Designing With Plastic

Designing With PlasticThe Fundamentals

TDM

-1D

esig

ning

Wit

h P

last

icT

he F

unda

men

tals


2

Table of Contents

Introduction

1. Plastics Materials – An Overview 10

1.1 Classification 111.1.1 Thermoplastics 121.1.2 Thermosets 121.1.3 Crystalline, Amorphous, and Liquid 13

Crystalline Polymers 131.1.4 Copolymers 141.1.5 Alloys 141.1.6 Elastomers 141.1.7 Additives, Reinforcements, and Fillers 14

2. Physical Properties and Terminology 15

2.1 Density 152.2 Specific Gravity 152.3 Water Absorption 152.4 Mold Shrinkage 152.5 Opacity/Transparency 162.6 Elasticity 162.7 Plasticity 162.8 Ductility 162.9 Toughness 162.10 Brittleness 172.11 Notch Sensitivity 172.12 Lubricity 172.13 Homogeneity 172.14 Heterogeneity 172.15 Isotropy 172.16 Anistropy 172.17 Significance of Elasticity, Homogeneity, 18

and Isotropy

3. Mechanical Properties 19

3.1 Basic Definitions 193.1.1 Stress 193.1.2 Normal Stress 203.1.3 Normal Strain 203.1.4 Modulus of Elasticity 213.1.4.1 Proportional Limit 213.1.4.2 Yield Point 213.1.4.3 Ultimate Strength 223.1.4.4 Elastic Limit 223.1.4.5 Secant Modulus 223.1.4.6 Yield Strength 223.1.5 Poisson’s Ratio 22


3

3.1.6 Shear Stress 233.1.7 Shear Modulus 233.2 Relating Material Constants 233.2.1 Direct Shear 233.2.2 True Stress 243.3 Other Measures of Strength and Modulus 243.3.1 Compression Strength and Modulus 243.3.2 Bending Strength and Modulus 253.4 Rate Dependence of Mechanical Properties 253.5 Time-Related Mechanical Properties 263.5.1 Creep 263.5.2 Apparent or Creep Modulus 273.5.3 Creep Rupture 273.5.4 Stress Relaxation 283.5.5 Extrapolating Creep and Relaxation Data 283.5.6 Impact Loading 283.5.7 Izod Impact 283.5.8 Charpy Impact 293.5.9 Tensile Impact 293.5.10 Falling Dart Impact Test 303.5.11 Fatigue Endurance 30

4. Thermal Properties 31

4.1 Melting Point 314.2 Glass Transition Temperature 314.3 Vicat Softening Point 314.4 Deflection Temperature Under Load 324.5 Coefficient of Linear Thermal Expansion 324.6 Thermal Conductivity 324.7 Aging at Elevated Temperatures 334.8 Temperature Index (UL) 334.9 Flammability 334.10 Effect of Temperature on Mechanical 33

Properties4.10.1 Strength, Modulus, and Elongation 334.10.2 Creep 354.10.3 Impact 354.10.4 Fatigue 35

5. Electrical Properties 36

5.1 Conductivity in Solids 365.2 Volume Resistivity 365.3 Surface Resistivity 375.4 Dielectric Strength 375.5 Dielectric Constant (Permittivity) 375.6 Dissipation Factor 385.7 Arc Resistance 385.8 Comparative Tracking Index (CTI) 38

1

2

3

4

5

6

7

8

9

10

Plastics Materials –An Overview

Physical Propertiesand Terminology

MechanicalProperties

ThermalProperties

Electrical Properties

EnvironmentalConsiderations

StructuralAnalysis

Design Considerations forInjection-Molded Parts

Assembly

Machining, FinishingAnd Decorating


4

6. Environmental Considerations 39

6.1 Factors Affecting Environmental Resistance 396.1.1 Stress Level 396.1.2 Temperature 396.1.3 Exposure 396.2 Chemical Compatibility 396.2.1 Reaction 396.2.2 Solvation 396.2.3 Plasticization 406.2.4 Environmental Stress Cracking 406.3 Weathering Resistance 40

7. Structural Analysis 42

7.1 Introduction 427.2 Defining the Structure 427.2.1 Loads 427.2.1.1 Directly Applied Loads 427.2.1.2 Strain-Induced Loads 437.2.2 Support Conditions 437.2.2.1 Free (Unsupported) 437.2.2.2 Guided 437.2.2.3 Simply Supported 437.2.2.4 Held (Pinned) 437.2.2.5 Fixed (also Clamped or Built-In) 437.2.3 Simplifications and Assumptions 447.3 Safety Factors 447.3.1 Failure Criteria 447.3.2 Beam Bending Stress 477.3.3 Shear Stress, Torsion 477.3.4 Shear Stress, Direct Shear 487.4 Pressure Vessels 497.5 Press-Fits 497.6 Thread Strength 507.7 Pipe Threads 517.8 Impact Loads 527.9 Thermal Stress 53

8. Design Considerations for 55Injection-Molded Parts

8.1 Injection Molding Process 558.2 Design Strategy 568.3 Efficient and Functional Design 568.4 Material Selection 578.5 Nominal Wall Thickness 588.5.1 Normal Ranges of Wall Thickness 588.5.2 Structural Requirements of the Nominal Wall 598.5.3 Insulation Characteristics of the 59

Nominal Wall8.5.4 Impact Response of the Nominal Wall 59


5

1

2

3

4

5

6

7

8

9

10




ThermalProperties



StructuralAnalysis


Assembly


8.5.5 Agency Approvals and the Nominal Wall 598.6 Draft 598.7 Structural Reinforcement 608.7.1 Ribs 608.7.2 Other Geometric Reinforcement 628.8 Bosses 628.9 Coring 628.10 Fillets and Radii 648.11 Undercuts 65

9. Assembly 66

9.1 General Types of Assembly Systems 669.1.1 Molded-In Assembly Systems 669.1.1.1 Snap-Fit Assembly 669.1.1.2 Molded-In Threads 699.1.1.3 Press-Fits 699.1.2 Chemical Bonding Systems 709.1.2.1 Solvent Welding 709.1.2.2 Adhesive Bonding 709.1.3 Thermal Welding Methods 719.1.3.1 Ultrasonic Welding 719.1.3.2 Vibration Welding 739.1.3.3 Spin Welding 749.1.3.4 Radio Frequency (RF) Welding 749.1.3.5 Electromagnetic or Induction Welding 749.1.4 Assembly with Fasteners 759.1.4.1 Bolted Assembly 759.1.4.2 Threaded Metal Inserts 779.1.4.3 Self-Tapping Screws 789.1.4.4 Riveted Assembly 789.1.4.5 Sheet Metal Nuts 789.1.4.6 Specialty Plastic Fasteners 79

10. Machining, Finishing and Decorating 80

10.1 Machining 8010.1.1 Drilling and Reaming 8010.1.2 Thread Tapping 8110.1.3 Sawing, Milling, Turning, Grinding 81

and Routing10.2 Finishing and Decorating 8110.2.1 Painting 8210.2.2 Vacuum Metallizing and Sputter Plating 8210.2.3 Electroplating 8210.2.4 Flame Spraying/Arc Spraying 8210.2.5 Hot Stamping 8210.2.6 Sublimation Printing 8210.2.7 Printing 8210.2.8 Decals and Labels 82

List of Symbols 83


6

List of TablesTable 1.01 Typical crystalline and amorphous 13

polymers Table 1.02 General relative polymer properties 14Table 1.03 Common plastic fillers, reinforcing 14

fibers, and other additivesTable 3.01 Typical values of Poisson’s ratio 22Table 4.01 Typical values for coefficient of linear 32

thermal expansion for thermoplasticsand other common materials

Table 5.01 Typical values of volume resistivity 36for thermoplastics

Table 5.02 Typical values of dielectric constant and 38dissipation factor for variousthermoplastics at room temperature

Table 6.01 Chemical resistance of various materials 41by chemical classes

Table 7.01 Design strength for preliminary 44part design

Table 8.01 Typical nominal thickness for various 58classes of thermoplastics

Table 8.02 Dimension difference versus draft 59angle

Table 8.03 Summary of the effect of rib and cross- 61section change in example, Figure 8.07

Table 10.01 Approximate drilling speed and feed 81rate for 1/4–3/8 in. hole in variousthermoplastics


7

List of FiguresFig. 1.01 Example of increasing molecular weight 11Fig. 1.02 Plastic classification 11Fig. 1.03 Thermoplastic molecular chains 12Fig. 1.04 Thermoset cross-linked molecules 12Fig. 1.05 Two dimensional representation of 13

crystalline, amorphous, and liquidcrystalline structure

Fig. 2.01 Density and specific gravity 15Fig. 2.02 Mold shrinkage 15Fig. 2.03 Material exhibiting elasticity / plasticity 16Fig. 2.04 Toughness can be measured by the area 16

under the stress-strain curveFig. 2.05 Anistropy of wood 17Fig. 2.06 Mold shrinkage of anisotropic plastics 18Fig. 3.01 Internal forces, stresses in a body 19Fig. 3.02 Simple tension load 20Fig. 3.03 Typical test setup and specimen 20Fig. 3.04 Plot of results of tensile test (stress- 21

strain curve)Fig. 3.05 Typical stress-strain curves 21Fig. 3.06 Loaded tensile bar showing dimensional 22

change in length and widthFig. 3.07 Shear stress 23Fig. 3.08 Direct shear stress test used in plastics 23

industryFig. 3.09 True stress 24Fig. 3.10 Tensile and compressive moduli 24Fig. 3.11 Beam in bending 25Fig. 3.12 A simple bending fixture 25Fig. 3.13 Typical presentation of creep data 26Fig. 3.14 Creep modulus 27Fig. 3.15 Creep rupture data – a curve showing 27

one cycle log time projectionFig. 3.16 Examples of constant strain loads 28Fig. 3.17 Izod and Charpy impact tests 29Fig. 3.18 Tensile impact 29Fig. 3.19 A typical dart impact apparatus 30Fig. 3.20 Typical S-N curve is shown along with 30

thermal effects which sometimes occurwhen plastics are fatigue tested

Fig. 4.01 Vicat softening point apparatus 31Fig. 4.02 Test apparatus for deflection temperature 32

under loadFig. 4.03 Effect of temperature or strain rate 33

on stress-strain curvesFig. 4.04 Modulus behavior of crystalline and 34

amorphous resin showing Tg and melttemperatures and the effect ofreinforcement on HDT

Fig. 4.05 Effect of temperature on tensile elongation 34

1

2

3

4

5

6

7

8

9

10




ThermalProperties



StructuralAnalysis


Assembly



8

Fig. 4.06 Apparent modulus curves at different 34temperatures showing time-temperatureshifting to estimate expanded time valuesat lower temperatures

Fig. 4.07 Creep rupture curves indicating danger of 34linear projection to longer times atlower temperatures

Fig. 5.01 Free electrons in a metal lattice 36Fig. 5.02 One type of text apparatus for volume 37

and surface resistivityFig. 5.03 A typical test configuration for dielectric 37

strengthFig. 5.04 A typical test configuration for arc 38

resistanceFig. 7.01 Directly applied loads 42Fig. 7.02 Strain-induced loads 42Fig. 7.03 Support conditions 43Fig. 7.04 Section properties for some common 45

cross-sections (na = neutral axis)Fig. 7.05 Maximum stress and deflection equations 46

for selected beamsFig. 7.06 Shaft of diameter, d, and length, L, 47

subjected to torque, T, with angulardeformation, θ

Fig. 7.07 Polar moments of inertia for common 48cross-sections

Fig. 7.08 Direct shear examples 48Fig. 7.09 Cylindrical pressure vessel, thin wall tube 49Fig. 7.10 Cylindrical pressure vessel, thick wall tube 49Fig. 7.11 Press-fit equations for two typical 50

situationsFig. 7.12 Illustration for torque-force relationships 50

for screw threadsFig. 7.13 Designing with pipe threads 52Fig. 7.14 Estimating impact stress and deflection 53

due to dropping a weight on the partFig. 7.15 Typical assemblies in which thermal 54

stresses can become a problemFig. 7.16 Equations of thermal expansions for 54

various conditionsFig. 8.01 Schematic of reciprocating screw injection 55

molding machineFig. 8.02 Injection-molded one piece housing 56

replaces fabricated sheet metal and manymiscellaneous parts and eliminates mostassembly operations

Fig. 8.03 Minimize material usage for plastic 56redesign

Fig. 8.04 Some typical spiral flow curves 58Fig. 8.05 Illustration of draft and associated 59

dimension change with draft angle


9

1

2

3

4

5

6

7

8

9

10




ThermalProperties



StructuralAnalysis


Assembly


Fig. 8.06 Interaction of wall thickness, load, 60stress, and deflection for a rectangularcross-section in bending

Fig. 8.07 Example of the effect of ribs on 60rectangular cross-section

Fig. 8.08 Rib design for reinforcing thermoplastic 61parts

Fig. 8.09 Modifications to nominal wall to improve 62structural response

Fig. 8.10 Design of bosses 63Fig. 8.11 Coring examples 64Fig. 8.12 Typical stress concentration factor 64Fig. 8.13 Creating “undercuts” with simple tooling 65Fig. 9.01 Snap-fit design for cantilever beam with 67

rectangular cross sectionFig. 9.02 Tooling for snap-fit fingers 68Fig. 9.03 Snap-on / Snap-in fits 68Fig. 9.04 Molded plastic threads 69Fig. 9.05 Alternative press-fit designs for metal 69

pins in plastic hubsFig. 9.06 Typical joint designs for solvent and 70

adhesive bondingFig. 9.07 Typical ultrasonic welding equipment 71Fig. 9.08 A basic shear/interference joint 71Fig. 9.09 Basic energy director joint design 71Fig. 9.10A Energy director 72Fig. 9.10B Shear joints 72Fig. 9.10C Scarf joints 73Fig. 9.11 Ultrasonic staking, swaging, and spot 73

weldingFig. 9.12 Typical joints for vibration welding 73Fig. 9.13 Preform in place for electromagnetic 74

weldingFig. 9.14 The electromagnetic welding process 75Fig. 9.15 Five basic joint designs for induction 75

weldingFig. 9.16 Bolt assembly, stress problems 76

and solutionsFig. 9.17 Common threaded metal inserts 77Fig. 9.18 Ultrasonic inserts designed for installation 77

standard ultrasonic equipmentFig. 9.19 Typical screws used with plastics 78Fig. 9.20 A typical boss cap 78Fig. 9.21 A typical push-nut assembly 79


10

IntroductionThis design manual was written to serve as a generalpurpose reference source for the experienced plasticproduct designer as well as the design engineer new toplastics. It should also be of interest to nondesignersand management personnel who need a generaloverview of the concepts and critical issues related tothe world of plastics. Although the manual is not aguide to injection molding, many of the designconsiderations are based upon molding criteria, sothose involved in the manufacturing and processing ofplastic parts should also find it useful.

Most design manuals deal with a specific family ofplastic resins, and present properties, design criteria,assembly, and other information related to theseresins. The product line of Ticona includes crystalline,amorphous, liquid crystalline, and elastomericpolymers. Due to this diversity, this manual deals withissues common to all injection-moldable thermoplasticresins. Tables, figures, and other descriptive methodsare used to help illustrate significant designdifferences.

The first part, Chapters 1 through 6, introduces thenature of plastic materials, their properties and testingmethods. Key advantages, as well as limitations, ofcertain thermoplastics are discussed. Whereappropriate, we compare the properties of variousengineering thermoplastics, often including theproperties of metals or other structural materials.However, our goal is to review fundamental conceptsrather than list the various testing methods.

Discussion of specific testing methods from thenumerous governmental, industrial, and standardsorganizations involved with testing is beyond thescope of this manual.

The second part, Chapters 7 through 10, deals withthe actual design of parts. It starts with structuralanalysis and injection molding considerations andconcludes with assembly, finishing, and decoratingtechniques. This part of the manual will serve as botha general reference and a “how-to” guide for thosenew to plastic design.

For further information on design-related topics, thereader is urged to consult the following individualproduct brochures: Designing with Celcon@ acetalcopolymer (CE-10), Designing with Fortron@

polyphenylene sulfide (FN-10) and Designing withVectra@ liquid crystal polymers (VC-10). These may be obtained by calling the Ticona ProductInformation Services Department at (800) 833-4882.Additional design information on Celcon@ acetalcopolymer, Celanese@ nylon 6/6 and Celanex@

thermoplastic polyester is also available.


11

1

1. Plastic Materials –An Overview

Plastics are synthetic materials called polymers, which are long-chain molecules made up of repeatingunits joined together. These units contain variouscombinations of oxygen, hydrogen, nitrogen, carbon,silicon, chlorine, fluorine, and sulfur. Althoughplastics are soft and moldable and approach a liquidcondition during manufacture, they are solid in theirfinished state. As more repeating units are added, theplastic’s molecular weight increases. Addition of morerepeating units to the chain makes the moleculeheavier. For example, Figure 1.01 shows the simplecompound, methane (CH4), a gas. As molecularweight increases, typical materials are pentane (a liquid), paraffin wax (a solid), and finallypolyethylene (a widely used thermoplastic material).The mechanical and physical properties of plastics are directly related to the bonds between molecularchains, as well as to the chain length and composition.Plastic properties can also be modified both byalloying and blending with various substances and reinforcements.

1.1 Classification The classification of plastics can be extensive andconfusing, as illustrated in Figure 1.02. However, twomajor groups can be identified: thermoplastics, whichare the main focus of this manual, and thermosets,which are discussed only in general terms here. In addition to the broad categories of thermoplasticsand thermosets, polymers can be classified in terms of their structure, i.e., crystalline, amorphous, andliquid crystalline. Other classes of plastics commonlyreferred to in the literature are copolymers, alloys,and elastomers. Finally, additives, reinforcements,and fillers play a major role in modifying properties.Each of these is discussed briefly.

METHANEGASCH4

H

H

H

H

C

PENTANELIQUIDC5H12

POLYETHYLENESOLIDC100H202

Fig 1.01 · Example of Increasing Molecular Weight

Fig 1.02 · Plastic classification


12

1.1.1 Thermoplastics Thermoplastics are resins that repeatedly soften whenheated and harden when cooled. Most thermoplasticsare soluble in specific solvents and can burn to somedegree. Softening temperatures vary with polymertype and grade. Because of thermoplastics’ heatsensitivity, care must be taken to avoid degrading,decomposing, or igniting the material. Nylon, acrylic,acetal, polystyrene, polyvinyl chloride, polyethylene,and cellulose acetate are just a few examples of themany rigid thermoplastic resins currently available.Also within this group are highly elastic, flexibleresins known as thermoplastic elastomers (TPEs) .

Most thermoplastic molecular chains can be thoughtof as independent, intertwined strings resemblingspaghetti (Figure 1.03). When heated, the individualchains slip, causing plastic flow. When cooled, thechains of atoms and molecules are once again heldfirmly. When subsequently heated, the chains slipagain. There are practical limitations to the number ofheating/cooling cycles to which thermoplastics can besubjected before appearance and mechanicalproperties are affected.

1.1.2 Thermosets Thermosets are plastics that undergo chemical changeduring processing to become permanently insolubleand infusible. Phenolic, amino, epoxy, and unsaturatedpolyester resins are typical thermoset plastics. Naturaland synthetic rubbers such as latex, nitrile, millablepolyurethane, silicone, butyl, and neoprene, whichattain their properties through a process known asvulcanization, are also thermoset polymers.

The structure of thermoset plastics is also chainlikeand, prior to molding, very similar to thermoplastics.However, cross-linking is the principal differencebetween thermoset and thermoplastic systems (Figure1.04). When thermosets are cured or hardened, cross-links are formed between adjacent molecules, resultingin a complex, interconnected network. These crossbonds prevent the individual chains from slipping,thus preventing plastic flow when heat is added. Ifexcessive heat is added to the thermoset resin after thecross-linking is complete, the polymer is degradedrather than melted. This behavior is somewhat similarto an egg when it is cooked; further heating does notreturn the egg to its liquid state, it only burns.

LINEAR POLYMER

Fig 1.03 · Thermoplastic molecular chains

CROSSED-LINKED POLYMER

Fig 1.04 · Thermoset cross-linked molecules


13

1

1.1.3 Crystalline, Amorphous, and LiquidCrystalline Polymers In some thermoplastics, the chemical structure allowsthe polymer chains to fold on themselves and packtogether in an organized manner (see Figure 1.05).The resulting organized regions show the behaviorcharacteristics of crystals. Plastics that have theseregions are called crystalline. Plastics without theseregions are called amorphous. All of the crystallineplastics have amorphous regions between andconnecting the crystalline regions. For this reason,crystalline plastics are often called semicrystalline in the literature. Table 1.01 gives some commonexamples of crystalline and amorphousthermoplastics.

Liquid crystalline polymers are best thought of as being a separate and unique class of plastics. The molecules are stiff, rodlike structures that areorganized in large parallel arrays or domains in both the melt and solid states. These large, ordereddomains provide liquid crystalline polymers withunique characteristics compared to those ofcrystalline and amorphous polymers.

Many of the mechanical and physical propertydifferences between plastics can be attributed to theirstructure. In general, the ordering of crystalline andliquid crystalline thermoplastics makes them stiffer,stronger, and less resistant to impact than theiramorphous counterparts. Crystalline and liquidcrystalline materials are also more resistant to creep,heat, and chemicals. However, crystalline materialsrequire higher melt temperatures to process, and theytend to shrink and warp more than amorphous

CRYSTALLINE

CRYSTALS

AMORPHOUS

AMORPHOUS

LIQUID CRYSTALLINE

MELT

SOLID

Fig 1.05 · Two dimensional representation of crystalline, amorphous, and liquid crystalline structure

Typical Crystalline Typical AmorphousThermoplastic Resins Thermoplastic Resins

Acetal PolystyreneNylon ABSPolyethylene SANPolypropylene PolycarbonatePolyester (PET, PBT) PVC

Table 1.01 · Typical crystalline and amorphous polymers


14

polymers. Amorphous polymers soften gradually andcontinuously as heat is applied, and in the moldingprocess they do not flow as easily as moltencrystalline polymers. Liquid crystalline polymershave the high melt temperature of crystalline plastics,but soften gradually and continuously likeamorphous polymers. They have the lowest viscosity,warpage, and shrinkage of all thermoplastics. Table1.02 generalizes the relative polymer properties.

1.1.4 Copolymers A copolymer is a polymer made with two or moredifferent kinds of repeating units. A terpolymer is acopolymer with three different kinds of repeatingunits. When a polymer family includes copolymers,as with acetal resins, the term homopolymer identitiesthe polymer with a single type of repeating unit.Copolymers can have different properties from thoseof the homopolymers making up either repeating unit.

1.1.5 Alloys Alloying is another way to create tailored variationsin plastics. The exact definition of a plastic alloy isnot clear, however it is generally accepted that alloysare combinations of polymers that are mechanicallyblended. They may depend on chemical bonds, butoften have special compatibilizers to join differentconstituent polymers together to improveperformance (e.g., impact strength and chemicalresistance), to lower cost, or to improveprocessability.

Generally, the properties of plastic alloys fall betweenthose of the starting polymers. However, some alloysare able to achieve a synergistic combination that isbetter than the properties of either component alone.

1.1.6 Elastomers Thermoplastic elastomers are generally lowermodulus, flexible materials that can be stretchedrepeatedly to at least twice their original length atroom temperature and are able to return to theirapproximate original length when stress is released.Thermoset rubber materials have been available for a long time, but currently many families of injection-moldable thermoplastic elastomers (TPE’s) arereplacing traditional rubbers. In addition, TPE’s are widely used to modify the properties of rigidthermoplastics, usually improving the impactstrength.

1.1.7 Additives, Reinforcements, and FillersThe physical and mechanical property profile ofplastics can also be modified by adding a wide varietyof fillers, fibers, and other chemical compounds. Ingeneral, mechanical properties are most significantlyincreased by adding reinforcing fibers. Particulatefillers usually increase modulus, and plasticizersusually decrease modulus and enhance flexibility.Flame retardants, thermal and UV stabilizers, andoxidation inhibitors are other common additives.Electrical properties may be affected by manyadditives, especially those that are conductive. When the mechanical properties are improved, the resin is called a reinforced resin. An example isglass-reinforced nylon. When the additive does notsignificantly improve the mechanical properties, butdoes affect the physical nature of the material, theresin is usually called a filled resin. An example ismineral-filled polyester. Table 1.03 lists a variety of common plastic fillers, reinforcements, and other additives.

Fillers Reinforcing Fibers Other AdditivesGlass Spheres Glass Fibers UV StabilizersCarbon Black Carbon Fibers PlasticizersMetal Powders Aramid Fibers LubricantsSilica Sand Jute ColorantsWood Flour Nylon Fibers Flame RetardantsCeramic Powders Polyester Fibers AntioxicantsMica Flakes AntistaticsMolybdenum Preservatives

Disulfide Processing AidsFungicidesSmoke SupressantsFoaming AgentsViscosity ModifiersImpact Modifiers

LiquidProperty Crystalline Amorphous Crystalline

Specific Gravity Higher Lower Higher

Tensile Strength Higher Lower Highest

Tensile Modulus Higher Lower Highest

Ductility, Elongation Lower Higher Lowest

Resistance to Creep Higher Lower High

Max. Usage Temp. Higher Lower High

Shrinkage and Warpage Higher Lower Lowest

Flow Higher Lower Highest

Chemical Resistance Higher Lower Highest

Table 1.02 · General relative polymer properties

Table 1.03 · Common plastic fillers, reinforcingfibers, and other additives


15

2

2. Physical Properties and TerminologyThe following is a discussion of some physicalproperties and fundamental concepts that apply toplastics. Most are familiar to readers who haveworked with other engineering materials. A few,however, are specific to plastics.

2.1 Density The density of any material is a measure of the massper unit volume, usually expressed as pounds percubic inch (lbs/in3) or grams per cubic centimeter(g/cm3) (see Figure 2.01). The density of a particularplastic resin is necessary to calculate the relationshipbetween the weight and volume of material in aparticular part.

2.2 Specific Gravity The specific gravity is the ratio of the mass of a givenvolume of material compared to the mass of the samevolume of water, both measured at 23°C. In otherwords, specific gravity is the density of a materialdivided by the density of water. Since it is adimensionless quality, it is convenient for comparingdifferent materials. Like density, specific gravity isused extensively in determining part cost, weight, andquality control.

2.3 Water Absorption Water absorption is the percentage increase in weightof a material due to absorption of water. Standard testspecimens are first dried for 24 hours and thenweighed before and after immersion in 23°C waterfor various lengths of time. Water absorption isimportant since it affects mechanical and electricalproperties, as well as part dimensions. Plastics withvery low water absorption rates tend to have betterdimensional stability.

2.4 Mold Shrinkage Mold shrinkage is the ratio of the expected reductionof the plastic part dimension as the part solidifies inthe mold and cools to room temperature to theoriginal mold dimensions (Figure 2.02). The moldmaker must know this value to properly size themold to allow for shrinkage. For a given material,mold shrinkage can vary with a number of design andmolding variables, such as wall thickness, flowdirection, and molding conditions. This measure isimportant to the design engineer, not only for new

designs, but also when considering substitutematerials for a specific application. In general,amorphous and liquid crystalline thermoplastics have lower mold shrinkage than do crystallinethermoplastics. In addition, glass-reinforced or filled materials have lower shrinkage than unfilled or “neat” resins.

DENSITY = MASS / VOLUMElbs/in.3 or gm/cc

DENSITY OF MATL. @ 23°CDENSITY OF H20 @ 23°C

Sp. Gr. =

1 1

1

a.)

b.)

Fig 2.01 · a. Density and b. specific gravity

MOLD CAVITY DIM.

MOLDED PART(AFTER COOLING)

MOLD SHRINKAGEEXPRESSED AS

in. cmin. cm OR % SHRINKAGE

Fig 2.02 · Mold shrinkage


16

2.5 Opacity/Transparency Opacity or transparency are important when lighttransmission is a consideration. These properties areusually measured as haze and luminous transmittance.Haze is measured as the percentage of transmittedlight through a test specimen that is scattered morethan 2.5° from the incident beam. Luminoustransmittance is measured as the ratio of transmittedlight to incident light.

2.6 Elasticity Elasticity is the ability of a material to return to itsoriginal size and shape after being deformed (Figure2.03). Most plastic materials have limited elasticity.Rubber and thermoplastic elastomers (TPEs) haveexcellent elasticity. Throughout this manual, it isgenerally assumed that plastics respond elasticallyunless otherwise indicated.

2.7 Plasticity Plasticity is the inverse of elasticity. A material thattends to stay in the shape or size to which it isdeformed has high plasticity. Plastic materials exhibitplasticity when they are stressed beyond the yieldpoint. This accounts for the ability of some plastics tobe cold-formed. Of course, when thermoplastics areheated to their softening point, they have almostperfect plasticity.

2.8 Ductility Ductility is the ability of a material to be stretched,pulled, or rolled into shape without destroying theintegrity of the material.

2.9 Toughness Toughness refers to a material’s ability to absorbmechanical energy without fracturing. A toughmaterial can absorb mechanical energy with eitherelastic or plastic deformation. In general, high-impactunfilled resins have excellent toughness. However,low- or moderate-impact resins can also displayconsiderable toughness if the material has sufficientlyhigh ultimate strength. This will become apparentwhen the stress-strain curve is discussed in Chapter 3. Toughness is often measured by the areaunder the stress-strain curve of the resin, as shown inFigure 2.04.

BEFORELOADING

LOADAPPLIED

AFTERLOADRELEASED

a)

b)

Fig 2.03 · a) Material exhibiting elasticityb) Material exhibiting plasticity

TOUGH MATERIALS

StrainStrain

Stre

ss

Stre

ss

Lower StrengthLower Modulus

Higher StrengthHigher Modulus

Strain

Stre

ss

Strain

Stre

ss

BRITTLE MATERIALS

Lower StrengthLower Modulus

Higher StrengthHigher Modulus

Fig 2.04 · Toughness can be measured bythe area under the stress-strain curve


17

2.10 Brittleness Brittleness is the lack of toughness. Plastic materialsthat are brittle frequently have lower impact andhigher stiffness properties. Many glass-reinforced andmineral-filled materials display brittleness.

2.11 Notch Sensitivity Not to be confused with brittleness, notch sensitivityis a measure of the ease with which a crack propagatesthrough a material from a pre-existing notch, crack,or sharp corner.

2.12 Lubricity Lubricity refers to the load-bearing characteristics ofa material under relative motion. Plastics with goodlubricity tend to have low coefficients of friction withother materials (or with themselves) and do not havea tendency to gall (wear away by friction).

2.13 Homogeneity The term homogeneous means uniform. As onemoves from point to point in a homogeneousmaterial, the material’s composition remains constant.Furthermore, the smallest sample of material cut from anywhere in a homogeneous body has the same physical properties as those of the whole body.An unfilled thermoplastic is a good example of areasonably homogeneous material.

2.14 Heterogeneity Heterogeneous means varying. The material’scomposition varies from point to point in aheterogeneous body. A glass-reinforced plastic isheterogeneous. However, for design purposes, manyheterogeneous materials are treated as homogeneous.This is because a “reasonably” small sample ofmaterial cut from anywhere in the body has the same physical properties as those of the body.

2.15 Isotropy The physical properties of an isotropic material at apoint are the same, independent of the direction inwhich they are measured.

2.16 Anisotropy The physical properties of an anisotropic materialdepend on the direction in which they are measured.The varying degrees of anisotropy depend on theamount of symmetry that exists in the material. A few examples should clarify.

Cast metals and plastics tend to be reasonablyisotropic. Samples cut in any direction within the cast body tend to have the same physical properties.However, rolled metals tend to develop crystalorientation in the rolling direction. Thus, rolled andsheet metal products have different mechanicalproperties in the rolling and transverse directions.Likewise, extruded plastic film can have very differentproperties in the material and transverse directions.Thus, these materials are oriented biaxially and areanisotropic. Composite materials, in which fiberreinforcements are carefully oriented in the directionsof applied load and surrounded by a plastic matrix,have a high degree of property orientation withdirection at various points in the structure and are anisotropic.

Wood is an anisotropic material with distinctproperties in three directions (Figure 2.05). It is verystiff and strong in the growth direction, and has fairproperties in one direction perpendicular to thegrowth direction. However, the mechanical propertiesare much lower in the third direction (perpendicularto the other two directions), as observed by anyonewho has seen a baseball bat shatter.

Y

X

Z

RELATIVE STRENGTH

Fig 2.05 · Anistropy of wood

2


18

While the above examples demonstrate primarilymechanical properties, anisotropy is also involvedwhen a material shrinks in the mold (Figure 2.06).Anisotropic mold shrinkage is an importantconsideration when crystalline and glass-fiber-reinforced materials are being molded. Thesematerials are usually listed with mold shrinkagevalues in the “flow direction” and “cross-flowdirection.” Although these values are mainly ofconcern to the toolmaker and molder, the existenceand severity of anisotropic shrinkage must beconsidered by the design engineer in choosing a resinfor a new part with tight tolerances on the drawing.

2.17 Significance of Elasticity, Homogeneity, and Isotropy In typical structural analyses performed whiledesigning parts, an engineer generally uses twoindependent constants to describe the mechanicalresponse of materials, Young’s modulus and Poisson’sratio (defined in Chapter 3). However, only elastic,isotropic materials that respond linearly to load (i.e.,load is proportional to deformation) can be analyzedby using these two constants. Furthermore, designersuse the same values of these constants everywhere inthe structure. This is only correct if the material ishomogeneous. Thus, in typical structural analyses,engineers generally assume that the material is linearlyelastic, homogeneous, and isotropic, which isreasonable for many analyses and is always a goodstarting point. However, it can lead to significanterrors when one is designing in plastics, particularlyglass-reinforced plastics and liquid crystallinepolymers, which tend to be highly anisotropic.

In this manual, it is generally assumed that plasticsmay be treated as linearly elastic, homogeneous, andisotropic. This affords a simpler presentation ofmechanical properties (Chapter 3), which is consistentwith the data typically provided in the plasticsmarketplace. This assumption is also required by thestandard equations of structural analysis, i.e., bending,torsion, pressure in a pipe, etc. (Chapter 7).

The designer and engineer must be aware that as the degree of anisotropy increases, the number ofconstants or moduli required to describe the materialincreases (to a maximum of 21). Uncertainty ofmaterial properties, as well as questionableapplicability of the simple analysis techniques that aregenerally used, provides justification for extensiveend-use testing of plastic parts before certification in aparticular application. However, please note that asfinite element analysis (FEA) methods become moreprevalent in plastic part design, the ability of FEAmethods to handle anisotropic materials demands abetter understanding of the anisotropic nature ofplastic materials.

GATE

MOLD CAVITY

BEINGMOLDED

(HIGHLYEXAGGERATED)

AFTEREJECTIONS &COOLING

FLOW DIRECTIONSHRINKAGE

CROSSFLOWSHRINKAGE

Fig 2.06 · Mold shrinkage of anisotropic plastics


19

3

3. Mechanical PropertiesMechanical properties are crucial since virtually all end-use applications involve some degree ofmechanical loading. Material selection for a variety of applications is often based on mechanicalproperties such as tensile strength, modulus,elongation, and impact strength. These values arenormally available in the product data sheets providedby material suppliers. More often than not, toomuch emphasis is placed on comparing thepublished values of different types and grades ofmaterials and not enough on determining the truemeaning of mechanical properties and theirrelation to end-use requirements.

In practical applications, materials are seldom, if ever,subjected to a single, steady deformation without thepresence of other adverse factors such as environmentand temperature. Since published values ofmechanical properties are generated from testsconducted in laboratories under standard conditions,the danger of selecting and specifying a material usingonly this information is obvious. A thoroughunderstanding of mechanical properties and testsemployed to determine such properties, as well as the effect of adverse (or beneficial) conditions onmechanical properties over long periods of time,is extremely important.

3.1 Basic Definitions The following review of the fundamental conceptsassociated with the mechanical properties of materialsshould be familiar to anyone who normally deals withother material systems. Here we will discuss linearelasticity in more detail, completing our review of thesimplified property assumptions used in this manual -that plastics, for design purposes, may be treated aslinearly elastic, homogeneous, isotropic materials. Inthe next section, properties of plastics that represent aconcession to their anisotropic and nonlinear natureis introduced. Finally, the last section deals with thetime-dependent properties of plastics.

3.1.1 Stress Consider a three-dimensional body with a balancedsystem of forces acting on it, F1-F5 in Figure 3.01a,such that the body is at rest. A body is subjected toexternal forces develops internal forces to transfer anddistribute the external load. Imagine that the body inFig. 3.01a is cut at an arbitrary cross-section and onepart is removed. To keep the body at rest, a systemof forces must be acting on the cut surface to balancethe external forces. The same system of forces existswithin the uncut body and is called stress. Stress mustbe described with both a magnitude and a direction.Consider an arbitrary point P on the cut surface inFigure 3.01b, where the stress, S, is as indicated. Foranalysis, it is more convenient to resolve the stress, S,into two stress components. One acts perpendicularto the surface and is called a normal or direct stress,σ. The second stress acts parallel to the surface and iscalled a shear stress, τ.

F1

F2

F3

F5F4

CUT

(a.)

FORCE VECTORS

F4F5

(b.)

CUT

F1P S

τ

σ

Fig 3.01 · Internal forces, stresses in a body


20

3.1.2 Normal Stress A basic understanding of load, deflection, and stressstarts with a simple tension test, shown in Figure3.02. Direct Stress is the ratio of applied load to theoriginal cross-sectional area, expressed in pounds persquare inch (lb/in.2), or psi. In the InternationalSystem of units (SI), stress is expressed as Newtonsper square meter, or Pascals (Pa).

Stress = load/area

or

If the load is applied as shown, the member is said tobe in tension. If the load is reversed, the member is in compression.

3.1.3 Normal Strain If a bar is subjected to a direct load, and thus a stress,the bar changes in length. If the bar has an originallength L and changes in length by an amount ∆L, thestrain produced is defined as:

strain = change in lengthoriginal length

or

ε = ∆LL

Strain is a measure of the deformation of the materialand is dimensionless; i.e., it has no units. It is simply aratio of two quantities with the same units.

In general, the extensions of materials under load arevery small. With most metals, it is convenient tomeasure and express strain in the form of microinchesper inch, or 10-6 inches per inch (10-6 cm/cm). Thesymbol µε, called microstrain, expresses this. Withplastics, which generally undergo greater deformationthan metals under the same loading, strain is normallyexpressed as 10-5 inches per inch. Another commonway to express strain is as percent strain. Theequivalence of the three is as follows:

1000µε = 1000 X 10-6 = 0.001 = 0.1% strain10000µε = 10000 x 10-6 = 0.01 = 1.0% strain

Figure 3.03 illustrates a typical tensile testingarrangement with a common test specimen. Theresults obtained from this testing may be plotted inthe form shown in Figure 3.04. This is a stress-straincurve, which characterizes the mechanical behavior ofa material in tension.

σ = F

A

(APPLIEDLOAD)

(ORIGINALLENGTH)

(INCREASE INLENGTH DUETO APPLIEDLOAD)

L

F

F

A(AREA)

∆L

Fig 3.02 · Simple tension load

FORCEMEASUREMENT

FIXED HEAD

GAUGE MARKS

MOVABLEHEAD

GRIPSFORHOLDINGSPECIMENFIRMLY

STRAINGAUGE

CONSTANTRATE OFMOTION

TEST SPECIMEN

THICKNESS 1/8"

1/2"

8 1/2"

3/4"

Fig 3.03 · Typical test setup and specimen


21

3X BREAK

STRESS

STRAIN

X

σ

ε ∆LL

FA

=

=

Fig 3.04 · Plot of results of tensile test (stress-strain curve)

3.1.4 Modulus of Elasticity Most materials, including metals and plastics, have a deformation that is proportional to theimposed loads over at least a range of loads. Since stress is proportional to load and strain isproportional to deformation, this also implies thatstress is proportional to strain. Hooke’s law is thestatement of that proportionality.

stress = constantstrain

The constant E is called the modulus of elasticity,Young’s modulus, or in the plastics industry, tensilemodulus. In terms of the bar in Figure 3.02, thetensile modulus is given by

E =F/A

=FL

∆L/L A∆L

The modulus is, therefore, the slope of the initialportion of the stress-strain curve. It must be notedthat an elastic material does not necessarily obeyHooke’s law. A material may return to its originalshape without the stress being proportional to thestrain. However, if a material obeys Hooke’s law, it iselastic. In many plastic materials, the straight regionof the stress-strain curve is so difficult to locate that astraight line tangent to the initial portion of the curvemust be constructed to obtain a “modulus.” Amodulus obtained in this manner is called the initialmodulus. For some plastic materials, the initialmodulus can be misleading due to the material’snonlinear elasticity. For this reason, some suppliers

provide the 1% secant modulus (see below) as abetter representation of the material’s behavior. Thedesigner is cautioned that the product data sheets donot always clarify whether the supplier is providingYoung’s modulus, an initial modulus, or a secantmodulus. Thus, the designer is reminded of thewarning at the beginning of this chapter on placingtoo much emphasis on the published data.

For metals, Young’s modulus is normally expressed interms of 106 psi, or Mpsi (MPa or GPa). For plastics,the tensile modulus is often expressed as 103 psi orkpsi (MPa).

A number of stress-strain curves are shown in Figure3.05. The explanation of points A through F on thecurves is provided below.

3.1.4.1 Proportional Limit, A With most materials, some point exists on the stress-strain curve where the slope begins to changeand the linearity ends. The proportional limit is thegreatest stress at which a material is capable ofsustaining the applied load without deviating from the proportionality of stress to strain. This limit isexpressed in psi (Pa) and is shown as Point A inFigure 3.05. Note that some materials maintain thisproportionality for large measures of stress and strainwhile others show little or no proportionality, aspreviously discussed.

3.1.4.2 Yield Point, B Yield point is the first point on the engineering stress-strain curve where an increase in strain occurswithout an increase in stress. This is shown as Point B in Figure 3.05. The slope of the curve is zero at this point. Note that some materials may nothave a yield point.

STRE

SS, σ

STRAIN, ε

X BREAK

O H

BC

C

C

B,C

AA

A D

D

D

D

F

F E

Fig 3.05 · Typical stress-strain curves


22

3.1.4.3 Ultimate Strength, C The ultimate strength is the maximum stress amaterial withstands when subjected to an appliedload. This is expressed in psi (Pa) and is denoted byPoint C in Figure 3.05.

3.1.4.4 Elastic Limit, D Many materials may be loaded beyond theirproportional limit and still return to zero strain whenthe load is removed. Other materials, particularlysome plastics, have no proportional limit in that noregion exists where the stress is proportional to strain(the material obeys Hooke’s law). However, thesematerials may also sustain significant loads and stillreturn to zero strain when the load is removed. Ineither case, the point on the stress-strain curve, D inFigure 3.05, represents the point beyond which thematerial is permanently deformed if the load isremoved. This point is called the elastic limit.

3.1.4.5 Secant Modulus, E The secant modulus is the ratio of stress tocorresponding strain at any point on the stress-straincurve. For instance, in Figure 3.05, the secantmodulus at Point E is the slope of the line OE.

3.1.4.6 Yield Strength, F Some materials do not exhibit a yield point. For suchmaterials, it is desirable to establish a yield strengthby picking a stress level beyond the elastic limit.Although developed for materials that do not exhibita yield point, this value is often used for plastics thathave a very high strain at the yield point to provide amore realistic yield strength. This is shown as Point Fon the curves on Figure 3.05. The yield strength isgenerally established by constructing a line parallel toOA at a specified offset strain, Point H. The stresswhere the line intersects the stress-strain curve atpoint F is the yield strength at H offset. For instance,if Point H were at 2% strain, then Point F would betermed the “yield strength at a 2% strain offset.”

3.1.5 Poisson’s Ratio Under the action of a tensile load, the bar shown inFigure 3.06 increases in length by an amount ∆L,giving a longitudinal strain in the bar of

ε = ∆LL

The bar also exhibits a reduction in dimensionslaterally; i.e., its breadth and depth both decrease. The associated lateral strains are opposite in sign

(contracting vs. stretching) to the longitudinal strain,and are given by

εlateral

= −∆b

= −∆d

b d

Provided the material deformation is within theelastic range, the ratio of the lateral to longitudinalstrains is always constant. This ratio is calledPoisson’s ratio, and is designated by the greek letter ν.

ν = lateral strain

= ∆d/d

longitudinal strain ∆L/L

For most engineering materials, the values of ν liebetween 0.20 and 0.40; however, a default value of0.35 is usually sufficient. Typically, ν is between zero(no lateral contraction) and 0.5 (constant volumedeformation). Table 3.01 shows typical values of νfor various structural materials. Poisson’s ratio is a necessary constant for the stress and deflectionanalysis of plastic structures such as plates, shells, and rotating discs.

Material Range of Poisson’s Ratio

Aluminum 0.33Carbon Steel 0.29Rubber 0.50Rigid Thermoplastics

Neat 0.20 – 0.40Filled or Reinforced 0.10 – 0.40

Structural Foam 0.30 – 0.40Rigid Thermosets

Neat 0.20 – 0.40Filled or Reinforced 0.20 – 0.40

Table 3.01 · Typical values of Poisson’s ratio

∆b = b-b'

∆d = d-d'

∆L

d'd

F

L

F

b'

b

Fig 3.06 · Loaded tensile bar showingdimensional change in length and width


23

3

3.1.6 Shear Stress A block of material shown in Figure 3.07a issubjected to a set of equal and opposite shearingforces, Q. If the material is imagined as an infinitenumber of infinitesimally thin layers, as shown inFigure 3.07b, a tendency exists for one layer of thematerial to slide over another to produce a shear formof deformation, or failure if the force is great enough.

The shear stress, τ, is defined as

τ = shear load

= Q

area resisting shear A

The shear stress is always tangential to the area onwhich it acts. The shearing strain is the angle ofdeformation, γ, and is measured in radians.

SHEAR STRAIN

SHEARING LOADAREA

γ (RADIANS)

Q

Q

Q

Q

a)

b)

Fig 3.07 · Shear stress

3.1.7 Shear Modulus For materials that behave according to Hooke’s law,the shear strain is proportional to the shear stressproducing it. Thus,

shear stress =

τ= constant = Gshear strain γ

The constant G is called the shear modulus ormodulus of rigidity, and is directly comparable to the modulus of elasticity used in direct stressapplications.

3.2 Relating Material Constants It was previously noted that only two materialconstants are required to characterize a material if oneassumes the material is linearly elastic, homogeneous,and isotropic. However, three material constants havebeen introduced: tensile modulus, E, Poisson’s ratio, ν;and shear modulus, G, Therefore, an equation relatingthese three constants is needed. On the basis ofelasticity principles that are beyond the scope of thismanual, the following equation may be developed:

E = 2 (1 + ν)G

This holds true for most metals and is generally appliedto injection-moldable thermoplastics. However, thedesigner is reminded of the inherently nonlinear,anisotropic nature of most plastics, particularly fiber-reinforced and liquid crystalline materials.

3.2.1 Direct Shear Figure 3.08 shows a typical shear strength setup usedfor plastics. Data obtained by this method is oftenreported in marketing data sheets as the shearstrength of the material. In strength of materialsliterature, this type of test is called direct shear. Thereader is cautioned to use the “shear strength”reported by this method only in similar direct shearsituations. This is not a pure shear test. The testcannot be used to develop shear stress-strain curvesor to determine the shear modulus because aconsiderable portion of the load is transferred bybending and/or compressing rather than by pureshear. In addition, the results can depend on thesusceptibility of the material to the sharpness of theload faces. When plastics are analyzed in a pure shearsituation or when the maximum shear stress iscalculated in a complex stress environment, the use ofa shear strength equal to half the tensile strength, orthe above reported shear strength, is recommended,whichever is less.

MALE PUNCH

TEST SAMPLE

LOAD

Fig 3.08 · Direct shear stress test used in plastics industry


24

3.2.2 True Stress Though infrequently used, the terms “true stress andstrain” are worth noting. In Figure 3.09, the normalstress is calculated on the basis of an increasing loadF, acting over a constant area, A. This form of thedirect stress, discussed previously, is often called“engineering stress.” With most materials, however, a“necking down” occurs in a critical area where failurewill eventually result. If the smaller cross-section, AI,were used in place of A, then the calculated stresswould be referred to as “true stress.” In addition, thedirect strain discussed previously, i.e., total change inlength over original length, is often called“engineering strain.” The true strain would be theinstantaneous deformation over the instantaneouslength. Therefore, the shape of a true stress-straincurve would not be the same as a simple stress-straincurve. Almost universally, however, modulus valuesand stress-strain curves are based on engineeringstress and strain.

3.3 Other Measures of Strength and Modulus For many engineering materials that are treated aslinearly elastic, homogeneous, and isotropic, thetensile and compression properties are consideredidentical. This eliminates the need to measureproperties in compression. Further, if tension andcompression properties are identical, there is no needto measure the properties in bending (under standardbeam bending theory). However, in concession to thenonlinear, anisotropic nature of most plastics, theseproperties, particularly flexural properties, are oftenreported on product data sheets.

STRAIN

TENSILECOMPRESSIVE

STRE

SS

CO

MPR

ESSI

VE

TEN

SILE

ET

EC

σε

=

σε

=

ε

ε

σ

σ

Fig 3.10 · Tensile and compressive moduli

ORIGINALAREA

A

F

F

AI

REDUCEDAREA DUE

TO NECKINGDOWN

F

F

Fig 3.09 · True stress

3.3.1 Compression Strength and Modulus Because of the relative simplicity of testing in tension,the elastic modulus of a material is usually measuredand reported as a tension value. A material can also be loaded in compression. However for design, thestress-strain curve for compression loading is often required.

With most elastic materials at low stress levels, thetensile and compressive stress-strain curves are nearlyequivalent, as depicted by the curve in Figure 3.10.However, at higher stress levels, the compressivestrain is less than the tensile strain. Unlike tensileloading, which usually results in a clear failure,stressing in compression produces a slow, indefiniteyielding, which seldom leads to a failure. Therefore,the compressive strength is customarily expressed asthe stress in psi (Pa) required to deform a standardplastic specimen to a certain strain.

Compression modulus is not always reported, sincedefining a stress at a strain is equivalent to reporting asecant modulus. However, if a compression modulusis reported, it is generally an initial modulus.


25

3

3.3.2 Bending Strength and Modulus If a piece of plastic or metal, most conveniently arectangular cross-section, is bent between one’sfingers, it is readily apparent that one surface of thematerial stretches in tension while the oppositesurface compresses (Figure 3.11). It follows that thereis a line or region of zero stress between the twosurfaces, called the neutral axis. In simple beambending theory, the following assumptions are made:

1 . The beam is initially straight, unstressed, and symmetric.

2. The material of the beam is linearly elastic, homogeneous, and isotropic.

3. The proportional limit is not exceeded.4. Young’s modulus for the material is the same

in tension and compression.5. All deflections are small, so that planar cross-

sections remain planar before and after bending.

Bending properties can be measured as shown inFigure 3.12. Using classical beam formulas andsection properties (see Chapter 7), the followingrelationships can be derived:

Bending Stress σ = 3FL2bh2

Bending or Flexural Modulus E = FL3

4bh3Y

where Y is the deflection at the load point.

With these relationships, the flexural strength andflexural modulus (of elasticity) can be determined in a testing laboratory. Again, the reported flexuralmodulus is usually the initial modulus from the loaddeflection curve. Since most plastic parts must beanalyzed in bending, flexural values should lead tomore accurate results than if the corresponding tensilevalues are used.

COMPRESSIVESTRESS

TENSILESTRESS

AXISNEUTRAL

M

Fig 3.11 · Beam in bending

L

bh

APPLIED LOAD

F

Fig 3.12 · A simple bending fixture

3.4 Rate Dependence of Mechanical Properties The tensile and flexural data reported in product datasheets are measured at a particular displacement rate.Unfortunately, this rate is rarely consistent with theend-use loading environment. The same plasticmaterial, under differing rate and/or environmentalconditions, can produce different stress-strain curves.The designer must be aware of the loading rate in aparticular application and request the appropriatedata. Often the data are not available. Therefore, theimportance of end-use testing must be kept in mind.


26

3.5 Time-Related Mechanical Properties In the previous discussion, the given mechanicalproperties involved a gradually applied, short-termload. In this section the effects of time-relatedloading, both long term and very short term, areconsidered. With high-performance thermoplasticmaterials, creep, impact, fatigue, and related issues areimportant considerations. Unfortunately, laboratorytest data are not always directly applicable inestimating the structural response of actual parts. The laboratory method of applying the time-dependent load is rarely consistent with the end-useenvironment. Furthermore, in the actual end use,other factors are usually involved which are notcovered by the laboratory test conditions.

This section reviews time-related loading phenomenawith regard to meaning, testing methods, and the useof data in the design and analysis of plastic parts.

Log time

Stra

in

IncreasingStress

Log time

Stre

ss

Strain

Stre

ss

CREEP CURVES, STRAINVS.

LOG TIME

ISOMETRIC STRESSVS LOG TIME

ISOCHRONOUS STRESSVS STRAIN

Increasing Time

IncreasingStrain

Fig 3.13 · Typical presentation of creep data

3.5.1 Creep When a part or structure is subjected to a given load,there is a corresponding predictable deformation. Ifthe deformation continues to increase without anyincrease in load or stress, the material is said to beexperiencing cold flow or creep. Thus, creep can bedefined as increasing strain over time in the presenceof a constant stress. The rate of creep for any givenmaterial depends on applied stress, temperature, and time.

In parts that are to be subjected to loads for extendedperiods of time and where the maximum deflection iscritical, the creep behavior of a material is veryimportant. Test samples may be loaded in tension,compression, or flexure in a constant-temperatureenvironment. With the load constant, the deflectionor strain is noted at regular intervals of hours, days,weeks, and/or months. Generally, results are obtainedat four or more stress levels. Stress-strain-time dataare usually presented as creep curves of strain vs. logtime. Sets of such creep curves, illustrated in Figure3.13, can be produced by smoothing and interpolatingon a computer. These same data may be presented inother ways to facilitate the selection of informationfor particular requirements. Sections may be takenthrough the creep curves at constant times, yieldingisochronous stress-strain curves; or at constant strain, giving isometric stress vs. log time curves.These derivations are shown schematically in Figure 3.13.

In general, crystalline materials have lower creep ratesthan amorphous resins. Glass reinforcement generallyimproves the creep resistance of a plastic material.


27

3

3.5.2 Apparent or Creep Modulus If the deflection of a part subject to continuousloading is calculated by using the modulus ofelasticity, E, the result is likely to be inaccurate sincethe effects of creep have not been considered.However, if the stress level and temperature areknown and the creep curves are available at thetemperature, an apparent or creep modulus, Eapp, canbe calculated by using the creep curves as follows:

Eapp

= σεc

where

σ is the calculated stress levelεc is the strain from the creep curve at the expected time and temperature.

This value, Eapp, can then be used instead of E topredict the maximum deflection, using the methodsdiscussed in Chapter 7.

Curves of creep modulus or log creep modulus vs. log time at either constant stress or strain are oftenderived from the creep data and plotted as in Figure3.14. Data may also be provided as tables at constantstress and temperature at various times. Manymaterial suppliers provide creep data in the form ofcreep modulus rather than by the various curves of Figure 3.13.

3.5.3 Creep Rupture Failure may occur in creep when a part exceedsallowable deformation or ruptures. In creep rupturefailures, the fracture may be brittle or ductile withsome degree of necking. Creep rupture data areobtained in the same manner as the creep data exceptthat higher stresses are used and time is measured tofailure. The strains are sometimes recorded, but arenot necessary for creep rupture. The results aregenerally plotted as log stress vs. log time to failure(Figure 3.15).

LOG TIME

INCREASINGSTRESS OR STRAIN

CRE

EP M

OD

ULU

S

Fig 3.14 · Creep modulus

LOG TIME

LOG

STR

ESS

CREEP RUPTURE

Fig 3.15 · Creep rupture data – a curve showingone cycle log time projection


28

3.5.4 Stress RelaxationIn many cases when plastic parts are assembled, theyare placed into a permanently deflected condition.Examples are press-fits, bolted assemblies, and someplastic springs (see Figure 3.16). In time, with thestrain constant, the stress level decreases due to thesame internal molecular movement that produces thecreep phenomenon. This gradual decay in stress atconstant strain is known as stress relaxation. Thisbecomes important in applications such as boltpreload and springs, where there is a concern for theretention of load. The relaxation can be measured byapplying a fixed strain to a sample and measuring theload with time. The resulting data can then bepresented as a series of curves very similar to theisometric stress curves in Figure 3.13. In addition; arelaxation modulus similar to the creep modulus, canbe derived from the relaxation data. Generally,relaxation data are not as available as creep data.However, the decrease in load due to stress relaxationcan be approximated by using the creep modulus,Eapp, calculated from the creep curves, as in Figure 3.14.

Plastic parts often fail due to excessive fixed strainsimposed on them for extended periods of time. An example would be the splitting of a plastic tube press fit over a steel shaft. Unfortunately, there is no“relaxation rupture” corollary to creep rupture. Forpurposes of initial design concept development, astrain limit of 20% of the strain at the yield point oryield strength is suggested for high-elongationplastics. Likewise, 20% of the elongation at break issuggested for low-elongation, brittle materials that do not have a yield point. However, this is only aguideline for initial design. Prototype parts should bethoroughly tested at end-use conditions to confirmthe design. In addition, data that suggest a higher orlower limit may be available on the specific materialof interest.

3.5.5 Extrapolating Creep and Relaxation Data It is sometimes necessary to predict material behaviorfor times greater than those for which creep andrelaxation data are available. This must be undertakenwith caution. However, there is generally a lesspronounced curvature when creep and relaxation dataare plotted log property vs. log time. This facilitatesextrapolation and is commonly practiced, particularlywith creep modulus and creep rupture data. Even so,extrapolation should not exceed one logarithmic unitof time, as illustrated in Figure 3.15. In addition, the

strain limit of 20% of the yield or ultimate strengthmentioned in the above paragraph is suggested .

3.5.6 Impact Loading Whenever a part is loaded rapidly, the part issubjected to impact loading. Any moving body haskinetic energy. When the motion is stopped due to acollision, energy must be dissipated. The ability of a plastic part to absorb energy is determined by itsshape, size, thickness, and the type of material.Unfortunately, the impact testing methods presentlyavailable do not provide the designer withinformation that can be used analytically. However,the tests are useful for comparing the relative notchsensitivity of materials or the relative impactresistance. This can be very useful in choosing a seriesof materials to be evaluated in an application or inrank ordering materials within a series.

3.5.7 Izod Impact Probably one of the most widely used tests for impactstrength is the notched Izod impact test. In this test apendulum arm swung from certain height is made toimpact a notched, cantilevered beam (see Figure 3.17).After fracturing the test specimen, the pendulumcontinues to travel in the same direction, but with less energy due to the impact with the specimen. This loss of energy, measured in foot-pounds per inch(ft-lb/in., or J/m) of beam thickness, is known as theIzod impact strength.

PRESS FITLIGHT INTERFERENCE

BOLTED ASSEMBLYLIGHT ASSEMBLED STRESS

Fig 3.16 · Examples of constant strain loads


29

3.5.9 Tensile Impact This test uses a swinging pendulum similar to thatused in the lzod impact test, except the samplespecimen is a tensile bar that is mounted (as shown inFigure 3.18) to measure energy required to fracture itdue to tensile impact loading.

IZOD

POINTOF IMPACT

IMPACT

IMPACT

IZOD

CHARPY

POINTOF IMPACT

CHARPY

CHARPY

Fig 3.17 · Izod and Charpy impact testsThis test can also be run with an unnotched specimenor with the notch reversed, in which case it isreported as unnotched or a reversed notch lzodimpact strength, respectively.

3.5.8 Charpy Impact Charpy impact is less common in the United States,but is widely used in Europe. The test is essentiallyidentical to the lzod test except that the test specimenis a simply supported beam that is impacted midwaybetween the supports. Like the lzod test, thespecimen can be notched or unnotched, and theresults are reported in foot-pounds/ inch (ft-lb/in., or J/m) of beam thickness.

ANVIL

TEST BAR

TENSILEIMPACT

Fig 3.18 · Tensile impact

3


30

3.5.10 Falling Dart Impact TestIn this test, a weight is dropped onto a flat disk of thematerial being tested. The leading edge of the dartwhere it impacts the specimen has a specific diameter.Figure 3.19 is one example of a falling dart apparatus.This test is valuable for ranking materials since ittends to better represent the impact on actual parts incertain applications.

3.5.11 Fatigue Endurance Generally, a material is subjected to fatigue when it isstressed repeatedly or in some defined cyclic manner.Examples are a snap-action plastic latch that isconstantly opened and closed, a reciprocatingmechanical part on a machine, a gear tooth, a bearing,any structural component subjected to vibration, andany part that is to be subjected to repeated impacts.Such cyclic loading can cause mechanicaldeterioration and progressive fracture of the material,leading to ultimate failure.

Typical fatigue tests are conducted on a machinewhich subjects a cantilever beam to reverse flexuralloading cycles at different maximum stress levels. Thenumber of cycles to failure is recorded for each stresslevel. The data are generally presented in a plot of logstress vs. log cycles, called an S-N curve. The cyclerate loading profile and environmental temperatureshould be reported with the curve. Figure 3.20illustrates a typical S-N curve.

With thermoplastic materials there is the addedcomplication of a heat buildup that can sometimescontribute to the actual failure. Figure 3.20 alsoillustrates the possible thermal failure that can occurin a fatigue test. Thermal failure is attributed to heatbuildup resulting from the frequency of the cyclicstress. Significant differences in the S-N curve canalso be produced by testing at different frequencies,different mean stresses, different waveforms, anddifferent test methods, i.e., tension rather thanbending.

Although fatigue test data give some indication of therelative ability of plastic materials to survive fatigue,the designer must be aware of the above variables.The tests are run on specially prepared samples in atest environment, which never resembles the actualloading and environment of the actual parts.Therefore, it is essential that tests be run on actualinjection-molded parts under end-use operatingconditions to determine the true fatigue endurance ofany part on which significant cyclic loading occurs.

W

DART1" DIA

Fig 3.19 · A typical dart impact apparatus

TEMPERATURE

TEM

PERA

TURE

LOG CYCLES

LOG

STR

ESS

S–N CURVE

Fig 3.20 · Typical S-N curve is shown alongwith thermal effects which sometimes occur

when plastics are fatigue tested


31

4

4. Thermal PropertiesTo select materials that will maintain acceptablemechanical properties and dimensional stability inend-use applications, the design engineer must beaware of the normal and extreme operatingenvironment to which the final product is to besubjected. Following are some of the basic thermalproperties that characterize thermoplastic materials.

4.1 Melting Point Thermoplastic materials become more fluid whentemperatures increase. While crystalline materialshave sharp, clearly defined melting points, amorphous and liquid crystalline thermoplasticssoften and become more fluid over a wider range of temperatures. There are many methods fordetermining melting points of plastics. However, thisproperty is of greater significance to molding andassembly operations than to product design.

4.2 Glass Transition TemperatureAt the glass transition temperature, Tg, a materialundergoes a significant change in properties.Generally, below the glass transition temperature, the material has a stiff, glassy, brittle response toloads, while above the Tg, it has a more ductile,rubbery response.

4.3 Vicat Softening Point This is the temperature at which a small, circular,heated, lightly loaded probe penetrates a specificdistance into a thermoplastic test specimen. Figure4.01 shows a typical test apparatus. This test is usefulfor crystalline materials, but is of somewhat limitedvalue for amorphous thermoplastics, which tend tocreep during the test. This test indicates the ability ofa thermoplastic to withstand a short-term contactwith a heated surface.

DIAL GAUGE

WEIGHT

TEST PIECE

INDENTING TIP

LIQUID LEVEL

LOADTRANSFER

ROD

Fig 4.01 · Vicat softening point apparatus


32

4.4 Deflection Temperature under LoadThe deflection temperature under load is thetemperature at which a 1/2 in. deep test bar, loaded toa specified bending stress, deflects by 0.010 in. Figure4.02 shows a schematic of a typical apparatus. Thistest is generally run at 66 psi and/or 264 psi. Sometimescalled the “heat distortion temperature” (HDT), thisvalue is useful to the design engineer as a relativemeasure of the ability of various materials to performat elevated temperatures while supporting loads. Sincea stress and deflection for a certain depth test bar isspecified, this test may be viewed as establishing thetemperature at which the flexural modulus decreasesto particular values, i.e., 35,200 psi at 66 psi stress and140,800 psi at 264 psi.

4.5 Coefficient of Linear Thermal Expansion Like metals, thermoplastic materials expand whenheated and contract when cooled. Generally, for agiven change in temperature, plastic materials changedimensionally much more than metals. Thecoefficient of linear thermal expansion, CLTE, is theratio of the change of a linear dimension to theoriginal dimension of the material for a unit change oftemperature. It is generally given in units of in./in./°For cm/cm/°C. Table 4.01 presents typical values formany common materials. This is a very importantconsideration if dissimilar materials are to beassembled. It should be noted that manythermoplastics, especially those with glassreinforcement or liquid crystalline resins, may havedifferent coefficients for the mold flow, cross-flow,and thickness directions. Since the CLTE of plasticscan vary with temperature over a broad temperaturerange, the extent of the variability must be known.This is especially important if the temperature rangeincludes a transition, such as Tg.

4.6 Thermal Conductivity Thermal conductivity is the rate at which a materialconducts heat energy along its length or through itsthickness. This is an important factor, since plasticsare often used as an effective heat insulator in heat-generating applications and in structures in whichheat dissipation is important.

Material in./in./F10-5 cm/cm/°C10-5

Liquid Crystal (GR) 0.3 0.6Glass 0.4 0.7Steel 0.6 1.1Concrete 0.8 1.4Copper 0.9 1.6Bronze 1.0 1.8Brass 1.0 1.8Aluminum 1.2 2.2Polycarbonate (GR) 1.2 2.2Nylon (GR) 1.3 2.3TP Polyester (GR) 1.4 2.5Magnesium 1.4 2.5Zinc 1.7 3.1ABS (GR) 1.7 3.1Polypropylene (GR) 1.8 3.2Epoxy (GR) 2.0 3.6Polyphenylene Sulfide 2.0 3.6Acetal (GR) 2.2 4.0Epoxy 3.0 5.4Polycarbonate 3.6 6.5Acrylic 3.8 6.8ABS 4.0 7.2Nylon 4.5 8.1Acetal 4.8 8.5Polypropylene 4.8 8.6TP Polyester 6.9 12.4Polyethylene 7.2 13.0

Table 4.01 · Typical values for coefficient of linearthermal expansion for thermoplastics and

other common materials

NOTES: Plastic materials are unfilled unless specified otherwise.

(GR) indicates typical glass-fiber-reinforced grade;TP indicates thermoplastic.

These are only typical, average values and do not account forvarious grade differences, molding conditions, wall thickness,flow direction, etc.

Request information for the specific resin of interest.

LOAD

GAUGE

THERMOMETER

Fig 4.02 · Test apparatus for deflection temperatureunder load


33

4

4.7 Aging at Elevated Temperatures Typically, aging at elevated temperatures involvesstoring a large number of samples at a particulartemperature for an extended time. Samples are thenremoved from storage at various time intervals andtested at standard conditions for whatever physical,mechanical, thermal, or electrical property is ofinterest. The data are generally presented as a plot of property vs. aging time for the various agingtemperatures. Thus, the results from aging at elevatedtemperatures may be used as a measure of the thermalstability of the resins, and must be considered indesign situations.

4.8 Relative Thermal Index (UL) Underwriters Laboratories tests and recognizesplastic materials used in electrical applications forcertain continuous operating temperatures. Theseratings are listed separately for electrical properties,mechanical properties including impact, andmechanical properties without impact. The RelativeThermal Index is important if the final product mustreceive UL recognition.

4.9 Flammability For many plastic applications, the consequences ofexposure to an actual flame must be considered. This is important not only in electrical applications,but also in any application where the plasticconstitutes a significant percentage of the exposedarea of a defined enclosed space. An example of thisis the plastic panels used in the interior of an aircraftcabin. Typical tests measure combustibility, smokegeneration, and ignition temperatures. Some of thefollowing tests are common:

1. UL 94 Flammability Class (V-0, V-1, V-2,5V, HB)In this test, specimens are subjected to a specifiedflame exposure, and the relative ability to maintaincombustion after the flame is removed becomes thebasis for classification. In general, the more favorableratings are given to materials that extinguishthemselves rapidly and do not drip flaming particles.Each rating is based on a specific material thickness.

2. Oxygen indexThis test measures the percentage of oxygen necessary to sustain combustion of the plasticmaterial. Obviously, the higher the value (moreoxygen needed), the lower the combustibility. Sinceair contains about 21 % oxygen, any material with arating below 21 will probably support combustion in a normal, open air environment.

4.10 Effect of Temperature on Mechanical Properties An important behavior characteristic of materials,which the designer must understand when working inplastics, is the inverse relationship between strain rateand temperature. Very high strain rates and very lowtemperatures produce similar material responses.Conversely, very low strain rate effects, i.e., creepeffects, can be seen much more quickly by testing atelevated temperatures. The designer might make useof the concept by testing at temperatures around andabove the highest values expected in end use. The datacan be used to estimate long-term performance of apart at times much longer than can be reasonablyused for testing. A number of theoretical andempirical models are used to describe this behavior ofmaterials. They have been used successfully, providedthey are not applied near or across the transitiontemperatures or near the melting point. However,these models are well beyond the scope of thismanual.

4.10.1 Strength, Modulus, and Elongation The strength, modulus, and elongation behavior aresimilar for tensile, compressive, flexural, and shearproperties. In general, strength and modulus decreaseand elongation increase with increasing temperature.Figure 4.03 illustrates a series of stress-strain curvesthat might be produced by testing a plastic material at one strain rate and several temperatures. The sameset of curves might also be generated by testing atconstant temperature and various strain rates, as indicated.

STRAIN

INCREASE TEMPERATURECONSTANT STRAIN RATE

INCREASE STRAIN RATECONSTANT TEMPERATURE

STRE

SS

Fig 4.03 · Effect of temperature or strain rateon stress-strain curves


34

Figure 4.04 illustrates modulus behavior astemperature increases. In general, a gradual drop inmodulus occurs as the temperature increases to theglass transition temperature. Above the glasstransition temperature, amorphous materialsexperience a rapid loss of modulus. However,crystalline materials maintain a significant, usablemodulus at temperatures approaching the crystallinemelting point. Glass fiber reinforcement cansignificantly improve the modulus of crystallinematerials above the glass transition temperature.Amorphous materials, even when reinforced withglass, still display a rapid drop in modulus above theglass transition temperature. Strength vs. temperaturecurves are similar to modulus curves. In general, theelongation of materials increases with increasingtemperature, as indicated in Figure 4.05.

AMORPHOUS

UNFILLED REINFORCED

CRYSTALLINE

UNFILLED REINFORCED

HDT

MO

DU

LUS

O=MELTX = Tg

TEMPERATURE

Fig 4.04 · Modulus behavior of crystalline andamorphous resin showing Tg and melt temperatures

and the effect of reinforcement on HDT

95431-1

LOG TIME (HOURS)

LOG

APP

ARE

NT

MO

DU

LUS

INCREASING TEMPERATURE

TIME-TEMPERATURE SHIFTING

87620

50101 YEARS

Fig 4.06 · Apparent modulus curves at differenttemperatures showing time-temperature shifting to

estimate expended time values at lower temperatures

43214-1

LOG TIME (HOURS)

LOG

CRE

EP R

UPT

URE

INCREASINGTEMPERATURE

Fig 4.07 · Creep rupture curves indicating danger of linear projection to longer times at

lower temperatures

TEMPERATURE

ELO

NG

ATI

ON

Fig 4.05 · Effect of temperature on tensile elongation


35

4

4.10.2 Creep Creep, isochronous stress, and isometric stress curves are produced from measurements at a fixedtemperature. Complete sets of the curves aresometimes available at temperatures other thanambient. It is common to find creep rupture orapparent modulus curves plotted vs. log time withtemperature as a parameter. Figure 4.06 illustratestime-temperature shifting of the apparent moduluscurves projecting to times beyond a normal testingrange. These curves suggest that it would bereasonable to estimate moduli at somewhat longertimes than the available data at the lowertemperatures. However, a set of creep rupture curvesat various temperatures (such as those in Figure 4.07)would suggest that projecting the lowest temperaturecurves to longer times as a straight line could producea dangerously high prediction of rupture strength.

4.10.3 Impact Testing impact properties at various temperaturesproduces a plot of impact property vs. temperature,which looks very much like the elongation vs.temperature curve in Figure 4.05. As temperaturesdrop significantly below ambient temperature, mostplastic materials lose much of their room-temperatureimpact strength. A few materials are on the lower,almost horizontal portion of the curve at roomtemperature and show only a gradual decrease inimpact property with decreases in temperature. Aninteresting exception is provided by the long-fiber-reinforced plastics, which have relatively high Izodimpact values at room temperature and essentially thesame values at -40°F (40°C).

4.10.4 Fatigue Occasionally, one finds S-N curves for plasticmaterials at various temperatures showing a decreasein the strength values with increasing temperature.However, as discussed in Chapter 3, laboratoryfatigue data rarely model the end-use fatigueenvironment adequately and should be used only as guides for the selection of materials for end-usetesting.


36

5. Electrical PropertiesPlastic materials have found widespread use in amultitude of electrical and electronic applicationsthroughout virtually all industries. The combinationof mechanical and electrical properties provides anideal choice for everything from tiny electroniccomponents to very large electrical equipmentenclosures. The most notable electrical property ofplastic materials is that they are good insulators, butas seen in this chapter, plastics have many otherimportant electrical properties that must beconsidered in the design of plastic parts.

5.1 Conductivity in Solids Solid materials vary in electrical conductivity,depending on the availability and mobility ofmovable charge carriers within the material. Metalsare excellent conductors largely due to the structureof a metal atom, which has a loosely held, outermostelectron. Metal atoms are in close proximity in aregular lattice. This allows the outer electron to easilybreak free and move within the lattice (Figure 5.01).The presence of these “free” electrons accounts forthe ability of a metal to conduct large currents evenwith low voltages.

In materials such as glass, porcelain, and plastics, theouter electrons are tightly bound to the atoms ormolecules and no free electrons are present. Thesematerials cannot readily conduct electrical current;thus, they are insulators. Other materials, such assilicon and germanium, have conductivities betweenthose of conductors and insulators, and are classifiedas semiconductors. These materials, in the pure stateand at low temperatures, are actually insulators.However, they can be made to conduct current byadding impurities or by raising the temperature.

5.2 Volume Resistivity A standard measure of conductivity is the electricalresistance of a material when a direct current potentialis applied to it. Figure 5.02 shows one of many testconfigurations used to measure the resistance across a volume of material. This test, known as volumeresistivity, is the resistance measured in ohms timesthe area of the smaller electrode divided by thethickness of the specimen. Thus, the volumeresistivity is generally given in ohm-cm. Testingconditions such as temperature, relative humidity, test voltage, and conditioning of the sample

can affect the results. Materials with values above 108

ohm-cm are considered to be insulators, while thosebetween 108 and 101 ohm-cm are considered to bepartial conductors. Table 5.01 shows the volumeresistivity values for various plastics.

FREE ELECTRONS

Fig 5.01 · Free electrons in a metal lattice

Material Volume Resistivity,ohm-cm

Acetal 1014 – 1016

Acrylic 1014 – 1018

ABS 1016

Nylon 1012 – 1016

Polycarbonate 1015 – 1017

TP Polyester 1014 – 1017

Polypropylene 1014 – 1017

Polysulfone 1015 – 1017

Modified PPO/PPE 1015 – 1017

Polyphenylene Sulfide 1016

Polyarylate 1016 – 1017

Liquid Crystal Polymer 1015

Table 5.01 · Typical values of volumeresistivity for thermoplastics


37

5

5.3 Surface Resistivity This test measures the ability of current to flow overthe surface of a material. Unlike the volume resistivitytest, the test electrodes are both placed on the sameside of the test specimen, as illustrated in Figure 5.02.While volume resistivity is a property of the material,surface resistivity is essentially a measure of thesusceptibility of the material to surfacecontamination, particularly moisture. Data from thistest are best used when materials are being evaluatedand selected for testing in applications in whichsurface leakage may be a problem. The data aresubject to large errors and should be used with largesafety factors in design analysis.

5.4 Dielectric Strength When an insulator is subjected to increasingly highvoltage, it eventually breaks down and allows acurrent to pass. The voltage reached just before itbreaks down divided by the sample thickness isknown as the dielectric strength of the material,measured in volts/mil. It is generally measured byputting electrodes on either side of a test specimenand increasing the voltage at a controlled rate (seeFigure 5.03). Factors that affect the test results aretemperature, sample thickness, conditioning ofsample, rate of increase in voltage, and duration oftest. Any contamination of, or internal voids in, thesample may cause premature failure in this test.

5.5 Dielectric Constant (Permittivity) When an electrical field is imposed across aninsulator, the molecules become polarized. If thepotential is reversed, the polarization of the moleculesalso reverses. The ease with which a material can bepolarized is measured by a material constant calledpermittivity. The ratio of a material’s permittivity tothe (negligible) permittivity in a vacuum is calledrelative dielectric constant or more commonly,dielectric constant. This is a dimensionless constantthat becomes an important factor when plastics areused as dielectric materials in high frequencyapplications. The value of this constant can vary withchanges in temperature, moisture level, frequency, andpart thickness. Table 5.02 gives the dielectricconstants for various insulating materials.

GAPELECTRODES

THICKNESS

VOLUME RESISTIVITY

GAP

SPECIMEN

SURFACE RESISTIVITY

Fig 5.02 · One type of text apparatus for volumeand surface resistivity

OILBATH

SPECIMEN

Fig 5.03 · A typical test configuration fordielectric strength


38

5.6 Dissipation Factor The dissipation factor can be measured along with thedielectric constant in the same apparatus. If thepolarizing reversals mentioned above occur rapidly, as with an alternating current, energy is dissipated asheat due to the rapid change in polarization. Thedissipation factor is a measure of that heat dissipation.It can also be viewed as the ratio of the energydissipated (lost) as heat compared to that transmitted,and is usually measured at 1 MHz (106 cycles/sec); see Table 5.02. A low dissipation factor becomesimportant when plastics are used as insulators inhigh-frequency applications such as radar andmicrowave equipment.

5.7 Arc Resistance If an electrical arc is imposed on the surface of aninsulating material, the material can develop aconductive path (Figure 5.04). The arc resistance testmeasures the time in seconds for this to occur. Thistest was developed for thermosets, since a conductivepath can be formed from the decomposition productsproduced by this kind of localized heating. While arcresistance times are often available for thermoplastics,the mechanisms can be very different and should beunderstood. The specimen for this test is dry and freeof contaminants. A material with a high value in thistest would be advantageous in electrical applicationswhere the possibility of arcing exists. Examples areswitches, circuit breakers, automotive ignitioncomponents, and high-voltage apparatus.

Material Dielectric DissipationDescription Constant Factor

Acetal 3.7 - 3.9 0.001 - 0.007Acrylic 2.1 - 3.9 0.001 - 0.060ABS 2.9 - 3.4 0.006 - 0.021Nylon 6/6 3.1 - 8.3 0.006 - 0.190Polycarbonate 2.9 - 3.8 0.0006 - 0.026TP Polyester 3.0 - 4.5 0.0012 - 0.022Polypropylene 2.3 - 2.9 0.003 - 0.014Polysulfone 2.7 - 3.8 0.0008 - 0.009Modified PPE (PPO) 2.4 - 3.1 0.0002 - 0.005Polyphenylene Sulfide 2.9 - 4.5 0.001 - 0.002Polyarylate 2.6 - 3.1 0.001 - 0.022Liquid Crystal Polymer 3.7 - 10 0.010 - 0.060

Table 5.02 · Typical values for dielectricconstant and dissipation factor for various

thermoplastics at room temperature

ELECTRODES

SAMPLE

Fig 5.04 · A typical test configuration forarc resistance

5.8 Comparative Tracking Index (CTI) This is an Underwriters Laboratories test run similarto the arc resistance test except that an electrolyte(solution of ammonium chloride) is put on thesurface. The CTI for a material is the numerical valueof the voltage required to cause a conductive path toform between the electrodes. This test is useful sinceit measures the arc resistance on a contaminatedsurface, which is often the case with actual electricaland electronic equipment.


39

6

6. Environmental Considerations

When specifying a particular material for use in anapplication, the design engineer must carefullyconsider the end-use environment. While rust andcorrosion can plague metals, cracking, crazing,discoloration, loss of properties, melting, ordissolution can affect thermoplastics in the presenceof chemical substances, energy sources, or radiation.An important aspect of the environmental assessmentthat is sometimes overlooked is the processing,assembly, finishing, and cleaning operations to whichthe product may be subjected even before it reachesits ultimate use.

6.1 Factors Affecting Environmental Resistance

6.1.1 Stress Level Stress level is a very significant factor affecting theperformance of a part. In general, as the stress levelincreases, resistance to a particular environmentdecreases. In addition to the obvious applied loads,stresses can result from injection molding, forming,and assembly operations.

6.1.2 Temperature In general, as temperature increases, the detrimentaleffect (if any) of a substance on a particular resin isgreater. Temperature often determines whether aparticular plastic will survive a specific environment.

6.1.3 Exposure The compatibility of a plastic material with aparticular environment is frequently determined bythe actual amount of exposure that exists. Among thefactors to be considered are the following:

1. Actual time of exposure to adverse environment2. Concentration of adverse substance3. State of substance (solid, liquid, or gas)4. Radiation level and intensity5. Protective barriers (coatings or paints), integral

“black” color

Data on the response of thermoplastics to stress andtemperature are usually available for established,commercially available materials. Generally availableenvironmental data include resistance to variouschemical substances and weathering information,particularly ultraviolet (UV) stability. For purposes of this discussion water is treated as simply anotherchemical. However, for many polymers, resistance towater receives separate, special treatment independentfrom general chemical compatibility.

6.2 Chemical Compatibility Part of the wide acceptance of thermoplastics is dueto their relative compatibility with the ambientenvironment, particularly moisture, as compared tometals. This is in spite of the fact that many plasticsare hygroscopic, thereby absorbing water. This resultsin dimensional and property changes. However, themechanism of chemical attack in plastics iscomplicated, making compatibility assessmentsdifficult. The same chemical may react differentlywith plastics in the same family or even with differentcompounds of the same base resin. Conversely, agiven resin compound may behave very differently inchemicals that are considered similar. In addition, avariety of mechanisms exists by which a chemicalmight attack a plastic.

6.2.1 Reaction The chemical can attack the polymer chain directlyproducing a progressive lowering of the molecularweight of the polymer. Changes in the short-termmechanical properties are evidence of reaction.

6.2.2 Solvation Most thermoplastics are soluble in some chemical.However, high molecular weight plastics usuallydissolve very slowly. Therefore, the solvation processusually appears similar to a chemical reaction.Generally, weight and dimensional changes andswelling are present along with a loss of properties.


40

6.2.3 Plasticization If a chemical is miscible with a polymer, absorptionand plasticization may result. Evidence ofplasticization of the plastic includes reduction ofstrength, stiffness, and creep resistance and increasedimpact resistance. The plastic tends to swell, andwarpage may occur due to relaxation of molded-instresses.

6.2.4 Environmental Stress Cracking An unstressed plastic may appear to be unaffected byexposure to a chemical. However, the same chemicalmay cause catastrophic failure when the plastic isstressed. This mechanism is called environmentalstress cracking.

Chemical compatibility data are generally obtained ina manner similar to “aging at elevated temperatures,”discussed in Chapter 4. That is, standard test bars areplaced in the chemical and stored at the desiredtemperature for some time interval. The test bars arethen removed from the exposure environment,cleaned, and tested for whatever properties are ofinterest, typically tensile strength, flexural modulus,dimensional change, weight, and discoloration. Table6.01 shows the chemical resistance of commonly usedthermoplastic materials. These are only generalguidelines. The design engineer must consultmarketing data sheets, test results, and the materialsupplier for accurate information concerning aparticular grade of resin. Even when the informationindicates the material is highly compatible, end-usetesting must be performed.

While this method is easy, it can be misleading sincethe response of stressed samples to the chemicalenvironment can be very different; i.e., it cannotgenerally detect agents causing environmental stresscracking. Some test procedures expose a test sampleto a chemical in the presence of either a fixed stress or fixed strain distribution along the length of thesample. The samples may then be examined for thestress or strain location at which damage begins.However, these tests are more difficult and expensiveto run. Therefore, such data for a resin in a particularchemical are often unavailable.

Finally, the above tests may provide informationabout the chemical compatibility of a plastic, but areof little value for establishing the performanceproperties of a plastic for design purposes. The onlytest that provides this information is the creep rupturetest conducted at appropriate temperatures in thechemical environment. This testing is difficult andexpensive to conduct and is infrequently done.

The design engineer should be prepared to pursuecreep rupture testing in the actual end-useenvironment on test bars or, preferably, prototypeparts to determine the suitability of a plastic in thatenvironment. Thus, it is advantageous for the designerto select several different plastics or components ofthe same plastic for evaluation in the end-useenvironment.

6.3 Weathering Resistance Many applications for plastics require that thematerial withstand exposure to “sunlight,” i.e., UV,when used in naturally lighted areas or outdoors. Allplastic materials are affected by ultraviolet lightexposure and suffer some degree of degradation. Thedegradation is usually noticed by fading, chalking,and embrittlement of the plastic material. Often, UV-resistant variants of popular grades of plastic materialsare available from suppliers.

UV test data are usually obtained through actualoutdoor exposure or in special test cabinets. Outdoorexposure tests may be as simple as attaching testsamples to a surface at a suitable angle for the latitudewhere the test is being conducted or as complicated as having mirrors and sun tracking equipment toaccelerate the effective exposure. Test cabinets areavailable to accelerate testing. Generally, they employhigh-intensity xenon or carbon-arc lamps to generatehigh levels of UV exposure in a relatively short time.


41

6

This information is presented for instructional purposes and is not intended for design. The data were extracted from numerous sources, making consistentrating assignments difficult. Further, the response of any given material to specific chemicals in any one class can vary significantly. Indeed, during thepreparation of the table, the effect of various chemicals in the same category on one plastic ranged from essentially no effect to total dissolution.Therefore, an “A” rating for a particular plastic exposed to a particular class of chemicals should not be interpreted as applying to all chemicals in thatclass. The rating simply means that for the chemicals in that class found in the literature reviewed, the rating was generally an “A”. There may be otherchemicals in the same class for which the rating would be a “C.” Finally, the typical chemicals listed do not necessarily correspond to the ones on whichthe individual ratings are based.

A – Minimal Effect B – Some Effect C – Generally Not RecommendedRoom temperature except hot water, steam, and “*”Generally, extended exposure (more than a week) data was used *200°F

Ace

tal Copoly

mer

Ace

tal H

om

opoly

mer

Alu

min

um

Carb

on S

teel

316 S

tain

less

Ste

el

Nylo

n 6

/6Th

erm

opla

stic

Poly

este

r (P

BT)

Ther

mopla

stic

Poly

este

r (P

ET)

Poly

este

r El

ast

om

erLi

quid

Cry

stal Poly

mer

*

ABS

Poly

pro

pyle

ne

Poly

phen

yle

ne

Sulf

ide

Poly

ary

late

Poly

sulf

one*

Modif

ied P

oly

phen

yle

ne

Ether

Poly

carb

onate

Examples

Acids and Bases

Acids, Weak A B C A A A A A A A B A A A A A C Dilute Mineral AcidsAcids, Strong C C C B – C B A – C C – A A B C C Concentrated Mineral AcidsBases, Weak A C A B B A B A – C A A A A A B C Dilute Sodium HydroxideBases, Strong A C B – – B C A – C A – A A B B C Concentrated Sodium HydroxideAcids, Organic, Weak A B C A A A A A A A B A A A A C C Acetic Acid, VinegarAcids, Organic, Strong C C C B – C B A – C C A A A B C C Trichloroacetic Acid

Automotive

Automotive Fuel A A A A A A A A C C A C C A A A AAutomotive, Lubricants A A A A A A A A C C A A A A A B AAutomotive, Hydraulic A A A A A – A A C C C A A A – – –

Solvents

Aliphatic Hydrocarbons A A A A A A A A A A A B C A A A A Heptane, HexaneAliphatic Hydrocarbons, Halogenated A B B B B A A A C C C – – – B B B Ethylene Chloride, ChloroformAlcohols A A B A A A A A A A A A A A A A B Ethanol, CyclohexanolAldehydes A A A A B B A A – C B – A – A B A Acetaldehyde, FormaldehydeAmines A C – – – – C B – C C – A – A B B Aniline, TriethanolamineAromatic Hydrocarbons A B A A B B A A C C C C C C A A A Tolulene, Xylene, NaphthaAromatic Hydrocarbons, Halogenated B B – – – C – A C C C – – – A A A ChlorobenzeneAromatic, Hydroxy A C C C – C A A – C C – A – B C A PhenolEsters A B A B B B A A C C C – C – B B B Ethyl Acetate, Dioctyl PhthalateEthers A – A A – – – A – A B – C – A A A Butyl Ether, Diethyl EtherKetones A B A B B B A A C C C – B C A A A Methyl Ethyl Ketone, Acetone

Miscellaneous

Detergents A – A – B – – A A A – B A – A A B Laundry and Dishwashing Detergents, Soaps

Inorganic Salts B B B – A – – A – A – – A A B B B Zinc Chloride, Cupric SulfateOxidizing Agents, Strong A C C – C – B B – C – – A – C C C 30% Hydrogen Peroxide,

Bromine (Wet)Oxidizing Agents, Weak C C C A – A A A – A – A A A B C A Sodium Hypochlorite SolutionWater, Ambient A A B A A A A A A A A A A – A C BWater, Hot B C B C C B A A – C – A C – A C BWater, Steam A C C C C C B A – C – – C – A C –

Table 6.01 · Chemical resistance of various materials by chemical classes


42

7. Structural Analysis7.1 IntroductionIn the design or analysis of any mechanicalcomponent, a systematic approach is desirable.Frequently the product is one in which no significantloads or deflection limitations occur. In such cases,the “gut feel” of the design engineer is usually all thatis required. This is especially true with small, load-free plastic parts in which processing requirementsdictate a minimum wall thickness that is more thanadequate for the part function. Still, even in thesecases, engineers new to plastic design often neglectthe effects of stresses caused by assembly, handling,shipping, processing, temperature changes, and otherenvironmental changes.

In this section, simple analysis techniques arepresented that assist the design engineer in developingnew parts to handle anticipated loading while keepingstress and deflections within acceptable limits. Thesetechniques are also useful in product improvement,cost reduction, and failure analysis of existing parts.Thus, this chapter is devoted to applying simplifiedclassical stress and deflection equations to plasticparts. As the complexity of the part increases or whenvery accurate results are required, more exact classicalmethods or computerized finite element analysis maybe required. However, such methods are beyond thescope of this manual.

7.2 Defining the Structure

7.2.1 Loads The first step in analyzing any part is to determinethe loads to which the part is to be subjected. Theseloads generally fall into the following categories:

7.2.1.1 Directly Applied LoadsThese loads, usually easy to understand, are defined loads applied to specific part areas, eitherconcentrated at a point, line, or boundary ordistributed over an area. The magnitude and directionof such loads are known or are easily determinedfrom service conditions. In larger plastic parts, theweight of the part itself is a significant load. Figure7.01 presents examples of directly applied loads.

LIFTING LOAD

PRESSUREIN A PIPE

PORTABLE TELEVISION HANDLE

Fig 7.01 · Directly applied loads

METAL PIN

PLASTIC BOSS

PLASTICHOUSING

METALSCREW

PRESS-FIT PRODUCES FIXED HOOP STRAIN IN BOSS

SCREW TORQUE PRODUCES COMPRESSIVESTRAIN IN HOUSING

Fig 7.02 · Strain-induced loads


43

7

7.2.1.2 Strain-Induced LoadsFrequently, a part becomes loaded when it issubjected to a defined deflection. The actual loadresults from the structural reaction of the part to theapplied strain. Unlike directly applied loads, strain-induced loads depend on the modulus of elasticity ofthe material and, in the case of thermoplasticmaterials, generally decrease in magnitude over time.Many assembly stresses and thermal stresses are aresult of strain-induced loads. Figure 7.02 shows twocommon examples.

7.2.2 Support Conditions When a load is applied to a part, for the part toremain in equilibrium there must be equal forcesacting in the opposite direction. These balancingforces are the reactions at the supports. For purposesof structural analysis, several support conditions mustbe defined (Figure 7.03):

7.2.2.1 Free (Unsupported)This is the support condition in which the edge of a body is totally free to translate or rotate in any direction.

7.2.2.2 GuidedThis condition is similar to a free end except that theedge is prevented from rotating.

7.2.2.3 Simply SupportedIn this support condition, transverse displacement in one direction is restricted, as illustrated in Figure 7.03.

7.2.2.4 Held (Pinned)This is similar to a simply supported condition exceptthat only rotations are allowed.

7.2.2.5 Fixed (also Clamped or Built-in)This is a support condition at the end of a beam orplate which prevents transverse displacements androtations. The condition can be viewed as an endsupport firmly embedded into a fixed solid wall. Inpractice, especially with plastic parts, this conditionrarely exists in its pure form since the mountingpoints of the parts usually have some give.

Free, simply supported, and fixed are the mostfrequently encountered support conditions.

FREE

GUIDED

SIMPLYSUPPORTED

HELD

FIXED

Fig 7.03 · Support conditions


44

7.2.3 Simplifications and Assumptions For the purposes of the discussion and examples inthis chapter, the following simplifications andassumptions are made:

1 . The part under load can be broken down into one or more simple structures, beams, plates, pressure vessels, etc. for analysis.

2. The material being analyzed may be considered to be linearly elastic, homogeneous, and isotropic. While not necessarily true for plastic materials, this assumption is fundamental to the equations that follow.

3. The equations assume that the load is a single concentrated or distributed static load, gradually applied for a short period and then removed. However, creep, relaxation, or fatigue loads may be analyzed by using the same equations, provided the appropriate modulus and/or rupture (strength) conditions are applied.

4. The part being analyzed has no residual or molded-in stresses.

5. The equations apply to regions that are remote both from the point of application of the load and from any shoulder, hole, or other sudden change in dimension of the structure.

6. The equations may be used at shoulders, holes, or other sudden dimensional changes as long as appropriate stress concentration factors are used (see Chapter 8 of this manual or almost any machine design text or mechanical engineering handbook for stress concentration factors).

7.3 Safety Factors No firm rules are available for setting safety factors forplastic parts. However, the most importantconsideration is, of course, the consequence of failure.For example, a little extra deflection in an outside wallor a crack in one of six internal screw bosses may be oflittle concern, but the failure of a pressure vessel orwater valve may have serious safety or product liabilityimplications. Before any product is put into the market,actual parts should be tested at the most extremeoperating conditions. For example, a maximumworking load should be applied at the maximumtemperature and in the presence of any chemicals thatmight be expected in the end use. Further, the loads,temperatures, and chemicals to which a product isexposed prior to reaching its end use must not beoverlooked, e.g., paint oven temperatures.

Impact loading should be applied at the lowestexpected temperature, including shipping and

assembly temperatures. The effects of variations inresin lots and molding conditions must also be considered.

Many failures in testing on preproduction parts can often be corrected with selective use of increasedwall thickness, ribs, gussets, or elimination of stressconcentrations. A material change to another grade of the same resin or to a different plastic with a suitablemechanical property profile may also solve the problem.

Engineers unfamiliar with plastic product designoften ask for a design strength for use during theinitial design phase. A review of product data anddiscussions with experienced engineers suggest thatthe ranges in Table 7.01 are generally appropriate.These ranges, applied to the strength on standardproduct data sheets, are suitable for use with theequations presented here only in the preliminarydesign phase, where general product dimensions arebeing evaluated. Any product designed by using theseguidelines must be thoroughly tested to ensuresatisfactory performance.

7.3.1 Failure Criteria Experienced designers not familiar with plastics oftenask, “What failure criterion should be used?” Thisquestion implies the rationalization of complex stressstates, which are beyond the scope and intent of thismanual. For the simple stress states considered here,the design strength given in Table 7.01 is adequate forpreliminary design analysis. However, for designerswho choose to perform the more complex analyses,the maximum shear (a.k.a. Coulomb, a.k.a. Tresca)theory of failure is suggested. It is further suggestedthat the shear strength be taken as the published shearstrength or, one-half the tensile strength, whichever islower. Even better, one-half the strength at the elasticlimit, if available, is preferred.

Failure FailureNoncritical Critical

Intermittent(nonfatigue) 25 – 50% 10 – 25%Loading

Continuous 10 – 25% 5 – 10%Loading

* Suggested percentage of strength values published in product datasheets, based on type of stress and maximum temperature. These areintended for preliminary design analysis only, and are not to be used inplace of thorough product testing.

Table 7.01 · Design strength* forpreliminary part design


45

7

d

b

c

RECTANGULAR

A = bd

c =d2

I =bd3

12

Z =bd2

6

na

CIRCULAR

d

c

c =d2

I =πd4

64

Z =πd3

32

A =πd2

4

I-BEAM

d

c

b

hnat

s

A = bd - h(b - t)

c =d2

I =bd3 - h3(b - t)

12

Z =bd3 - h3(b - t)

6d

H-BEAM

h

b

d

cna

A = bd - h(b - t)

c =b2

I =2sb3 + ht3

12

Z =2sb3 + ht3

6b

na

TUBE

do

c

dl

c =do

2

I =π(do

4 - d l4 )

64

Z =π(do

4 - d l4 )

32do

A =π(do

2 - d l2)

4

b

t

c

na h

s

d

C-BEAM

A = bd - h(b - t)

c =d2

I =bd3 - h3(b - t)

12

Z =bd3 - h3(b - t)

6d

b

h

s

na

cd

T-BEAM OR RIB

c = d - d2t + s2(b - t)2 (bs + ht)

I =tc3 + b(d - c)3 - (b - t)(d - c - s)3

3

A = bs + ht

Z =Ic

Z l =I

(d - c)

U-BEAM

c

A = bd - h(b - t)

c = b - 2b2s + ht2

2A

I =2b3s + ht3

3bna

h

d

- A(b - c)2

Z =Ic

Z l =I

(b - c)

s

t

t

t

na

s

Fig 7.04 · Section properties for some common cross-sections (na = neutral axis)


46

SIMPLY SUPPORTED BEAMCONCENTRATED LOAD AT CENTER

L

F

Y

(at load) σ =

(at load) Y =

FL4Z

FL3

48EI

CANTILEVERED BEAM (ONE END FIXED)CONCENTRATED LOAD AT FREE END

F

L Y

(at support) σ =

(at load) Y =

FLZ

FL3

3EI

CANTILEVERED BEAM (ONE END FIXED)UNIFORMLY DISTRIBUTED LOAD

LY

(at support) σ =

(at free end) Y =

FL2Z

FL3

8EI

F (total load)

SIMPLY SUPPORTED BEAMUNIFORMLY DISTRIBUTED LOAD

L

F (total load)

Y

(at center) σ =

(at center) Y =

FL8Z

5FL3

384EI

BOTH ENDS FIXEDCONCENTRATED LOAD AT CENTER

L

F

(at supports) σ =

(at load) Y =

FL8Z

FL3

192EI

BOTH ENDS FIXEDUNIFORMLY DISTRIBUTED LOAD

L

(at supports) σ =

(at center) Y =

FL12Z

FL3

384EI

Y

F (total load)

L2

Y

L2

Fig 7.05 · Maximum stress and deflection equations for selected beams


47

7

7.3.2 Beam Bending Stresses As indicated in Chapter 3, in simple beam bendingtheory, the following additional assumptions are made:

1 . The beam is initially straight, unstressed, and symmetric.

2. The proportional limit is not exceeded.3. Young’s modulus for the material is the same in

tension and compression.4. All deflections are small, so that plane cross-

sections remain planar before and after bending.

The maximum stress occurs at the surface of the beamfarthest from the neutral surface and is given by

σ = Mc

= M

I Z

where

M = bending moment, inch pounds c = distance from neutral axis to outer surface where

maximum stress occurs, inches (see Figure 7.04)I = moment of inertia, inches4 (see Figure 7.04) Z = l/c, section modulus, inches3 (see Figure 7.04);

this is a geometric property not to be confused with the modulus of the material, a material property.

I, c, Z, and the cross-sectional area, A, for somecommon cross-sections are given in Figure 7.04. Themaximum stress and deflection equations for somecommon beam loading and support geometries aregiven in Figure 7.05. Note that for the T- and U-sections in Figure 7.04, the distance from the neutralsurface to the outermost surface is not the same forthe top and the bottom of the beam. Occasionally, itmay be desirable to determine the maximum stress onthe “other” surface, particularly if it is in tension. Forthis reason, ZI is provided for these two sections.

7.3.3 Shear Stress, Torsion A shaft subjected to a torque (see Figure 7.06) isgenerally considered to have failed when the strengthof the material in shear is exceeded. For a torsionalload, the recommended shear is the published valueor one-half the tensile strength, whichever is less. Themaximum shear stress on a shaft in torsion is given by

τ = TcJ

where

T= applied torque, inch-pounds c = distance from the center of the shaft to the

location on the outer surface of the shaft where maximum stress occurs, inches (see Figure 7.07)

J = polar moment of inertia, inches4 (see Figure 7.07)

The angular rotation of the shaft due to the torque isgiven by

θ = TLGJ

where

L= length of the shaft, inchesG= shear modulus, psi

= E

2 (1 + ν)

E= Young’s modulus (tensile modulus), psi ν= Poisson’s ratio (if unknown, use 0.35)

θ

L

d T

Fig 7.06 · Shaft of diameter, d, and length, L,subjected to torque, T, with angular deformation, θ


48

d

di

do

b

h

h

b

h

h

CROSS-SECTION

POLAR MOMENTOF INERTIA,

J

πd4

32

π(do4 – di

4)

32

πb3h32

bh(b2 + h2)

12

h4

6

LOCATION OF MAXSHEAR STRESS,

C

d2

do

2

b2

b2

h2

Fig 7.07 · Polar moments of inertia forcommon cross-sections

PARTSTO BE

ASSEMBLED

SPRING CLIPIN PLACE

SHEARLOAD

F

F

d

SPRING CLIP ASSEMBLY

ASSEMBLY BYSTAKING

d

F

F

SHEARLOAD

SHEAR AREA

A = πd2

4

AFTER STAKING

SPRING CLIP

Fig 7.08 · Direct shear examples

7.3.4 Shear Stress, Direct ShearMany situations occur in which direct shear is appliedto a plastic part. Figure 7.08 illustrates assembly bystaking and by a spring clip in which a load may beapplied in direct shear. Other examples include spotwelds (Figure 9.11) and pinned structures such ashinges and conveyor links. For direct shear (as inFigure 7.08), the shear stress is simply the appliedload divided by the shear area:

τ = FA

Direct shear situations, such as those illustrated, arethe only appropriate times to use the shear strengthdata reported on marketing data sheets. However,since the load is not only transferred by shear, butalso contains a considerable bending and/orcompressing component, all of which depend greatlyon geometry, the actual strength can vary greatly.Therefore, it is suggested that safety factors beincreased significantly in direct shear situations.


49

7

7.4 Pressure Vessels The most common plastics pressure vessel applicationis a tube with internal pressure. In the selection of thewall thickness for the tube, the thin-wall tube hoopstress equation is convenient (see Figure 7.09). It isvery useful in determining an approximate wallthickness, even when the condition t < d/10 is notmet. After the thin-wall equations have been applied,the thick-wall equation in Figure 7.10 can be used toverify the design.

7.5 Press-Fits Although the use of press-fit assemblies can bedangerous with thermoplastic parts, their low cost,assembly, speed, and convenience often result in theiruse. A common use is with a plastic hub or bossaccepting either a plastic or metal shaft or pin. Thepress-fit operation tends to expand the hub, creating atensile or hoop stress. If the interference is too great,a very high strain and stress develops. The plastic part either:

1 . fails immediately by developing a crack parallel tothe axis of the hub relieving the stress, a typical hoopstress failure, or

2. survives assembly but fails prematurely when thepart is in use for a variety of reasons related to thehigh induced stress levels, or

3. undergoes stress relaxation sufficient to reduce thestress to a lower level that can be maintained.

Hoop stress equations for two typical press-fitsituations are shown in Figure 7.11. The allowabledesign stress depends on the particular plastic resin,temperature, and other environmental considerations.

A simpler, although less accurate, method ofevaluating these press-fits is to assume that the shaftwill not deform when pressed into the hub. This isreasonably accurate when a metal shaft is used in aplastic hub. The hoop strain developed in the hub isthen given by

ε = i

d i

The hoop stress can then be obtained by multiplyingby the appropriate modulus. For high strains, thesecant modulus gives the initial stress. However, forlonger term stresses, the apparent or creep modulusshould be used. The main point is that the maximum

strain or stress must be below the value that producescreep rupture in the material. It must be noted that aweld line is usually present in the hub, which cansignificantly affect the creep rupture strength of mostplastic materials. An additional frequent complicationwith press-fits is that a round hub or boss is oftendifficult to mold. There is a tendency for the hub tobe slightly elliptical in cross-section, increasing thestresses on the part. In view of the preceding, allpress-fits must be given end-use testing under actualoperating conditions to assure product reliability.Design alternatives to a standard press-fit are given in Chapter 9.

d

t

UNIFORM INTERNAL PRESSURE, P

HOOP STRESS

Pd2t

σ =

This equation is reasonably accurate for t < d/10. As the wallthickness increases, the error becomes quite large.

Fig 7.09 · Cylindrical pressure vessel, thin wall tube

UNIFORM INTERNAL PRESSURE, P

HOOP STRESS

1 + R1 - R

σ = P

This equation is for the maximum hoop stress which occurs at the surface of the inside wall of the tube.

di

do

WHERE

di

doR = ( )

2

Fig 7.10 · Cylindrical pressure vessel, thick wall tube


50

ds

dl

do

GEOMETRY FACTOR

Γ =

ds

do( )

2

1 +

ds

do( )

2

1 –

Ep = MODULUS OF ELASTICITY OF PLASTIC HUB OR BOSSνp = POISSON'S RATIO OF PLASTICσa = ALLOWABLE DESIGN STRESS FOR PLASTICi = ds – di = DIAMETRAL INTERFERENCEia = ALLOWABLE INTERFERENCE

CASE ASHAFT AND HUB ARE BOTH THE SAME OR ESSENTIALLYSIMILAR MATERIALS. HOOP STRESS GIVEN i IS

OR, THE ALLOWABLE INTERFERENCE IS

CASE BSHAFT IS METAL, HUB IS PLASTICHOOP STRESS GIVEN i IS

OR, THE ALLOWABLE INTERFERENCE IS

ids

σ = Ep ΓΓ + 1

σaEp

ia = ds Γ + 1 Γ

ids

σ = Ep ΓΓ + νp

σaEp

ia = ds Γ + νp

Γ

Fig 7.11 · Press-fit equations for two typical situations 7.6 Thread Strength When threads are either molded or tapped into aplastic part, the assembly torque must be controlledto prevent excessive shear stress, which results instripped threads, and to limit hoop stress, which canresult in tensile failures. Although the mechanics ofstress analysis for screw threads are readily available,the equations are complicated and few wish to takethem on. For this reason, the following simple,approximate equations are provided (see Figure 7.12).

The relationship between the torque on a screw andthe generated axial force is approximately

T = 4

µdF3

where

µ = dynamic coefficient of friction between the sliding surfaces.

The shear area of the threads, A, is approximatelyone-half the thread engagement cylinder or

A = 1

πdL2

Therefore, the thread-stripping shear stress is

τ = F

= 2F

A πdL

L

F

F AXIAL FORCE CREATED BYTORQUE

t

d

dp

NOMINAL THREAD OD

PITCH DIA. OF THREAD

T TIGHTENING TORQUE

EFFECTIVETHREADENGAGEMENTLENGTH

Fig 7.12 · Illustration for torque-forcerelationships for screw threads


51

7

For standard 60° unified threads, the radialcomponent of force is approximately 60% of F. Thisforce is spread over the thread engagement cylinder,producing an internal pressure in the boss. Therefore,the internal pressure in the boss is approximately

P = 0.6FπdL

This pressure can then be used in the simple thin-wall, hoop stress equation:

σ = P(d + 2t)

2t

to obtain the hoop stress produced in the boss.

A major problem with the thread torque-forceequations is that the coefficient of friction variessignificantly with material and surface finish. Inaddition, most published data for coefficient offriction are generated from high speeds and low loadconditions rather than the low speed and high loadsinvolved in thread engagement. Thus, theseapproximate equations are probably adequate.Furthermore, for initial design purposes, it isprobably worthwhile to pick a median value for thecoefficient of friction. A value of 0.15 is suggested.When this value is used, the previous equations canbe further simplified as follows:

The torque-force relationship is

T = 0.2dF or F = 5Td

The thread-stripping shear stress equation is

τ = 2F

= 10T

orπdL πd2L

The pressure generated on the inside of the boss is

P = 0.2F

= T

dL d2L

The hoop stress generated in the boss is

σ = 0.2F

=T

Lt dLt

if it is assumed that d + 2t ≈ 2d.

10TdπdL

( )

Finally, the following observations should be clearlyunderstood concerning screw threads:

1. The torque values are based on the coefficient offriction of the mating parts and can vary significantly.The use of any compatible lubricant that reducesfriction also increases the shear and hoop stresses ifthe torque remains the same. Therefore, the allowabletorque must be reduced when lubricants are present.

2. High assembly torque for the purpose ofpreventing vibrational loosening is frequentlyineffective since creep of the plastic material reducesthe effective assembly torque even if the fastener doesnot rotate. Vibration-proof screws, lock washers, locknuts, and thread-locking adhesives, all specificallydesigned for thermoplastics, are usually a betteralternative when loosening is considered a problem.

3. Self-tapping screws require additional torque tocut or form the thread. This torque must usually beadded to the allowable safe assembly torque only forthe first assembly. The appropriate hole design forself-tapping screws depends greatly on the materialand screw design. Consult the screw manufacturerand Ticona for design recommendations.

7.7 Pipe Threads Pipe threads are commonly used in plastic plumbingand pneumatic devices. Properly designed plastic pipethreads usually require only hand-tight assembly toeffect a good seal, especially if a compatible sealanttape or compound is used. Assembling a tapered malepipe thread into a mating female thread in a plasticpart is analogous to driving a cone into a round hole.Many split bosses are a result of improper fieldinstallation. Figure 7.13 shows some alternatives.Sometimes the use of straight threads and an O-ringseal can avoid the need for pipe threads. When pipethreads must be used, torque control is critical. Thefollowing general recommendations should befollowed whenever possible:

1 . When metal is mated with plastic using pipethreads, make the plastic-threaded member the malethread so that the plastic is in compression.

2. When torque can be controlled during assembly offemale plastic pipe threads, specify fluoroplastic tapeand hand-tighten only.


52

3. When torque cannot be controlled, such as duringfield assembly, consider the use of an external ormolded-in hoop ring.

4. Never design flats into plastic parts with threadsfor assembly purposes. This only encouragesovertightening. If necessary, add “wings” or atextured surface for improved grip.

An approximate formula for the hoop stressproduced in a plastic boss with internal pipe threadsas illustrated in 7.13 is

σ = 3TtdL

As with the screw thread equations in the previoussection, this equation assumes certain geometricrelationships and a coefficient of friction of 0.15.Obviously, the torque must be reduced if compatiblethread lubricants are used during assembly. As withall threaded assemblies, long-term testing under theextremes of operating pressure, temperature, andinstallation procedure is essential to ensure reliability.

7.8 Impact Loads One of the assumptions made at the beginning of thischapter is that the load is gradually applied. In reality,loads are often applied abruptly. This has the effect ofsignificantly increasing the stress and strain. However,due to the elasticity of most thermoplastic materials,recovery is usually complete and the assumption isvalid. Therefore, the steady-state stress and deflectionof the plastic part is, for all practical purposes,identical to that of the part loaded gradually.

However, when impact loading becomes severe,impact failure can result. Figure 7.14 presents amethod for estimating an “impact stress” and “impact deflection” for the case of a falling weight. (It is interesting to note that when the drop height is zero, the stress and deflection double.)

Many high-impact materials can survive very largedeflections or strains during impact without thepermanent deformations or failure one would expectfrom the stress-strain curve of the material measuredat standard loading rates. Therefore, the calculatedimpact stress of successful parts often appearsunreasonably high. For many of these materials, the

stress-strain characteristics are very different underrapid loading conditions as compared to the slow,steady loading conditions used in normal testing ofplastic resin specimens. This partially accounts for theability of many plastic parts to successfully dissipatelarge amounts of mechanical energy when subjectedto impact.

BETTER DESIGNS

METAL FITTING

PLASTICMALEPIPETHREAD

PLASTICBOSS

METALPIPE

d

t

L

POOR DESIGN

PRESSED-ONOR MOLDED-IN METALHOOP RING

Fig 7.13 · Designing with pipe threads


53

7

W

ys

W

DROP HEIGHT

H

Calculate static stress anddeflection from placingweight on part.

Static Stress

σa

Static Deflection

ys

Impact Stress

σs= σs (1 + 1 + 2Hys )

yi = ys (1 + 1 + 2Hys )

NOTES:— This is only an estimate; actual testing is essential.— Calculated impact stress will often exceed ultimate stress listed in data sheets for high-imact materials.

Fig 7.14 · Estimating impact stress and deflectiondue to dropping a weight on the part

7.9 Thermal Stresses When materials with different coefficients of thermalexpansion are bolted, riveted, bonded, crimped,pressed, welded, or fastened by any method thatprevents relative movement between the parts, thereis a potential for thermal stress. The typical casewhere problems develop (see Figure 7.15) involves thejoining of nonreinforced thermoplastic parts withmaterials such as metals, glass, or ceramics, whichgenerally have much lower coefficients of thermalexpansion.

Figure 7.16 illustrates the equations for thermalexpansion that apply in various situations. The basicrelationship for the thermal expansion of a part is

∆L = αL∆T

where

∆L = change in lengthα = coefficient of linear thermal expansion

(see Table 4.01)L = linear dimension under consideration

(including hole diameters)∆T = temperature change

If the part is confined so that it cannot expand orcontract, the strain induced by the temperaturechange is

εT = ∆L

= α∆TL

The stress can then be calculated by multiplying thestrain by the tensile modulus of the material at thegiven temperature. However, a typical situationoccurs where a plastic part is mounted to a metal part.While both expand due to a change in temperature,the plastic usually applies insignificant load to themetal, while a considerable stress is generated in theplastic. In this case, the approximate thermal stress, σT, in the plastic is

σT = (αm - αp) Ep∆T

where

αm = coefficient of thermal expansion of the metal

αp = coefficient of thermal expansion of the plastic

Ep = tensile modulus of the plastic at the temperature in question


54

Note that as the temperature increases, most plasticsexpand more than metals while their modulus drops.This produces a compressive load in the plastic part,which often results in buckling. Conversely, as thetemperature drops, the plastic part shrinks more thanthe metal part and develops an increased tensilemodulus. This can result in tensile rupture of theplastic part.

In many assemblies, clearances around fasteners,failure or yield of adhesives, warpage, or creep tend torelieve the thermal stress. Good design practice allowsfor these temperature changes, especially with largeparts, which can experience wide temperature swings.Automotive interior and body panels are typicalexamples of such parts. This is often done byallowing for relative motion, as illustrated in Figure7.16. The relative motion, ∆Lrel, between twoattachment points of joined plastic and metal parts in which motion is allowed is as follows:

∆Lrel = (αp - αm) L∆T

PLASTIC

CAST IRON

PARTS ARE TIGHTLY BOLTED OR BONDED TO METAL

GLASS WINDOWBONDED TOPLASTIC

PLASTICFRAME

ADHESIVE

PLASTICASSEMBLY

STEELBOLT

Fig 7.15 · Typical assemblies in which thermalstress can be a problem

d = DIAMETER OF HOLE

d ∆d

L

FREE EXPANSION

L = LENGTH OF BEAM CHANGE IN ANYLINEAR DIMENSION∆L = αL∆T ∆d = αd∆T

L

RESTRICTED EXPANSION

STRAINSTRESS

εT =∆LL

σT = EεT = α∆T

ASSEMBLY WITH METAL

STRESS IN PLASTICσT = (αp – αm)Ep∆T

PLASTIC

METAL

SHOULDER SCREW

WASHER

RELATIVE MOTION

RELATIVE MOTION BETWEENPARTS AT SHOULDER SCREW

∆Lrel = (αp – αm) L∆T

Fig 7.16 · Equations of thermal expansions forvarious conditions


55

8

8. Design Considerationsfor Injection Molded Parts

The most prevalent method for producingthermoplastic parts in large quantities is injectionmolding. This is a highly economical, efficient, andprecise manufacturing method, which can be highlyautomated and produces almost no waste. It does,however, require expensive machinery and tooling. In addition, certain established geometric and other design considerations must be followed toconsistently produce quality molded parts. Thesedesign considerations, or rules for injection-moldedparts, are the subject of this chapter.

8.1 Injection Molding Process A brief explanation of the injection molding processis useful at this point to help understand theimportance of proper design (see Figure 8.01). Themolder receives plastic resin in the form of smallchopped pellets (usually 1/8 in. long). These are fedinto the hopper of an injection molding machine,where they fall into an augur-type screw channel,which feeds the pellets forward inside the heatedbarrel. As the mass of plastic moves toward the frontof the barrel, it is plasticized or melted. The screw is

allowed to travel back until a sufficient quantity ofmolten plastic accumulates in front of the screw to fillthe cavity in the mold. The screw is then pushedforward under high pressure to force the moltenplastic through the machine nozzle into the closedmold. Once in the mold, the plastic flows through adistribution system called runners and then throughgates into the part cavities. As soon as the plasticcools and solidifies in the mold cavity, the mold isopened and the part is removed.

The mold is usually heated or cooled to provide theproper temperature for plastic solidification. Themold also has some type of mechanical assist, calledejection, to help extract the part from the mold. Whilethe part is cooling in the mold, the next shot is beingplasticized within the barrel. The mold then closesand the process is repeated. Although this is a veryelementary explanation of the injection moldingprocess, it should be observed that the cooling time inthe mold is usually the controlling factor in the totalmolding cycle time and, thus, a key factor indetermining the production rate of the machine. Anoverly thick wall section, even in just a small portionof the part, can significantly lengthen the cooling timeand hurt the overall economics of the part.

MOLDBACK FLOWCHECK VALVE

MOTORS, PUMPS, VALVES, OIL TANK,HEAT EXCHANGERS, ETC.

MOLD CLAMPSYSTEM

HEATER BANDS

HOPPER

SCREWMOTOR INJECTION

PISTON

CONTROLS

SCREW TRAVELLIMIT SWITCHES

Fig 8.01 · Schematic of reciprocating screw injection molding machine


56

8.2 Design Strategy In the design of any injection-molded part, thedesigner should strive for the following goals:

1. Maximize FunctionalitySince injection molds are costly, the engineer shoulddesign as much function as possible into each part.This means that a single part may take the place ofmany individual separate parts, eliminating assemblyoperations, reducing weight, and frequentlyimproving the overall structural integrity. Figure 8.02illustrates this point by comparing a sheet metalelectronics enclosure with a well-designed injectionmolded plastic replacement.

2. Optimize Material SelectionOn the basis of the desired functionality of the part,the operating environment, cost constraints, and anyspecial requirements, several candidate materialsshould be selected for the proposed part.

3. Minimize Material UseThe minimum volume of plastic that satisfies thestructural, functional, appearance, and moldingrequirements of the application is usually the bestchoice. This contrasts sharply with machiningoperations, where one starts with a solid block ofmaterial and machines away until only the materialnecessary to make the part function remains. Figure8.03 illustrates this contrast between a machined valvebody and a properly designed injection molded,equally functional part.

The remainder of this chapter presents guidelines forimplementing these goals.

8.3 Efficient and Functional Design Injection molding should be considered when a new part is being redesigned or an existing part isredesigned. The key advantage is injection molding’sability to accurately and repeatedly producemultifunctional or complex molded plastic parts in asingle, highly automated operation. Keeping this inmind during the initial planning stage, one shouldconsider the following questions:

INJECTION MOLDEDTHERMOPLASTIC

SHEET METAL

Fig 8.02 · Injection-molded one piece housingreplaces fabricated sheet metal and manymiscellaneous parts and eliminates most

assembly operations

MACHINED VALVE BODY

INJECTION MOLDED

Fig 8.03 · Minimize material usage for plastic redesign


57

8

1 . If the product consists of more than one part, canthey be combined into a single molded part,eliminating extra materials, molds, moldingoperations, and assembly procedures?

2. Can hardware items or other components beeliminated from the proposed part by integratingthem with the molded part?

If the answer to either of the above questions is yes,more work is probably needed in planning the mostefficient and cost-effective product. Of course, insome cases, trying to design multiple features into asingle part or putting too much complexity into a partcauses the tooling to be prohibitively expensive or thepart too difficult to mold. This is especially true whenthe mold must open in many directions to eject themolded part. The product engineer must weigh the increased cost for a more complex tool against the cost savings per unit over the expected production volume.

8.4 Material Selection The wide variety of injection moldable thermoplasticmaterials often makes selection of a plastic resin verydifficult. The chosen material is often one used beforeor one that a molder, toolmaker, or material supplierrecommends. Unfortunately, the person making theselection may not fully understand the requirementsof the application or the properties of the plastic.Therefore, the material selected is not always optimal.The designer should carefully consider the followingitems, not necessarily in order of importance.

1. Temperature Determine the capability of the plastic material towithstand the normal, as well as extreme, operatingtemperature of the product. Be sure to check shippingtemperatures, sources of internal heat, and anyassembly or finishing operations. Keep in mind thatmechanical and electrical properties usually dependon temperature. (Consult Chapter 4 of this manual.)

2. Environment Every substance, solid, liquid, or gas, that can come incontact with the plastic part over its expected life span must be considered for chemical compatibility. If the compatibility information is not obvious fromthe product literature, the engineer should consult the material supplier and request or conduct specific testing if necessary. (Consult Chapter 6 of this manual.)

3. Agency Approvals Refer to the material product data sheets to ensure that the resin being considered meets therequirements of the end product. These approvalscould include Underwriters Laboratories (UL),Canadian Standards Association (CSA), US Food and Drug Administration (FDA), National SanitationFoundation (NSF), United States Department ofAgriculture (USDA), a variety of MilitarySpecifications, industry-specific requirements, andmany others. Many of these approvals for any specificresin grade can depend on such factors as minimumwall thickness, color additives, fillers, and theoperating temperature.

4. Assembly Check to ensure that the proposed material lendsitself to the expected assembly operations. Forexample, certain classes of materials are difficult tosolvent bond, while others may not work well withultrasonic methods. (Consult Chapter 9 of thismanual.)

5. Finish Determine whether plastic resin can be easily moldedinto the proposed product with the desired finalappearance. If not, can the part be easily andeconomically finished? (Consult Chapter 10 of this manual.)

6. Cost To determine the economics of using a particularresin, consider the cost per pound in the applicablecolor and purchase volume. In addition, look at thespecific gravity and typical molding cycle times of theparticular resin grade, since these affect the finalmolded part cost. The cost per unit volume of aparticular plastic resin (or any material) is as follows:

Cost/in.3 =0.0361 x Specific Gravity x Resin Cost/Pound

The material cost of a part is obtained by multiplyingthe cost/cubic inch by the part volume. Finally, a veryrough part cost estimate is obtained by doubling thematerial cost of the part.

7. Availability Check to see that the chosen resin will be readilyavailable in sufficient quantities when needed for production.


58

8.5 Nominal Wall Thickness Of all the issues in plastic design, selecting the propernominal wall thickness is probably the mostimportant and all-encompassing topic. Just aboutevery subject addressed in this manual somehowrelates to, affects, or is influenced by the wallthickness. Choosing proper wall sections sometimesdetermines the ultimate success or demise of aproduct. While an inadequate wall section can lead topoor performance or structural failure, a section thatis too heavy, even in just certain regions, can make theproduct unattractive, overweight, or too expensive.Although some problems can be corrected after themold is built, such solutions are often expensive.

The following discussion on determining wall sectionthickness should help the design or productionengineer eliminate potential problems on paper (orcomputer screen) rather than in tool steel. In manyparts, only some of the guidelines can be followeddue to geometric, structural, or functionalrequirements, but at least the potential existence of aparticular problem is known in advance and remedialaction can be planned. For example, if a surface defectis discovered to be likely to appear in a visible areaduring molding, a texture, logo, or label can beplanned for that region.

8.5.1 Normal Ranges of Wall Thickness Just as metals have normal working ranges of wallthicknesses based on their processing methods (ie.,casting, bending, forging, and extruding), so doinjection-molded plastics. The vast majority ofinjection-molded plastic parts probably range from1/32 in. (0.80 mm) to 3/16 in. (4.8 mm), with thethickness within that range generally related to thetotal size of the part. That does not mean that partscannot be molded to be thinner or thicker, or that abig part cannot be thin or a tiny part thick. However,these norms can act as a starting point for the design.Table 8.01 shows some general guidelines for variousclasses of thermoplastics.

The design engineer should also refer to data relatedto the ability of a plastic resin to flow into the moldcavity. This information, usually shown in the formof spiral flow curves, gives a relative measure of how far one can expect the plastic resin to flow fromthe gate. Figure 8.04 shows some typical spiral flow curves.

Thermoplastic Resin Family Typical ThicknessRanges (inches)

ABS, Acrylonitrile-Butadiene-Styrene 0.045 - 0.140Acetal 0.030 - 0.120Acrylic 0.025 - 0.150Liquid Crystal Polymer 0.008 - 0.120Long-Fiber Reinforced Plastics 0.075 - 1.000Modified Polyphenylene Ether 0.045 - 0.140Nylon 0.010 - 0.115Polyarylate 0.045 - 0.150Polycarbonate 0.040 - 0.150Polyester 0.025 - 0.125Polyester Elastomer 0.025 - 0.125Polyethylene 0.030 - 0.200Polyphenylene Sulfide 0.020 - 0.180Polypropylene 0.025 - 0.150Polystyrene 0.035 - 0.150Polysulfone 0.050 - 0.150Polyurethane 0.080 - 0.750PVC, Polyvinyl Chloride 0.040 - 0.150SAN, Styrene-Acrylonitrile 0.035 - 0.150

Table 8.01 · Typical Nominal Thickness forVarious Classes of Thermoplastics

FLO

W L

ENG

TH

INJECTION PRESSURE

1 2 3

4

Fig 8.04 · Some typical spiral flow curves

1. Nylon 6/62. Thermoplastic polyester, PBT

Liquid crystal — glass reinforcedPolyphenylene sulfide — glass reinforced

3. Acetal copolymer4. PBT — glass reinforced


59

8

8.5.2 Structural Requirements of the Nominal WallIf a part is subjected to any significant loading, theload-bearing areas should be analyzed for stress anddeflection. If the stress or deflection is too high, thefollowing alternatives should be considered:

1 . Use ribs or contours to increase the sectionmodulus. This is often the most economical solutionand is discussed in detail in the structuralreinforcement section.

2. Use a higher strength, higher modulus material.

3. Increase the wall section if it is not already too thick.

8.5.3 Insulation Characteristics of the Nominal WallPlastic materials are good insulators for electrical andheat energy. They can also serve as barriers and filtersfor sound and light. In general, insulating ability isdirectly related to the thickness of the plastic. In thecase of sound transmission, change in thickness maybe needed to change the resonant frequency of aplastic housing (Consult Chapter 5 of this manual for electrical properties.)

8.5.4 Impact Response of the Nominal Wall The impact resistance of a particular part is directlyrelated to its ability to absorb mechanical energywithout fracture or plastic deformation. This, in turn,depends on the properties of the plastic resin and thegeometry of the part. Increasing wall thicknessgenerally improves the impact resistance of moldedparts. However, increased wall thickness could hurtimpact resistance by making the part overly stiff,unable to deflect and distribute the impact. Both ofthese methods of absorbing impact energy should be examined when the nominal wall thickness is being selected.

8.5.5 Agency Approvals and the Nominal Wall When materials are recognized for flammability, heat resistance, electrical properties, or othercharacteristics by agencies, these ratings are usuallybased on a certain wall thickness. When agencyapprovals are important, the design engineer may belocked into a thicker wall section than would benecessary for mechanical purposes.

8.6 Draft On most molded parts, some part features must becut into the surface of the mold perpendicular to theparting line. To properly release an injection-moldedpart from the tool, plastic parts are almost alwaysdesigned with a taper in the direction of moldmovement. This is commonly referred to as draft inthe line of draw. This allows the molded part to breakfree by creating a clearance as soon as the mold startsto open. Since plastic materials shrink as they cool,they grip cores (male parts) of the mold very tightly.Without sufficient draft, normal ejection from themold can be difficult. Although exceptions occur, adraft of 1/2° per side is considered a minimum, with1.5-3° per side being frequently recommended.

Draft Angle (degrees)

1/8 1/4 1/2 1 2 3 4 5

1 0.002 0.004 0.009 0.017 0.035 0.052 0.070 0.0872 0.004 0.009 0.017 0.035 0.070 0.105 0.140 0.1753 0.007 0.013 0.026 0.053 0.105 0.157 0.210 0.2634 0.009 0.018 0.035 0.070 0.140 0.210 0.280 0.3505 0.011 0.022 0.044 0.087 0.175 0.262 0.350 0.4376 0.013 0.026 0.052 0.105 0.209 0.314 0.420 0.5257 0.015 0.031 0.061 0.123 0.244 0.367 0.490 0.6128 0.018 0.035 0.070 0.140 0.279 0.419 0.559 0.7009 0.020 0.040 0.078 0.158 0.314 0.472 0.629 0.78710 0.022 0.044 0.087 0.175 0.379 0.524 0.669 0.875

Table 8.02 · Dimensional Difference versus DraftAngle Dimensional Change in Inches* forVarious Draft Angles and Draw Depths

Dra

w D

epth

(inc

hes)

DEPTH OF DRAW

DIMENSIONDIFFERENCE

DRAFT ANGLE

PARALLELDRAFT

Fig 8.05 · Illustration of draft and associateddimension change with draft angle


60

The required amount of draft depends on the surfacefinish on the mold. A highly polished mold requiresless draft than an unpolished mold, and any surfacetexture increases the draft required by at least 1° perside for every 0.001 in. depth of texture. When deepdraws are required with only minimal draft, theproduct engineer should consult with a toolmakerand molder to see if the part can be properly ejected.Sometimes special mold surface treatments or regularapplication of mold release spray is necessary duringproduction.

Figure 8.05 and Table 8.02 provide a quick referencefor determining the amount of dimensional changerequired due to the draft angle. A general rule ofthumb used by designers and toolmakers is:

1° of draft yields 0.017 inches of taper per inch length.

One of the difficulties encountered when draft isapplied to a part is the creation of very heavy walls.Sometimes these conditions can be remedied by theuse of parallel draft as shown in Figure 8.05. Withparallel draft, the wall sections can be kept uniform.

8.7 Structural Reinforcement To increase the load-carrying ability or stiffness of aplastic structure, an increase is required either in theproperties of the plastic material or in the sectionproperties of the structure. Improving the material,by changing grade or by going to a glass-reinforcedversion of the same material, is sometimes adequate,but is often not practical or economical. Increasingthe section properties, i.e., the moment of inertia andthe section modulus, is usually necessary. Aspreviously discussed, thickening wall sections is oftena practical means of increasing section properties, buteconomic limitations can exist in terms of material useand molding cycles. The relative effect on load-carrying capacity (maximum allowable stress) andstiffness (maximum allowable deflection) for materialchanges and wall section changes are shown in Figure 8.06.

8.7.1 Ribs If sufficient space is available, the use of ribs is apractical and economical means of increasing thestructural integrity of injection-molded plastic partswithout creating thick walls. The efficiency of aribbed structure can be seen in the following example.Consider the loaded beam shown in Figure 8.07 to bea simplification of a section of an actual plastichousing subjected to a concentrated load of 5 lbs.

Based on a constant wall section of 0.080 in., themaximum stress is calculated to be 6250 psi, and themaximum deflection, if the flexural modulus is300,000 psi, is calculated at 0.694 in. It is determinedthat both the stress and deflection are far too highand, thus, unacceptable for the plastic resin chosen.To reinforce the structure, a 0.040 in. wide by 0.400in. high rib with 1/2° draft per side is added to theunderside of the part. By use of the new sectionproperties, the calculated stress drops to 2273 psi andthe deflection drops to 0.026 in., which is nowdeemed acceptable for both the resin and theapplication. To achieve the same reduction in

% O

F O

RIG

INA

L ST

RESS

AN

DD

EFLE

CTI

ON

AT

CO

NST

AN

T LO

AD

0 20 40 60 80 100% INCREASE IN WALL SECTION

100

80

60

40

20

0

STRESSDEFLECTION

% IN

CRE

ASE

IN L

OA

D C

APA

CIT

Y A

TC

ON

STA

NT

STRE

SS A

ND

DEF

LEC

TIO

N

0 20 40 60 80 100% INCREASE IN WALL THICKNESS

800

STRESS

DEFLECTIO

N700600500400

300

200

100

Fig 8.06 · Interaction of wall thickness, load,stress, and deflection for a rectangular

cross-section in bending

0.080

PLASTICE = 300,000 psi

LOAD = 5 LBS

4"

2"

.75

Fig 8.07 · Example of the effect of ribs onrectangular cross-section


61

8

deflection by thickening walls alone, the wallthickness would be an uneconomical 0.239 in. thick.Besides being impractical to injection mold, the addedthickness would triple the weight of the part. Bycontrast, the rib only adds 25% to the total sectionweight. Table 8.03 shows a summary of the results ofthis example.

Although the use of ribs gives the design engineergreat latitude in efficiently tailoring the structuralresponse of a plastic part, the use of ribs can result inwarping and appearance problems. In general,experienced design engineers do not use ribs if there isdoubt as to whether they are structurally necessary.Adding ribs after the tool is built is usually simple andrelatively inexpensive since it involves removing steel.

Certain basic guidelines should be followed. Themost general is to make the rib thickness at its baseequal to one-half the adjacent wall thickness. Withribs opposite appearance areas, the width should bekept as thin as possible. In areas where structure ismore important than appearance, or when dealingwith very low shrinkage materials, ribs are often 75%,or even 100%, of the outside wall thickness. As seenin Figure 8.08, a goal in rib design is to preventformation of a heavy mass of material that can resultin a sink, void, distortion, long cycle time, or anycombination of these problems.

All ribs should have a minimum of 1/2° of draft perside and should have a minimum radius of 0.005 in, at the base. Generally, the draft and thicknessrequirements limit the height of the rib. Therefore,multiple, evenly spaced ribs are preferred to single,large ribs. Wherever possible, ribs should besmoothly connected to other structural features suchas bosses, side walls, and component mounting pads.Ribs need not be constant in height or width and areoften matched to the stress distribution in the part.

VOID

SINK RIB TOO THICK

POOR

BETTER

2t, MINIMUM

USE MULTIPLE RIBS

CORE THICK RIBFROM BACK

RIB THICKNESS: tr = 1/2 t TYPICAL FOR APPEARANCE PARTS

RIB HEIGHT: L = 1 1/2 t TO 5t

RADIUS: r = 0.005 INCHES, MINIMUM

tr

t

L r

1/2° MIN. DRAFT

Fig 8.08 · Rib design for reinforcing thermoplastic parts

Cross-Section Max. Max.Geometry Area Stress Deflection

(square inches) (psi) (inches)

0.0600 6250 0.694

0.0746 2273 0.026

0.1793 699 0.026

Table 8.03 · Summary of the Effect of Rib andCross-Section change in example, Figure 8.07


62

8.7.2 Other Geometric Reinforcement Besides ribs, there are other acceptable methods ofimproving section properties. Many of these canoften be worked into functional or appearancefeatures of the part. Some typical examples are shown in Figure 8.09.

8.8 Bosses Bosses and other projections from the nominal wallare commonly found in injection-molded plasticparts. These often serve as mounting or fasteningpoints. Figure 8.10 shows some typical boss designsalong with common problems. As with rib design,avoiding overly thick wall sections is important forminimizing the chance of appearance or moldingproblems. When bosses are designed to accommodateself-tapping screws, the inside diameter and wallthickness must be controlled to avoid excessivebuildup of hoop stresses in the boss. Ribs arefrequently used in conjunction with bosses whenlateral forces are expected. Special care must be usedwith tapered pipe threads, since they can create awedging action on the boss. If there is a choice, themale rather than the female pipe thread should be theone molded into the plastic.

8.9 Coring The term coring in injection molding refers to theaddition of steel to the mold for the purpose ofeliminating plastic material in that area. Usually,

coring is necessary to create a pocket or opening inthe part, or simply to reduce an overly heavy wallsection (see Figure 8.11). For simplicity and economyin injection molds, cores should be parallel to the lineof draw of the mold. Cores placed in any otherdirection usually create the need for some type of side action (either a cam or hydraulic cylinder) ormanually loaded and unloaded loose cores.

Blind holes in molded plastic parts are created by acore supported by only one side of the mold. Thelength of the core and depth of the hole are limited bythe ability of the core to withstand the bending forcesproduced by the flowing plastic without excessivedeflection. For this reason, the depth of a blind holeshould not exceed three times its diameter orminimum cross-sectional dimension. For small blindholes with a minimum dimension below 1/2 in., theL/d ratio should be kept to 2. With through holes, thecores can be longer since they are supported by theopposite side of the mold cavity. Sometimes the corescan be split between the two sides and interlockedwhen the mold is closed, allowing long through holesto be created. With through holes, the overall lengthof a given size core generally can be twice as long asthat of a blind hole. Sometimes even longer cores arenecessary. The tool can be designed to balance thehydraulic pressure on the core pin, thus limiting the deflection.

HAT SECTION

METALREINFORCEMENT BI-DIRECTIONAL

CORRUGATION DOMING

CROWNINGCORRUGATION

Fig 8.09 · Modifications to nominal wall to improve structural response


63

8

Fig 8.10 · Design of bosses

CONNECTING BOSSES TOOUTSIDE WALLS WITH RIBS

USE GUSSETS RATHER THANVERY THICK BOSSES WHENRESISTANCE TO LOADING IS REQUIRED

t

1/2 t LOAD

HEAVYSECTION

SINK

VOID

1/2° MIN

1/2 t

1/2° MIN

0.005 MIN

1/2° MIN

1/2° MIN

CORE FROMBELOW –(PARALLELDRAFT)

POOR BETTER BETTER

POOR BETTER BETTER

t

1/2 tt t


64

Fig 8.11 · Coring examples

8.10 Fillets and Radii In the design of injection-molded parts, sharpcorners should always be avoided. Although sharpcorner designs are common with sheet metal andmachined parts, good design practice in anymaterial dictates the use of generous radii to reducestress concentrations. Metal parts often toleratesharp corners. Either the stresses at the corners arevery low compared to the strength of the materialor localized yielding redistributes the load. Neitherof these phenomena is reliable in injection-moldedparts. Sharp corners, particularly inside corners,cause poor flow patterns, reduced mechanicalproperties, and increased tool wear, as well assevere molded-in stresses as the material shrinksonto the corner.

Besides molding problems, sharp corners are oftenthe cause of premature failure in plastic parts dueto the stress concentration. To avoid theseproblems, inside corner radii should be equal tohalf of the nominal wall thickness. A 0.020 in.radius is considered a minimum for parts subjectedto stress, and a 0.005 in. minimum for stress-freeregions of a part. Inside radii less than 0.005 in. arenot recommended for most materials. Outsidecorners should have a radius equal to the insidecorner plus the wall thickness. Figure 8.12dramatically shows how the radius/thickness ratioaffects stress concentration factors.

LOADr

t

r = 1/2 tGOOD DESIGN STANDARD

RATIO, r t

0.5 1.0 1.51.0

1.5

2.0

2.5

3.0

STRE

SS C

ON

CEN

TRA

TIO

N F

AC

TOR,

K

TO FIND STRESS DUE TO SMALL RADIUS,MULTIPLY CALCULATED BENDING STRESS BY K

Fig 8.12 · Typical stress concentration factor

POOR RECTANGULARPART WITH ROUND HOLES

POOR DESIGN

1/2 tt

SUGGESTED ALTERNATIVES

SINK MARKS

WARPAGE

VOID

ttt

MATCH OUTSIDE CONFIGURATIONTO INSIDE CORES

CORES FROMBOTH SIDESIF POSSIBLE

t

1/2 t


65

8

With an inside radius equal to half of the nominalwall, a stress concentration of 1.5 is a reasonableassumption. For radii down to about 10% of thenominal wall, a stress concentration of 3.0 is a goodfirst approximation. Standard tables of stressconcentration factors should always be consulted for critical applications.

8.11 Undercuts Ideally, plastic parts are designed so that the mold canopen easily, and eject the molded part without anypart of the tool needing to open in any directionother than parallel to the movement of the machineplaten. Some complex parts require other mechanicalmovement known as side actions, side coring, cams,pullers, collapsing cores, loose cores, etc., but certaindesign techniques can achieve the desired geometrywith simple tooling. Figure 8.13 shows someexamples of these techniques.

Fig 8.13 · Creating “undercuts” with simple tooling

SIDE WALL OPENING UNDERCUT

CORE & CAVITY KISS-OFFCREATING OPENING IN SIDE WALL

LOCKING TAB UNDERCUT

CAVITY

CORE

USUALLY REQUIRES5° DRAFT MINIMUM

DIRECTION OF MOLD TRAVEL

INJECTION-MOLDEDPLASTIC PART

CAVITY

CORE

CORE EXTENDSTHROUGH OPENINGIN PART

CORE & CAVITY KISS-OFF

LOCKING TAB



HINGED DOOR WITH MOLDED-IN“THRU-HOLE” UNDERCUT

ALTERNATE CORINGFROM BOTH SIDESCREATED THRUHOLE FOR HINGE PIN


66

9. AssemblyOne of the goals in plastic product design is toeliminate as much assembly as possible by combiningcomponents and building as much function aspossible into the design. However, most applicationsfor plastics do require some assembly, and manymethods, which can be broken down into thefollowing general categories, are available to the designer:

9.1 General Types of Assembly Systems

1. Molded-in Assembly SystemsMolded-in assembly systems are generally economicalmethods of assembly, since no additional fastener,adhesive, solvent, or special equipment is required.Examples are snap-fit, press-fit, pop-on, and molded-in threads. Key advantages of these methods are thatassembly is fast and inexpensive, and does not requireadditional parts or substances. They also effectivelyminimize the chances of improper assembly, since theproper fits can be fine tuned into the tooling. The keydisadvantage is that the tooling can be complex andexpensive, and some of the methods may not besuitable for parts that need to be disassembled.

2. Chemical Bonding SystemsChemical bonding systems require no additionalfasteners, but they do require chemical substances,fixtures, and usually safety equipment. Thesemethods use solvents and adhesives to create a joiningof similar and dissimilar materials, and are well suitedto applications where a liquid or gas seal is requiredand where the use of fasteners would create aproblem. Furthermore, these joining methodsgenerally do not create assembly stresses.Disadvantages of such systems are that adhesives orsolvents could be toxic and dangerous, and must beproperly matched to the particular grades of plastic orother materials being joined. In addition, preparationand cure times could be time consuming, dependingon the system used.

3. Thermal Welding MethodsThermal welding methods generally require nochemicals or joining materials, but do requirespecialized equipment. With the right equipment,assembly is fast, economical and generally consideredsafer than chemical methods. The most commonlyused method is ultrasonic welding. Others include

thermal welding methods such as hot plate, spin,vibration, induction, and radio frequency welding. All of these methods depend on the interface or bondline between the two parts melting sufficiently tocreate a weld between the two parts. These thermalmethods are also used for staking, swaging, and withother methods using heat deformation to assemblecomponents. A limitation of these methods is that thematerials being welded must be compatible and havesimilar melting temperatures.

4. Assembly with FastenersMost of the mechanical fasteners used with metals can be successfully used with plastics. In addition, amultitude of fasteners are specifically designed for usewith particular types of thermoplastics. Necessaryequipment ranges from simple hand tools tosophisticated automatic machinery. Typical methodsinclude bolted assemblies, self-tapping screws, rivets,threaded inserts, spring clips, and any othermechanical devices that join parts together. The keyadvantages of these mechanical fasteners is that theyare readily available, easy to use, do not requirecomplex tooling or special preparation, and mostallow for simple nondestructive disassembly.Disadvantages are that extra parts must be stocked,and care must be taken to prevent the assembled partsfrom becoming overstressed. Furthermore, creep canresult in loss of pre-load if poorly designed.

Many plastic assemblies employ a combination ofmethods. The most notable example is ultrasonicinjection used to install a threaded insert that engagesa machine screw. Other examples are the use ofelastomeric adhesives to create an air-tight seal in abolted assembly and solvents used to facilitate orstrengthen a snap-fit or interference fit assembly. This section examines the various methods used withthermoplastics and presents specific guidelines fortheir use.

9.1.1 Molded In Assembly Systems

9.1.1.1 Snap-Fit Assembly For high-volume production, molded-in snap-fitdesigns provide economical and rapid assembly. Inmany products, such as inexpensive housewares orhand-held appliances, snapfits are designed for onlyone assembly without any nondestructive means fordisassembly. Where servicing is anticipated, provisionis made to release the assembly with a tool. Other


67

9

snap-fit designs, such as those used in batterycompartment covers for calculators and radios, are designed for easy release and reassembly overhundreds or even thousands of cycles.

In all snap-fit designs, some portion of the moldedpart must flex like a spring, usually past a designed-ininterference, and quickly return, or nearly return, toits original position to create an assembly betweentwo or more parts. The key to successful snap-fitdesign is having sufficient holding power withoutexceeding the elastic or fatigue limits of the material.

Figure 9.01 shows a typical snap-fit design. Using thebeam equations, we can calculate the maximum stressduring assembly. If we stay below the yield point ofthe material, the flexing finger returns to its originalposition. However, for certain designs there is notenough holding power due to low forces or smalldeflections. With many plastic materials, thecalculated bending stress can far exceed the yieldpoint stress if the assembly occurs rapidly. In otherwords, the flexing finger just momentarily passesthrough its maximum deflection or strain, and thematerial does not respond as if the yield stress hasbeen greatly exceeded. Thus, a common way toevaluate snap-fits is by calculating strain rather than stress. Compare this value with the allowabledynamic strain limit (if available) for the particularmaterial. In designing the finger, it is extremelyimportant to avoid any sharp corners or structuraldiscontinuities, which can increase stress. A taperedfinger provides a uniform stress distribution and isadvisable where possible.

h Lho

L

TAPERED BEAM

Y (MAX DEFLECTION)

DYNAMIC STRAIN

3Yho

2L2Kε =

L

ho

Y (MAX DEFLECTION)

DYNAMIC STRAIN

3Yho

2L2ε =

STRAIGHT BEAM

0.4 0.5 0.60.3 0.7 0.8 0.9 1.01.0

1.1

1.2

1.3

1.4

1.5

1.6

1.7

1.8

1.9

2.0

2.1

2.2

2.3

K

PROPORTIONED CONSTANT, K,FOR TAPERED BEAM

h L/ho

Fig 9.01 · Snap-fit design for cantilever beamwith rectangular cross section


68

Since the snap-fit generally requires an undercut, a mold with side action is frequently required, asshown in Figure 9.02A. Figure 9.02B shows analternative that works well when an opening at thebase of the flexing finger is acceptable. In certaincases, the snap-finger can simply be popped off the mold.

Another type of snap-fit assembly system, which cansometimes be molded into the part, is known as snap-on or snap-in. It is used most often on round parts.Often larger portions of the part or even the entirepart flexes, but the deflections are usually very small.Figure 9.03 shows a typical example of a snap-onassembly.

CORE

CAVITY

PROJECTION FROM CAVITY, THROUGH PART, AND MATINGWITH CORE FORMS UNDERCUT WITHOUT NEED FOR SIDECORE. IT DOES, HOWEVER, LEAVE SMALL "WINDOW" OROPENING IN MOLDED PART.

B) MOLDING SNAP-FIT FLEXING FINGER WITHOUT SIDE CORE

CAVITY

SIDE CORE MOVES DOWN TO MOLD UNDERCUT, AND ISACTIVATED UP TO ALLOW MOLD TO OPEN

A) MOLDING SNAP-FIT FLEXING FINGER WITH MOLD USING SIDE CORE

SIDE CORE

CORE

Fig 9.02 · Tooling for snap-fit fingers Fig 9.03 · Snap-on / Snap-in fits

SNAP-ON FIT

PROLONGED SNAP-IN

PRONGS

FULL PERIMETER SNAP-IN

DIM A

DIM B

BALL OR CYLINDER SNAP-IN

R

R


69

9

9.1.1.2 Molded-in Threads In this type of assembly, mating male and femalethreads are molded into the parts to be assembled(Figure 9.04). Molded internal threads usually requiresome type of unscrewing or collapsing mechanism,which complicates the tooling. Many external threadscan be molded by splitting them across the partingline, as shown. The most common applications formolded threads are in containers and caps, moldedplastic hardware, and liquid handling applications.Molding very fine threads, greater than 28 pitch, isusually not practical.

Fig 9.04 · Molded plastic threads

9.1.1.3 Press-Fits In a press-fit assembly, parts or components of amaterial are assembled to a plastic part, usinginterference fits to maintain the assembly. The mainadvantage of this system is that the tooling is keptrelatively simple, however the method can create veryhigh stresses in the plastic part. The consequences of this stress depend on many factors, such astemperature during and after assembly, modulus ofthe mating material, type of stress, use environment,and probably most important, the type of materialbeing used. Some materials creep or stress relax, while

others fracture or craze if the strain is too high.Except for very light press-fits, this type of assemblycan be very risky due to the hoop stress in the boss,which might already be weakened by a knit-line.Figure 9.05 presents alternative methods of designingpress-fits that have a lower risk of failure.

Fig 9.05 · Alternative press-fit designs formetal pins in plastic hub

INTERNAL THREAD

THREADED CORE PIN IN TOOL MUSTUNSCREW AS MOLD OPENS

EXTERNAL THREAD

SPLIT CAVITY ATPARTING LINE

MOLD CAVITY

MOLD CAVITY

METAL PIN

ALTERNATE PRESS-FITDESIGNS FOR LOWER STRESS

STRAIGHT (INTERFERENCE)PRESS-FIT CAN PRODUCEHIGH STRAINS

INTERFERENCE

PIN DIAMETERSTRAIN =

ADD METAL “HOOP”RING PREVENTINGEXPANSION OFPLASTIC BOSS

USE BARBS OR SPLINES ONTHE METAL PIN TO CREATEINTERFERENCE FIT ANDRETENTION

CREATEINTERFERENCEPRESS-FIT BY ADDING“CRUSH RIBS” TO THEINSIDE DIAMETER OFTHE BOSS


70

9.1.2 Chemical Bonding Systems

9.1.2.1 Solvent Welding This is a fast and economical method for joining likeor similar types of plastics. The principle involved isto apply a substance, usually a liquid solvent, todissolve the surfaces of the joint areas sufficiently toallow the parts to be joined in a true weld after thesolvent evaporates. The key advantages of thismethod are that it is inexpensive and requires little or no part preparation or special equipment.

Of course, the use of solvent bonding is limited tomaterials that are compatible and dissolve in the samesolvent or combination of solvents. The chemicalresistance of many plastic resins, particularlycrystalline resins, limits the applicability of thismethod.

The key limitation of solvent welding, in applicationswhere it is otherwise suitable, is the requirement forprecautions in handling the solvents. Strict regulationsmust be observed regarding worker protection,ventilation, and solvent recovery.

9.1.2.2 Adhesive Bonding This method differs from solvent welding in that athird substance, which has suitable adhesion to bothparts being joined, is introduced at the interface ofboth parts. This third substance, called an adhesive, is capable of joining plastics together with otherplastics, metals, rubber, ceramics, glass, wood, or anyother bondable surface.

Adhesives frequently recommended for use withthermoplastics are epoxies, acrylics, polyurethanes,phenolics, rubbers, polyesters, vinyl, and manyothers, depending on the plastic being used. Becauseof their rapid adhesion to many materials,cyanoacrylates have become very popular in plasticassemblies. Specific adhesive recommendations can beobtained from both the plastic and adhesive suppliers.However, the designer must test the performance ofthe joint in the actual end-use environment.

Many adhesive systems also use solvents that partiallydissolve the plastic surface, thus improving adhesion.It is important to check for compatibility since someadhesives can strongly attack certain plastics and leadto part deterioration and failure.

The key disadvantages of adhesive bonding are that itis generally slow, requires longer clamp times, oftenrequires more fixtures, and sometimes special ovensor curing conditions are necessary. Furthermore,surface preparation can be critical since anycontamination such as grease, oil, mold release, oreven fingerprints can spoil a bond. With many hard tobond materials, the surfaces must be mechanicallyroughened or chemically etched to allow the adhesiveto gain a firm grip.

Figure 9.06 shows typical joint designs used for bothsolvent and adhesive bonding.

SIMPLE LAP

JOGGLE LAP

DOUBLE SCARF LAP

TONGUE-AND-GROOVE

LANDED SCARF TONGUE-AND-GROOVE

BUTT SCARF LAP JOINT

Fig 9.06 · Typical joint designs for solventand adhesive bonding


71

9

9.1.3 Thermal Welding Methods

9.1.3.1 Ultrasonic Welding Ultrasonic welding is an economical method forjoining small- to medium-sized plastics parts to partsof the same or similar plastic material. This techniqueis very rapid and can be fully automated. Weldingoccurs when high-frequency (20-40 kHz) vibrationalenergy is directed to the interface between the twoparts. This creates localized molecular excitation,causing the plastic to melt. Pressure is maintainedbetween the two parts after vibrations stop, and themelted resin immediately solidifies. The entirewelding process normally takes place in less than 2 seconds and has high strength, sometimesapproaching the strength of the base material itself,provided the joint design is correct and equipment isproperly set. Figure 9.07 illustrates typical ultrasonicwelding equipment.

The principle involved in ultrasonic joint design is toconcentrate the energy in an initial small contact area,creating rapid melting and melt flow, which progressesalong the joint as the parts are pressed together. Thebottom of the plastic part is held rigidly in a speciallymade nest. The top part is properly aligned with thebottom, usually by the joint design, but has freedom tocouple acoustically when contacted by the horn, whichtransmits the ultrasonic energy.

Figure 9.08 illustrates a basic shear/interference jointboth before and after welding. Melting starts at thesmall initial contact area, and melt flow continuesalong the vertical wall as the parts telescope together,creating a continuous, leakproof joint whose strengthoften exceeds that of the parts being joined. This jointdesign is preferred with liquid crystal polymers andwith crystalline materials such as nylon, acetal, andthermoplastic polyester and for any material wherehigh strength and a hermetic seal are required.

Before Weld After Weld

45°

0.01

0-0.

025

in.

Fig 9.09 · Basic energy director joint design

The simple butt/energy director design that worksvery well with amorphous materials is shown inFigure 9.09 (also see Fig. 9. 10A). Before joining, a V-shaped projection, known as an energy director,concentrates the ultrasonic energy in a small area,which quickly melts and creates a melt flow as theparts are pressed together.

B

C

A

(A) Ultrasonic assembly stand(B) Horn(C) Work piece area

Fig 9.07 · Typical ultrasonic welding equipment

Fig 9.08 · A basic shear/interference joint

INTERFERENCE0.005-0.012 in.

BEFORE

AFTER


72

Figure 9.10 shows some typical joint designscommonly used with injection-molded thermoplasticparts. Exact dimensions vary with the material used,equipment available, and part requirements. Bothplastic resin suppliers and ultrasonic equipmentmanufacturers offer literature and design assistancewith equipment, fixtures, horns, part configuration,joint design settings, cycle times, and workerprotection. They should be consulted during the partdesign process.

0.005-0.012 in.

FLASH TRAP

0.005-0.012 in.

TONGUE ANDGROOVE SHEARJOINT

0.005-0.012 in.DOUBLE SHEAR JOINT

0.005-0.012 in.

0.001-0.004 in.

Fig 9.10B · Shear joints

Fig 9.10A · Energy director


73

9

With hygroscopic materials, it is best to weld theparts as soon as possible after molding, since theabsorption of moisture can lead to weaker bonds. Insome cases, the parts may require drying just prior towelding. Although ultrasonics provide a nearly idealwelding method, following are some drawbacks.

1 . Design and quality control of the parts along withmaintenance and settings of equipment are critical inensuring consistent high-strength welds.

2. Ultrasonic equipment is expensive and can becomeuneconomical with large parts due to the powerrequired.

3. Parts to be joined must be made from the same orvery similar material, and there are some limitationswith filled and reinforced grades.

It should be noted that the operating frequencies ofultrasonic welding, 20-40 kHz, are above humanhearing. However, sometimes a shrill sound isproduced when a section of the part vibrates at alower frequency. This can result in a need for acousticenclosures or ear protection in the immediate area.There is seldom any sound when welding isperformed at 40 kHz.

Finally, ultrasonic equipment can also be used forstaking, swaging, and spot welding, as illustrated inFigure 9.11.

9.1.3.2 Vibration Welding In vibration welding, two parts are rubbed together,producing frictional heat, which results in a melt atthe interface of the two parts. The movement is in theform of high-amplitude, low-frequency, reciprocatingmotion. When vibration stops, the melt cools and theparts become permanently welded in the properalignment. Typical frequencies are 120 and 240 Hz,and amplitudes range between 0.10 and 0.20 in. oflinear displacement. Vibration welding equipment can also produce angular displacement when thegeometry or assembly of the product prevents linear displacement.

Vibration welding, like ultrasonic welding, produceshigh-strength joints and is much better suited to large parts and irregular joint interfaces. Moisture in hygroscopic materials such as nylon usually has no adverse effect on the weld, as is the case withultrasonics. Figure 9.12 illustrates typical joints usedwith vibration welding.

FLASH TRAP

30° - 60°

Fig 9.10C · Scarf joints

Fig 9.11 · Ultrasonic staking, swaging, and spot welding

FLASH TRAP

Fig 9.12 · Typical joints for vibration welding

SPOT WELDING

SPOT WELD HORN TIP

WELDED AREA

SWAGING

FORMING DIE

ATTACHMENT

PLASTICPART

PLASTICPART

FORMING DIE

ATTACHMENT

STAKING


74

9.1.3.3 Spin Welding Spin welding is a rapid and economical method forjoining parts with circular joint interfaces. As withvibration welding, frictional heat is generated,creating a melt and subsequent weld. While one partis held fixed, the rotating part contacts it with aspecified pressure, generating frictional heating. Afterthe melting occurs, rotation is halted, but pressure ismaintained to create a weld when the interfacesolidifies. The entire process usually takes less than 3 seconds and can be easily automated. Compared toultrasonic and vibration welding, the equipment isvery simple, sometimes just a properly equipped drillpress for low production runs.

9.1.3.4 Radio Frequency (RF) Welding This joining method, often called heat sealing, is widelyused with flexible thermoplastic films and sheets suchas vinyl (plasticized PVC) and polyurethane. However,it can be used with injection-molded parts, most oftenin joining them to flexible films. With this process,welding occurs due to the heat created by theapplication of a strong radio frequency field to theselected joint region. The field is usually applied by aspecially formed metal die in the shape of the desiredjoint, which also applies the clamping pressure neededto complete the weld. However, some plastics aretransparent to radio frequency energy and cannot bewelded by this method.

9.1.3.5 Electromagnetic or Induction WeldingElectromagnetic or induction welding uses theprinciple of inductive heating to generate fusiontemperatures in thermoplastic materials. A radiofrequency (RF) magnetic field is used to excite fine,magnetically sensitive particles, either metallic orceramic, to welding temperatures. The particles can beincorporated into a gasket, preform, filament, ribbon,adhesives, coextruded film, or the molded part.

The most common electromagnetic welding involvesan additional part added to the joint interface, asillustrated in Figure 9.13. The additional part is a“preform” containing the magnetically activeparticles. It is designed to provide a specific volumeof material to the joint during fusion. The preform isplaced at the joint interface and exposed to anelectromagnetic field. Electromagnetically inducedheat is conducted from the particles through thepreform and to the part joint as the parts are pressedtogether. As shown in Figure 9.14, once fusiontemperatures are achieved, the preform material,under pressure, flows through the joint interface andbecomes an integral part of the weld.

A proper joint design is essential to the ultimatesuccess of the weld. Figure 9.15 illustrates five basicjoint designs. Since the preform material located atthe joint interface becomes molten when activated, itflows under pressure into voids and irregular surfacesto produce a reliable weld. Ideally, the molten flowshould be contained and subjected to an internalpressure against the abutting weld surface.

While the addition of the magnetic preform addssome cost to the piece price, this increase is oftenoffset by lower reject rates and gains in weldreliability. Induction welding can produce structural,hermetic welds in most thermoplastic materials and can generally be automated for large volumeproduction applications with greater latitude in jointsize, configuration, tolerance requirements, and theability to bond some dissimilar materials.

However, a significant disadvantage of the method isthat no metal components should be present near theweld line during welding. Therefore, for example, ifelectromagnetic welding is used to seal the housing ofa mechanism, either all internal components must benonmetallic or all metallic components must be in aposition so as not to be subjected to the RF field.

WELD MATERIAL

Fig 9.13 · Preform in place forelectromagnetic welding


75

9

WINDUCTIONCOIL

W

BEFORE WELDING

DURING WELDING

AFTER WELDING

W

Fig 9.14 · The electromagnetic welding process

JOINT BEFORE AFTER

FLATTOFLAT

FLATTOGROOVE

TONGUEANDGROOVE

SHEAR

STEP

Fig 9.15 · Five basic joint designs for induction welding

9.1.4 Assembly with Fasteners

9.1.4.1 Bolted Assembly Standard fasteners, screws, bolts, nuts, lock nuts, lockwashers, and other specially designed products areused frequently with injection-molded plastic parts.Since these threaded fasteners are usually made fromsteel, brass, or other high-strength metal, the plasticparts being assembled can likely be overstressed wellbefore these fasteners break or strip. Therefore, gooddesign practice dictates that certain precautions betaken to prevent excessive stresses when boltedassemblies are used.

The most obvious method for preventing a high-stress assembly is to control carefully the tighteningof all fasteners with properly adjusted torque-limitingdrivers. This approach can work well whenoperations are confined to a factory assembly line,but field or customer assembly presents a difficult


76

control problem. Even when torque can becontrolled, sometimes a poorly designed boltedassembly has an impractically low assembly torque.In this case design modifications are called for.

Figure 9.16 presents common examples of highassembly stress problems that often occur withbolted assembly, along with some practicalsolutions.

LARGE WASHERS SHOULDER WASHERSFLANGE-HEAD SCREWS

SHOULDER SCREW

PLASTIC PART

METALCASTING

STANDARD SCREW

WHERE TORQUE CANNOT BE CONTROLLED, AS IS THE CASE WITHFIELD ASSEMBLY, SHOULDER SCREWS WILL LIMIT COMPRESSION ONPLASTIC PART; OTHER SOLUTIONS ARE:

POTENTIAL HIGH STRESS DUE TO WEDGING ACTION OFSCREW HEAD

ALTERNATIVE RECESSED HEADAVOIDS POTENTIALLY DANGEROUSWEDGING ACTION

POTENTIAL HIGH BENDINGSTRESS BOLT IS TIGHTENED

PLANNED GAP BETWEEN ADDEDBOSSES PREVENTS EXCESSIVEBENDING OF HOUSING AS BOSSESTOUCH AND GO INTO COMPRESSION

POOR DESIGNS PREFERRED DESIGNS

TRUSS OR ROUND HEADSCREW

FLAT-HEAD SCREW

PLASTIC PART

METAL SUB-FRAME

Fig 9.16 · Bolt assembly, stress problems and solutions


77

9

PUSH-IN TYPE INSERTS ADVANTAGES — SPEED AND LOW EQUIPMENT COST

DISADVANTAGES — HIGH INDUCED STRESS AND ONLYFAIR HOLDING POWER

PUSH-IN TYPE INSERTS ADVANTAGES — NO INSTALLATION EQUIPMENT

DISADVANTAGES — LOWER PERFORMANCE AND MODERATELYHIGH INDUCED STRESS

PUSH-IN TYPE INSERTS ADVANTAGES — INSTALLED WITH MINIMAL EQUIPMENT

DISADVANTAGES — SLOW AND THEY CAN CREATEHIGH STRESS

Fig 9.17 · Common threaded metal inserts 9.1.4.2 Threaded Metal Inserts More common than bolted assemblies, threaded metalinserts provide metallic machine threads permanentlyinstalled in the plastic part. These inserts come in awide variety of types and sizes and permit variousinstallation methods. Inserts are typically installed inmolded bosses whose internal diameter is specificallydesigned for the particular insert to be used. Someinserts are simply forced into the boss, while othersare installed by methods that create stronger andlower stress installations. Figure 9.17 illustrates someof these inserts and the relative advantages anddisadvantages of each. They all provide a permanentmetallic machine thread, eliminating the need for anut, allowing assembly with access to only one side of the product. In addition to female threads, insertscan provide threaded male studs, locating pins, andbushings. Follow insert supplier and resinmanufacturer recommendations for suitability, boss dimensions, and installation procedures.

A very popular and preferred type of insert for justabout any thermoplastic material is the ultrasonicallyinstalled type shown in Figure 9.18. This type ofinsert is installed with the same equipment used forultrasonic welding. The resulting installation is strongand relatively free of stress compared to many otherstyles of inserts because the plastic melts around theinsert as it is installed. Installation is fast, economical,and often performed by the operator at the moldingmachine.

ULTRASONICTYPE

INSERTSSTUD TYPE

ULTRASONICINSERT

ADVANTAGES — EXCELLENT PERFORMANCE VERY FAST (OFTEN DONE AT MOLDING MACHINE) VERY LITTLE INDUCED STRESS

DISADVANTAGES— REQUIRES EXPENSIVE EQUIPMENT SOMETIMES NOISY

Fig 9.18 · Ultrasonic inserts designed forinstallation with standard ultrasonic equipment


78

9.1.4.3 Self-Tapping Screws Screws that create their own threads are widely usedwith all thermoplastic parts. They are economical andrequire no extra operations or special unscrewingcores since the thread is cut or formed by the screwduring normal assembly operations. A cutting screwactually removes material, like a thread-cutting tap,and works with most materials. The forming screw, asthe name implies, actually displaces material and mustbe used with caution, since very high hoop stressescan be developed in attempting to form threads inhigh-modulus, low-creep materials. Some screwdesigns create the threads through a combination ofcutting and forming. Figure 9.19 illustrates some of the many varieties of screws used withthermoplastics. Two recent designs that are verypopular due to their excellent holding power andlower levels of induced stress are screws that are notround but have multiple lobes and screws withalternating thread heights. These are also illustrated in Figure 9.19. With all selftapping screws, certainguidelines should be followed:

1. The diameter of the molded hole in the plastic part must be properly sized for the screw design andgrade of plastic material. An undersized hole can lead to excessive hoop stresses in the boss andeventual fracturing.

2. The depth of the molded hole should be sufficientto prevent bottoming of the leading edge of the screw.A planned clearance is usually recommended,especially for thread-cutting screws.

3. The wall section of the boss must be sufficient toresist the stresses created by the screw. As a generalrule, the outside diameter of the boss should be atleast twice the major diameter of the screw.

4. Avoid the use of thread-forming screws on glass-reinforced and other very high modulusthermoplastics.

5. To avoid stripping or high-stress assemblies,torque controlled drivers should be used on assembly lines.

9.1.4.4 Riveted Assembly For fast, permanent assemblies, metal rivets arefrequently used. Hollow aluminum rivets are oftenchosen. The major considerations in using rivets are1) to use rivets with large heads to distribute the load,and 2) to form the rivet against the metal part of the

assembly or against a properly sized metal washer ifboth parts being joined are plastic.

9.1.4.5 Sheet Metal Nuts A wide variety of stamped sheet metal fasteners areavailable to provide light-duty threads or push-ontype of assembly for thermoplastic parts. Boss caps,as illustrated in Figure 9.20, are cup-shaped parts thatcan simply be pushed onto a plastic boss. In additionto providing a partial metal thread, usually for a self-tapping or sheet metal screw, they also tend toreinforce the boss against expansion forces exerted by the screw.

THREAD-CUTTING SCREWS

THREAD-FORMING SCREWS

MULTIPLELOBES

HIGH THREAD

LOW THREAD

ALTERNATE THREAD HEIGHT

Fig 9.19 · Typical screws used with plastics

Fig 9.20 · A typical boss cap


79

9

Figure 9.21 illustrates a typical push-nut, which isusually used in permanent light-duty assemblies. Pushnuts are similar to metal shaft retainers and are simplypushed on a plain, molded plastic stud or boss. Theyare easy to install, inexpensive, vibration-proof, andwork with most plastics.

9.1.4.6 Specialty Plastic Fasteners There are many varieties of molded plastic fasteners,which are satisfactory for many light-duty plasticassemblies. They are especially useful for attachingexternal appearance items such as trim, escutcheons,and faceplates where metal fasteners would beunsatisfactory. In addition to plastic screws and rivets,a wide variety of push-in plastic fasteners is alsoavailable.

PUSH-NUT FASTENER

TOP VIEW

SIDE VIEW

STUD

SHEET-METAL

PUSH-NUTFASTENER

PART

Fig 9.21 · A typical push-nut assembly


80

10. Machining, Finishing,and Decorating

Ideally, injection-molded thermoplastic parts arefinished as molded. For example, almost any type oftexture or surface finish can be molded into the part,as can almost any geometric shape, hole, orprojection. In some situations it is not possible,practical, or economical to have every feature orfinish molded into the part. Typical examples wheremachining may be required are certain undercuts,complicated side coring, or places where parting lineirregularity is unacceptable. Another commonmachining/finishing operation with plastics is theremoval of the remnant of the sprue or gate if it is in an appearance area or critical tolerance region ofthe part.

Many plastic parts are decorated to make themmulticolored, to add distinctive logos, or to allowthem to imitate wood, metal, and other materials.Some plastic parts are painted since their as-moldedappearance is not satisfactory, as may be the case withreinforced, filled, or foamed thermoplastics. Paintingor coating is also used for part protection. Thissection discusses some of the secondary operationsfrequently used with thermoplastic materials. Sinceplastics vary widely in their ability to be machinedand to accept finishes, this discussion is general innature, with details left to other literature dealingwith specific resins.

10.1 Machining All thermoplastic materials can be shaped andfinished with common equipment used for machiningmetals. In addition, many tools specifically used forwoodworking, such as routers, shapers, and sanders,are well suited for thermoplastic materials. Sincemany materials are available in the form of sheets,blocks, slabs, rods, tubes, and other cast and extrudedshapes, initial prototypes are frequently made entirelyby machining.

The main problems encountered when machiningthermoplastic materials are due to the heat built up byfriction. As the resin and cutting tools begin to heatup, the plastic can distort or melt. This can produce apoor surface finish, tearing, localized melting, weldingtogether of stacked parts, and jamming of cutters. It isimportant to prevent the part and cutting tool fromheating up to the point where significant softening or

melting occurs. Some plastic materials machine mucheasier and faster than others due to their physical andmechanical properties. Generally, a high meltingpoint, inherent lubricity, and good hardness andrigidity are factors that improve machinability.

10.1.1 Drilling and Reaming In addition to building prototype parts, drilling and reaming are often required to enlarge, deepen, orremove the draft from a molded hole. In some cases,secondary drilling is a more economical solution thanside cores in a mold. Although specific requirementsvary with material, the following guidelines apply toalmost all thermoplastic resins:

1 . Standard drill presses, as well as other drillingequipment used for metals and wood, are appropriatefor drilling and reaming thermoplastics. Speeds andfeeds must be controlled to avoid heat buildup.

2. Wood and metal drill bits can usually be used, butbest results are obtained with commercially availablebits designed for plastics. These special drills usuallyhave one or two highly polished or chrome-platedflutes, narrow lands, and large helix angles to quicklyexpel chips and minimize frictional heating. For holesin thin sections, circle cutters, which are drills thatonly cut the circumference and eject a round thinplug of material, are often preferred for production.

3. Table 10.01 gives the approximate drilling speedfor thermoplastics. In practice, the drill speed andfeed rate can be increased for maximum productionprovided that there is no melting, burning,discoloration, or poor surface finish For deep drilling,frequent withdrawal of the drill may be necessary for chip ejection.

4. Drill bits and reamers must be kept sharp and cool for good results. For high-volume production,carbide tools are sometimes preferred, especially withglass-reinforced materials. The first choice for coolingis clean, compressed air, since no part contaminationoccurs and chip removal is improved. If a liquidcoolant lubricant is required for deep drilling, wateror some aqueous solution can be used. Metal-cuttingfluids and oils should be avoided since they maydegrade or attack the plastic and create a cleaningproblem.

5. Plastic parts must be firmly held, fixed, or clampedduring drilling and reaming operations to preventdangerous grabbing and spinning of the work.


81

10

10.1.2 Thread Tapping Many plastic parts use self-tapping screws, threadedmetal inserts, molded-in threads, or other fastenersystems. When a machine thread must be added aftermolding, standard metal-cutting taps and dies may beused, provided the same precautions regarding heat,chip removal, tool maintenance, and lubricationdiscussed for drilling are observed. For production of high volumes or with filled resins, carbide taps are recommended. Drilled or molded holes shouldgenerally be larger than those specified for steel, and threads finer than 28 threads per inch should be avoided.

10.1.3 Sawing, Milling, Turning, Grinding, and RoutingThese cutting operations are usually used only formachined prototypes or very low-volume productionof simple shapes. High-speed routing is sometimesused for slotting or gate removal on injection-moldedparts. Standard end mills (two-flute), circular cutters,tool bits, wood saw blades, router bits, files, rasps,and sandpaper can usually be used. As with drilling,tools must be kept sharp and cool, and feeds andspeeds may be increased until overheating, gumming,or poor finish becomes a problem. All machiningoperations should provide for dust control, adequateventilation, safety guards, and eye protection. Inquireabout machining information for each specific resin.

10.2 Finishing and Decorating Since most injection-molded plastic parts areattractive as well as inherently resistant to corrosionand rust, special paints, coatings, and other surfacetreatments are used mainly to enhance eye appeal.Many reinforced and filled resin parts, as well asstructural foam molded parts, emerge from the moldwith an uneven appearance, and paint may benecessary in critical appearance applications.Common decorative finishes applied to plastic partsare spray painting, vacuum metallizing, hot stamping,silk screening, metal plating, printing, and theapplication of self-adhesive labels, decals, and borderstripping. In some cases the finish gives the partadded protection from heat, ultraviolet radiation,chemicals, scratching, or abrasion.

Some conductive coatings are applied to the inside of the part for dissipation of static electricity and/orelectromagnetic shielding. Such coatings are commonin computer and other electronic equipment housings.With all coatings and finishes, a clean surface isessential for a good bond. Care must be used to avoidcontamination. Common sources of contaminationinclude oils, mold releases, the environment, andhandling. In addition to cleaning with solvents anddetergents, some plastics require primers, etching,sanding, or flame treatment to enhance adhesion.Following is a brief description of several widely used processes.

Drill Speed Feeding SpeedMaterial (RPM) (Low, Med., Comments

High)

Polyethylene 1000 – 2000 H Easy tomachine

Polyvinyl 1000 – 2000 M Tends toChloride become

gummy

Acrylic 500 – 1500 M – H Easy todrill withlubricant

Polystyrene 500 – 1500 H Must havecoolant forgood hole

ABS 500 – 1000 M – HPolytetra-

fluoroethylene 1000 L – M Easily drilled

Nylon 6/6 1000 H Easy to drill

Polycarbonate 500 – 1500 M – H Easy to drill,some

gumming

Acetal 1000 – 2000 H Easy to drill

Polypropylene 1000 – 2000 H Easy to drill

Polyester 1000 – 1500 H Easy to drill(PET, PBT)

Table 10.01 · Approximate drilling speed and feedrate for 1/4 – 3/8 in. hole in various thermoplastics


82

10.2.1 Painting Most plastics can be painted, though some are much more difficult than others. Materials such aspolyethylene, polypropylene, and acetal, which havewaxy surfaces, or other crystalline resins that are very resistant to solvents, can be difficult to paint and require special primers or pretreatments forsatisfactory adhesion. Many amorphous plastics easily accept a wide variety of coatings.

Although rolling and dipping are sometimes used,power spray painting is the usual method of paintapplication. Among the coatings used for plastics arepolyurethane-, epoxy-, acrylic-, alkyd-, and vinyl-based paints. Since many paints are oven cured,materials must have sufficient heat resistance tosurvive this treatment without distortion.

10.2.2 Vacuum Metallizing and Sputter PlatingIn these processes, a special base coat is applied to thesurface of the plastic part to be metallized. The part is then placed in a vacuum chamber in which ametallic vapor is created and deposited on the part. A protective clear top coat is then applied over thethin metal layer for abrasion and environmentalresistance. The simplest vacuum metallizing processesuse resistance heating to melt and vaporize the metal.These processes are generally limited to pure metals,typically aluminum, but also silver, copper, and gold.A newer vacuum metallizing process uses an electronbeam to vaporize the metal. The sputter platingprocess uses a plasma to produce the metallic vapor.Both the electron beam and plasma heating methodscan be used satisfactorily with alloys such as brass,and are economical metallizing processes that canproduce attractive high-gloss finish. However, theadhesion is generally low.

10.2.3 Electroplating Many chromed automotive exterior parts areelectroplated injection-molded plastic parts. Afterspecial pretreatments, specific grades of plastics canbe put through electroplating processes similar tothose used in plating metals. Electroplated plasticparts are very durable and provide lightweightreplacements for die castings and sheet metal indemanding applications such as automotive grillesand wheel covers.

10.2.4 Flame Spraying/Arc Spraying In these processes, specialized equipment actuallydeposits a fine spray of molten metal on the plasticsurface. The relatively thick, rough surface isgenerally used in unseen internal surfaces for

electromagnetic and radio frequency shielding, as wellas static electricity dissipation.

10.2.5 Hot Stamping This is a one-step economical process for selectivelytransferring a high-quality image to a plastic part. Aheated die transfers the pattern from selected transfertape to a flat plastic surface. Lettering of decorativedesigns can be transferred in pigmented, woodgrain,or metallic finishes.

10.2.6 Sublimation Printing Sublimation (diffusion) printing is a textile process in which color patterns in dry dye crystals aretransferred from a release film to the fabric underhigh heat and pressure. The process has been adaptedto plastics. The equipment is very similar to that usedfor hot stamping. Under heat and pressure, the dyecrystals sublime (go directly to the vapor phase fromthe solid phase without melting) and the vaporpenetrates the plastic part. As a result, the decorationis very durable and resistant to wear. It is also costcompetitive against other processes such as two-stageinjection molding or silk screening. The process isgenerally limited to polyester and polyester-basedalloys due to the availability of dye technology fromthe textile industry. However, new dyes are underdevelopment and the process is being applied to more plastics.

10.2.7 Printing Lettering and decoration can be applied to mostmaterials by using various printing methods. Offsetprinting, silk screening, and pad printing are amongthe methods adapted to plastics.

10.2.8 Decals and Labels These are usually self-adhesive, precut, printedpatterns on a substrate that simply adhere to thesurface of a part. Decals generally use a transparentplastic film, while labels normally use an opaqueplastic, metallic, multilayer sandwich base. Whensufficiently thick, labels are useful for hidingoccasionally unavoidable appearance problems, suchas gate and sprue removal areas, sink marks, blush,splay, and knit lines.


83

NOTICE TO USERS: This design manual was written to serve as a general purpose reference source for theexperienced plastic product designer as well as the designengineer new to plastics. It should also be of interest tonondesigners and management personnel who need ageneral overview of the concepts and critical issues relatedto the world of plastics. Although the manual is not a guideto injection molding, many of the design considerations arebased upon molding criteria, so those involved in themanufacturing and processing of plastic parts should alsofind it useful.

Most design manuals deal with a specific family of plasticresins, and present properties, design criteria, assembly, andother information related to these resins. The product lineof Ticona includes crystalline, amorphous, liquidcrystalline, and elasto-meric polymers. Due to this diversity,this manual deals with issues common to all injection-moldable thermoplastic resins. Tables, figures, and otherdescriptive methods are used to help illustrate significantdesign differences.

The first part, Chapters 1 through 6, introduces the nature of plastic materials, their properties and testingmethods. Key advantages, as well as limitations, of certain thermoplastics are discussed. Where appropriate, we compare the properties of various engineeringthermoplastics, often including the properties of metals orother structural materials. However, our goal is to reviewfundamental concepts rather than list the various testing methods.

Discussion of specific testing methods from the numerousgovernmental, industrial, and standards organizationsinvolved with testing is beyond the scope of this manual.

The second part, Chapters 7 through 10, deals with theactual design of parts. It starts with structural analysis andinjection molding considerations and concludes withassembly, finishing, and decorating techniques. This part ofthe manual will serve as both a general reference and a“how-to” guide for those new to plastic design.

For further information on design-related topics, the reader is urged to consult the following individual productbrochures: Designing with Celcon® Acetal Copolymer (CE-10), Designing with Fortron® Polyphenylene Sulfide(FN-10), Vectra® Liquid Crystal Polymers Global Brochure(VC-7) and Designing with Celanex, Vandar®, Impet®, & Riteflex® Thermoplastic Polyesters (PE-10). These maybe obtained by calling the Ticona Product InformationServices Department at (800) 833-4882. Additional designinformation on Celcon® Acetal Copolymer and Celanese®

Nylon 6/6.

List of Symbols

A – areaE – modulus of elasticityEapp – apparent or creep modulusEc – modulus of elasticity in compressionEm – modulus of elasticity in metalEp – modulus of elasticity in plasticEt – modulus of elasticity in tensionF – force or loadG – shear modulusH – drop heightI – area moment of inertiaJ – polar moment of inertiaK – stress concentration factor or

proportionality constantL – lengthM – bending momentP – pressureQ – shear forceR – square of the ratio of inside to

outside diameterT – torque, temperatureTg – glass transition temperatureY – deflectionW – weightZ – section modulus, a geometric propertyZI – section modulus, unsymmetrical beamb – dimensionc – distance from axis or neutral surface

to maximum stressd – diameter or dimensiondi – inside diameterdo – outside diameterds – shaft diameterh – dimensionho – height of tapered cantilever beam at

fixed endhL – height of tapered cantilever beam at

free endi – diametrical interferenceia – allowable interferencena – neutral axisr – fillet radiuss – dimensiont – nominal wall thickness, section thicknesstr – rib thicknessyi – impact deflectionys – static deflectionΓ – geometry factor for press-fits∆L – change in length∆Lrel – relative change in length∆b – change in width∆d – change in depth∆T – change in temperatureα – coefficient of linear thermal expansionαm – coefficient of linear thermal expansion

of metalαp – coefficient of linear thermal expansion

of plasticγ – shear strainε – normal strainεc – creep strain

εT – thermal strainθ – angle of twist in torsionµ – coefficient of frictionν – Poisson’s ratioσ – normal or direct stressσa – allowable stressσi – impact stressσs – static stressσT – thermal stressτ – shear stress

Celcon® and Hostaform® acetal copolymer (POM)

GUR® ultra-high molecular weight polyethylene (UHMWPE)

Celanex® thermoplastic polyester

Impet® thermoplastic polyester

Vandar® thermoplastic polyester alloy

Riteflex® thermoplastic polyester elastomer

Vectra® liquid crystal polymer (LCP)

Celstran® long fiber reinforced thermoplastics

Fortron® polyphenylene sulfide (PPS)

Celanese® Nylon 6/6 (PA 6/6)

Topas® cyclic-olefin copolymer (COC)

Encore® recycled thermoplastic molding resins

Products offered by Ticona

Des

igni

ng W

ith

Pla

stic

The

Fun

dam

enta

lsNorth AmericaTicona90 Morris AvenueSummit, NJ 07901

Technical Information1-800-833-4882

Customer Service1-800-526-4960

EuropeTicona GmbHIndustriepark Höchst,Building C65765926 Frankfurt

Technical Information+49(0)69-305-4653

Customer Service+49(0)69-305-31949

Asia PacificPolyplastics Co., Ltd.Kasumigaseki Bldg (6th Fl.)2-5 Kasumigaseki 3-chomeChiyoda-ku, Tokyo, 100Japan

Telephone Number+(81) 3-3593-2411

Fax Number+(81) 3-3593-2455

Visit our Internet Site at http://www.ticona.comfor further information. ©

200

0 Ti

cona

Pr

inte

d in

USA

0

0-33

2/3M

/070

0

TiconaA business of Celanese AG

Documents

Designing With Plastic