Effect of Different Tensile Testing Speeds on Strength: Testing Guide, Standards, and Methods

The tensile rate applied during a strength test has a measurable effect on tensile strength, yield strength, and elongation results, especially for rate-sensitive materials such as plastics, rubber, and thin films. This technical guide explains the mechanisms behind strain-rate sensitivity, the applicable ASTM and ISO standards, specimen requirements, test procedures, and practical applications of multi-speed tensile testing.

Technical Overview

In tensile testing, the "testing speed" refers either to the crosshead separation rate of the universal testing machine, expressed in mm/min, or to the strain rate applied to the specimen gauge section, expressed in s⁻¹. Because most engineering materials are strain-rate sensitive to some degree, the measured strength values change systematically when the rate changes, which is why international standards treat test speed as a mandatory controlled and reported parameter.

Polymeric materials are viscoelastic, meaning their molecular chains need a finite relaxation time to untangle and slip past one another. When the tensile rate is increased, the chains have less time to relax, so the material responds in a stiffer and stronger manner. For many thermoplastics, raising the speed from 5 mm/min to 500 mm/min can increase the measured tensile strength by roughly 15–30% while sharply reducing elongation at break, and in some cases shifting the failure mode from ductile to brittle.

Metallic materials show a smaller but real effect governed by the thermally activated motion of dislocations. At higher strain rates, dislocations have less time to overcome obstacles with the help of thermal energy, so the flow stress rises. At quasi-static rates typical of laboratory machines (10⁻⁴ to 10⁻² s⁻¹), the strain-rate sensitivity exponent m for common steels and aluminum alloys is usually only 0.005–0.05, but the effect becomes significant at dynamic rates encountered in impact and crash loading.

Consequently, comparing tensile data obtained at different speeds without correction can lead to false conclusions about material quality. ISO 6892-1 explicitly requires the test speed and the method of speed control to be stated in the test report, and strain-rate control (Method A) is increasingly preferred over simple crosshead speed control because it improves interlaboratory reproducibility.

Testing Samples

Metallic Materials

Metallic specimens are machined as proportional dog-bone bars or flat coupons in accordance with ASTM E8/E8M or ISO 6892-1, with typical round specimens of 10–12.5 mm diameter and 50 mm gauge length, or rectangular specimens of 12.5–20 mm width. The gauge section must be free of burrs, tool marks, and residual stresses that could initiate premature fracture, and the cross-sectional area S0 is measured to calculate stress.

Plastics and Polymers

Rigid and semi-rigid plastics are tested using ISO 527-2 type 1A/1B dumbbell specimens or ASTM D638 type I specimens, with a preferred thickness of approximately 3.2 mm for the ASTM geometry. Specimens are preferably injection molded in a matched die; when machined, surfaces must be smooth and free of notches. Reinforced and filled compounds require careful specimen orientation relative to the fiber direction.

Rubbers and Elastomers

Elastomers are die-cut from flat sheets of 1–3 mm thickness into standard dumbbells such as ASTM D412 type C or ISO 37 type 2, using a sharp, well-maintained die to avoid micro-tears at the edges that would lower the measured tensile strength and produce scatter.

Films and Composites

Thin plastic films below 0.25 mm are tested as straight strips per ASTM D882, while fiber-reinforced composite laminates use rectangular coupons with bonded tabs per ASTM D3039. All specimen classes should be conditioned at 23 ± 2 °C and 50 ± 10% relative humidity (at least 40 hours for plastics per ASTM D618) before testing.

Testing Items

A multi-speed tensile testing program typically evaluates the following properties at each selected rate:

• Ultimate tensile strength (UTS)
• Yield strength, including upper and lower yield points for mild steels
• 0.2% offset proof strength (Rp0.2)
• Tensile modulus (Young's modulus, chord modulus, or 1% secant modulus)
• Elongation at yield and elongation at break
• Uniform elongation and total fracture strain
• Strain hardening exponent (n-value) at a defined rate
• Strain-rate sensitivity exponent (m-value)
• Energy absorbed to break (area under the stress–strain curve)
• Complete engineering stress–strain curves for each speed

Testing Methods

Equipment and Calibration

Testing is performed on a screw-driven or servo-hydraulic universal testing machine calibrated to ISO 7500-1 or ASTM E4, with the load cell selected so that measured forces fall within its verified working range. For metals, an extensometer of ISO 9513 class 1 or ASTM E83 class B-1 accuracy is attached directly to the gauge length, because crosshead displacement alone cannot resolve the small strains needed for accurate modulus and proof strength. For rubber, films, and high-elongation plastics, a long-travel or non-contact video extensometer is recommended. Wedge grips suit metals, while pneumatic side-action grips prevent slippage and jaw breaks in polymers and films.

Standard Test Speed Selection

• ISO 6892-1 (metals, Method A): nominal strain rates of 0.00007 s⁻¹, 0.00025 s⁻¹, and 0.0025 s⁻¹ with ±20% tolerance for yield properties, and rates up to 0.008 s⁻¹ for tensile strength determination.
• ASTM E8/E8M (metals): stress rate limited to approximately 12 MPa/s for yield determination, with strain-rate or crosshead control within stated limits and the speed reported.
• ISO 527-1 (plastics): preferred crosshead speeds of 1, 2, 5, 10, 20, 50, 100, 200, and 500 mm/min, with modulus measured at 1 mm/min between strains of 0.0005 and 0.0025.
• ASTM D638 (plastics): speeds from 1 to 500 mm/min, most commonly 5 mm/min for stiffness and 50 mm/min for strength.
• ASTM D412 / ISO 37 (rubber): 500 mm/min as the standard speed unless otherwise agreed.
• ASTM D882 (thin films): grip separation rate selected from the standard table so that rupture occurs within the prescribed time window.

Test Procedure

• Condition the specimens and measure the width, thickness, or diameter at the gauge section to compute the initial cross-sectional area.
• Select a rate series covering at least one decade, with a minimum of three speeds (for example 1, 10, and 100 mm/min) to reveal the trend.
• Mount the specimen in aligned grips and apply a small preload to remove slack without plastically straining the material.
• Attach the extensometer, start data acquisition, and pull the specimen to complete fracture without changing the speed during the post-yield stage unless the standard directs otherwise.
• Test at least five specimens per speed for plastics and rubbers, or a minimum of three to five for metals in comparative studies.
• Convert force–extension data to engineering stress–strain, extract all properties, and plot strength values against the logarithm of strain rate.
• Calculate the strain-rate sensitivity exponent as m = Δln(σ) / Δln(ε̇) from the slope of the log-log plot, where σ is flow or yield stress and ε̇ is strain rate.

Data Interpretation and Reporting

For polymers, yield stress frequently increases approximately linearly with the logarithm of strain rate over moderate ranges, consistent with Eyring-type activated flow, while ductility decreases; time–temperature superposition can be used to extend the measured trend. For metals, the quasi-static rate effect is small, and rates above roughly 10 s⁻¹ require specialized dynamic techniques such as split Hopkinson pressure bar testing. The report must state the control mode (crosshead speed, strain rate, or stress rate), all speeds used, mean values, standard deviations, and coefficients of variation for each rate, because ISO 6892-1 results differ between Method A, B, and C control schemes.

Applications

• Generating rate-dependent material cards for crashworthiness and drop-test simulation in LS-DYNA, Abaqus, and similar CAE platforms.
• Calibrating constitutive models such as Johnson–Cook, Cowper–Symonds, and Eyring equations from measured multi-rate data.
• Qualifying incoming materials against supplier datasheets, which are referenced to specific standard test speeds.
• Evaluating packaging films and seals for performance at high converting line speeds rather than only at slow laboratory rates.
• Supporting failure analysis by reproducing the actual loading rate experienced in service.
• Providing objective, speed-matched data for supplier comparison, material selection, and new formulation screening.

FAQ

Why does tensile strength increase when the test speed increases?

In viscoelastic polymers, molecular chains cannot relax and reorient quickly enough at high rates, so the material deforms through stiffer chain stretching and localized yielding, which raises the measured strength and lowers elongation at break. In metals, higher rates reduce the contribution of thermal activation to dislocation motion, raising the flow stress.

Is the speed effect the same for metals and plastics?

No. Plastics and rubbers are far more rate-sensitive and can show strength gains of 15–30% or more across a two-decade speed change, whereas common structural metals show only a few percent change at quasi-static rates, with pronounced effects appearing only at dynamic rates.

What is the difference between crosshead speed and strain rate?

Crosshead speed is the commanded displacement rate of the machine and is affected by load-frame compliance, grip slippage, and specimen geometry. Strain rate is the local deformation rate of the gauge section measured by an extensometer and is the physically meaningful quantity, which is why ISO 6892-1 Method A controls strain rate directly.

Which test speed should be selected for comparative testing?

Use the speed specified by the governing product or material standard, and test all comparison groups at the identical speed with the same control mode. The speed must always be stated in the test report so results can be legitimately compared.

How many specimens are required for each test speed?

Standards such as ASTM D638 and ISO 527 require a minimum of five specimens per condition, and five or more per rate is also recommended for metals when statistical comparison across rates is the objective.

Can results obtained at different speeds be compared directly?

Direct comparison is not valid without qualification, because rate-induced differences reflect material behavior rather than quality differences. If conversion is necessary, it must rely on validated rate-sensitivity models or on dedicated tests performed at the same standard speed.