Written and maintained by CASRAI Editorial Board
Last updated
Short answer
ASTM D638 is the ASTM test method for the tensile properties of plastics, both unreinforced and reinforced. It tests dumbbell-shaped specimens under defined pretreatment, temperature, humidity and testing-machine speed. It covers material up to 14 mm thick; thin sheeting under 1.0 mm is handled by ASTM D882 instead.
Definition and scope
The full title is Standard Test Method for Tensile Properties of Plastics. ASTM’s store lists the current designation as D638-22 (status Active, last updated 21 July 2022), published in Volume 08.01 by subcommittee D20.10. The scope says the method covers the determination of the tensile properties of unreinforced and reinforced plastics, tested as standard dumbbell-shaped specimens under defined conditions of pretreatment, temperature, humidity and testing-machine speed. A measurement of Poisson’s ratio at room temperature is an optional part of the method.
On thickness, the scope states that the method is applicable to materials of any thickness up to 14 mm (0.55 in.). Thicker material has to be machined down to fit within that limit, and thin sheeting and film below 1.0 mm is routed to ASTM D882, covered below.
The scope and significance statements above are paraphrased from ASTM’s public page. The standard itself is licensed by ASTM, and anyone running the test needs the current edition from ASTM.
How it is used in practice
The significance statement in the standard gives the purpose: the method is designed to produce tensile property data for the control and specification of plastic materials. In practice that makes it a material-qualification and quality-control test. A lab pulls dumbbell-shaped specimens of a given material in a testing machine and records the tensile properties the standard defines, so that two lots, or a supplier’s datasheet and a lab’s own result, can be compared on the same basis.
The standard also states that tensile properties are known to vary with specimen preparation and with speed and environment of testing. That sentence explains why the scope fixes pretreatment, temperature, humidity and speed: a number is only comparable to another number produced under the same conditions.
D638 vs D882: which tensile standard fits the thickness
Two ASTM pages state the handoff between these methods in their own words, so the choice is a matter of specimen thickness rather than of material type.
| Item | ASTM D638 | ASTM D882 |
|---|---|---|
| Title | Tensile Properties of Plastics | Tensile Properties of Thin Plastic Sheeting |
| Thickness covered | Any thickness up to 14 mm (0.55 in.) | Thin sheeting and film less than 1.0 mm (0.04 in.) |
| What the other standard says | Scope names D882 as the preferred method for thin sheeting under 1.0 mm | Note 2 says plastics 1.0 mm (0.04 in.) or greater in thickness are tested under D638 |
Both standards give 1.0 mm as the boundary. ASTM’s D882 page also notes that D882 is similar to ISO 527-3 but is not considered technically equivalent, so a D882 result should not be treated as an ISO 527-3 result without checking.
The D882 page ASTM shows for the 2018 edition is marked Historical, with a later edition (D882-26A) listed as Active, so check which edition applies before citing it.
Where it matters for labs and materials decisions
Because the standard ties its results to preparation, speed and environment, a tensile result is only as useful as the record that goes with it. Drawing on the factors the standard itself names, a lab report for a D638 test should state:
- Edition: the designation and year of the standard used (for example D638-22).
- Material and thickness: the material tested and its thickness, which decides whether D638 or D882 applies, and whether thicker stock was machined down.
- Specimen preparation: how the dumbbell specimens were produced, since the standard says properties vary with it.
- Pretreatment, temperature and humidity: the conditioning and test environment the scope lists as defined conditions.
- Testing-machine speed: the speed used, which is the third factor the standard names as affecting results.
- Whether Poisson’s ratio was measured: the optional room-temperature measurement.
Two results that differ on any of these lines may not be directly comparable. The same discipline applies to the instrument itself: a testing machine’s force and speed performance depends on calibration, which is covered in what is calibration and how to set a calibration interval, and a calibration provider’s paperwork is interpreted in ISO 17025 calibration certificates. Writing the test down as a controlled procedure is the subject of the standard operating procedure (SOP) entry.
Common misunderstandings
“D638 gives design data”
The standard describes itself as producing data for the control and specification of plastic materials. It also states that data from the test cannot be considered valid for applications involving load-time scales or environments widely different from those of the test. A D638 value is a standardized comparison point, not a guarantee of how a molded part behaves in service.
“D638 covers all plastic forms”
It covers unreinforced and reinforced plastics up to 14 mm, but not thin film or sheeting below 1.0 mm, which belongs to D882.
“Any dumbbell will do”
The scope calls for standard dumbbell-shaped specimens. The specimen dimensions are set in the licensed standard text, which this page does not reproduce; take them from the current edition.
“Two labs reporting D638 results are comparable”
Only if preparation, conditioning, humidity, temperature and speed match, which the standard names as factors that change the result.
Related reading
For the wider field these tests sit in, see what is polymer science and what is materials science.
Sources
- ASTM International, ASTM D638-22, Standard Test Method for Tensile Properties of Plastics (store page, scope and significance text), checked 29 October 2026.
- ASTM International, ASTM D882-18, Standard Test Method for Tensile Properties of Thin Plastic Sheeting (store page, scope and Notes 2 and 4), checked 29 October 2026.
