Anda di halaman 1dari 53

release 05//1997

Design guide

contents update
GE Plastics

Design guide
release 05//1997

Design guide 1 Contents

C o n t e n t s

1 General Introduction

1 Design Development ............... 9

1.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
1.2 Material design factors . . . . . . . . . . . . . . . . . . . . . . . 9
1.3 Development steps . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
1.3.1 End-use requirements . . . . . . . . . . . . . . . . . . . . . . 9
1.3.2 Preliminary design . . . . . . . . . . . . . . . . . . . . . . . . . 10
1.3.3 Material selection . . . . . . . . . . . . . . . . . . . . . . . . . 10
1.3.4 Design modification . . . . . . . . . . . . . . . . . . . . . . . . 10
1.3.5 CAD/CAE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10 Flow analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10 Stress analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
1.3.6 Prototyping and testing . . . . . . . . . . . . . . . . . . . . . 11
1.3.7 End-use testing . . . . . . . . . . . . . . . . . . . . . . . . . . . 11

2 Design for stiffness . . . . . . . . . . . . . . . . 13

2.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
2.2 How to determine stiffness . . . . . . . . . . . . . . . . . . . . 13
2.2.1 Material . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
2.2.2 Geometry & loading . . . . . . . . . . . . . . . . . . . . . . . . 14
2.2.3 Other factors influencing stiffness . . . . . . . . . . . . 14
2.2.4 Calculating the stiffness . . . . . . . . . . . . . . . . . . . . 15
2.2.5 Modal analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
2.2.6 Safety factors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
2.3 Increasing part stiffness . . . . . . . . . . . . . . . . . . . . . . 15
2.3.1 Ribs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
2.3.2 V-grooves . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
2.3.3 Corrugation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
2.4 Optimization of stiffness . . . . . . . . . . . . . . . . . . . . . . 16
2.4.1 Optimization for part weight . . . . . . . . . . . . . . . . . 16 Material Selection . . . . . . . . . . . . . . . . . . . . . . . . 16 Geometric considerations . . . . . . . . . . . . . . . . . . 16
2.4.2 Optimization for stiffness to cost ratio . . . . . . . . . 16

3 Design for strength . . . . . . . . . . . . . . . . 19

3.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
3.2 Material Strength . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
3.2.1 Ultimate strength . . . . . . . . . . . . . . . . . . . . . . . . . . 20
3.2.2 Yield strength . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
Design guide 1 Contents

3.2.3 Strain to failure . . . . . . . . . . . . . . . . . . . . . . . . . . . 20

3.2.4 Proportional limit . . . . . . . . . . . . . . . . . . . . . . . . . . 20
3.2.5 Material toughness . . . . . . . . . . . . . . . . . . . . . . . . 20
3.2.6 Other measures . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
3.3 Effects of various factors on strenght . . . . . . . . . . . 20
3.3.1 Other factors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
3.4 Part Strength . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
3.4.1 Ultimate part strength . . . . . . . . . . . . . . . . . . . . . . 21
3.4.2 Part yield strength . . . . . . . . . . . . . . . . . . . . . . . . . 22
3.4.3 Part toughness . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
3.5 Improving part strength . . . . . . . . . . . . . . . . . . . . . . . 22
3.5.1 Material choice . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
3.5.2 Geometry optimization . . . . . . . . . . . . . . . . . . . . . . 23
3.6 Design considerations . . . . . . . . . . . . . . . . . . . . . . . . 23

4 Design for behaviour over time . . 25

4.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
4.2 Static time dependent phenomena . . . . . . . . . . . . . . 25
4.2.1 Creep . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25 Creep recovery . . . . . . . . . . . . . . . . . . . . . . . . . . . 26 Creep failure modes . . . . . . . . . . . . . . . . . . . . . . . 26 Apparent modulus . . . . . . . . . . . . . . . . . . . . . . . . 26
4.2.2 Stress relaxation . . . . . . . . . . . . . . . . . . . . . . . . . . 27
4.2.3 Design considerations . . . . . . . . . . . . . . . . . . . . . . 28
4.3 Dynamic time dependent phenomena . . . . . . . . . . . 28
4.3.1 Fatigue . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28 Fatigue and endurance limits . . . . . . . . . . . . . . . . 28 Factors affecting fatigue . . . . . . . . . . . . . . . . . . . 29
4.3.2 Wear resistance . . . . . . . . . . . . . . . . . . . . . . . . . . . 29 Types of wear . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29 Factors affecting wear . . . . . . . . . . . . . . . . . . . . . 29

5 Design for impact performance . . 31

5.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
5.2 Common impact testing methods . . . . . . . . . . . . . . . 31
5.2.1 Pendulum methods . . . . . . . . . . . . . . . . . . . . . . . . . 32
5.2.2 Falling weight methods . . . . . . . . . . . . . . . . . . . . . 32
5.3 Design considerations . . . . . . . . . . . . . . . . . . . . . . . . 33
Design guide 1 Contents

6 Design for appearance . . . . . . . . . . . . 35

6.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
6.1.1 General remarks . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
6.2 Surface defects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
6.2.1 Sink marks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
6.2.2 Weld lines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
6.2.3 Air traps . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
6.2.4 Voids . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
6.2.5 Streaks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
6.2.6 Delamination . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
6.2.7 Jetting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
6.2.8 Gate marks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
6.2.9 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37

7 Design for precision . . . . . . . . . . . . . . . 39

7.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
7.1.1 Mould shrinkage . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
7.1.2 Secondary effects . . . . . . . . . . . . . . . . . . . . . . . . . 39
7.2 Shrinkage phenomena . . . . . . . . . . . . . . . . . . . . . . . . . 39
7.2.1 Cooling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
7.2.2 Packing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
7.2.3 Orientation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
7.3 Materials and shrinkage . . . . . . . . . . . . . . . . . . . . . . . 40
7.3.1 Amourphous materials . . . . . . . . . . . . . . . . . . . . . . 40
7.3.2 Crystalline materials . . . . . . . . . . . . . . . . . . . . . . . 41
7.3.3 Reinforced materials . . . . . . . . . . . . . . . . . . . . . . . 41
7.4 Design related factors . . . . . . . . . . . . . . . . . . . . . . . . 41
7.4.1 Part wall thickness . . . . . . . . . . . . . . . . . . . . . . . . . 41
7.4.2 Ribs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
7.5 Mould related factors . . . . . . . . . . . . . . . . . . . . . . . . . 41
7.5.1 Gate location . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
7.5.2 Gate type . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42
7.5.3 Gate size . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42
7.6 Processing related factors . . . . . . . . . . . . . . . . . . . . . 42
7.6.1 Melt temperature . . . . . . . . . . . . . . . . . . . . . . . . . . 42
7.6.2 Mould temperature . . . . . . . . . . . . . . . . . . . . . . . . . 42
7.6.3 Injection time . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42
7.6.4 Packing pressure . . . . . . . . . . . . . . . . . . . . . . . . . . 42
7.7 Secondary effects . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42
7.7.1 Thermal expansion . . . . . . . . . . . . . . . . . . . . . . . . . 42
7.7.2 Moisture absorption . . . . . . . . . . . . . . . . . . . . . . . . 42
Design guide 1 Contents

7.7.3 Post crystallization . . . . . . . . . . . . . . . . . . . . . . . . . 42

7.7.4 Creep under load . . . . . . . . . . . . . . . . . . . . . . . . . . 42
7.8 Simulation techniques . . . . . . . . . . . . . . . . . . . . . . . . 42
7.9 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43

8 Design for mouldability . . . . . . . . . . . 45

8.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
8.2 Material issues . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
8.2.1 Melt flow length . . . . . . . . . . . . . . . . . . . . . . . . . . . 45 Viscosity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45 Thermal properties . . . . . . . . . . . . . . . . . . . . . . . . 46 Shear properties . . . . . . . . . . . . . . . . . . . . . . . . . . 46
8.2.2 Melt temperature . . . . . . . . . . . . . . . . . . . . . . . . . . 46
8.3 Shrinkage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
8.3.1 General remarks . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
8.3.2 Warpage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
8.4 Cooling time . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
8.5 Design considerations . . . . . . . . . . . . . . . . . . . . . . . . 48
8.5.1 General remarks . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
8.5.2 Nominal wall thickness . . . . . . . . . . . . . . . . . . . . . 48
8.5.3 Projections . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
8.5.4 Radii . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
8.5.5 Ribs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
8.5.6 Support ribs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
8.5.7 Bosses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50
8.5.8 Undercuts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50
8.5.9 Coring . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
8.5.10 Draft angles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
8.5.11 Textures & lettering . . . . . . . . . . . . . . . . . . . . . . . 51
8.5.12 Flow leaders . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
8.5.13 Moulded-in-stress . . . . . . . . . . . . . . . . . . . . . . . . . 52
8.5.14 Weld lines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
8.6 Processing considerations . . . . . . . . . . . . . . . . . . . . . 52
8.6.1 Venting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
8.6.2 Gating . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
8.7 Ejection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54

9 Design for recyclability . . . . . . . . . . . 55

10 Design for automation . . . . . . . . . . . 57

11 Appendix . . . . . . . . . . . . . . . . . . . . . . . . . . . 59

The intention of this guide is to This design guide differs from most by
provide the design and engineering virtue of its ‘Designing for’ concept,
communities with an insight into the helping the reader move quickly to the
considerations necessary when issue that needs addressing.
designing applications in engineering Consequently, discussion of certain
thermoplastics. Many of these aspects can occur in more than one
considerations relate to moulding section, which demonstrates how
criteria, so those involved in the integral the process of designing for
manufacturing and processing of plastics is. Following the explanatory
plastics components should also chapters is a section showing typical
find it useful. engineering material performance
graphs for a range of GE Plastics
Typically, design manuals deal with a thermoplastic polymers.
specific resin family, presenting
properties, design criteria, assembly and Supplementary publications covering
other related information. GE Plastics’ assembly techniques, polymer
product line includes crystalline, processing considerations and overviews
amorphous, thermoplastic elastomers of specific resins in the GE Plastics
and glass mat reinforced polymers. product lines are available on request.
Because of this diversity, this brochure In addition, monographs discussing
will concentrate on issues common to all detailed design studies are released
injection mouldable thermoplastic periodically, covering these issues in
resins. greater depth than is possible in this
generalized guide.

Design development

1.1 Introduction

Throughout the process of product Specific information needed to establish weighed against the financial aspects
design, both functional and material end-use requirements includes: of the product. This exercise is often
aspects must be considered. Functional called ‘Product Risk Analysis’.
design factors relate to production and (a) Anticipated structural requirements Additional details are not given here
assembly. Material design factors since each company applies its own
concern the performance of a material LOADS measures.
in service. This performance, which These dictate the stresses a material
includes strengths, weaknesses and will be subjected to, and they define General Electric Plastics is prepared
limitations, is investigated to provide the component deflections. to provide assistance in this matter
starting point of the design process. and, if required, contact should be
RATE OF LOADING made with the nearest GE Plastics’
1.2 Material design factors A thermoplastic may demonstrate sales office.
different behaviour with changes in
Considerable information is needed by loading rate. Therefore, in addition (b) Anticipated environments
the design engineer to develop a to its magnitude, the rate at which
product design from the initial concept. loading is applied should be TEMPERATURE EXTREMES
This can be a smooth process if careful investigated. All materials possess a working
attention is given to each step involved. temperature range. Outside this
The designer must know the end-use DURATION OF LOADING range the component cannot
performance requirements of the Initial negligible deflections properly perform its intended
proposed application. To determine resulting from a small load may function. In addition, the properties
whether a material can meet these become unacceptably large if the of the material may vary considerably
requirements, the designer must be able load is maintained. within working limits.
to rely on information provided by the
raw material supplier, indicating the IMPACT FORCES As all thermoplastics are subject to
environmental and physical capabilities Because the application of high attack by certain chemical agents,
of the material. loads for short periods of time may the service environment of the
result in premature failure, the proposed component must be
1.3 Development steps nature of impact forces to which the established.
component will be subjected should
1.3.1 Establish end-use requirements be determined. Outdoor exposure for prolonged
periods may result in material
The development of any component VIBRATION degradation.
starts with careful consideration of This induces stress and deflection
anticipated end-use requirements. changes. Though these may be (c) Assembly and secondary operations
In general, the lower strength small, component failure may occur Usually a plastic component is not
properties of polymers compared with through constant repetition. used in isolation but is just one of a
metals and woods require that products number of components making up
be designed to utilize a larger FORESEEABLE MISUSE the end-product. The technique
percentage of their available strength. Though structural requirements may used for assembly, such as
have been satisfied, and an mechanical fastening, welding and
appropriate design proposed, failure adhesive bonding, needs to be
can still occur as a result of misuse. considered at the initial design stage
To establish the anticipated risk to optimize the component for ease
level, therefore, a ‘risk assessment’ of of assembly, (and ease of handling in
the product in use should be made. the case of automated assembly), or
ease of disassembly for maintenance
The ‘relevance tree’ method is ideal and recycling.
for this purpose, providing data to
facilitate assessment of foreseeable
misuse. The results should then be
Design guide 1 Design Development

1.3.2 Establish a preliminary design 1.3.4 Modify the design

Secondary operations, such as
painting, printing and hot stamping, A preliminary concept sketch of the If this is necessary, four areas in
also have to be given early proposed component can help the particular should be considered:
consideration in order to design the designer to determine which aspects are
best surface profile. This means inflexible, and which can be modified to (a) The specific property balance of the
avoiding for example sink marks and achieve required performance. selected grade, (e.g. tensile strength,
sharp changes in shape to achieve a The preliminary sketch, therefore, impact resistance)
high quality smooth surface. should include both fixed and variable
dimensions. (b)Processing limitations, (e.g. wall
(d) Cost limits thickness vs flow lengths)
1.3.3 Select the material
The following should be established: (c) Assembly methods, (e.g. snap-fits,
A GE Plastics’ material should be adhesives)
· Component cost resulting in selected that will satisfy initially defined,
profitable sales end-use requirements. The comparative (d)Cost of modification and its impact
· Annual volume property data sheet should be the first on component and/or project bud-
· Economic processing method(s) document which is consulted. Initial get, (refer to ■ F I G U R E 1 ).
with estimated cycle time(s) selection may subsequently be refined
· Tooling cost(s) for selected by reviewing the time, temperature and Strength of materials formulae should
processing method(s) environment-dependent properties be used in conjunction with material
· The expected service life of relevant to the particular application. property data to calculate necessary
the component Supplementary data, such as abrasion dimensions such as wall thickness.
resistance or ductility, may be needed (Refer to Chapter 3 ‘Design for
(e) Regulations/standards compliance to confirm the selection. strength’). Design calculations of a
repetitive or iterative nature, however,
Check which standards or Material properties can be divided into may warrant a computer-aided
regulations apply or can be applied two main categories. approach.
in the market place to the
component, product or appliance, (a) Mechanical properties used 1.3.5 CAD/CAE
for example: essentially for component design
calculations: Two particularly relevant computer-
STANDARDS aided systems are Flow Analysis of an
IEC/CEE International Electrical · Elastic limit injection moulding, and Stress Analysis
Committee/Commité Européan · Tensile strength of a final component. Both generally
d’Electricité · Modulus vs temperature use the Finite Element Method. This
· Poisson’s ratio approach considers the geometry and
ISO/CEN International Standards · Apparent (creep) modulus physical properties of the component as
Organization/Commité Européan · Fatigue limit a continuum of small manageable parts,
de Normalisation · Coefficient(s) of thermal or finite elements. Each element of the
expansion structure is individually investigated in
DIN Deutsche Industrie für · Coefficient of friction relation to its neighbouring elements,
Normungen · Thermal conductivity the total structure, and the physical
· Density constraints on the system. A large
BSI British Standards Institute · Mould shrinkage number of simultaneous equations
result, the solution of which is
NF Normes Francais (b) Other relevant properties: particularly suited to the repetitive
capabilities of a computer.
ASTM American Society for Testing · Hardness
of Materials · Impact strength Flow analysis
· Chemical resistance
REGULATIONS · Weathering resistance Because the performance of an
UL Underwriters Laboratories · Abrasion resistance injection moulded component is largely
· Ductility dependent on the moulding process,
CSA Canadian Standards Association · Flammability consideration of the service conditions
· Heat deflection temperature of the component in isolation is
CEE Publications of the Commité · Electrical properties insufficient to ensure a successful
Européan d’Electricité product. Simple shapes should not give
To design a plastic component, material flow problems to the
Factory and building codes. information concerning any experienced tool-maker and moulder.
combination of material properties However, larger components of
may be required. If the data are complex geometry often present
unavailable, or assistance is needed difficulties, for example positioning and
in interpretation, contact should be number of gates, runner dimensions
made with the nearest GE Plastics’ and location of weld lines.
sales office.
Design guide 1 Design Development

1.3.6 Prototyping and testing

In the flow analysis process, a computer Product performance tests can be
model of the component is produced, At this point in the procedure, conducted on functional prototypes or
and initial gating positions selected. a prototype should be constructed. production parts. Since functional
The predicted manner in which the The prototype and its testing can help prototypes may be produced using non-
material will fill the cavity is then the designer by: production tooling or part modelling,
presented graphically and numerically. caution must be exercised during
Isochronous temperatures and · Establishing confidence in the design testing and interpreting the results.
pressures throughout the system are by confirming that component require- The prototype may not behave in
calculated, in addition to weld line ments do not exceed design limits. exactly the same manner as a
locations and undesirable conditions · Developing preliminary product production component. The initial
such as overpacking. Moulding variables performance information. production components should also be
and/or gating positions are subsequent- · Identifying potential problem areas tested to confirm product performance
ly changed, if necessary, in order to in performance, manufacturing or testing.
achieve an optimum flow pattern. assembly.
· Allowing pre-launch assessment and Several specialized techniques,
By means of this iterative approach, feedback from consumer trials. examples of which follow, can be used
analyses of different options are possible for product performance testing:
before commitment to actual tool In order to obtain useful results,
production. Moreover, processing particular attention should be paid to · Strain gauge analysis
difficulties are identified and may be certain aspects of the testing: · Brittle coating analysis
rectified at the design stage. · Photoelasticity
· Proper analysis of component · Stress analysis by thermal emission
Nowadays, flow simulation is not the requirements. · Infra-red light banks for radiant heat
only software available to the designer. · Close similarity to the proposed effect measurement
Other injection moulding related product, particularly in critical or
software prediction tools include: suspect performance areas. · Environmental chambers for thermal
warpage, mould cooling, fibre · Development of realistic simulated use cycling
orientation, (for example for glass-filled and storage tests · Life testing under simulated use cond-
thermoplastics), moulded-in stresses · Commitment to the time and effort itions
and many more. required for testing before product · Accelerated ageing under elevated tem-
introduction perature, high humidity, or ultraviolet Stress analysis radiation conditions
1.3.7 End-use testing · Holography
A component in service will be
subjected to forces which induce stresses Tests should be conducted to simulate Although computer-aided techniques
in the material. To ensure that failure use and storage of the component. It allow accurate modelling of a proposed
resulting from overstressing does not should be established which tests reflect design, they should not exclude or
occur, it is essential that the stresses do defined component requirements, and replace finished part testing.
not exceed recommended design limits. whether they can be conducted in the The construction of a functional
laboratory or in situ. Frequently, the test prototype is therefore advisable.
As with Flow Analysis, the investigation equipment developed can be used for
of a simple geometry should not present future quality control testing.
any difficulty, since equations which can ■ FIGURE 1
often be solved by substitution have Influence of modification
been derived for many commonly to designs
encountered situations. However, the
analysis of a complex geometry, though
not generally suitable for this approach,
may be solved by the finite element
method. A mathematical model of the
component, defining the geometry by x,
y and z co-ordinates, is first described,
together with the properties of the
material. The boundary and loading
conditions are then entered and specific
output requested. Stresses in a
particular area may be of interest for

Numerical values of deflections,

diagrams of the distorted structure and Specification Part Mould Mould Mould Production
of stress distribution may also contribute Design Design Construction Commisioning
to a useful investigation. By this method, Time
areas of unacceptably high stress or
deflection may be identified and Cost of modification
suitably modified. Ease of modification
Design for stiffness

2.1 Introduction

The stiffness of a part is defined as the is assumed to be linear. Especially for strain curves are heavily dependent on
relationship between the load and the thermoplastics, the range in which the temperature. It is advised to consider
deflection of a part. This Chapter will stress/strain curve can be estimated with the stress/strain curve at the
discuss what modifications can be made a straight line is limited. For this reason, temperature at which the load is applied
to a part in order to influence and when a stiffness calculation of a part is for the calculation of part stiffness.
optimize stiffness. It will also give some made, it is necessary to check if the
guidelines of how the stiffness of a part occurring stresses and strains still allow Time also plays a role in the
can be calculated. a linear approach. If this is not the case, determination of the stiffness of a part.
it is advised to use a secant modulus for It can influence stiffness in one of two
2.2 How to determine the stiffness the stiffness calculation. ways:

In general the stiffness of a part is Consider ■ F I G U R E 2 . Suppose that the (a) the material is loaded for a
determined by its material and its stiffness of a part is calculated using the long time,
geometry. Young’s modulus Y. Suppose that a (b)the material is loaded in cycles,
verification of the occurring stress (c) the material is loaded during
2.2.1 Material results in the value s. This value is a very short time.
clearly out of the range in which a linear
The most important material property approximation of the stress/strain curve The phenomena (a) and (b)are
for stiffness is the stress/strain curve. is justified. In this case it is better to typically known as creep and fatigue.
In general, the Young’s modulus, which recalculate the stiffness of the part with How these effects must be taken into
is determined from the stress/strain the secant modulus Y*. account can be found in Chapter 4
curve, is the best parameter to be used ‘Design for behaviour over time’.
when comparing the stiffness of Furthermore, it is important to consider Phenomenon (c) is known as impact.
materials. However, when the Young’s the temperature at which the load is Refer to Chapter 5 ‘Design for
modulus is used, the stress/strain curve applied. For thermoplastics the stress/ impact’.


stress strain relation with Young’s modulus Y

Accounting for material

measured stress strain

A characteristic


Design guide 2 Design for stiffness

2.2.2 Geometry and loading

As mentioned earlier, the stiffness of
■ FIGURE 3 Besides the material, the geometry is a construction is determined by the
also important for part stiffness. Which combination of the material and the
A part loaded in tension
factors of the geometry are important is geometry. The following two examples
mainly determined by the type of load- illustrate how the stiffness of a part can
c ing. It should be noted that, in general, be calculated.
F F a part is loaded in more than one of the
following types at the same time. EXAMPLE 1
A part under tension
area A (a) Tensile loading
For a part loading in tension, For a part of length c, and constant
( ■ F I G U R E 3 ), the cross-sectional cross-sectional area A , the deflection
area A and the length of the part c f can be calculated with the formula:
are important.

(b) Compressive loading f = Fc/ yA

■ FIGURE 4 When a part is loaded under
compression, again the cross-
Cross section of a part
sectional area A and the length of where F is the tension force and Y is the
the part are the geometrical Young’s modulus of the material.
parameters that determine the From this formula it can be seen that
dA stiffness. In this case, buckling of the the stiffness of this part can be increased
y part is something that has to be by decreasing I, increasing Y or
considered separately. increasing A.

(c) Flexural loading

For flexural loading, the moment of EXAMPLE 2
inertia and the length are the geo- A part under bending
metrical parameters that determine
the stiffness. As can be seen in For a part of length c and moment of
■ F I G U R E 4 , the moment of inertia I inertia I, loaded by a force F at the end,
■ FIGURE 5 is defined as: ( ■ F I G U R E 6 ), the deflection f at the
end can be calculated using:
A rectangular cross-section
I = e y2 dA
A f = Fc 3/ 3YI
where y is the direction
b dA
y perpendicular to the neutral where F is the force and Y the Young’s
bending axis and A is the cross- modulus of the material. The stiffness of
sectional area. For example, for a this part can be increased by decreasing
rectangular cross-section, with the c, increasing Y and increasing I.
dimension a and b ( ■ F I G U R E 5 ), Suppose that the part has a rectangular
I can be calculated with the formula cross-section. Then I is given by 1/12 ab 3,
mentioned above: ( ■ F I G U R E 5 ). This means that I can be
a increased by increasing either a or b,
though increasing b has a much larger
_b _b effect.
2 2
I = e y 2 dA = e ay 2 dy = 1/3 a y 3 d = 1/12 a b 3 2.2.3 Other factors influencing
A -b_ -b_
■ FIGURE 6 2 2 Besides the geometry and the material,
other important factors have an
14 Clamped part loaded
(d) Torsional loading influence on stiffness, such as the type
at the end
When a part is loaded in torsion, of loading or the restraints of the part.
c the polar moment of inertia and the A load can, for instance, be
F length are the important concentrated on a point, but it can also
geometrical parameters that be a pressure on an area. Different
determine stiffness. The polar restraints that are used in calculations
moment of inertia Ip can be are clamped or simple supports. Note
calculated with a similar formula to that, in reality, the fixing system will
the moment of inertia. always be something in between fully
clamped and simply supported.
Design guide 2 Design for stiffness

2.2.4 Calculating the stiffness 2.2.5 Modal Analysis

· The part is used at temperatures above
Similar formulae to those given above Vibration resistance is important for the expected maximum.
are available for different cross-sections, many applications such as automotive · The part is used in a manner other than
for changing cross-sections, and so on. components. All parts designed in what it was designed for.
(Please refer to the Appendix for plastic or any other material will have · The part is assembled poorly or ‘forced’,
additional sources of reference). eigenfrequencies. These eigenfrequen- incurring a pre-load
cies will, amongst other factors, depend · Excessive moulded-in stresses resulting
The application of formulae is limited upon the stiffness of the part. If a part is from high moulding pressures, too small
due to different factors: loaded with a vibration load with a gates and too low tool temperatures.
frequency close to, or equal to, one of
(a) The material behaviour is assumed its eigenfrequencies, a potential danger A safety factor is typically decided upon
to be linear, which means that the of part damage exists. based on prior experience with similar
stress/strain curve of the material is designs. In general, the greater the
a straight line. In automotive components, it is often potential damage from failure, the
desirable to have the lowest greater the factor of safety that should
(b)The geometrical effects are assumed eigenfrequency of any part to be above be used. In many industries, safety codes
to be linear. For large deflections the normal operating frequencies of the and test procedures exist, which give
this leads to considerable vehicle. Therefore, although a part may standards and recommendations for
inaccuracies. be stiff enough to meet static loading safety factors.
requirements, it may require additional
(c) The formulae given above assume a stiffness to increase the 2.3 Increasing part stiffness
relatively simple shape. eigenfrequencies.
Features like ribs and V-grooves can be
(d)In general, a part is not purely Finite element analysis can be used to used efficiently to improve the stiffness
loaded in one mode, but in a predict the eigenfrequencies of the part and the load bearing possibilities of a
combination of modes. as well as the vibration modes shape for part. This is accomplished by locating as
each of the eigenfrequencies. This type much material as possible as far as pos-
A simple way of accounting for material of analysis is called modal analysis. sible from the neutral axis of the part.
nonlinearities is already given above. Modal analysis results are very sensitive This increases the moment of inertia.
Taking into account the other factors is to the type of loads, restraints and their As can be seen from Example 2 in
more complicated. The simple formulae locations on the part. Often it is possible Section 2.2.2 (d), the moment of inertia
cannot be used anymore when these to significantly change the vibration increases to the third power of the
effects play an important role. If the behaviour of a plastic part by distance of the material to the neutral
part stiffness has to be determined for redesigning the way it is mounted or axis, and only linearly with the distance
this type of problem, the only possibility restrained. along the neutral axis. This is a
is computer simulation using the finite principle that has to be kept in mind
element method. A computer model of 2.2.6 Safety factors when designing stiffeners into a part.
the part has to be made and evaluated
which requires much more time than The use of safety factors in engineering 2.3.1 Ribs
hand calculations. Also, an appropriate design is common practice with almost
computer system and software must be all types of material. Plastics are no Ribs are the most commonly used
available. exception to this rule. Conservative stiffeners. Processing and tooling
assumptions should be made when determine certain requirements for the
In general, the following guidelines can possible and worst case loads should be shape and the thickness of the ribs,(see
be given: considered. It is the responsibility of the Chapter 8). For a part under bending,
design engineer to anticipate that the ribs should run perpendicular to the
(a) For simple geometries and small products will not always be bending moment. For parts under
deflections, hand calculations can be manufactured or used as intended or torsion, ribs most efficiently increase the
made. planned. In some cases, product failure stiffness if they are placed diagonally.
is acceptable under unusual loading, or For pure stiffness reasons, it is
(b)For more complex geometries with for an occasional poorly manufactured important to note that high ribs are
small deflections, linear finite part. However, there are many cases in more efficient than thick ribs.
element analysis techniques can be which safety factors must be considered.
used. 2.3.2 V-grooves 15
There are numerous reasons why a
(c) Only nonlinear geometrical finite product can have less than its originally V-grooves, ( ■ F I G U R E 7 ), are often in-
element analysis can account for intended properties. Some of these corporated into parts where significant
large deflections on complex reasons are beyond the control of the increases in stiffness are necessary, and
geometries. designer and include: other requirements permit their use.
V-grooves are very efficient stiffeners
Note that the analysis costs increase · Exposure to chemicals which does not because they do not use large amounts
dramatically when going from step (a) normally occur of additional material and they do not
to step (c). · Moulding issues such as insufficient require additional cooling time.
drying or excessive heating of the resin However, V-grooves often cannot be
· Tool wear resulting in dimensional used because they provide an uneven
changes of the part top and bottom surface.
Design guide 2 Design for stiffness

As with ribs, V-grooves provide Another important factor for part

■ FIGURE 7 additional stiffness by increasing the weight optimization is density. It is often
average distance of material from the recommended to choose a material with
V-grooved structure
neutral axis of the part. For bending the highest ratio of elastic modulus to
M1 problems, V-grooves should be oriented density, which meets all the other
in the direction perpendicular to the requirements. A material with a very
bending moment. V-grooves can good stiffness to density ratio is
decrease the bending stiffness when the engineering structural foam.
M2 bending moment is directed along the Furthermore, the larger wall thickness
V-groove. The V-grooves given in at equivalent weight is a plus, since it
■ F I G U R E 7 are efficient for bending gives a greater moment of inertia for
moment M 1 , but inefficient for bending bending loads.
moment M 2 .
Note that when a glass fibre-reinforced
2.3.3 Corrugation material is used, the stiffness is mainly
increased in the direction, (orientation),
■ FIGURE 8 Corrugation is similar to the V-groove. of the glass fibres. This means that the
It also does not use large amounts of design engineer must ensure that in the
Corrugated structure
additional material and does not critical areas the fibres are aligned in
M1 require additional cooling time, but, the right direction in order to optimize
like the V-groove, often cannot be used the performance of the glass-filled
since it provides an uneven top and material. The direction of the glass fibres
bottom surface. A stiffness increase is is dependent on the flow direction of
M2 only achieved for a bending moment the material during injection moulding.
perpendicular to the corrugated axis, The gating of the part therefore has to
and can even be reduced for bending be designed in such a way that the
moments along that axis. increased material stiffness due to glass
The corrugation shown in ■ F I G U R E 8 is fibre reinforcement is used.
efficient for bending moment M 1 but
inefficient for bending moment M 2 . Geometric considerations

Note that for V-grooves and corrugated It is not possible to give a general rule
structures, when larger loads are of how the geometry of a part should be
applied, the stiffness can decrease due optimized, since it heavily depends on
to the fact that the profile becomes the type of loading and other factors.
flatter, and hence the moment of inertia Furthermore, design limitations
becomes smaller. This can be avoided by imposed by the material and the process
putting ribs perpendicular on the axis will limit the designer’s freedom.
of the corrugated structure. In general, it can be stated that material
should be added where it most effective-
2.4 Optimization of stiffness ly enhances the stiffness of the part.
Finite element analysis techniques are
Before an optimization study is started, particularly useful for geometrical part
the parameter that must be minimized weight optimization. Localized areas of
must be determined. In most cases this high and low stress can be easily
is either part weight or part costs. identified and redesigned. Material
situated in an area of low stress does not
2.4.1 Optimization for stiffness to contribute significantly to the part
weight ratio stiffness. Areas of high stress are areas
where also large strains occur which Material selection induce large deflections. An optimum
part design for stiffness, and strength,
When designing for an optimum would result in the most uniform stress
stiffness to weight ratio, for example distribution possible, subject to all other
aircraft components, often a material limitations.
with a high Young’s modulus and yield
strength is recommended. Of course the 2.4.2 Optimization for stiffness to
material also has to meet other cost ratio
requirements like the moulding
capability, aesthetics, environmental Designing for an optimum stiffness to
resistance, and appropriate regulatory cost ratio is often critical in applications
or standard requirements. Often glass- such as material handling pallets and
fibre reinforced materials are selected, building and construction components.
together with materials such as Ultem® The situation is very similar to designing
polyetherimide resin because of their for optimum stiffness to weight, except
strength and stiffness. that the final part weight must be
Design guide 2 Design for stiffness

multiplied by the cost per kg. of the ribbed and a corrugated structure have
material. All of the techniques been chosen. ■ T A B L E I summarizes ■ FIGURE 9
mentioned above still apply. some geometrical properties of the
The restraint and the loads
three different sections. As can be seen, on the beam
When designing an application for for the load on the beam, which results
which the cost is critical, all costs should
be considered, including material costs,
in bending moment around the y-axis,
the corrugated beam has the highest
processing costs, tooling costs, stiffness.
secondary costs and inventory costs. _c F _c
For the beams with the three different 2 2
EXAMPLE 3 cross-sections, a hand calculation and a
Comparison of a ribbed, V-grooved linear finite element simulation using
and corrugated structure. shell elements have been performed.
For the finite element simulation, the
This example shows how different levels mid-lines of the cross-sections are
of stiffness can be reached using the modelled and the thickness is
same amount of material. Furthermore, superposed on that. The results are
differences between hand calculations, presented in ■ T A B L E I I . The formula
linear finite element calculations, and for the hand calculation to calculate the ■ F I G U R E 10
geometrically nonlinear finite element deflection f is:
The cross-sections of the beam
calculations are shown and discussed.

Suppose that a beam, supported on two f = Fc 3/(48 YI ) V-groove

sides as shown in ■ F I G U R E 9 is loaded
with a force in the middle. The length is
assumed to be 800 mm. Suppose that for where F is the force, c the length of the
the cross-sectional area a rectangle with beam, Y the Young’s modulus of the
a width of 20 mm and a height of 10 mm material and I the moment of inertia.
y 10
is available. 2
Three possible cross-sections are shown x
in ■ F I G U R E 10 , each of them using the
same amount of material. A V-groove, a Ribs

■ TABLE I y 1.41 10
Some geometrical properties of the cross-section

Quantity V-groove Ribs Corrugated

J (mm 4 ) 348.6 480.6 668.0
x-coord c.g. (mm) 10 10 10 20
y-coord c.g. (mm) 5.896 6.983 4.403

Area (mm 2 ) 52.57 52.57 52.57

y 1.41 10

Hand calculation and linear finite element results

Deflection at 10N V-groove Ribs Corrugated

Hand calc. (mm) 139 101 72.5

Linear f.e.m. (mm) 148 99.6 74.0

Design guide 2 Design for stiffness

The deviations between the results from In order to study the effect of a linear
■ F I G U R E 11 the hand calculation and those from the f.e.m. calculation versus a nonlinear
linear finite element method, (f.e.m.), f.e.m. calculation, geometrically
The cross-sections of the beams
in a finite element analysis with
are due to the fact that for f.e.m. I is nonlinear f.e.m. simulations have also
shell elements calculated from the thickness been performed. The results are shown
superposed on the mid-line of the cross- in ■ F I G U R E 12 . It can be seen that for
section,( ■ F I G U R E 11 ). This means that larger deflections, there is a
V-groove the stiffness is underestimated by the considerable difference between the
f.e.m. for the V-groove and to a lesser linear and the nonlinear simulation.
extent for the corrugated section.
The stiffness is slightly over-estimated
for the ribs.

Force deflection curve of
■ F I G U R E 12 the V-groove beam
The difference between 40

Force (N)
a linear and a non-linear 35
finite element analysis
Corrugated 15
0 50 100 150 200 250 300
Deflection (mm)

Force deflection curve of

the ribbed beam
Force (N)

0 50 100 150 200 250 300
Deflection (mm)

Force deflection curve of

the corrugated beam
Force (N)

18 30
0 50 100 150 200 250 300
Deflection (mm)
Design for strength

3.1 Introduction 3.2 Material strength

The strength of a part is defined as the From a materials point of view, strength soon nonlinearities occur; a close
maximum load that can be applied to a is a stress/strain related property observation of the stress/strain curve
part without causing part failure under inherent in the material. There are shows that actually a proportional part
given conditions. In order to be able to many different stress/strain issues which does not exist. For larger strains,
determine the strength of a part, first can relate to strength and which must yielding occurs and the maximum stress
failure has to be defined. The right be understood in order to design for is reached. If the strain is increased
definition of failure depends on the strength. A variety of standard tests is further, necking will occur and the neck
application. In some cases a part has available to study the stress/strain will propagate through the structure
failed if it shows a certain amount of behaviour of a material under various until the material fails.
permanent deformation, while in other conditions. In general, the tensile test
cases this can be allowed and failure is provides the most useful information for For a glass-filled thermoplastic, the
defined as breakage of the part. In other engineering design. Other standard stress increases faster with an increasing
more critical applications, failure must tests that are performed to obtain strain; a glass-filled thermoplastic has a
be defined as the load at which the first strength data are flexural testing, shear higher Young’s or elastic modulus. For
crack occurs in the material. testing and compression testing. larger strains, nonlinearities occur and
the part fails in brittle mode when the
This chapter explains how to obtain Two typical stress/strain curves failure strain is reached. The following
strength measures from standard obtained from a tensile test are shown in strength measurements can be obtained
material data. Different failure ■ F I G U R E 13 . One curve is from an from these curves:
definitions are discussed and ways of unfilled thermoplastic and the other
determining the strength of a part are curve is from a filled thermoplastic.
presented. Furthermore, some general For the unfilled thermoplastic, for small
guidelines to design for strength are strains the stress increases
given. proportionally with the strain however

strain at fracture
■ F I G U R E 13

Typical stress-strain
curves for thermoplastics

Proportional Limit
maximum stress
yield region propagation

unfilled thermoplastic

slope = elastic modulus

Design guide 3 Design for strength

3.2.1 Ultimate strength 3.2.6 Other measures

fact that, for the glass-filled material
The ultimate strength, (tensile, flexural, brittle failure occurs, and for the Other measures that are equally
compressive, or shear), of a material is unfilled material ductile failure occurs. important for strength, but are not
the maximum stress level in a sample directly related to the stress/strain
prior to failure. The ultimate tensile 3.2.4 Proportional limit curve, are impact strength and fatigue
strength of a material is the maximum strength. For impact strength, readers
load per unit area that the material will The proportional limit of a material is are referred to Chapter 5, ‘Design for
bear before failure. This value can be the point at which the stress/strain impact performance’.
obtained from the maximum stress curve becomes nonlinear,
value on the stress/strain curve (see (see ■ F I G U R E 13 ). Although the The fatigue strength of a material can
■ F I G U R E 13 ). Simple formulae are often proportional limit of a material is often be measured in a number of ways.
used to obtain strength values from the considered to be the same as the Typically, it is expressed as the number
load-displacement output of non-tensile yielding point, the two are distinctly of cycles to failure of a standard test
type tests. Ultimate strength is a different. For many conventional specimen when subjected to tensile or
function of temperature and strain rate. materials like metals, the stress/strain flexural cyclic loading, in accordance
curve is so highly linear that the two with test standards. It can also be
3.2.2 Yield strength values are very close. For plastics, expressed as the stress which a material
however, the viscoelastic nature of the can withstand up to a given number of
The classical definition of the yield material can push the yield point of the cycles without failing.
strength of a material is the stress level material well out into the nonlinear
which will cause a small amount of region of the stress/strain curve. 3.3 Effects of various factors on
permanent deformation, (standard is Also the proportional limit is heavily the strength measures
0.2 % strain). This definition originates temperature- and strain rate-dependent.
from the steel industry, but it also Note that, strictly speaking, due to the As mentioned earlier, temperature and
applies to plastics. However, the yield viscoelastic nature of plastics, a strain rate heavily influence the strength
point of thermoplastics is difficult to proportional part of the stress/strain measures of materials. In addition, the
determine because of their viscoelastic curve does not exist for these materials. use of glass fibres alters the strength of
nature. The yield point for classical materials. ■ T A B L E I I I summarizes the
materials such as metals can easily be 3.2.5 Material toughness effects of temperature increase, strain
determined and tested by loading the rate increase and addition of glass fillers
sample to a point slightly beyond the Material toughness is a useful property on the strength measures.
proportional limit, (linear portion of for comparing material strengths.
the stress/strain curve), and then As with most material properties, it is 3.3.1 Other factors
unloading it and measuring the highly dependent on temperature and
deflection. Therefore, with some strain rate. Toughness is the energy (a) Chemicals
experience it is easy to predict. absorbed by the material. The area The chemical compatibility of
When plastic materials are tested in a under the stress/strain curve is equal to engineering thermoplastics is a complex
similar manner, what originally appears the absorbed energy per unit volume. phenomenon dependent on the
to be permanent deformation can be material, type of chemical, mode and
recovered slowly. The amount of
recovery is dependent on the
temperature and the rate at which the
sample is tested. The yield point of a Effect of temperature, strain rate and glass fibres on the strength of materials
thermoplastic cannot be determined
Temperature ➶ Strain rate ➶ Add glass fillers
from a stress/strain curve. For this
reason, thermoplastic yield points are Ultimate strength ➴ ➶ ➶
typically estimated or shown as a yield
range, (see ■ F I G U R E 13 ). Yield strength ➴ ➶ ➶
Strain to fracture ➶ ➴ ➴
3.2.3 Strain to failure
Proportional limit ➴ ➶ ➶ stress, ➴ strain
Another material property which is Material toughness *1 *2 ➴
related to strength is the percentage of
strain at which the material fails,
(see ■ F I G U R E 13 ). This can easily be *1 A measure of material toughness is the area under the stress/strain curve.
obtained from the stress/strain curve. An increase in temperature decreases the maximum strength. This effect will
As with many other plastic material increase toughness. But an increase of temperature also decreases the strain
properties, the strain to failure is highly to failure which will decrease the toughness. Hence, the effect of a
dependent on the temperature, the temperature increase on material toughness depends on the balance of the
strain rate and the ductility of the two factors described.
material. When the stress/strain curves
* 2 The influence on material toughness of an increased strain rate also depends
of the glass-filled material and the
on the balance of two different effects. Material toughness is increased at
unfilled material are compared, the
larger strain rates due to the increased ultimate strength but, on the other
strain to failure of the glass-filled
hand, it is decreased due to the lower strain to failure.
material is much smaller than that of
the unfilled material. This is due to the
Design guide 3 Design for strength

duration of exposure, temperature, and of what part strength really means. This concept of strength is common in
levels and state of stress present during In some cases, stress limits like yield engineering design, especially when
exposure. Different types of chemicals strength and ultimate strength are most factors of safety and worst case loadings
can embrittle a thermoplastic material important, while in other cases yield are considered.
or can cause it to become softened. strain and ultimate strain are important.
However, it is not possible to classify the An example of this type of strength
general effect of chemicals on materials. The meaning of the term strength of a concept on a load controlled
For more specific information readers part will depend upon the type of application is the design of load bearing
are invited to contact their GE Plastics’ application, the function of the part, aircraft components. In these designs,
representative. loading conditions, restraint conditions, loads of many times the expected load
and the performance of competitive or are considered in order to assure safety.
(b) Moisture comparable parts. In this section three Other part strength concepts like
Moisture can significantly affect the measures of part strength will be permanent deformations are not of
strength of engineering thermoplastics, discussed: ultimate part strength, part concern. In this case, the ultimate part
much in the same way as chemicals. yield strength and part toughness. strength is dependent on the ultimate
The hydrolytic stability of engineering The material parameters that are strength of the material and on the
thermoplastics is a complex issue important for part strength depend on geometry of the part.
dependent upon material type, mode both the measure of part strength and
and duration of exposure, temperature, on the type of loading. Two types of Another example is a flexible bearing
and stress level and stress states in the loading will be distinguished: cover in a car. In this case, the
material. Some materials like Noryl ® oad (or force) controlled applications application is deflection controlled and
PPO ® resin have an excellent hydrolytic and deflection controlled applications. it is possible that an extreme loading
stability, while others, such as Lexan® A load controlled application is defined condition on the bearing could produce
polycarbonate resin, can have hydrolytic as an application which must bear a load a deflection far beyond the normal
stability problems at elevated tempera- of a certain magnitude. A deflection operating range, which can cause the
tures. GE Plastics’ representatives can controlled application is an application part to rupture. The ultimate strength
supply more specific information. to which a deflection of a certain of a part is dependent on the strain to
magnitude is applied. failure of the material and the
(c) Processing conditions geometry.
Processing conditions can have a 3.4.1 Ultimate part strength
significant effect on the strength of a In summary, when ultimate strength is
material in the final product. Moisture, Consider the force deflection curve as considered, for load controlled
small gate sizes, long residence times, depicted in ■ F I G U R E 14 . Suppose this applications the ultimate strength is the
(material at melt temperature in the curve is the result of a test on an actual governing material parameter, while for
barrel of the machine), sharp corners or plastic part. When the part is loaded to deflection controlled applications the
wall transitions, or excessive use of point A, the part breaks. The force and strain to failure is the governing
regrind can all cause degradation of the deflection level at which the part breaks material parameter. Materials should be
polymer. Effectively this means that the can be defined as the strength of the selected accordingly.
average length of the molecular chains part. The maximum force level is usually
which comprise the material is reduced. referred to as the ultimate strength,
This can embrittle the material or cause while the maximum deflection level is
it to have a reduced modulus, yield referred to as the ultimate deflection.
strength or ultimate strength.

Other moulding parameters such as low

tool temperatures or high injection
pressures can cause a significant ■ FIGURE 14
Force (N)

increase in the levels of moulded-in A Failure

Sample load-
stresses. These, in turn, add to the deflection curve
stresses induced in the part by the load of a part
and hence decrease the load bearing
capabilities of the part.

Another important aspect of processing

is the occurrence and the location of
weld lines. The material strength on a
weld line is significantly lower. This
means that it is important to avoid weld
lines in areas where large stresses are
expected due to the applied loads.

3.4 Part Strength

permant deformation Deflection (mm)
The strength of a plastic part can have
different meanings depending on the
part. There are many different concepts
Design guide 3 Design for strength

3.4.2 Part yield strength 3.4.3 Part toughness 3.5 Improving the part strength

Suppose that the part, (■ F I G U R E 14 ) Part toughness is often considered to Once the strength criterion is decided,
is loaded to point B and then unloaded. be a good measure of the strength of the strength of a part can be improved.
After a certain time, the part regains its the part, and depends on material This can be done by selecting the right
original shape. If the part is loaded toughness and geometry. It is measured material and/or by optimizing the part
further, (for example to point C ) and by calculating the area under the load- geometry.
then unloaded, it does not retain its deflection curve from initial load to part
original shape and a permanent failure. The toughness of a part is an 3.5.1 Material choice
deformation will occur. issue for both load and deflection
controlled applications, and it is a Based on the strength criterion and the
If at a certain level of permanent measure of the amount of energy that loading type of the application,
deformation the part ceases to be useful a part can absorb. ■ T A B L E I V can be used to decide which

and must be replaced, then this is again material parameter should be increased
a new strength criterion which is in order to improve the strength of the
referred to as the part yield strength. part. Based on this information the
An example of a load controlled optimal material can be selected.
application is a thermoplastic spring.
Permanent deformation of the spring ■ F I G U R E 15
usually makes it non-functional. In this
Strain recovery
case, the part strength depends on both
the yield strength of the material and on
the geometry of the part. Another
example is a plastic snap-fit. A snap-fit is PC PEI
Percent after recovery 10 min.

a deflection controlled feature where

over-extension can cause permanent
deflection. This can make the snap-fit
useless if it cannot return to engage the mineral-filled
modified PPE PC - PBT
stop. In this case, the part strength
depends on both the yield stress, 80
(or yield strain), of the material and on
the geometry of the part.
A refinement of this failure criterion
could be the allowance of a certain
amount of permanent deformation. 60
If this is the case, strain recovery plays
an important role. When a material is
loaded and afterwards unloaded, (for 50
example up to point C in ■ F I G U R E 14 ), 3 4 5 6 7 8 9 10
the permanent strain will partly Applied Strain (%)
disappear in time.

In order to acquire material data, ■ TABLE IV

a laboratory test was developed.
Tensile specimens were loaded to Governing material properties depenent on failure criterion and load type
various strain levels at a rate of 10%/s
and then rapidly unloaded, while the
strength part yield strength part toughness
strain recovery was monitored.
Since the rate of recovery was very small no permanent some
after 10 minutes, the percent recovery, deformation permanent
which is defined as the ratio between allowed deformation
the residual strain and the applied allowed
strain, was plotted after this time as a
function of the applied strain. load ultimate yield stress maximum material
The results are depicted in ■ F I G U R E 15 . controlled material strain toughness
As an example, a design rule of 90% strength according to
recovery is shown in this Figure. ■ F I G U R E 15
It can be seen that in the case of PC a
maximum strain of approximately 9.3% deflection strain to strain at yield maximum material
will be allowed, compared to controlled failure strain toughness
approximately 5.8% for PC-PBT. according to
■ F I G U R E 15
Design guide 3 Design for strength

3.5.2 Geometry optimization 3.6 Design considerations

(g) Use safety factors
Since the strength of a part depends on Some general guidelines for design for As with all materials, it is impossible to
many factors, like loads and restraint strength are as follows: control all aspects of production and
conditions, it is very difficult to give use. For this reason, conservative
some general guidelines on how to (a) Avoid stress concentrators estimates and risk dependent safety
change a design in order to improve the Stress concentrators can significantly factors should be applied when
strength of a part. However, some reduce the strength of a part and should appropriate.
general remarks can be made be avoided. These include sharp
dependent on the type of loading on notches, or internal corners, sharply (h) Consider worst case
the part. Of course when a design is angled wall intersections, large wall In addition to analyzing the expected
modified to improve the strength, the thickness transitions and surface use for a proposed part, a worst case
design rules that are summarized in the interruptions such as holes and inserts. scenario should also be studied.
following section have to be respected.
(b) Avoid uncontrollable
For load controlled applications, the loading situations
amount of load that has to be carried by Examples include pipe threads where
the part is known. In the material this the designer can have no control over
load will be transformed to stresses. how much it is tightened, causing hoop
Areas of large stresses have to be stress.
identified and material should be added
to the appropriate areas. The goal is to (c) Design for compressive stress
have the load carried by a maximum As with many materials, unfilled plastic
amount of material. This will decrease materials tend to be stronger in
the occurring stress levels. compression than in tension, as long as
buckling does not become critical.
For deflection controlled applications
the situation is different. In this case, a (d) In tension, design for a uniform
certain deflection is prescribed to the cross-sectional area
part and this results in a given strain in In a plastic part which is primarily in
the material. The strength can be tension, a uniform cross-sectional area
increased by removing material in the should be used to obtain a more
areas where the largest strains occur. uniform stress distribution.
This will significantly decrease the
stiffness of the part, so it must be (e) In bending, design for moment
verified if this is allowed. of inertia
In a plastic part which is primarily
For load controlled applications the subjected to flexural loading, the
area of maximum stresses and for moment of inertia is critical in
deflection controlled applications the determining the load bearing capability.
area of maximum strains have to be For maximum load bearing strength,
identified. In general, hand calculations material should be located as far away
are not sufficient here, since the stress from the neutral axis as possible. For
and strain levels are usually far beyond maximum deflection without yield,
the point at which the linear material should be concentrated along
calculations are valid. Even the results of the neutral axis.
detailed finite element analyses have to
be interpreted with care. The reason for (f) Consider processing aspects
this is that the accuracy of especially the Processing can be critical for the
stresses and the strains is dependent on strength of a part. Recommendations
the type of element used and on the include:
mesh density. This means that the · Weld lines should be avoided or located
accuracy of the stress and strain levels is in non-critical areas.
not always as high as desired. However, · The frozen-in stresses should be as low
finite element analysis is very well-suited as possible.
to identifying areas of the largest · For fibre-reinforced plastics, the
stresses and to comparing the effect of difference in strength in the flow and
different possible design changes. the cross-flow direction should be
Design for behavior
over time

4.1 Introduction 4.2.1 Creep

When discussing time related behav- Under the action of a constant stress, deformation due to the applied load on
iour, two types of phenomena should be (load), a viscoelastic material undergoes a specimen. Up to this point, the
considered. Static time dependent a time dependent increase in strain response is elastic in nature and
phenomena, such as creep, are caused called creep or cold-flow. Creep is there- therefore the specimen will fully recover
by a single long-term loading of an fore the result of increasing strain over after the load is removed. However,
application. Dynamic time dependent time under a constant load. The creep continued application of the load will
phenomena, such as fatigue, are rate for any material is dependent upon result in a gradual increase in
produced by cyclic loading of an applied stress, temperature and time. deformation over time to point B ; in
application. Both types of behaviour are other words, it ‘creeps’. In the flexural
heavily influenced by material choice, Creep behaviour is initially examined and compressive modes, the specimen is
operating environment and component using plots of strain as a function of subjected to either a constant bending
design. time, over a range of loads at a given or compressive load and the deflection
temperature. Measurement may be is measured as a function of time.
The biggest problem experienced when taken in the tensile, flexural or
attempting to predict long-term loading compressive mode. In the tensile mode, The data may also be presented in other
effects, is the availability of data specific a test specimen is subjected to a forms to suit particular requirements.
to an application; most data are based constant tensile stress and the change in Sections taken through the creep curves
upon standard tests performed upon length is measured as a function of time. at constant times produce isochronous
material specimens. However, this The resultant stress/strain/time creep stress/strain curves, whilst isometric
information can be used to give an data are normally presented as curves of stress versus log time can be derived
indication of the performance of a strain versus log time. Point A in from constant strain.
certain design and can be verified by ■ F I G U R E 16 illustrates the initial

actual physical testing of the

thermoplastic components for the
required performance.

4.2 Static time dependent B ■ FIGURE 16


phenomena Typical creep behaviour

of an amorphous
Traditional structural calculations

centre around short-term material data assuming no plastic
in which the material exhibits perfectly deformation
elastic behaviour. However, another constant stress
important property exhibited by
thermoplastics is that of viscoelasticity.
This is the simultaneous demonstration
of both viscous flow and elastic
deformation under an applied load over A
a long time period. Short-term stress/
strain behaviour usually occurs within

less than an hour of the load being

applied and is often considered
instantaneous. Viscoelastic behaviour,
however, may continue throughout the
working life of a component, often over Log Time
a period of years.
Design guide 4 Design for behaviour over time Creep recovery

illustrated by point D in ■ F I G U R E 17 , creep, except that higher stresses are
If the long-term process of creep is whilst in others, a significant amount of used and time measured to failure.
interrupted by removing the load, an permanent plastic deformation will Results are usually presented as log
immediate but partial elastic recovery in occur, as illustrated by point E . stress versus log time to failure, as in
the deformation of the specimen can be ■ F I G U R E 18 .

seen, as illustrated in ■ F I G U R E 17 from Creep failure modes

point B to point C . This represents the Apparent Modulus
release of energy stored elastically by There are two failure modes
the material when it was first deformed encountered under the action of creep. Though creep and creep rupture data
to point A . give an indication of the long-term
(a) Excessive deformation behaviour of a material, for practical
As time progresses, recovery continues; When the deformation of a component design purposes the corresponding
in fact, the shape of the curve from exceeds the allowable limit for that reduction in modulus over time is of
point B is almost an inverted copy of the application. greater value. If a calculation is
shape up to point B . The amount of performed on a component subjected to
recovery depends upon the type of (b) Creep Rupture a continuous load, and short-term
material, the applied stress, the tem- This may result in either a brittle or moduli such as E , (modulus ofelasticity),
perature and the duration of loading. ductile fracture of the component. or G , (shear modulus), are used, the
In some cases, there will eventually be Measurement of creep rupture is result is likely to be misleading since
total recovery of the deformation, as performed in the same manner as neither E nor G reflect the effects of



Typical creep recovery
diagram of an amorphous

thermoplastic, assuming no
plastic deformation C
constant stress


Log Time


Typical creep rapture data,

with one cycle projection

one cycle projection


measured data

Log Time to failure

Design guide 4 Design for behaviour over time

4.2.2 Stress relaxation

creep. When the stress level and Stress relaxation data can be generated
temperature are known and creep Unlike creep where the strain increases by applying a fixed strain to a sample
curves are available at the given over time, stress relaxation is the and measuring the gradual decay of
temperature, an apparent or creep reduction in stress in a component over stress over time. The resultant data can
modulus, E app . may be calculated using time under a constant strain. The area then be used to generate stress
the creep curves; where stress relaxation is of greatest relaxation curves which are very similar
impact is in component assembly. This to isometric strain curves.
s1 includes threaded assembly, inserts,
E app . = ec
press-fits and snap-fits, although any Stress relaxation data can be used to
component undergoing long-term generate a relaxation modulus, similar
deformation may be affected to some to the creep modulus.
Where; degree. However, relaxation data are not as
s is the calculated stress level. commonly available as creep data.
ec is the strain from the creep curve Initially, the material is subjected to a In such instances, however, it is possible
at a given temperature and time. strain e1 which is maintained for a long to approximate the decrease in load due
period, that is to e 2, (see ■ F I G U R E 2 0 ). to stress relaxation by using the creep
The value E app . can then be substituted The immediate response of the material modulus, Eapp.
for E , or the like, in standard structural when the strain is applied is an increase
design equations. Creep modulus curves in stress from zero to s 1. This value is The use of standard moduli, ( E or G ),
or log creep versus log time at either temperature dependent and tends to in structural calculations, is inapprop-
constant stress or strain are usually de- decrease as a function of time, that is to riate for the long-term structural
rived from creep data, see ■ F I G U R E 19 . s 2,(see ■ F I G U R E 2 1 ). performance of assembled components
as it does not reflect the effects of stress

Apparent Modulus (GPa)

Typical Creep
Modulus curve of
2.0 Apparant Modulus
versus Log Time for
a given material at
1.5 temperatures


100 101 102 103 104 10 5 106
Log Time (s)


Graph illustrating Graph illustrating

constant strain stress relaxation
e1 e2 over time s1 over time for
constant strain

Log Time Log Time

Design guide 4 Design for behaviour over time

4.2.3 Design considerations Fatigue and endurance limits

Fatigue tests are usually conducted
Creep is often a critical concern when under flexural conditions, though The most important features of an
designing a structural part in thermo- tensile and torsional approaches are S - N curve are;
plastics. Although data exist for specific also possible. A specimen of material is
times, stress levels and temperatures, repeatedly subjected to a constant · The greater the applied stress or strain,
it is often difficult for engineers to deflection at constant frequency, and the fewer the cycles a specimen will
translate this data into structural the number of cycles to failure is survive.
analyses relevant to their application. recorded. The procedure is then
Since the process of individual testing repeated over a range of deflections. · The curve gradually approaches a con-
over long time periods is not feasible, The test data are usually presented as a stant value, known as the fatigue limit,
methods of interpolating and plot of log stress versus log cycles; this is as illustrated in ■ F I G U R E 2 2 . Below this
extrapolating data from short-term commonly referred to as a S - N curve. value of stress or strain failure is very
behaviour are necessary. unlikely. Some materials do not exhibit
a fatigue limit and for these materials an
Several methods for extrapolating endurance limit is specified. This is the
time/strain data exist, each with the value of stress or strain at a stated
objective of obtaining the most accurate ■ FIGURE 22 number of cycles beyond which the
fit to the actual data, while giving Typical fatigue curve specimen will fail, (see ■ F I G U R E 2 3 ).
reasonable extrapolation predictions for for a material with a
longer time periods. From work fatigue limit
performed within GE Plastics, it was
determined that for its materials the
best fit for the data appeared to be given

by a quadratic function of log time.

However, engineering judgement must
be used concerning the extent of the
extrapolation in time. It is not recom- S-N curve
mended that the extrapolation exceed
more than one unit of logarithmic time
and a strain/elongation limit of 20%
of the yield or ultimate stress/strength
value for the material being analyzed.
It should be noted that although Fatigue Limit
amorphous materials give a single curve
when interpolating to other tempera-
tures, for semi-crystalline materials two
plots are required to create the curve,
one above and one below the glass
transition temperature (Tg) of the Log Time to failure

4.3 Dynamic time dependent

phenomena ■ FIGURE 23
Typical fatigue curve
4.3.1 Fatigue for a material with an
endurance limit
This is the process by which a material is
stressed repeatedly or in some defined
cyclical manner. The magnitude of the

loading is usually of such low value that

failure would not normally be expected
were it applied only once. Additionally,
the loading and its frequency of
application may vary in value.
Structural components subjected to
vibration, components subjected to
repeated impacts, reciprocating
mechanical components, plastic snap-fit
Endurance Limit
latches and moulded-in plastic hinges
are all examples of applications where
fatigue can play an important factor.
Cyclic loading can result in mechanical
deterioration and fracture propagation
through the material, leading to Log Cycles to failure
ultimate failure.
Design guide 4 Design for behaviour over time Factors affecting fatigue 4.3.2 Wear resistance

S - N curves, obtained under laboratory Wear can be described as the reduction ■ FIGURE 24
conditions, may be regarded as ‘ideal’. of a bearing surface due to the loss of Microscopic view of
However, practical conditions usually molecular adhesion as a result of surfaces in contact
necessitate the use of a modified fatigue mechanical friction, (see ■ F I G U R E 2 4 ).
or endurance limit, as other factors may There is no apparent consistent
affect performance. These include: relationship between friction and wear; normal reaction
friction may be high and wear low, and between surfaces
Type of loading vice versa.
Results of tests performed under flexure
may not be directly applicable to Types of wear
components subjected to axial or
torsional loading. Adhesive wear wear bond plane of motion
Surface shearing, deformation and
Size of component removal of material at the points of
A deterioration in fatigue properties is adhesion.
usually experienced with increasing
physical dimensions. Abrasive wear
Surfaces of different hardness in
Mean stress contact, with grooves being ploughed in
Amplitudes about a tensile mean stress the softer material, the presence of an
reduce fatigue and endurance limits, abrasive medium and erosion abrasion
whereas an increase occurs when the due to particles impacting the surfaces.
mean stress is compressive.
Loading frequency Surface fatigue failure as a result of
At high frequencies, an increase in surface or sub-surface stress exceeding
temperature may be experienced if the endurance limit of the material.
inadequate provision for heat
dissipation exists. This may lead to Factors affecting wear
deterioration in fatigue properties.
The wear resistance of a thermoplastic is
Amplitude dependent on the environmental
The onset of fatigue failure may also be conditions of the end-use application.
accelerated at low frequency under Temperature, surface contamination,
conditions of high amplitude. (chemicals, grease, etc.), surface
structure and contouring, all have an
Fatigue testing provides only an influence on wear resistance. Changes
indication as to a given plastic material’s in the wear process occur with the
relative ability to survive fatigue. degree of surface roughness of the
Therefore, it is essential that tests are bearing materials. In the case of plastics,
performed on actual moulded this is strongly dependent on the
components under actual end-use modulus of elasticity, (E-modulus), of
operating conditions, in order to the materials being used.
determine the true fatigue endurance
of that component. Optimally moulded surfaces of Valox®
PBT resin components have a relative
smoothness which makes them ideally
suited to low friction, high wear
resistance applications over a wide
range of operating conditions.
In general, Valox resins exhibit very low
coefficients of static and dynamic
friction against metals and also against
similar Valox resins. Other semi-
crystalline resins generally exhibit an
appropriate wear resistance against
metals but invariably may not be
suitable for applications where they bear
against themselves.
It is important to remember that the
operating environment of the end-use
application should be taken into
consideration when using frictional
values for engineering design purposes.
Design for impact

5.1 Introduction 5.2 Common impact testing methods

Impact strength can be described as the A number of tests are commonly used to They can be useful in applications for
ability of a material to withstand an provide insight into the response of quality control and initial material
impulsive loading. By definition, any plastics to impact loadings. Two generic comparisons, but even in this latter role
body in motion possesses kinetic energy types of testing exist, pendulum and different tests will often rank materials
and, when this motion is stopped, the falling weight methods. It is important in a different order. Typical traces for
energy must be dissipated. to note that none of these traditional both ductile and brittle response to an
test methods results in real, geometry instrumented impact measurement
There are several factors which independent material data which can be system are illustrated in ■ F I G U R E 2 6 .
determine the ability of a plastic used in engineering design calculations.
component to absorb impact energy:

· Type of material
· Wall thickness
· Geometric shape of component ■ FIGURE 25
Impact value

· Size of component Graph illustrating

· Operating temperature effect of tempera-
· Rate of loading ture upon impact
· Stress state induced by loading Ductile behaviour response

For ductile polymers, the load at which Ductile/brittle transition

yielding occurs in a component is
affected by the operating environment
as illustrated by the last three factors
detailed above. Of even more
significance to design is the fact that,
under the appropriate circumstances, Brittle behaviour
the impact behaviour of a ductile
polymer will undergo a transition from
ductile and forgiving in its response, to
brittle and catastrophic. Temperature

Usually this change in behaviour is

described in terms of a transition
temperature above which the failure is
Ductile polymer Brittle polymer
more ductile in nature, and below ■ FIGURE 26


which it is more brittle, as illustrated in Typical force/time

■ F I G U R E 2 5 . However, both rate of
curves for ductile
loading and stress state have an effect and brittle material
on this transition temperature. responses, as 31
produced by an

Time Time
Energie absorbed by specimen
Design guide 5 Design for impact performance

5.2.1 Pendulum methods

(c) Tensile test
■ FIGURE 27 In this form of testing, a specimen of This test uses a similar swinging
material is struck by a pendulum under pendulum to the previously mentioned
Illustration of Charpy impact
defined conditions, and the energy test methods. The test specimen is
test configuration
required for fracture is measured. mounted so as to facilitate measurement
It should be noted that the results of of the energy required to fracture the
Path of pendulum these tests are highly dependent upon bar under tensile impact loading,
the geometry of the test specimen and (see ■ F I G U R E 2 9 ).
that the results do not define actual
material properties. 5.2.2 Falling weight methods

(a) Charpy impact In this test a weight (dart) is allowed to

In this test, a notched specimen is fall onto a specimen, under defined
simply supported. When released, conditions, as illustrated in ■ F I G U R E 3 0 .
the pendulum strikes the specimen The procedure is then repeated with
centrally, resulting in fracture successive increases as a product of dart
The energy absorbed by the specimen mass and drop height.
is displayed by a system activated by the Preferred methods are those which keep
motion of the pendulum, a constant height and vary the dart
(see ■ F I G U R E 2 7 ). mass, since speed at the point of impact
remains equal.
■ FIGURE 28 (b) Izod impact
This is similar to Charpy though A modern, instrumented form of the
Illustration of Izod impact
differing more in notch geometry and dart puncture impact test is the Dynatup
test configuration
method of specimen support. In this impact test. In this test, head velocity
test, the specimen is mounted vertically, and force are measured continuously
Path of pendulum impact consequently occurring distant and are output along with a computer
to the notch. The test can also be calculated energy absorption curve.
performed with an unnotched specimen For ductile polymers, the energy to
or the notch reversed. failure measured in a falling dart test is
Test results are always reported with a complicated function of the yield
respect to notch presence and position, stress, draw-strain, post-yield modulus
(see ■ F I G U R E 2 8 ). and ultimate failure stress of the
material, none of which can be
fundamentally measured by the test.

Illustration of tensile Illustration of falling dart
impact test configuration impact test configuration
Dart in guides

Test bar

32 Height of fall


Anvil Specimen usually in sheet form

Support centred
under specimen
Design guide 5 Design for impact performance

5.3 Design considerations

Although the brittle failure stress usually
It is well-established that strain rate and shows only a small dependence on rate,
temperature have a distinct effect upon the yield stress is affected much more
the yield stress, (sigma-yield), of a significantly. Furthermore, the yield
polymer. The relationship between stress usually also shows a stronger
stress and strain rate takes the form: dependence on temperature than the
brittle failure stress as reflected in the
sigma-yield = B1 + B2 * In (strain rate) steeper yield curves as a function of
temperature. The net effect is that, as
where B1 and B2 depend upon the the strain rate is increased, the
polymer and the temperature. transition temperature defining the
In general, higher rates and lower boundary between ductile and brittle
temperatures lead to high yield stresses, failure moves to a higher level.
and lower rates and higher tempera-
tures lead to lower yield stresses. Due to the many factors which can
influence the ability of a plastic
The interrelationship of strain rate and component to absorb impact energy,
temperature with regard to ductile- combined with the problem of relating
brittle transitions can be seen in standard test results to application
■ F I G U R E 3 1 . Two sets of curves are requirements, it is recommended that
shown, one describing the brittle stress all final designs be fully tested for
at two strain rates for a hypothetical applications where impact is a serious
polymer as a function of temperature; performance issue.
the other describing the yield stress of
the same material for the same strain


Graph illustrating ductile-

to-brittle transition in
Brittle stress O- B failure


Yield stress O- Y
H - High strain rate
L - Low strain rate


Design for

6.1 Introduction

This Chapter looks at applying guide- that a slight sink mark will occur,
lines to achieve the best possible surface especially far away from the gating ■ FIGURE 32
appearance for parts produced by the point. A textured finish to the cavity wall Styling features can be applied
injection moulding process with might hide the defects but it is seldom to hide sinkmarks caused by ribs
engineering thermoplastic materials. possible to create a high quality surface or bosses
when applying ribs, bosses or sudden
6.1.1 General remarks wall thickness transitions.

The nature of the polymer chosen is an A compromise can be found by using

important factor in obtaining a good styling features like grooves or a step in
surface finish. Glass-filled materials in the surface where a rib is mating the
general have a relatively poor surface critical surface, (■ F I G U R E 3 2 ).
finish compared to unfilled polymers. Gas-assisted injection moulding is a
The discussion about surface quality of technique which can greatly reduce sink
injection moulded parts is not really an marks, but gloss differences are difficult
objective one, because the perception of to avoid, (■ F I G U R E 3 3 ). It has to be
quality is different from person to realized that locations far away from the
person. The topics discussed in this gating points, combined with thin walls
chapter therefore should be considered and a relatively small gating system,
as guidelines to create surfaces with increase the risk of unacceptable
minimized defects rather than absolute sinkmarks.
quality criteria. ■ FIGURE 33
6.2.2 Weld lines Gas-assist Injection Moulding
6.2 Surface defects Technology can be used to
A further phenomenon causing possible reduce the sinkmark behind the
The surface defects discussed below are surface defects is the formation of weld boss. Gloss differences however
only a selection of the wide variety lines which occur where flow fronts may occur!
which is possible, and are more or less meet. It is recommended to locate the
related to design. It should be noted injection points in the least sensitive
that mould steel quality and surface areas, (■ F I G U R E 3 4 ), considering
finish of the cavity play a major role in critical locations such as around holes
hiding surface defects. or between gating points.

6.2.1 Sink marks Flow simulation is a very useful way of

predicting the location of weld lines
Sink marks typically occur where before the gating system is cut into the
projections such as ribs or bosses meet mould. Adequate venting in the weld
the main surface of a plastic compo- line areas at the outer edge can reduce ■ FIGURE 34
nent. The cause of this type of sink mark the visibility of weld lines, but, in most The position of the gate 35
is the local increase of thickness at the cases, will not eliminate them. determines the location of the
location of the projection in the visible weldline. Around apertures a
main wall, which suffers a higher than 6.2.3 Air traps corner is quite often the best
average shrinkage. (Refer to Chapter 8 location to position a weldline
‘Design for Mouldability’, ■ F I G U R E 5 0 Air traps are undesired areas in a from cosmetic point of view.
and 5 1 ). To minimize this effect, the moulding which occur when there is
mating wall thickness should be kept insufficient venting at the edge of a weldline
down to 50% of the main wall. moulding, or when air is trapped as the
outer area of a component fills before
The amount of shrinkage is also the area closest to the gate. gate
influenced by gate design, material type This can result in burned spots due
and process conditions. The risk remains to diesel effects.
Design guide 6 Design for appearance

6.2.6 Delamination
Quite often this can be avoided by a
■ FIGURE 35 slight wall thickness increase between Delamination occurs when there is
the gating point and the air tra flow insufficient adhesion between the
Sharp transitions might cause
leader. frozen skin and the molten inner
the plastic to loose contact with
material of a plastic component and,
the mould for a short period of
Rib shapes blindly spark-eroded into a consequently, layers can be peeled off
time. This can result in
mould can also cause air traps. easily. The main causes are processing
This might cause problems when filling with material which is too hot and/or
the rib, and the material might overheat with injection rates which are too high.
by diesel effects. Venting has to be If the problem cannot be solved by
flow direction provided in this case by ejector pins, slower injection rates and suitable melt
porous inserts or other venting conditions, the gating system can be
constructions. modified with larger gate and runner
dimensions. This will reduce shear rate
Air traps can also occur on the surface which can be one of the causes of local
of a part when sudden changes in wall material degradation.
thickness occur. The melt stream locally
possible airtrap or material damage
loses contact with the cavity surface and, 6.2.7 Jetting
as pressure builds up, is then forced
Improved design against the surface with a volume of air Small gate openings directing the
between the plastic and the mould. polymer flow into an open space can
flow direction This air is compressed at very high cause jetting. The polymer expands
speeds, which can raise the temperature after passing the gate, cools rapidly and
and degrade the surface locally, (refer builds a strand of relatively cold
to ■ F I G U R E 3 5 ). In practice this can material in the cavity, (see ■ F I G U R E 3 6 ).
cause for example bad paint adhesion After a certain time, pressure builds up
or poor impact behaviour. in the cavity and the remaining
injection volume fills the cavity
6.2.4 Voids normally. The initial cold strand is
visibly embedded in the rest of the
Voids can be described as air bubbles polymer and causes a reduction in the
which are only visible in transparent mechanical strength of the material.
materials such as Lexan® resin. To avoid jetting, gates should be
They can occur in very thick areas of a positioned so that the polymer stays in
component and are caused by excessive contact with the cavity wall after passing
shrinkage. Normally this phenomenon the gate opening, (■ F I G U R E 3 7 ).
can be avoided by not using thick
sections relative to the rest of the part. Injection speed and melt temperature
If these cannot be avoided, special care can also influence jetting to a certain
should be taken to design a large extent; the lower the injection speed
enough gating system and apply and melt temperature, the lesser the
sufficient packing pressure and time. risk of jetting.

Voids can also occur due to insufficient 6.2.8 Gate marks

predrying or too much decompression
in the barrel in front of the screw. When small gate openings are used in
combination with high injection speeds,
6.2.5 Streaks there is a risk of gate marks. These will
show up in general as matt spots around
Streaks on the surface of a plastic the gate area. The explanation of this
component can be caused by moisture, phenomenon is that the material is
degradation in the machine barrel or extremely highly oriented in and
hot runner system or by overshearing. directly after the gate. The outer layer is
Looking at the relationship with design, frozen in this highly oriented situation.
overshearing can occur due to a gate After filling, the inner layer can relax
being too small. Computer simulation much more than the oriented thin outer
can help to determine the level of layer. This creates high stress levels
shearing and adapt the gate or runner during cooling, causing micro cracks.
dimensions to achieve an acceptable These will again show up as matt spots,
level of shear. Typical safe shear levels in because the light refraction is different
gates are 20.000 - 25.000 sec-1 for GE compared to the environment of the
Plastics’ engineering thermoplastics. gate area. This problem can also be
reduced by larger gate and runner
dimensions and slower injection speeds.
Design guide 6 Design for appearance

6.2.9 Summary

Surface defects can be divided into ■ FIGURE 36

the following categories:
Small gates positioned opposite
open cavity area can produce
(a) Material degradation
jetting. The strand cools down
Material degradation can occur in the
and later on is embedded in the
machine barrel or the hot runner
part, causing bad cosmetics and
system, (refer to Chapter 7 for process- lower mechanical properties.
related factors). Hot runner systems are
quite often the cause of material
degradation. Wrong dimensions, Side view
inadequate temperature control, and
hang up areas are the most important
factors. Flow simulation can help to
determine the correct hot runner
channel diameters. They should be big
enough to prevent overshearing and
small enough to achieve reasonable
Top view
residence times.

(b) Moisture in the material

This problem can be prevented by
proper predrying of the material before
moulding. Water leakage in the mould
can also be a cause of moisture streaks.

(c) Component design

Wrong wall thickness ratios of bosses,
ribs and other projections are common
causes of sink marks. Basic design ■ FIGURE 37
principles can help to avoid certain
To avoid jetting put the gate in
surface defects. A typical ratio for
such a way that pressure is
features perpendicular to the visible
build up immediately in the
surface is 50% of the main wall
beginning of the filling stage.
thickness. The location of the feature
relative to the gate position is
important: the further away from the
gate, the bigger the risk of sink marks.

Wall thickness transitions are another

cause of visible marks on the surface of
the part. These should be smooth and
not sudden.

Sharp corners in a mould can cause air If the design can not be changed
traps or damage the passing material. according the example
This should be avoided whenever illustrated above, an ejectorpin
possible. can be used for gating.

(d) Runner and gate design

Small cold runners and gates create
high shear rates and temperatures in
the melt during the filling stage which,
as discussed, can cause several types of
surface defect. Flow simulation can
assist in determining runner and gate
sizes to achieve acceptable shear and
pressure levels.
Design for precision

7.1 Introduction

Designing for precision can be defined ejection. As the part cools down to room This problem can be minimized by
as applying guidelines when designing a temperature, final mould shrinkage will optimizing the cooling circuit.
component and the corresponding be reached. Quite often thisphenome- This means minimizing the temperature
mould, to mass produce thermoplastic non creates warped parts, because the difference of the cooling medium
components using the injection shrinkage of the component during between the beginning and the end of
moulding process, within as narrow as cooling to room temperature is not cooling circuits, and selecting the right
possible dimensional tolerances. even, depending on the design and the distance to the cavity wall and the right
The discussion of precision is strongly injection moulding process. distance between channels.
related to the way a plastic component
shrinks after the injection moulding 7.1.2 Secondary effects Areas which are difficult to cool can be
process. This chapter deals with various avoided by part design or through the
shrinkage mechanisms which occur Sometimes components are heated after use of highly conductive metals in the
both during moulding and as secondary moulding for example during paint relevant sections. Corners are difficult
effects. curing. This operation can also cause to cool evenly, as there is always the
both permanent and temporary tendency for the outside to cool faster
7.1.1 Mould shrinkage dimensional changes. than the inside. As can be seen in
External loads and moisture absorption ■ F I G U R E 3 9 , this phenomenon can

Shrinkage or mould shrinkage can be have an added influence on dimensions. cause walls of boxes to warp inwards.
defined as the difference between the
moulded component dimensions and 7.2 Shrinkage phenomena
the corresponding mould cavity
dimensions. Normally this phenomenon There are three key factors governing ■ FIGURE 39
is expressed as an average percentage. shrinkage behaviour: cooling, packing
Corners are difficult to cool at
This is used by the mould maker to add and orientation. the same rate internally and
to the desired component dimensions as externally. Tendency to distort.
a target for the mould dimensions. 7.2.1 Cooling

As can be seen in ■ F I G U R E 3 8 , the Uneven cooling can cause differential

shrinkage of a component is a time/ shrinkage. Uneven cooling is caused by
temperature related process. Most of mould surface temperature differences
the shrinkage occurs directly after part during the cooling process.

Shrinkage %

Most of the shrinkage

occurs directly after
part ejection. As part Applying a generous radius the
cools down to room cooling can be more even, hence 39
temperature final less problems with distortion
mould shrinkage will
be reached

Design guide 7 Design for precision

7.3 Materials and shrinkage

This effect can be reduced by giving This can be explained by the PVT
the corners an internal radius and by relationship of thermoplastics, 7.3.1 Amorphous materials
keeping wall thickness constant. (see ■ F I G U R E 4 0 ).
Location of cooling lines inside the part Amorphous materials exhibit lower
can be optimized by positioning the In order to really influence the packing shrinkage than semi-crystalline
channel closer to the corner. stage, it is important to create materials. Furthermore, the levels of
Computer simulation programmes can sufficiently large gate openings and cold shrinkage in flow and cross-flow
assist in this process. runners as the moment of gate or direction are closer for amorphous
runner freeze-off determines the end of materials.
Depending on the nature of the plastic effective packing. It is therefore easier to produce precise
used, the level of mould temperature parts with amorphous, unfilled
determines the degree of crystallinity, 7.2.3 Orientation materials than with fibre-filled or semi-
which has an influence on the level of crystalline materials.
volumetric shrinkage. The third important factor related to Typical examples of GE Plastics’
shrinkage behaviour is the orientation materials with a mainly amorphous
7.2.2 Packing of molecules and fibres. Orientation character are Lexan ® polycarbonate,
effects cause different shrinkage in flow Ultem ® polyetherimide resin, Cycolac ®
The packing or holding pressure phase and cross-flow direction, which can ABS resin, Cycoloy ® polymer blend and
has a significant effect on shrinkage. influence the flatness of parts Noryl ® PPO ® resin.
In general, the higher the holding significantly.
pressure and the longer it is effective, Apart from material choice, the method
the smaller the shrinkage. of gating plays a major role.

■ FIGURE 40 P 0 MPa
Specific Volume m3/kg

Typical PVT P 50 MPa

relationship for
amorphous plastics. P 100 MPa
Higher holding P 150 MPa
pressure reduces
volumetric shrinkage


Troom Tfreeze Temperature

Specific Volume

Typical PVT
relationship for semi-
crystalline plastics.
Crystalline materials
shrink more than
40 amorphous ones due
to crystallinity.

Design guide 7 Design for precision

7.3.2 Semi-crystalline materials 7.4.2 Ribs

Semi-crystalline materials exhibit Uneven shrinkage applies in particular ■ FIGURE 41

higher shrinkage than amorphous ones. to parts with ribs, where the ribs are
Centre gated disk with thicker
Therefore parts made out of this type of sometimes 50% of the nominal wall centre section allows for more
material, or similarly fibre-filled thickness to prevent sink marks. controlled holding pressure to
materials, often suffer distortion due to The ribs cool much faster than the main get a flatter part
differential shrinkage. wall, which will shrink more due to a
Typical examples of GE Plastics’ longer available cooling time.
materials with a semi-crystalline In practice this means that it is almost
behaviour are Valox ® thermoplastic impossible to make flat parts with thin,
polyester resin, Lomod ® flexible long ribs, (see ■ F I G U R E 4 2 ).
engineering thermoplastic resin,
Noryl ® GTX modified PPO ® alloy, and Gas-assisted injection moulding is an
some grades of Xenoy ® thermoplastic option which is increasingly used to
alloy. control the differential shrinkage
between the rib and main wall, by
7.3.3 Reinforced materials. creating a hollow channel in the
crossing point of rib and wall. In such ■ FIGURE 42
Both amorphous and semi-crystalline instances, the rib can have the same
Thin rib shrinks less than
materials have a more orthotropic thickness as the main wall,
thick mainwall, resulting in
character when glass-fibre filled; in (see ■ F I G U R E 4 3 ).
permanent deformation
other words, the difference between
flow and cross-flow shrinkage increases. 7.5 Mould related factors
The most difficult polymers for
precision moulding are glass-filled semi- 7.5.1 Gate location
crystalline resins because they already
have a higher shrinkage level, they are The choice of gate location is extremely
sensitive to crystallinity levels and the important when designing for precision
fibres increase the sensitivity to parts. Often the choice is a compromise
orientation. between precision and productivity. For ■ FIGURE 43
To reduce warpage tendency example, an ideal gate choice for a plate
Using gas-assisted injection
thermoplastics are sometimes filled with type of component is a film gate along
moulding technology, ribs
minerals, which provides good one of the sides of the component, (see
can be used with minimised
temperature resistance combined with a ■ F I G U R E 4 4 ). However, this option is
warpage and sinkmarks.
lower warpage tendency, but lower frequently not chosen due to the Gloss differences can occur.
mechanical strength. disadvantages of removing such a gate.

7.4 Design related factors In summary, the chosen gate location

should provide a short flow length,
7.4.1 Part wall thickness uniform orientation, and optimum
pressure distribution at fill.
Wall thickness is an important factor
when designing for precision. Thin
parts are more sensitive to orientation
because they have to be filled quickly,
and the cooling time available is very ■ FIGURE 44
short. This provides minimal time for film gate
Film gate produces more
thin parts to be corrected in the holding
uniform orientation
pressure phase.
comparred to point gate
and therefor less warpage
Parts with different thicknesses suffer tendency
from distortion, because the different
thicknesses exhibit different degrees of
shrinkage. However, the phenomenon
of differential shrinkage can also be
point gate
used as a benefit. A centre-gated disk
type of part, for instance, with warpage
tendency, can be forced into a
repeatable shape by making the centre
section thicker. This is in order to
control the holding pressure for a
longer time and so force a lower level of
shrinkage in the centre of the
component, (refer to ■ F I G U R E 4 1 ).
Design guide 7 Design for precision

7.5.2 Gate type 7.6.2 Mould temperature 7.7.1 Thermal expansion

The gate type is often predetermined Mould temperature has an influence on When a component has to perform at
for cosmetic or economic reasons. shrinkage. A higher mould temperature elevated temperatures, thermal
As with gate location, a compromise is allows for more crystallization of semi- expansion will cause the part to become
often necessary in its selection, because crystalline materials. Furthermore, the larger. Glass- or mineral-filled materials
of the difficulty in combining optimum relaxation time is longer, which means will suffer less from this phenomenon.
precision, cosmetics and cost efficiency. some orientation effects are reduced. If critical stress levels are reached,
permanent deformation can occur,
There are basically two gate types: the (refer to section 7.7.4 - Creep under
round gate like tunnels, direct sprues However the cooling times will be load).
and pinpoints, and the rectangular- longer and extremely high mould
shaped gate such as tab gates and film temperatures require special equipment 7.7.2 Moisture absorption
gates. It can be said that, in general, like oil heaters or high pressure hot
point gates create high orientation water devices. Another effect on dimensions will occur
levels in the gate area, which will through water absorption. PA-based
frequently lead to distortion, whereas 7.6.3 Injection time materials such as Noryl ® GTX resin are
wider film or tab gates tend to bring a prone to this phenomenon.
more uniform orientation into a The time to fill a cavity influences the
component. orientation level in a moulded 7.7.3 Post crystallization
component: the higher the injection
If a point gate is really essential, speed, the more orientation occurs. Semi-crystalline materials, such as
sometimes an increase in the number of At the same time, the filling speed has Valox ®, Noryl ® GTX and Lomod ® resins,
gating points can bring a more uniform an influence on the temperature can exhibit post shrinkage, especially
orientation pattern to a component, distribution at the moment the cavity is when used at elevated temperatures due
as well as giving better control in the filled and the packing pressure starts to to post crystallization.
holding pressure stage due to shorter be effective. Depending on geometry
flow lengths. and chosen gating, both the orientation 7.8 Simulation techniques
and the temperature of the material
7.5.3 Gate size have an effect on the final shape of the Today several computer programs are
part. The injection phase therefore available to calculate the expected
Small gates create more shear, plays an important part in controlling shrinkage phenomena due to the
orientation and pressure loss than the shrinkage and warpage of an injection moulding operation. These
larger ones. They also limit the effect in injection moulded component. programs can assist the design engineer
the packing stage. In general, the gate in evaluating different alternatives of
diameter or thickness should be at least 7.6.4 Packing pressure gating and processing combinations,
60% the thickness of the main wall, as well as comparing the shrinkage
although preferably larger, in order to Packing pressure is one of the most behaviour of different materials.
provide the largest possible processing important variables to consider when
window. Film gates have the advantage trying to control shrinkage. The way a Thermal expansion and, to a lesser
of producing a low shear rate and component is packed determines the extent creep, can be computer-
uniform orientation, but are more final specific volumetric shrinkage in simulated with commercially available
difficult to remove after moulding. each area of the component. The more non-linear finite element analysis
uniform the volumetric shrinkage over programs, (see Chapter 2).
7.6 Processing related factors the whole part, the more likely the part
will come close to the desired shape with
7.6.1 Melt temperature a minimum of distortion.

Melt temperature has an effect on 7.7 Secondary effects

shrinkage as an absolute value. Looking
at shrinkage differences is less After ejection and cool down, a
significant. Of course a higher melt may component can still suffer unwanted
lead to a reduced average shrinkage deformation.
percentage, and hence to less In particular when temperature changes
differences, but it is not a major factor occur during usage, there are a number
in shrinkage control within a of phenomena which can temporarily or
component. permanently cause dimensional changes
due to thermal expansion and post
Design guide 7 Design for precision

7.9 Summary gates, this effect is considerable.

The end result will be increased
Major factors influencing shrinkage differential shrinkage between flow
include: and cross-flow direction.

(a) Material: (g) Holding pressure

Amorphous materials exhibit less The effect of holding pressure level
shrinkage than semi-crystalline and time is one of the most critical
materials; glass fibres increase factors with respect to shrinkage.
sensitivity for orientation but have a High and long holding pressures
lower shrinkage. create less shrinkage than short and
low holding pressures. The size of
(b) Gate location the gating system has to be taken
The choice of gate location dictates into account; a small gating system,
the orientation of material, which (gates and cold runners), can reduce
can have a dramatic effect on the effects of holding pressure
shrinkage and warpage behaviour. dramatically.

(c) Gate shape (h) Mould cooling

The size and type of gate can Mould temperature differences have
influence the level of orientation a further effect on shrinkage.
and hence the shrinkage behaviour. Temperature variations in the cavity
In addition, the gate and cold wall can cause different cooling rates
runner size determines the and hence different degrees of
effectiveness of the packing phase, shrinkage. Therefore mould cavity
which has an impact on shrinkage temperature should be kept as
and warpage. constant as possible.

(d) Part thickness (i) Process consistency

Wall thickness influences the In addition to the basic principles
orientation level of plastic parts: the regarding shrinkage and warpage,
thinner the wall, the more the quality and consistency achieved
orientation due to shorter injection during processing are critical. In
times. Minimum influence on the particular, temperature of melt and
packing stage limits the control over mould, injection profiles and
shrinkage. pressure control play a major role in
assuring shot-to-shot consistency for
(e) Ribs accurate dimensions.
Ribs are normally thinner than the
main wall thickness to reduce sink (j) Secondary effects
marks. This causes a different The impact of additional loads on
shrinkage in rib and main wall, dimensional changes of injection
resulting in distorted parts. In moulded parts during usage should
particular this is a problem with long not be overlooked. Heat loads,
ribs. mechanical loads and water
absorption are typical examples of
(f) Injection speed detrimental external influences.
Injection speed is related to the
orientation of the material: the
higher the injection speed, the more
orientation. Especially in
combination with small pinpoint

Design for

8.1 Introduction

Optimum design for mouldability shown in a graph as flow length versus flow is caused by the internal shear
provides the possibility to obtain wall thickness at a variety of initial melt stresses within the melt. From a
thermoplastic parts in their final temperatures, (see ■ F I G U R E 4 5 ). chemical or molecular point of view, the
finished shape without any secondary viscosity of a material is a function of the
operations and with no waste of Melt flow data are generated using a molecular chain length of adjacent
material. spiral flow tool under processing polymers and the strength of the
conditions which are considered bonding between them. In other words,
The injection moulding process is common for the material. The initial it is a function of the energy required to
widely used for the production of plastic melt temperature, mould temperature cause relative motion between the
components. In the injection moulding and injection pressure must be recorded adjacent molecules.
machine the material is melted and together with the data curve because
homogeneously plasticized by means of they significantly influence the distance Factors affecting viscosity:
a screw inside a heated cylinder. The of flow. The viscosity of a material will largely be
molten and homogeneous plastic mass a function of its temperature and of the
is injected under high pressure via the The melt flow length of a material is a amount and type of any fillers or
machine nozzle into the cavity of the function of viscosity, thermal properties additives present. Viscosity generally
mould. In the mould cavity the material and shear properties. functions inversely with respect to
is cooled down and the part is ejected temperature: an increase in
from the mould when sufficiently rigid. Viscosity temperature usually causes a decrease in
material viscosity. Fillers and mineral
The injection moulding process offers Viscosity determines the resistance of additives tend to increase viscosity, while
many advantages. In particular it offers the material flow due to internal plasticizers, impact modifiers and wear
the possibilities for: resistance at a given melt temperature. enhancers tend to decrease viscosity.
For example, water has a low viscosity, Molecular weight also significantly
· Highly complex parts molasses a high viscosity and molten influences viscosity: higher molecular
· Repeatability of tight tolerances plastic an even higher viscosity. weight polymers generally have higher
· Moulded-in features like bosses, The internal resistance of the material viscosities.
snap-fits, ribs, undercuts and holes
· High quality surface properties
· Ease of automation for large
production runs.
Flow length

8.2 Material issues Melt flow length

melt temp = T1
To obtain large parts from an injection
moulding machine, the material’s flow melt temp = T2
properties are critical.
These are measured based on melt flow
length and melt temperature.
8.2.1 Melt flow length

The melt flow length of a material is a

measure of how easily a material can
flow to fill a part. It indicates the length
that a material can flow under a specific
set of common moulding conditions.
Determination of melt flow length is Wall thickness
Injection pressure: xxx
important when trying to predict if a
Mold temperature: xxx
part can be filled or not. It is commonly
Material: xxxx
Design guide 8 Design for mouldability

8.3 Shrinkage
Common viscosity tests: Conversely, most of GE Plastics’
Some of the common viscosity tests are materials are shear sensitive as they can 8.3.1 General remarks
as follows: be significantly damaged by being
exposed to large amounts of shear. All thermoplastic materials contract
MELT VISCOSITY (MV): Shear curves showing the relationship during a temperature transition from a
This measures the time it takes for a between shear rate and viscosity at molten state to room temperature as
specific amount of molten material different temperatures are required for they cool down and solidify after being
at a fixed temperature to be extruded accurate injection moulding simulation, plasticized. As a result of this, when
through an orifice at a constant (see ■ F I G U R E 4 6 ). designing a mould, it is necessary to
pressure. make the core and cavity slightly larger
8.2.2 Melt temperature in dimension than the finished
MELT FLOW INDEX (MFI): component size. The difference in
This measures the amount of molten The melt temperature of the injection dimensions which exists between the
material which can be extruded through moulded material is also an important mould and component is known as the
an orifice under a fixed pressure over a parameter. Thermoplastic materials mould shrinkage. Mould shrinkage can
specific amount of time. become less viscous as temperature vary considerably, depending on the
increases. Crystalline materials have a mould geometry and processing Thermal properties clearly defined melt temperature, (Tm), conditions. For example, thin walled
above which they lose all structural components exhibit less shrinkage than
The thermal conductivity and specific integrity and flow freely. Amorphous those with thicker walls.
heat of the material will determine how materials soften over a wide
quickly the flow front will cool down, temperature range above their glass When screening materials for particular
which in turn greatly affects material transition temperatures, and therefore applications it is important to
viscosity. do not actually have a specific melting understand the different shrinkage
point or temperature. Above the glass characteristics of amorphous and Shear properties transition temperature the material is crystalline types of resin. The degree of
more ductile and behaves in a rubbery crystallinity of the resin has a major
When shearing occurs in the melt, the manner. influence on shrinkage: crystalline
temperature of the material will rise resins tend to shrink more than
causing the material to flow more easily. amorphous ones which can be
Some materials, such as nylon, heat up explained by examining the
and flow much more easily when morphology of both materials.
exposed to large amounts of shear and Amorphous polymers are isotropic and
yet retain much of their original display approximately equal shrinkage
mechanical properties. These materials in both flow and cross-flow direction.
are said to be shear insensitive. Crystalline polymers are anisotropic and
therefore display different shrinkage in
flow and cross-flow direction.
(For more detail see Chapter 7).

10 3
Apparent viscosity Pa.sec

Factors affecting

10 2


10 2
102 103 104
Apparent shear rate (sec -1)
Viscosity vs shear rate
curves represent different processing temperatures
Design guide 8 Design for mouldability

8.3.2 Warpage
the core of the mould if adequate
Warpage typically occurs in anisotropic draft is not included. The shrinkage ■ FIGURE 47
materials and is caused by different can pull the sides of the part onto
Mould shrinkage
shrinkage in flow and cross-flow the core and lock a vacuum which
direction. prevents ejection.
Differential shrinkage can be caused by
any of the following: 8.4 Cooling time

(a) orientation glass fibres The cooling rate is important to

With unfilled crystalline resins the consider when designing a plastic part Mould cavity aim
greatest shrinkage is usually because it is the primary determining
encountered in the flow direction. factor of cycle time and therefore
With glass-reinforced polymers the processing costs. In general, cooling
shrinkage in the flow direction can time is strongly affected by the thermal
Moulded component
become less than in cross-flow conductivity and specific heat of the (after cooling)
direction due to the orientation of material and the part geometry.
the fibres. Illustrations of mould
shrinkage can be seen in (a) specific heat Mould shrinkage
■ F I G U R E 4 7 and 4 8 . The specific heat of the material expressed as %
dictates the thermal energy per
(b) geometric asymmetry degree of temperature change which
Warpage can also be caused by the is desired.
part geometry. If a number of ■ FIGURE 48
parallel ribs are used on just one side (b) thickness
Shrinkage of anistropic resins
of a part which is intended to be flat, The thickest area of the part will be
shrinkage in the ribs can cause the the last to fully cool down. It will
top surface to warp. therefore dictate the amount of time being moulded
which the part must remain in the
(c) asymmetric cooling tool while cooling.
If cooling is not well-distributed in
the part, differential shrinkage can (c) tooling
lead to warpage. The tooling for the part should be
designed with cooling lines in
(d) highly aligned flow critical areas to remove heat. Up to a
In semi-crystalline polymers, regions point, the greater the capacity of the
of order tend to align themselves cooling system, the faster the part
with flow in parts where there is a can be cooled and removed from the
mould cavity
highly oriented flow pattern. tool. Once a certain point is reached,
the conductivity between the steel
(e) voids and the plastic in the tool will dictate
If a thick section exists in the part, the rate of cooling regardless of a After ejection & cooling
the material on the outside skin will further increase in the cooling (Highly exaggerated)
cool down first and consolidate. system’s capacity. It should be noted
As the centre area over the part wall that excessively cooled cavities can
thickness finally cools down and tries cause the part surface to shrink away
to shrink, the outer skin cannot from the cavity wall. This prevents
accommodate it because it has thermal conduction and required
already frozen off. As it shrinks, the thermal convection from being the Cross flow
material in the centre area pulls primary modes of heat transfer, shrinkage
towards the outer skin, forming a thereby imposing much longer
vacuum bubble or void in the centre. cooling times.

(f) sink marks The cooling rate can be increased by

flow direction
If a thick section exists in the part in thinning part walls and providing shrinkage 47
areas such as intersections of thick improved cooling in the tool. Cooling
ribs with the nominal part wall time for a part is often predicted using
thickness, the centre area in this material cooling data. This is presented
thick section can cause the surface of in the form of a graph showing cooling
the outer wall to be distorted or time as a function of maximum part
pulled in. thickness for a specific material. This is
calculated at the recommended melt
(g) ejection problems temperature, under standard moulding
If a circular, square or cylindrical conditions and with adequate mould
shape is moulded, it will shrink onto cooling.
Design guide 8 Design for mouldability

8.5 Design considerations 8.5.3 Projections

■ FIGURE 49 8.5.1 General remarks Due to shrinkage and cooling time

related issues associated with thick
Wall section transition
When designing thermoplastic sections, the projection determines the
components, general guidelines should size of the isolated mass or thickness.
be considered. Exact recommendations The projection can be determined by
are sometimes impractical, since obtaining a large scale cross-sectional
variables such as component geometry, drawing of the intersection and
production process and individual inscribing the largest possible circle in
requirements may dictate certain design that area, as shown in ■ F I G U R E 5 0 .
Sharp - not recommended features.
When considering alternative designs
8.5.2 Nominal wall thickness for this area, the smaller this circle is,
the better for cooling. A recommended
The nominal wall thickness forms the approach is to estimate the radius as the
basic frame or shell of a component. largest distance in all directions from a
All other details are added to this shell small area of material to reach mould
Tapered - better structure and will affect, or be related steel. This gives the distance which the
to, the part wall thickness. thermal energy must conduct away from
It is important that the nominal part that area of material in order to get to
wall thickness should be correctly the steel where it can be removed.
designed. A wall section which is too
thin can lead to structural failure or 8.5.4 Radii
poor insulation characteristics, whilst a
Gradual - recommended wall section which is too thick, even Chapters 2 and 3, stiffness and strength,
locally, can result in appearance defects discussed a large number of mechanical
and an overweight or over-engineered and performance related reasons for
component. incorporating large radii. They are also
Direction of flow during injection important to the reduction of stress
moulding of compact resins The actual wall thickness is generally concentration which can be defined as
related to the overall size of the part and the magnification of the level of an
gate locations. Also the flow characteris- applied stress in the region of a notch,
tics of the particular material being void or hole. In most cases, the stresses
specified should be taken into account. in these regions will be far greater than
Ideally the nominal wall thickness is calculated stresses which are based on
kept constant due to shrinkage and assumptions as to stress distribution.
■ FIGURE 50 cooling related issues, but in reality this
is usually impractical. Because of the general notch sensitivity
In most applications, a thin, uniform of plastic resins, care should be taken to
wall with ribs is preferred to a thick wall, include generous fillets and radii in
for optimum part requirements, areas which include sharp changes in
strength to weight ratio and cost section or direction. In addition to these
part wall
R effectiveness. structural reasons, the avoidance of
sharp corners:
Where changes in thickness are
involved, care should be taken that the * facilitates part ejection: Sharp internal
direction of the melt flow during the corners tend to stick in the mould as the
moulding process is always from a thick part shrinks onto the core.
area into a thinner section. This flow
direction promotes higher cavity * provides smooth flow: Resin flows
pressures, which minimize sink marks much more easily and less turbulently
and reduce the risk that a part cannot around smooth uniformly radiused
be completely filled. ■ F I G U R E 4 9 shows corners than sharp corners.
a series of wall section transitions, with This reduces the pressure drop around
the arrows indicating the melt flow the corner, thereby reducing the total
direction during injection moulding. required injection pressure.

Wall thickness variation influences * reduces shear: Shear induced by flow

cooling rates of the moulded compo- around sharp corners can degrade the
nent and unequal thickness causes an polymer’s mechanical properties and
imbalance of cooling and possible flow cause blush and other aesthetic
problems during moulding, which can problems.
result in warping and appearance
defects. Additionally, sharp corners act
as stress concentrators which can often
lead to premature failure.
Design guide 8 Design for mouldability

8.5.5 Ribs 8.5.6 Support ribs

To increase the load carrying ability or Support ribs may be considered as a ■ FIGURE 51
stiffness of a plastic structure, it is form of reinforcement for corners, side
Guidelines for proportioning
necessary to increase either the walls or bosses. For the successful
properties of the plastic material or the introduction of support ribs the
sectional properties of the structure. following guidelines are suggested, q
One method of increasing the (refer also to ■ F I G U R E 5 2 ):
component stiffness, without increasing
the overall wall thickness or involving a (a) The thickness of the support rib
large weight increase, is the should be between 50% and 70% of
incorporation of ribs. the component wall thickness.
Ribs offer structural advantages but they (b)The minimum distance between
can also result in warping and faces of successive support ribs T
appearance problems. To achieve a should be twice the component wall
successful rib design the following thickness.
guidelines are suggested, (refer also to
■ F I G U R E 5 1 ): (c) The minimum length of the support t
rib face attached to the component
(a)In order to reduce sink marks on wall should be twice the wall 1. Base tickness t ² 0.5 T
prime appearance surfaces, the base thickness. 2. Height h²3T
thickness of the rib should not 3. Corner radius r ³ 0.25 - 0.4 T
4. Draft angle q ³ 0.5°
exceed 50% of the adjoining wall (d)Generous radii should be
5. Spacing S³2T
thickness. This however, may be incorporated at the ends of the rib.
increased when appearance is less
critical. (e)A minimum draft angle of 0.5
degrees should be incorporated.
(b)To reduce possible overstressing, ■ FIGURE 52
filling and ejection problems, the (f) The minimum length of the support
Guidelines for introducing
height of the ribs should not exceed rib face attached to a boss should be
of support ribs
three times the adjoining wall four times the wall thickness.
thickness. When increased strength D
is required, more ribs of the
specified proportions are
recommended in preference to an reinforcing edge T
increase in rib height.

(c) A minimum radius of 25% of the

adjoining wall thickness should be
incorporated at the base of the ribs,
since sharp corners act as stress
concentrators. Radii larger than 50%
of the adjoining wall thickness only C
component wall
give marginal improvements and
may result in sink marks on the
opposing surface.
1. Component wall thickness T
(d)Ribs are most effective when placed
2. Rib thickness A
down the length of the area 3. Distance between successive
subjected to bending. rib faces B
4. Length of rib face attached
(e) Rib spacing should be at least twice to component C
the nominal wall thickness. 5. Thickness of reinforcing edge D
· 0.7 T ³ A ³ 0.5 T 49
(f) A draft angle of at least 0.5 degrees · B ³2T
on each side should be incorporated · C ³2T
in order to facilitate release from the

(g)Care should be taken to ensure

adequate tool venting where gas
traps are likely.
Design guide 8 Design for mouldability

8.5.7 Bosses
(e)In order to minimize sink marks
■ FIGURE 53 Bosses are features added to the opposite the boss, it is usually
nominal wall thickness of the necessary to ensure that the core pin
Boss proportions that usually
component and are usually used to partially penetrates the nominal wall.
provide sufficient strength,
facilitate mechanical assembly. To avoid stress concentrations and to
though accompanied by sink
marks and residual stress
Under service conditions, bosses are minimize material turbulence
often subjected to loadings not during mould filling, the head of the
encountered in other sections of a core pin should incorporate a
2D component. Hollow bosses may receive generous radius.
self-tapping screws, ultrasonic welding,
D press-fits or moulded-in inserts, any of 8.5.8 Undercuts
which may exert an excessive hoop
stress on the boss wall. This can be Undercuts of part geometry in the
alleviated by the application of mould should be avoided if possible
suggested boss design principles: through redesign. Ideally, mould tools
T should open in a direction parallel to
(a) General recommendations for the the movement of the injection
thickness of projections from a moulding machine. Depending on the
nominal wall suggest a boss wall component shape, slight undercuts can
W thickness of 50% to 70% of the be stripped from the tool.
nominal wall. This may not, however, This is provided that sufficient taper is
provide sufficient strength to given to avoid scuffing the surface of the
Strong possibility of sink marks withstand the stresses imposed by an component. Certain design techniques
when W > 0.6 T insert. The increased strength can give the desired geometry without
achieved by increasing the section is having to use mechanical devices.
accompanied by sink marks and high Examples of these techniques are
residual stresses. For these reasons a illustrated in ■ F I G U R E 5 6 .
■ FIGURE 54 compromise between aesthetics and
strength is frequently required,
Reduced boss section with
(refer to ■ F I G U R E 5 3 ).
strengthening gussets and
incorporating radius ■ FIGURE 56
(b)Since external forces imposed on a
boss also act on the wall from which
it projects, a minimum radius of 25%
of the wall thickness at the base of
the boss is recommended.

T (c) Further strength may be achieved by
gusset plate supports, as illustrated
in ■ F I G U R E 5 4 .

(d)Attaching the boss to a nearby wall

with a rib results in increased
R = 0.25 T strength. Furthermore, this assists in
E³4T venting during mould filling.
Examples are outlined in
■ FIGURE 55.

Attaching bosses to nearby

Design guide 8 Design for mouldability

For some complex components it is rejoins, it forms a weld- or a flow line. expected temperature fluctuations and
necessary to use side cores, This is always a weak spot in the differences between expansion
mechanically operated cams or loose component and may also be coefficients of the materials.
cores. They can be dealt with if aesthetically undesirable.
necessary as follows: 8.5.10 Draft angles
To minimize these problems, and
(a) cams: encourage the formation of strong weld In order to facilitate component
Cams or pneumatic cylinders move part lines, the following recommendations removal from the mould and hence
of the mould out of the way to permit should be considered. Firstly, the reduce cycle time, a design should
ejection. They are quite expensive in shortest distance between the edges of incorporate appropriate draft angles.
that the mould layout is much more any two holes or slots should be greater For untextured surfaces, 0.25 degrees to
complicated to machine. Controllers are than twice the nominal wall thickness. 2 degrees per side for both inner and
required to operate in the moulding Secondly, when positioning a hole or outer wall is usually sufficient. In certain
cycle. Also the cycle time will be slot near to the edge of a component, applications the use of draw polish on
increased. the shortest distance between the edges the mould surface may allow a smaller
of the hole and the component should angle.
(b) slides: exceed twice the nominal wall thickness.
Angled pins or rods are mounted in the The mould parting line position on the
mould. Part of the mould forming an Blind holes are created by a core part can often be relocated in order to
undercut or hole in the component will supported on only one side of the tool. change or split the required draft.
slide in the direction of the angled pin The core’s ability to withstand the If absolutely no draft is permitted due to
when the mould opens. Proper ejection bending forces induced by the flowing dimensional requirements, a cam or
is then possible. polymer determines the maximum hole slide in the mould may be required.
depth. In general, the depth of a blind An illustrated example of draft angles is
(c) deflect: hole should not exceed three times its given in ■ F I G U R E 5 7 .
Small undercuts can often be deflected diameter or minimum cross-sectional
by bending out the part from the dimensions. For small holes with a 8.5.11 Textures and lettering
mould. diameter of less than 6 mm, the length:
diameter ratio should be kept to 1:2. Textures and lettering can often be
(d) inserts: moulded into the surface of a part. This
Some undercuts can be produced using With through holes it is possible to use can be very helpful in that it can serve as
removable inserts which eject with the longer cores as they are supported on an aesthetic or decorative surface at no
part. This requires an extra operation to both sides of the mould cavity. additional cost. It can also help to hide
replace an insert in the mould for the In general, for through holes the overall surface imperfections such as weld lines.
next cycle and to remove the one from length of a core can be twice as long as
the ejected part. that for a blind hole. Textured side walls require an
additional 0.4 degrees draft per 0.01
(e) stepped parting line: If the hole is to be used to mechanically mm depth of texture, though each
Often the parting line can be relocated fasten the component to a dissimilar individual case should be discussed with
so that there is no longer an undercut. material, allowances should be made for the mould texturing supplier. For larger
This can add some complexity to the expansion and contraction. drafts, up to as much as 10 degrees may
mould design but is the most It is recommended that the diameter of be required for particularly complex
recommended solution if possible. the hole is greater than that of the mouldings with textured finishes.
fastener by 20%-40%, depending on
8.5.9 Coring

The term coring in the injection

moulding process refers to the addition
of steel to the tool in order to reduce or q1 ■ FIGURE 57
eliminate material from a particular
Guidelines for introducing
area of the component.
of support ribs

A core which forms a hole, however,

tends to limit natural material shrinkage q 1 0,25° - 2° for smooth 51
and results in an area of relatively high surfaces
residual stress around the hole. q 2 Larger angle required
To minimize stress concentration for textured surfaces and
points, irregularly shaped holes such as complex geometries
ventilation slots incorporating sharp
internal corners should be avoided. q2
Round holes are generally less
susceptible, since the residual stresses
tend to be evenly distributed. The
inclusion of any hole results in an
interruption of the melt as it flows
around the core. Where the melt
Design guide 8 Design for mouldability

8.5.12 Flow leaders

· difficulty in painting or plating No fill:
Ribs and thickened sections can be uniformly A trapped air pocket in the mould can
designed into a part to function as flow · reduction in impact strength cause the section of the part not to fill.
leaders. While these can be used to help · increase in ductile-brittle failure
fill a part in a predetermined way, they transition temperature High pressure:
can also lead to problems such as If air must be forced out through
backfilling and inappropriate weld line 8.5.14 Weld lines inadequate venting, excessive pressure
locations. may be required to fill the mould, and
Weld lines are areas where two melt high levels of moulded-in stress can
Gate location and size play a big role in fronts come together and bond during result.
determining the effectiveness of flow the filling of a part. Weld lines will
leaders. For example, a large centre always occur where the flow front is Slower fill cycle:
sprue gate positioned in a thick rib on a forced to go around an obstacle, like a If venting is inadequate, the fill cycle
large flat part will cause the resin to flow hole, or where flow fronts meet, for can be slowed down.
rapidly down the rib and then fan out to example coming from two different
fill the part. gates at opposite ends of the part.
Because melt fronts can be marginally Tool corrosion:
A further example is if a large flat part cooled by contact with the mould, they Air which is trapped in a mould heats up
has two crossing diagonal support ribs are often slightly less than fully molten. while retaining its atmospheric
and the gate is located onto one of Thus the strength of a part in a weld line moisture. This can result in tool
them. The flow will often first proceed area and the levels of moulded-in stress corrosion in those areas.
all of the way along the ribs. When the in a weld line area are reduced and
flow front reaches the end of the ribs, increased respectively. Burn marks:
resistance to flow and hence pressure Trapped air pockets can heat up and
will rise. In this case, the material will Parts should be designed with respect to parts can suffer burning or
begin to fan out into the flat portion of weld lines as follows: discolouration.
the part all of the way along both ribs.
The flow fronts from this fanning out · Locate weld lines in structurally There are a number of common types of
can meet in the centre of the sections non-critical areas. venting:
between the ribs forming four large · Locate weld lines in aesthetically
weld lines in the part surface. This effect non-critical areas. Parting line:
can be reduced by thinning the ribs or · Texture part surface in weld line A venting system can be designed into
relocating the gates. area if possible to hide aesthetic the parting line of the mould so that
imperfection. trapped air can escape all of the way
8.5.13 Moulded-in stress · Locate weld lines in areas where around the part. Parting line gates are
gussets or thickened ribs can be used to relatively inexpensive because they can
Moulded-in stress is present in all strengthen them if required. be cut into the surface of the mould
moulded plastic parts to some degree. For example, if a weld line occurs on a along with the part geometry.
In general, high levels of moulded-in cylindrical boss, locate a support gusset
stress are undesirable and are caused by at the weld line location. Ejector pins:
any of the following factors: · Use diaphragm or ring gates for Flats can be ground onto ejector pins to
cylindrical parts when possible to produce venting by allowing the gases to
· high pressure in the moulding cycle eliminate weld lines. escape through the gap in the round
· cold tool temperature · If weld lines cannot be tolerated, pin hole. Ejector pin gates are often
· thin sections in high flow areas, locate them in a removable portion of used in blind pockets in the mould,
especially in gates the part, as is done for overhead such as those for part projections.
· fine details at the extreme of flow fluorescent light covers.
· cold flow front from long flow lengths, Dummy pins:
including runners 8.6 Processing considerations Dummy pins can be used similarly to
· sharp corners in flow path ejector pins to provide venting in
· high viscosity materials 8.6.1 Venting critical areas.
· cold metal inserts in the mould
· very fast cooling of plastic parts Adequate venting is essential to the Vent sizes will depend upon the
production of consistently high quality characteristics of the individual material
High levels of moulded-in stress can moulded parts. In general, it is better to and moulding parameters. Vents which
result in: put in more vents than the minimum are too small can be ineffective and
required because this will facilitate flow easily clog and become useless. Vents
· lack of optical clarity and assure removal of gases from the which are too large can flash under
· acceleration of chemical attack mould. There are a number of problems some types of moulding conditions.
· reduction in load bearing capacity which can result from inadequate The number of required vents will
· warpage venting: largely depend on the process,
· inability to hold tight tolerances - geometry, material and mould quality.
anisotropic shrinkage
Design guide 8 Design for mouldability

8.6.2 Gating
from the part after moulding which (g) diaphragm gate:
As illustrated in ■ F I G U R E 5 8 , there is a leaves a mark on the edge or side of the A circular gate inside the end of a
wide variety of gate types which can be part. Due to sharp angles in the gate cylindrical part section, the diaphragm
used for an injection moulded area, melt shear can be more severe for gate is ideal for the filling of circular
component. a tab gate in comparison to the similar parts. This is because no weld line forms
but preferred fan gate. when the resin enters all of the way
(a) sprue gate: around the cylinder. However, this gate
Sprue gates are usually located in the (d) fan gate: is among the most expensive to remove
top of the centre of the part with a cold- The fan gate is similar to a tab gate but since it requires a mechanical
slug-well opposite the gate. They are in provides large blending radii to reduce operation.
general considered to be the best type shear. A fan gate must be trimmed or
of gate with respect to filling machined off and will leave a mark. (h) ring gate
possibilities. However, they must be However, it is the best type of side gate In this construction, a large diameter
machined off as a secondary operation for material flow. runner goes all of the way around the
to the injection moulding process. outside end of a cylindrical part and
(e) tunnel or submarine gate: material enter from all sides. As with
(b) pin point gate: These are very small gates on the sides diaphragm gates, ring gates are very
Similar to a sprue gate, the centre gate of a part. They are self-degating and good for circular section parts. They
has a reverse taper and is self-degating. leave very small marks, but cause large contain the weld line in the gate area
It is only feasible in a 3-plate tool amounts of shear in the material. and do not typically produce any weld
because it must be ejected separately Tunnel gates should have an included lines in the part. However, they must
from the part, in the opposite direction. angle of approximately 45 degrees. also be mechanically removed. Ring
The gate must also be weak enough so gates are relatively inexpensive because
that it will break off without damaging (f) jump gate: they can be put in the parting line of the
the part. The jump gate provides gate location on mould and easily machined in the
the bottom of the edge of the part. mould half.
(c) tab gate: This is very desirable from an aesthetic
A tab gate is a straight gate into the side point of view, but can be expensive to
of a part. It is very inexpensive to machine and cause problems with shear.
machine. However, it must be removed ■ FIGURE 58
Common gate types

(a) Direct centre sprue (b) Pin point gate (c) Tab gate
‘0’ minimum 5 mm sucker pin
taper 3-5 degrees
keep sprue length
break sharp runner plate
short as possible
corners minimum R 0,5 gate
retainer plate

cold slug well gate diameter

(d) Fan gate (e) tunnel gate (f) Jump gate

part 10 mm maximum
non-impinging sprue gate
2x runner
50% T
full round runner minimum
wall thickness
part thickness = 53
gate thickness runner

(g) Diaphragm gate (g) Ring gate


L cross section
t = minimum gate thickness - 10 mm
L = maximum gate land length - 1 mm

runner gate
Design guide 8 Design for mouldability


As general recommendations, gates

should be:

· As large as possible to permit high

speed and low shear material flow
· Radiused to reduce shear and
facilitate flow
· Designed to include a cold slug well
to catch the cooled melt front
· Located strategically to control weld
line locations and flow length
· In thicker areas for injection
moulded parts

8.7 Ejection

There is a variety of ejector mechanisms

used in injection moulding machines
for plastic components:

(a) pins:
Pins are the most common types of
ejector. Vents can often be machined on
round ejector pins. Care must be taken
to ensure that there are enough ejector
pins of sufficient diameter to properly
eject the part without damage. It is also
important that the ejection force is
equally spread over the part for proper
ejection from the mould.

(b) blades:
Blades are very poor ejectors for a
number of reasons. They often damage
parts, they bend or break and require a
lot of maintenance. Furthermore,
rectangular slots must be machined into
the mould to fit the blade ejector. Blade
ejectors are most commonly used in the
base of pockets which produce ribs on
the part.

(c) plates:
Plates can be put in the mould to
produce large ejection surfaces. Ejector
plates reduce part damage on ejection.

(d) stripper bars:

A stripper bar is a type of plate which
takes the form of a bar attached to some
heavy duty ejector pins. The back end of
a stripper bar should be tapered to
prevent alignment problems.

(e) air valves:

Air valves are commonly used to assist in
ejection but not as a primary ejection
system. They release pressurized air
when opening the valve.
Design for

Manufacturing economics combined This is achieved by using fewer compo- induction coils or powdered metals.
with environmental concerns are nents through, for example, component When applied to both bonded and
forcing the design engineer to re-think integration, and by utilizing assembly welded assemblies the ‘break-out’
the approach to product design, and features that can be ‘designed-in’ to technique also facilitates rapid
consider the disposal issues arising after thermoplastic components. separation of the components.
the product is discarded. Because of the
waste disposal problems associated with The following points are general (d)Where possible make the
plastics, it is necessary to think in terms guidelines for the design of new components from the same material
of recyclability. This means that the products which are intended to be type and grade. Where this is not
material from which the ‘initial’ compo- recyclable. possible, code the components for
nent was manufactured is reground and easy identification of the material
used for several generations of other (a) Avoid using ‘moulded-in’ metal type, by, for example, using barcodes
components, each requiring a reduced reinforcements in plastic or standard moulded-in colour
material property profile compared to components. These are difficult to identification chips. When
that of the preceding component. separate and make recycling dismantling assemblies which use
A typical example of a totally recyclable uneconomical. This also applies to different resin types and grades, a
plastic design would be one that uses no metal inserts for bushings or coding procedure for materials
metal components, is easily machine screws. enables them to be easily identified
disassembled, uses no adhesives or and sorted.
decorative/protective finishes, and has (b)Where possible avoid using
components which are manufactured self-tapping screws, and take (e)Avoid where possible using for
from the same resin type and grade. advantage of the inherent properties example decorative paints, lacquers
of thermoplastics to design snap-fit and protective coatings. Carefully
To achieve a totally recyclable design is, assembly features. selected resin colours in
of course, a monumental task when combination with surface texturing
considering legislative, regulatory, (c) Bonding and cementing with can sometimes give an ideal cosmetic
technical and cosmetic demands, which polyester- and polyurethane-based appearance without having to use
all impose restrictions on component/ materials should be avoided as these such coatings.
product design and material selection. contaminate the recycled material.
The key criterion for ease of If unavoidable then design ‘break- (f) When hot-foil decoration or printing
recyclability is ease of disassembly which out’ features in the area of the bond is needed, separate, easy-to-remove
in turn is dependent on simplified line to facilitate removal. This also secondary mouldings can be used as
assembly. applies to welded assemblies using the carrier/base.

Design for automation

Automation is a generic term used to Automation is an integral part of * Snap-fit features should be designed
describe all production methods which product design which must be into components wherever possible, as
utilize mechanized assembly techniques. considered at the initial design stage. screws and bayonet fixings are among
These include pick and place devices, The following points should be viewed the most difficult fastenings to assemble.
manual manipulators and fully as guidelines when designing for
robotized production systems such as automation: * Components should incorporate
computer-aided manufacturing guide surfaces and location features to
technology or advanced manufacturing * Assembly can be rationalized through help position components for ease of
techniques. product design by simplifying assembly.
component assembly or by eliminating
Efficiency, flexibility, economy, quality it through component integration. * It can be generally assumed that
and reliability of end products are all products which are well-designed for
factors in the drive towards automation. * Integration of functions reduces automated assembly are equally easy to
The key to automation is design. components and sub-assemblies and assemble manually.
High performance engineering creates larger, more easily handled
thermoplastics provide valuable parts. * Components that are equally easy to
opportunities for greater integration of assemble and dismantle provide a good
functions than can be achieved with * Reducing the number of components basis for recycling.
traditional construction materials such to a minimum is essential to facilitate
as metal, wood or glass. rapid assembly.
Such integration results in fewer
components and reduced inventory, but * It is good design practice to create a
requires a new approach to product chassis as the carrier, with all
design. components being fixed by snap-fit
features ‘on-line’.


Complementary reading

Roark’s formulae for Stress and Strain Polymer Engineering

Warren C. Young H. Leverne Williams
McGraw-Hill Book Company, New York Elsevier Scientific Publishing, Oxford

Modern Plastics Encyclopedia Konstruieren mit Kunststoffen

McGraw-Hill Book Company, New York G. Schreger
Hanser Verlag, Munich
Mechanics of Materials
H.J. Hearn Engineering Thermoplastics
Longman Press, London J. Margous
Marcel Dekker Inc., New York
Design Engineering
R.J. Matousec Thermoplastics: Materials Engineering
Hermes Verlag, Berlin L. Mascia
Applied Science Publishing Ltd.,
Mechanical testing of materials London
A.J. Ferner George Newnes Ltd.,
London Plastics materials
J.A. Brydson
Structural design with Plastics Butterworth Scientific, London
B.S. Benjamin
Van Norstrand, New York Machinery’s Handbook
E. Oberg, F.D. Jones, H.L. Horton
Designing with Plastics Edited by H.L. Ryffel
G.W. Ernstein and G. Erhard Industrial Press Inc., New York
McMillan Publishing Company,
New York

GE Plastics

Design guide
GE Plastics in Europe GE Plastics in India Hong Kong Singapore
GE Plastics Hong Kong Ltd. GE Plastics Singapore Pte Ltd.
European Headquarters GE Plastics India Ltd. Room 1088 - Tower 1 c/o 23 Benoi Road
General Electric Plastics B.V. 405-B, Sector 20 The Gateway, Tshimshatsui Singapore 2262
1 Plasticslaan, PO Box 117 Udyog Vihar Phase - III Kowloon Tel. 65 846 3290
NL - 4600 AC Bergen op Zoom Gurgaon, Haryana - 122 016, India Hong Kong Fax 65 861 3063
The Netherlands Tel. (91) 124 341 801 to 806 Tel. (852) 2629 0827
Tel. (31) (164) 29 29 11 Fax (91) 124 341 817 or 815 Fax (852) 2629 0800 Taiwan
Fax (31) (164) 29 29 40 GE Plastics Taiwan
Japan 9/F 37 Min Chuan East Road Sec 3
Headquarters Sales Region GE Plastics in South Africa GE Plastics Japan Ltd. Taipei 10462
General Electric Plastics B.V. Nihombashi Hamacho Park Building Taiwan, Rep. of China
Gagelboslaan 4 GE Plastics South Africa 2-35-4, Nihombashi-Hamacho Tel. (886) 2 509 2124/6
NL - 4623 AD Bergen op Zoom General Electric South Africa (Pty) Ltd. Chuo-ku, Tokyo 103 Fax (886) 2 509 1625
The Netherlands 15th floor Sandton Office Tower Japan
Tel. (31) (164) 29 23 91 Sandton 2146 Tel. (81) 3 5695 4888 Thailand
Fax (31) (164) 29 17 25 Johannesburg, South Africa Fax (81) 3 5695 4859 GE Plastics Thailand
Tel. (27) 11 784 2108 21st Floor
Benelux Sales Region Fax (27) 11 784 2216 Korea Thaniya Plaza Building
General Electric Plastics B.V. GE Plastics Korea Co. Ltd. 52 Silom Road
Gagelboslaan 4 231-8 Nonhyun-Dong Bangkok 10500
NL - 4623 AD Bergen op Zoom GE Plastics in the Americas Kangnam-Ku Thailand
The Netherlands Seoul 135-010 Tel. (66) (2) 231 2323
Tel. (31) (164) 29 11 92 Worldwide Headquarters Republic of Korea Fax (66) (2) 231 2322
Fax (31) (164) 29 17 25 United States Tel. (82) 2 510 6250/1
GE Plastics Fax (82) 2 510 66 66/7
United Kingdom 1 Plastics Avenue
GE Plastics Limited Pittsfield, MA 01201
Old Hall Road USA
Sale Tel. (1) (413) 448 7110
Cheshire M33 2HG United Kingdom Fax (1) (413) 448 7493
Tel. (44) (161) 905 50 00
Fax (44) (161) 905 51 19 Canada
GE Plastics Canada Ltd.
Germany 2300 Meadowvale Boulevard
General Electric Plastics GmbH Mississauga, Ontario L5N 5P9
Eisenstraße 5 Canada
D - 65428 Rüsselsheim Germany Tel. (1) (905) 858 5774
Tel. (49) (6142) 6010 Fax (1) (905) 858 5798
Fax (49) (6142) 65746
France GE Plastics - Mexico S.A. de C.V.
General Electric Plastics France S.à.R.L. Av. Prolongacion Reforma #490, 4o. piso
Z.I. St. Guénault B.P. 67 Colonia Santa Fe
F - 91002 Evry-Cedex France 01207 Mexico, D.F.
Tel. (33) (1) 60 79 69 00 Tel. (11) 525 257 6060
Fax (33) (1) 60 77 56 53 Fax (11) 525 257 6070
Italy Brazil
General Electric Plastics Italia S.p.A. GE Plastics South America S.A.
Viale Brianza, 181 Av. Nações Unidas, 12995 - 20° andar -
I - 20092 Cinisello Balsamo (Milano) Italy Cep 04578.000
Tel. (39) (02) 61 83 41 São Paulo - SP
Fax (39) (02) 61 83 42 11 Brazil
Tel. (55) 11 5505 2800
Russia Fax (55) 11 5505 1757
General Electric International A/O
Kosmodamianskaia Nab, 52
Tel. (7) (095) 935 7312 Pacific Headquarters AS PROVIDED IN GEP’S STANDARD CONDITIONS OF SALE, GEP AND ITS
Spain GE Tower #09-00 HEREIN. Each user bears full responsibility for making its own determination as to
General Electric Plastics Ibérica S.A Singapore 0208 the suitability of GEP’s materials, products, recommendations, or advice for its own
Avenida Diagonal, 652-656 Tel. (65) 326 3301 particular use. Each user must identify and perform all tests and analyses necessary to
Edificio D. Planta 3 Fax (65) 326 3303/(65) 326 3290 assure that its finished parts incorporating GEP materials or products will be safe and
08034 Barcelona Spain suitable for use under end-use conditions. Nothing in this or any other document, nor
Tel. (34) (93) 252 16 00 Australia any oral recommendation or advice, shall be deemed to alter, vary, supersede, or waive
Fax (34) (93) 280 26 19 GE Plastics Australia any provision of GEP’s Standard Conditions of Sale or this Disclaimer, unless any such
175 Hammond Road modification is specifically agreed to in a writing signed by GEP. No statement
Sweden Dandenong, Victoria 3175 contained herein concerning a possible or suggested use of any material, product or
GE Plastics Limited Australia design is intended, or should be construed, to grant any license under any patent or
Box 1242, Skeppsbron 44 Tel. (61) 3 794 4201 other intellectual property right of General Electric Company or any of its subsidiaries
S-11182 Stockholm Sweden Fax (61) 3 794 8563 or affiliates covering such use or design, or as a recommendation for the use of such
Tel. (46) (8) 402 40 24 material, product or design in the infringement of any patent or other intellectual
Fax (46) (8) 723 12 92 China property right.
GE Plastics China
Turkey Beijing, 3rd floor * Company not connected with the English company of a similar name.
GE Plastics Turkey CITIC Bldg. No.19
Is Bankasi Bloklari Jian Guo Men Wai Avenue Lexan®, Noryl®, Noryl® GTX®, Noryl® Xtra, Valox®, Ultem®, Xenoy®, Cycolac®, Cycoloy®
Cemil Topuzlu Caddesi Beijing 100004 and Enduran® are Registered Trademarks of General Electric Co., USA.
A Blok, Daire 18 China
81030 Feneryolu Tel. (86) (21) 270 6789 Gelon™ is a Trademark of General Electric Co., USA.
Istanbul Turkey Fax (86) (1) 512 7345
Tel. (90) (216) 386 7104
Fax (90) (216) 386 5812 Design Guide Eng/4M/0597
For your convenience, the preference

! for ÔMax ÒFit VisibleÓ MagnificationÕ

should be changed from 800% to 150%.

set preference continue