Designation: D 638 – 98 An American National Standard

AMERICAN SOCIETY FOR TESTING AND MATERIALS

100 Barr Harbor Dr., West Conshohocken, PA 19428
Reprinted from the Annual Book of ASTM Standards. Copyright ASTM

Standard Test Method for

Tensile Properties of Plastics1
This standard is issued under the fixed designation D 638; the number immediately following the designation indicates the year of
original adoption or, in the case of revision, the year of last revision. A number in parentheses indicates the year of last reapproval. A
superscript epsilon (e) indicates an editorial change since the last revision or reapproval.

This standard has been approved for use by agencies of the Department of Defense.

1. Scope * safety concerns, if any, associated with its use. It is the

1.1 This test method covers the determination of the tensile responsibility of the user of this standard to establish appro-
properties of unreinforced and reinforced plastics in the form priate safety and health practices and determine the applica-
of standard dumbbell-shaped test specimens when tested under bility of regulatory limitations prior to use.
defined conditions of pretreatment, temperature, humidity, and 2. Referenced Documents
testing machine speed.
1.2 This test method can be used for testing materials of any 2.1 ASTM Standards:
thickness up to 14 mm (0.55 in.). However, for testing D 229 Test Methods for Rigid Sheet and Plate Materials
specimens in the form of thin sheeting, including film less than Used for Electrical Insulation2
1.0 mm (0.04 in.) in thickness, Test Methods D 882 is the D 412 Test Methods for Vulcanized Rubber and Thermo-
preferred test method. Materials with a thickness greater than plastic Rubbers and Thermoplastic Elastomers— Tension3
14 mm (0.55 in.) must be reduced by machining. D 618 Practice for Conditioning Plastics and Electrical
1.3 This test method includes the option of determining Insulating Materials for Testing4
Poisson’s ratio at room temperature. D 651 Test Method for Tensile Strength of Molded Electri-
cal Insulating Materials5
NOTE 1—This test method and ISO 527-1 are technically equivalent. D 882 Test Methods for Tensile Properties of Thin Plastic
NOTE 2—This test method is not intended to cover precise physical
Sheeting4
procedures. It is recognized that the constant rate of crosshead movement
type of test leaves much to be desired from a theoretical standpoint, that D 883 Terminology Relating to Plastics4
wide differences may exist between rate of crosshead movement and rate D 1822 Test Method for Tensile-Impact Energy to Break
of strain between gage marks on the specimen, and that the testing speeds Plastics and Electrical Insulating Materials4
specified disguise important effects characteristic of materials in the D 3039/D 3039M Test Method for Tensile Properties of
plastic state. Further, it is realized that variations in the thicknesses of test Polymer Matrix Composite Materials6
specimens, which are permitted by these procedures, produce variations in D 4000 Classification System for Specifying Plastic Mate-
the surface-volume ratios of such specimens, and that these variations may
rials7
influence the test results. Hence, where directly comparable results are
desired, all samples should be of equal thickness. Special additional tests D 4066 Specification for Nylon Injection and Extrusion
should be used where more precise physical data are needed. Materials7
NOTE 3—This test method may be used for testing phenolic molded D 5947 Test Methods for Physical Dimensions of Solid
resin or laminated materials. However, where these materials are used as Plastic Specimens8
electrical insulation, such materials should be tested in accordance with E 4 Practices for Force Verification of Testing Machines9
Test Methods D 229 and Test Method D 651. E 83 Practice for Verification and Classification of Exten-
NOTE 4—For tensile properties of resin-matrix composites reinforced
with oriented continuous or discontinuous high modulus >20-GPa
someters9
(>3.0 3 106-psi) fibers, tests shall be made in accordance with Test E 132 Test Method for Poisson’s Ratio at Room Tempera-
Method D 3039/D 3039M. ture9
1.4 Test data obtained by this test method are relevant and E 691 Practice for Conducting an Interlaboratory Study to
Determine the Precision of a Test Method10
appropriate for use in engineering design.
2.2 ISO Standard:
1.5 The values stated in SI units are to be regarded as the
standard. The values given in parentheses are for information
only. 2
Annual Book of ASTM Standards, Vol 10.01.
3
1.6 This standard does not purport to address all of the Annual Book of ASTM Standards, Vol 09.01.
4
Annual Book of ASTM Standards, Vol 08.01.
5
Discontinued; see 1994 Annual Book of ASTM Standards, Vol 10.01.
6
Annual Book of ASTM Standards, Vol 15.03.
1 7
This test method is under the jurisdiction of ASTM Committee D-20 on Plastics Annual Book of ASTM Standards, Vol 08.02.
8
and is the direct responsibility of Subcommittee D 20.10 on Mechanical Properties. Annual Book of ASTM Standards, Vol 08.03.
9
Current edition approved September 10, 1998. Published March 1999. Originally Annual Book of ASTM Standards, Vol 03.01.
10
published as D 638 – 41 T. Last previous edition D 638 – 97. Annual Book of ASTM Standards, Vol 14.02.

*A Summary of Changes section appears at the end of this standard.

1
 D 638
ISO 527-1 Determination of Tensile Properties11 4.4 Poisson’s Ratio—When uniaxial tensile force is applied
to a solid, the solid stretches in the direction of the applied
3. Terminology force (axially), but it also contracts in both dimensions lateral
3.1 Definitions—Definitions of terms applying to this test to the applied force. If the solid is homogeneous and isotropic,
method appear in Terminology D 883 and Annex A2. and the material remains elastic under the action of the applied
force, the lateral strain bears a constant relationship to the axial
4. Significance and Use strain. This constant, called Poisson’s ratio, is defined as the
4.1 This test method is designed to produce tensile property negative ratio of the transverse (negative) to axial strain under
data for the control and specification of plastic materials. These uniaxial stress.
data are also useful for qualitative characterization and for 4.4.1 Poisson’s ratio is used for the design of structures in
research and development. For many materials, there may be a which all dimensional changes resulting from the application
specification that requires the use of this test method, but with of force need to be taken into account and in the application of
some procedural modifications that take precedence when the generalized theory of elasticity to structural analysis.
adhering to the specification. Therefore, it is advisable to refer NOTE 6—The accuracy of the determination of Poisson’s ratio is
to that material specification before using this test method. usually limited by the accuracy of the transverse strain measurements
Table 1 in Classification D 4000 lists the ASTM materials because the percentage errors in these measurements are usually greater
standards that currently exist. than in the axial strain measurements. Since a ratio rather than an absolute
4.2 Tensile properties may vary with specimen preparation quantity is measured, it is only necessary to know accurately the relative
and with speed and environment of testing. Consequently, value of the calibration factors of the extensometers. Also, in general, the
value of the applied loads need not be known accurately.
where precise comparative results are desired, these factors
must be carefully controlled. 5. Apparatus
4.2.1 It is realized that a material cannot be tested without
also testing the method of preparation of that material. Hence, 5.1 Testing Machine—A testing machine of the constant-
when comparative tests of materials per se are desired, the rate-of-crosshead-movement type and comprising essentially
greatest care must be exercised to ensure that all samples are the following:
prepared in exactly the same way, unless the test is to include 5.1.1 Fixed Member—A fixed or essentially stationary
the effects of sample preparation. Similarly, for referee pur- member carrying one grip.
poses or comparisons within any given series of specimens, 5.1.2 Movable Member—A movable member carrying a
care must be taken to secure the maximum degree of unifor- second grip.
mity in details of preparation, treatment, and handling. 5.1.3 Grips—Grips for holding the test specimen between
4.3 Tensile properties may provide useful data for plastics the fixed member and the movable member of the testing
engineering design purposes. However, because of the high machine can be either the fixed or self-aligning type.
degree of sensitivity exhibited by many plastics to rate of 5.1.3.1 Fixed grips are rigidly attached to the fixed and
straining and environmental conditions, data obtained by this movable members of the testing machine. When this type of
test method cannot be considered valid for applications involv- grip is used extreme care should be taken to ensure that the test
ing load-time scales or environments widely different from specimen is inserted and clamped so that the long axis of the
those of this test method. In cases of such dissimilarity, no test specimen coincides with the direction of pull through the
reliable estimation of the limit of usefulness can be made for center line of the grip assembly.
most plastics. This sensitivity to rate of straining and environ- 5.1.3.2 Self-aligning grips are attached to the fixed and
ment necessitates testing over a broad load-time scale (includ- movable members of the testing machine in such a manner that
ing impact and creep) and range of environmental conditions if they will move freely into alignment as soon as any load is
tensile properties are to suffice for engineering design pur- applied so that the long axis of the test specimen will coincide
poses. with the direction of the applied pull through the center line of
the grip assembly. The specimens should be aligned as per-
NOTE 5—Since the existence of a true elastic limit in plastics (as in fectly as possible with the direction of pull so that no rotary
many other organic materials and in many metals) is debatable, the
motion that may induce slippage will occur in the grips; there
propriety of applying the term “elastic modulus” in its quoted, generally
accepted definition to describe the “stiffness” or “rigidity” of a plastic has is a limit to the amount of misalignment self-aligning grips will
been seriously questioned. The exact stress-strain characteristics of plastic accommodate.
materials are highly dependent on such factors as rate of application of 5.1.3.3 The test specimen shall be held in such a way that
stress, temperature, previous history of specimen, etc. However, stress- slippage relative to the grips is prevented insofar as possible.
strain curves for plastics, determined as described in this test method, Grip surfaces that are deeply scored or serrated with a pattern
almost always show a linear region at low stresses, and a straight line similar to those of a coarse single-cut file, serrations about 2.4
drawn tangent to this portion of the curve permits calculation of an elastic
modulus of the usually defined type. Such a constant is useful if its
mm (0.09 in.) apart and about 1.6 mm (0.06 in.) deep, have
arbitrary nature and dependence on time, temperature, and similar factors been found satisfactory for most thermoplastics. Finer serra-
are realized. tions have been found to be more satisfactory for harder
plastics, such as the thermosetting materials. The serrations
should be kept clean and sharp. Breaking in the grips may
11
Available from American National Standards Institute, 11 W. 42nd St., 13th occur at times, even when deep serrations or abraded specimen
Floor, New York, NY 10036. surfaces are used; other techniques must be used in these cases.

2
 D 638
Other techniques that have been found useful, particularly with meet at least Class C (Practice E 83) requirements, which
smooth-faced grips, are abrading that portion of the surface of include a fixed strain error of 0.001 strain or 61.0 % of the
the specimen that will be in the grips, and interposing thin indicated strain, whichever is greater.
pieces of abrasive cloth, abrasive paper, or plastic, or rubber- 5.2.3 High-Extension Measurements—For making mea-
coated fabric, commonly called hospital sheeting, between the surements at elongations greater than 20 %, measuring tech-
specimen and the grip surface. No. 80 double-sided abrasive niques with error no greater than 610 % of the measured value
paper has been found effective in many cases. An open-mesh are acceptable.
fabric, in which the threads are coated with abrasive, has also 5.2.4 Poisson’s Ratio—Bi-axial extensometer or axial and
been effective. Reducing the cross-sectional area of the speci- transverse extensometers capable of recording axial strain and
men may also be effective. The use of special types of grips is transverse strain simultaneously. The extensometers shall be
sometimes necessary to eliminate slippage and breakage in the capable of measuring the change in strains with an accuracy of
grips. 1 % of the relevant value or better.
5.1.4 Drive Mechanism—A drive mechanism for imparting
NOTE 8—Strain gages can be used as an alternative method to measure
to the movable member a uniform, controlled velocity with axial and transverse strain; however, proper techniques for mounting
respect to the stationary member, with this velocity to be strain gages are crucial to obtaining accurate data. Consult strain gage
regulated as specified in Section 8. suppliers for instruction and training in these special techniques.
5.1.5 Load Indicator—A suitable load-indicating mecha- 5.3 Micrometers—Suitable micrometers for measuring the
nism capable of showing the total tensile load carried by the width and thickness of the test specimen to an incremental
test specimen when held by the grips. This mechanism shall be discrimination of at least 0.025 mm (0.001 in.) should be used.
essentially free of inertia lag at the specified rate of testing and All width and thickness measurements of rigid and semirigid
shall indicate the load with an accuracy of 61 % of the plastics may be measured with a hand micrometer with ratchet.
indicated value, or better. The accuracy of the testing machine A suitable instrument for measuring the thickness of nonrigid
shall be verified in accordance with Practices E 4. test specimens shall have: (1) a contact measuring pressure of
NOTE 7—Experience has shown that many testing machines now in use 25 6 2.5 kPa (3.6 6 0.36 psi), (2) a movable circular contact
are incapable of maintaining accuracy for as long as the periods between foot 6.35 6 0.025 mm (0.250 6 0.001 in.) in diameter, and (3)
inspection recommended in Practices E 4. Hence, it is recommended that a lower fixed anvil large enough to extend beyond the contact
each machine be studied individually and verified as often as may be foot in all directions and being parallel to the contact foot
found necessary. It frequently will be necessary to perform this function within 0.005 mm (0.0002 in.) over the entire foot area. Flatness
daily.
of the foot and anvil shall conform to Test Method D 5947.
5.1.6 The fixed member, movable member, drive mecha- 5.3.1 An optional instrument equipped with a circular con-
nism, and grips shall be constructed of such materials and in tact foot 15.88 6 0.08 mm (0.625 6 0.003 in.) in diameter is
such proportions that the total elastic longitudinal strain of the recommended for thickness measuring of process samples or
system constituted by these parts does not exceed 1 % of the larger specimens at least 15.88 mm in minimum width.
total longitudinal strain between the two gage marks on the test
specimen at any time during the test and at any load up to the 6. Test Specimens
rated capacity of the machine. 6.1 Sheet, Plate, and Molded Plastics:
5.2 Extension Indicator (extensometer)—A suitable instru- 6.1.1 Rigid and Semirigid Plastics—The test specimen shall
ment shall be used for determining the distance between two conform to the dimensions shown in Fig. 1. The Type I
designated points within the gage length of the test specimen as specimen is the preferred specimen and shall be used where
the specimen is stretched. For referee purposes, the extensom- sufficient material having a thickness of 7 mm (0.28 in.) or less
eter must be set at the full gage length of the specimen, as is available. The Type II specimen may be used when a
shown in Fig. 1. It is desirable, but not essential, that this material does not break in the narrow section with the preferred
instrument automatically record this distance, or any change in Type I specimen. The Type V specimen shall be used where
it, as a function of the load on the test specimen or of the only limited material having a thickness of 4 mm (0.16 in.) or
elapsed time from the start of the test, or both. If only the latter less is available for evaluation, or where a large number of
is obtained, load-time data must also be taken. This instrument specimens are to be exposed in a limited space (thermal and
shall be essentially free of inertia at the specified speed of environmental stability tests, etc.). The Type IV specimen
testing. Extensometers shall be classified and their calibration should be used when direct comparisons are required between
periodically verified in accordance with Practice E 83. materials in different rigidity cases (that is, nonrigid and
5.2.1 Modulus-of-Elasticity Measurements—For modulus- semirigid). The Type III specimen must be used for all
of-elasticity measurements, an extensometer with a maximum materials with a thickness of greater than 7 mm (0.28 in.) but
strain error of 0.0002 mm/mm (in./in.) that automatically and not more than 14 mm (0.55 in.).
continuously records shall be used. A Class B-2 extensometer 6.1.2 Nonrigid Plastics—The test specimen shall conform
(Practice E 83) meets this requirement. to the dimensions shown in Fig. 1. The Type IV specimen shall
5.2.2 Low-Extension Measurements—For elongation-at- be used for testing nonrigid plastics with a thickness of 4 mm
yield and low-extension measurements (nominally 20 % or (0.16 in.) or less. The Type III specimen must be used for all
less), the same above extensometer, attenuated to 20 % exten- materials with a thickness greater than 7 mm (0.28 in.) but not
sion, may be used. In any case, the extensometer system must more than 14 mm (0.55 in.).

3
 D 638

Specimen Dimensions for Thickness, T, mm (in.)A

7 (0.28) or under Over 7 to 14 (0.28 to 0.55), incl 4 (0.16) or under
Dimensions (see drawings) Tolerances
Type I Type II Type III Type IVB Type VC,D
W—Width of narrow sectionE,F 13 (0.50) 6 (0.25) 19 (0.75) 6 (0.25) 3.18 (0.125) 60.5 (60.02)B,C
L—Length of narrow section 57 (2.25) 57 (2.25) 57 (2.25) 33 (1.30) 9.53 (0.375) 60.5 (60.02)C
WO—Width overall, minG 19 (0.75) 19 (0.75) 29 (1.13) 19 (0.75) ... + 6.4 ( + 0.25)
WO—Width overall, minG ... ... ... ... 9.53 (0.375) + 3.18 ( + 0.125)
LO—Length overall, minH 165 (6.5) 183 (7.2) 246 (9.7) 115 (4.5) 63.5 (2.5) no max (no max)
G—Gage lengthI 50 (2.00) 50 (2.00) 50 (2.00) ... 7.62 (0.300) 60.25 (60.010)C
G—Gage lengthI ... ... ... 25 (1.00) ... 60.13 (60.005)
D—Distance between grips 115 (4.5) 135 (5.3) 115 (4.5) 65 (2.5)J 25.4 (1.0) 65 (60.2)
R—Radius of fillet 76 (3.00) 76 (3.00) 76 (3.00) 14 (0.56) 12.7 (0.5) 61 (60.04)C
RO—Outer radius (Type IV) ... ... ... 25 (1.00) ... 61 (60.04)
A
Thickness, T, shall be 3.26 0.4 mm (0.13 6 0.02 in.) for all types of molded specimens, and for other Types I and II specimens where possible. If specimens are
machined from sheets or plates, thickness, T, may be the thickness of the sheet or plate provided this does not exceed the range stated for the intended specimen type.
For sheets of nominal thickness greater than 14 mm (0.55 in.) the specimens shall be machined to 14 6 0.4 mm (0.55 6 0.02 in.) in thickness, for use with the Type III
specimen. For sheets of nominal thickness between 14 and 51 mm (0.55 and 2 in.) approximately equal amounts shall be machined from each surface. For thicker sheets
both surfaces of the specimen shall be machined, and the location of the specimen with reference to the original thickness of the sheet shall be noted. Tolerances on
thickness less than 14 mm (0.55 in.) shall be those standard for the grade of material tested.
B
For the Type IV specimen, the internal width of the narrow section of the die shall be 6.00 6 0.05 mm (0.2506 0.002 in.). The dimensions are essentially those of Die
C in Test Methods D 412.
C
The Type V specimen shall be machined or die cut to the dimensions shown, or molded in a mold whose cavity has these dimensions. The dimensions shall be:
W 5 3.18 6 0.03 mm (0.125 6 0.001 in.),
L 5 9.53 6 0.08 mm (0.375 6 0.003 in.),
G 5 7.62 6 0.02 mm (0.300 6 0.001 in.), and
R 5 12.7 6 0.08 mm (0.500 6 0.003 in.).
The other tolerances are those in the table.
D
Supporting data on the introduction of the L specimen of Test Method D 1822 as the Type V specimen are available from ASTM Headquarters. Request RR:D20-1038.
E
The width at the center Wc shall be +0.00 mm, −0.10 mm ( +0.000 in., −0.004 in.) compared with width W at other parts of the reduced section. Any reduction in W
at the center shall be gradual, equally on each side so that no abrupt changes in dimension result.
F
For molded specimens, a draft of not over 0.13 mm (0.005 in.) may be allowed for either Type I or II specimens 3.2 mm (0.13 in.) in thickness, and this should be taken
into account when calculating width of the specimen. Thus a typical section of a molded Type I specimen, having the maximum allowable draft, could be as follows:
G
Overall widths greater than the minimum indicated may be desirable for some materials in order to avoid breaking in the grips.
H
Overall lengths greater than the minimum indicated may be desirable either to avoid breaking in the grips or to satisfy special test requirements.
I
Test marks or initial extensometer span.
J
When self-tightening grips are used, for highly extensible polymers, the distance between grips will depend upon the types of grips used and may not be critical if
maintained uniform once chosen.

FIG. 1 Tension Test Specimens for Sheet, Plate, and Molded Plastics

6.1.3 Reinforced Composites—The test specimen for rein- machining operations, or die cutting, from materials in sheet,
forced composites, including highly orthotropic laminates, plate, slab, or similar form. Materials thicker than 14 mm (0.55
shall conform to the dimensions of the Type I specimen shown in.) must be machined to 14 mm (0.55 in.) for use as Type III
in Fig. 1. specimens. Specimens can also be prepared by molding the
6.1.4 Preparation—Test specimens shall be prepared by material to be tested.

4
 D 638
will be significantly weakened by cutting on a bias, resulting in lower
laminate properties, unless testing of specimens in a direction other than
parallel with the reinforcement constitutes a variable being studied.
NOTE 11—Specimens prepared by injection molding may have different
tensile properties than specimens prepared by machining or die-cutting
because of the orientation induced. This effect may be more pronounced
in specimens with narrow sections.
6.2 Rigid Tubes—The test specimen for rigid tubes shall be
as shown in Fig. 2. The length, L, shall be as shown in the table
in Fig. 2. A groove shall be machined around the outside of the
specimen at the center of its length so that the wall section after
machining shall be 60 % of the original nominal wall thick-
ness. This groove shall consist of a straight section 57.2 mm
(2.25 in.) in length with a radius of 76 mm (3 in.) at each end
joining it to the outside diameter. Steel or brass plugs having
diameters such that they will fit snugly inside the tube and
having a length equal to the full jaw length plus 25 mm (1 in.)
shall be placed in the ends of the specimens to prevent
crushing. They can be located conveniently in the tube by
separating and supporting them on a threaded metal rod.
Details of plugs and test assembly are shown in Fig. 2.
6.3 Rigid Rods—The test specimen for rigid rods shall be as
shown in Fig. 3. The length, L, shall be as shown in the table
in Fig. 3. A groove shall be machined around the specimen at
the center of its length so that the diameter of the machined
portion shall be 60 % of the original nominal diameter. This
groove shall consist of a straight section 57.2 mm (2.25 in.) in
length with a radius of 76 mm (3 in.) at each end joining it to
the outside diameter.
6.4 All surfaces of the specimen shall be free of visible
flaws, scratches, or imperfections. Marks left by coarse ma-
chining operations shall be carefully removed with a fine file or
DIMENSIONS OF TUBE SPECIMENS abrasive, and the filed surfaces shall then be smoothed with
Standard Length, L,
abrasive paper (No. 00 or finer). The finishing sanding strokes
Length of Radial Total Calculated shall be made in a direction parallel to the long axis of the test
Nominal Wall of Specimen to Be
Sections, Minimum
Thickness
2R.S. Length of Specimen
Used for 89-mm specimen. All flash shall be removed from a molded specimen,
(3.5-in.) JawsA
taking great care not to disturb the molded surfaces. In
mm (in.) machining a specimen, undercuts that would exceed the
0.79 (1⁄32) 13.9 (0.547) 350 (13.80) 381 (15) dimensional tolerances shown in Fig. 1 shall be scrupulously
1.2 (3⁄64) 17.0 (0.670) 354 (13.92) 381 (15)
1.6 (1⁄16) 19.6 (0.773) 356 (14.02) 381 (15)
avoided. Care shall also be taken to avoid other common
2.4 (3⁄32) 24.0 (0.946) 361 (14.20) 381 (15) machining errors.
3.2 (1⁄8) 27.7 (1.091) 364 (14.34) 381 (15) 6.5 If it is necessary to place gage marks on the specimen,
4.8 ( ⁄16)
3 33.9 (1.333) 370 (14.58) 381 (15)
6.4 (1⁄4) 39.0 (1.536) 376 (14.79) 400 (15.75) this shall be done with a wax crayon or India ink that will not
7.9 (5⁄16) 43.5 (1.714) 380 (14.96) 400 (15.75) affect the material being tested. Gage marks shall not be
9.5 (3⁄8) 47.6 (1.873) 384 (15.12) 400 (15.75) scratched, punched, or impressed on the specimen.
11.1 (7⁄16) 51.3 (2.019) 388 (15.27) 400 (15.75)
12.7 (1⁄2) 54.7 (2.154) 391 (15.40) 419 (16.5) 6.6 When testing materials that are suspected of anisotropy,
A
For other jaws greater than 89 mm (3.5 in.), the standard length shall be duplicate sets of test specimens shall be prepared, having their
increased by twice the length of the jaws minus 178 mm (7 in.). The standard
length permits a slippage of approximately 6.4 to 12.7 mm (0.25 to 0.50 in.) in each
long axes respectively parallel with, and normal to, the
jaw while maintaining the maximum length of the jaw grip. suspected direction of anisotropy.
FIG. 2 Diagram Showing Location of Tube Tension Test
Specimens in Testing Machine 7. Number of Test Specimens
7.1 Test at least five specimens for each sample in the case
NOTE 9—Test results have shown that for some materials such as glass of isotropic materials.
cloth, SMC, and BMC laminates, other specimen types should be 7.2 Test ten specimens, five normal to, and five parallel
considered to ensure breakage within the gage length of the specimen, as
with, the principle axis of anisotropy, for each sample in the
mandated by 7.3.
NOTE 10—When preparing specimens from certain composite lami-
case of anisotropic materials.
nates such as woven roving, or glass cloth, care must be exercised in 7.3 Discard specimens that break at some obvious fortuitous
cutting the specimens parallel to the reinforcement. The reinforcement flaw, or that do not break between the predetermined gage

5
 D 638
the grips or test fixtures during the test. The rate of motion of
the driven grip or fixture when the testing machine is running
idle may be used, if it can be shown that the resulting speed of
testing is within the limits of variation allowed.
8.2 Choose the speed of testing from Table 1. Determine
this chosen speed of testing by the specification for the material
being tested, or by agreement between those concerned. When
the speed is not specified, use the lowest speed shown in Table
1 for the specimen geometry being used, which gives rupture
within 1⁄2 to 5-min testing time.
8.3 Modulus determinations may be made at the speed
selected for the other tensile properties when the recorder
response and resolution are adequate.
8.4 Poisson’s ratio determinations shall be made at the same
speed selected for modulus determinations.
9. Conditioning
9.1 Conditioning—Condition the test specimens at 23 6
2°C (73.4 6 3.6°F) and 50 6 5 % relative humidity for not less
than 40 h prior to test in accordance with Procedure A of
Practice D 618, for those tests where conditioning is required.
In cases of disagreement, the tolerances shall be6 1°C (1.8°F)
and 62 % relative humidity.
9.1.1 Note that for some hygroscopic materials, such as
nylons, the material specifications (for example, Specification
D 4066) call for testing “dry as-molded specimens.” Such
requirements take precedence over the above routine precon-
ditioning to 50 % relative humidity and require sealing the
specimens in water vapor-impermeable containers as soon as
DIMENSIONS OF ROD SPECIMENS
molded and not removing them until ready for testing.
9.2 Test Conditions—Conduct tests in the Standard Labora-
Standard Length, L, of
Nominal Diam- Length of Radial
Total Calculated
Specimen to Be Used tory Atmosphere of 23 6 2°C (73.4 6 3.6°F) and 50 6 5 %
Minimum
eter Sections, 2R.S.
Length of Specimen
for 89-mm (31⁄2-in.) relative humidity, unless otherwise specified in the test meth-
JawsA ods. In cases of disagreement, the tolerances shall be 6 1°C
mm (in.) (1.8°F) and 62 % relative humidity.
3.2 (1⁄8) 19.6 (0.773) 356 (14.02) 381 (15)
4.7 (1⁄16) 24.0 (0.946) 361 (14.20) 381 (15)
NOTE 13—The tensile properties of some plastics change rapidly with
6.4 (1⁄4) 27.7 (1.091) 364 (14.34) 381 (15)
9.5 (3⁄8) 33.9 (1.333) 370 (14.58) 381 (15) TABLE 1 Designations for Speed of TestingA
12.7 (1⁄2) 39.0 (1.536) 376 (14.79) 400 (15.75)
15.9 (5⁄8) 43.5 (1.714) 380 (14.96) 400 (15.75) Nominal
19.0 (3⁄4) 47.6 (1.873) 384 (15.12) 400 (15.75) StrainC Rate at
Speed of Testing,
22.2 ( ⁄8)
7 51.5 (2.019) 388 (15.27) 400 (15.75) ClassificationB Specimen Type Start of Test,
mm/min (in./min)
25.4 (1) 54.7 (2.154) 391 (15.40) 419 (16.5) mm/mm· min
31.8 (11⁄4) 60.9 (2.398) 398 (15.65) 419 (16.5) (in./in.·min)
38.1 (11⁄2) 66.4 (2.615) 403 (15.87) 419 (16.5) Rigid and Semirigid I, II, III rods and 5 (0.2) 6 25 % 0.1
42.5 (13⁄4) 71.4 (2.812) 408 (16.06) 419 (16.5) tubes
50.8 (2) 76.0 (2.993) 412 (16.24) 432 (17) 50 (2) 6 10 % 1
A
For other jaws greater than 89 mm (3.5 in.), the standard length shall be 500 (20) 6 10 % 10
increased by twice the length of the jaws minus 178 mm (7 in.). The standard IV 5 (0.2) 6 25 % 0.15
length permits a slippage of approximately 6.4 to 12.7 mm (0.25 to 0.50 in.) in each 50 (2) 6 10 % 1.5
jaw while maintaining the maximum length of the jaw grip. 500 (20) 6 10 % 15
FIG. 3 Diagram Showing Location of Rod Tension Test Specimen V 1 (0.05) 6 25 % 0.1
in Testing Machine 10 (0.5) 6 25 % 1
100 (5)6 25 % 10
marks, and make retests, unless such flaws constitute a variable Nonrigid III 50 (2) 6 10 % 1
to be studied. 500 (20) 6 10 % 10
IV 50 (2) 6 10 % 1.5
NOTE 12—Before testing, all transparent specimens should be inspected 500 (20) 6 10 % 15
in a polariscope. Those which show atypical or concentrated strain A
Select the lowest speed that produces rupture in 1⁄2 to 5 min for the specimen
patterns should be rejected, unless the effects of these residual strains geometry being used (see 8.2).
B
constitute a variable to be studied. See Terminology D 883 for definitions.
C
The initial rate of straining cannot be calculated exactly for dumbbell-shaped
8. Speed of Testing specimens because of extension, both in the reduced section outside the gage
length and in the fillets. This initial strain rate can be measured from the initial slope
8.1 Speed of testing shall be the relative rate of motion of of the tensile strain-versus-time diagram.

6
 D 638
small changes in temperature. Since heat may be generated as a result of TABLE 3 Tensile Stress at Yield, 103 psi, for Eight Laboratories,
straining the specimen at high rates, conduct tests without forced cooling Three Materials
to ensure uniformity of test conditions. Measure the temperature in the Mean Sr SR Ir IR
reduced section of the specimen and record it for materials where
Polypropylene 3.63 0.022 0.161 0.062 0.456
self-heating is suspected. Cellulose acetate butyrate 5.01 0.058 0.227 0.164 0.642
Acrylic 10.4 0.067 0.317 0.190 0.897
10. Procedure
10.1 Measure the width and thickness of rigid flat speci-
TABLE 4 Elongation at Yield, %, for Eight Laboratories, Three
mens (Fig. 1) with a suitable micrometer to the nearest 0.025 Materials
mm (0.001 in.) at several points along their narrow sections.
Mean Sr SR Ir IR
Measure the thickness of nonrigid specimens (produced by a
Cellulose acetate butyrate 3.65 0.27 0.62 0.76 1.75
Type IV die) in the same manner with the required dial Acrylic 4.89 0.21 0.55 0.59 1.56
micrometer. Take the width of this specimen as the distance Polypropylene 8.79 0.45 5.86 1.27 16.5
between the cutting edges of the die in the narrow section.
Measure the diameter of rod specimens, and the inside and 10.5 Record the load-extension curve of the specimen.
outside diameters of tube specimens, to the nearest 0.025 mm 10.6 Record the load and extension at the yield point (if one
(0.001 in.) at a minimum of two points 90° apart; make these exists) and the load and extension at the moment of rupture.
measurements along the groove for specimens so constructed.
NOTE 15—If it is desired to measure both modulus and failure proper-
Use plugs in testing tube specimens, as shown in Fig. 2. ties (yield or break, or both), it may be necessary, in the case of highly
extensible materials, to run two independent tests. The high magnification
TABLE 2 Modulus, 106 psi, for Eight Laboratories, Five Materials
extensometer normally used to determine properties up to the yield point
Mean Sr SR Ir IR may not be suitable for tests involving high extensibility. If allowed to
Polypropylene 0.210 0.0089 0.071 0.025 0.201 remain attached to the specimen, the extensometer could be permanently
Cellulose acetate butyrate 0.246 0.0179 0.035 0.051 0.144 damaged. A broad-range incremental extensometer or hand-rule technique
Acrylic 0.481 0.0179 0.063 0.051 0.144 may be needed when such materials are taken to rupture.
Glass-reinforced nylon 1.17 0.0537 0.217 0.152 0.614
Glass-reinforced polyester 1.39 0.0894 0.266 0.253 0.753 11. Calculation
11.1 Tensile Strength—Calculate the tensile strength by
10.2 Place the specimen in the grips of the testing machine, dividing the maximum load in newtons (or pounds-force) by
taking care to align the long axis of the specimen and the grips the original minimum cross-sectional area of the specimen in
with an imaginary line joining the points of attachment of the square metres (or square inches). Express the result in pascals
grips to the machine. The distance between the ends of the (or pounds-force per square inch) and report it to three
gripping surfaces, when using flat specimens, shall be as significant figures as tensile strength at yield or tensile strength
indicated in Fig. 1. On tube and rod specimens, the location for at break, whichever term is applicable. When a nominal yield
the grips shall be as shown in Fig. 2 and Fig. 3. Tighten the or break load less than the maximum is present and applicable,
grips evenly and firmly to the degree necessary to prevent it may be desirable also to calculate, in a similar manner, the
slippage of the specimen during the test, but not to the point corresponding tensile stress at yield or tensile stress at break
where the specimen would be crushed. and report it to three significant figures (see Note A2.8).
10.3 Attach the extension indicator. When modulus is being 11.2 Percent Elongation—If the specimen gives a yield load
determined, a Class B-2 or better extensometer is required (see that is larger than the load at break, calculate percent elonga-
5.2.1). tion at yield. Otherwise, calculate percent elongation at break.
NOTE 14—Modulus of materials is determined from the slope of the Do this by reading the extension (change in gage length) at the
linear portion of the stress-strain curve. For most plastics, this linear moment the applicable load is reached. Divide that extension
portion is very small, occurs very rapidly, and must be recorded automati- by the original gage length and multiply by 100. Report percent
cally. The change in jaw separation is never to be used for calculating elongation at yield or percent elongation at break to two
modulus or elongation. significant figures. When a yield or breaking load less than the
10.3.1 Poisson’s Ratio Determination: maximum is present and of interest, it is desirable to calculate
10.3.1.1 When Poisson’s ratio is determined, the speed of and report both percent elongation at yield and percent
testing and the load range at which it is determined shall be the elongation at break (see Note A2.2).
same as those used for modulus of elasticity. 11.3 Modulus of Elasticity—Calculate the modulus of elas-
10.3.1.2 Attach the transverse strain measuring device. The ticity by extending the initial linear portion of the load-
transverse strain measuring device must continuously measure extension curve and dividing the difference in stress corre-
the strain simultaneously with the axial strain measuring sponding to any segment of section on this straight line by the
device. corresponding difference in strain. All elastic modulus values
10.3.1.3 Make simultaneous measurements of load and shall be computed using the average initial cross-sectional area
strain and record the data. The precision of the value of of the test specimens in the calculations. The result shall be
Poisson’s ratio will depend on the number of data points of expressed in pascals (pounds-force per square inch) and
axial and transverse strain taken. reported to three significant figures.
10.4 Set the speed of testing at the proper rate as required in 11.4 Secant modulus—At a designated strain, this shall be
Section 8, and start the machine. calculated by dividing the corresponding stress (nominal) by

7
 D 638

FIG. 4 Plot of Strains Versus Load for Determination of Poisson’s Ratio

the designated strain. Elastic modulus values are preferable and 11.8 See Annex A1 for information on toe compensation.
shall be calculated whenever possible. However, for materials
where no proportionality is evident, the secant value shall be TABLE 5 Tensile Strength at Break, 103 psi, for Eight
calculated. Draw the tangent as directed in A1.3 and Fig. A1.2, Laboratories, Five MaterialsA
and mark off the designated strain from the yield point where Mean Sr SR Ir IR
the tangent line goes through zero stress. The stress to be used Polypropylene 2.97 1.54 1.65 4.37 4.66
in the calculation is then determined by dividing the load- Cellulose acetate butyrate 4.82 0.058 0.180 0.164 0.509
extension curve by the original average cross-sectional area of Acrylic 9.09 0.452 0.751 1.27 2.13
Glass-reinforced polyester 20.8 0.233 0.437 0.659 1.24
the specimen. Glass-reinforced nylon 23.6 0.277 0.698 0.784 1.98
11.5 Poisson’s Ratio—The axial strain, ea, indicated by the A
Tensile strength and elongation at break values obtained for unreinforced
axial extensometer, and the transverse strain, e, indicated by propylene plastics generally are highly variable due to inconsistencies in necking
the transverse extensometers, are plotted against the applied or “drawing” of the center section of the test bar. Since tensile strength and
elongation at yield are more reproducible and relate in most cases to the practical
load, P, as shown in Fig. 4. A straight line is drawn through usefulness of a molded part, they are generally recommended for specification
each set of points, and the slopes, dea / dP and det / dP, of these purposes.
lines are determined. Poisson’s ratio, µ, is then calculated as
follows: TABLE 6 Elongation at Break, %, for Eight Laboratories, Five
µ 5 2~det / dP!/~dea / dP! (1) MaterialsA
Mean Sr SR Ir IR
where: Glass-reinforced polyester 3.68 0.20 2.33 0.570 6.59
det 5 change in transverse strain, Glass-reinforced nylon 3.87 0.10 2.13 0.283 6.03
dea 5 change in axial strain, and Acrylic 13.2 2.05 3.65 5.80 10.3
dP 5 change in applied load; Cellulose acetate butyrate 14.1 1.87 6.62 5.29 18.7
Polypropylene 293.0 50.9 119.0 144.0 337.0
or A
Tensile strength and elongation at break values obtained for unreinforced
µ 5 2~det! / ~dea! (2) propylene plastics generally are highly variable due to inconsistencies in necking
or “drawing” of the center section of the test bar. Since tensile strength and
11.5.1 The errors that may be introduced by drawing a elongation at yield are more reproducible and relate in most cases to the practical
straight line through the points can be reduced by applying the usefulness of a molded part, they are generally recommended for specification
method of least squares. purposes.

11.6 For each series of tests, calculate the arithmetic mean

of all values obtained and report it as the “average value” for 12. Report
the particular property in question. 12.1 Report the following information:
11.7 Calculate the standard deviation (estimated) as follows 12.1.1 Complete identification of the material tested, includ-
and report it to two significant figures: ing type, source, manufacturer’s code numbers, form, principal
s 5 =~ (X 2 2 nX̄ 2! / ~n 2 1! (3) dimensions, previous history, etc.,
12.1.2 Method of preparing test specimens,
where: 12.1.3 Type of test specimen and dimensions,
s 5 estimated standard deviation, 12.1.4 Conditioning procedure used,
X 5 value of single observation, 12.1.5 Atmospheric conditions in test room,
n 5 number of observations, and 12.1.6 Number of specimens tested,
X̄ 5 arithmetic mean of the set of observations.
12.1.7 Speed of testing,

8
 D 638
TABLE 7 Tensile Yield Strength, for Ten Laboratories, Eight TABLE 9 Tensile Break Strength, for Nine Laboratories, Six
Materials Materials
Test Values Expressed in psi Units Test Values Expressed in psi Units
Material Speed, Material Speed,
in./min Average Sr SR r R in./min Average Sr SR r R
LDPE 20 1544 52.4 64.0 146.6 179.3 LDPE 20 1592 52.3 74.9 146.4 209.7
LDPE 20 1894 53.1 61.2 148.7 171.3 LDPE 20 1750 66.6 102.9 186.4 288.1
LLDPE 20 1879 74.2 99.9 207.8 279.7 LLDPE 20 4379 127.1 219.0 355.8 613.3
LLDPE 20 1791 49.2 75.8 137.9 212.3 LLDPE 20 2840 78.6 143.5 220.2 401.8
LLDPE 20 2900 55.5 87.9 155.4 246.1 LLDPE 20 1679 34.3 47.0 95.96 131.6
LLDPE 20 1730 63.9 96.0 178.9 268.7 LLDPE 20 2660 119.1 166.3 333.6 465.6
HDPE 2 4101 196.1 371.9 549.1 1041.3
HDPE 2 3523 175.9 478.0 492.4 1338.5
TABLE 10 Tensile Break Elongation, for Nine Laboratories, Six
Materials
12.1.8 Classification of extensometers used. A description Test Values Expressed in Percent Units
of measuring technique and calculations employed instead of a Material Speed,
in./min Average Sr SR r R
minimum Class-C extensometer system,
12.1.9 Tensile strength at yield or break, average value, and LDPE 20 567 31.5 59.5 88.2 166.6
LDPE 20 569 61.5 89.2 172.3 249.7
standard deviation, LLDPE 20 890 25.7 113.8 71.9 318.7
12.1.10 Tensile stress at yield or break, if applicable, LLDPE 20 64.4 6.68 11.7 18.7 32.6
average value, and standard deviation, LLDPE 20 803 25.7 104.4 71.9 292.5
LLDPE 20 782 41.6 96.7 116.6 270.8
12.1.11 Percent elongation at yield or break, or both, as
applicable, average value, and standard deviation,
12.1.12 Modulus of elasticity, average value, and standard each material. Data from some laboratories could not be used
deviation, for various reasons, and this is noted in each table.
12.1.13 Date of test, and 13.1.2 In Tables 2-10, for the materials indicated, and for
12.1.14 Revision date of Test Method D 638. test results that derived from testing five specimens:
13.1.2.1 Sr is the within-laboratory standard deviation of
13. Precision and Bias 12 the average; Ir 5 2.83 Sr. (See 13.1.2.3 for application of Ir.)
13.1 Precision—Tables 2-6 are based on a round-robin test 13.1.2.2 SR is the between-laboratory standard deviation of
conducted in 1984, involving five materials tested by eight the average; IR 5 2.83 SR. (See 13.1.2.4 for application of IR.)
laboratories using the Type I specimen, all of nominal 0.125-in. 13.1.2.3 Repeatability—In comparing two test results for
thickness. Each test result was based on five individual the same material, obtained by the same operator using the
determinations. Each laboratory obtained two test results for same equipment on the same day, those test results should be
each material. judged not equivalent if they differ by more than the Ir value
for that material and condition.
TABLE 8 Tensile Yield Elongation, for Eight Laboratories, Eight 13.1.2.4 Reproducibility—In comparing two test results for
Materials the same material, obtained by different operators using differ-
Test Values Expressed in Percent Units ent equipment on different days, those test results should be
Material Speed, judged not equivalent if they differ by more than the IR value
in./min Average Sr SR r R
for that material and condition. (This applies between different
LDPE 20 17.0 1.26 3.16 3.52 8.84
LDPE 20 14.6 1.02 2.38 2.86 6.67
laboratories or between different equipment within the same
LLDPE 20 15.7 1.37 2.85 3.85 7.97 laboratory.)
LLDPE 20 16.6 1.59 3.30 4.46 9.24 13.1.2.5 Any judgment in accordance with 13.1.2.3 and
LLDPE 20 11.7 1.27 2.88 3.56 8.08
LLDPE 20 15.2 1.27 2.59 3.55 7.25
13.1.2.4 will have an approximate 95 % (0.95) probability of
HDPE 2 9.27 1.40 2.84 3.91 7.94 being correct.
HDPE 2 9.63 1.23 2.75 3.45 7.71 13.1.2.6 Other formulations may give somewhat different
results.
13.1.1 Tables 7-10 are based on a round-robin test con- 13.1.2.7 For further information on the methodology used in
ducted by the polyolefin subcommittee in 1988, involving eight this section, see Practice E 691.
polyethylene materials tested in ten laboratories. For each 13.1.2.8 The precision of this test method is very dependent
material, all samples were molded at one source, but the upon the uniformity of specimen preparation, standard prac-
individual specimens were prepared at the laboratories that tices for which are covered in other documents.
tested them. Each test result was the average of five individual 13.2 Bias—There are no recognized standards on which to
determinations. Each laboratory obtained three test results for base an estimate of bias for this test method.
14. Keywords
12
Supporting data are available from ASTM Headquarters. Request RR:D20- 14.1 modulus of elasticity; percent elongation; plastics;
1125 for the 1984 round robin and RR:D20-1170 for the 1988 round robin. tensile properties; tensile strength

9
 D 638

ANNEXES

(Mandatory Information)

A1. TOE COMPENSATION

A1.1 In a typical stress-strain curve (Fig. A1.1) there is a elastic modulus can be determined by dividing the stress at any
toe region, AC, that does not represent a property of the point along the line CD (or its extension) by the strain at the
material. It is an artifact caused by a takeup of slack and same point (measured from Point B, defined as zero-strain).
alignment or seating of the specimen. In order to obtain correct
values of such parameters as modulus, strain, and offset yield A1.3 In the case of a material that does not exhibit any
point, this artifact must be compensated for to give the linear region (Fig. A1.2), the same kind of toe correction of the
corrected zero point on the strain or extension axis. zero-strain point can be made by constructing a tangent to the
maximum slope at the inflection point (H8). This is extended to
A1.2 In the case of a material exhibiting a region of- intersect the strain axis at Point B8, the corrected zero-strain
Hookean (linear) behavior (Fig. A1.1), a continuation of the point. Using Point B8 as zero strain, the stress at any point (G8)
linear (CD) region of the curve is constructed through the on the curve can be divided by the strain at that point to obtain
zero-stress axis. This intersection (B) is the corrected zero- a secant modulus (slope of Line B8 G8). For those materials
strain point from which all extensions or strains must be with no linear region, any attempt to use the tangent through
measured, including the yield offset (BE), if applicable. The the inflection point as a basis for determination of an offset
yield point may result in unacceptable error.

NOTE 1—Some chart recorders plot the mirror image of this graph. NOTE 1—Some chart recorders plot the mirror image of this graph.
FIG. A1.1 Material with Hookean Region FIG. A1.2 Material with No Hookean Region

A2. DEFINITIONS OF TERMS AND SYMBOLS RELATING TO TENSION TESTING OF PLASTICS

A2.1 elastic limit—the greatest stress which a material is A2.2 elongation—the increase in length produced in the
capable of sustaining without any permanent strain remaining gage length of the test specimen by a tensile load. It is
upon complete release of the stress. It is expressed in force per expressed in units of length, usually inches (millimetres). (Also
unit area, usually pounds-force per square inch (megapascals). known as extension.)
NOTE A2.1—Measured values of proportional limit and elastic limit NOTE A2.2—Elongation and strain values are valid only in cases where
vary greatly with the sensitivity and accuracy of the testing equipment,
uniformity of specimen behavior within the gage length is present. In the
eccentricity of loading, the scale to which the stress-strain diagram is
case of materials exhibiting necking phenomena, such values are only of
plotted, and other factors. Consequently, these values are usually replaced
by yield strength.

10
 D 638
qualitative utility after attainment of yield point. This is due to inability to A2.7 percent elongation—the elongation of a test specimen
ensure that necking will encompass the entire length between the gage expressed as a percent of the gage length.
marks prior to specimen failure.
A2.8 percent elongation at break and yield:
A2.3 gage length—the original length of that portion of the
specimen over which strain or change in length is determined. A2.8.1 percent elongation at break
the percent elongation at the moment of rupture of the test
A2.4 modulus of elasticity—the ratio of stress (nominal) to specimen.
corresponding strain below the proportional limit of a material. A2.8.2 percent elongation at yield
It is expressed in force per unit area, usually megapascals the percent elongation at the moment the yield point (A2.21)
(pounds-force per square inch). (Also known as elastic modu- is attained in the test specimen.
lus or Young’s modulus).
A2.9 percent reduction of area (nominal)—the difference
NOTE A2.3—The stress-strain relations of many plastics do not con- between the original cross-sectional area measured at the point
form to Hooke’s law throughout the elastic range but deviate therefrom of rupture after breaking and after all retraction has ceased,
even at stresses well below the elastic limit. For such materials the slope expressed as a percent of the original area.
of the tangent to the stress-strain curve at a low stress is usually taken as
the modulus of elasticity. Since the existence of a true proportional limit A2.10 percent reduction of area (true)—the difference
in plastics is debatable, the propriety of applying the term “modulus of
between the original cross-sectional area of the test specimen
elasticity” to describe the stiffness or rigidity of a plastic has been
seriously questioned. The exact stress-strain characteristics of plastic and the minimum cross-sectional area within the gage bound-
materials are very dependent on such factors as rate of stressing, aries prevailing at the moment of rupture, expressed as a
temperature, previous specimen history, etc. However, such a value is percentage of the original area.
useful if its arbitrary nature and dependence on time, temperature, and
other factors are realized. A2.11 proportional limit—the greatest stress which a
material is capable of sustaining without any deviation from
A2.5 necking—the localized reduction in cross section proportionality of stress to strain (Hooke’s law). It is expressed
which may occur in a material under tensile stress. in force per unit area, usually megapascals (pounds-force per
square inch).
A2.6 offset yield strength—the stress at which the strain
exceeds by a specified amount (the offset) an extension of the A2.12 rate of loading—the change in tensile load carried
initial proportional portion of the stress-strain curve. It is by the specimen per unit time. It is expressed in force per unit
expressed in force per unit area, usually megapascals (pounds- time, usually newtons (pounds-force) per minute. The initial
force per square inch). rate of loading can be calculated from the initial slope of the
NOTE A2.4—This measurement is useful for materials whose stress- load versus time diagram.
strain curve in the yield range is of gradual curvature. The offset yield
strength can be derived from a stress-strain curve as follows (Fig. A2.1): A2.13 rate of straining—the change in tensile strain per
On the strain axis lay off OM equal to the specified offset. unit time. It is expressed either as strain per unit time, usually
Draw OA tangent to the initial straight-line portion of the stress-strain metres per metre (inches per inch) per minute, or percent
curve. elongation per unit time, usually percent elongation per minute.
Through M draw a line MN parallel to OA and locate the intersection of The initial rate of straining can be calculated from the initial
MN with the stress-strain curve. slope of the tensile strain versus time diagram.
The stress at the point of intersection r is the “offset yield strength.” The
specified value of the offset must be stated as a percent of the original gage NOTE A2.5—The initial rate of straining is synonymous with the rate of
length in conjunction with the strength value. Example: 0.1 % offset yield crosshead movement divided by the initial distance between crossheads
strength 5 ... MPa (psi), or yield strength at 0.1 % offset ... MPa (psi). only in a machine with constant rate of crosshead movement and when the
specimen has a uniform original cross section, does not “neck down,” and
does not slip in the jaws.

A2.14 rate of stressing (nominal)—the change in tensile

stress (nominal) per unit time. It is expressed in force per unit
area per unit time, usually megapascals (pounds-force per
square inch) per minute. The initial rate of stressing can be
calculated from the initial slope of the tensile stress (nominal)
versus time diagram.
NOTE A2.6—The initial rate of stressing as determined in this manner
has only limited physical significance. It does, however, roughly describe
the average rate at which the initial stress (nominal) carried by the test
specimen is applied. It is affected by the elasticity and flow characteristics
of the materials being tested. At the yield point, the rate of stressing (true)
may continue to have a positive value if the cross-sectional area is
decreasing.

A2.15 secant modulus—the ratio of stress (nominal) to

FIG. A2.1 Offset Yield Strength corresponding strain at any specified point on the stress-strain

11
 D 638
curve. It is expressed in force per unit area, usually megapas- NOTE A2.9—Only materials whose stress-strain curves exhibit a point
cals (pounds-force per square inch), and reported together with of zero slope may be considered as having a yield point.
the specified stress or strain. NOTE A2.10—Some materials exhibit a distinct “break” or discontinu-
ity in the stress-strain curve in the elastic region. This break is not a yield
NOTE A2.7—This measurement is usually employed in place of modu- point by definition. However, this point may prove useful for material
lus of elasticity in the case of materials whose stress-strain diagram does characterization in some cases.
not demonstrate proportionality of stress to strain.
A2.22 yield strength—the stress at which a material exhib-
A2.16 strain—the ratio of the elongation to the gage length its a specified limiting deviation from the proportionality of
of the test specimen, that is, the change in length per unit of stress to strain. Unless otherwise specified, this stress will be
original length. It is expressed as a dimensionless ratio. the stress at the yield point and when expressed in relation to
the tensile strength shall be designated either tensile strength at
A2.17 tensile strength (nominal)—the maximum tensile yield or tensile stress at yield as required in A2.17 (Fig. A2.3).
stress (nominal) sustained by the specimen during a tension (See offset yield strength.)
test. When the maximum stress occurs at the yield point A2.23 Symbols—The following symbols may be used for
(A2.21), it shall be designated tensile strength at yield. When the above terms:
the maximum stress occurs at break, it shall be designated Symbol Term
tensile strength at break. W Load
DW Increment of load
A2.18 tensile stress (nominal)—the tensile load per unit L Distance between gage marks at any time
Lo Original distance between gage marks
area of minimum original cross section, within the gage Lu Distance between gage marks at moment of rupture
boundaries, carried by the test specimen at any given moment. DL Increment of distance between gage marks 5 elongation
It is expressed in force per unit area, usually megapascals A Minimum cross-sectional area at any time
Ao Original cross-sectional area
(pounds-force per square inch). DA Increment of cross-sectional area
Au Cross-sectional area at point of rupture measured after
NOTE A2.8—The expression of tensile properties in terms of the breaking specimen
minimum original cross section is almost universally used in practice. In AT Cross-sectional area at point of rupture, measured at the
the case of materials exhibiting high extensibility or necking, or both moment of rupture
(A2.15), nominal stress calculations may not be meaningful beyond the t Time
yield point (A2.21) due to the extensive reduction in cross-sectional area Dt Increment of time
s Tensile stress
that ensues. Under some circumstances it may be desirable to express the
Ds Increment of stress
tensile properties per unit of minimum prevailing cross section. These sT True tensile stress
properties are called true tensile properties (that is, true tensile stress, etc.). sU Tensile strength at break (nominal)
sUT Tensile strength at break (true)
A2.19 tensile stress-strain curve—a diagram in which e Strain
De Increment of strain
values of tensile stress are plotted as ordinates against corre- eU Total strain, at break
sponding values of tensile strain as abscissas. eT True strain
%El Percentage elongation
A2.20 true strain (see Fig. A2.2) is defined by the follow- Y.P. Yield point
E Modulus of elasticity
ing equation for eT:
A2.24 Relations between these various terms may be
* dL/L 5 ln L/L
L
eT 5 o (A2.1) defined as follows:
Lo
s 5 W/Ao
where: sT 5 W/A
sU 5 W/Ao (where W is breaking load)
dL 5 increment of elongation when the distance between sUT 5 W/AT (where W is breaking load)
the gage marks is L, e 5 DL/Lo 5 (L − Lo)/Lo
Lo 5 original distance between gage marks, and eU 5 (Lu − Lo)/Lo
L 5 distance between gage marks at any time. eT 5 *LLo dL/L 5 ln L/Lo
%El 5 [(L − Lo)/Lo] 3 100 5 e 3 100
Percent reduction of area (nominal) 5 [(Ao − Au)/Ao] 3 100
A2.21 yield point—the first point on the stress-strain curve Percent reduction of area (true) 5 [(Ao − AT)/Ao] 3 100
at which an increase in strain occurs without an increase in Rate of loading 5 DW/Dt
stress (Fig. A2.2). Rate of stressing (nominal) 5 Ds/D 5 (DW]/Ao)/Dt
Rate of straining 5 De/Dt 5 (DL/Lo)Dt
For the case where the volume of the test specimen does not
change during the test, the following three relations hold:
sT 5 s~1 1 e! 5 sL/Lo (A2.2)
sUT 5 sU ~1 1 eU! 5 sU Lu /Lo
A 5 Ao /~1 1 e!
FIG. A2.2 Illustration of True Strain Equation

12
 D 638

FIG. A2.3 Tensile Designations

SUMMARY OF CHANGES

This section identifies the location of selected changes to this test method. For the convenience of the user,
Committee D-20 has highlighted those changes that may impact the use of this test method. This section may
also include descriptions of the changes or reasons for the changes, or both.

D 638–98: (2) Added 12.1.14.

(1) Revised 10.3 and added 12.1.8 to clarify extensometer (3) Replaced reference to Test Methods D 374 with Test
usage. Method D 5947 in 2.1 and 5.3.

The American Society for Testing and Materials takes no position respecting the validity of any patent rights asserted in connection
with any item mentioned in this standard. Users of this standard are expressly advised that determination of the validity of any such
patent rights, and the risk of infringement of such rights, are entirely their own responsibility.

This standard is subject to revision at any time by the responsible technical committee and must be reviewed every five years and
if not revised, either reapproved or withdrawn. Your comments are invited either for revision of this standard or for additional standards
and should be addressed to ASTM Headquarters. Your comments will receive careful consideration at a meeting of the responsible
technical committee, which you may attend. If you feel that your comments have not received a fair hearing you should make your
views known to the ASTM Committee on Standards, 100 Barr Harbor Drive, West Conshohocken, PA 19428.

13