Submitted by

Er. Devavnshi Kapadia

Submitted to
Institute of Polymer
Injection Moulding and
Process Automation

Supervisor

Shrinkage and Warpage Univ.-Prof. DI. Dr.

Georg Steinbichler

in Polymer Injection Co-Supervisor

Thomas Mitterlehner,
MSc.

Moulding: Models and September 2019

Simulation

Master Thesis
to obtain the academic degree of
Diplom-Ingenieurin
in the Master’s Programme
Managment in Polymer Technologies

JOHANNES KEPLER
UNIVERSITÄT LINZ
Altenbergerstraße 69
4040 Linz, Österreich
www.jku.at
DVR 0093696
Acknowledgements i

Acknowledgements
I would like to express my gratitude to the head of the Institute of Polymer Injection Mould-
ing and Process Automation (IPIM) Univ. - Prof. DI Dr. Georg Steinbichler for having been
given the opportunity to work on this important project and for all his valuable teachings all
through the course of my master program. I am also very grateful to Thomas Mitterlehner,
MSc for helping me a great deal in the practical work and showing me the correct path and
the helpful feedback for the work that needed to be done. I am very thankful to all the people
from the institute who supported me without whom this would have been an impossible task
to complete, so a big thank you to Petra Zimmerhansl for co-ordinating all the meetings and
helping out with the administrative work; Ing. Alfred Mayr for imparting a lot of technical
knowledge and helping with the practical work, this project would not have been possible
without your infinite support and finally DI Dominik Altmann for your important contribu-
tion.

I am also deeply thankful to the head of Institute of Polymer Product Engineering (IPPE)
Univ. - Prof. Dr. Zoltan Major for having been allowed to work with and taught some
important software all through the course and been given the chance to work freely in the
IPPE laboratory. Special thanks to DI Philipp Stelzer for his timely and much needed doubt
clearing sessions and also to DI Florian Kiehas for the endless support and help he extended.

I sincerely thank all the other teaching and non-teaching members for their knowledge and
support all through the course.

I express a deep sense of appreciation to my family for their blessings, love and support and
my friends for the continuous motivation. You have all together helped me in ways to move
ahead in life always.
Statutory Declaration ii

Statutory Declaration
I hereby declare that the thesis submitted is my own unaided work, that I have not used
other than the sources indicated, and that all direct and indirect sources are acknowledged
as references. This printed thesis is identical with the electronic version submitted.

Place, Date Signature

Abstract iii

Abstract
Today the world is moving towards zero waste, sustainability and circular economy. Plastic
waste is a critical issue tackled on a global scale. It is everybody’s responsibility to help
contribute to this ever raging issue.

Injection moulding is the most widely used technique for the production of parts made out
of polymers. Although very versatile there are various shortcomings to this process which
need to be worked upon to get an optimized process. Two of the most common and critical
defects of this process are shrinkage and warpage, which require a trial and error methods
to reach the most optimum process settings. This means investing a lot of capital, labour,
time, material, energy and efforts. While optimizing the process, there is plenty of waste
generated along the process. This calls for a dire need of finding a solution that reduces not
only efforts and resources being utilized but also reduces the usage of energy being put in
and the waste generated thereby.

The aim of this thesis is to assess and analyse the efficiency of the simulations, to find out
how close or how accurate a simulation study could be in order to reduce utilization of re-
sources, increase efficiency and reduce wastage by getting a particular process setting right.
In this study, first a practical experiment on the machine was made, to have observations of
the behaviour of the parts; after which the exact same parameters from the practical inputs
and outputs were used for simulations. This gives us a comparable data.

The materials used were all unfilled PP, ABS and PA 6 with a simple plate geometry with
a sprue gate in a single cavity cold runner mould. Each material was processed at their
respective process parameters, and six samples were produced for measurement. A filling
and gate freezing study was carried out before setting the process.

As expected it was observed that the semi crystalline materials - PP and PA shrink more than
the amorphous material ABS. The results obtained by the simulation were very comparable
to the experimental results and were about 70% accurate. Experimental shrinkage values
were in the range of 0.75 − 0.9% for ABS, 1.5 − 1.7% for PP and 1.5 − 1.8% for PA; whereas,
the results predicted by the simulation were 0.6% for ABS, 1.4 − 1.5% for PP and 1.2 − 1.5%
for PA.
Table of contents iv

Acknowledgements i

Statutory Declaration ii

Abstract iii

Table of contents v

Abbreviations vi

1 Introduction 1

2 General Background 2
2.1 Injection moulding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
2.2 Modelling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
2.3 PvT behaviour . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
2.4 Shrinkage and Warpage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
2.4.1 Inside the mould . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
2.4.2 Residual stresses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
2.5 Sensors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
2.5.1 Kistler sensors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
2.5.2 Cavity Eye sensors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
2.6 Simulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
2.6.1 Shrinkage and warpage . . . . . . . . . . . . . . . . . . . . . . . . . . . 16

3 Experiments 22
3.1 Specifications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
3.2 Procedure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
3.2.1 Material Testing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
3.2.2 Injection moulding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
3.2.3 Simulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31

4 Results and discussion 33

4.1 Pressure curves . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
4.2 Shrinkage and warpage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39

5 Conclusion 42

6 Summary 43

References 44
Table of contents v

Appendix 47
6.1 Datasheet For HG385MO . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
6.2 Datasheet for NOVODURP2H-ATDS . . . . . . . . . . . . . . . . . . . . . . . 50
6.3 Datasheet for Durethan B30S . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
6.4 Datasheet for Kistler pressure sensor . . . . . . . . . . . . . . . . . . . . . . . 56
6.5 Datasheet for CavityEye pressure cell . . . . . . . . . . . . . . . . . . . . . . . 58
Abbreviations vi

Abbreviations

Abbreviation Expansion

PP Polypropylene

ABS Acrylonitrile butadiene styrene

PA6 Polyamide 6

PvT Pressure - Volume - Temperature

MSDS Material Safety Data Sheet

CRIMS Corrected Residual In-Mould Stress

1 Introduction 1

1 Introduction
Shrinkage and warpage lie among the top defects in the injection moulding process. These
are difficult to tackle and control, and need a lot of time and energy to be spent for opti-
mization of the process and for avoiding the defects. There are many factors and parameters
that need adjustments to maintain shrinkage and warpage of injection moulded parts. These
factors are mostly interlinked which may cause a compromise in the desired properties of
the final product. The process needs to be optimized to control the unavoidable shrinkage
which goes hand-in-hand with polymer processing. Process optimization in itself is a long
and tedious task with usage of a huge amount of capital, energy, time and other resources.
Not only that, a lot of unusable parts are produced during optimization by trial and error
thus creating a lot of mostly "non-degradable polymer waste".

An attempt to improve process optimization, a shrinkage study is aimed at with simulation

runs to make the entire optimization more efficient. In this study, three unfilled materials
PP, ABS and PA were used. The design of the specimen for this project was a simple thin
plate with a sprue gate which was moulded from a single cavity cold runner mould on an
injection moulding machine. Initially the materials were preconditioned to remove moisture
if required and then used for injection moulding the specimen samples. At the beginning
of injection moulding, a filling and packing study was done to ensure complete filling of the
cavity. After achieving a stable process for each material, the initial ten parts were discarded
after which six samples were taken for further measurement. The process inputs and outputs
were carefully noted along with all other observations. Two different types of sensors were
fitted into the mould to measure the cavity pressures additional to the other sensors present
on the machine for measurement of pressure, temperature, speed, position etc.. The sam-
ple dimensions were measured manually after 48 hours of processing in order to calculate
shrinkage. Warpage was a qualitative observation only made visually.

The next step was to use all the data input in the injection moulding process as well as some
other outputs received from the practical experiment on the moulding machine. The part
and mould design used for the simulation and moulding are exactly the same. The simulation
software used was Moldex 3D Designer R15.0 for the preparation and meshing of the design
input. The Moldex 3D R15.0 was used to run simulations to find out the predicted shrinkage
and warpage results using the same data as for injection moulding.

To compare shrinkage and warpage data predicted by the software, the process curves from
both injection moulding and simulation were compared as well. Final results were then
plotted into graphs with relevant deductions. The results did not match 100%, however were
quite close and are usable. There is a continuous development in this area of production and
simulation as it seems quite promising and an ever booming technology, as an attempt to
make it easier for the generations to come.
2 General Background 2

2 General Background

Figure 2.1: Global production of plastics in million metric tons [Garside 2018]

The last century has seen a very fast growth [Garside 2018] of polymers and plastics and
their penetration in every market. Even though a century old, however as compared to other
materials plastics are relatively new and very extensively used, outperforming the utilization
of metals. Products manufactured from plastics are benefiting various fields like packaging,
textiles, commodities, construction among many others as shown in figure 2.2.

Figure 2.2: Sector-wise growth of polymers in tons per year [Geyer; Jambeck, and Law 2017]
2 General Background 3

2.1 Injection moulding

Injection moulding is the most widely used technique for the production of parts made out
of polymers having different complexities and precision. It is a process where polymer melt
is injected in a hot or a cold mould with single or multiple cavities, held under pressure
and cooled up to a temperature suitable enough for ejection, thereby replicating the cavity
in which it is formed. The basic steps in injection moulding include plasticization of the
material inside the injection unit, injection and solidification of the material and ejection by
mould opening. This method of manufacturing is particularly widely used as it is capable of
producing functional, intricate and complex components which may have tight tolerances or
thin cross sections, repeatedly in a single relatively short cycle which is accurate.

Figure 2.3: Injection moulding machine [Frizelle 2017]

All the stages of injection moulding are shown in 2.4.

Figure 2.4: Injection moulding cycle [Zhou 2013]

Although this process is very versatile there are various types of defects which occur in the
part due to many reasons pertaining to the part design, mould design, material properties
and processing conditions which need to be worked upon to get an optimized process. The
target is not only to produce components in bulk quantities but also with good quality and
in the most affordable way, to stay put to the competition posed in the industry. To sur-
vive the crucial competition, it is essential to have a deep understanding of the mechanisms
and laws of the process to have the best possible quality of output. There is quite little
know-how about what actually takes place in the moulding process and there are enormous
2 General Background 4

possibilities for development in this area which must be regarded in co-operation with experts.

Quality influencers

The development of any product initiates with the design of the part and the selection of
the material for that part. This is followed by the designing of the mould, manufacturing,
performance control and testing in the sequence stated. The manufacturing technique being
the most important of them all as it has the biggest impact on the quality of the production.
The factors that influence the quality are part and mould design, process parameters and the
machine itself.

1. Design - part and mould

Design plays a very important role as the part must meet the requirements structurally,
functionally, aesthetically, with respect to manufacturing and also cost wise. It is
expected out of the designer to be immensely efficient to be able to face the challenge
of meeting the requirements by not complicating too much, affecting cost and the ability
to manufacture.
A few criteria to be considered while designing a mould or a part: dimensions of the
part taking into account shrinkage, complete cavity filling, sufficient and even cooling,
adequate venting, withstanding high clamping pressures and cyclic loading internally
due to pressures from injection and machinability.
Selecting the correct material for the application is also a vital part of designing. Poly-
mers are known as viscoelastic materials. This viscous characteristic of the melt is
essential to injection moulding. Owing to rheology, where flow and deformation of the
melt under the processing conditions is dealt with, the polymer while being processed
influences the end performance based on the morphology and the micro-structure dur-
ing injection moulding. Thus, crystalline and amorphous materials act differently as
the micro-structures vary considerably.

2. Process parameters
For a stable production with high quality output the following points are crucial:

• optimizing the plasticization by adjusting the screw speed and the back pressure
to inject the required amount of melt uniformly;
• the injection speed affecting pressure drop, shear rate and temperature variation
after filling;
• optimizing switch-over of filling phase to packing phase can avoid defects like sink
marks, helps maintain correct dimensions and prevent underweight;
• cooling time;
• temperature of the barrel and mould;
• holding pressure

A process window must be developed to optimize the process parameters to have a

stable process.
2 General Background 5

3. Machine
It is of utmost importance that the right size of the mould is set up on the correct
machine to get the best output. A very small mould on a high tonnage machine and
vice-versa will not give the best results and may lead to higher process defects and
may also cause damages to the mould and the machine. The machine type and the
size depend upon the size and volume of the component. The injection moulding
machine with a reciprocating screw amongst others is generally used because of better
performance.

Various defects like sink marks, air entrapment, short-shot, burning etc. are seen during the
entire process. An example of the defects is shown in the figures 2.5, 2.6 and 2.7.

Figure 2.5: Defects observed for injection moulded parts - Sink marks and air entrapment [Zhou
2013]

Figure 2.6: Defect - Short-shot observed in injection moulding [Li 2018]

Figure 2.7: Defect - Warpage observed in the blue crate against a non warped orange crate
[Macplas 2017]
2 General Background 6

Two of the most common and critical defects that occur are shrinkage and warpage shown
in figure 2.7, which are mainly influenced by parameters like packing, cooling, geometry and
so on and therefore require a trial and error method to reach the most optimum process set-
tings in actual production and development practices. This means investing a lot of capital,
labour, time, material, energy and efforts. This calls for a dire need of finding a solution that
reduces the efforts and energy being put in.

A complete overview of the process and its components are shown in the figure 2.8. There
are correlations among all the factors for example, a mould designed for a specific material
may not be used with another material due to difference in flow properties and shrinkage
values. It would hence be a good idea to combine simulations and computer modelling to
optimize processes, attempting to solve the problems that would otherwise be solved by trial
and error or intuition.

Figure 2.8: Fishbone diagram of quality influencers for injection moulding [Zhou 2013]

In this study the focus will be on the defects shrinkage and warpage where the parts are
produced by injection moulding and then the moulding parameters are used for simulations
to have a comparison of results.

2.2 Modelling
It is undeniable that, when properly utilised, technologies like computer simulations and
modelling increase efficiency of a process. It is usually the case that after a mould is manu-
factured, several iterations and corrections are made to achieve the desired properties, shape
and dimensions. This is because the first few parts manufactured in injection moulding do
not meet the specifications. Thus, it is highly preferable that computer simulations be used
2 General Background 7

to understand the basic behavioural attributes for moulds and materials, which would usu-
ally bring down the development cycle time considerably. The diagram 2.9 depicts the basic
structure of computer modelling which helps improve product quality.

Figure 2.9: Basic structure for computer modelling for injection moulding [Zhou 2013]

A few attempts have been made to develop mathematical models by for various geometries
since many years. A model developed to understand the behaviour of the melt inside the
cavity, with energy, motion and continuity equations, for every stage in the injection mould-
ing process viz. filling stage, packing stage and holding stage; while applying boundary
conditions; with considerations

• the polymer melt is non-Newtonian and that temperature has an effect on the viscosity
and the density;

• the latent heat;

• change in solid and melt properties to get results at different locations and different
times in the mould cavity to measure flow rate, velocity, temperature and pressure
changes for all stages of the process.

This model was tested on injection moulding to get results which proved the model, however
some issues which arose while applying the model. The assumptions of viscoelasticity and
end effects considered in the model did not apply well at the entry of material in the cavity.
[Kamal and Kenig 1972b]

A study of a disk-shaped cavity was carried out to find out pressures and time of filling the
cavity using the power law model for various materials to clarify the non-isothermal and
transient issues. [Wu; Huang, and Gogos 1974] Another study was made to find out the
clamping force and injection pressure using the same model. [Stevenson 1978]

An analysis was developed to find out the distribution of pressure, velocity and temperature
which when associated with the analysis for filling of the cavity was able to model the process
of injection moulding and also helped in design of mould, to assess mouldability, geometry,
2 General Background 8

conditions for operation and problem solving in case of moulding. [Lord and Williams
1975]

Mathematical models were proposed to assess the cooling time. [Busch; Rosato, and
Field 1988] Additionally, systems on the computer to evaluate the process settings, an an-
alytical model was for calculating clamp force, filling pressure, temperature, shear rate and
shear stress at different points in the cavity and time was proposed. [Tan and Yuen 1996]
Several models for simulation of behaviour of the injection moulding process were also devel-
oped. [Zhou 2013] A consolidated model was developed based on finite element analysis to
simulate the stages of filling and post-filling, using Hele-Shaw generalizations for flow of com-
pressible viscous fluids for non-isothermal conditions. [Chiang; Hieber, and Wang 1991]
This path was further extended by others for simulations of residual stresses, mould cooling
[Himasekhar; Lottey, and Wang 1992] as well as warpage and shrinkage. [Santhanam
et al. 1991 ]

Some commercial simulation softwares are now available in the recent years like Moldex3D
and Moldflow, which when used could be advantageous in eliminating the trial and error
methods of production, in understanding material behaviour and improving design of com-
ponents and moulds. It was seen that when software-based approach was compared with
the manual trial and error method, a better quality of parts were manufactured using the
software-based approach with more consistency in dimensions and weight of parts. [Nirkhe
and Barry 2003] Even though such mathematical models and software are quite effective,
care should be taken as to not overestimate their capability. The simplifications and as-
sumptions tend to cause variations between the actual versus the results obtained from the
models.

2.3 PvT behaviour

Different polymeric materials have their individual rheology and flow properties; however
polymers are viscoelastic in nature. Viscoelasticity in polymers is caused due to two com-
ponents of flow: viscous - where there is dissipation of energy which causes deformation and
elastic - in which there is storage of that energy. Polymers respond to stress viscoelastic be-
haviour which means that strain takes time to accommodate the stress which is applied. The
relaxation time only differs depending on material characteristics and processing methods.

As polymers are materials that are compressible in character, under loading they tend to
compress more than other materials. Density has a high dependence on the pressure and the
temperature which is known as PvT or thermodynamic behaviour of the material. This be-
haviour plays an important role during the process of injection moulding, most importantly
in the packing stage, which particularly has an influence on phenomenon like shrinkage as
well as warpage.

The figures 2.10 and 2.11 show the PvT behaviours of amorphous and semi-crystalline ma-
terials respectively. It can be seen that both type of materials behave differently. For both,
in the melting temperature range, there is a linear relationship between the specific volume
and temperature and there is a decrease in the specific volume with increase in pressure.
2 General Background 9

However, the range in which both materials are solid the behaviour differs; specific volume
depends linearly on temperature for amorphous polymers whereas the dependence of specific
volume to temperature is exponential for semi-crystalline polymers. At temperatures below
the Tg , the semi-crystalline polymers are packed more densely due to the orderly fashion of
the molecular arrangement as compared to the amorphous polymers, hence specific volume is
low. Also, specific volume has a dependence on the rate at which cooling takes place. Slower
the cooling rate, higher the crystallinity, hence specific volumes are lower. [Poetsch and
Michaeli 2008]

Figure 2.10: Examples of PvT curves for amorphous materials [Zhou 2013]

Figure 2.11: Examples of PvT curves for semi-crystalline materials [Zhou 2013]
2 General Background 10

2.4 Shrinkage and Warpage

Shrinkage, which is part and parcel of polymer processing, occurs when the part reaches
from moulding temperatures to the point where the temperature crosses the glass transition
temperature Tg . It is a phenomenon which although cannot be avoided but can be controlled.
Warpage on the other hand is a result of differential shrinkage.

In injection moulding shrinkage takes place in three types: shrinkage during processing which
occurs under extreme conditions which is known as "in-mould shrinkage"; "as-moulded shrink-
age" which occurs just after the mould opens; and "post-moulding shrinkage" which occurs
during storage of the part moulded. [Jansen; Dijk, and Husselman 1998] This report
will talk about the "as-moulded shrinkage"; the "post-moulding shrinkage" and the effects of
moulding conditions. The "as-moulded shrinkage" occurs as a result of processing conditions
and the geometry of the mould. Holding pressure is known to be the most influential pro-
cess parameter which affects shrinkage; higher the holding pressure, lower is the shrinkage.
In case of the experiments carried out for this study, the selected holding pressure was ap-
proximately 80% of the injection pressure. Another important processing parameter is the
injection temperature [Jansen; Dijk, and Husselman 1998] which in this case was set as
recommended by the material MSDS. Gate freezing time is also important which also has an
impact on shrinkage. [Jansen; Dijk, and Husselman 1998]

There are various causes of shrinkage and warpage individually or in combinations, few of
them as stated below:-

• Geometry of the part and the mould may affect the orientation, be it amorphous or
crystalline morphology thereby affecting flow of the material. Presence of geometri-
cal constraints like ribs and bosses also have an impact on the shrinkage behaviour.
[Jansen; Dijk, and Husselman 1998]

• Residual stresses built-up in the part due to prevention of shrinkage of the layers that
are solidified in the mould, are released at ejection resulting in shrinkage. Initial starins
that are proportionate to residual stress in the cavity should be uniform, to have uniform
shrinkage and avoid post-moulding residual stresses. [Autodesk 2017c]

• Cavity pressure causes a variation in shrinkage along the path of flow depending upon
the morphology. [Jansen; Dijk, and Husselman 1998]

• Uneven cooling results in warpage caused due to variations in temperature at different

regions on the mould developing a bending moment. [Yang and Xu 2012]

• Process settings used to mould the part have overall a huge impact on the produced
part, and thus are of utmost importance.

2.4.1 Inside the mould

The material flows into the mould after the plasticization of the material. Pressure is the
force that helps to overpower the resistance of the melt to fill-up the sprue, runner, gate and
finally the cavity. The pressure in the cavity reduces gradually as it fills as illustrated by the
2 General Background 11

figure 2.12. A typical cavity pressure curve is illustrated in figure 2.13

Figure 2.12: Cavity pressure profile at the beginning of injection [Zhou 2013]

Figure 2.13: Typical curve for pressure inside the cavity [Zhou 2013]

In the recent time, higher pressure during injection is required to make sure that the cavity
has been filled up completely, this is because of the following reasons:
• Since the material constitutes 50-80% of the entire cost, it is aimed to reduce as much
material usage as possible. This may be done by reducing the thickness of the walls to
as low as possible. An immense pressure is necessary to make the material flow through
these thin cross sections.
• The cycle time also constitutes a big portion of the entire cost. The material must
therefore flow quickly and cool quickly, this means higher temperatures for heating and
cooling. Material flowing with a higher temperature would need a longer time to cool
inside the mould before being ejected, which in turn means a lengthy process. It is thus
cheaper to utilize higher pressures rather than temperatures. [Zhou 2013]
2 General Background 12

For the duration that the part remains inside the mould cavity, it is constrained, and it does
not deform, rather residual stresses build up. On demoulding the stresses release and defor-
mation takes place causing shrinkage. In the mould there is non-Newtonian, non-isothermal
and transient flow. As the polymeric material flows into the mould, the material coming in
contact with the cooler walls of the mould freezes and forms a layer. This skin layer is not
oriented. The material that is below the skin layer, let us call it as the sub-skin layer does
not freeze immediately but gets oriented because of the high shear rate. As we move close
to the gate, the stresses are lower because of gate-freezing which allows relaxation, therefore
the material close to the gate is less oriented than the sub-skin layer. PvT data is useful to
understand the process going on in the cavity physically. The figure 2.14 shows the devel-
opment of the material inside the cavity with respect to pressure and temperature against
time.

Figure 2.14: PvT diagram illustrating the development inside the cavity w.r.t. pressure and tem-
perature against time [Zhou 2013]
2 General Background 13

2.4.2 Residual stresses

The geometry used in this thesis is a rectangular plate. The fig. 2.15 shows how the material
flows in such a geometry.

Figure 2.15: Flow pattern in a rectangular cavity [Zhou 2013]

When shrinkage takes place all through the length and the cross section, internal stresses
build up. These are the residual stresses which have an effect as if force has been applied
externally, which is actually absent. Residual stresses are built up mostly because of the
different rates of cooling in the different layers formed along the cross section of the part.
The skin layer slows down the cooling process of the sandwich layers thereby slowing the
contraction. All the layers formed are linked to each other mechanically thereby holding
down each other for the thermal contraction. There are tensile stresses in the sandwiched
layers and compressive stresses in the skin layers, both being in equilibrium. This is explained
diagrammatically in fig. 2.16.

Figure 2.16: Development of residual stresses in the cavity [Zhou 2013]

This explains the role of residual stresses affecting the shrinkage and warpage.
2 General Background 14

2.5 Sensors
There are various sensors present in the machine which retrieve the data for important
parameters like temperature, flow rate, screw speed and position, pressures for filling and
packing etc. These sensors help in maintaining the quality and efficiency of the process.
Each cycle is recorded by the machine and a statistical study is possible to generate a process
window for the least number of defects and rejections in the process. By interpreting the
data, it is easy to understand the things gone wrong in the process which may have caused
deviations and therefore improve the final output. In addition to the sensors available in the
machine, pressure sensors were used to study the pressure profiles being built up inside the
cavity. The work of the pressure sensor is to convert applied pressure to electrical energy.
Calibration of the sensors is necessary after being fit into the system to establish a relationship
between the pressure applied and the signal received.
The output received from pressure sensors is a graph of pressure vs time which is measured
throughout each moulding cycle, a typical curve is as shown in figure 2.17. Two different
types of sensors were fit into the mould for this study - each type in either of the mould
halves. Both sensor types are explained briefly below.

Figure 2.17: Typical pressure curve plotted against time [Kamal and Kenig 1972a]

2.5.1 Kistler sensors

Four Kistler sensors of the type 6157B had been fit inside the fixed half of the mould and
have a diameter of 4mm. Fig 2.18 shows the type of sensors used.

These are piezoelectric, corrosion resistant sensors used for a direct pressure measurement
which can withstand a 2000 bar pressure. When the material flows into the mould with
tremendous pressure, it comes in contact with the front of the sensor. Form the front,
the pressure is transmitted to the measuring element made from quartz. The measuring
2 General Background 15

Figure 2.18: 4mm kistler sensor [Kistler 2011]

element then produces an electric charge which is proportional to the pressure applied which
is eventually converted to voltage in an amplifier. These sensors are especially used for
process monitoring, optimization and control of injection moulding. [Kistler 2011]

2.5.2 Cavity Eye sensors

Cavity eye PC-15 sensors shown in 2.19 were fit in the moving half of the mould.

Figure 2.19: Cavity eye sensor [CavityEye 2019]

Figure 2.20: Cavity eye sensor placement in the mould [CavityEye 2019]
2 General Background 16

Cavity eye sensors are resistant to shock and are water proof which are usable upto a pressure
of 40 kN . These no maintenance sensors are used for indirect pressure measurement and are
claimed to be equally accurate as direct pressure measurement sensors. In the current mould
there are 14 sensors placed behind each ejector pin. When the material flows, pressure gets
transferred from the ejector pin to the transfer pin and eventually to the sensing top of the
sensor (placement as shown in figure 2.20). The signal given out by the sensor is in direct
proportion to the force acting on the ejector pin. Since the surface area of the pin in the
mould is known, the pressure is calculated. [CavityEye 2019]

2.6 Simulation
Simulation for injection moulding has developed to a great extent, such that it allows to pre-
dict not only the filling behaviour and cooling behaviour but also helps in making predictions
for magnitude and directions in which the part may shrink and/or warp. This is particularly
important because these parameters influence the quality of the final part to a huge extent.
We may say that the results in the simulations are correct qualitatively, however there are
deviations as compared to the part produced in practice. The basic reason for the deviations
is that the simulations utilize a macro scale which does not take into consideration certain
criteria like the formation of microstructures and degree of crystallization, which should not
be neglected. [Hopmann et al. 2015]

The models predict the trends for moulding problems like pressure increase for filling; filling
temperatures; and decrease in the temperature of the mould - correctly. For certain special
cases the predictions agree to the actual results. [Lord and Williams 1975] At the time
of designing of the mould it is necessary for the designer to have a clear understanding of
the heat transfer and the flow behaviour of the material. This will help in reducing to rely
on intuition as also avoid the practice of trial and error for mould design; bring down cost
and time thereby reduction in the overall process and will also aid in recognising problems
occurring in moulding. [Lord and Williams 1975]

Mathematical models are required to predict the behaviour and calculate the various at-
tributes for filling, packing, cooling, shrinkage and warpage. In this section we will only talk
about the models used for shrinkage and warpage.

2.6.1 Shrinkage and warpage

The prediction of shrinkage and warpage in an injection moulded part depends on the de-
velopment of residual stresses in the process and their calculations. An important aspect in
injection moulding is how the holding pressure affects the stresses and a thermo-viscoelastic
model which took into account both stresses that are induced by pressure and thermal stresses
in the holding phase. [Baaijens 1991] This model was however isotropic, whereas semi-
crystalline polymers shrink differently in flow and transverse direction. Many researches
were done to overcome the problems faced in predicting the shrinkage and warpage of a wide
range of materials. This section will describe the residual strain model, the residual stress
model and a hybrid model for determining shrinkage.
2 General Background 17

1. Residual strain model

The residual strain model is given by equations (1) and (2)
k
S k = Σi=5
i=1 ai Mi (1)

S ⊥ = Σi=10 ⊥
i=6 ai Mi (2)
Where,
S k and S ⊥ = in plane strains in the directions parallel and transverse to the direction
of flow respectively,
a(i = 1...10) = shrinkage coefficients,
k
Mi and Mi⊥ = measures for - crystallization/ material orientation / volumetric shrink-
age / relaxation due to mould restraint / a constant - in the parallel and transverse
direction of flow respectively.

The calculation of the measures is done using finite element and finite difference which
analyses the stages of the injection moulding like filling, packing and cooling. This
model is the oldest methods used by simulation software for shrinkage prediction. The
following empirical model is the basis of the residual strain model:
S k = a1 Mv + a2 Mc + a3 Mok + a4 Mr + a5 (3)

S ⊥ = a6 Mv + a7 Mc + a8 Mo⊥ + a9 Mr + a10 (4)

Where,
S k and S ⊥ = predicted values of linear shrinkage in directions parallel and transverse
to the direction of flow respectively,
a(1...10) = material constants,
Mv = measure for volumetric shrinkage,
Mc = measure for crystallization,
Mok and Mo⊥ = measures for molecular orientation in the directions parallel and trans-
verse to the direction of flow respectively,
Mr = measure for mould restraint

The terms used in the above equation are briefly explained below as they constitute an
important aspect of the equation.
• Volumetric shrinkage

The volumetric shrinkage (VS) is an integral part for the calculations which is
mainly affected by the temperature of the material and the packing pressure. The
term used is as given in equation 5
v
V S = 100(1 − ) (5)
vf rz
Where,
V S = volumetric shrinkage
2 General Background 18

v = specific volume
vf rz = specific volume at room temperature and atmospheric pressure

• Crystallization
In case of semi-crystalline materials, the degree of crystallization also needs to be
taken into consideration for shrinkage calculations and is highly affected by the
mould temperature. With this term the volumetric contraction is calculated due
to the degree of crystallization. Higher the rate of cooling lesser is the crystalliza-
tion and vice versa.

• Orientation

The molecules in the polymer are aligned in the flow direction at the time of
shear flow. Orientation of the molecules are affected by melt temperature and
shear rate. When the flowing of the material stops molecular orientation relaxes
depending on the relaxation time of the material. If the polymer gets frozen faster
than its relaxation time it results in "frozen-in molecular orientation" which will
ultimately cause shrinkage variation in the flow and transverse directions. The
dT
level of orientation Φi is a function of the rate of cooling ( )i , shear stress τi
dt
and the angle of flow θi from the first to the next node with respect to the local
X-axis. Thus, for the x-axis
Φxi = Φi cos(θi ) (6)
Similarly, the calculations are done for the y-axis.
Therefore, for all elements,
Φx = Σni=1 Φxi (7)
and
Φy = Σni=1 Φyi (8)
Where,
n = number of nodes
Then the direction of orientation with respect to the local x-axis Ψ is

Φy
!
Ψ = arctan x (9)
Φ

• Mould constraint

There is an assumption that when the part is constrained inside the mould, it does
not contract. There is however a possibility of contraction the thickness direction.
During contraction there is a built up of residual stresses. The relaxation rate of
these stresses is affected by the filling phase, packing phase and the cooling phase.
The temperature increments from which the relaxation rate can be determined
helps in the calculation for mould restraint.
2 General Background 19

• Corresponding thermal strains

Different values for shrinkages are taken from the elements in the part, the resid-
ual strain model averages the values for both shrinkages in flow and transverse
direction. In actual practice there is variation in the shrinkage along the thick-
ness, which may be asymmetric creating a bending moment to cause warpage in
the part. Knowing the distribution of temperature along the thickness, an ap-
proximation can be made to calculate the distribution of shrinkage SH(z) along
thickness.
SH(z) = α(T (z) − Troom ) (10)

SH
α= (11)
Tav − Troom
Where,
T (z) = temperature distribution when the centre is frozen,
Tav = average of T (z) along thickness,
Troom = room temperature
SH = average shrinkage predicted by the model,
α = input for warpage analysis

Because software can deal only with linear strain distribution along the thickness,
thus there needs to be linearisation of SH(z) in such a manner as to save the
bending effect. Bending effect can be defined by equation 12

12 Z z=h/2
= zSH(z)dz (12)
h3 z=h/2

Where,
h = thickness. [Kennedy and Zheng 2001][Autodesk 2017a]

2. Residual stress model

This model holds the assumption that the material has a linear thermo-viscoelastic
behaviour. Instead of strain calculations, there is a direct calculation of distribution of
residual stresses, when the material is cooling inside the mould under certain pressure.
Stress along the thickness in the flow and transverse directions is provided by the dis-
tribution of residual stresses. The model takes into account stresses that are induced
thermally due to freezing of the material followed by shrinkage and also stresses induced
by pressure. [Autodesk 2017b]

This is a general form of Hooke’s Law for elastic solids having the form:

σij = ceijkl εkl (13)

Where,
σij = stress tensor,
2 General Background 20

εkl = total strain tensor,

ceijkl = tensor for elastic constants.

The tensor for strain is obtained by differentiation of displacement components ~u which

is defined as:

1
εij = (ui,j + uj,i ) (14)
2
[Kennedy and Zheng 2001]

The general relation for anisotropic linear thermo-viscoelasticity can be given as:
Z 1
∂εkl ∂T
σij = cijkl (ξ(t) − ξ(t0 ))( 0
− αkl (ξ(t) − ξ(t0 )) 0 )dt0 (15)
0 ∂t ∂t
Where,
cijkl = modulus for viscoelastic relaxation,
t = time,
T = temperature,
αkl = tensor for expansion coefficients,
R 1
ξ(t) = scale for pseudo time = 0t
aT
Where,
aT = factor for shift of time temperature.

The equation 15 relies on the assumption that the material is simple with respect to
thermo-rheology. For semi-crystalline materials, relaxation functions in the above equa-
tion have a huge dependence on the structures internally which themselves depend on
the processing conditions. The relationship between relaxation functions and internal
structure thus is neither theoretically nor experimentally exactly known. [Kennedy
and Zheng 2001]

Certain assumptions are accounted for in the viscoelastic model:

• Stress is only built up when the material goes above transition temperature.
• Shear stresses σ13 = σ23 = 0 when the direction x3 is normal to mid-plane locally.
• Through the thickness, σ33 is constant.
• Until σ33 < 0, the material will stick on the wall of the mould.
• Elasticity of the mould is not considered.
• After ejection, the behaviour of the material is like an elastic solid.

The accuracy of this model depends upon the following factors:

• The dependence of shrinkage on the PvT and transition temperature does not
replicate the behaviour in the moulding conditions.
2 General Background 21

• Predicting anisotropy is not possible as there is no arrangement for determining

the orientation of the molecules.
• Effects of crystallinity are not considered.
• Inadequate relaxation data range available to calculate visco-elasticity. [Au-
todesk 2017b]

3. Hybrid model

To reduce the inconsistency for predicted and the measured shrinkage hybrid model is
used which is also known as the Corrected Residual In-Mould Stress (CRIMS) model.
The formulation for unfilled materials is given by the equations 16 and 17

σck = b1 σp + b2 τ + b3 (16)

σc⊥ = b4 σp + b5 τ + b6 (17)

Where,
σck and σc⊥ = corrected principal stresses in directions parallel and transverse to the
direction of flow respectively,
σp = predicted residual stress,
τ = measure of orientation,
bi (1...6) = constants to be found.

To find the constants the values of shrinkage that have been measured in directions
parallel and transverse to the direction of the flow are used. By scaling the average of
the predicted residual stress and using modulus, the measured values of shrinkage are
converted to equivalent stresses in order to obtain the measured strain. Thus, the LHS
of the equations 16 and 17 above is known and the constants can be achieved using
regression. The stresses predicted and thermomechanical properties are regarded as
isotropic, thereby simplifying the material data. However, orthotropy is presented in
corrected residual stresses. [Kennedy and Zheng 2001] In a study it was concluded
that CRIMS gives more accurate results for unfilled crystalline materials, however it is
not of much significance for amorphous or filled crystalline materials that were used in
that study. [Kuppuswamy et al. 2010]

In the current study, the Residual stress model has been utilized by the simulation software
to calculate the shrinkage and warpage output.
3 Experiments 22

3 Experiments

3.1 Specifications
• Part geometry: A simple rectangular plate with a sprue gate with dimensions 200mm
x 100mm x 2mm was the specimen of the study

• Mould: Single cavity cold runner mould was used to mould the specimen. Cooling
channels were machined in the mould, water being the cooling medium. Temperature
of the mould was set as per the specifications required for each material.

Figure 3.1: Cross section of the single cavity mould used for the study
3 Experiments 23

Figure 3.2: Cooling channel

• Materials:

– Polypropylene HG385MO (unfilled) - Borealis (PP)

– Durethan B30S (unfilled) - Lanxess (ABS)
– Novodur P2H-AT (unfilled) - Ineos (PA6)

• Machine: E-motion 100 from Engel GmbH

– EM310/110T US
– Screw diameter 35 mm

• Measurement: Height gauge with dial gauge (accuracy 0.01 mm) was used to measure
the linear dimensions i.e. length and width. Thickness of the samples was measured
by a micrometer. The sample size was 6 numbers and each measurement was checked
at three locations on the part for a specific dimension.

Figure 3.3: Micrometer to measure thickness of the produced samples

[Mitutoyo]
3 Experiments 24

Figure 3.4: Height gauge used for linear measurements of the produced samples

• Process:

– Maximum injection pressure was set at 800 bar

– Barrel temperatures in ◦ C are as stated in the matrix 3.1 and the other processing
conditions are stated in the matrix 3.2

3.2 Procedure

3.2.1 Material Testing

All the materials used were initially tested for their flow properties. The viscosity and the
PvT data thus calculated has been used for simulations for this study. The PvT curves are
shown in the figures 3.5, 3.7 and 3.9 and the viscosity curves for the materials are given in

Barrel zones PP PA6 ABS

Nozzle (◦ C) 240 275 260

Metering (◦ C) 240 270 250

Compression (◦ C) 240 270 240

Feed (◦ C) 240 260 230

Throat (◦ C) 30 30 30

Table 3.1: Material specific barrel temperature zones

3 Experiments 25

Parameters PP PA6 ABS

Raw material pre-drying N.A. 6 hours at 80 ◦ C 4 hours at 80 ◦ C

Injection rate (cm3 /sec) 50 40 45

Switchover by screw position (mm) 21.83 17.15 15.07

Clamping force (kN ) 1000 1000 900

Holding pressure (bar) 400 400 400

Holding time (sec) 9 8 6

Mould temperature (◦ C) 30 80 70

Table 3.2: Material specific processing conditions

the figures 3.6, 3.8 and 3.10.

Figure 3.5: PvT curve for PP Figure 3.6: Viscosity curve for PP
(semi-crystalline) (semi-crystalline)
3 Experiments 26

Figure 3.7: PvT curve for ABS Figure 3.8: Viscosity curve for ABS
(amorphous) (amorphous)

Figure 3.9: PvT curve for PA Figure 3.10: Viscosity curve for PA
(semi-crystalline) (semi-crystalline)

3.2.2 Injection moulding

In order to start with the experiment injection moulding was carried out first to get a basis
for the simulation. The materials were pre-dried as per the requirements and standards as
mentioned in their respective data sheets. The specifications are mentioned in the matrix
3.2. The mould was mounted on the injection moulding machine and testers were moulded to
ensure complete and smooth functioning of the mould and the machine. The temperatures
were set with reference to the processing temperatures suitable for each material. A filling
study was performed followed by a packing study.

• Filling study
3 Experiments 27

1. A volumetric switch-over is intended by this filling study.

2. With the processing temperatures and other processing parameters set on the
machine the holding pressure is set to zero.
3. The switch-over volume is set to a value where there is minimum filling of the
cavity. In this case the starting volume was set to 50cm3 .
4. After the first shot is injected in the mould, the switch-over volume is then reduced
further to gradually increase the amount of material injected into the mould e.g.
45cm3 so as to observe and understand the filling pattern inside the cavity and
contemplate the amount of cavity filled.
5. With each increasing cycle, the cavity is aimed to fill more than the previous cycle.
6. This procedure is continued till we get a switch-over volume at which approxi-
mately 98% of the cavity is filled up.
7. This is the final switch-over volume.
8. The corresponding screw position was noted which was eventually used for the
simulations.

Figure 3.11: Illustrative example of filling study for PP - clockwise

3 Experiments 28

Figure 3.12: Illustrative example of filling study for ABS - clockwise

Figure 3.13: Illustrative example of filling study for PA - clockwise

• Packing study
3 Experiments 29

1. The aim of this study is to be able to set an appropriate holding time for the
production of the specimens for further measurement.
2. With the switch-over volume set as per the filling study, the holding pressure is
set to approximately about 80% of the filling pressure.
3. At this pressure, the holding time is initially set to zero and a shot is injected.
4. The weight of the part ejected form the mould is measured after the part is cooled
to room temperature.
5. This procedure is continued by gradually increasing holding time, while measuring
the weight for each part that is moulded until an almost constant weight profile is
observed. This is an indicator that the part has been sufficiently filled and packed.
6. A suitable packing time is thus chosen for the further processing of the parts.

Figure 3.14: Example for selection of packing time for PP

7. In case flash is observed, it is an indication of over packing of the part, in which

case the packing time is too much.

Since the filling and packing study are an important part for injection moulding for any
geometry, these must be done at the beginning of any new process. The parts are then pro-
duced with the parameters stated in tables 3.1 and 3.2 for all three materials.

The moulded parts are allowed to cool for 48 hours after which they are measured for lin-
ear shrinkages in the direction length, width and thickness using a height gauge with a dial
gauge. From the dimensional values measured, shrinkage in each direction is calculated.
3 Experiments 30

Figure 3.15: Illustrative example of packing

study for PP - clockwise

Figure 3.16: Illustrative example of packing

study for ABS - clockwise

Figure 3.17: Illustrative example of packing

study for PA - clockwise
3 Experiments 31

3.2.3 Simulation
Once the injection moulding is completed, the simulations are carried out. To perform the
simulations the model is prepared using the designer followed by simulation. The detailed
procedure is as following:

• Moldex 3D Designer

1. The geometry of the exact part produced in injection moulding is imported in the
Moldex 3D Designer software in the eDesign mode.
2. Points were defined to be able to measure the different attributes in the process
as well as the part, these are defined as measure nodes and sensor nodes.
3. Measure nodes help measuring the dimensional attributes and sensor nodes help
measuring the attributes related to the flow of the material, just like the sensors
present in the mould for sensing the temperature, pressure etc. against time. The
sensor nodes have been defined at the exact same locations as the sensors in the
mould.
4. In the next step, the type of gate from where the material enters the mould and
material entry point are defined. This step is followed by defining the runner
system. To achieve closest possible results, the original runner design is imported
and finally, the melt inlet is specified.
5. The next step is to define a mould base. The original mould design is also imported
to avoid discrepancies. The mould cooling channel design is then imported with
the defined inlet and outlet. Finally, the software checks for any errors in the
cooling channel to complete this step.
6. The next step is to generate the mesh. Here level 3 of the mesh was generated as
the part is quite thin and simple and is not able to generate a finer mesh. The
mesh element count was 536, 126 for the part geometry.
7. The mesh file is then saved and ready to use in the Moldex 3D simulation.

Figure 3.18: Meshed part

3 Experiments 32

• Moldex 3D Simulator

1. Open the simulator in the classic mode and import the mesh filed saved with the
designer.
2. Give an appropriate name to the project. Select the 3D model solver for injection
moulding in general and finish by checking the project setting.
3. Input the process settings to start the run by selecting new run, import the mesh
created in the Moldex 3D designer, select the material used for the process and
finally put in the process settings used in the injection moulding process carried
out earlier to finish the setting.
4. Select the Transient analysis 3 in the analysis sequence setting and hit run.
5. The software takes a few hours to complete the run at the end of which the results
are displayed.
4 Results and discussion 33

4 Results and discussion

This chapter will give an overview of the results for each material with respect to various
parameters viz. sensor positions comparing practical and simulated curves, injection time,
screw position and injection speed. The target of this study was to understand the accuracy
of the simulation software when the parameters from the practical injection moulding process
are input into the simulation.

The aim is to compare the shrinkage values and warpage behaviour of the practical versus the
simulated predictions. To be able to infer correctly, the cavity pressures from the practical
experiment need to be compared with the cavity pressures predicted by the simulation.
The figure 4.1 shows the location of the sensors in the mould. Here the fourteen red circles
indicate the position of the cavity eye sensors placed in the moving half of the mould, these
are also the positions of as the sensor nodes in the simulation; and the four blue crosses
indicate the position of the Kistler sensors placed in the fixed half of the mould. The gate is
located exactly at the center of the plate where the material gets injected into the cavity.

Figure 4.1: Sensor positions

As mentioned in the experimental a filling and packing study was carried out. During the
4 Results and discussion 34

filling study it was observed that the melt front contacts the sensors at position 5 and a first;
next it contacts the position 7; followed by 6 and b; and finally further to position c and
1. The sensor at position d was not under working condition and thus no data is recorded
at that position in the fixed half of the mould. This gives us an idea of the flow front and
pressures at the respective positions.

Figure 4.2: Sensor nodes to measure process attributes

Figure 4.3: Measure nodes to measure dimensional attributes

The figure 4.2 shows the positions of the sensor nodes which are placed at the exact same
positions as the sensors in the mould. These nodes are useful to obtain cavity pressure in
the simulation. Figure 4.3 shows the measure nodes. These are the positions at which the
dimensions have been checked in the simulation as well as in practice. Each dimension is
checked at three locations and an average value is calculated.
4 Results and discussion 35

4.1 Pressure curves

The figures 4.4, 4.5 and 4.6 show a comparison of the cavity pressures measured by the Kistler
sensors, the Cavity eye sensors and that predicted by the softare shown in shades of green,
blue and orange respectively. The curves for the screw position, injection speed and injection
pressure have been plotted using readings from the injection moulding machine during the
practical process.

Figure 4.4: Compilation of experimental and simulated process curves for PP

Figure 4.5: Compilation of experimental and simulated process curves for ABS
4 Results and discussion 36

Figure 4.6: Compilation of experimental and simulated process curves for PA

For all materials, the measurement of the curves in the practical was triggered at the begin-
ning of injection and stops at mould opening for ejection. The black curve on the primary
Y-axis, plots the injection pressure against time, which shows a low pressure at the start of
the process, the first peak in figure 4.4 indicated the end of the filling phase and the start of
the packing phase for PP where it is held at 400 bar for 9 sec. The sudden dip indicates the
end of the packing phase. The small land seen in this curve at a little less than 100 bar close
to 10 sec indicates a screw delay and the plateau observed between 11 − 15 sec indicates a
presence of back pressure. [Bozzelli 2019]

Similar inferences can be made for figures 4.5 and 4.6, except that two peaks at are observed
in the curve at the initial stage of the cycle. This initial peak at a higher pressure may appear
due to the presence of a cold slug which has been blown out of the nozzle. [Bozzelli 2012]
The holding phase parameters for ABS being 400 bar for 6 sec and for PA being 400 ba4 for
8 sec.

The pink curve in figures 4.4, 4.5 and 4.6 indicates the injection speed or the volumetric
flow rate plotted on the secondary Y-axis. It can be observed that, there is a quick filling
of the cavity in the filling phase. At the switchover to the packing phase, there is a sudden
drop of speed of injection. This maybe because the cavity is completely filled and now the
material is injected for the purpose of shrinkage compensation. This phenomenon lasts till
the end of the packing phase. At the end of the packing phase a small peak is observed due
to decompression after which the screw retracts and fills up the barrel with new material
form the hopper and begins plasticization for the next cycle.

The grey curve in figures 4.4, 4.5 and 4.6 is the indicator for the screw position for the du-
ration of the entire cycle, plotted on the secondary Y-axis. During the filling phase, a steep
downward slope indicates the quick forward movement of the screw to fill the material in
4 Results and discussion 37

the cavity. At switchover, the downward slope reduces considerably depicting a rather slow
forward movement of the screw, injecting much less material than before. For ABS, figure
4.5 it goes on to be an almost flat curve. Followed by the packing phase, a screw delay
is observed after which there is a steep increasing slope which indicates the refilling of the
material from the hopper into the barrel as the screw retracts upto a point where it is at the
most retracted position when a flat horizontal profile is seen. During the cooling phase of
the cycle the material is refilled into the barrel and plasticization takes place to prepare the
material to be injected for the next cycle.

The cavity pressures are recorded at four different positions in the cavity for each sensor and
simulation - closest to farthest. At the position where the cavity gets filled up completely, is
indicated by the sudden change in slope of the cavity pressure curves recorded. Descriptively,
for PP i.e. figure 4.4 this occurs at about 150 bar close to 1sec of the cycle time; for ABS
i.e. figure 4.5 approximately at a pressure of a little more than 100 bar at about 1 sec of the
cycle time and finally for PA the cavity completely fills up at about 80 bar a little more than
1 sec.

The parameters set for switchover were derived from the filling and packing study with respect
to the screw position, where the metering stroke was 65.48 mm. These were:

• PP: 21.83 mm

• ABS: 15.07 mm

• PA: 17.15 mm

At the position where the switchover actually takes place, a peak is observed in the cavity
pressure curves. This is the beginning of the packing phase. Further there is a gradual drop
in cavity pressures, and the curves begin to have a steeper downward slope at the end of the
packing phase. Further the gate freezes and the part is cooled in the mould till ejection takes
place. Ideally, the pressure at the position closest to the gate should experience maximum
pressure and thus have the highest pressure profile for the longest period of time in the cycle.
However, since the part in this study has a very thin cross section, the pressure profiles ob-
served for all the positions are quite close to each other for each means of recording in terms
of pressure. Another aspect is that the date of calibration of the sensors was not known and
the Kistler sensor at position d did not function, hence only three curves were recorded for
the same in the fixed half of the mould. There may also be some pressure loss due to the
freezing of the skin layer of the thin plate, thus the actual pressure is not sensed and the
error unknown. The simulation sotware utilizes the residual stress model for 3D designs. For
the simulation, the cavity pressure drops rather quickly in case of PP. This maybe because,
the simulation only works at a macro scale and ignores the effects of crsytallization, friction,
flow effects and the micro-structural formation as in the actual injection moulding process.
Also the errors in the model used for simulation are not known which could have an impact
on the results.

Individual curves for the material PP for position 5 for Cavityeye sensors and simulation,
and position a for Kistler sensors are shown in the figures 4.7, 4.9 and 4.8 respectively.
4 Results and discussion 38

Figure 4.7: Position 5 - Cavityeye sensor for PP Figure 4.8: Position a - Kistler sensor for PP

Figure 4.9: Position 5 - Simulation for PP

Figure 4.11: Magnified view for ABS

Figure 4.10: Magnified view for PP

Figure 4.12: Magnified view for PA

When magnified into the initial stage of the cycle and taking a closer look at the process
curves for the materials the switchover time can be inferred as in figures 4.10 4.11 and 4.12.
It can be seen in figure 4.10 that the switchover for PP occured at around 12 mm screw
position occuring at about 1 sec of the cycle time instead of the set value of 21.83 mm. For
ABS the set value was 15.07 mm occuring at a little more than 1 sec as in figure 4.11. For
4 Results and discussion 39

PA switchover is set at 17.15 mm which occurs at about 1.2 sec of the cycle time.

4.2 Shrinkage and warpage

After 48 hours of processing and cooling in atmospheric conditions the moulded parts were
measured for their dimensions in order to calculate the shrinkage. Warpage on the other hand
was only visually observed and not measured quantitatively. The semi-crystalline materials
PA and PP showed higher shrinkage values as compared to ABS which is an amorphous ma-
terial. The theoretical values for overall shrinkages for PA are in the range of 1.34% − 1.56%
[Lanxess Durethan B30S 000000 ], for PP are in the range of 1% − 2% [HG385MO - Borealis]
and for ABS the range is 0.4% − 0.6% [Ineos Technical Data Sheet], as per the data sheets
provided by the material manufacturers (refer appendix).

The graphs shown in figures 4.14, 4.13 and 4.15 have been plotted for the shrinkage values
for PP, ABS and PA respectively. The dimensions in the direction of width, length and
thickness were measured. In all cases the simulated shrinkage values appear to be less than
the shrinkage of the injection moulded parts.

Figure 4.13: Shrinkage comparison for ABS along the length, width and thickness

ABS being an amorphous material shows the least shrinkage compared to the other two
materials as seen in the figure 4.13. The molecules in an amorphous morphology are entangles
randomly and with increasing temperature start to uncoil and arrange themselves in the
direction of flow during injection due to experienced shear. Once the flow of the material
has stopped, relaxation of the molecules takes place and the molecules begin to go back
to orienting themselves randomly. Intermolecular forces tend to pull the molecules closer
till there sufficient drop in temperature for the molecules to freeze in that place. In case
of the samples moulded for this study, the material flows in the direction of thickness.The
4 Results and discussion 40

material tends to have the highest shrinkage in the direction of flow which is expected due
to the relaxation effect, however a uniform shrinkage results due to the intermolecular forces.
[Technology 2019] This can be very well seen in the injection moulded samples. The
simulation predicts more or less similar shrinkage values between 0.5% − 0.6% as the software
utilizes the residual stress model and does not predict anisotropy thereby cannot replicate
the actual moulding conditions as also it does not take into consideration the effects of
crystallinity.

Figure 4.14: Shrinkage comparison for PP along the length, width and thickness

Figure 4.15: Shrinkage comparison for PA along the length, width and thickness

The materials PP and PA are semi-crystalline in nature. Figures 4.14 and 4.15 show the
shrinkage observations for these two materials. The semi-crystalline morphology has and
4 Results and discussion 41

orderly structure of molecules which are firmly bundled together. As they come in contact
with higher temperatures, molecules start to unbind themselves and start to orient in the
direction of flow, quite like the amorphous materials. However, when cooling, relaxation of
the molecules does not occur, rather they begin to orient themselves in the direction of flow
and recrystallize thereby causing much higher shrinkages. The shrinkages are theoretically
higher in the transverse flow rather than the direction of flow [Technology 2019] which
can be seen for PP in figure 4.14 in case of injection moulded samples observed in the width
of the sample. Simulation predicts a higher length-wise shrinkage, however the values lie in a
close range of 1.4% − 1.5% due to the same reasons mentioned above. In case of PA a higher
shrinkage is observed in the direction of flow. The inability of the software to account for
the effects of crystallinity may be one of the reasons for such a result.

Warpage is caused due to uneven shrinkage. When there is uniform shrinkage in the moulded
part, a reduction in size is observed without distortion in its shape. Internal stress are
generated when there is a variation of shrinkage in any region of the part. These stresses
when exceed the structural integrity of the part, warpage is observed at ejection from the
mould. Warpage is not seen distinctly in the parts moulded for this project neither is it
observed in the simulation. This maybe as the part geometry is rather simple and they are
produced under rather high stresses.
5 Conclusion 42

5 Conclusion
Controlling shrinkage and warpage is a tricky business. It is a complicated aspect which is
interlinked with various factors that are involved; where each factor may affect some other
aspect or property of the final product. Simulation software are rather handy tools to get
an idea of the behaviour of shrinkage and warpage on given materials and parameters. It is
advantageous when a designer has access to simulation software while designing a product,
tool or process. This will help the designer tackle the problems which would have otherwise
occurred during production. Utilization of simulation software helps one to run different pro-
gramme settings to have a better understanding of and clearer visualizations of what could
be expected out of the design, tool, material and process parameters. Faster iterations with
a wide range of possible solutions with multiple combinations can be tested using limited
resources.

As per the study conducted it may be concluded that the simulations did give an idea of
how the shrinkage and warpage in the parts for the given materials would result in. The
simulated results are 70 − 75% accurate when compared to the experimental results. The
amorphous material ABS has the least shrinkage values both in practice and simulation. A
higher shrinkage is observed in the semi-crystalline materials PP and PA. The simulated
results show lower values of shrinkage as compared to the practical results as they are cal-
culated on a macro scale not taking into consideration aspects like crystallinity, anisotropy
etc.. Although not completely accurate, the simulated data can be used in cases of new de-
velopments or to access the effects of design changes, material changes or any other process
changes without much wastage of capital, time, material, labour, energy and other resources.

It is a cost effective method of getting deeper insights and improving one’s learning curve,
not only for shrinkage and warpage but also other analysis like filling, packing, cooling etc.
for a wide range of materials. Even though the software have some limitations, the future of
this technology looks quite promising.
6 Summary 43

6 Summary
To summarize shortly, three different thermoplastics PP and PA (semi-crystalline) and ABS
(amorphous) were used in this study. The materials were first injection moulded thereafter
simulations were performed for a simple plate geometry. To obtain closest possible results
the same inputs were used for injection moulding and the simulation, some outputs received
from the moulding process were also used for simulations. Practical and the simulated results
were compared to get an estimation of the accuracy of simulations. Comparison was made
using pressure curves for the process and linear dimensions for shrinkage.

The cavity pressure curves obtained from the sensors present in the mould for all three ma-
terials are quite similar in terms of pressure and vary by about 25 − 30%, where, curves from
Cavity eye sensors have higher values than the Kistler sensors and simulated curves show
the highest pressure values. However, with respect to time, the cavity pressures from the
simulations have a quicker pressure drop for the semi-crystalline materials PP and PA but
are quite similar for the amorphous material ABS.

The injection moulded specimens were measured for linear shrinkage in the width, length
and thickness direction at three different points on the sample. As expected ABS being
amorphous showed the least shrinkage in the range of 0.75 − 0.9% for the injection moulded
samples whereas the predicted shrinkage was about 0.6%. Semi-crystalline materials have a
higher shrinkage; for injection moulded PP samples the shrinkage was in the range 1.5−1.7%
against the predicted range of 1.4 − 1.5%. PA showed shrinkage values of practical samples
in the range 1.5 − 1.8% against the prediction of 1.2 − 1.5%. Thus it can be observed that
the predictions vary by a maximum of 30%.

Even though not 100% accurate owing to the ambiguities in the simulations, it would be
useful to have a basis for process optimization by using simulated data to make a difference
not only economically but also ecologically.
References 44

References
Autodesk (2017a): Residual strain shrinkage prediction method (MP and DD) (Concept).
— (2017b): Residual stress shrinkage prediction method (MP and DD) (Concept).
— (2017c): Shrinkage prediction method for 3D models (Concept).
Baaijens, F.p.t. (1991): “Calculation of residual stresses in injection molded products”. In:
Rheologica Acta 30.3, 284 to 299.
Bozzelli, John (2012): Improve Profits by Graphing Injection Pressure.
— (2019): How to Quickly Verify Your Machine & Process Setpoints.
Busch, J.; D. Rosato, and F. Field (1988): “Computer aided part cost analysis for injection
molding”. In: SPE RETEC.
CavityEye (2019): Industry 4.0 Connection of external devices in real time.
Chiang, H. H.; C. A. Hieber, and K. K. Wang (1991): “A unified simulation of the filling
and postfilling stages in injection molding. Part I: Formulation”. In: Polymer Engineering
and Science 31.2, 116 to 124.
Frizelle, William G. (2017): Injection Molding Technology. Ed. by Kutz, M. second.
Plastics Design Library. William Andrew Publishing, 191 to 202.
Garside, M. (2018): Global plastic production.
Geyer, Roland; Jenna R. Jambeck, and Kara Lavender Law (2017): “Production, use, and
fate of all plastics ever made”. In: Science Advances 3.7.
HG385MO - Borealis.
Himasekhar, K.; J. Lottey, and K. K. Wang (1992): “CAE of Mold Cooling in Injection
Molding Using a Three-Dimensional Numerical Simulation”. In: Journal of Engineering
for Industry 114.2, 213 to 221.
Hopmann, Ch. et al. (2015): “Simulation of shrinkage and warpage of semi-crystalline ther-
moplastics”. In: 050009: 1 to 5.
Ineos Technical Data Sheet.
Jansen, K. M. B.; D. J. Van Dijk, and M. H. Husselman (1998): “Effect of processing
conditions on shrinkage in injection molding”. In: Polymer Engineering and Science 38.5,
838 to 846.
References 45

Kamal, M. R. and S. Kenig (1972a): “The injection molding of thermoplastics part I:

Theoretical model”. In: Polymer Engineering and Science 12.4, 294 to 301.
Kamal, M. R. and S. Kenig (1972b): “The injection molding of thermoplastics part II:
Experimental test of the model”. In: Polymer Engineering and Science 12.4, 302 to 308.
Kennedy, P and R Zheng (2001): “High accuracy shrinkage and warpage prediction for
injection molding.” In: ANTEC 2002 Annual Technical Conference, pp. 1–7.
Kistler (2011): Mold Cavity Pressure Sensor. company magazine. Eulachstrasse 22, 8408
Winterthur, Switzerland: Kistler Holding AG.
Kuppuswamy, Venkatesh Aungadu et al. (2010): “IMPACT OF CRIMS DATA ON THE
ACCURACY OF MOLDFLOW WARPAGE PREDICTION”. In: ANTEC, 777 to 780.
Lanxess Durethan B30S 000000.
Li, Bo (2018).
Lord, H. A. and G. Williams (1975): “Mold-filling studies for the injection molding of
thermoplastic materials. Part II: The transient flow of plastic materials in the cavities of
injection-molding dies”. In: Polymer Engineering and Science 15.8, 569 to 582.
Macplas (2017): Deleting shrinkage and warpage in injection moulding and blow moulding.
Mitutoyo.
Nirkhe, C.P. and C.M.F. Barry (2003): “Comparison of approaches for optimizing molding
parameters”. In: Annual Technical Conference - ANTEC, Conference Proceedings 3, 3534
to 3538.
Poetsch, Gerd and Walter Michaeli (2008): Injection molding: an introduction. Carl
Hanser Publishers.
Santhanam, N. et al. (1991): “Postmolding and load-induced deformation analysis of plastic
parts in the injection molding process”. In: Advances in Polymer Technology 11.2, 77 to
89.
Stevenson, James F. (1978): “A simplified method for analyzing mold filling dynamics Part
I: Theory”. In: Polymer Engineering and Science 18.7, 577 to 582.
Tan, Kim H. and M. M. F. Yuen (1996): “Computer-aided system for the initial setting
of injection molding machine”. In: Fourth International Conference on Computer-Aided
Design and Computer Graphics.
Technology, Plastics (2019): Injection Molding: The Causes of Warpage.
References 46

Wu, P.-C.; C. F. Huang, and C. G. Gogos (1974): “Simulation of the mold-filling process”.
In: Polymer Engineering and Science 14.3, 223 to 230.
Yang, Jun Kai and Yun Jie Xu (2012): “Warpage Analysis of Injection Molding Based on
Mold Flow”. In: Advanced Materials Research 538 to 541, 1192–1196.
Zhou, Huamin (2013): Computer modeling for injection molding: simulation, optimization,
and control. A JOHN WILEY and SONS, INC., PUBLICATION.
Appendix 47

6.1 Datasheet For HG385MO

This document was obtained from the official website for Borealis AG [HG385MO - Bore-
alis]
 IM 0001 05.07.2004 Ed. 7

Polypropylene

HG385MO
Polypropylene Homopolymer for Injection Moulding

Description
HG385MO is a polypropylene homopolymer intended for injection moulding. It is characterised
by excellent flow properties combined with a narrow molecular weight distribution well suited for
low distortion products.

HG385MO is produced using the Borealis Controlled Crystallinity Polypropylene (CCPP)

technology. This provides a polypropylene with a very effective nucleation. Combined with an
antistatic and slip formulation, that allows short cycle times, good demoulding properties, and
low dust attraction.

Products moulded with HG385MO exhibit excellent dimension consistency (e.g. with different
colours) combined with high stiffness.

Applications
HG385MO is developed for injection moulding, and especially intended for caps and closures
products.

Typical application areas are:

- Caps and closures of various design, including hinge-caps, screw caps, caps with
complicated flow path, child-proof caps
- Items requiring good antistatic properties

Physical Properties**
Typical Value* Unit Test Method
3
Density 910 kg/m ISO 1183
Melt Flow Rate (230ºC/2.16kg) 25 g/10 min ISO 1133
Tensile Stress at Yield (50 mm/min) 36.5 MPa ISO 527-2
Tenisle Strain at Yield (50 mm/min) 8 % ISO 527-2
Tensile Modulus (1 mm/min) 1750 MPa ISO 527-2
2
Charpy Impact Strength, notched (+23ºC) 3.5 kJ/m ISO 179/1eA
Hardness, Rockwell 98 R-scale ISO 2039-2
2
Heat Deflection Temperature (0.45 N/mm ) 115 ºC ISO 75-2
* Data should not be used for specification work
** Values determined on injection moulded specimens acc. to ISO 1873-2 (97), based on 7 days conditioning time.

Borealis A/S
Parallelvej 16
DK-2800 Kongens Lyngby (Denmark)
Telephone: +45 45 96 60 00
Fax : +45 45 96 61 23
www.borealisgroup.com
 Polypropylene

HG385MO

Processing Guidelines
The grade can be processed on standard injection moulding machines. Following moulding parameters
should be used as guidelines.

Melt temperature 220 – 260ºC

Injection speed Highest possible
Holding pressure Minimum required to avoid sink marks (typical values are 200 - 500 bars)
Mould temperature 10 – 30ºC
Shrinkage 1 - 2%, depending on wall thickness and moulding parameters.

Storage and Handling

The product should be stored in dry conditions at temperatures below 50ºC and protected from UV-light.

Improper storage can initiate degradation, which results in odour generation and colour changes and can
have negative effects on the physical properties of the product.

Safety
HG385MO is not classified as dangerous preparation.

Dust and fines from the product carry a risk of dust explosion. All equipment should be properly earthed.
Inhalation of dust should be avoided as it may cause irritation of the respiratory system.
Small amounts of fumes are generated during processing of the product. Proper ventilation is therefore
required.

Recycling
The product is suitable for recycling using modern methods of shredding and cleaning. In-house
production waste should be kept clean to facilitate direct recycling.

A Safety Data Sheet is available on request. Please contact your Borealis representative for more details
on various aspects of safety, recovery and disposal of the product.
Appendix 50

6.2 Datasheet for NOVODURP2H-ATDS

This document was obtained from the official website for Ineos Holdings Limited [Ineos
Technical Data Sheet]
 Data Sheet

Novodur P2H-AT
standard impact strength, easy flowing, high gloss, contains antistatic additive
ISO Shortname: ISO 2580-1 -ABS 0, MGZ, 095-30-16-25

Property Test Condition Unit Standard Value

Rheological properties
Molding shrinkage, normal 60x60x2 % ISO 294-4 0.4 - 0.6
Melt volume-flow rate 220 °C; 10 kg cm³/(10 min) ISO 1133 37
Molding shrinkage, parallel 60x60x2 % ISO 294-4 0.4 - 0.6

Mechanical properties (23 °C/50 % r. h.)

Yield stress 50 mm/min MPa ISO 527-1,-2 44
Tensile Strain at break 50 mm/min % acc. ISO 527-1,-2 >15
Tensile modulus 1 mm/min MPa ISO 527-1,-2 2500
Flexural strength 2 mm/min MPa ISO 178 70
Flexural modulus 2 mm/min MPa ISO 178 2400
Izod notched impact strength 23 °C kJ/m² ISO 180-1A 16
Izod notched impact strength -30 °C kJ/m² ISO 180-1A 7
Yield strain 50 mm/min % ISO 527-1,-2 2.1
Tensile creep modulus 1h MPa ISO 899-1 2200
Tensile creep modulus 1000 h MPa ISO 899-1 1500
Charpy impact strength 23 °C kJ/m² ISO 179-1eU 100
Charpy impact strength -30 °C kJ/m² ISO 179-1eU 80
Charpy notched impact strength 23 °C kJ/m² ISO 179-1eA 16
Charpy notched impact strength -30 °C kJ/m² ISO 179-1eA 7
Ball indentation hardness N/mm² ISO 2039-1 110

Thermal properties
Temperature of deflection under load 1.80 MPa °C ISO 75-1,-2 93
Temperature of deflection under load 0.45 MPa °C ISO 75-1,-2 97
Vicat softening temperature 50 N; 50 °C/h °C ISO 306 98
Burning behavior UL 94 (1.6 mm) [UL listed] 1.6 mm Class UL 94 HB
Coefficient of linear thermal expansion, parallel 23 to 55 °C 10-4/K ISO 11359-1,-2 0.9
Burning rate (US-FMVSS) 2.0 mm mm/min ISO 3795 60
Glow wire test (GWFI) 2.0 mm °C IEC 60695-2-12 700

Electrical properties (23 °C/50 % r. h.)

Relative permittivity 100 Hz - IEC 60250 3.0
Relative permittivity 1 MHz - IEC 60250 2.9
-4
Dissipation factor 100 Hz 10 IEC 60250 55
Dissipation factor 1 MHz 10-4 IEC 60250 90
Volume resistivity Ohm·m IEC 60093 1E13
Surface resistivity Ohm IEC 60093 1E15
Electric strength 1 mm kV/mm IEC 60243-1 34
Comparative tracking index CTI Solution A Rating IEC 60112 600

Page 1 of 2
Edition 10.07.2009
 Data Sheet

Novodur P2H-AT
Property Test Condition Unit Standard Value

Other properties (23 °C)

Density g/cm³ ISO 1183 1.050

Processing conditions for test specimens

Injection molding-Melt temperature °C ISO 294 240
Injection molding-Mold temperature °C ISO 294 70
Injection molding-Injection velocity mm/s ISO 294 240

C These property characteristics are taken from the CAMPUS plastics data bank and are based on the international catalogue of basic data for
plastics according to ISO 10350.

Disclaimer
Disclaimer for sales products
This information and our technical advice - whether verbal, in writing or by way of trials - are given in good faith but without warranty, and this also applies where
proprietary rights of third parties are involved. Our advice does not release you from the obligation to verify the information currently provided - especially that contained
in our safety data and technical information sheets - and to test our products as to their suitability for the intended processes and uses. The application, use and
processing of our products and the products manufactured by you on the basis of our technical advice are beyond our control and, therefore, entirely your own
responsibility. Our products are sold and our advisory service is given in accordance with the current version of our General Conditions of Sale and Delivery.
Test values styrenics
Unless specified to the contrary, the values given have been established on standardised test specimens at room temperature. The figures should be regarded as guide
values only and not as binding minimum values. Kindly note that, under certain conditions, the properties can be affected to a considerable extent by the design of the
mould/die, the processing conditions and the colouring. This is valid especially for CTI.
Processing note
Under the recommended processing conditions small quantities of decomposition product may be given off during processing. To preclude any risk to the health and
well-being of the machine operatives, tolerance limits for the work environment must be ensured by the provision of efficient exhaust ventilation and fresh air at the
workplace in accordance with the Safety Data Sheet. In order to prevent the partial decomposition of the polymer and the generation of volatile decomposition products,
the prescribed processing temperatures should not be substantially exceeded. Since excessively high temperatures are generally the result of operator error or defects in
the heating system, special care and controls are essential in these areas.

INEOS ABS (Deutschland) GmbH, D-50769 Köln

© INEOS ABS (Deutschland)

Page 2 of 2
Edition 10.07.2009
Appendix 53

6.3 Datasheet for Durethan B30S

This document was obtained from the official website for Lanxess [Lanxess Durethan B30S
000000 ]
Durethan B30S 000000
PA 6, non-reinforced, injection molding
ISO Shortname: ISO 1874-PA 6, GR, 14-030

Property Test Condition Unit Standard guide

value
d.a.m. cond.

Rheological properties
Molding shrinkage, parallel 150x105x3; 270 °C / WZ % acc. ISO 2577 1.02
80 °C; 500 bar
Molding shrinkage, transverse 150x105x3; 270 °C / WZ % acc. ISO 2577 1.16
80 °C; 500 bar
Post- shrinkage, parallel 150x105x3; 120 °C; 4 h % acc. ISO 2577 0.32
Post- shrinkage, transverse 150x105x3; 120 °C; 4 h % acc. ISO 2577 0.4

Mechanical properties (23 °C/50 % r. h.)

C Tensile modulus 1 mm/min MPa ISO 527-1,-2 3200 1000
C Yield stress 50 mm/min MPa ISO 527-1,-2 80 40
C Yield strain 50 mm/min % ISO 527-1,-2 4.0 20
C Nominal strain at break 50 mm/min % ISO 527-1,-2 20 > 50
C Tensile creep modulus 1h MPa ISO 899-1 800
C Tensile creep modulus 1000 h MPa ISO 899-1 600
C Charpy impact strength 23 °C kJ/m² ISO 179-1eU N N
C Charpy impact strength -30 °C kJ/m² ISO 179-1eU N N
C Charpy notched impact strength 23 °C kJ/m² ISO 179-1eA < 10 20
C Charpy notched impact strength -30 °C kJ/m² ISO 179-1eA <10 <10
Izod impact strength 23 °C kJ/m² ISO 180-1U N N
Izod impact strength -30 °C kJ/m² ISO 180-1U N N
Izod notched impact strength 23 °C kJ/m² ISO 180-1A < 10 30
Flexural modulus 2 mm/min MPa ISO 178-A 2900 850
Flexural strength 2 mm/min MPa ISO 178-A 110 35
Flexural strain at flexural strength 2 mm/min % ISO 178-A 6.0 8.0
Flexural stress at 3.5 % strain 2 mm/min MPa ISO 178-A 95 25
C Puncture maximum force 23 °C N ISO 6603-2 5900
C Puncture maximum force -30 °C N ISO 6603-2 7300
C Puncture energy 23 °C J ISO 6603-2 90
C Puncture energy -30 °C J ISO 6603-2 85
Ball indentation hardness N/mm² ISO 2039-1 140 50

Thermal properties
C Melting temperature 10 °C/min °C ISO 11357-1,-3 222
C Temperature of deflection under load 1.80 MPa °C ISO 75-1,-2 55
C Temperature of deflection under load 0.45 MPa °C ISO 75-1,-2 160
C Temperature of deflection under load 8.00 MPa °C ISO 75-1,-2 45
Vicat softening temperature 50 N; 120 °C/h °C ISO 306 200
C Coefficient of linear thermal expansion, parallel 23 to 55 °C 10-4/K ISO 11359-1,-2 1.0

Page 1 of 3
Edition 28.04.2017
Durethan B30S 000000
Property Test Condition Unit Standard guide
value
d.a.m. cond.

-4
C Coefficient of linear thermal expansion, transverse 23 to 55 °C 10 /K ISO 11359-1,-2 1.1
C Burning behavior UL 94 1.5 mm Class UL 94 V-2
C Burning behavior UL 94 0.4 mm Class UL 94 V-2
C Oxygen index Method A % ISO 4589-2 26
Glow wire test (GWFI) 2.0 mm °C IEC 60695-2-12 750
Burning behavior US-FMVSS302 >=1.0 mm ISO 3795 passed
C Vicat softening temperature 50 N; 50 °C/h °C ISO 306 200

Electrical properties (23 °C/50 % r. h.)

C Relative permittivity 100 Hz - IEC 60250 4.0 15
C Relative permittivity 1 MHz - IEC 60250 3.5 4.0
C Dissipation factor 100 Hz 10-4 IEC 60250 170 2000
C Dissipation factor 1 MHz 10-4 IEC 60250 200 1200
C Volume resistivity Ohm·m IEC 60093 1E13 1E10
C Surface resistivity Ohm IEC 60093 1E14 1E13
C Electric strength 1 mm kV/mm IEC 60243-1 30 30
C Comparative tracking index CTI Solution A Rating IEC 60112 600
Comparative tracking index CTI M Solution B Rating IEC 60112 600 M

Other properties (23 °C)

C Water absorption (Saturation value) Water at 23 °C % ISO 62 10
C Water absorption (Equilibrium value) 23 °C; 50 % RH % ISO 62 3
C Density kg/m³ ISO 1183 1140
Bulk density kg/m³ ISO 60 700

Processing conditions for test specimens

C Injection molding-Melt temperature °C ISO 294 270
C Injection molding-Mold temperature °C ISO 294 80

Processing recommendations
Drying temperature dry air dryer °C - 80
Drying time dry air dryer h - 2-6
Residual moisture content % Acc. to Karl 0.03-0.12
Fischer
Melt temperature (Tmin - Tmax) °C - 260-280
Mold temperature °C - 80-100

C These property characteristics are taken from the CAMPUS plastics data bank and are based on the international catalogue of basic data for
plastics according to ISO 10350.

Page 2 of 3
Edition 28.04.2017
Appendix 56

6.4 Datasheet for Kistler pressure sensor

This document was obtained from the official website for Kistler [Kistler 2011]
 Pressure

Mold Cavity Pressure Sensor Type 6157B...

Unisens® with Front ø4 mm Patent No. US 6,212,963

Quartz sensor for cavity pressures up to 2 000 bar for injection

molding of plastics.

• ideally suited for industrial applications

• sensor front can be machined to adapt to the cavity wall
(except for coated versions of the sensor)
• exchangeable cable

Description
The Unisens quartz sensor for mold cavity pressure Type
6157BE... has a front diameter of 4 mm. An O-ring seals the
annular gap of <10 µm between sensor and mounting bore
and thereby also center aligns the sensor in the bore.

The pressure acts over the entire front of the sensor and is
transmitted to the quartz measuring element, which produc- For abrasive melts (e.g. filled with glass fibers or carbon fibers,
es a proportional electric charge (pC = Picocoloumb). This is thermosets, BMC/SMC), these sensors are available as Types
­converted into a voltage 0 ... 10 V in the amplifier and is then 6157BCE (0 ... 200 °C)/BD... (0 ... 300 °C) with a hardcoated
available as an amplifier output. front.

All parts of the sensor are corrosion-resistant. The exchange-

able cable is screwed to the sensor with a tight seal. The con-
nector is self-locking and splash-proof. Technical Data

For multi cavity applications the sensor Types 6157BA... and Range bar 0 ... 2 000
6157BC... are used without the single-wire connector Typ Overload bar 2 500
1839. The Multi Cavity Set Type 6829A... and the Multi Sen- Sensitivity pC/bar –9,4
sor System Type 6831B... are described in the appropriate data Linearity, all ranges % FSO ≤±1
sheets. Operating temperature range
Mold (sensor, cable)
This sensor is available with several Types of connecting cables Type 6157BA.../BC... °C 200
(see page 2). Type 6157BB.../BD... °C 300
Melt (at the front of the sensor) °C <450
Application Connector °C 0 ... 200*
This diaphragm-free sensor measures mold cavity pressures Insulation resistance
up to 2 000 bar during injection molding. It is particularly suit- at 20 °C Ω >1013
6157B_000-030e-12.11

able for optimizing, monitoring and controlling the injection at 200 °C Ω >1012
molding process of thermoplastics, elastomers, thermosets and at 300 °C Ω >1010
SMC.
* During machine down time the mold temperature may rise up to
240 °C, without causing any damage to the sensor. Note that meas-
uring errors may temporarily result.

Page 1/6
This information corresponds to the current state of knowledge. Kistler reserves ©1994 ... 2011, Kistler Group, Eulachstrasse 22, 8408 Winterthur, Switzerland
the right to make technical changes. Liability for consequential damage resulting Tel. +41 52 224 11 11, Fax +41 52 224 14 14, [email protected], www.kistler.com
from the use of Kistler products is excluded. Kistler is a registered trademark of Kistler Holding AG.
Appendix 58

6.5 Datasheet for CavityEye pressure cell

This document was obtained from the official website for Cavityeye [CavityEye 2019]
Pressure Cell Data Sheet
PC15
PC26

Description
The Cavity Eye developed water and shock
resistant sensors. The company offers sensors
in different diameters (15 and 26 mm), which
have a maximum load range of 1 to 40 kN.

Cavity Eye uses only indirect sensors. From our

experiences, the indirect measurement is much
more precise, than the direct measurement.
Using indirect measurement results in no need
for sensor maintenance during the whole
Technical Data
lifetime of the mould. Thanks to the sensor
geometry we designed, the sensor installation PC15
is as easy as it gets!
Weight g 21

Application Diameter mm 15
The sensors offered by Cavity Eye were Height mm 8
developed for cavity pressure measurement in Measuring pin diameter mm 3
injection moulds.
Hardness HRC 45
Fulfilling the industry’s requirements, the Operating temperature
sensors are designed heavy-duty, with water range
and shock resistance, thus applicable for the PC … °C -40 - +90
everyday use of pressure measuring.
PC … -MT °C -40 - +150

How does it work PC … -HT °C -40 - +200

The pressure sensor’s emitted signals – caused Excitation voltage V 5…12
by the excitation – are directly proportional to
Protection rating IEC 60529:1989 IP67
the force acting on the sensor’s measuring pin.
Knowing the area of the measuring ejector pin’s Maximum load
surface in the cavity, whereat the cavity PC15-1-… kN 1
pressure is acting on, the pressure inside the
PC15-5-… kN 5
cavity can be calculated.
Wire color codes
GND Black
Excitation Red
Differential signal + White
Differential signal - Yellow
Wire cross section
PC … AWG 30
PC … -MT AWG
PC … -HT AWG
Linearity deviation % <1
Bend at max. load mm <0,05

Cavity Eye Hungary Kft.,

PC15 Szellő str. 1, H-6000 Kecskemét, Hungary
PC26 PHONE: +36 76 951 966 │[email protected]│ WW.CAVITYEYE.COM