Designation: D 5628 – 96 (Reapproved 2001)e1

Standard Test Method for

Impact Resistance of Flat, Rigid Plastic Specimens by
Means of a Falling Dart (Tup or Falling Mass)1
This standard is issued under the fixed designation D 5628; 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.

e1 NOTE—Editorial changes were made throughout in November 2001.

1. Scope Tup Falling Weight5

1.1 This test method covers the determination of the relative D 3763 Test Method for High-Speed Puncture Properties of
ranking of materials according to the energy required to crack Plastics Using Load and Displacement Sensors6
or break flat, rigid plastic specimens under various specified D 4066 Classification System for Nylon Injection and
conditions of impact of a free-falling dart (tup). Extrusion Materials PA6
1.2 The values stated in SI units are to be regarded as the E 177 Practice for Use of the Terms Precision and Bias in
standard. The values in parentheses are for information only. ASTM Test Methods7
1.3 This standard does not purport to address all of the E 691 Practice for Conducting an Interlaboratory Study to
safety concerns, if any, associated with its use. It is the Determine the Precision of a Test Method7
responsibility of the user of this standard to establish appro- 2.2 ISO Standards:
priate safety and health practices and determine the applica- ISO 291 Standard Atmospheres for Conditioning and Test-
bility of regulatory limitations prior to use. Specific hazard ing8
statements are given in Section 8. ISO 6603-1 Plastics-Determination of Multiaxial Impact
Behavior of Rigid Plastics—Part 1: Falling Dart Method8
NOTE 1—This test method and ISO 6603-1-1985 are technically
equivalent only when the test conditions and specimen geometry required 3. Terminology
for Geometry FE and the Bruceton Staircase method of calculation are
used. 3.1 Definitions:
3.1.1 For definitions of plastic terms used in this test
2. Referenced Documents method, see Terminologies D 883 and D 1600.
2.1 ASTM Standards: 3.2 Definitions of Terms Specific to This Standard:
D 374 Test Methods for Thickness of Solid Electrical Insu- 3.2.1 failure (of test specimen)—the presence of any crack
lation2 or split, created by the impact of the falling tup, that can be
D 618 Practice for Conditioning Plastics for Testing3 seen by the naked eye under normal laboratory lighting
D 883 Terminology Relating to Plastics2 conditions.
D 1600 Terminology for Abbreviated Terms Relating to 3.2.2 mean-failure energy (mean-impact resistance)—the
Plastics2 energy required to produce 50 % failures, equal to the product
D 1709 Test Method for Impact Resistance of Plastic Film of the constant drop height and the mean-failure mass or the
by the Free Falling Dart Method2 product of the constant mass and mean-failure height.
D 1898 Practice for Sampling of Plastics4 3.2.3 mean-failure height (impact-failure height)—the
D 2444 Test Method for Determination of the Impact Re- height at which a standard mass, when dropped on test
sistance of Thermoplastic Pipe and Fittings by Means of a specimens, will cause 50 % failures.
NOTE 2—Cracks usually start at the surface opposite the one that is
struck. Occasionally incipient cracking in glass-reinforced products, for
1
This test method is under the jurisdiction of ASTM Committee D20 on Plastics
and is the direct responsibility of Subcommittee D20.10 on Mechanical Properties.
Current edition approved March 10, 1996. Published July 1996. Originally 5
Annual Book of ASTM Standards, Vol 08.04.
published as D 5628 – 94. Last previous edition D 5628 – 95. 6
Annual Book of ASTM Standards, Vol 08.02.
2
Annual Book of ASTM Standards, Vol 08.01. 7
Annual Book of ASTM Standards, Vol 14.02.
3
Annual Book of ASTM Standards, Vol 10.01. 8
Available from American National Standards Institute, 25 W. 43rd St., 4th
4
Discontinued; see 1998 Annual Book of ASTM Standards, Vol 08.01. Floor, New York, NY 10036.

Copyright © ASTM International, 100 Barr Harbor Drive, PO Box C700, West Conshohocken, PA 19428-2959, United States.

1
 D 5628 – 96 (2001)e1
example, may be difficult to differentiate from the reinforcing fibers. In 5.3.1 The conical configuration of the 12.7-mm diameter
such cases, a penetrating dye may be used to confirm the onset of crack tup used in Geometry FB minimizes problems with tup
formation. penetration and sticking in failed specimens of some ductile
3.2.4 mean-failure mass (impact-failure mass)—the mass of materials.
the dart (tup) that, when dropped on the test specimens from a 5.4 The test conditions of Geometry FC are the same as
standard height, will cause 50 % failures. those of Test Method A of Test Method D 1709. They have
3.2.5 tup—a dart with a hemispherical nose. See 7.2 and been used in specifications for extruded sheeting. A limitation
Fig. 1. of this geometry is that considerable material is required.
5.5 The test conditions of Geometry FD are the same as for
4. Summary of Test Method Test Method D 3763.
4.1 A free-falling dart (tup) is allowed to strike a supported 5.6 The test conditions of Geometry FE are the same as for
specimen directly. Either a dart having a fixed mass may be ISO 6603-1.
dropped from various heights, or a dart having an adjustable 5.7 Because of the nature of impact testing, the selection of
mass may be dropped from a fixed height. (See Fig. 2). a test method and tup must be somewhat arbitrary. While any
4.2 The procedure determines the energy (mass 3 height) one of the tup geometries may be selected, knowledge of the
that will cause 50 % of the specimens tested to fail (mean final or intended end-use application should be considered.
failure energy). 5.8 Clamping of the test specimen will improve the preci-
4.3 The technique used to determine mean failure energy is sion of the data. Therefore, clamping is recommended. How-
commonly called the Bruceton Staircase Method or the Up- ever, with rigid specimens, valid determinations can be made
and-Down Method (1).9 Testing is concentrated near the mean, without clamping. Unclamped specimens tend to exhibit some-
reducing the number of specimens required to obtain a reason- what greater impact resistance.
ably precise estimate of the impact resistance. 5.9 Before proceeding with this test method, reference
4.4 Each test method permits the use of different tup and test should be made to the specification of the material being tested.
specimen geometries to obtain different modes of failure, Any test specimens preparation, conditioning, dimensions, or
permit easier sampling, or test limited amounts of material. testing parameters or combination thereof covered in the
There is no known means for correlating the results of tests relevant ASTM materials specification shall take precedence
made by different impact methods or procedures. over those mentioned in this test method. If there are no
relevant ASTM material specifications, then the default condi-
5. Significance and Use tions apply.
5.1 Plastics are viscoelastic and therefore may be sensitive
to changes in velocity of the mass falling on their surfaces. 6. Interferences
However, the velocity of a free-falling object is a function of 6.1 Falling-mass-impact-test results are dependent on the
the square root of the drop height. A change of a factor of two geometry of both the falling mass and the support. Thus,
in the drop height will cause a change of only 1.4 in velocity. impact tests should be used only to obtain relative rankings of
Hagan et al (2) found that the mean-failure energy of sheeting materials. Impact values cannot be considered absolute unless
was constant at drop heights between 0.30 and 1.4 m. This the geometry of the test equipment and specimen conform to
suggests that a constant mass-variable height method will give the end-use requirement. Data obtained by different procedures
the same results as the constant height-variable mass tech- within this test method, or with different geometries, cannot, in
nique. On the other hand, different materials respond differ- general, be compared directly with each other. However, the
ently to changes in the velocity of impact. Equivalence of these relative ranking of materials may be expected to be the same
methods should not be taken for granted. While both constant- between two test methods if the mode of failure and the impact
mass and constant-height techniques are permitted by these velocities are the same.
methods, the constant-height method should be used for those 6.1.1 Falling-mass-impact types of tests are not suitable for
materials that are found to be rate-sensitive in the range of predicting the relative ranking of materials at impact velocities
velocities encountered in falling-weight types of impact tests. differing greatly from those imposed by these test methods.
5.2 The test geometry FA causes a moderate level of stress 6.2 As cracks usually start at the surface opposite the one
concentration and can be used for most plastics. that is struck, the results can be greatly influenced by the
5.3 Geometry FB causes a greater stress concentration and quality of the surface of test specimens. Therefore, the com-
results in failure of tough or thick specimens that do not fail position of this surface layer, its smoothness or texture, levels
with Geometry FA (3). This approach may produce a punch of and type of texture, and the degree of orientation introduced
shear failure on thick sheet. If that type of failure is undesir- during the formation of the specimen (such as may occur
able, Geometry FC may be used. Geometry FB is suitable for during injection molding) are very important variables. Flaws
research and development because of the smaller test area in this surface will also affect results.
required. 6.3 Impact properties of plastic materials can be very
sensitive to temperature. This test can be carried out at any
reasonable temperature and humidity, thus representing actual
9
The boldface numbers in parentheses refer to a list of references at the end of use environments. However, this test method is intended
the text. primarily for rating materials under specific impact conditions.

2
 D 5628 – 96 (2001)e1

Dimensions of Conical Dart (Not to scale.)—Fig. 1(b)

NOTE 1—Unless specified, the tolerance on all dimensions shall be 62 %.

Position Dimension, mm Dimension, in.
A 27.2 1.07
B 15 0.59
C 12.2 0.48
D 6.4 0.25
E 25.4 1
F 12.7 0.5
R 6.35 6 0.05 0.250 6 0.002
(nose radius)
r (radius) 0.8 0.03
S (diameter)A 6.4 0.25
u 25 6 1° 25 6 1°
A
Larger diameter shafts may be used.
FIG. 1 Tup Geometries for Geometries FA (1a), FB (1b), FC (1c), FD (1d), and FE (1e)

3
 D 5628 – 96 (2001)e1

FIG. 2 One Type of Falling Mass Impact Tester

7. Apparatus 7.1.2 Tup Support, capable of supporting a 13.5-kg (30-lb)

mass, with a release mechanism and a centering device to
7.1 Testing Machine—The apparatus shall be constructed
ensure uniform, reproducible drops.
essentially as is shown in Fig. 2. The geometry of the specimen
clamp and tup shall conform to the dimensions given in 7.1.1 NOTE 3—Reproducible drops may be ensured through the use of a tube
and 7.2. or cage within which the tup falls. In this event, care should be exercised
so that any friction that develops will not reduce the velocity of the tup
7.1.1 Specimen Clamp—For flat specimens, a two-piece appreciably.
annular specimen clamp similar to that shown in Fig. 3 is
recommended. For Geometries FA and FD, the inside diameter 7.1.3 Positioning Device—Means shall be provided for
positioning the tup so that the distance from the impinging
should be 76.0 6 3.0 mm (3.00 6 0.12 in.). For Geometry FB,
surface of the tup head to the test specimen is as specified.
the inside diameter should be 38.1 6 0.80 mm (1.5 6 0.03 in.).
7.2 Tup:
For Geometry FC, the inside diameter should be 127.0 6 2.5
7.2.1 The tup used in Geometry FA shall have a 15.86 6
mm (5.00 6 0.10 in.). For Geometry FE an annular specimen
0.10-mm (0.625 6 0.004-in.) diameter hemispherical head of
clamp similar to that shown in Fig. 4 is required. The inside
tool steel hardened to 54 HRC or harder. A steel shaft about 13
diameter should be 40 6 2 mm (1.57 6 0.08 in.) (see Table 1). mm (0.5 in.) in diameter shall be attached to the center of the
For Geometries FA, FB, FC, and FD, the inside edge of the flat surface of the head with its longitudinal axis at 90° to that
upper or supporting surface of the lower clamp should be surface. The length of the shaft shall be great enough to
rounded slightly; a radius of 0.8 mm (0.03 in.) has been found accommodate the maximum mass required (see Fig. 1(a) and
to be satisfactory. For Geometry FE this radius should be 1 mm Table 1).
(0.04 in.). 7.2.2 The tup used in Geometry FB shall be made of tool
7.1.1.1 Contoured specimens shall be firmly held in a jig so steel hardened to 54 HRC or harder. The head shall have a
that the point of impact will be the same for each specimen. diameter of 12.76 0.1 mm (0.500 6 0.003 in.) with a conical

4
 D 5628 – 96 (2001)e1

FIG. 3 Support Plate/Specimen/Clamp Configuration for Geometries FA, FB, FC, and FD

FIG. 4 Test-Specimen Support for Geometry FE

TABLE 1 Tup and Support Ring Dimensions axis at 90° to that surface. The length of the shaft shall be great
Dimensions, mm (in.) enough to accommodate the maximum mass (see Fig. 1(c) and
Geometry
Tup Diameter Inside Diameter Support Ring Table 1).
FA 15.86 6 0.10 76.0 6 3.0 7.2.4 The tup used in Geometry FD shall have a 12.70 6
(0.625 6 0.004) (3.00 6 0.12) 0.25-mm (0.500 6 0.010-in.) diameter hemispherical head of
FB 12.7 6 0.1 38.1 6 0.8
(0.500 6 0.003) (1.5 6 0.03) tool steel hardened to 54 HRC or harder. A steel shaft about 8
FC 38.1 6 0.4 127.0 6 2.5 mm (0.31 in.) in diameter shall be attached to the center of the
(1.5 6 0.010) (5.00 6 0.10)
FD 12.70 6 0.25 76.0 6 3.0
flat surface of the head with its longitudinal axis at 90° to the
(0.500 6 0.010) (3.00 6 0.12) surface. The length of the shaft shall be great enough to
FE 20.0 6 0.2 40.0 6 2.0 accommodate the maximum mass required (see Fig. 1(d) and
(0.787 6 0.008) (1.57 6 0.08)
Table 1).
7.2.5 The tup used in Geometry FE shall have a 20.0 6
0.2-mm (0.787 6 0.008-in.) diameter hemispherical head of
(50° included angle) configuration such that the conical surface tool steel hardened to 54 HRC or harder. A steel shaft about 13
is tangent to the hemispherical nose. A 6.4-mm (0.25-in.) mm (0.5 in.) in diameter shall be attached to the center of the
diameter shaft is satisfactory (see Fig. 1(b) and Table 1). flat surface of the head with its longitudinal axis at 90° to the
7.2.3 The tup used for Geometry FC shall be made of tool surface. The length of the shaft shall be great enough to
steel hardened to 54 HRC or harder. The hemispherical head accommodate the maximum mass required (see Fig. 1(e) and
shall have a diameter of 38.1 6 0.4 mm (1.5 6 0.015 in.). A Table 1).
steel shaft about 13 mm (0.5 in.) in diameter shall be attached 7.2.6 The tup head shall be free of nicks, scratches, or other
to the center of the flat surface of the head with its longitudinal surface irregularities.

5
 D 5628 – 96 (2001)e1
TABLE 2 Minimum Size of Specimen 10.3 When the approximate mean failure mass for a given
Geometry Specimen Diameter, mm (in.) Square Specimen, mm (in.) sample is known, 20 specimens will usually yield sufficiently
FA 89 (3.5) 89 by 89 precise results. If the approximate mean failure mass is
(3.5 by 3.5) unknown, six or more additional specimens should be used to
FB 51 (2.0) 51 by 51
(2.0 by 2.0)
determine the appropriate starting point of the test. For
FC 140 (5.5) 140 by 140 compliance with ISO 6603-1 a minimum of 30 specimens must
(5.5 by 5.5) be tested.
FD 89 (3.5) 89 by 89
(3.5 by 3.5)
10.4 Carefully examine the specimen visually to ensure that
FE 58 (2.3) 58 by 58 samples are free of cracks or other obvious imperfections or
(2.3 by 2.3) damages, unless these imperfections constitute variables under
study. Samples known to be defective should not be tested for
specification purposes. Production parts, however, should be
7.3 Masses—Cylindrical steel masses are required that have tested in the as-received condition to determine conformance to
a center hole into which the tup shaft will fit. A variety of specified standards.
masses are needed if different materials or thicknesses are to be 10.5 Select a suitable method for making the specimen that
tested. For a material of low impact resistance, the tup mass will not affect the impact resistance of the material.
may need to be adjusted by increments of 10 g or less. 10.6 Specimens may have flat smooth surfaces on both
Materials of high impact resistance may require increments of sides, be textured on one side and smooth on the other side, or
1 kg or more. be textured on both surfaces. Both surfaces may have the same
7.4 Micrometer, for measurement of specimen thickness. It texture or two different levels and types of texture. When
should be accurate to within 1 % of the average thickness of testing, special attention must be paid to how the specimen is
the specimens being tested. See Test Methods D 374 for positioned on the support.
descriptions of suitable micrometers. NOTE 4—As few as ten specimens often yield sufficiently reliable
7.5 The mass of the tup head and shaft assembly and the estimates of the mean-failure mass. However, in such cases the estimated
additional mass required must be known to within an accuracy standard deviation will be relatively large (1).
of 61 %. 11. Conditioning
8. Hazards 11.1 Unless otherwise specified, condition the test speci-
mens at 23 6 2°C (73.4 6 3.6°F) and 50 6 5 % relative
8.1 Safety Precautions:
humidity for not less than 40 h prior to test, in accordance with
8.1.1 Cushioning and shielding devices shall be provided to
Procedure A of Test Methods D 618, for those tests where
protect personnel and to avoid damage to the impinging surface
conditioning is required. In cases of disagreement, the toler-
of the tup. A tube or cage can contain the tup if it rebounds after
ances shall be 61°C (61.8°F) and 62 % relative humidity. For
striking a specimen.
compliance with ISO requirements, the specimens must be
8.1.2 When heavy weights are used, it is hazardous for an
conditioned for a minimum of 16 h prior to testing or post
operator to attempt to catch a rebounding tup. Figure 2 of Test
conditioning in accordance with ISO 291, unless the period of
Method D 2444 shows an effective mechanical “rebound
conditioning is stated in the relevant ISO specification for the
catcher” employed in conjunction with a drop tube.
material.
9. Sampling 11.1.1 Note that for some hygroscopic materials, such as
nylons, the material specifications (for example, Specification
9.1 Unless otherwise agreed upon between the manufacturer D 4066) call for testing “dry as-molded specimens”. Such
and the producer, sample the material in accordance with requirements take precedence over the above routine precon-
Sections 9 through 14 of Practice D 1898. ditioning to 50 % RH and require sealing the specimens in
water vapor-impermeable containers as soon as molded and not
10. Test Specimens
removing them until ready for testing.
10.1 Flat test specimens shall be large enough so that they 11.2 Conduct tests in the standard laboratory atmosphere of
can be clamped firmly if clamping is desirable. See Table 2 for 23 6 2°C (73.4 6 3.6°F) and at 50 6 5 % relative humidity,
the minimum size of specimen that can be used for each test unless otherwise specified.
geometry. 11.3 When testing is desired at temperatures other than
10.2 The thickness of any specimen in a sample shall not 23°C, transfer the materials to the desired test temperature
differ by more than 5 % from the average specimen thickness within 30 min, preferably immediately, after completion of the
of that sample. However, if variations greater than 5 % are preconditioning. Hold the specimens at the test temperature for
unavoidable in a sample that is obtained from parts, the sample no more than 5 h prior to test, and, in no case, for less than the
may be tested, but the data shall not be used for referee time required to ensure thermal equilibrium in accordance with
purposes. For compliance with ISO 6603-1 the test specimen Section 10 of Test Method D 618.
shall be 60 6 2 mm (2.4 6 0.08 in.) in diameter or 60 6 2 mm
(2.4 6 0.08 in.) square with a thickness of 2 6 0.1 mm (0.08 12. Procedure
6 0.004 in.). Machining specimens to reduce thickness varia- 12.1 Determine the number of specimens for each sample to
tion is not permissible. be tested, as specified in 10.3.

6
 D 5628 – 96 (2001)e1
12.2 Mark the specimens and condition as specified in 11.1. 12.14 Keep a running plot of the data, as shown in Appendix
12.3 Prepare the test apparatus for the geometry (FA, FB, X1. Use one symbol, such as X, to indicate a failure and a
FC, FD, FE) selected. different symbol, such as O, to indicate a non-failure at each
12.4 Measure and record the thickness of each specimen in mass or height level.
the area of impact. 12.15 For any specimen that gives a break behavior that
appears to be an outlier, the conditions of that impact shall be
12.5 Choose a specimen at random from the sample. A
examined. The specimen may be discarded only if a unique
random-numbers table may be used if desired. cause for the anomaly can be found, such as an internal flaw
12.6 Clamp or position the specimen. The same surface or visible in the broken specimen. Note that break behavior may
area should be the target each time (see 6.2). When clamping vary widely within a set of specimens. Data from specimens
is employed, the force should be sufficient to prevent motion of that show atypical behavior shall not be discarded simply on
the clamped portion of the specimen when the tup strikes. the basis of such behavior.
12.7 Unless otherwise specified, initially position the tup
0.660 6 0.008 m (26.0 6 0.3 in.) from the surface of the 13. Calculation
specimen. 13.1 Mean-Failure Mass—If a constant-height procedure
12.8 Adjust the total mass of the tup or the height of the tup, was used, calculate the mean-failure mass from the test data
or both, to that amount expected to cause half the specimens to obtained, as follows:
fail. w 5 wo 1 dw ~A/N 6 0.5! (1)
NOTE 5—If failures cannot be produced with the maximum available 13.2 Mean-Failure Height—If a constant-mass procedure
missile mass, the drop height can be increased. The test temperature could was used, calculate the mean-failure height from the test data
be reduced by (a) use of an ice-water mixture, or (b) by air-conditioned obtained, as follows:
environment to provide one of the temperatures given in 3.3 of Test
Methods D 618. Conversely, if the unloaded tup causes failures when h 5 ho 1 dh ~A/N 6 0.5! (2)
dropped 0.660 m, the drop height can be decreased. A moderate change in
dart velocity will not usually affect the mean-failure energy appreciably. where:
Refer to 5.1. w = mean-failure mass, kg,
h = mean-failure height, mm,
12.9 Release the tup. Be sure that it hits the center of the dw = increment of tup weight, kg,
specimen. If the tup bounces, catch it to prevent multiple dh = increment of tup height, mm,
impact damage to the specimen’s surface (see 8.1.2). N = total number of failures or non-failures, whichever is
12.10 Remove the specimen and examine it to determine smaller. For ease of notation, call whichever are used
whether or not it has failed. Permanent deformation alone is events,
not considered failure, but note the extent of such deformation wo = smallest mass at which an event occurred, kg
(depth, area). For some polymers, for example, glass- ho = lowest height at which an event occurred, mm (or
reinforced polyester, incipient cracking may be difficult to in.),
determine with the naked eye. Exposure of the stressed surface A = (ik5 0 ini,
to a penetrating dye, such as gentian violet, may be used to i = 0, 1, 2... k (counting index, starts at ho or wo ),
determine the onset of cracking. As a result of the wide range ni = number of events that occurred at hi or wi ,
wi = wo + idw , and
of failure types that may be observed with different materials, hi = ho + idh .
the definition of failure defined in the material specification
In calculating w or h, the negative sign is used when the
shall take precedence over the definition stated in 3.2.1. Other events are failures. The positive sign is used when the events
definitions of failure may be used if agreed upon by supplier are non-failures. Refer to the example in Appendix X1.
and user. 13.3 Mean-Failure Energy—Compute the mean-failure en-
12.11 If the first specimen fails, remove one increment of ergy as follows: MFE = hwf
mass from the tup while keeping the drop height constant, or
decrease the drop height while keeping the mass constant (see where:
12.12). If the first specimen does not fail, add one increment of MFE = mean-failure energy, J,
mass to the tup or increase the drop height one increment, as h = mean-failure height or constant height as appli-
above. Then test the second specimen. cable, mm
w = mean-failure mass or constant mass as applicable,
12.12 In this manner, select the impact height or mass for
kg, and
each test from the results observed with the specimen just f = factor for conversion to joules.
previously tested. Test each specimen only once. Use f = 9.80665 3 10−3 if h = mm and w = kg.
12.13 For best results, the mass or height increment used 13.4 Estimated Standard Deviation of the Sample—If de-
should be approximately equivalent to s, the estimated standard sired for record purposes, the estimated standard deviation of
deviation of the test for that sample. An increment of 0.5 to 2 the sample for either variable mass or variable height can be
times s is satisfactory (see section 13.4). calculated as follows:
NOTE 6—An increment of 10 % of the estimated mean-failure mass or sw 5 1.62dw @B/N 2 ~A/N!2 # 1 0.047dw or (3)
mean-failure height has been found to be acceptable in most instances.

7
 D 5628 – 96 (2001)e1
sh 5 1.62dh @B/N 2 ~A/N!2 # 1 0.047dh (4) TABLE 3 Precision, Method FB
Values Expressed as Percent
Material Mean, J of the Mean
where: Vr r
Polymethyl Methacrylate (PMMA) 0.35 12.6 35.7
sw = estimated standard deviation, mass, kg
Styrene–Butadiene (SB)A 9.26 18.7 52.9
sh = estimated standard deviation, height, mm, and Acrylonitrile–Butadiene–Styrene 11.8 14.9 42.2
(ABS)A
B 5 (ik 5 0 i2 ni (5)
A
Data generated in three laboratories.
Vr = within-laboratory coefficient of variation of the mean.
The above calculation is valid for [B/N − (A/N)2 ] > 0.3. If r = 2.83 Vr .
the value is <0.3, use Table I from Ref (3).
13.5 Estimated Standard Deviation of the Sample Mean—
Calculate the estimated standard deviation of the sample that penetrate the entire thickness (water would probably
mean-failure height or weight as follows: penetrate through the plaque), (c) brittle shatter (the plaque is
in several pieces after impact), or (d) ductile failure (the plaque
Sw̄ 5 Gsw / =N (6) is penetrated by a blunt tear). Report other observed deforma-
or tion due to impact, whether the specimens fail or not,
14.1.11 If atypical deformation for any specimen within a
Sh¯ 5 Gsh / =N (7)
sample for that material is observed, note the assignable cause,
where: if known,
sh̄ = estimated standard deviation of the mean height, mm, 14.1.12 Date of test and operator’s identification,
sw̄ = estimated standard deviation of the mean mass, kg, 14.1.13 Test temperature,
and 14.1.14 In no case shall results obtained with arbitrary
G = factor that is a function of s/d (see Appendix X2). geometries differing from those contained in these test methods
A sample computation of sw may be found in Appendix X1. be reported as values obtained by this test method (D 5628),
and
NOTE 7—For values of G at other levels of s/d, see Fig. 22 in Ref (4).
14.1.15 The test method number and published/revision
13.6 Estimated Standard Deviation of the Mean-Failure date.
Energy—Calculate the estimated standard deviation of the
mean-failure energy as follows: 15. Precision and Bias
SMFE 5 sh¯ w f (8) 15.1 Tables 3 and 4 are based on a round robin10 conducted
in 1972 involving three materials tested by six laboratories.
or Data from only four laboratories were used in calculating the
values in these tables. Each test result was the mean of multiple
SMFE 5 Sw̄ h f, as applicable (9)
individual determinations (Bruceton Staircase Procedure).
where: Each laboratory obtained one test result for a material.
SMFE = estimated standard deviation of the mean-failure NOTE 8—The number of laboratories participating in the 1972 round
energy. robin and the number of results collected do not meet the minimum
requirements of Practice E 691. Data in Tables 3 and 4 should be used
14. Report only for guidance, and not as a referee when there is a dispute between
14.1 Report the following information: users of this test method.
14.1.1 Complete identification of the sample tested, includ- 15.1.1 Polymethylmethacrylate (PMMA)—Specimens were
ing type of material, source, manufacturer’s code, form, cut from samples of 3.18-mm (0.125-in.) thickness extruded
principal dimensions, and previous history, sheet.
14.1.2 Method of preparation of specimens, 15.1.2 Styrene-Butadiene (SB)—Specimens were cut from
14.1.3 Whether surface of the specimen is smooth or samples of 2.54-mm (0.100-in.) thickness extruded sheet.
textured, the level of and type of texture if known, and whether 15.1.3 Acrylonitrile-Butadiene-Styrene (ABS)—Specimens
texture is on only one or both surfaces, were cut from samples of 2.64-mm (0.104-in.) thickness
14.1.4 If the specimen is textured, report whether textured extruded sheet.
surface faces upward towards the dart or downward away from
the dart, NOTE 9—Caution: The following explanations of r and R (15.2-15.2.3)
are only intended to present a meaningful way of considering the
14.1.5 Means of clamping, if any,
approximate precision of this test method. The data in Tables 3 and 4
14.1.6 Statement of geometry (FA, FB, FC, FD, FE) and should not be rigorously applied to acceptance or rejection of material, as
procedure used—constant mass or constant height, those data are specific to the round robin and may not be representative of
14.1.7 Thickness of specimens tested (average and range). other lots, conditions, materials, or laboratories. Users of this test method
14.1.8 Number of test specimens employed to determine the should apply the principles outlined in Practice E 691 to generate data
mean failure height or mass,
14.1.9 Mean-failure energy,
14.1.10 Types of failure, for example: (a) crack or cracks on 10
Supporting data are available from ASTM Headquarters.
one surface only (the plaque could still hold water), (b) cracks Request RR:D 20–1030.

8
 D 5628 – 96 (2001)e1
TABLE 4 Precision, Method FC 15.2.2 Reproducibility, R—In comparing two test results for
Values Expressed as Percent the same material obtained by different operators using differ-
Material Mean, J of the Mean ent equipment in different laboratories, reproducibility statis-
Vr r
Polymethyl Methacrylate (PMMA) 1.33 4.13 11.7 tics were not calculated because data from only four and three
Styrene–Butadiene (SB) 48.3 18.3 51.8 laboratories do not justify making these calculations.
Vr = within-laboratory coefficient of variation of the mean. 15.2.3 Any judgment in accordance with 15.2.1 would have
r = 2.83 Vr . an approximate 95 % (0.95) probability of being correct.
specific to their laboratory and materials, or between specific laboratories.
15.3 Bias—There are no recognized standards by which to
The principles of 15.2-15.2.3 would then be valid for such data. estimate bias of this test method.
15.2 Concept of r and R—If Vr and VR have been calculated 15.4 Efforts to form a task group to address between
from a large enough body of data, and for test results that were laboratory reproducibility of this test method has been unsuc-
means from testing multiple individual specimens (Bruceton cessful. Persons interested in participating in such a task group
Staircase Procedure), the following applies: should contact ASTM Headquarters.
15.2.1 Repeatability, r—In comparing two test results for
the same material obtained by the same operator using the 16. Keywords
same equipment on the same day, the two test results should be 16.1 dart impact; falling-mass impact; impact; impact resis-
judged not equivalent if they differ by more than the r value for tance; mean-failure energy; mean-failure height; mean-failure
that material. mass; rigid plastic; tup

APPENDIX

(Nonmandatory Information)

X1. SAMPLE CALCULATIONS

9
 D 5628 – 96 (2001)e1
TABLE X1.1 Values of G for Obtaining the Estimated Standard Deviation of the Mean
s/d 0.00 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09
0.40 1.18 1.175 1.17 1.16 1.155
0.50 1.15 1.145 1.14 1.135 1.13 1.125 1.12 1.11 1.105 1.10
0.60 1.095 1.09 1.085 1.08 1.075 1.07 1.07 1.065 1.06 1.06
0.70 1.055 1.055 1.05 1.05 1.045 1.04 1.04 1.035 1.035 1.03
0.80 1.03 1.025 1.025 1.02 1.02 1.02 1.015 1.015 1.015 1.01
0.90 1.01 1.01 1.005 1.005 1.005 1.00 1.00 1.00 0.995 0.995
1.00 0.995 0.99 0.99 0.99 0.985 0.985 0.985 0.985 0.98 0.98
1.10 0.98 0.98 0.98 0.975 0.975 0.975 0.975 0.975 0.975 0.97
1.20 0.97 0.97 0.97 0.97 0.97 0.97 0.965 0.965 0.965 0.965
1.30 0.965 0.965 0.965 0.965 0.96 0.96 0.96 0.96 0.96 0.96
1.40 0.96 0.96 0.96 0.955 0.955 0.955 0.955 0.955 0.955 0.955
1.50 0.955 0.955 0.955 0.95 0.95 0.95 0.95 0.95 0.95 0.95
1.60 0.95 0.95 0.95 0.95 0.945 0.945 0.945 0.945 0.945 0.945
1.70 0.945 0.945 0.945 0.945 0.945 0.945 0.94 0.94 0.94 0.94
1.80 0.94 0.94 0.94 0.94 0.94 0.94 0.94 0.94 0.94 0.935
1.90 0.935 0.935 0.935 0.935 0.935 0.935 0.935 0.935 0.935 0.935
2.00 0.935 0.935 0.935 0.93 0.93 0.93 0.93 0.93 0.93 0.93

REFERENCES

(1) Brownlee, K. A., Hodgest, J. L., Jr., and Rosenblatt, Murray, “The ods for Plastics Parts Used in Appliances, the Society of the Plastics
Up-and-Down Method with Small Samples,” American Statistical Industry, New York, NY, January 1965.
Association Journal, Vol 48, 1953, pp. 262–277. (4) Weaver, O. R., “Using Attributes to Measure a Continuous Variable in
(2) Hagan, R. S., Schmitz, J. V., and Davis, D. A., “Impact Testing of High Impact Testing Plastic Bottles,” Materials Research and Standards,
Impact Thermoplastic Sheet,” Technical Papers, 17th Annual Techni- MR & S, Vol 6, No. 6, June 1966, pp. 285–291.
cal Conference of SPE, SPPPB, Vol VIII, January 1961.
(3) “Test Method A—Falling Dart Impact, Proposed Method of Test for (5) Natrella, M. G., Experimental Statistics, National Bureau of Standards
Impact Resistance of Fabricated Plastics Parts,” Proposed Test Meth- Handbook 91, October 1966, pp. 10–22 and 10–23.

ASTM International 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 International 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, at the address shown below.

This standard is copyrighted by ASTM International, 100 Barr Harbor Drive, PO Box C700, West Conshohocken, PA 19428-2959,
United States. Individual reprints (single or multiple copies) of this standard may be obtained by contacting ASTM at the above
address or at 610-832-9585 (phone), 610-832-9555 (fax), or [email protected] (e-mail); or through the ASTM website
(www.astm.org).

10