ASTM D638 Testing Tailored to Your Project Needs

ASTM D638 applies principally to plastics. ASTM D412 applies to vulcanized thermoset rubbers and thermoplastic elastomers and uses procedures and specimen forms suited to their characteristically large extensions.

ASTM D412 provides dumbbell, straight-section, and cut-ring specimen and procedure options. The different D412 methods do not necessarily produce identical results, so the required specimen and procedure should be established before testing.

Where the material is a rubber or rubber-like thermoplastic elastomer rather than a conventional plastic, ASTM D412 will generally provide the more appropriate framework.

ASTM D638 is intended for plastics; ASTM E8/E8M covers tension testing of metallic materials at room temperature.

ASTM E8/E8M addresses properties including yield strength, tensile strength, yield-point elongation, elongation, and reduction of area. It also uses metallic-material specimen geometries, gripping practices, extensometry requirements, and reporting conventions that differ substantially from those of ASTM D638.

A metallic material or component should therefore normally be tested under ASTM E8/E8M or an applicable metal product specification rather than ASTM D638.

ASTM D1708 uses a smaller microtensile specimen and may be appropriate where an applicable specification, customer requirement, or historical dataset specifically calls for that method.

Where limited material availability is the only reason for considering a smaller specimen, ASTM D638 Type V will often provide the more direct standards-based option while retaining the D638 method and reporting framework.

ASTM D1708 does not provide tensile modulus under the standard. DatapointLabs can perform D1708 by technical review, but it is not presently promoted as a routine catalog service.

ASTM D638 and ISO 527 address the same general subject – tensile properties of plastics – but they are not technically interchangeable.

Within the ISO system:

• ISO 527-1 establishes the general principles.
• ISO 527-2 provides test conditions for molding and extrusion plastics.
• ISO 527-3 addresses films and sheets.
• ISO 527-4 addresses isotropic and orthotropic fiber-reinforced plastic composites.
• ISO 527-5 addresses unidirectional fiber-reinforced plastic composites.

The governing product specification, customer requirement, supply-chain convention, or existing comparison dataset should determine whether ASTM D638 or the applicable ISO 527 part is used.

DatapointLabs performs ISO 527-2 tensile testing for molding and extrusion plastics in accordance with the general principles of ISO 527-1. ISO 527-4 and ISO 527-5 testing is also available, although those two parts are not presently within the accredited scope. ISO 527-3 thin-film testing is not offered.

ASTM D638 applies to plastics up to 14 mm thick, but ASTM D882 is generally preferred for thin plastic sheeting and film below 1 mm.

Very thin specimens present different gripping, dimensional-measurement, alignment, and strain-measurement considerations from standard D638 dumbbell specimens. The methods should therefore not be treated as interchangeable merely because both measure tensile properties of plastics.

DatapointLabs does not currently offer ASTM D882 or ISO 527-3 testing. Where the submitted material is a thin film or sheeting, the applicable external film-testing standard should be identified rather than routing the work to ASTM D638 solely for convenience.

ASTM D638 covers reinforced plastics generally, but it directs resin-matrix composites reinforced with oriented high-modulus fibers to ASTM D3039/D3039M.

ASTM D3039/D3039M determines the in-plane tensile properties of continuous- or discontinuous-fiber polymer-matrix composites whose laminates are balanced and symmetric with respect to the test direction. Composite coupon geometry, layup, tabs, fiber orientation, gripping, conditioning, strain measurement, and acceptable failure mode require a different test framework from the molded dumbbell specimens used under ASTM D638.

ASTM D5766/D5766M is more specialized. It determines the open-hole tensile strength of balanced, symmetric, multidirectional polymer-matrix composite laminates. The test is performed using the ASTM D3039/D3039M tensile framework, but with a centrally located hole that creates the required stress concentration.

The practical distinction is therefore:

• Use ASTM D638 for applicable unreinforced and reinforced plastics.
• Use ASTM D3039/D3039M for the unnotched tensile properties of oriented high-modulus-fiber composite laminates.
• Use ASTM D5766/D5766M when the engineering requirement is specifically open-hole tensile strength.

ASTM D638 specimens may be prepared by injection molding or by machining or die cutting from suitable sheet, plate, slab, plaque, or similar material forms.

DatapointLabs routinely molds ASTM D638 specimens from supplied pellets. Specimens can also be machined or cut from suitable plaques, sheet, plate, or other stock. Where a required molding operation cannot be completed internally, specimen molding can be subcontracted.

The selected preparation method should be documented because injection-molded specimens may not produce the same tensile properties as specimens machined or die-cut from stock. ASTM D638 specifically notes that orientation induced during injection molding can affect the result and that the effect may be more pronounced in specimens with narrow sections.

Materials thicker than 14 mm must be reduced by machining for testing under ASTM D638. For material between 14 and 51 mm thick, approximately equal amounts are removed from both surfaces. For still thicker material, both surfaces are machined and the specimen’s location through the original thickness should be recorded.

The preparation method can materially affect the result. Molding conditions, flow-induced orientation, residual stress, machining or cutting direction, moisture history, and specimen thickness can all influence the measured tensile properties.

Where specimens are taken from molded plaques or finished parts, specimen location may also matter. Local molding features – such as weld or knit lines formed where separate melt fronts meet – may affect the observed behavior and should be considered when selecting and documenting the specimen location.

A comparison among formulations, suppliers, molding conditions, or manufacturing processes is most meaningful when the specimens are prepared by a consistent method. Conversely, specimen location, orientation, or processing history may be varied deliberately where their effects are the subject of the study.

Plastics may exhibit direction-dependent tensile behavior because of molecular orientation, molding flow, extrusion direction, reinforcement, or the way specimens are removed from a plaque or finished part.

Where anisotropy is suspected, ASTM D638 requires duplicate specimen sets with their long axes prepared parallel and normal to the suspected principal direction. When applicable, at least five specimens are tested in each direction.

For reinforced plastics tested under ASTM D638, the cutting direction relative to the reinforcement should be controlled carefully. Cutting certain composite laminates on a bias to the reinforcement can weaken the specimen and produce lower measured properties unless off-axis behavior is intentionally being investigated.

Orientation should therefore be established before specimen preparation and identified in the test report. Terms such as machine direction, transverse direction, flow direction, longitudinal direction, or other customer-defined axes should be used consistently throughout the testing program.

Specimen surfaces should be free from visible flaws, scratches, and preparation damage. Coarse machining marks should be removed without creating undercuts or altering the required dimensions, and final sanding strokes should run parallel to the specimen’s long axis. Flash should be removed carefully from molded specimens without disturbing the molded surfaces.

Gauge marks must not be scratched, punched, or impressed into the specimen because those markings can create local damage. The standard permits non-damaging marking methods such as wax crayon or India ink where gauge marks are required.

Specimens that break at a flaw or outside the narrow test section are ordinarily discarded and replaced unless the flaw or failure location is itself a variable being studied.

DatapointLabs performs ASTM D638 tensile testing from −54°C to 200°C, subject to specimen, fixture, extensometry, and test requirements.

The appropriate temperature or series of temperatures depends on the intended use of the results. A room-temperature test may be sufficient for specification compliance or an initial material comparison. Product-development, design, or simulation programs may require tensile behavior at several temperatures representative of manufacturing, storage, or service conditions.

Results obtained at one temperature should not automatically be assumed to describe behavior at another. ASTM D638 itself cautions that many plastics are highly sensitive to environmental conditions and that substantially different application environments may require broader testing.

Crosshead speed and specimen strain rate are not the same quantity.

Crosshead speed describes the motion of the testing machine. The strain rate developed within the specimen also depends on factors such as:

• Specimen geometry and gauge length.
• Grip separation.
• Machine and fixture compliance.
• Specimen deformation outside the measured gauge region.
• Material stiffness and nonlinear response.
• Localization or necking.

ASTM D638 expressly recognizes that wide differences may exist between crosshead speed and the rate of strain measured between the specimen’s gauge marks. The standard speed should therefore be reported as the prescribed machine rate, not represented as a precisely controlled material strain rate unless strain is measured and the test is specifically designed for that purpose.

A single ASTM D638 test speed provides tensile behavior under one defined loading condition. Where the material’s rate dependence matters, testing may be performed at several speeds or through a dedicated rate-dependent characterization program.

DatapointLabs offers tensile testing across multiple strain rates, including high- and very-high-rate characterization using video extensometry. These enhanced testing options may use D638-type material or specimen geometries as a starting point, but they extend beyond routine ASTM D638 execution and should not be represented as properties or procedures established by the standard itself.

Multiple-rate testing may be appropriate for:

• Comparing quasi-static and faster material response.
• Impact- or crash-related product development.
• Evaluating rate-sensitive yield, strength, and failure behavior.
• Calibrating rate-dependent CAE material models.
• Determining whether a standard-speed D638 curve adequately represents the intended application.

Tensile modulus is calculated from relatively small strains near the beginning of the stress–strain curve. Accurate measurement in this region requires substantially greater strain resolution than measurements made later in the test.

For modulus determination, ASTM D638 requires an extensometer with a maximum strain error of 0.0002 mm/mm. A system meeting the Class B-2 requirements of ASTM E83 over the range used for the modulus measurement satisfies this requirement.

Machine crosshead displacement should not be substituted for the required gauge-length measurement when reporting ASTM D638 tensile modulus.

For elongation at yield and other measurements at nominal extensions of approximately 20% or less, the measurement system must meet at least the applicable Class C requirements of ASTM E83.

The modulus extensometer may also be used where its range and configuration permit the required extension to be measured without damage to the instrument or interruption of the result.

For elongations greater than approximately 20%, ASTM D638 permits broader-range measurement techniques having an error no greater than 10% of the measured value.

The appropriate technique depends on the material. A rigid plastic that fails at low strain presents a different measurement problem from a ductile polymer that yields, necks, and continues to deform through large extensions.

Where deformation becomes strongly localized, conventional gauge-length elongation may no longer describe the local material response adequately. ASTM D638 addresses this partly through the distinction between elongation and nominal strain, while enhanced characterization may be needed where localized or true stress–strain behavior is the engineering objective.

Depending on the test requirement and expected material behavior, DatapointLabs can use:

• Contact extensometry for accurate small-strain measurements.
• Noncontact video extensometry for broader deformation ranges and high-rate testing.
• Measurement systems configured for full engineering stress–strain curves.
• Transverse and axial strain measurement for Poisson’s ratio.
• 2D digital image correlation for full-field surface strain and localized deformation.
• 3D digital image correlation for true stress–strain characterization through localized deformation and failure.

The measurement approach should be established before testing because the system appropriate for modulus may differ from the system needed to follow large extension, localized yielding, or high-speed deformation.

ASTM D638 includes the option of determining Poisson’s ratio at room temperature. This requires synchronized measurement of longitudinal and transverse strain over a strain interval in which the material response is sufficiently proportional.

Where Poisson’s ratio is requested, the report should include the average value, standard deviation, and whether proportionality was observed within the strain range used.

Tensile modulus describes the material’s initial tensile stiffness. It is determined from the slope of the initial linear portion of the stress–strain curve.

Because the calculation uses very small strains, modulus is especially sensitive to extensometer accuracy, alignment, specimen seating, and correction of any initial toe region caused by slack or settling in the test system.

A higher tensile modulus indicates greater resistance to tensile deformation under the particular test conditions. It does not by itself indicate greater strength, toughness, or resistance to failure.

Some plastics do not exhibit a sufficiently clear proportional or linear region for a conventional tensile-modulus calculation.

In such cases, a secant modulus may be reported at a designated strain. It is the ratio of nominal stress to the corresponding strain at that specified point. The strain or stress used must accompany the reported value because secant modulus is not a single intrinsic number independent of the selected point.

Poisson’s ratio relates transverse contraction to longitudinal extension during tensile loading.

ASTM D638 includes an optional room-temperature procedure using synchronized longitudinal and transverse strain measurements. The reported result includes the average, standard deviation, and a statement indicating whether proportionality was observed over the strain range used.

Poisson’s ratio may be important for engineering calculations and CAE material models, but it requires a measurement approach specifically configured to capture both axial and transverse deformation.

The yield point is the first point on the stress–strain curve at which strain increases without a corresponding increase in stress. Only materials exhibiting the required curve behavior can properly be described as having a yield point under the ASTM definition.

Yield stress describes the nominal tensile stress at that point. Yield strain or elongation at yield describes the corresponding deformation.

Not every plastic produces a distinct yield point. A change in slope or a discontinuity in the elastic region should not automatically be labeled as yielding.

Tensile strength is the maximum nominal tensile stress sustained during the test.

Where the maximum occurs at the yield point, the result is designated tensile strength at yield. Where the maximum occurs at specimen rupture, it is designated tensile strength at break.

Tensile strength should not be confused automatically with stress at break. For a material that yields and then undergoes necking or strain softening, the maximum engineering stress may occur well before rupture.

Tensile stress is calculated from the applied tensile load divided by the specimen’s minimum original cross-sectional area within the gauge region.

Stress at yield and stress at break refer to the calculated nominal stress at those particular events. Either may differ from the maximum tensile stress sustained during the test.

Because nominal or engineering stress continues to use the original area, it may cease to represent the local material stress accurately after substantial necking or reduction of cross-sectional area.

Percent elongation is the change in gauge length divided by the original gauge length, expressed as a percentage.

Elongation at yield is measured at the yield point. Elongation at break is measured at specimen rupture.

ASTM D638 states that elongation results are quantitatively relevant where deformation remains sufficiently uniform within the gauge length.

Where substantial necking or other nonuniform deformation occurs, a gauge-length elongation value may no longer describe the specimen response adequately.

ASTM D638 then calls for nominal strain at break, calculated from the change in grip separation relative to the original grip separation. The standard characterizes nominal strain under these conditions as having qualitative rather than quantitative engineering utility.

This distinction is important: elongation at break and nominal strain at break are not interchangeable measurements derived from the same reference length.

An engineering stress–strain curve plots nominal tensile stress against corresponding tensile strain.

Engineering stress is based on the specimen’s original cross-sectional area, while engineering strain is based on the change in gauge length relative to the original gauge length. The curve can show initial stiffness, yield behavior, strain hardening or softening, maximum engineering stress, elongation, and failure behavior.

A full curve generally provides more engineering information than a small set of summary properties, but its usefulness still depends on specimen preparation, test conditions, measurement quality, and similarity to the intended application.

A true stress–strain curve accounts for changes in specimen geometry as deformation proceeds rather than continuing to divide force by the original cross-sectional area.

This distinction becomes particularly important after yielding and localized necking, when the specimen’s instantaneous cross-sectional area may differ substantially from its original area. ASTM D638 defines true stress and true strain terminology, but routine D638 testing does not by itself establish the enhanced optical measurement and post-processing approach needed to characterize localized true stress–strain behavior through failure.

For enhanced characterization, DatapointLabs uses 3D digital image correlation to track localized deformation. Synchronized force and optical deformation data are post-processed to obtain true stress–strain behavior through localized yielding and subsequent failure.

A meaningful comparison begins with complete identification of the material and its condition before testing. Relevant information may include:

• Resin grade, formulation, reinforcement, and supplier.
• Lot, batch, or manufacturer code.
• Colorant, filler, additives, or recycled content.
• Molding or processing history.
• Previous thermal, mechanical, moisture, or environmental exposure.
• Age and storage conditions where relevant.

Two specimens identified only by a broad polymer family may not represent equivalent materials or processing states.

Measured properties may be influenced by:

• Molding, machining, or die-cutting method.
• Specimen Type I, II, III, IV, or V.
• Width and thickness.
• Surface finish and edge quality.
• Specimen location within a plaque or component.
• Flow or reinforcement direction.
• Residual stress.
• Weld or knit lines.
• Damage introduced during specimen preparation.

ASTM D638 notes that variations in specimen thickness can affect the surface-to-volume ratio and therefore influence the result. Where directly comparable data are required, specimens should be of equal thickness whenever possible.

Results obtained from injection-molded specimens should not automatically be treated as equivalent to results from specimens machined from sheet, plate, or a molded plaque.

Molecular orientation, extrusion direction, molding flow, fiber alignment, and laminate construction can produce direction-dependent tensile properties.

Comparisons should therefore use consistent specimen orientation. Where anisotropy is being evaluated, the principal directions should be defined clearly and tested separately.

Terms such as machine direction, transverse direction, flow direction, longitudinal direction, and customer-defined product axes should be used consistently in the specimen plan, data files, and report.

Moisture content, temperature, humidity, and previous exposure can materially affect the tensile response of plastics.

Comparable testing ordinarily requires consistency in:

• Conditioning procedure.
• Time between conditioning and testing.
• Test-room temperature and humidity.
• Specimen moisture state.
• Test temperature.
• Prior thermal, fluid, humidity, or chemical exposure.

Room-temperature results and nonambient results describe different material conditions and should not be combined into a single comparison without accounting for the test environment.

Tensile properties may vary with the prescribed crosshead speed and with the strain rate actually developed in the specimen.

Comparable programs should use the same:

• ASTM D638 speed selection.
• Specimen geometry and gauge length.
• Grip separation.
• Measurement approach.
• Temperature and conditioning state.

Crosshead speed should not be treated as though it guarantees an identical specimen strain rate across different geometries, machines, fixtures, or material responses.

Misalignment, specimen slippage, grip damage, machine compliance, and unsuitable extensometry can alter the measured curve and reported properties.

Comparisons should use compatible:

• Specimen alignment and gripping practices.
• Force-measurement range.
• Extensometer classification and gauge length.
• Strain-measurement method.
• Data-acquisition and calculation procedures.
• Toe-compensation treatment where applicable.

A modulus calculated from accurate gauge-length extensometry should not be compared indiscriminately with one derived from machine crosshead displacement.

A reported average is meaningful only in relation to the specimen count, scatter, failure behavior, and treatment of invalid specimens.

The report should identify:

• Number of specimens tested.
• Individual results.
• Average values and standard deviations.
• Failure locations and observations.
• Specimens excluded or replaced.
• Applicable material or customer specification.
• Revision of ASTM D638 used.

The practical principle is:

Comparable results require comparable material states, specimens, test conditions, measurement methods, and reporting conventions.

ASTM D638 data can support:

• Verification against a material or product specification.
• Supplier or incoming-material qualification.
• OEM or customer approval requirements.
• Confirmation of minimum tensile properties.
• Comparison with established acceptance criteria.
• Documentation of conditioned or environmentally exposed properties.

Where a governing material specification modifies the general ASTM D638 procedure, the specification requirements should be identified before testing because they may take precedence.

The method can be used to monitor:

• Lot-to-lot or batch-to-batch consistency.
• Changes in formulation or reinforcement.
• Effects of processing conditions.
• Material degradation or contamination.
• Differences among manufacturing sites or suppliers.
• Changes after aging or environmental exposure.

For these uses, consistent specimen preparation and test conditions are often as important as the individual property values.

ASTM D638 testing can help compare:

• Alternative material grades or formulations.
• Virgin and recycled-content materials.
• Molding or processing conditions.
• Specimen orientations.
• Conditioned and unconditioned states.
• Room-temperature and nonambient behavior.

The comparison should be designed so that unintended differences in specimen geometry, preparation, conditioning, speed, or measurement method do not obscure the variable being studied.

Full stress–strain curves and individual specimen data can provide more insight than summary values alone.

ASTM D638 testing may help engineers investigate:

• Initial tensile stiffness.
• Onset of yielding.
• Maximum engineering stress.
• Ductility and failure behavior.
• Effects of formulation or reinforcement.
• Process-induced orientation.
• Environmental or temperature sensitivity.
• Unexpected production or field-performance differences.

These data can support material screening and development decisions, but the test conditions should remain relevant to the question being investigated.

ASTM D638 tensile modulus, Poisson’s ratio, yield behavior, strength, and stress–strain data may support selected engineering calculations where:

• The material condition is representative.
• The loading rate is reasonably comparable.
• The temperature and environment are appropriate.
• The deformation and loading mode fall within the useful range of the measurement.
• The limitations of engineering stress and strain are understood.

ASTM D638 cautions that plastics can be highly sensitive to strain rate, temperature, environment, and previous history. Results obtained under one set of conditions cannot automatically be assumed valid for applications with widely different load-time scales or environments.

A well-executed ASTM D638 test can provide an initial material-screening result and quasi-static tensile baseline, help select additional temperatures or strain rates, and serve as one part of a broader product-development or CAE material-characterization dataset. Its suitability depends on the intended next engineering step rather than on the standard designation alone.

Additional testing may be needed where the application involves:

• Subambient or elevated temperatures.
• Temperature transitions.
• Moisture or humidity effects.
• Chemical or fluid exposure.
• Thermal aging.
• Weathering or other long-term environmental exposure.

A room-temperature ASTM D638 curve may not represent the material under manufacturing, storage, or service conditions.

Polymers frequently exhibit rate-dependent stiffness, yield, strength, elongation, and failure behavior.

Testing across multiple strain rates may be appropriate for:

• Rapid loading.
• Appliance or consumer-product drops.
• Automotive impact and crash.
• Protective structures.
• Dynamic product loading.
• Rate-dependent CAE material models.

DatapointLabs provides rate-dependent tensile characterization from quasi-static through high-strain-rate and very-high-rate regimes using dedicated high-speed testing and optical video extensometry.

Learn more: High Strain Rate Testing

Engineering stress and engineering strain are based on original specimen dimensions and a defined gauge length. They become less representative of the local material response after substantial necking or localization.

Enhanced measurement may be appropriate where the objective includes:

• Local yielding.
• Neck development.
• Nonuniform deformation.
• Post-yield behavior.
• True stress–strain through failure.
• Spatial strain distributions.
• Validation of localized material-model response.

DatapointLabs can use 2D DIC to measure full-field surface deformation and 3D DIC to obtain true stress–strain behavior through localized yielding and subsequent failure.

A single tensile direction may be insufficient for materials affected by:

• Molding flow.
• Fiber reinforcement.
• Additive manufacturing.
• Sheet or plaque orientation.
• Anisotropic processing history.

Additional tensile directions – or other loading modes – may be required to characterize the directional material response.

ASTM D638 is a relatively short-duration tensile test. It does not characterize long-term deformation under sustained load.

Creep or stress-relaxation testing may be required where the application involves:

• Sustained stress.
• Sustained deformation.
• Long service times.
• Dimensional stability.
• Retention of load or clamping force.
• Time-dependent material models.

A monotonic tensile curve does not establish behavior under repeated or reversible loading.

Additional cyclic characterization may be needed to investigate:

• Stiffness change.
• Permanent set.
• Loading–unloading response.
• Progressive damage.
• Energy dissipation.
• Stabilization after repeated cycles.

A uniaxial tensile test does not fully characterize behavior in:

• Biaxial loading.
• Volumetric deformation.
• Complex multiaxial stress states.
• Notched or open-hole configurations.

The appropriate additional modes depend on the material model, product geometry, failure mechanism, and engineering objective.

Routine tensile strength and elongation may not be sufficient where the objective is to predict failure under complex stress states.

Broader programs may require:

• Multiple specimen geometries.
• Different stress triaxialities.
• Notched and unnotched testing.
• Multiple loading modes.
• High-rate failure data.
• Full-field strain measurement.
• Physical validation against a component or representative structure.

An ASTM D638 engineering stress–strain curve can be an important input for CAE or FEA, but it does not necessarily provide every parameter required by the selected material model.

The necessary characterization depends on:

• The solver and material-model formulation.
• Elastic, plastic, viscoelastic, or damage behavior represented.
• Expected temperature and strain-rate range.
• Material orientation.
• Loading modes.
• Unloading or cyclic response.
• Large-strain behavior.
• Failure criteria.

A simple model for small-strain, quasi-static behavior may use modulus, Poisson’s ratio, yield, and an appropriate tensile curve. More advanced models may require multiple temperatures, strain rates, directions, loading modes, time-dependent response, true stress–strain data, or failure characterization.

DatapointLabs can help determine whether routine ASTM D638 data are sufficient or whether the engineering objective calls for a broader characterization scope. TestPaks can include material testing selected for a particular CAE material model, conversion of raw data into model parameters, solver-formatted material files, and optional material-card validation.

The key distinction is:

ASTM D638 defines a valuable tensile test. It does not, by itself, define a complete material-characterization program for every product-development, design, or simulation objective.