Tc = 120s Tc = 150s Tc = 120s
5 Reps Pp = 0% Pp = 0% Pp = 0% Pp = 0% Pp = 5% Pp = 5% Pp = 5% Pp = 5%
SET 6
Run 1
Run 2
Run 3
Run 4
Run 5
Run 6
Run 7
Run 8
SL 5170 Tp = 2 s Tp = 2s
Tp = 6 s Tp = 6 s Tp = 2s
Tp = 2s
Tp = 6 s Tp = 6 s
HIPS
Tc = 150s Tc = 120s Tc = 150s Tc = 120s Tc = 150s Tc = 120s Tc = 150s Tc = 120s
5 Reps Pp = 0% Pp = 0% Pp = 0% Pp = 0% Pp = 5% Pp = 5% Pp = 5% Pp = 5%
Table 4.7: Experimental design for six sets, including process parameters.
86
4.3.4 Experimental Procedure
An experimental set was defined by the mold insert and thermoplastic material.
The procedure followed for each experimental set is as follows:
1) Begin Set
2) Setup hardware and load thermoplastic material
3) Load parameter settings
4) Begin Run
i) Adjust DOE settings for packing time, cooling time and packing pressure
ii) Process test parts to bring insert to temperature
iii) Begin Iteration
(a) Inject part
(b) Capture ejection force and temperature during ejection
(c) Digitally photograph part immediately after ejection
(d) Wait for core to return to desired temperature
iv) Repeat iteration 8 times (5 times with SL 5170 insert)
5) Repeat run for all DOE combinations
6) Go to next set
Ejection force data were used to determine the initial force required to release
each part from the core. Thermal data were used to determine the elastic modulus of
each thermoplastic at ejection. Inside and outside diameters of the canister parts,
87
measured from digital photographs, were used to determine part thickness. All of these
data were used with the Menges ejection force model to calculate the ejection force and
the apparent coefficient of static friction. Statistical analysis of variance, as described
above, was conducted using packing time, cooling time, and packing pressure as input
factors, and ejection force as the response. Experimental results and conclusions are
included in Chapters 5 and 6. Experimental data are included in Appendix A.
4.4 Set-up and Data Acquisition
The data required from the injection molding experiments included thermal (core
temperature at ejection), load (ejection force), and dimensional (part diameter and
thickness at ejection) measurements. Type J thermocouples were used to measure
temperature, subminiature load cells were used to measure force, and digital imaging was
used to measure diameter. The three thermal and four load sensors collected data through
a National Instruments SC-2311 signal conditioner with associated thermocouple and
strain gage input modules. A LabVIEW™ program was used to read the ejection force
and temperature data and write them to a Microsoft Excel® spreadsheet. Pictures of each
part were taken with an Olympus Camedia C-740 digital camera and processed using
Adobe PhotoShop software. The inside and outside diameters of each part were
measured in pixels and converted to inches and millimeters by reference to a scale, a
picture of which was taken during each set of experiments. The data acquisition
equipment is shown in Figure 4.10.
88
Figure 4.10: Signal conditioner and computer with front panel for data acquisition
(left), and core side of mold with thermocouple and load cell sensor wires (right).
4.4.1 Temperature Measurement and Thermal Model
Three NANMAC Type J thermocouples were used to measure temperature in the
core at ejection. The thermocouples were installed by friction-fitting them into holes
drilled at three depths into the core (Figure 4.11). Representative thermal traces of the
injection molding cycle are shown in Figure 4.12.
89
Figure 4.11: Thermocouple placement within core insert.
90
Core Temperature (C) vs. Time
End of Core
Mid Core
Base of Core
55
50
45
Cooling Time
Injection
Mold Open
40
35
Core Temperature (C) vs. Time
End of Core
Mid Core
Base of Core
55
50
45
Cooling Time
Mold Open
40
Injection
35
Figure 4.12: Representative thermal traces of the injection molding cycle.
91
Since the distance between each thermocouple and the surface of the core was
more than a millimeter, there was a time lag from heating of the surface to heating of the
thermocouple. A thermal analysis was run to determine the convergence time between
the thermocouple reading and the actual temperature at the surface of the core (Carpenter
2004). The analysis simulated injection of the thermoplastic into the mold insert for each
material combination. The simulation was run using ABAQUS, and the results show the
time required for temperature at the thermocouple to match the temperature at the surface
of the core (Table 4.8). Initial and boundary conditions for the thermal analysis are
shown in Table 4.9, and graphs of the results are shown in Figure 4.13. The table of
temperature values is included in Appendix A. The graphs show that the P-20 and ST100 thermocouple readings and surface temperatures converge within 5 seconds for
HDPE, and within 10 seconds for HIPS. Convergence in the SL 5170 insert requires 110
seconds for HDPE and 120 seconds for HIPS. All processing times allow for these
convergence times so that thermocouple readings are accurate. It can be seen from the
simulation of the SL 5170 core with the P-20 cavity that convergence times are
approximately 10 seconds for HDPE and 20 seconds for HIPS. The longer cooling times,
therefore, are not necessary. In the experiments with this combination insert, however,
the longer, more conservative cooling times were used.
92
Insert Material
P-20 Steel
P-20 Steel
ST-100
ST-100
SL 5170
SL 5170
SL 5170/P-20
SL 5170/P-20
Thermoplastic
Material
HDPE
HIPS
HDPE
HIPS
HDPE
HIPS
HDPE
HIPS
Convergence Time
(s)
4.3
9.3
4.3
9.3
120.4
121.1
10.4
14.9
Table 4.8: Resulting convergence times from the thermal simulation.
P-20
Density
(kg/m3)
Specific Heat
(J/kg.C)
Thermal
Conductivity
(W/m.C)
Mold
Temperature
Polymer
Injection
Temperature
Material Properties
ST-100
SL 5170
HDPE
HIPS
7,870
7,700
1,220
958
1,050
486
475.2
1,674
2,200
2,000
47.6
49
0.2
0.39
0.16
P-20 and
HDPE
P-20 and
HIPS
50 °C
50 °C
50 °C
50 °C
30 °C
33 °C
30 °C
33 °C
210 °C
221 °C
210 °C
221 °C
177 °C
210 °C
177 °C
210 °C
Initial Temperature Condition
SL 5170
SL/P-20
SL/P-20
ST-100 and ST-100 and SL 5170
HIPS
and HDPE and HIPS and HDPE and HIPS
HDPE
Table 4.9: Input conditions for the thermal analysis.
93
P-20 and HDPE
Thermocouple Reading
Core Surface Temperature
250
Temperature (C)
200
150
100
50
0
0
5
10
15
20
25
30
Time (s)
P-20 and HIPS
Thermocouple Reading
Core Surface Temperature
250
Temperature (C)
200
150
100
50
0
0
5
10
15
20
25
30
Time (s)
Figure 4.13: Graphs of the thermal analysis results for each material combination.
(continued)
94
Figure 4.13 (continued.)
ST-100 and HDPE
Thermocouple Reading
Core Surface Temperature
250
Temperature (C)
200
150
100
50
0
0
5
10
15
20
25
30
Time (s)
ST-100 and HIPS
Thermocouple Reading
Core Surface Temperature
250
Temperature (C)
200
150
100
50
0
0
5
10
15
20
25
30
Time (s)
(continued)
95
Figure 4.13 (continued.)
SL 5170 and HDPE
Thermocouple Reading
Core Surface Temperature
200
180
Temperature (C)
160
140
120
100
80
60
40
20
0
0
20
40
60
80
100
120
140
160
180
Time (s)
SL 5170 and HIPS
Thermocouple Reading
Core Surface Temperature
250
Temperature (C)
200
150
100
50
0
0
50
100
150
200
Time (s)
(continued)
96
Figure 4.13 (continued.)
SL Core P-20 Cavity and HDPE
Thermocouple Reading
Core Surface Temperature
200
180
Temperature (C)
160
140
120
100
80
60
40
20
0
0
10
20
30
40
50
60
Time (s)
SL Core P-20 Cavity and HIPS
Thermocouple Reading
Core Surface Temperature
250
Temperature (C)
200
150
100
50
0
0
10
20
30
Time (s)
97
40
50
60
4.4.2 Ejection Force Measurement
Four Sensotec subminiature load cells, rated at 100 lbs each, were used to
measure ejection force. Each load cell was installed between one of four ejector pins and
the ejector plate. The four pins, mounted on bearings to reduce friction, were attached to
the stripper plate, which removed the canister from the core. Total load was measured
during ejection of each part, and the ejection force was determined from the initial peak
load required to release the part from the insert core. Sample ejection force traces are
shown in Figure 4.14.
After each experimental run, the ejection force was measured without a part on
the core. This no-load ejection force, i.e., the force required to simply move the ejection
mechanism, was subtracted from the peak load measurement of every canister to
determine the actual force required to release the part.
98
Total Load (lbs) vs. Time (0.1s)
60
Part Release
50
40
Part Sliding off Core
30
20
10
0
Total Load (lbs) vs. Time (0.1 sec)
100
Part Release
90
80
70
Part Sliding off Core
60
50
40
30
20
10
0
Figure 4.14: Representative ejection force traces
99
4.4.3 Diameter and Thickness Measurement
The inside and outside diameters of each canister were measured immediately
after ejection to determine the relative change in diameter due to shrinkage and the
canister wall thickness. The relative change in diameter and the thickness measurements
are required in order to use the Menges equation (see Chapter 3). A digital picture of
each canister was taken immediately after ejection (Figure 4.15). A picture including a
scale was taken for each set of experiments. Using Adobe Photoshop® software, the
pictures were magnified and the inside and outside diameters of each canister were
measured in four places: vertically, horizontally, and at 45-degree angles. The four
measurements were averaged for each diameter value. The scale reference picture was
also magnified and measured to determine the number of pixels per inch. The resolution
for each set of measurements varied between 0.0009 and 0.0010 inches per pixel.
100
Figure 4.15: Digital pictures of HDPE canisters for measuring inside and outside
diameter.
4.4.4 Calculation of Static Friction Coefficient
Ejection forces were calculated using equations 3.18 and 3.25, as derived in
Chapter 3, and the data described above. The apparent coefficients of static friction were
calculated using equation 3.18. Spreadsheets for these data are included in Appendix A,
and results are given in Chapter 5.
101
CHAPTER 5
RESULTS AND ANALYSIS
This chapter summarizes results from the injection molding experiments and
standard friction tests. First, the data from each are discussed individually. Next,
experimental ejection forces are compared with those calculated from the ejection force
model, using friction coefficients from the standard tests. Then, calculations of the
coefficient of static friction, using data from the injection molding experiments, are
presented and compared to standard test results. Analysis of variance results from the
designed experiment are also included. The last section of the chapter presents a
qualitative analysis of the rapid tooled injection mold inserts, i.e., some observations of
how these tools performed.
102
5.1 Injection Molding Experiments
5.1.1 Experimental Results and Discussion
Ejection forces from the injection molding experiments for HDPE and HIPS are
shown in Table 5.1. Measured ejection force results are listed by levels of packing time,
cooling time, and packing pressure. Other experimental data, including diameter and
temperature measurements, are included in Appendix A. The following discussion
comments on the ejection force results according to thermoplastic material and mold
insert material.
103
Packing
Time
s
2
2
2
2
2
2
2
2
2
6
6
6
6
6
6
6
6
6
Cooling
Time
s
5
5
5
10
10
10
15
15
15
5
5
5
10
10
10
15
15
15
Packing
Pressure
%
0
5
10
0
5
10
0
5
10
0
5
10
0
5
10
0
5
10
Packing
Time
s
2
2
2
2
Cooling
Time
s
120
150
180
150
Packing
Pressure
%
0
0
0
5
2
2
2
2
6
6
6
6
120
120
150
150
120
120
150
150
0
5
0
5
0
5
0
5
Experimental Ejection Force
HDPE
HIPS
P-20
LaserForm
P-20
LaserForm
Steel
ST-100
Steel
ST-100
N
N
N
N
177.15
182.30
343.97
366.29
183.29
190.02
376.37
389.51
186.52
209.79
401.48
375.28
176.51
177.44
346.11
375.71
185.89
196.61
385.56
393.54
172.87
194.33
408.26
394.47
191.90
196.15
384.60
366.31
174.68
201.68
381.94
393.56
173.08
208.75
403.70
398.83
184.69
173.95
376.97
363.86
193.58
185.81
395.23
378.14
173.88
184.41
393.38
388.07
171.28
172.12
369.19
369.54
174.63
178.83
390.81
360.58
175.19
180.87
391.73
374.38
170.00
170.97
351.62
340.67
180.05
186.82
394.77
370.57
185.53
184.13
424.46
399.88
HDPE
SL 5170
SL 5170
w/P-20
N
N
239.06
193.21
HIPS
SL 5170
SL 5170
w/P-20
N
N
1334.27
1136.12
1512.25
274.21
299.65
258.76
278.18
313.33
297.38
321.09
317.93
695.76
826.28
610.12
845.02
770.24
939.13
702.28
892.15
Table 5.1: Experimental ejection force results for HDPE and HIPS according to
packing time, cooling time, and packing pressure parameters.
104
5.1.2 HDPE Experimental Ejection Force Results
Figure 5.1 shows experimental ejection force results for HDPE. Ejection forces
for HDPE from the P-20 core, averaged per run, are generally lower than from the ST100 core, which are lower than from the SL 5170 core. Ejection forces from the ST-100
core at low level packing time are higher than at high level packing time. The parts with
higher ejection force also had lower shrinkage values, as measured by the relative change
in diameter immediately after ejection.
Ejection forces for HDPE from the SL 5170 core with the P-20 cavity are higher
than from the SL 5170 insert because the P-20 cavity draws away much of the heat,
resulting in more shrinkage of the HDPE against the SL 5170 core.
105
Ejection Force HDPE
330
310
290
Ejection Force (N)
270
250
230
210
190
170
150
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
Run
P-20
ST-100
SL 5170
SL/P-20
Figure 5.1: Experimental ejection force results for HDPE, all runs.
106
16
17
18
5.1.3 HIPS Experimental Ejection Force Results
Figure 5.2 shows experimental ejection force results for HIPS. In contrast to
HDPE, ejection forces from the P-20 core, averaged per run, are greater than from the
ST-100 core for 12 out of 18 runs. Ejection forces from the SL 5170 core are much
higher than from the other two. Also in contrast to HDPE, ejection forces for HIPS from
the SL 5170 insert are larger than those from the combination SL 5170/P-20 insert. This
is because, as will be seen in the standard friction tests, HIPS and SL 5170 react more
strongly with each other at higher temperatures.
107
Ejection Force HIPS
1600
1400
Ejection Force (N)
1200
1000
800
600
400
200
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
Run
P-20
ST-100
SL 5170
SL/P-20
Figure 5.2: Experimental ejection force values for HIPS, all runs.
108
15
16
17
18
5.1.4 Experimental Ejection Force Results from the P-20 and ST-100 Inserts
From both the P-20 and ST-100 inserts, ejection forces for HDPE were lower than
for HIPS for all experimental runs (Figure 5.3).
5.1.5 Experimental Ejection Force Results from the SL 5170 and SL 5170/P-20 Inserts
Due to the problems with cavity deformation and parts sticking in the core, only
two HDPE runs and three HIPS runs were completed with the SL 5170 core and cavity.
With this insert, ejection forces for HIPS were much higher than for HDPE (Figure 5.4).
The same result can be seen from the SL 5170 insert with the P-20 cavity.
109
Ejection Force P-20
450
Ejection Force (N)
400
350
300
250
200
150
100
50
0
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
Run
HDPE
HIPS
Ejection Force (N)
Ejection Force ST-100
0
1
2
3
4
5
6
7
8
9 10 11 12 13 14 15 16 17 18
Run
HDPE
HIPS
Figure 5.3: Experimental ejection force results from the P-20 and ST-100 inserts.
110
Ejection Force SL 5170
1600
1500
1400
1300
1200
Ejection Force (N)
1100
1000
900
800
700
600
500
400
300
200
100
0
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
Run
HDPE
HIPS
HDPE with SL/P-20
HIPS with SL/P-20
Figure 5.4: Experimental ejection force results from the SL 5170 insert and the
combination SL 5170/P-20 insert (all completed runs are shown).
111
5.2 Statistical Analysis
5.2.1 DOE Results
Analysis of variance results for the designed experiment described in Chapter 4
are shown in Table 5.2. Numbers are given for the slope of the line that defines the effect
(in Newtons per unit) and the correlation coefficient. The slope of the line indicates the
magnitude of the effect. The correlation coefficient applies only for three-level
parameters and indicates how well the line (defined by the slope) fits the data. If the
correlation coefficient is low, then the effect is not linear. A check mark indicates that
there was an interaction between the parameters shown.
5.2.2 Main Effects and Interactions
As shown in Table 5.2, ejection force increased with packing time for the SL
5170 insert, and decreased with packing time for the ST-100 insert. Packing time also
had an effect from the P-20 insert on the HIPS ejection force, but not on the HDPE
ejection force. Ejection force decreased with cooling time for the P-20 insert. Note that,
for HDPE, the correlation coefficient was low, indicating that the effect was not linear.
Packing pressure had an effect on ejection force for HIPS with the SL 5170 and P-20
inserts, but did not for HDPE. There were no interaction effects for the SL 5170 insert.
The effects on ejection force from these parameters differed between the baseline
steel insert and the rapid tooled inserts. Furthermore, there were differences in effects
112
between the two thermoplastics, especially with the P-20 insert. Main effects and
interaction plots for each set of experiments are shown in Figures 5.5 through 5.10.
HDPE
Insert
Material
Main Effects on Ejection Force
Packing
Packing
Cooling Pressure
Time Tp Time Tc
Pp
P-20 Steel
Sintered
ST-100
-3.86
SL 5170/
P-20
8.68
HIPS
Insert
Material
P-20 Steel
-.40
.31
-1.10
1.00
Sintered
ST-100
-2.99
SL 5170/
P-20
20.46
-2.90
.96
Tp-Tc
Tp-Pp
Tc-Pp
Tp-Tc-Pp
a
a
Main Effects on Ejection Force
Packing
Packing
Cooling Pressure
Time Tp Time Tc
Pp
1.82
Interactions
.88
.82
a
Interactions
Tp-Tc
Tp-Pp
Tc-Pp
Tp-Tc-Pp
a
a
a
a
a
a
a
36.21
Table 5.2: Results from the designed experiment indicating which factors had a
significant effect on ejection force.
113
Main Effects Plot - Data Means for EF Set 1
Packing Time
Cooling Time
Packing Pres
EF Set 1
41.5
41.0
40.5
40.0
39.5
2
6
5
10
0
15
5
10
Interaction Plot - Data Means for EF Set 1
2
6
5
10
15
0
5
10
42.0
Packing Time
6
40.5
39.0
2
Cooling Time
42.0
15
40.5
10
39.0
5
Packing Pres
42.0
10
40.5
5
0
39.0
Figure 5.5: Main effects and interactions for HDPE with the P-20 insert. (Ejection
force is shown in pounds.)
114
Main Effects Plot - Data Means for EF Set 2
Packing Time
Cooling Time
Packing Pres
89.0
EF Set 2
87.5
86.0
84.5
83.0
2
6
5
10
0
15
5
10
Interaction Plot - Data Means for EF Set 2
2
6
5
10
15
0
5
10
90
Packing Time
6
85
80
2
90
Cooling Time
15
85
10
80
5
Packing Pres
90
10
85
5
80
0
Figure 5.6: Main effects and interactions for HIPS with the P-20 insert. (Ejection
force is shown in pounds.)
115
Main Effects Plot - Data Means for EF Set 3
Packing Time
Cooling Time
Packing Pres
43.7
EF Set 3
42.9
42.1
41.3
40.5
2
6
5
10
0
15
5
10
Interaction Plot - Data Means for EF Set 3
2
6
5
10
15
0
5
Packing Time
10
45.0
6
42.5
40.0
2
Cooling Time
45.0
15
42.5
10
40.0
5
Packing Pres
45.0
10
5
0
42.5
40.0
Figure 5.7: Main effects and interactions for HDPE with the ST-100 insert.
(Ejection force is shown in pounds.)
116
Main Effects Plot - Data Means for EF Set 4
Packing Time
86.4
Cooling Time
Packing Pres
EF Set 4
85.8
85.2
84.6
84.0
2
6
5
10
15
0
5
10
Interaction Plot - Data Means for EF Set 4
2
6
5
10
15
0
5
10
88
Packing Time
6
84
80
2
88
Cooling Time
15
84
10
80
5
Packing Pres
88
10
84
5
0
Figure 5.8: Main effects and interactions for HIPS with the ST-100 insert.
(Ejection force is shown in pounds.)
117
80
Main Effects Plot - Data Means for EF Set 5b
Packing Time
70.6315
Cooling Time
Packing Pres
EF Set 5b
68.4793
66.3271
64.1748
62.0226
2
6
0
12
0
15
0
5
Interaction Plot - Data Means for EF Set 5b
2
6
0
12
0
15
0
5
Packing Time
70
6
65
60
2
Cooling Time
70
150
65
60
120
Packing Pres
70
5
65
0
60
Figure 5.9: Main effects and interactions for HDPE with the SL 5170/P-20 insert.
(Ejection force is shown in pounds.)
118
Main Effects Plot - Data Means for EF Set 6a
Packing Time
Cooling Time
Packing Pres
195
EF Set 6a
185
175
165
155
2
6
0
12
0
15
0
5
Interaction Plot - Data Means for EF Set 6a
2
6
0
12
0
15
0
5
200
Packing Time
6
175
150
2
200
Cooling Time
150
175
150
120
Packing Pres
200
5
175
0
150
Figure 5.10: Main effects and interactions for HIPS with the SL 5170/P-20 insert.
(Ejection force is shown in pounds.)
119
5.3 Standard Friction Testing Results
Coefficient of friction results from the standard tests are shown in Figures 5.11
and 5.12. In general the data show some expected trends. For example, at room
temperature, the friction coefficient of HIPS is larger than that of HDPE on all three plate
materials. Also, the friction coefficients of both thermoplastics on the SL 5170 plate are
higher than those on the metal plates.
5.3.1 HDPE Standard Friction Results
Temperature did not make a dramatic difference in friction coefficient for HDPE
on P-20, ST-100, or SL 5170. For the P-20 plate, the difference in coefficient between
temperatures was not statistically significant. With the ST-100 and SL 5170 plates, there
was an increase in the friction coefficient in the ejection temperature tests as compared to
the room temperature tests. This is probably because the adhesion component of friction
became more apparent in the heated tests. From ejection temperature to elevated
temperature for the ST-100 plate, the difference in the coefficient of friction was not
statistically significant. For the SL 5170 plate, there was actually a slight decrease in
friction coefficient from ejection temperature to elevated temperature, an unexpected
result.
In all cases, elevating the initial temperature to imprint the surface of the
specimen was presumed to increase the friction coefficient. This was not the case for any
of the HDPE tests. No change in coefficient (in the case of P-20 and ST-100) and the
slight decrease in coefficient (in the case of SL 5170) may be due to shrinkage of the sled
120
specimen from the plate and a reduction in the area of contact due to the imprinted
pattern. HDPE is a crystalline polymer and will shrink more than HIPS, which is an
amorphous polymer. As discussed in Chapter 3, compared to amorphous materials, a
crystalline structure can arrange itself into a tighter, more orderly fashion as the polymer
cools. It is noted, however, that the imprinted pattern was not very pronounced on the
HDPE specimens. Therefore, the temperature to which the specimens were heated was
probably not high enough to sufficiently soften the polymer.
The relationship of HDPE friction coefficients among the three plate materials
was as expected. The coefficients for HDPE on SL 5170 were highest because of the
nature of polymer on polymer materials and because the surface roughness of this plate
was higher than that of the other two. The coefficients of HDPE on ST-100 were lowest
because the surface roughness of this plate was lower than that of the other two. The
conductivity of the ST-100 plate is also highest, which may have contributed to a
reduction in adhesion by dissipating heat at a faster rate in the heated tests.
121
Plate Material
Surface
Roughness
P-20 Steel
0.7 microns
LaserForm
ST-100
SL5170
Resin
0.2 microns
3.6 microns
HDPE Static Friction Coefficient
Room Temp
Ejection Temp Elevated Temp
+0.08
+0.04
+0.06
0.26 -0.07 0.31 -0.02 0.28 -0.02
+0.02
+0.04
+0.01
0.21 -0.03 0.26
0.25
-0.05
-0.01
+0.04
+0.02
+0.07
0.37 -0.07 0.45 -0.05 0.38 -0.02
Static CoF of HDPE
0.60
0.50
0.40
0.30
0.20
0.10
0.00
Room Temp
On P-20 Steel
Ej Temp
On ST-100
Elev Temp
On SL5170
Static CoF of HDPE
0.6
0.5
0.4
Room Temp
0.3
Ej Temp
0.2
Elev Temp
0.1
0
On P-20
Steel
On ST-100
On
SL5170
Figure 5.11: Standard friction test results for HDPE; means and ranges shown in
the table.
122
Plate Material
Surface
Roughness
P-20 Steel
0.7 microns
LaserForm
ST-100
SL5170
Resin
0.2 microns
3.6 microns
HIPS Static Friction Coefficient
Room Temp
Ejection Temp Elevated Temp
+0.04
+0.04
+0.11
0.36 -0.04 0.32 -0.04 0.35 -0.21
+0.08
+0.03
+0.17
0.32
0.13
0.54 -0.29
-0.05
-0.02
+0.05
+0.11
+2.65
0.43 -0.03 0.56 -0.13 5.47 -2.08
Static CoF of HIPS
5.47
SL 5170 elev
temp
0.6
0.5
0.4
0.3
0.2
0.1
0
Room Temp
On P-20 Steel
Ej Temp
On ST-100
Elev Temp
On SL5170
Static CoF of HIPS
5.47
0.6
0.5
0.4
Room Temp
0.3
Ej Temp
0.2
Elev Temp
0.1
0
On P-20
Steel
On ST-100
On
SL5170
Figure 5.12: Standard friction test results for HIPS; means and ranges shown in the
table.
123
5.3.2 HIPS Standard Friction Results
The HIPS friction results were much more diverse than the HDPE results and will
be discussed by individual plate material.
HIPS on P-20 Steel – Temperature had only a slight effect in this case. While one
would expect the friction coefficient to increase with temperature, the coefficient actually
decreased in the ejection temperature test compared to the room temperature test. This
may be explained by a softening of the polymer to allow asperities in the two surfaces to
slide over each other more easily. Adhesion was probably not dominant in this case. The
elevated temperature test resulted in a small increase in average friction, but this
difference is not statistically significant.
HIPS on ST-100 – The results from these tests show the same trend by
temperature as the P-20 results, except that it is much more pronounced. As in the case
of the P-20 Steel plate, the decrease in coefficient from room temperature to ejection
temperature may be explained by a softening of the polymer to allow asperities in the two
surfaces to slide over each other more easily. The marked increase in coefficient at the
elevated temperature may be caused by the increase in area of contact due to the
imprinted pattern, from an increase in adhesion, or both. Note also that the surface
roughness of the P-20 is larger than that of the ST-100. At room temperature and
ejection temperature, this may have contributed to higher friction on the P-20 plate, while
at elevated temperature it may have caused more adhesion on the ST-100 plate.
HIPS on SL 5170 – Adhesion had a very prominent effect in the heated tests on
the SL 5170 plate. At ejection temperature, the friction coefficient is significantly higher
124
than at room temperature. Whereas the softened polymer slid over the metal plates with
less resistance, it adhered to the resin plate, causing a larger peak frictional force. The
elevated temperature test shows an extreme case of this adhesion, due to a strong
interaction between the two materials. Secondary forces such as dispersion forces,
polarity, or hydrogen bonding, contribute to this interaction. Adhesion between two
materials, for example, will be maximized when their polarities are similar. While the
imprinted pattern of the SL 5170 plate into the softened HIPS material may also have
contributed to the spike in frictional force, the mechanical interaction was not a dominant
influence.
Figure 5.13 shows a sample plot from the HIPS on SL 5170 test. The graph
shows an initial large peak followed by a lower peak as the specimen begins to slide. The
initial peak may be explained by the strong force required to overcome adhesion,
followed by a lesser force required to overcome roughness once the specimen is
“unstuck.” The HIPS on SL 5170 at elevated temperature was the only test to show this
phenomenon. The secondary molecular forces that caused HIPS and SL 5170 to adhere
to each other were not a prominent factor with other material pairs. Polarity has no effect
between HIPS and the two metal plates, because the metal plates become oxidized, and
there is an insulating layer between the polymer specimen and the plate. Polarity has no
effect between HDPE and any of the plates because HDPE is a nonpolar material. Other
test results looked more like the sample HIPS on P-20 graph shown in Figure 5.14.
125
4.000
3.500
Friction Force (lbf)
3.000
2.500
2.000
1.500
1.000
0.500
0.000
Time
12
24
36
48
60
72
84
Time (sec)
Figure 5.13: Sample plot of load vs. time for HIPS on SL 5170 from elevated
temperature tests.
126
0.250
Friction Force (lbf)
0.200
0.150
0.100
0.050
0.000
Time
12
24
36
48
60
72
84
Time (sec)
Figure 5.14: Sample plot of load vs. time for HIPS on P-20 from elevated
temperature tests.
The relationship of HIPS friction coefficients among the three plate materials was
as expected, except for the elevated temperature test on ST-100 compared to P-20 Steel.
One would anticipate that the coefficient of friction on ST-100 would be lower because
its surface roughness is lower than that of the P-20. Instead, however, the smoother
finish may have allowed the adhesion component of friction to dominate in this case. In
the room temperature and ejection temperature cases, the friction coefficient of HIPS on
ST-100 was lowest, and in all cases the friction coefficient of HIPS on SL 5170 was
highest.
127
5.4 Reliability of the Data
In conjunction with the experimental and test discussions above, a few parameter
discrepancies and data variations must be noted. First, there are differences in injection
molding process parameters among experimental sets. These are necessary because of
the nature of the materials used. For example, HDPE and HIPS have different processing
temperatures and injection velocities. Also, the temperatures and pressures used with the
SL 5170 resin mold insert are as low as the processing window will allow so that minimal
deformation or degradation occurs. A complete list of process parameters is included in
Chapter 4.
Second, there may be some variation in the data due to the following:
•
HDPE parts were flared slightly at the rim in ejections from the P-20 and ST-100
cores. This flaring occurred due to the magnitude of the ejection force applied by
the stripper plate and the softness of the warm thermoplastic material. This
flaring may have increased canister diameter measurements slightly, and can
affect ejection force calculations.
•
The SL 5170 core was susceptible to swelling. Experiments with the SL 5170
core and the P-20 cavity were run with significantly lower ejection temperature
because, if the heat was increased to raise ejection temperature, then core swelling
was excessive. Core swelling may have affected diameter measurements as well.
128
•
The shape of the SL 5170 cavity was modified to include a taper to facilitate
ejection. Those few experimental parts from the SL 5170 cavity required
adjustments to the ejection force equation based on this geometry. Shrinkage and
ejection forces may have been affected as well.
•
Surface roughnesses vary among the three insert cores. The plates used in the
friction tests were intended to have the same surface roughness as their
corresponding injection molding core, but this was not accomplished in all cases
(Table 5.3). Comparisons of friction coefficients among materials and between
friction test data and experimental injection molding data must take this into
account.
•
The length of the part was not measured at the time of ejection. This introduces a
small amount of error, especially for HDPE, in the calculation of ejection force
(see next section) because the lateral shrinkage is not taken into account.
Surface Roughness, Ra (microns)
P-20 Plate
0.7
P-20 Core
0.7
ST-100 Plate
0.2
ST-100 Core
0.3
SL 5170 Plate SL 5170 Core
3.6
0.7
Table 5.3: Surface roughnesses of all plates (friction tests) and cores (injection
molding experiments).
129
5.5 Calculation of Ejection Force Using the Model
Using the values for coefficient of static friction from the standard tests at
elevated temperature, ejection forces have been calculated using the Menges model
derived in Chapter 3. These values, along with the experimental values for ejection
force, are shown in Tables 5.4 and 5.5. The difference between calculated values and
actual experimental values for ejection force is significant, and excessive in some of the
HIPS cases.
130
HDPE
P-20
ST-100
Packing Cooling Packing
Experiment Calculation Experiment Calculation
Time
Time Pressure
s
s
%
N
N
N
N
2
5
0
177.15
84
182.30
33
2
5
5
183.29
78
190.02
39
2
5
10
186.52
82
209.79
46
2
10
0
176.51
92
177.44
36
2
10
5
185.89
84
196.61
43
2
10
10
172.87
77
194.33
41
2
15
0
191.90
114
196.15
82
2
15
5
174.68
80
201.68
46
2
15
10
173.08
86
208.75
45
6
5
0
184.69
89
173.95
59
6
5
5
193.58
82
185.81
71
6
5
10
173.88
67
184.41
64
6
10
0
171.28
79
172.12
59
6
10
5
174.63
72
178.83
68
6
10
10
175.19
63
180.87
64
6
15
0
170.00
82
170.97
67
6
15
5
180.05
86
186.82
74
6
15
10
185.53
76
184.13
78
SL 5170
SL Core with P-20 Cavity
Packing Cooling Packing
Experiment Calculation Experiment Calculation
Time
Time Pressure
s
s
%
N
N
N
N
2
150
0
239.06
73
2
180
0
193.21
83
2
2
2
2
6
6
6
6
120
120
150
150
120
120
150
150
0
5
0
5
0
5
0
5
274.21
299.65
258.76
278.18
313.33
297.38
321.09
317.93
286
251
355
248
261
155
255
163
Table 5.4: Calculated values of ejection force for HDPE from the Menges equation
and experimental data.
131
HIPS
P-20
ST-100
Packing Cooling Packing
Experiment Calculation Experiment Calculation
Time
Time Pressure
s
s
%
N
N
N
N
2
5
0
343.97
218
366.29
700
2
5
5
376.37
235
389.51
639
2
5
10
401.48
205
375.28
603
2
10
0
346.11
237
375.71
753
2
10
5
385.56
258
393.54
698
2
10
10
408.26
210
394.47
612
2
15
0
384.60
284
366.31
767
2
15
5
381.94
220
393.56
592
2
15
10
403.70
273
398.83
655
6
5
0
376.97
234
363.86
353
6
5
5
395.23
240
378.14
330
6
5
10
393.38
73
388.07
330
6
10
0
369.19
205
369.54
541
6
10
5
390.81
168
360.58
438
6
10
10
391.73
131
374.38
343
6
15
0
351.62
182
340.67
541
6
15
5
394.77
228
370.57
555
6
15
10
424.46
264
399.88
560
SL Core with P-20 Cavity
SL 5170
Packing Cooling Packing
Experiment Calculation Experiment Calculation
Time
Time Pressure
s
s
%
N
N
N
N
2
120
0
1334.27
2268
2
150
0
1136.12
4094
2
150
5
1512.25
1495
2
2
2
2
6
6
6
6
120
120
150
150
120
120
150
150
0
5
0
5
0
5
0
5
695.76
826.28
610.12
845.02
770.24
939.13
702.28
892.15
13059
6277
19679
6825
5405
2812
8383
1905
Table 5.5: Calculated values of ejection force for HIPS from the Menges equation
and experimental data.
132
5.5.1 Calculated Ejection Force for HDPE
Figure 5.15 shows that the calculated values for ejection force for HDPE from the
Menges model are lower than the measured values by 50 to 70 percent on average, except
for those from the SL 5170 core with the P-20 cavity. Based on results from the standard
friction test at elevated temperature, the calculated values for ejection force were
expected to be lower than the actual values because, as previously mentioned, the
elevated temperature was probably not high enough to measure an accurate value of the
coefficient of static friction for HDPE. Furthermore, the standard test environment was
not identical to the injection molding environment in terms of temperatures and pressures
and their respective histories.
Calculated ejection force values for HDPE from the SL 5170 core with the P-20
cavity were closer to actuals, i.e., within 16 percent on average. This is because the
ejection temperatures during these experiments were lower than the others, so the friction
coefficient from the standard test is more comparable.
133
Ejection Force HDPE
400
350
300
N
250
200
150
100
50
0
P-20
SL/P-20
ST-100
Experimental
SL 5170
Calculated
Figure 5.15: Calculated values for ejection force for HPDE using the Menges model
compared with experimental values, averaged across all runs.
5.5.2 Calculated Ejection Force for HIPS
Calculated values for ejection force, averaged across all runs, for HIPS parts from
the P-20 and ST-100 cores are shown with experimental values in Figure 5.16. Those
from the SL 5170 insert, and the SL 5170 core with the P-20 cavity are shown in Figure
5.17. The model estimated ejection force for HIPS on P-20 to be 44 percent lower than
134
actuals, and on ST-100 to be 47 percent higher on average. For the SL 5170 core,
calculated values were higher on average than actual values by 97 percent. (Note that
only three runs of data were collected for the SL 5170 insert.) For the SL 5170 core with
the P-20 cavity, calculated values were extremely high (924 percent higher than actuals).
As was shown for HDPE, differences between calculated and actual ejection force
values for HIPS are due in part to friction coefficients, which vary between standard
measurements and actual ejection. Assuming this to be a primary reason for the
differences in calculated and actual ejection forces, the measured friction coefficient for
HIPS with P-20 was low, and the measured friction coefficients for HIPS with ST-100
and HIPS with SL 5170 were high. In any case, the surface interactions of each material
pair were different during the standard tests as compared to during the injection molding
experiments. Therefore the friction coefficient measured in the standard test caused some
error in the calculation of ejection force when applied to the injection molding case.
5.5.3 Possible Sources of Error
One source of error in the calculation of ejection force using the model developed
by Menges is the coefficient of friction measurement. As mentioned above, the
environment of the standard friction test is not identical to the environment of the
injection molding experiment. Therefore, it is likely that the standard measurement of
the coefficient of friction, even at elevated temperature, is not accurate.
Another possible source of error in the ejection force calculation is the elastic
modulus measurement. Again, the environment of the standard modulus measurement
135
does not exactly simulate the injection molding environment, so there is likely to be some
error. Furthermore, the elastic modulus of the HIPS material is more sensitive to
temperature as compared with HDPE. A small difference between the measured and
actual temperatures in the modulus measurements described in Chapter 4 or in the
injection molding experiments would cause a significant change in elastic modulus and a
resultant change in the calculation of ejection force.
A third source of error in the ejection force calculation is the measurement of the
relative change in diameter of the part. The digital imaging approach used to measure the
inside and outside diameters of the part was probably the most accurate method short of a
laser-based, real time, and much more expensive system. However, digital imaging
required manual transfer of the part from the injection molding machine to the camera
fixture and most likely introduced some error by expanding the time between ejection
and capture of the digital data.
136
Ejection Force HIPS
800
700
600
N
500
Experimental
400
Calculated
300
200
100
0
P-20
ST-100
Figure 5.16: Calculated values for ejection force for HIPS parts from the P-20 and
ST-100 cores compared with experimental values, averaged across all runs.
Ejection Force HIPS
10000
8000
6000
N
Experimental
Calculated
4000
2000
0
SL/P-20
SL 5170
Figure 5.17: Calculated values for ejection force for HIPS parts from the SL 5170
core compared with experimental values, averaged across all runs.
137
5.6 Calculation of Apparent Friction Coefficients using the Menges Model
By rearranging the Menges model equation, and using experimental ejection force
and shrinkage data, the apparent coefficient of static friction was calculated for each
material combination. Table 5.7 lists the calculated values by experimental parameters.
138
Packing
Time
s
2
2
2
2
2
2
2
2
2
6
6
6
6
6
6
6
6
6
Cooling
Time
s
5
5
5
10
10
10
15
15
15
5
5
5
10
10
10
15
15
15
Packing
Pressure
%
0
5
10
0
5
10
0
5
10
0
5
10
0
5
10
0
5
10
Packing
Time
s
2
2
2
2
Cooling
Time
s
120
150
180
150
Packing
Pressure
%
0
0
0
5
2
2
2
2
6
6
6
6
120
120
150
150
120
120
150
150
0
5
0
5
0
5
0
5
Calculated Friction Coefficient
HDPE
HIPS
P-20
LaserForm
P-20
LaserForm
Steel
ST-100
Steel
ST-100
0.59
0.65
0.64
0.54
0.62
0.63
0.47
0.61
0.56
0.58
0.66
0.73
0.61
0.68
0.78
0.58
0.59
0.68
1.40
1.21
1.15
1.24
1.14
1.20
0.60
1.10
1.16
0.73
0.66
0.73
0.73
0.66
0.71
0.64
0.63
0.59
HDPE
SL 5170
SL 5170
w/P-20
N
N
1.24
0.88
0.55
0.56
0.68
0.51
0.52
0.68
0.47
0.61
0.52
0.56
0.58
1.88
0.63
0.81
1.05
0.68
0.61
0.56
0.28
0.33
0.34
0.27
0.30
0.35
0.26
0.36
0.33
0.56
0.62
0.63
0.37
0.44
0.59
0.34
0.36
0.39
HIPS
SL 5170
SL 5170
w/P-20
N
N
3.22
1.52
5.53
0.36
0.45
0.28
0.43
0.46
0.73
0.48
0.74
0.29
0.72
0.17
0.68
0.78
1.83
0.46
2.56
Table 5.6: Calculated apparent coefficient of friction results according to packing
time, cooling time, and packing pressure parameters.
139
5.6.1 HDPE Apparent Coefficient of Friction Results
Figure 5.18 shows calculated values of the static friction coefficient by run for
HDPE. The average calculated value of friction coefficient for HDPE from the P-20 core
is lower than that from the ST-100 core, which is lower than that from the SL 5170 core.
The calculated value of friction coefficient from the SL 5170 core with the P-20 cavity is
lower than the other calculated values due to the lower ejection temperature.
Coefficient of Static Friction HDPE
1.6
1.4
1.2
CoF
1
0.8
0.6
0.4
0.2
0
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
Run
P-20
ST-100
SL 5170
SL/P-20
Figure 5.18: Calculated values of the apparent coefficient of static friction for
HDPE, all runs.
140
5.6.2 HIPS Apparent Coefficient of Friction Results
Figure 5.19 shows calculated values of the static coefficient of friction for HIPS
from the experimental data. Results from all four inserts are shown. The HIPS
coefficient of friction on P-20 is generally higher than on ST-100. Two of the friction
values from the P-20 insert are much higher than the others, i.e., 1.88 and 1.05. These
values correspond to high packing time, high packing pressure, and lower cooling time,
and therefore imply that higher pressure and temperature cause higher friction between
the HIPS material and the steel. This phenomenon is also seen with the SL 5170/P-20
insert, but not with ST-100. Friction values for HIPS with the SL 5170 core and cavity
are very high, as expected. In general, the coefficient of friction values on SL 5170 core
with P-20 cavity are low because ejection temperatures were much lower.
141
CoF
Coefficient of Static Friction HIPS
6.0
5.8
5.6
5.4
5.2
5.0
4.8
4.6
4.4
4.2
4.0
3.8
3.6
3.4
3.2
3.0
2.8
2.6
2.4
2.2
2.0
1.8
1.6
1.4
1.2
1.0
0.8
0.6
0.4
0.2
0.0
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
Run
P-20
ST-100
SL 5170
SL/P-20
Figure 5.19: Apparent coefficients of friction calculated from experimental results
for HIPS.
142
5.6.3 Apparent Friction Coefficient Results from the P-20 and ST-100 Inserts
From the P-20 insert, the friction coefficient was lower for HDPE than for HIPS
for 10 out of 18 runs, most of which correspond to the higher level of packing time
(Figure 5.20). From the ST-100 insert, the friction coefficient for HDPE was higher than
for HIPS for all runs (Figure 5.21). The HDPE coefficients corresponding to low packing
time were higher than those corresponding to high packing time, whereas the HIPS
coefficients showed the opposite relation to a lesser degree.
Coefficient of Static Friction P-20
2
CoF
1.8
1.6
1.4
1.2
1
0.8
0.6
0.4
0.2
0
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
Run
HDPE
HIPS
Figure 5.20: Apparent coefficient of static friction for parts from the P-20 insert.
143
Coefficient of Static Friction ST-100
1.6
1.4
1.2
CoF
1
0.8
0.6
0.4
0.2
0
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
Run
HDPE
HIPS
Figure 5.21: Apparent coefficient of static friction for parts from the ST-100 insert.
5.6.4 Apparent Friction Coefficient Results from the SL 5170 and SL 5170/P-20 Inserts
With SL 5170 core and cavity, calculated values for coefficient of friction for
HIPS are much higher than for HDPE (Figure 5.22). From the SL 5170 insert with the P20 cavity, HIPS friction coefficients were higher than HDPE for the runs at higher
packing pressure, and lower for the runs with zero packing pressure (Figure 5.23). Note,
once again, that the coefficients of friction on SL 5170 core with P-20 cavity were low
because ejection temperatures were much lower.
144
Coefficient of Static Friction SL 5170
6.0
5.0
CoF
4.0
3.0
2.0
1.0
0.0
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
Run
HDPE
HIPS
Figure 5.22: Apparent coefficient of static friction for parts from the SL 5170
insert.
145
18
Coefficient of Static Friction SL/P-20
3
2.5
CoF
2
1.5
1
0.5
0
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
Run
HDPE
HIPS
Figure 5.23: Apparent coefficient of static friction for parts from the SL 5170 core
with the P-20 cavity.
5.6.5 Comparing Calculated Friction Results to Standard Friction Test Results
Figures 5.24 and 5.25 show the average calculated friction coefficient results for
HDPE and HIPS, respectively, along with those measured in the standard tests. The
standard test values for HDPE are all lower than the calculated values because the
temperatures and pressures in the standard tests did not match those in the injection
molding experiments, and because of the sources of error discussed in Section 5.5.3. The
values shown for the SL 5170 core with the P-20 cavity are reasonably close because the
146
ejection temperature during this injection molding experiment was much lower compared
with experiments using the other inserts. Calculated friction coefficients for HIPS on P20 were higher than standard test results. Those for HIPS on ST-100, however, were not
all higher than standard test results at elevated temperature.
Recall that only two runs of HDPE parts and three runs of HIPS parts were
completed with the SL 5170 insert. These average and range calculations, then, are based
on only a small amount of data. Note for HIPS that the calculated friction coefficient
value corresponding to higher packing pressure (5.53) is comparable to the friction
coefficient obtained in the standard test (5.47).
147
Static Friction Coefficient for HDPE
1.6
1.4
1.2
1.0
P-20
ST-100
0.8
SL 5170
SL/P20
0.6
0.4
0.2
0.0
STD RT
STD Elev
Calculated
Figure 5.24: Average apparent coefficient of static friction for HDPE compared to
standard test results at room temperature and elevated temperature.
148
Static Friction Coefficient for HIPS
9.0
8.0
7.0
6.0
5.0
ST-100
SL 5170
P-20
4.0
SL/P20
3.0
2.0
1.0
0.0
STD RT
STD Elev
Calculated
Figure 5.25: Average apparent coefficient of static friction for HIPS compared to
standard test results at room temperature and elevated temperature.
149
5.7 Other Observations of Rapid Tooled Inserts
In addition to the quantitative analyses, a number of qualitative observations were
made during experiments with the ST-100 and SL 5170 tools. These are summarized in
the paragraphs below.
The ST-100 insert seemed to operate as well as the P-20 insert. There were no
problems with core swelling or parts adhering to the sintered material. The insert held up
well for the number of parts that were processed, and had every indication that it would
last for many more. There were, however, differences in data between the ST-100 and P20 experiments, indicating that there are thermal and adhesion differences that affect
ejection force and friction coefficient.
The SL 5170 insert, on the other hand, operated quite differently from the other
two metal inserts, as one would expect. Surprisingly, the resin core held up for more than
105 full parts, with minimal flashing and no catastrophic failure. The following are some
of the anomalies of the SL 5170 insert discovered during processing.
It was obvious during molding of both HDPE and HIPS that there was adhesion
of the parts to the SL 5170 core (no matter which cavity was used). The adhesion was
both visible and audible. After the mold opened and before the part was ejected, the
portion of the part that adhered was more translucent than the rest. At ejection, the part
would audibly snap upon release, then slide off the core. It is quite possible that the force
from the stripper plate, in conjunction with the adhesion, forced the part to bow outward
somewhat, thus reducing the amount of shear along the core surface.
150
Core swelling was a problem with the SL 5170 material. Control of the core
temperature became especially important when using this insert. During the first
experimental set, the core diameter was reduced by 0.13 mm (0.005 in) to avoid
interference between the swelled core and the stripper plate. During experiments with the
SL 5170 core and the P-20 cavity, the planned temperature at ejection could not be
maintained. This was because the steel cavity would conduct much of the heat away
from the insert, and allowing the temperature to build would cause excessive swelling of
the resin core. The reason for the core swelling is not entirely clear. There may have
been some resin material that was not completely cured. Or swelling may be
characteristic of this material at certain temperature, humidity, and/or pressure
conditions.
After the SL 5170 core processed about 40 parts, it began to show some internal
defects. The defects were barely visible at first, but gradually increased in number and
size as more and more parts were processed. Shining a light through the end of the core
highlighted the defects, and one could see that they looked like internal delaminations
along the layers of the tool (Figure 5.26). The core, however, lasted through the entire
experiment without these defects propagating into failure.
151
Figure 5.26: Defects in the SL 5170 core.
The issue with the SL 5170 core that most affected experimental results was
deflection of the cavity wall during injection. Only a few experimental runs were
completed with the SL 5170 core and cavity when the parts began to stick in the cavity
upon mold opening. The wall of the cavity apparently gave under injection pressure and
elastically deformed. Then, although the part began to shrink, too much material had
been forced into the mold, and was held fast by the cavity wall.
In a preliminary study of this phenomenon, the SL 5170 insert was modeled with
the two thermoplastic materials using ABAQUS finite element analysis (Carpenter 2004).
Using the thermoplastic material characteristics from the software database and the
geometry of the SL 5170 core and cavity, the injection was simulated for HDPE and
HIPS with no packing, and for HIPS with 6 seconds at 5 percent packing. Results show
that the walls of the cavity do indeed elastically deform in a way that would cause parts
to stick. In Figure 5.27, deformation of the SL 5170 by HDPE injection, no packing, is
152
concentrated near the base of the cavity, with a maximum magnitude of 0.06 millimeters
(0.002 inches). In Figure 5.28, similar deformation results with HIPS, although the
maximum magnitude is slightly less at 0.05 millimeters (0.002 inches). When packing is
included, Figure 5.29, the magnitude of the deformation is greater at 0.07 millimeters
(0.003 inches), and it encompasses a greater area along the cavity wall.
Figure 5.27: Simulation results of HDPE injection into SL 5170 insert, no packing.
153
Figure 5.28: Simulation results of HIPS injection into SL 5170 insert, no packing.
154
Figure 5.29: Simulation results of HIPS injection into SL 5170 insert, with packing.
155
CHAPTER 6
CONCLUSIONS
This chapter presents the conclusions of this research project. First, conclusions
regarding the molding of HDPE and HIPS parts using the rapid tooled inserts are
presented, including the benefits and limitations of rapid tooled injection mold inserts.
Next, the use of a model for determining ejection force and the coefficient of friction is
discussed. Implications and future work are also included, and the chapter concludes
with an overall summary.
6.1 Molding HDPE and HIPS with ST-100 and SL 5170 Inserts
6.1.1 Benefits and Limitations of Using Rapid Tooled Injection Mold Inserts
The general benefits of rapid tools were discussed in Chapter 1. These include
the ability to build complex geometries and incorporate conformal cooling lines, and
potential savings in lead times and material and labor costs. As an injection molding
insert, the ST-100 tool was very capable. Few process problems were experienced in this
156
work. This particular insert withstood the processing of 288 parts plus many test parts
with no signs of wear or damage. For these quantities, the ST-100 performed just as well
as the P-20, and even showed some advantages for processing HIPS in terms of friction
coefficients. For the scope of this work, no limitations of the ST-100 insert are noted.
The SL 5170 tool shares the same general benefits as the ST-100 and other rapid
tools as mentioned above. Some unexpected benefits, as seen in this work, include its
ability to mold both HDPE and HIPS, and durability of the core for processing more than
105 parts. The use of SL 5170 for injection molds, however, is limited because of the
deformation that occurs with high pressure and swelling of the material at high
temperature. Although the SL 5170 core did not fail catastrophically during these
experiments, it is assumed that the fatigue life of the core is limited because of the defects
that developed after approximately 40 parts. The interaction of the surface of SL 5170
with those of some thermoplastic materials, such as HIPS, is also a drawback. Adhesion
often occurs between the two surfaces, which can accelerate failure of the core and
potentially affect the quality of the part.
The statistical analysis shows the effects of processing parameters on ejection
forces for all three inserts. Packing time, cooling time, and packing pressure affect
ejection force differently between the baseline steel and the rapid tooled inserts. Effects
are also different between thermoplastic materials. Conclusions from the statistical
analysis that relate directly to the insert material are as follows:
•
Ejection force increases with packing time for the SL 5170 insert.
157
•
Ejection force decreases with packing time for the ST-100 insert.
•
Ejection force decreases with cooling time for the P-20 insert; but this is a nonlinear effect.
6.1.2 Friction and Ejection Force Considerations
In the standard tests, the friction coefficients of HDPE were similar on P-20, ST100, and SL 5170 at all temperature conditions. HIPS showed a different friction
response than HDPE, and its friction coefficients varied significantly between plate
materials in heated tests. Both polymers showed highest coefficients on SL 5170 at all
three temperature conditions. The HIPS test on SL 5170 showed the interplay of the
adhesion and deformation components of friction and how this affects the friction
coefficient.
Although the standard friction tests at elevated temperatures may have given a
more accurate estimate of the friction coefficient during ejection than those at room
temperature, they still did not exactly simulate the actual process. Additional adjustments
might be made to the temperatures and normal forces applied in the standard tests to
render the results more similar to actual molding conditions.
In the injection molding experiments, ejection forces for parts from the ST-100
core were generally similar to those from the P-20 baseline core (170 – 200 N for HDPE
and 340 – 430 N for HIPS). HDPE parts from the SL 5170 core had slightly higher
ejection forces (190 to 240 N), and those from the SL 5170 core with the P-20 cavity
were higher still (250 – 330 N). Conversely, HIPS parts had higher ejection forces from
158
the SL 5170 core with the P-20 cavity (600 – 950 N), and much higher ejection forces
from the SL 5170 core (1100 – 1600 N). This seems to indicate that, when ejecting
HPDE parts from the SL 5170 core, a lower ejection temperature will increase shrinkage
and increase ejection force, and, when ejecting HIPS parts from the SL 5170 core, a
higher ejection temperature will increase adhesion and increase ejection force.
Given the discussion above and all the results of this research, the following
conclusions are drawn:
§
ST-100 inserts can be used to mold HDPE parts. This insert material performed
similarly to P-20, but was affected differently by process parameters.
Calculations of apparent coefficient of static friction indicated that friction can be
high when packing time is low, but these values did not cause extremely large
ejection forces.
§
ST-100 inserts can be used to mold HIPS parts. Once again, ST-100 performed
similarly to P-20, and in some cases had lower ejection forces.
§
SL 5170 can be used to mold HDPE parts, but with adjusted process parameters
or alternative cavity materials to minimize cavity deformation. Ejection
temperatures should be relatively high to minimize the load on the core.
Minimizing this load may extend core life prior to defect formation.
159
§
SL 5170 is not recommended for molding HIPS due to adhesion and very high
ejection forces. The coefficient of friction will increase with higher ejection
temperatures and packing times due to adhesion, which is enhanced by the
secondary forces between the two materials. Maintaining a lower ejection
temperature (e.g., by using a P-20 core) reduces the ejection force somewhat.
Core life, however, will probably be minimal.
6.2 Using a Model to Determine Ejection Force and the Coefficient of Friction
In this work the Menges model was used to determine ejection force for
comparison to experimental measurements. This model requires values for the
coefficient of static friction between the part and the core, the elastic modulus of the part
material at the time of ejection, and the relative change in diameter of the part
immediately after ejection. Each of these values is difficult to obtain and introduces error
into the calculation. The differences between the calculated ejection force and the actual
ejection force varied from 16 percent to 70 percent for HDPE, and from 44 percent to 924
percent for HIPS. While some of the calculations provided good ballpark estimates,
others did not, and only one was within 20 percent.
The static friction coefficient measurement may have been the largest contributor
to the lack of accuracy in the ejection force calculations. Standard friction tests provided
values for static friction coefficients that were an improvement over room temperature
160
values, but were still not an exact simulation of the injection molding experiment. More
standard testing at a wider range of temperature and pressure environments would be
required to determine more accurate values for the static friction coefficient that occur
during ejection of an injection molded part.
The Menges model was also used to determine the apparent coefficient of friction
for comparison to results from the standard friction tests. The specific description of
friction coefficient is still largely a mystery. The term includes deformation and adhesion
in unknown proportions and affected by certain conditions to unknown extents. The
calculation of apparent coefficient of friction from ejection force models, as was done
here, results in a value that encompasses a complete surface interaction under the given
processing conditions. How this value compares with standard measures of friction
coefficient has not been entirely clear. Furthermore, the calculated value of the apparent
coefficient of friction includes error from the measurements of elastic modulus and the
relative change in diameter, and so is not an apples-to-apples comparison with the
standard friction test values.
The apparent friction coefficient calculation, however, can be potentially useful in
testing a new material for an injection mold insert application. For example, to estimate
the ejection forces that will occur, a simple cylindrical mold insert can be built with the
new material, and the apparent coefficient of friction can be calculated using the Menges
model and injection molding data from the cylindrical mold. This calculated value can
then be applied to the model to estimate ejection force for molds having different
geometries but the same materials and similar processing parameters.
161
Results from this work, however, are insufficient to validate this application of the
model. Additional experimentation with a different part geometry would be required to
provide comparison data and prove this concept.
6.3 Implications and Future Work
This work has included friction testing of thermoplastics against rapid prototyped
materials, following a standard procedure and including higher temperatures; direct
measurement of ejection force from steel and rapid tooled injection mold inserts;
calculation of ejection forces using a model developed by Menges; and determination of
the apparent coefficient of static friction from experimental data using the same Menges
model.
A good indication of the processing capability of a mold material are its ejection
force requirements. The experimental results provide these data for all material
combinations. The results were compared among thermoplastic and mold insert
materials, and then compared to calculated ejection force values. These results give an
indication of the usefulness of the ejection force model.
The static friction coefficient results from the standard tests were also compared
among materials, and then compared with calculations of apparent coefficients of
friction. The friction test data are a useful reference for understanding the basic friction
conditions between the thermoplastics, HDPE and HIPS, and the mold insert materials,
162
P-20, ST-100 and SL 5170. Friction results from the standard tests have also pointed out
the adhesion phenomenon that occurs between HIPS and SL 5170. The adhesion results
from the molecular forces between the two materials and is enhanced by higher ejection
temperatures.
The statistical results are useful for determining those process parameters that can
be adjusted to optimize ejection force for the material pairs studied. Analysis of variance
has shown which parameters affect ejection force and how strong those effects are for the
given process window. This information can be used, once a decision has been made to
use one of these inserts, to design a process that meets ejection force requirements.
The experiments have shown the molding of HDPE and HIPS parts with P-20 and
ST-100 mold inserts as rather routine. The more interesting results come from the use of
the SL 5170 insert. While HDPE parts can be molded with SL 5170 inserts, higher
friction coefficients and ejection forces will result. HIPS parts were molded with the SL
5170 insert as well, but with extremely high friction coefficients and ejection forces.
These forces caused high cyclic loads on the SL 5170 core, leading to the formation of
internal defects. Additionally, deformation of the cavity occurred during injection.
These considerations must be taken into account in the application of this rapid tooled
material for injection molds.
Overall, the data are useful for choosing mold insert materials, for deciding
whether or not to use rapid tooled inserts for small quantity production, for development
of rapid prototyping materials and processes, or for injection molding part or machine
design.
163
As mentioned in the previous section, additional work in static friction coefficient
measurement and ejection force comparison with a different part geometry would
enhance the current work. Friction testing under various temperature and pressure
conditions may provide more accurate values of the friction that exists during ejection.
These values could then be used with the ejection force model. Further injection molding
experiments with a different part shape would provide ejection force data to compare to
those from the cylindrical part. This comparison would indicate whether or not the
Menges model could be used to determine ejection forces for new mold insert materials.
Other possible areas for future work include further study on the deformation and
swelling of the SL 5170 material, failure testing and analysis of the SL 5170 core,
optimization of injection molding process parameters, and materials characterization for
adhesion. The deformation and swelling of the SL 5170 resin can be investigated to
determine their actual causes. If curing of the material is an issue, improvements in the
stereolithography process or its post-cure may have an effect. The selection of alternative
materials may also be a solution, including more recently developed materials and resins
that contain fillers. The defects that developed in the SL 5170 core can be analyzed to
confirm whether or not they are in fact delaminations and to determine why they
occurred. Improvements in the building or curing of the stereolithography tool might be
required, or it may be characteristic of the process or material. Further testing to failure
would provide useful data on the actual life expectancy of the core.
Any of the designed experiments could be expanded to encompass a broader
processing window, leading to the optimization of process parameters. This might be
164
especially useful for the SL 5170 core and the HDPE thermoplastic. Further study of the
temperature and pressure parameters would further delineate the effects of the adhesion
component of friction and clarify a feasible processing space.
Since the role of adhesion and friction in injection molding is not explicitly
understood, further study in this area would also be useful. This would include materials
characterization and research into the interfaces between the thermoplastics and rapid
prototyped materials.
Areas that were not addressed in this work include part quality, as-built rapid
tools, and conformal cooling lines. First, the performance of rapid tooled inserts cannot
be completely assessed without consideration for part quality; this would include
dimensional and surface finish quality. Second, the rapid tools used in this work were
finish machined. The advantage to using rapid tools is maximized, however, if they are
inserted as built and not post-processed. And third, the addition of conformal cooling
lines would allow more control of processing temperatures. The nature of rapid
prototyping processes is such that they facilitate the incorporation of conformal cooling
lines. Further research in each of these areas would contribute to the potential use of
rapid tools for injection molding.
6.4 Summary
The application of rapid prototyped tools for injection molding, if technically
feasible, may allow for small quantity production by reducing the cost of tooling. This
165
work has investigated one aspect of the technical feasibility through testing and
experimentation to determine ejection force requirements and coefficients of friction.
Friction coefficients between thermoplastics and rapid tooled materials were measured
using a modified standard testing process. Injection molding experiments were
conducted using three mold insert materials, P-20 steel, laser sintered ST-100, and
stereolithography SL 5170 resin. Ejection forces for cylindrical parts molded with high
density polyethylene and high impact polystyrene were measured directly and then
compared with values calculated from an ejection force model. Process parameters
affected the adhesion and deformation components of friction differently, depending on
the materials characteristics. Results show that ST-100 is a good candidate for injection
molding tools, and that SL 5170 may be a good candidate for molding some
thermoplastics, but only in very small quantities.
166
LIST OF REFERENCES
Anderson, D. M., Pine, B. J. II 1997, Agile Product Development for Mass
Customization, Irwin Professional Publishing, Chicago.
ASM International 1992, “Static and Kinetic Friction Coefficients for Selected
Materials,” ASM Handbook, vol. 18, p. 73.
Balsamo, R., Hayward, D., Malloy, R. 1993, “An Experimental Evaluation of Ejection
Forces: Frictional Effects,” ANTEC, pp. 2515-2521.
Beaman, J.J., Barlow, J.W., Bourell, D.L., Crawford, R.H., Marcus, H.L., McAlea, K.P
1997, Solid Freeform Fabrication: A new Direction in Manufacturing, Kluwer
Academic Publishers, Dordrecht.
Beer, F.P., Johnston, E.R. 1981, Mechanics of Materials, McGraw-Hill, New York, pp
325-6.
Benabdallah, H., Fisa, B. 1989, “Static Friction Coefficient of Some Thermoplastics,”
ANTEC, pp. 1055-1058.
Benabdallah, S.H. 1993, “Static Shear Strength and Adhesion Friction of Some
Thermoplastics,” Polymer Engineering and Science, vol. 33, no. 9, pp. 543-548.
Bourell, D.L., Crawford, R.H., Marcus, H.L., Beaman, J.J., Barlow, J.W. 1994,
“Selective Laser Sintering of Metals,” Proceedings of the 1994 ASME Winter
Annual Meeting, November, pp. 519-528.
Briscoe, B.J. 1981, “Wear of polymers: an essay on fundamental aspects,” Tribology
International, August, pp. 231-243.
Brydson, J.A. 1999, Plastics Materials, Butterworth-Heinemann, [On line] Available
from www.knovel.com, [June 2004].
Burke, C.T. 1991, Ejection Force Measurement for the Injection Molding Process,
Master Thesis, University of Lowell, Appendix E.
167
Burke, C., Malloy, R. 1991, “An Experimental Study of the Ejection Forces Encountered
During Injection Molding,” ANTEC 1991, pp. 1781-1787.
Cagle, C.V. (ed.) 1973, Handbook of Adhesive Bonding, McGraw-Hill, New York.
Callister, W.D., Jr. 1985, Materials Science and Engineering An Introduction, John
Wiley & Sons, New York.
Campbell, C. 2000, “Rapid Tooling for Real World Applications,” Rapid Tooling
Technologies for Plastic Applications, Society of Manufacturing Engineers
Symposium, October, Detroit, MI.
Carpenter, B. 2004, personal communication.
Castle Island, “Worldwide Guide to Rapid Prototyping,” [Online]. Copyright Castle
Island Co. Available: home.att.net/~castleisland/home.htm [May 2003].
Cedorge, T., Colton, J. 2000, “Draft Angle and Surface Roughness Effects on
Stereolithography Molds,” Polymer Engineering and Science, vol. 40, no. 7, pp.
1581-1588.
Cho, W., Sachs, E. M., Patrikalakis, N., M. 2001, “Solid Freeform Fabrication with Local
Composition Control,” Rapid Prototyping, Society of Manufacturing Engineers,
vol. 7, no. 2, pp 1-5.
Colton, J. S., Crawford, J., Pham, G., Rodet, V. 2001, “Failure of Rapid Prototype Molds
During Injection Molding,” CIRP Annals – Manufacturing Technology, vol. 50, no.
1, pp. 129-132.
Colton, J.S., LeBaut, Y. 2000, “Thermal Effects on Stereolithography Injection Mold
Inserts,” Polymer Engineering and Science, vol. 40, no. 6.
Czichos, H. 1983, “Influence of Adhesive and Abrasive Mechanisms on the Tribological
Behaviour of Thermoplastic Polymers,” Wear, vol. 88, pp. 27-43.
Czichos, H. 1985, “Contact Deformation and Static Friction of Polymers, Influences of
Viscoelasticity and Adhesion,” Polymer Wear and Its Control, American Chemical
Society, pp. 3-26.
Dalgarno, K. W., Stewart, T. D. 2001, “Manufacture of production injection mould
tooling incorporating conformal cooling channels via indirect selective laser
sintering,” Proceedings of the Institution of Mechanical Engineers Part B, Journal
of Engineering Manufacture, vol. 215, pp.1323-1332.
Dearnley, P.A. 1999, “Low friction surfaces for plastic injection moulding dies – an
experimental case study,” Wear, vol. 225-229, pp. 1109-1113.
168
Dickens, P., Rudgley, M. 2001, “DirectAIMTM in a New Direction,” Rapid Prototyping
and Manufacturing Conference [CD-ROM], Society of Manufacturing Engineers,
Cincinnati.
Eley, D.D. (ed.) 1961, Adhesion, Oxford University Press, London.
Ferreira, E.C., Neves, N.M., Muschalle, R., Pouzada, A.S. 2001, “Friction Properties of
Thermoplastics in Injection Molding,” ANTEC Portugal, [Online]. Available:
www.dep.uminho.pt/NP/2001_03.pdf, [July 2003].
Ferreira, E. C., Muschalle, R., Neves, N. M., Pouzada, A. S. 2002, “Analysis of
parameters determining the friction properties of thermoplastics in injection
molding,” ANTEC 2002, vol. 1, pp. 878-881.
Glanvill, A.B. 1971, The Plastics Engineer’s Data Book, Machinery Publishing Co. Ltd.,
The Whitefriars Press Ltd., London, p. 30.
Harris, R. A., Dickens, P. M. 2001, “Mould Variables in Stereolithography Injection
Moulds,” ANTEC 2001, vol. 1, pp 107-111.
Harris, R. A., Dickens, P. M. 2003, “Determining, Understanding, and Controlling the
Morphology of Injection Moulded Parts Produced in Stereolithography Moulds,”
ANTEC 2003, pp. 889-893.
Hopkinson, N., Dickens, P. 1999, “Study of ejection forces in the AIM™ process,”
Materials and Design, vol. 20, pp. 99-105.
Hopkinson, N., Dickens, P. 2000a, “Conformal Cooling and Heating Channels using
Laser Sintered Tools,” Proceedings of the Solid Freeform Fabrication Symposium,
Austin, TX, pp. 490-497.
Hopkinson, N., Dickens, P. 2000b, “A comparison between stereolithography and
aluminium injection moulding tools,” Rapid Prototyping Journal, vol. 6, no.4, pp.
253-258.
Hopkinson, N., Dickens, P. 2000c, “Predicting stereolithography injection mould tool
behaviour using models to predict ejection force and tool strength,” International
Journal of Production Research, vol. 38, no. 16, pp. 3747-3757.
Hopkinson, N., Dickens, P. 2000d, “Using stereolithography tools for injection moulding:
research into tensile tool failure and unexpected benefits of the process,”
Proceedings of the Institution of Mechanical Engineers, vol. 214, part B, pp. 891899.
Jacobs, P. F. 1996, Stereolithography and other RP&M Technologies, Society of
Manufacturing Engineers, Dearborn, MI, ASME Press, New York.
169
Jansen, K.M.B, Titomanlio, G. 1996, “Effect of Pressure History on Shrinkage and
Residual Stresses – Injection Molding with Constrained Shrinkage,” Polymer
Engineering and Science, vol. 36, no. 15, pp 2029-2040.
Kabanemi, K.K., Vaillancourt, Wang, H., Salloum, G. 1998, “Residual Stresses,
Shrinkage and Warpage of Complex Injection Molded Products: Numerical
Simulation and Experimental Validation,” Polymer Engineering and Science, vol.
38, no. 1, pp. 21-37.
Kai, C.C., Fai, L.K 1997, Rapid Prototyping, Principles and Applications in
Manufacturing, John Wiley & Sons, New York.
Kaminsky, P.G. 1996, “Lean Logistics: Better, Faster, Cheaper,” Defense Issues, vol. 11,
no. 99, [Online] Available: http://www.defenselink.mil/speeches/index.html.
Karapatis, N.P., van Griethuysen, J.-P.S., Glardon, R. 1998, “Direct rapid tooling: a
review of current research,” Rapid Prototyping Journal, vol. 4, no. 2, pp. 77-89.
Keicher, D., Gorman, P., Taute, L. 2001, “A Novel Method for Fabrication and Repair of
Tooling?” National Tooling and Machining Association Record, January.
Kinsella, M.E. 2000, “Piloted Concepts for Commercial-Military Integration Ready for
Implementation,” Program Manager, July-August, p. 30.
Li, Y., Gargiulo, E.P., Keefe, M. 2000, “Studies in Direct Tooling Using
Stereolithography,” Journal of Manufacturing Science and Engineering, vol. 122,
May, pp. 316-322.
Malloy, R., Majeski, P. 1989, “Design of Pin Ejector Systems for Injection Molds,”
ANTEC 1989, pp. 1231-1235.
MatWeb Material Property Data (March 2004), [Online], Available from:
www.matweb.com [6 June 2004].
McAlea, K., Booth, R., Forderhase, P., Lakshminarayan, U. 1995, “Materials for
Selective Laser Sintering Processing,” Proceedings of the 27th International
SAMPE Technical Conference, vol. 27, pp. 949-961.
Menges, G., Bangert, H. 1981, “Measurement of Coefficients of Static Friction as a
Means of Determining Opening and Demoulding Forces in Injection Moulds,”
Kunststoffe, vol. 71, no. 9, pp. 552-557.
Menges, G., Michaeli, W., Mohren, P. 2001, How to Make Injection Molds, Hanser
Gardner Publications, Munich, Cincinnati.
170
Michaeli, W., Gutberlet, D., Hoffmann, S., Zachert, J. 1999, “Development steps for
integrated material modeling for a semicrystalline polymer,” Computational
Materials Science, vol.16, pp. 151-157.
Michalski, D.D. 2000, A Parametric Study of the Ejection Forces Encountered During
the Injection-Molding Process, Master Thesis, The Ohio State University.
Muschalle, R. 2001, “Friction Behaviour of Thermoplastics,” Kunststoffe, vol. 91, no. 12,
pp. 48-54.
Nelson, C., Kepler, J., Booth, R., Conner, P. 1998, “Direct Injection Molding Tooling
Inserts from the SLS Process with Copper Polyamide,” Proceedings of the Solid
Freeform Fabrication Symposium, August, Austin TX, pp. 451-459.
Optomec, Inc. 2001, company literature, Albuquerque.
Osswald, T.A. 1998, Polymer Processing Fundamentals, Hanser Publishers, Munich.
Palmer, A.E., Colton, J.S., 2000, “Failure Mechanisms in Stereolithography Injection
Molding Tooling,” Polymer Engineering and Science, vol. 40, no. 6, pp. 13951404.
Pantani, R., Titomanlio, G. 1999, “Analysis of Shrinkage Development of Injection
Moulded PS Samples,” Intern. Polymer Processing, vol. XIV, no.2, pp. 183-190.
Petrie, E.M. 2000, Handbook of Adhesives and Sealants, McGraw-Hill, [On line]
Available from www.knovel.com, [June 2004].
Pham, G.T., Colton, J.S. 2002, “Ejection Force Modeling for Stereolithography Injection
Molding Tools,” Polymer Engineering and Science, vol. 42, no. 4, pp.681-693.
Pham, D. T., Dimov, S. S., Lacan, F. 2000, “The RapidTool process: technical
capabilities and applications,” Proceedings of the Institution of Mechanical
Engineers Part B, Journal of Engineering Manufacture,vol. 214, pp 107-116.
Pine, B. J. II 1993, Mass Customization – The new Frontier in Business Competition,
Harvard Business School Press, Boston.
Pontes, A.J., Pinho, A.M., Miranda, A.S., Pouzada, A.S. 1997, “Effect of Processing
Conditions on Ejection Forces in Injection Moulds,” O Molde, OctoberDecember, pp. 25-34.
Pontes, A.J., Pantani, R., Titomanlio, G., Pouzada, A.S. 2001, “Prediction of Ejection
Forces in Tubular Moldings in Amorphous Polymers,” ANTEC, vol. I, pp. 927931.
171
Pontes, A.J., Brito, A.M., and Pouzada, A.S. 2002, “Assessment of the Ejection Force in
Tubular Injection Moldings,” Journal of Injection Molding Technology, vol. 6, no.
4, pp. 343-352.
Popov, E.P. 1976, Mechanics of Materials, second edition, Prentice-Hall, Inc.,
Englewood Cliffs, New Jersey.
Rosato, D.V., Rosato, D.V. 1986, Injection Molding Handbook, Van Nostrand Reinhold
Company, New York.
Rubin, I.I. (ed.) 1990, Handbook of Plastic Materials and Technology, John Wiley &
Sons, Inc., New York.
Sachs, E., Wylonis, E., Allen, S., Cima, M., Guo H. 2000, “Production of Injection
Molding Tooling With Conformal Cooling Channels Using the Three Dimensional
Printing Process,” Polymer Engineering and Science, May, vol. 40, no. 5, pp. 12321247.
Sasaki, T., Koga, N., Shirai, K., Koayashi, Y., Toyoshima, A. 2000, “An experimental
study on ejection forces of injection molding,” Precision Engineering, vol. 24, pp.
270-273.
Shen, K., Chen, L.-M., Jiang, L. 1999, “Calculation of ejection force of hollow, thin
walled, and injection molded cones,” Plastics, Rubber, and Composites, vol. 28, no.
7, pp. 341-345.
Titomanlio, G., Jansen, K.M.B. 1996, “In-Mold Shrinkage and Stress Prediction in
Injection Molding,” Polymer Engineering and Science, vol. 36, no. 15, pp. 20412049.
Trantina, G.G., Nimmer, R. 1994, Structural Analysis of Thermoplastic Components,
McGraw-Hill, NewYork.
University of Southern Mississippi, “The Macrogalleria,” [Online]. Copyright 1995-2002
by the Department of Polymer Science, Available:
www.psrc.usm.edu/macrog/index.htm [November 2003].
Victor, B., Boynton, A. 1998, Invented Here – Maximizing Your Organization’s Internal
Growth and Profitability, Harvard Business School Press, Boston.
Wang, H., Kabanemi, K. K., Salloum, G. 1997
Wang, H., Kabanemi, K. K., Salloum, G. 2000, “Numerical and Experimental Studies on
the Ejection of Injection-Molded Plastic Products,” Polymer Engineering and
Science, vol. 40, no. 3, pp. 826-840.
172
Weaver, T.J., Thomas, J.A., Atre, S.V., German, R.M. 2000, “Time compression-rapid
steel tooling for an ever-changing world,” Materials and Design, vol. 21, pp. 409415.
Wu, S. 1982, Polymer Interface and Adhesion, Marcel Dekker, New York.
Wypych, G. 2004, Handbook of Plasticizers, Noyes Publications, [On line] Available
from www.knovel.com, [June 2004].
173
APPENDIX A
DATA TABLES
174
HDPE at Temperature - Sample
250
RT
200
Load lbs
30 C
150
100
40 C
50 C
50
60 C
0
-0.05
0
0.05
0.1
0.15
0.2
0.25
0.3
Extension i/i
HIPS at Temperature - Sample
400
350
Room Temp
Load lbs
300
250
30 C
200
150
40 C
100
50 C
50
60 C
0
0
0.05
0.1
0.15
0.2
Elongation i/i
Type I Specimen - HDPE
Thickness
3.09 mm
Width 12.35 mm
2
Area 38.22 mm
0.12 in
0.49 in
2
0.06 in
Type I Specimen - HIPS
Thickness
3.12 mm
Width 12.76 mm
2
Area 39.85 mm
Figure A.1: Sample plots from tensile test data.
175
0.12 in
0.50 in
2
0.06 in
A.1 Tensile Test Data Table
Material
HDPE
Lutene-H
ME9180
HIPS
BASF
PS495F
Temp
degC
20.5
20.5
20.5
20.5
20.5
30
30
30
30
30
40
40
40
40
40
40
50
50
50
50
50
60
60
60
60
60
60
70
70
70
70
70
20.5
20.5
20.5
20.5
Sample
PERT1
PERT2
PERT3
PERT4
PERT5
PE301
PE302
PE303
PE304
PE305
PE401
PE402
PE403
PE404
PE405
PE406
PE501
PE502
PE503
PE504
PE505
PE601
PE602
PE603
PE604
PE605
PE606
PE701
PE702
PE703
PE704
PE705
PSRT1
PSRT2
PSRT3
PSRT4
30
30
30
40
40
40
50
50
50
60
60
60
PS30A
PS30B
PS30C
PS40A
PS40B
PS40C
PS50A
PS50B
PS50C
PS60A
PS60B
PS60C
Pull Speed
mm/min
(in/min)
50(2)
50(2)
50(2)
50(2)
50(2)
50(2)
50(2)
50(2)
50(2)
50(2)
50(2)
50(2)
50(2)
50(2)
50(2)
50(2)
5(0.2)
5(0.2)
5(0.2)
5(0.2)
5(0.2)
5(0.2)
5(0.2)
5(0.2)
5(0.2)
5(0.2)
5(0.2)
5(0.2)
5(0.2)
50(2)
50(2)
50(2)
50(2)
50(2)
50(2)
50(2)
50(2)
50(2)
50(2)
5(0.2)
5(0.2)
5(0.2)
5(0.2)
5(0.2)
5(0.2)
Slope
Slopes
Included In Excluded from
Analysis
Analysis
4469.30
3956.00
3179.90
3783.30
3386.60
2796.60
2054.50
1783.10
2708.60
2832.00
1499.60
1730.50
2233.00
1342.50
1412.80
1654.40
616.06
401.10
677.94
430.26
617.97
377.76
416.34
375.86
235.52
336.33
463.26
200.00
178.57
30871.0
30481.0
31497.0
8011.4
10513.0
10636.0
11191.0
14663.0
14753.0
5874.0
5627.6
176
Modulus
Modulus
(MPa)
(psi)
520
75439
461
66775
370
53675
440
63860
394
57164
326
47205
0
0
315
45720
330
47803
175
25312
201
29210
0
156
22661
164
23847
193
27925
72
10399
0
79
11443
0
72
10431
44
6376
48
7028
44
6344
27
3975
39
5677
54
7820
23
3376
21
3014
0
3447
3403
3517
894
1174
1187
1249
1637
1647
656
0
628
0
0
0
0
499768
493455
509903
129696
170194
172185
181170
237378
238835
95094
91105
0
0
0
Average
(MPa)
Average
(psi)
437
63383
324
46909
178
25791
74
10758
43
6203
22
3195
3455
501042
1085
157359
1511
219128
642
93099
A.2 Modulus Look-up Table
Estimate from Modulus Graph
HDPE
HIPS
temp
Mpa
Mpa
35
252
36
238
37
220
38
209
39
190
40
178
1511
40.1
176.7
1499.9
40.2
175.4
1488.8
40.3
174.1
1477.7
40.4
172.8
1466.6
40.5
171.5
1455.5
40.6
170.2
1444.4
40.7
168.9
1433.3
40.8
167.6
1422.2
40.9
166.3
1411.1
41
165
1400
41.1
163.9
1391
41.2
162.8
1382
41.3
161.7
1373
41.4
160.6
1364
41.5
159.5
1355
41.6
158.4
1346
41.7
157.3
1337
41.8
156.2
1328
41.9
155.1
1319
42
154
1310
42.1
152.6
1302
42.2
151.2
1294
42.3
149.8
1286
42.4
148.4
1278
42.5
147
1270
42.6
145.6
1262
42.7
144.2
1254
42.8
142.8
1246
42.9
141.4
1238
43
140
1230
44
131
1150
45
119
1050
46
110
970
47
100
890
48
90
800
48.1
89.2
792
48.2
88.4
784
48.3
87.6
776
48.4
86.8
768
48.5
86
760
48.6
85.2
752
48.7
84.4
744
48.8
83.6
736
48.9
82.8
728
Estimate from Modulus Graph
HDPE
HIPS
temp
Mpa
Mpa
49
82
720
49.1
81.2
712.2
49.2
80.4
704.4
49.3
79.6
696.6
49.4
78.8
688.8
49.5
78
681
49.6
77.2
673.2
49.7
76.4
665.4
49.8
75.6
657.6
49.9
74.8
649.8
50
74
642
50.1
73.5
635.8
50.2
73
629.6
50.3
72.5
623.4
50.4
72
617.2
50.5
71.5
611
50.6
71
604.8
50.7
70.5
598.6
50.8
70
592.4
50.9
69.5
586.2
51
69
580
51.1
68.1
573
51.2
67.2
566
51.3
66.3
559
51.4
65.4
552
51.5
64.5
545
51.6
63.6
538
51.7
62.7
531
51.8
61.8
524
51.9
60.9
517
52
60
510
52.1
59.9
504
52.2
59.8
498
52.3
59.7
492
52.4
59.6
486
52.5
59.5
480
52.6
59.4
474
52.7
59.3
468
52.8
59.2
462
52.9
59.1
456
53
59
450
53.1
58.8
445
53.2
58.6
440
53.3
58.4
435
53.4
58.2
430
53.5
58
425
53.6
57.8
420
53.7
57.6
415
53.8
57.4
410
53.9
57.2
405
177
Estimate from Modulus Graph
HDPE
HIPS
temp
Mpa
Mpa
54
57
400
54.1
56.7
391
54.2
56.4
382
54.3
56.1
373
54.4
55.8
364
54.5
55.5
355
54.6
55.2
346
54.7
54.9
337
54.8
54.6
328
54.9
54.3
319
55
54
310
55.1
53.8
307
55.2
53.6
304
55.3
53.4
301
55.4
53.2
298
55.5
53
295
55.6
52.8
292
55.7
52.6
289
55.8
52.4
286
55.9
52.2
283
56
52
280
56.1
51.8
275
56.2
51.6
270
56.3
51.4
265
56.4
51.2
260
56.5
51
255
56.6
50.8
250
56.7
50.6
245
56.8
50.4
240
56.9
50.2
235
57
50
230
58
48
170
59
45
120
60
43
100
61
41
70
62
39
60
63
36
30
64
34
65
32
A.3 Thermal Analysis Convergence Table
P20 and HDPE
Time
sec
0
3.39E-03
6.77E-03
1.02E-02
1.35E-02
1.69E-02
2.03E-02
2.37E-02
3.05E-02
3.73E-02
4.40E-02
5.08E-02
5.76E-02
7.11E-02
8.47E-02
9.82E-02
1.12E-01
1.25E-01
1.39E-01
1.66E-01
1.93E-01
2.20E-01
2.47E-01
2.74E-01
3.01E-01
3.29E-01
3.56E-01
3.83E-01
4.37E-01
4.91E-01
5.45E-01
6.00E-01
6.54E-01
7.08E-01
7.62E-01
8.16E-01
8.70E-01
9.25E-01
9.79E-01
1.08727
1.19566
1.30405
1.41244
1.52082
1.7376
1.95438
2.17115
2.60471
3.03826
3.47181
4.33892
5.33892
6.33892
7.33892
8.33892
9.33892
10.3389
11.3389
12.3389
13.3389
14.3389
15.3389
16.3389
17.3389
18.3389
19.3389
20.3389
P20 and HIPS
Thermocouple Peak Polymer
Reading
Temperature
C
50
50.0009
50.0039
50.0114
50.0256
50.0484
50.0807
50.1228
50.2409
50.3853
50.5482
50.7229
50.9043
51.2692
51.623
51.9591
52.2751
52.5704
52.8457
53.3268
53.7491
54.1225
54.4549
54.7528
55.0215
55.2652
55.4872
55.6904
56.0362
56.3323
56.5867
56.8053
56.9931
57.154
57.2911
57.4074
57.5051
57.5864
57.6531
57.7399
57.7896
57.8099
57.8069
57.7859
57.6994
57.5874
57.463
57.2144
56.9887
56.7936
56.5142
56.2969
56.1476
56.0401
55.9575
55.8894
55.8297
55.7748
55.7225
55.6718
55.622
55.5727
55.5238
55.4753
55.4269
55.3789
55.331
C
210
209.958
210.002
210.054
210.087
210.096
210.085
210.06
210.001
209.954
209.924
209.906
209.887
209.755
209.465
208.977
208.279
207.371
206.267
203.43
200.123
196.485
192.631
188.652
184.616
180.576
176.567
172.619
165.075
157.939
151.228
144.941
139.062
133.574
128.453
123.678
119.226
115.074
111.203
104.436
98.518
93.338
88.8004
84.8222
78.5953
73.708
69.8605
64.8203
61.4853
59.2533
56.9357
55.5814
54.8542
54.4353
54.1744
53.9982
53.8696
53.7694
53.687
53.6164
53.5541
53.4978
53.4461
53.398
53.3528
53.31
53.2692
Time
sec
0
0.007719
0.015438
0.023156
0.030875
0.038594
0.046313
0.054032
0.06175
0.077188
0.092625
0.108063
0.123501
0.138938
0.169813
0.200688
0.231563
0.262439
0.293314
0.324189
0.385939
0.447689
0.50944
0.57119
0.63294
0.69469
0.756441
0.818191
0.879941
0.941692
1.00344
1.12694
1.25044
1.37394
1.49744
1.62094
1.74444
1.86795
1.99145
2.11495
2.23845
2.36195
2.60895
2.85595
3.10295
3.34995
3.59695
3.84395
4.33796
4.83196
5.32596
6.31396
7.30197
8.30197
9.30197
10.302
11.302
12.302
13.302
14.302
15.302
16.302
17.302
18.302
19.302
20.302
21.302
Sintered and HDPE
Thermocouple Peak Polymer
Reading
Temperature
C
50
50.0074
50.031
50.0758
50.1419
50.2262
50.3247
50.4331
50.5479
50.7851
51.0197
51.246
51.4614
51.6647
52.0251
52.3443
52.6277
52.8806
53.1078
53.313
53.6575
53.9519
54.2074
54.4321
54.632
54.8115
54.974
55.122
55.2575
55.3821
55.497
55.695
55.8664
56.0149
56.1436
56.2551
56.3514
56.4344
56.5057
56.5665
56.6183
56.6619
56.7238
56.766
56.7928
56.8077
56.8137
56.8129
56.7958
56.7707
56.742
56.6856
56.6351
56.5907
56.5522
56.5182
56.4875
56.4591
56.4322
56.4062
56.3809
56.3559
56.3312
56.3065
56.2819
56.2572
56.2326
C
221
220.954
220.995
221.049
221.088
221.104
221.099
221.078
221.048
220.987
220.942
220.914
220.897
220.88
220.764
220.5
220.051
219.4
218.545
217.495
214.75
211.506
207.897
204.036
200.014
195.905
191.762
187.628
183.534
179.502
175.548
168.021
160.901
154.195
147.895
141.986
136.449
131.265
126.412
121.87
117.62
113.642
106.646
100.491
95.0735
90.304
86.1032
82.4017
76.5642
71.9549
68.3107
63.5261
60.3707
58.2612
56.8584
55.9176
55.28
54.8421
54.5364
54.3188
54.1601
54.0413
53.9497
53.8769
53.8171
53.7667
53.723
Time
sec
0
0.003387
0.006774
0.01016
0.013547
0.016934
0.020321
0.023708
0.030481
0.037255
0.044029
0.050802
0.057576
0.071123
0.08467
0.098218
0.111765
0.125312
0.138859
0.165954
0.193048
0.220143
0.247237
0.274332
0.301426
0.328521
0.355615
0.38271
0.436899
0.491088
0.545277
0.599466
0.653654
0.707843
0.762032
0.816221
0.87041
0.924599
0.978788
1.08717
1.19554
1.30392
1.4123
1.52068
1.73743
1.95419
2.17095
2.60446
3.03797
3.47148
4.3385
5.3385
6.3385
7.3385
8.3385
9.3385
10.3385
11.3385
12.3385
13.3385
14.3385
15.3385
16.3385
17.3385
18.3385
19.3385
20.3385
178
Sintered and HIPS
Thermocouple Peak Polymer
Reading
Temperature
C
50
50.0011
50.005
50.0146
50.0324
50.0602
50.0989
50.1486
50.2838
50.4457
50.6253
50.8155
51.0109
51.398
51.7692
52.1191
52.4457
52.7493
53.0311
53.5207
53.9488
54.3259
54.6608
54.9605
55.2303
55.4747
55.6973
55.9007
56.2464
56.5421
56.7957
57.0134
57.2001
57.3597
57.4956
57.6104
57.7067
57.7866
57.8519
57.9359
57.9831
58.0011
57.9962
57.9735
57.8848
57.7719
57.6476
57.4023
57.1819
56.9929
56.7253
56.5184
56.3756
56.2713
56.1893
56.1201
56.0579
55.9997
55.9436
55.8888
55.8348
55.7811
55.7279
55.6749
55.6222
55.5698
55.5177
C
210
209.958
210.002
210.054
210.087
210.096
210.085
210.06
210.001
209.954
209.924
209.906
209.887
209.756
209.465
208.978
208.28
207.373
206.27
203.434
200.128
196.492
192.641
188.664
184.63
180.592
176.585
172.639
165.099
157.967
151.26
144.975
139.1
133.615
128.497
123.725
119.275
115.125
111.256
104.493
98.5779
93.4006
88.8653
84.8891
78.6654
73.7805
69.9351
64.8984
61.5669
59.3384
57.0278
55.6808
54.9592
54.5444
54.286
54.1109
53.9824
53.8816
53.7982
53.7262
53.6622
53.6041
53.5505
53.5005
53.4533
53.4085
53.3658
Time
sec
0
7.72E-03
1.54E-02
2.32E-02
3.09E-02
3.86E-02
4.63E-02
5.40E-02
6.17E-02
7.72E-02
9.26E-02
1.08E-01
1.23E-01
1.39E-01
1.70E-01
2.01E-01
2.32E-01
2.62E-01
2.93E-01
3.24E-01
3.86E-01
4.48E-01
5.09E-01
5.71E-01
6.33E-01
6.95E-01
7.56E-01
8.18E-01
8.80E-01
9.42E-01
1.00337
1.12686
1.25035
1.37384
1.49734
1.62083
1.74432
1.86781
1.9913
2.11479
2.23829
2.36178
2.60876
2.85574
3.10273
3.34971
3.59669
3.84368
4.33764
4.83161
5.32558
6.31351
7.30144
8.30144
9.30144
10.3014
11.3014
12.3014
13.3014
14.3014
15.3014
16.3014
17.3014
18.3014
19.3014
20.3014
21.3014
Thermocouple
Reading
C
50
50.0091
50.0373
50.0894
50.1647
50.2591
50.3676
50.4857
50.6094
50.8615
51.108
51.3439
51.567
51.7766
52.1456
52.4706
52.7583
53.0143
53.2436
53.4505
53.7972
54.0932
54.35
54.5758
54.7767
54.9572
55.1207
55.2696
55.4061
55.5317
55.6476
55.8476
56.021
56.1716
56.3024
56.416
56.5144
56.5996
56.6731
56.7363
56.7903
56.8364
56.9033
56.9507
56.9828
57.0031
57.0144
57.0189
57.0119
56.996
56.9755
56.9315
56.8898
56.8514
56.8165
56.7844
56.7543
56.7256
56.6977
56.6704
56.6433
56.6164
56.5896
56.5628
56.5359
56.5091
56.4821
Peak
Polymer
Temperature
C
221
220.954
220.995
221.049
221.088
221.104
221.099
221.078
221.048
220.987
220.942
220.914
220.897
220.88
220.764
220.5
220.052
219.401
218.546
217.497
214.753
211.51
207.902
204.042
200.022
195.914
191.773
187.641
183.548
179.518
175.566
168.041
160.924
154.221
147.924
142.017
136.483
131.3
126.45
121.91
117.662
113.686
106.694
100.543
95.129
90.363
86.1656
82.4674
76.6362
72.0331
68.3946
63.6202
60.4733
58.3708
56.9734
56.0368
55.4022
54.9664
54.662
54.4451
54.2867
54.1678
54.0759
54.0025
53.9421
53.891
53.8464
A.3 Thermal Analysis Convergence Table (continued)
P20/SLA and HDPE
Time
sec
0
0.001915
0.003831
0.005746
0.007661
0.009576
0.011492
0.013407
0.017237
0.021068
0.024898
0.028729
0.03639
0.044051
0.051712
0.059373
0.074695
0.090017
0.10534
0.120662
0.135984
0.166628
0.197272
0.227916
0.258561
0.289205
0.319849
0.381137
0.442426
0.503714
0.565003
0.626291
0.68758
0.748868
0.810156
0.932733
1.05531
1.17789
1.30046
1.42304
1.54562
1.79077
2.03593
2.28108
2.52623
3.01654
3.50685
3.99716
4.48746
5.46808
6.44869
7.44869
8.44869
9.44869
10.4487
11.4487
12.4487
Thermocouple Peak Polymer
Reading
Temperature
C
30
29.9969
29.9948
29.9934
29.9927
29.9926
29.9928
29.9934
29.9954
29.9978
30.0004
30.003
30.0074
30.0106
30.0127
30.0138
30.0129
30.0104
30.0072
30.0041
30.0015
29.9989
29.9986
29.9997
30.0014
30.0032
30.0048
30.0063
30.0068
30.0068
30.007
30.0078
30.0094
30.0118
30.0151
30.0259
30.0422
30.0655
30.0978
30.1407
30.1958
30.3603
30.5795
30.8496
31.1638
31.903
32.6928
33.4856
34.2472
35.5264
36.5324
37.292
37.8225
38.1658
38.361
38.442
38.4366
C
177
176.942
176.946
176.979
177.022
177.061
177.092
177.112
177.112
177.087
177.05
177.011
176.958
176.935
176.929
176.921
176.788
176.471
175.946
175.215
174.294
171.891
169.106
166.084
162.933
159.733
156.537
150.385
144.573
139.137
134.079
129.382
125.024
120.978
117.218
110.656
104.896
99.8088
95.2896
91.2527
87.6279
81.631
76.6035
72.3389
68.6838
63.0562
58.6276
55.0774
52.1857
48.0636
45.0333
42.7113
40.9253
39.5234
38.4031
37.4934
36.7442
Time
sec
13.4487
14.4487
15.4487
16.4487
17.4487
18.4487
19.4487
20.4487
21.4487
22.4487
23.4487
24.4487
25.4487
26.4487
27.4487
28.4487
29.4487
30.4487
31.4487
32.4487
33.4487
34.4487
35.4487
36.4487
37.4487
38.4487
39.4487
40.4487
41.4487
42.4487
43.4487
44.4487
45.4487
46.4487
47.4487
48.4487
49.4487
50.4487
51.4487
52.4487
53.4487
54.4487
55.4487
56.4487
57.4487
58.4487
59.4487
60
P20/SLA and HIPS
Thermocouple Peak Polymer
Reading
Temperature
C
38.3671
38.2508
38.1014
37.9291
37.7419
37.5457
37.3451
37.1432
36.9426
36.7451
36.5517
36.3635
36.181
36.0045
35.8343
35.6703
35.5126
35.3611
35.2155
35.0758
34.9417
34.813
34.6895
34.5711
34.4573
34.3481
34.2433
34.1425
34.0457
33.9527
33.8632
33.7771
33.6942
33.6144
33.5375
33.4634
33.392
33.3231
33.2565
33.1923
33.1303
33.0704
33.0124
32.9564
32.9022
32.8497
32.7989
32.7714
C
36.1194
35.5922
35.1429
34.7562
34.4207
34.1274
33.869
33.6399
33.4357
33.2526
33.0877
32.9383
32.8025
32.6786
32.5651
32.4607
32.3645
32.2754
32.1928
32.116
32.0444
31.9775
31.9149
31.8561
31.8008
31.7488
31.6996
31.6532
31.6093
31.5676
31.5281
31.4905
31.4548
31.4207
31.3882
31.3571
31.3275
31.2991
31.2719
31.2458
31.2208
31.1968
31.1737
31.1515
31.1301
31.1095
31.0897
31.079
179
Time
sec
0
3.73E-03
7.46E-03
1.12E-02
1.49E-02
1.87E-02
2.24E-02
2.61E-02
3.36E-02
4.11E-02
4.85E-02
5.60E-02
6.34E-02
7.84E-02
9.33E-02
1.08E-01
1.23E-01
1.38E-01
1.68E-01
1.98E-01
2.28E-01
2.58E-01
2.87E-01
3.17E-01
3.47E-01
4.07E-01
4.67E-01
5.26E-01
5.86E-01
6.46E-01
7.05E-01
7.65E-01
8.85E-01
1.00396
1.12339
1.24282
1.36225
1.48168
1.60111
1.72054
1.83997
1.9594
2.19827
2.43713
2.67599
2.91485
3.15371
3.39257
3.63143
4.10915
4.58687
5.0646
5.54232
6.02004
6.97548
7.93092
8.93092
Thermocouple Peak Polymer
Reading
Temperature
C
33
32.9975
32.9964
32.9962
32.9967
32.9977
32.999
33.0005
33.0035
33.0061
33.0083
33.0098
33.0107
33.0106
33.0092
33.0073
33.0051
33.0031
33.0008
33.0001
33.0005
33.0016
33.003
33.0043
33.0054
33.0064
33.0067
33.0068
33.0071
33.0079
33.0094
33.0116
33.0194
33.0316
33.0492
33.074
33.1075
33.1513
33.2069
33.2752
33.357
33.4529
33.7007
34.0014
34.3499
34.7404
35.166
35.6202
36.0965
37.0936
38.0999
39.0884
40.0395
40.9398
42.4921
43.7884
44.881
C
210
209.931
209.922
209.946
209.986
210.03
210.071
210.105
210.138
210.14
210.119
210.084
210.044
209.976
209.934
209.916
209.912
209.913
209.841
209.64
209.277
208.74
208.03
207.158
206.137
203.634
200.791
197.713
194.482
191.163
187.804
184.444
177.887
171.589
165.596
159.924
154.575
149.54
144.803
140.348
136.159
132.216
125.19
118.91
113.279
108.214
103.642
99.504
95.7464
89.4384
84.097
79.5367
75.6134
72.214
66.912
62.7018
59.1744
Time
sec
9.93092
10.9309
11.9309
12.9309
13.9309
14.9309
15.9309
16.9309
17.9309
18.9309
19.9309
20.9309
21.9309
22.9309
23.9309
24.9309
25.9309
26.9309
27.9309
28.9309
29.9309
30.9309
31.9309
32.9309
33.9309
34.9309
35.9309
36.9309
37.9309
38.9309
39.9309
40.9309
41.9309
42.9309
43.9309
44.9309
45.9309
46.9309
47.9309
48.9309
49.9309
50.9309
51.9309
52.9309
53.9309
54.9309
55.9309
56.9309
57.9309
58.9309
59.9309
60
Thermocouple Peak Polymer
Reading
Temperature
C
45.7366
46.3867
46.8622
47.192
47.4015
47.5125
47.5437
47.5107
47.4265
47.3018
47.1453
46.9643
46.7648
46.5514
46.3281
46.0981
45.864
45.6277
45.391
45.1552
44.9213
44.6902
44.4624
44.2386
44.0191
43.8041
43.5939
43.3886
43.1882
42.9927
42.8023
42.6168
42.4361
42.2603
42.0892
41.9227
41.7607
41.6031
41.4498
41.3007
41.1557
41.0145
40.8772
40.7436
40.6136
40.4871
40.3639
40.244
40.1272
40.0135
39.9028
39.8951
C
56.3059
53.9398
51.963
50.2924
48.8659
47.6362
46.5674
45.6312
44.8054
44.0724
43.4179
42.8306
42.3009
41.8211
41.3847
40.9862
40.621
40.2854
39.9758
39.6897
39.4243
39.1777
38.9479
38.7334
38.5327
38.3446
38.1679
38.0017
37.8451
37.6973
37.5576
37.4254
37.3001
37.1812
37.0682
36.9607
36.8584
36.7608
36.6677
36.5788
36.4938
36.4124
36.3344
36.2597
36.1881
36.1193
36.0531
35.9896
35.9284
35.8695
35.8128
35.8089
A.3 Thermal Analysis Convergence Table (continued)
SLA and HDPE
Time
sec
0
0.001915
0.003831
0.005746
0.007661
0.009576
0.011492
0.013407
0.017237
0.021068
0.024898
0.028729
0.03639
0.044051
0.059373
0.074695
0.090017
0.105339
0.135983
0.166628
0.197272
0.25856
0.319848
0.381137
0.442425
0.503713
0.62629
0.748866
0.871443
0.994019
1.23917
1.48433
1.72948
1.97463
2.46494
2.95524
3.44555
4.42616
5.40678
6.38739
7.38739
8.38739
9.38739
10.3874
11.3874
12.3874
13.3874
14.3874
15.3874
16.3874
17.3874
18.3874
19.3874
20.3874
21.3874
22.3874
23.3874
24.3874
25.3874
26.3874
27.3874
28.3874
29.3874
30.3874
31.3874
32.3874
33.3874
Thermocouple Peak Polymer
Reading
Temperature
C
30
29.9969
29.9948
29.9934
29.9927
29.9926
29.9928
29.9934
29.9954
29.9978
30.0004
30.003
30.0074
30.0106
30.013
30.0125
30.0102
30.0073
30.0023
29.9996
29.9991
30.0014
30.004
30.0057
30.0065
30.0069
30.0091
30.0144
30.0238
30.0384
30.0998
30.2049
30.3615
30.5729
31.1932
31.9665
32.8439
34.7255
36.5589
38.253
39.7962
41.147
42.3143
43.3146
44.1668
44.8896
45.5007
46.0156
46.448
46.8098
47.1112
47.3607
47.5658
47.7326
47.8666
47.9722
48.0534
48.1134
48.1551
48.1809
48.193
48.1931
48.1828
48.1634
48.1363
48.1022
48.0623
C
177
176.939
176.944
176.978
177.022
177.064
177.096
177.117
177.118
177.092
177.054
177.014
176.961
176.94
176.91
176.784
176.499
176.03
174.47
172.471
170.206
165.42
160.784
156.469
152.525
148.947
142.997
137.993
133.722
130.023
124.133
119.284
115.184
111.644
106.025
101.419
97.5524
91.6781
87.0491
83.2962
80.1313
77.4634
75.178
73.1934
71.45
69.9031
68.5185
67.2699
66.1363
65.101
64.1507
63.2742
62.4624
61.7077
61.0037
60.3449
59.7267
59.1451
58.5965
58.0781
57.587
57.1211
56.6781
56.2563
55.8541
55.47
55.1026
Time
sec
44.3874
45.3874
46.3874
47.3874
48.3874
49.3874
50.3874
51.3874
52.3874
53.3874
54.3874
55.3874
56.3874
57.3874
58.3874
59.3874
60.3874
61.3874
62.3874
63.3874
64.3874
65.3874
66.3874
67.3874
68.3874
69.3874
70.3874
71.3874
72.3874
73.3874
74.3874
75.3874
76.3874
77.3874
78.3874
79.3874
80.3874
81.3874
82.3874
83.3874
84.3874
85.3874
86.3874
87.3874
88.3874
89.3874
90.3874
91.3874
92.3874
93.3874
94.3874
95.3874
96.3874
97.3874
98.3874
99.3874
100.387
101.387
102.387
103.387
104.387
105.387
106.387
107.387
108.387
109.387
110.387
Thermocouple Peak Polymer
Reading
Temperature
C
47.3999
47.3293
47.2581
47.1864
47.1143
47.042
46.9695
46.897
46.8245
46.7521
46.6799
46.6079
46.5362
46.4649
46.3939
46.3233
46.2532
46.1835
46.1143
46.0456
45.9775
45.9099
45.8428
45.7763
45.7104
45.645
45.5803
45.5161
45.4525
45.3895
45.3271
45.2653
45.204
45.1434
45.0834
45.0239
44.965
44.9067
44.849
44.7919
44.7353
44.6793
44.6238
44.5689
44.5145
44.4607
44.4074
44.3547
44.3024
44.2507
44.1995
44.1488
44.0986
44.0489
43.9997
43.951
43.9027
43.855
43.8076
43.7608
43.7144
43.6684
43.6229
43.5778
43.5332
43.489
43.4452
C
51.8966
51.663
51.4368
51.2176
51.0051
50.7989
50.5987
50.4043
50.2154
50.0317
49.853
49.6791
49.5097
49.3447
49.1839
49.0271
48.8741
48.7248
48.5791
48.4368
48.2978
48.1619
48.029
47.8991
47.772
47.6476
47.5259
47.4067
47.2899
47.1755
47.0634
46.9535
46.8457
46.7401
46.6364
46.5347
46.4348
46.3368
46.2406
46.1461
46.0532
45.962
45.8724
45.7843
45.6976
45.6125
45.5287
45.4463
45.3652
45.2855
45.207
45.1297
45.0536
44.9787
44.905
44.8323
44.7607
44.6902
44.6208
44.5523
44.4848
44.4183
44.3527
44.288
44.2242
44.1613
44.0993
180
Time
sec
124.387
125.387
126.387
127.387
128.387
129.387
130.387
131.387
132.387
133.387
134.387
135.387
136.387
137.387
138.387
139.387
140.387
141.387
142.387
143.387
144.387
145.387
146.387
147.387
148.387
149.387
150.387
151.387
152.387
153.387
154.387
155.387
156.387
157.387
158.387
159.387
160.387
161.387
162.387
163.387
164.387
165.387
166.387
167.387
168.387
169.387
170.387
171.387
172.387
173.387
174.387
175.387
176.387
177.387
178.387
179.387
180.387
181.387
182.387
183.387
184.387
185.387
186.387
187.387
188.387
189.387
190.387
Thermocouple
Reading
Peak
Polymer
Temperature
C
42.873
42.8349
42.7971
42.7596
42.7225
42.6857
42.6492
42.613
42.5772
42.5416
42.5064
42.4715
42.4368
42.4025
42.3684
42.3347
42.3012
42.268
42.2351
42.2024
42.17
42.1379
42.106
42.0745
42.0431
42.012
41.9812
41.9506
41.9203
41.8902
41.8603
41.8307
41.8013
41.7721
41.7432
41.7144
41.6859
41.6577
41.6296
41.6018
41.5741
41.5467
41.5195
41.4925
41.4656
41.439
41.4126
41.3864
41.3604
41.3345
41.3089
41.2834
41.2581
41.233
41.2081
41.1834
41.1588
41.1344
41.1102
41.0862
41.0623
41.0386
41.0151
40.9917
40.9685
40.9454
40.9225
C
43.3108
43.2596
43.209
43.159
43.1096
43.0608
43.0125
42.9648
42.9176
42.871
42.8249
42.7793
42.7342
42.6896
42.6455
42.6018
42.5587
42.516
42.4737
42.4319
42.3905
42.3496
42.3091
42.269
42.2293
42.19
42.1511
42.1126
42.0745
42.0368
41.9994
41.9624
41.9258
41.8895
41.8535
41.8179
41.7826
41.7477
41.7131
41.6788
41.6448
41.6112
41.5778
41.5448
41.512
41.4796
41.4474
41.4155
41.3839
41.3526
41.3216
41.2908
41.2603
41.2301
41.2001
41.1703
41.1409
41.1116
41.0826
41.0539
41.0254
40.9971
40.9691
40.9412
40.9136
40.8863
40.8591
A.3 Thermal Analysis Convergence Table (continued)
SLA and HIPS
Time
Thermocouple
Reading
Peak Polymer
Temperature
sec
C
C
33
32.9975
32.9964
32.9962
32.9967
32.9977
32.999
33.0005
33.0035
33.0061
33.0083
33.0098
33.0106
33.0099
33.0082
33.0062
33.0028
33.0008
33.0003
33.0009
33.0032
33.005
33.0061
33.0073
33.0098
33.015
33.0239
33.0374
33.0567
33.0834
33.1189
33.235
33.4027
33.6242
33.8989
34.2235
34.5935
35.4695
36.4382
37.4618
38.5097
39.558
41.5477
43.3751
45.0859
46.5935
47.9085
49.0472
50.0281
50.8695
51.5886
52.2012
52.7212
53.161
53.5313
53.8414
54.0994
54.3121
54.4855
54.6249
54.7345
54.8184
54.8798
54.9216
54.9463
54.956
54.9526
210
209.928
209.918
209.943
209.985
210.031
210.075
210.111
210.146
210.148
210.126
210.09
210.013
209.958
209.929
209.922
209.909
209.819
209.603
209.237
207.925
206.137
203.998
199.048
193.892
188.794
183.889
179.243
174.877
170.788
166.967
160.247
154.31
149.033
144.313
140.066
136.223
129.766
124.261
119.51
115.365
111.714
105.825
100.975
96.7511
93.1789
90.1158
87.4574
85.1256
83.0613
81.2185
79.5614
78.0616
76.6961
75.4464
74.2973
73.236
72.2521
71.3368
70.4824
69.6825
68.9317
68.2251
67.5587
66.9287
66.332
65.7657
0
0.003732
0.007464
0.011197
0.014929
0.018661
0.022393
0.026125
0.03359
0.041054
0.048518
0.055983
0.070911
0.08584
0.100769
0.115698
0.145555
0.175412
0.20527
0.235127
0.294842
0.354557
0.414272
0.533701
0.653131
0.772561
0.89199
1.01142
1.13085
1.25028
1.36971
1.60857
1.84743
2.08629
2.32515
2.56401
2.80287
3.28058
3.7583
4.23602
4.71374
5.19146
6.1469
7.10233
8.10233
9.10233
10.1023
11.1023
12.1023
13.1023
14.1023
15.1023
16.1023
17.1023
18.1023
19.1023
20.1023
21.1023
22.1023
23.1023
24.1023
25.1023
26.1023
27.1023
28.1023
29.1023
30.1023
Time
sec
41.1023
42.1023
43.1023
44.1023
45.1023
46.1023
47.1023
48.1023
49.1023
50.1023
51.1023
52.1023
53.1023
54.1023
55.1023
56.1023
57.1023
58.1023
59.1023
60.1023
61.1023
62.1023
63.1023
64.1023
65.1023
66.1023
67.1023
68.1023
69.1023
70.1023
71.1023
72.1023
73.1023
74.1023
75.1023
76.1023
77.1023
78.1023
79.1023
80.1023
81.1023
82.1023
83.1023
84.1023
85.1023
86.1023
87.1023
88.1023
89.1023
90.1023
91.1023
92.1023
93.1023
94.1023
95.1023
96.1023
97.1023
98.1023
99.1023
100.102
101.102
102.102
103.102
104.102
105.102
106.102
107.102
Thermocouple
Reading
Peak Polymer
Temperature
C
54.4061
54.3307
54.2533
54.1742
54.0935
54.0117
53.9288
53.8451
53.7608
53.6761
53.591
53.5057
53.4203
53.3349
53.2497
53.1645
53.0797
52.9951
52.9109
52.8271
52.7437
52.6608
52.5785
52.4966
52.4154
52.3347
52.2547
52.1752
52.0964
52.0183
51.9408
51.864
51.7879
51.7124
51.6376
51.5635
51.4901
51.4173
51.3453
51.2739
51.2032
51.1332
51.0638
50.9951
50.9271
50.8598
50.7931
50.727
50.6616
50.5968
50.5327
50.4691
50.4062
50.3439
50.2823
50.2212
50.1607
50.1008
50.0415
49.9827
49.9246
49.8669
49.8099
49.7533
49.6974
49.6419
49.587
181
C
60.9927
60.6557
60.3308
60.0171
59.7141
59.4212
59.1378
58.8634
58.5976
58.34
58.09
57.8474
57.6118
57.3828
57.1603
56.9437
56.733
56.5279
56.328
56.1333
55.9434
55.7582
55.5775
55.4011
55.2288
55.0606
54.8961
54.7354
54.5782
54.4244
54.274
54.1267
53.9825
53.8413
53.703
53.5674
53.4346
53.3043
53.1766
53.0513
52.9283
52.8077
52.6892
52.5729
52.4587
52.3466
52.2364
52.1281
52.0216
51.917
51.8142
51.713
51.6135
51.5156
51.4193
51.3245
51.2312
51.1394
51.049
50.96
50.8723
50.7859
50.7008
50.617
50.5343
50.4529
50.3726
Time
sec
121.102
122.102
123.102
124.102
125.102
126.102
127.102
128.102
129.102
130.102
131.102
132.102
133.102
134.102
135.102
136.102
137.102
138.102
139.102
140.102
141.102
142.102
143.102
144.102
145.102
146.102
147.102
148.102
149.102
150.102
151.102
152.102
153.102
154.102
155.102
156.102
157.102
158.102
159.102
160.102
161.102
162.102
163.102
164.102
165.102
166.102
167.102
168.102
169.102
170.102
171.102
172.102
173.102
174.102
175.102
176.102
177.102
178.102
179.102
180.102
181.102
182.102
183.102
184.102
185.102
186.102
187.102
Thermocouple Peak Polymer
Reading
Temperature
C
48.8698
48.822
48.7747
48.7278
48.6813
48.6352
48.5895
48.5442
48.4993
48.4548
48.4107
48.367
48.3236
48.2806
48.238
48.1958
48.1539
48.1124
48.0712
48.0303
47.9898
47.9497
47.9099
47.8703
47.8312
47.7923
47.7538
47.7155
47.6776
47.64
47.6027
47.5657
47.529
47.4925
47.4564
47.4205
47.385
47.3497
47.3146
47.2799
47.2454
47.2112
47.1772
47.1435
47.11
47.0768
47.0439
47.0112
46.9787
46.9465
46.9145
46.8828
46.8513
46.82
46.789
46.7581
46.7275
46.6971
46.667
46.637
46.6073
46.5778
46.5485
46.5193
46.4904
46.4617
46.4332
C
49.3572
49.2916
49.2268
49.1627
49.0995
49.037
48.9753
48.9143
48.854
48.7945
48.7356
48.6774
48.6199
48.5631
48.5069
48.4513
48.3964
48.342
48.2883
48.2351
48.1826
48.1306
48.0791
48.0282
47.9778
47.928
47.8787
47.8299
47.7816
47.7338
47.6865
47.6396
47.5933
47.5474
47.5019
47.4569
47.4124
47.3682
47.3245
47.2812
47.2384
47.1959
47.1538
47.1122
47.0709
47.03
46.9895
46.9493
46.9096
46.8701
46.8311
46.7924
46.754
46.7159
46.6782
46.6409
46.6038
46.5671
46.5307
46.4946
46.4588
46.4234
46.3882
46.3533
46.3187
46.2844
46.2504
A.4 Sample Experimental Data Set, All Runs
Date-Time and Part ID
20040304-102314 Set 1 Run 1 Rep 1.xls
20040304-102433 Set 1 Run 1 Rep 2.xls
20040304-102552 Set 1 Run 1 Rep 3.xls
20040304-102716 Set 1 Run 1 Rep 4.xls
20040304-102838 Set 1 Run 1 Rep 5.xls
20040304-103000 Set 1 Run 1 Rep 6.xls
20040304-103124 Set 1 Run 1 Rep 7.xls
20040304-103247 Set 1 Run 1 Rep 8.xls
20040304-103857 Set 1 Run 2 Rep 1.xls
20040304-104021 Set 1 Run 2 Rep 2.xls
20040304-104145 Set 1 Run 2 Rep 3.xls
20040304-104314 Set 1 Run 2 Rep 4.xls
20040304-104449 Set 1 Run 2 Rep 5.xls
20040304-104622 Set 1 Run 2 Rep 6.xls
20040304-104756 Set 1 Run 2 Rep 7.xls
20040304-104929 Set 1 Run 2 Rep 8.xls
20040304-105106 Set 1 Run 3 Rep 1.xls
20040304-105240 Set 1 Run 3 Rep 2.xls
20040304-105416 Set 1 Run 3 Rep 3.xls
20040304-105552 Set 1 Run 3 Rep 4.xls
20040304-105730 Set 1 Run 3 Rep 5.xls
20040304-105908 Set 1 Run 3 Rep 6.xls
20040304-110046 Set 1 Run 3 Rep 7.xls
20040304-110226 Set 1 Run 3 Rep 8.xls
20040304-111421 Set 1 Run 4 Rep 1.xls
20040304-111558 Set 1 Run 4 Rep 2.xls
20040304-111738 Set 1 Run 4 Rep 3.xls
20040304-111917 Set 1 Run 4 Rep 4.xls
20040304-112102 Set 1 Run 4 Rep 5.xls
20040304-112246 Set 1 Run 4 Rep 6.xls
20040304-112433 Set 1 Run 4 Rep 7.xls
20040304-112619 Set 1 Run 4 Rep 8.xls
20040304-113245 Set 1 Run 5 Rep 1.xls
20040304-113431 Set 1 Run 5 Rep 2.xls
20040304-113623 Set 1 Run 5 Rep 3.xls
20040304-113819 Set 1 Run 5 Rep 4.xls
20040304-114011 Set 1 Run 5 Rep 5.xls
20040304-114207 Set 1 Run 5 Rep 6.xls
20040304-114404 Set 1 Run 5 Rep 7.xls
20040304-114603 Set 1 Run 5 Rep 8.xls
20040304-114802 Set 1 Run 6 Rep 1.xls
20040304-115008 Set 1 Run 6 Rep 2.xls
20040304-115216 Set 1 Run 6 Rep 3.xls
20040304-115426 Set 1 Run 6 Rep 4.xls
20040304-115636 Set 1 Run 6 Rep 5.xls
20040304-115851 Set 1 Run 6 Rep 6.xls
20040304-120107 Set 1 Run 6 Rep 7.xls
20040304-120326 Set 1 Run 6 Rep 8.xls
20040304-120544 Set 1 Run 7 Rep 1.xls
20040304-120805 Set 1 Run 7 Rep 2.xls
20040304-121025 Set 1 Run 7 Rep 3.xls
20040304-121247 Set 1 Run 7 Rep 4.xls
20040304-121511 Set 1 Run 7 Rep 5.xls
20040304-121737 Set 1 Run 7 Rep 6.xls
20040304-122003 Set 1 Run 7 Rep 7.xls
20040304-122231 Set 1 Run 7 Rep 8.xls
20040304-123344 Set 1 Run 8 Rep 1.xls
20040304-123610 Set 1 Run 8 Rep 2.xls
20040304-123837 Set 1 Run 8 Rep 3.xls
20040304-124108 Set 1 Run 8 Rep 4.xls
20040304-124339 Set 1 Run 8 Rep 5.xls
20040304-124615 Set 1 Run 8 Rep 6.xls
20040304-124851 Set 1 Run 8 Rep 7.xls
20040304-125129 Set 1 Run 8 Rep 8.xls
20040304-125659 Set 1 Run 9 Rep 1.xls
20040304-125945 Set 1 Run 9 Rep 2.xls
20040304-130231 Set 1 Run 9 Rep 3.xls
20040304-130808 Set 1 Run 9 Rep 4.xls
20040304-131055 Set 1 Run 9 Rep 5.xls
20040304-131343 Set 1 Run 9 Rep 6.xls
20040304-131631 Set 1 Run 9 Rep 7.xls
20040304-131921 Set 1 Run 9 Rep 8.xls
Peak Load
46.62883
45.436356
45.279495
50.681374
49.785763
49.105278
49.205067
44.488358
47.062386
48.801445
44.790001
48.562714
44.295425
47.351768
45.579613
44.489647
48.056576
42.174576
44.498764
41.299881
41.452736
43.930874
44.221973
41.120594
42.379463
46.715652
41.865509
41.553627
47.430531
41.915798
44.105892
43.67466
44.537437
44.807816
46.723701
42.515388
40.387726
40.945396
42.459793
43.998917
42.385052
50.552158
44.403934
41.823353
45.487274
45.608482
47.657516
47.201088
44.425724
43.235306
42.173809
44.286232
43.368046
41.627769
48.082321
46.882805
45.507973
41.00259
45.06089
43.363491
44.254284
45.699806
45.69239
42.352139
43.582703
49.186348
45.506218
47.338581
46.109573
44.55669
47.892349
45.624733
Net Max Load
42.19622471
41.00375071
40.84688971
46.24876871
45.35315771
44.67267271
44.77246171
40.05575271
42.62978071
44.36883971
40.35739571
44.13010871
39.86281971
42.91916271
41.14700771
40.05704171
43.62397071
37.74197071
40.06615871
36.86727571
37.02013071
39.49826871
39.78936771
36.68798871
37.94685771
42.28304671
37.43290371
37.12102171
42.99792571
37.48319271
39.67328671
39.24205471
40.10483171
40.37521071
42.29109571
38.08278271
35.95512071
36.51279071
38.02718771
39.56631171
37.95244671
46.11955271
39.97132871
37.39074771
41.05466871
41.17587671
43.22491071
42.76848271
39.99311871
38.80270071
37.74120371
39.85362671
38.93544071
37.19516371
43.64971571
42.45019971
41.07536771
36.56998471
40.62828471
38.93088571
39.82167871
41.26720071
41.25978471
37.91953371
39.15009771
44.75374271
41.07361271
42.90597571
41.67696771
40.12408471
43.45974371
41.19212771
182
Temp at Load
49.884712
49.936985
49.973155
49.772371
49.96387
49.982246
50.054485
50.088226
51.658398
51.695778
51.745364
51.559713
51.123327
51.096231
51.17956
51.206355
50.105169
49.97516
49.960869
49.957606
49.962041
50.061731
50.097313
50.055912
50.518533
50.521381
50.503502
50.593096
50.363425
50.376292
50.316962
50.506229
50.957099
51.223117
51.09591
50.86924
51.022172
51.004598
50.932648
50.915217
51.439566
51.171801
51.05388
50.916729
50.910329
50.807639
50.696154
50.641937
51.049432
51.008214
51.035702
51.00606
50.960645
50.942015
50.965229
50.867827
50.820927
50.47199
50.43103
50.381209
50.405053
50.390062
50.404541
50.394667
50.101435
50.1542
50.120294
50.079656
50.213548
50.190848
50.191024
50.181164
Run Avg Load
43.1437
Run Avg Temp
50.0
41.9340
51.4
38.9119
50.0
39.2725
50.5
38.8644
51.0
41.2073
51.0
39.8276
51.0
39.6841
50.5
41.7920
50.2
A.4 Sample Experimental Data Set, All Runs (continued)
Date-Time and Part ID
20040305-091132 Set 1 Run 10 Rep 1.xls
20040305-091217 Set 1 Run 10 Rep 2.xls
20040305-091301 Set 1 Run 10 Rep 3.xls
20040305-091349 Set 1 Run 10 Rep 4.xls
20040305-091433 Set 1 Run 10 Rep 5.xls
20040305-091521 Set 1 Run 10 Rep 6.xls
20040305-091609 Set 1 Run 10 Rep 7.xls
20040305-091658 Set 1 Run 10 Rep 8.xls
20040305-092152 Set 1 Run 11 Rep 1.xls
20040305-092303 Set 1 Run 11 Rep 2.xls
20040305-092421 Set 1 Run 11 Rep 3.xls
20040305-092535 Set 1 Run 11 Rep 4.xls
20040305-092653 Set 1 Run 11 Rep 5.xls
20040305-092816 Set 1 Run 11 Rep 6.xls
20040305-092938 Set 1 Run 11 Rep 7.xls
20040305-093058 Set 1 Run 11 Rep 8.xls
20040305-093332 Set 1 Run 12 Rep 1.xls
20040305-093449 Set 1 Run 12 Rep 2.xls
20040305-093609 Set 1 Run 12 Rep 3.xls
20040305-093730 Set 1 Run 12 Rep 4.xls
20040305-093852 Set 1 Run 12 Rep 5.xls
20040305-094011 Set 1 Run 12 Rep 6.xls
20040305-094129 Set 1 Run 12 Rep 7.xls
20040305-094249 Set 1 Run 12 Rep 8.xls
20040305-094533 Set 1 Run 13 Rep 1.xls
20040305-094652 Set 1 Run 13 Rep 2.xls
20040305-094807 Set 1 Run 13 Rep 3.xls
20040305-094919 Set 1 Run 13 Rep 4.xls
20040305-095033 Set 1 Run 13 Rep 5.xls
20040305-095150 Set 1 Run 13 Rep 6.xls
20040305-095307 Set 1 Run 13 Rep 7.xls
20040305-095422 Set 1 Run 13 Rep 8.xls
20040305-100925 Set 1 Run 14 Rep 1.xls
20040305-101047 Set 1 Run 14 Rep 2.xls
20040305-101219 Set 1 Run 14 Rep 3.xls
20040305-101340 Set 1 Run 14 Rep 4.xls
20040305-101504 Set 1 Run 14 Rep 5.xls
20040305-101632 Set 1 Run 14 Rep 6.xls
20040305-101806 Set 1 Run 14 Rep 7.xls
20040305-101943 Set 1 Run 14 Rep 8.xls
20040305-102447 Set 1 Run 15 Rep 1.xls
20040305-102617 Set 1 Run 15 Rep 2.xls
20040305-102747 Set 1 Run 15 Rep 3.xls
20040305-102919 Set 1 Run 15 Rep 4.xls
20040305-103050 Set 1 Run 15 Rep 5.xls
20040305-103224 Set 1 Run 15 Rep 6.xls
20040305-103358 Set 1 Run 15 Rep 7.xls
20040305-103532 Set 1 Run 15 Rep 8.xls
20040305-103715 Set 1 Run 16 Rep 1.xls
20040305-103855 Set 1 Run 16 Rep 2.xls
20040305-104036 Set 1 Run 16 Rep 3.xls
20040305-104221 Set 1 Run 16 Rep 4.xls
20040305-104407 Set 1 Run 16 Rep 5.xls
20040305-104556 Set 1 Run 16 Rep 6.xls
20040305-104735 Set 1 Run 16 Rep 7.xls
20040305-104916 Set 1 Run 16 Rep 8.xls
20040305-105845 Set 1 Run 17 Rep 1.xls
20040305-110035 Set 1 Run 17 Rep 2.xls
20040305-110236 Set 1 Run 17 Rep 3.xls
20040305-110443 Set 1 Run 17 Rep 4.xls
20040305-110647 Set 1 Run 17 Rep 5.xls
20040305-110854 Set 1 Run 17 Rep 6.xls
20040305-111102 Set 1 Run 17 Rep 7.xls
20040305-111310 Set 1 Run 17 Rep 8.xls
20040305-112848 Set 1 Run 18 Rep 1.xls
20040305-113056 Set 1 Run 18 Rep 2.xls
20040305-113310 Set 1 Run 18 Rep 3.xls
20040305-113522 Set 1 Run 18 Rep 4.xls
20040305-113735 Set 1 Run 18 Rep 5.xls
20040305-113954 Set 1 Run 18 Rep 6.xls
20040305-114208 Set 1 Run 18 Rep 7.xls
20040305-114427 Set 1 Run 18 Rep 8.xls
Peak Load
47.874283
46.989723
46.881813
45.736076
45.029144
46.069248
48.469646
42.096375
44.725861
44.587673
42.717182
46.337166
44.815929
45.658592
49.946003
48.852619
51.720848
45.144958
43.349007
51.384926
50.378567
41.728935
48.978333
50.947769
41.8722
43.347034
44.334808
45.608547
49.756042
45.773117
40.27779
48.326248
43.114841
41.929817
40.761803
44.940529
44.202026
45.258545
39.655685
43.660015
43.201645
44.945435
44.624767
41.508282
43.016941
40.242275
48.910042
43.095409
45.12886
45.127762
42.580982
41.723831
44.49881
37.737038
37.382568
47.043259
41.832466
40.071449
43.794182
41.143127
39.029778
46.641609
47.707619
47.969162
39.171555
48.006432
47.755684
39.958481
40.581181
47.601276
44.750404
42.72242
Net Max Load
43.44167771
42.55711771
42.44920771
41.30347071
40.59653871
41.63664271
44.03704071
37.66376971
40.29325571
40.15506771
38.28457671
41.90456071
40.38332371
41.22598671
45.51339771
44.42001371
47.28824271
40.71235271
38.91640171
46.95232071
45.94596171
37.29632971
44.54572771
46.51516371
37.43959471
38.91442871
39.90220271
41.17594171
45.32343671
41.34051171
35.84518471
43.89364271
38.68223571
37.49721171
36.32919771
40.50792371
39.76942071
40.82593971
35.22307971
39.22740971
38.76903971
40.51282971
40.19216171
37.07567671
38.58433571
35.80966971
44.47743671
38.66280371
40.69625471
40.69515671
38.14837671
37.29122571
40.06620471
33.30443271
32.94996271
42.61065371
37.39986071
35.63884371
39.36157671
36.71052171
34.59717271
42.20900371
43.27501371
43.53655671
34.73894971
43.57382671
43.32307871
35.52587571
36.14857571
43.16867071
40.31779871
38.28981471
183
Temp at Load
49.551394
49.373647
49.661218
49.484529
49.625281
49.707837
49.691736
49.545989
50.649783
50.050031
49.423058
49.432578
49.194352
48.901438
48.908986
49.013739
49.580786
49.472202
49.345601
49.332055
49.245178
49.557564
49.720498
49.871513
48.736814
48.674564
48.834702
49.227592
49.39116
49.28816
49.367914
49.558032
50.717812
50.769615
50.150435
50.775567
50.935811
50.822927
50.449383
50.229616
50.501971
50.609038
50.686845
50.747482
50.830825
50.822244
50.772311
50.923402
50.471526
50.413878
50.514117
50.402253
50.351899
50.22589
50.763444
50.892408
51.569878
51.321208
50.669801
50.379514
50.390362
50.404251
50.409625
50.403584
49.835092
50.093459
49.993261
50.178301
50.237394
50.096854
50.242349
50.195285
Run Avg Load
41.7107
Run Avg Temp
49.6
41.5225
49.4
43.5216
49.5
40.4794
49.1
38.5078
50.6
39.2605
50.7
38.2203
50.5
39.0911
50.7
39.3858
50.1
A.5 Sample Experimental Part Dimensions (2 Runs Shown)
Part No.
1-1-1
1-1-2
1-1-3
1-1-4
1-1-5
1-1-6
1-1-7
1-1-8
1-2-1
1-2-2
1-2-3
1-2-4
1-2-5
1-2-6
1-2-7
1-2-8
ID
pixels
1110
1124
1118.64
1114.4
1116
1114
1120.06
1115.81
1113
1118
1112.99
1115.81
1115
1118
1115.81
1112.99
1115
1117
1115.81
1118.64
1110
1123
1114.4
1117.23
1115
1120
1118.64
1112.99
1115
1119
1114.4
1111.57
1121
1119
1118.64
1120.06
1118
1121
1117.23
1122.89
1123
1121
1117.23
1122.89
1122
1119
1117.23
1120.06
1121
1125
1118.64
1118.64
1120
1121
1122.89
1118.64
1119
1119
1115.81
1122.89
1123
1118
1120.6
1120.06
Average
pixels
1116.76
ID
Run Avg Run Avg
inches
inches
m
1.151299 1.150522 0.029223
1116.468
1.150997
1114.95
1.149433
1115.45
1.149948
1116.613
1.151147
1116.158
1.150678
1116.658
1.151193
1114.993
1.149477
1119.675
1.154304 1.154781 0.029331
1119.78
1.154412
1121.03
1.155701
1119.573
1.154198
1120.82
1.155485
1120.633
1.155291
1119.175
1.153789
1120.415
1.155067
OD
pixels
1197
1204
1199.25
1195.01
1202
1196
1199.25
1197.84
1195
1201
1197.84
1200.67
1198
1202
1200.67
1196.42
1195
1203
1197.84
1199.25
1195
1203
1197.84
1199.25
1198
1204
1200.67
1196.42
1199
1203
1199.25
1195.01
1208
1206
1206.32
1204.91
1204
1206
1203.5
1207.74
1210
1204
1204.91
1206.32
1209
1206
1204.91
1204.91
1205
1209
1206.32
1203.5
1205
1207
1207.74
1203.5
1206
1204
1202.08
1209.15
1209
1205
1206.32
1203.5
184
Average
OD
Run Avg Run Avg rel ∆ dia thickness
pixels
inches
inches
m
m/m
m
1198.815 1.235892 1.236066 0.031396 0.016648 0.001086
1198.773 1.235848
1198.628 1.235698
1199.273 1.236363
1198.773 1.235848
1198.773 1.235848
1199.773 1.236879
1199.065 1.236149
1206.308 1.243616 1.243190 0.031577 0.013008 0.001123
1205.31 1.242588
1206.308 1.243616
1206.205 1.243510
1205.955 1.243253
1205.81 1.243103
1205.308 1.242585
1205.955 1.243253
A.6 Experimental Data and Calculated Coefficient of Friction (Menges), Run Average
-1
-1
-1
-1
-1
-1
-1
-1
-1
1
1
1
1
1
1
1
1
1
-1
-1
-1
0
0
0
1
1
1
-1
-1
-1
0
0
0
1
1
1
-1
0
1
-1
0
1
-1
0
1
-1
0
1
-1
0
1
-1
0
1
Modulus at
Ejection
Ejection
Ejection
o
Force (N) Temp ( C) Temp (Pa)
177.1534
51.0 69000000
183.2899
51.0 69000000
186.5225
51.4 65400000
176.5148
50.5 71500000
185.891
50.2 73000000
172.8689
51.0 69000000
191.9032
50.0 74000000
174.6842
50.5 71500000
173.0801
50.0 74000000
184.6922
49.4 78800000
193.5839
49.5 78000000
173.8771
50.7 70500000
171.2827
50.6 71000000
174.6307
50.7 70500000
175.1881
50.1 73500000
170.0038
50.5 71500000
180.0522
49.1 81200000
185.5291
49.6 77200000
P-20
HIPS
-1
-1
-1
-1
-1
-1
-1
-1
-1
1
1
1
1
1
1
1
1
1
-1
-1
-1
0
0
0
1
1
1
-1
-1
-1
0
0
0
1
1
1
-1
0
1
-1
0
1
-1
0
1
-1
0
1
-1
0
1
-1
0
1
343.9689
376.3673
401.4804
346.1149
385.5599
408.2617
384.5955
381.943
403.7004
376.9719
395.2322
393.3786
369.1949
390.8133
391.7333
351.6246
394.7653
424.4616
51.0
51.1
51.1
50.8
50.8
51.0
50.4
50.2
50.2
50.5
51.1
51.5
50.6
50.8
50.9
50.4
49.6
48.9
580000000
573000000
573000000
592400000
592400000
580000000
617200000
629600000
629600000
611000000
573000000
545000000
604800000
592400000
586200000
617200000
673200000
728000000
0.003089
0.003377
0.002935
0.003301
0.003565
0.002989
0.003788
0.002896
0.00354
0.003199
0.003444
0.001107
0.002796
0.002337
0.001826
0.002424
0.002791
0.003016
0.001138
0.00114
0.001144
0.001134
0.001146
0.001133
0.001139
0.001131
0.00115
0.001124
0.001138
0.00114
0.001139
0.001137
0.001146
0.001141
0.001137
0.001127
0.553423
0.559850
0.684961
0.511977
0.522718
0.681704
0.473993
0.607589
0.516881
0.563075
0.577328
1.876760
0.628923
0.814746
1.047471
0.675713
0.606239
0.562590
ST-100
HDPE
-1
-1
-1
-1
-1
-1
-1
-1
-1
1
1
1
1
1
1
1
1
1
-1
-1
-1
0
0
0
1
1
1
-1
-1
-1
0
0
0
1
1
1
-1
0
1
-1
0
1
-1
0
1
-1
0
1
-1
0
1
-1
0
1
182.3003
190.0239
209.7913
177.4434
196.6074
194.3292
196.1544
201.6826
208.7471
173.9534
185.8118
184.4146
172.1205
178.8341
180.8674
170.97
186.8184
184.1281
50.9
50.9
50.3
50.3
50.2
51.4
49.1
50.2
49.6
50.8
50
50.6
50.4
50.2
50.2
50.2
49.3
48.6
69500000
69500000
72500000
72500000
73000000
65400000
81200000
73000000
77200000
70000000
74000000
71000000
72000000
73000000
73000000
73000000
79600000
85200000
0.005901
0.006708
0.007395
0.006204
0.006976
0.007264
0.012154
0.007462
0.006877
0.010785
0.011344
0.010469
0.010469
0.010979
0.010258
0.011443
0.01105
0.010809
0.00104
0.001108
0.001116
0.001045
0.001109
0.001122
0.001093
0.001105
0.001113
0.00103
0.001103
0.001122
0.001031
0.001109
0.001119
0.001055
0.001104
0.001114
1.402205
1.207056
1.150452
1.238469
1.142516
1.196203
0.596783
1.099385
1.159108
0.733839
0.658192
0.725377
0.726403
0.659969
0.707842
0.636309
0.631115
0.589072
Insert/
Thermoplastic
P-20
HDPE
Tpack
Tcool
Ppack
185
Relative
Change in
Dia (m/m)
0.013299
0.012021
0.013008
0.013940
0.012097
0.011691
0.016648
0.011819
0.012125
0.012311
0.011043
0.009859
0.012184
0.010684
0.008880
0.012579
0.011084
0.010279
Thickness
(m)
0.001075
0.001109
0.001123
0.00108
0.001114
0.001121
0.001086
0.00111
0.001127
0.00107
0.001118
0.001123
0.001066
0.001114
0.001125
0.001074
0.001115
0.001126
CoF
0.589265
0.653811
0.640622
0.537958
0.619705
0.627208
0.470372
0.610927
0.561318
0.583684
0.659497
0.730678
0.609197
0.682887
0.782797
0.577123
0.588728
0.681465
A.6 Experimental Data and Calculated Coefficient of Friction, Run Average (continued)
Insert/
Thermoplastic
ST-100
HIPS
Tpack
Tcool
Ppack
Modulus at Relative
Ejection
Ejection
Ejection
Change in Thickness
o
Force (N) Temp ( C) Temp (Pa) Dia (m/m)
(m)
CoF
366.2944
49.9 649800000
0.00575 0.001138 0.282649
389.5081
50 642000000 0.005295 0.001143 0.328955
375.2821
50.4 617200000 0.005225 0.001135
0.33628
375.7127
49.6 673200000 0.005929 0.001146 0.269402
393.5372
49.8 657600000 0.005626 0.001147 0.304293
394.4723
50.5 611000000 0.005295 0.001149 0.348226
366.3106
49.5 681000000 0.006008 0.001139 0.257966
393.5601
49.6 673200000 0.004871 0.001097 0.358898
398.8342
49.6 673200000 0.005179 0.001141 0.328905
363.8594
50.6 604800000
0.00307 0.001153 0.557383
378.138
50.4 617200000 0.002821 0.001151 0.618891
388.0728
49.6 673200000 0.002605 0.001144
0.63481
369.5402
49.7 665400000 0.004337
0.00114 0.368619
360.5764
50.1 635800000 0.003636 0.001152 0.444142
374.3817
49.8 657600000 0.002766 0.001145 0.589803
340.6728
50 642000000 0.004431 0.001155 0.340328
370.5696
49.4 688800000 0.004128 0.001186 0.360363
399.8762
49.8 657600000 0.004506 0.001147 0.385893
-1
-1
-1
-1
-1
-1
-1
-1
-1
1
1
1
1
1
1
1
1
1
-1
-1
-1
0
0
0
1
1
1
-1
-1
-1
0
0
0
1
1
1
-1
0
1
-1
0
1
-1
0
1
-1
0
1
-1
0
1
-1
0
1
SL 5170
HDPE
-1
-1
-1
1
-1 239.0608
-1 193.2119
53.1
51.2
SL/P-20
HDPE
-1
-1
-1
-1
1
1
1
1
-1
-1
1
1
-1
-1
1
1
-1
1
-1
1
-1
1
-1
1
42.2
42
40
40.7
41.3
42
40.5
40.9
SL 5170
HIPS
-1
-1
-1
-1
1
1
-1 1334.275
-1 1136.124
1 1512.254
55.2 304000000 0.003412 0.001311 3.217736
52.8 462000000
0.00406 0.001309 1.518087
54.1 391000000 0.001753 0.001308 5.533213
SL/P-20
HIPS
-1
-1
-1
-1
1
1
1
1
-1
-1
1
1
-1
-1
1
1
-1
1
-1
1
-1
1
-1
1
40.7
42.6
38.1
41.1
42.2
42.7
40.2
42.1
274.2134
299.6501
258.7609
278.1758
313.3349
297.3824
321.087
317.9256
695.756
826.2838
610.1186
845.0178
770.2405
939.1286
702.2841
892.146
186
58800000 0.009477 0.001138 1.237096
67200000 0.009388 0.001137 0.883788
151200000
154000000
178000000
168900000
161700000
154000000
171500000
166300000
1433300000
1262000000
1675000000
1391000000
1294000000
1254000000
1488800000
1302000000
0.013628
0.01166
0.014343
0.010554
0.011732
0.007237
0.01081
0.006996
0.004444
0.002462
0.00579
0.002444
0.002077
0.001119
0.002788
0.000735
0.001197
0.001205
0.001202
0.0012
0.001189
0.001204
0.001187
0.001206
0.00123
0.001212
0.001217
0.001204
0.001206
0.001202
0.001211
0.001194
0.364731
0.454231
0.276662
0.426762
0.45566
0.727173
0.478619
0.743436
0.291433
0.720073
0.169589
0.677233
0.779572
1.826759
0.458249
2.561299
A.7 Analysis of Variance Tables by Set
General Linear Model: P-20 and HDPE
EF Set 1 versus Packing Time, Cooling Time, Packing Pressure
Factor
Packing
Cooling
Packing
Type Levels Values
fixed
2 2 6
fixed
3 5 10 15
fixed
3 0 5 10
Analysis of Variance for EF Set 1, using Adjusted SS for Tests
Source
Packing
Cooling
Packing
Packing*Cooling
Packing*Packing
Cooling*Packing
Packing*Cooling*Packing
Error
Total
DF
1
2
2
2
2
4
4
126
143
Seq SS
3.836
149.135
13.938
36.255
3.204
40.352
106.985
899.771
1253.478
Adj SS
3.836
149.135
13.938
36.255
3.204
40.352
106.985
899.771
Adj MS
3.836
74.568
6.969
18.128
1.602
10.088
26.746
7.141
F
0.54
10.44
0.98
2.54
0.22
1.41
3.75
P
0.465
0.000
0.380
0.083
0.799
0.234
0.006
F
8.84
49.29
5.30
7.32
11.79
16.03
20.01
P
0.004
0.000
0.006
0.001
0.000
0.000
0.000
General Linear Model: P-20 and HIPS
EF Set 2 versus Packing Time, Cooling Time, Packing Pressure
(One Outlier Removed)
Factor
Packing
Cooling
Packing
Type Levels Values
fixed
2 2 6
fixed
3 5 10 15
fixed
3 0 5 10
Analysis of Variance for EF Set 2, using Adjusted SS for Tests
Source
Packing
Cooling
Packing
Packing*Cooling
Packing*Packing
Cooling*Packing
Packing*Cooling*Packing
Error
Total
DF
1
2
2
2
2
4
4
125
142
Seq SS
89.91
1059.86
119.56
166.61
245.27
689.01
863.76
1349.12
4583.09
187
Adj SS
95.44
1063.93
114.30
158.03
254.54
691.83
863.76
1349.12
Adj MS
95.44
531.97
57.15
79.01
127.27
172.96
215.94
10.79
A.7 Analysis of Variance Tables by Set (continued)
General Linear Model: ST-100 and HDPE
EF Set 3 versus Packing Time, Cooling Time, Packing Pressure
Factor
Packing
Cooling
Packing
Type Levels Values
fixed
2 2 6
fixed
3 5 10 15
fixed
3 0 5 10
Analysis of Variance for EF Set 3, using Adjusted SS for Tests
Source
Packing
Cooling
Packing
Packing*Cooling
Packing*Packing
Cooling*Packing
Packing*Cooling*Packing
Error
Total
DF
1
2
2
2
2
4
4
126
143
Seq SS
435.039
146.848
28.902
86.973
6.791
113.575
122.030
1064.275
2004.434
Adj SS
435.039
146.848
28.902
86.973
6.791
113.575
122.030
1064.275
Adj MS
435.039
73.424
14.451
43.487
3.395
28.394
30.508
8.447
F
51.50
8.69
1.71
5.15
0.40
3.36
3.61
P
0.000
0.000
0.185
0.007
0.670
0.012
0.008
F
17.51
2.62
3.76
6.97
3.75
9.00
6.70
P
0.000
0.076
0.026
0.001
0.026
0.000
0.000
General Linear Model: ST-100 and HIPS
EF Set 4 versus Packing Time, Cooling Time, Packing Pressure
Factor
Packing
Cooling
Packing
Type Levels Values
fixed
2 2 6
fixed
3 5 10 15
fixed
3 0 5 10
Analysis of Variance for EF Set 4, using Adjusted SS for Tests
Source
Packing
Cooling
Packing
Packing*Cooling
Packing*Packing
Cooling*Packing
Packing*Cooling*Packing
Error
Total
DF
1
2
2
2
2
4
4
126
143
Seq SS
261.17
78.28
112.18
207.85
111.73
536.86
399.58
1879.43
3587.07
188
Adj SS
261.17
78.28
112.18
207.85
111.73
536.86
399.58
1879.43
Adj MS
261.17
39.14
56.09
103.92
55.87
134.21
99.89
14.92
A.7 Analysis of Variance Tables by Set (continued)
Fractional Factorial Fit: SL 5170/P-20 and HDPE
EF Set 5b versus Packing Time, Cooling Time, Packing Pressure
Estimated Effects and Coefficients for EF (coded units)
Term
Constant
Packing
Cooling
Packing
Packing*Cooling
Packing*Packing
Cooling*Packing
Packing*Cooling*Packing
Effect
7.809
-0.485
1.447
3.666
-3.595
0.380
1.057
Coef
66.337
3.904
-0.243
0.723
1.833
-1.798
0.190
0.529
SE Coef
0.9830
0.9830
0.9830
0.9830
0.9830
0.9830
0.9830
0.9830
T
67.49
3.97
-0.25
0.74
1.86
-1.83
0.19
0.54
P
0.000
0.000
0.807
0.467
0.071
0.077
0.848
0.594
Analysis of Variance for EF (coded units)
Source
Main Effects
2-Way Interactions
3-Way Interactions
Residual Error
Pure Error
Total
DF
3
3
1
32
32
39
Seq SS
633.01
265.08
11.18
1236.78
1236.78
2146.06
Adj SS
633.01
265.08
11.18
1236.78
1236.78
Adj MS
211.00
88.36
11.18
38.65
38.65
F
5.46
2.29
0.29
P
0.004
0.098
0.594
Fractional Factorial Fit: SL 5170/P-20 and HIPS
EF Set 6a versus Packing Time, Cooling Time, Packing Pressure
Estimated Effects and Coefficients for EF (coded units)
Term
Constant
Packing
Cooling
Packing
Packing*Cooling
Packing*Packing
Cooling*Packing
Packing*Cooling*Packing
Effect
18.358
-10.220
40.702
-2.700
-0.375
7.045
-4.687
Coef
176.511
9.179
-5.110
20.351
-1.350
-0.188
3.523
-2.344
SE Coef
2.401
2.401
2.401
2.401
2.401
2.401
2.401
2.401
T
73.50
3.82
-2.13
8.47
-0.56
-0.08
1.47
-0.98
P
0.000
0.001
0.041
0.000
0.578
0.938
0.152
0.336
Analysis of Variance for EF (coded units)
Source
Main Effects
2-Way Interactions
3-Way Interactions
Residual Error
Pure Error
Total
DF
3
3
1
32
32
39
Seq SS
20981.5
570.6
219.7
7381.2
7381.2
29153.1
Adj SS
20981.5
570.6
219.7
7381.2
7381.2
189
Adj MS
6993.8
190.2
219.7
230.7
230.7
F
30.32
0.82
0.95
P
0.000
0.490
0.336
APPENDIX B
MOLD AND CANISTER DRAWINGS
190
B.1 Part Drawing
191
B.2 Mold Insert Drawings
192
B.2 Mold Insert Drawings (continued)
193
B.2 Mold Insert Drawings (continued)
194
B.2 Mold Insert Drawings (continued)
195
B.2 Mold Insert Drawings (continued)
196
B.2 Mold Insert Drawings (continued)
197
B.2 Mold Insert Drawings (continued)
198
B.2 Mold Insert Drawings (continued)
199
B.2 Mold Insert Drawings (continued)
200
B.2 Mold Insert Drawings (continued)
201
B.2 Mold Insert Drawings (continued)
202
B.2 Mold Insert Drawings (continued)
203
B.2 Mold Insert Drawings (continued)
204
B.3 Mold Assembly Drawings
205
B.3 Mold Assembly Drawings (continued)
206