This international standard was developed in accordance with internationally recognized principles on standardization established in the Decision on Principles

for the
Development of International Standards, Guides and Recommendations issued by the World Trade Organization Technical Barriers to Trade (TBT) Committee.

Designation: D2444 − 21

Standard Practice for

Determination of the Impact Resistance of Thermoplastic
Pipe and Fittings by Means of a Tup (Falling Weight)1
This standard is issued under the fixed designation D2444; 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 (´) indicates an editorial change since the last revision or reapproval.

1. Scope* F412 Terminology Relating to Plastic Piping Systems

1.1 This practice covers the determination of the impact F725 Practice for Drafting Impact Test Requirements In
resistance of thermoplastic pipe and fittings under specified Thermoplastic Pipe And Fittings Standards
conditions of impact by means of a tup (falling weight). Three 3. Terminology
interchangeable striking noses are used on the tup, differing in
geometrical configuration. Two specimen holders are de- 3.1 Definitions—Definitions are in accordance with Termi-
scribed. nology F412, and abbreviations are in accordance with Termi-
NOTE 1—Appendix X1 shows the procedure to determine impact nology D1600, unless otherwise specified.
strength.
3.2 Definitions of Terms Specific to This Standard:
1.2 The values stated in inch-pound units are to be regarded 3.2.1 tup—The striking part of a falling hammer like mecha-
as standard. The values given in parentheses are mathematical nism.
conversions to SI units that are provided for information only
and are not considered standard. 4. Significance and Use
1.3 This standard does not purport to address all of the 4.1 The impact resistance of thermoplastic pipe and fittings
safety concerns, if any, associated with its use. It is the relates to suitability for service and to quality of processing.
responsibility of the user of this standard to establish appro- Impact resistance may also provide a relative measure of a
priate safety, health, and environmental practices and deter- material’s resistance to breakage during handling and installa-
mine the applicability of regulatory limitations prior to use. tion and, for non-buried applications, to in-service breakage.
1.4 This international standard was developed in accor- See Appendix X5 for guidelines for selecting testing combi-
dance with internationally recognized principles on standard- nations.
ization established in the Decision on Principles for the 4.2 Results obtained by use of this practice can be used in
Development of International Standards, Guides and Recom- three ways:
mendations issued by the World Trade Organization Technical 4.2.1 As the basis for establishing impact test requirements
Barriers to Trade (TBT) Committee. in product standards,
4.2.2 To measure the effect of changes in materials or
2. Referenced Documents processing, and
2.1 ASTM Standards:2 4.2.3 To measure the effect of the environment.
D618 Practice for Conditioning Plastics for Testing
D1600 Terminology for Abbreviated Terms Relating to Plas- 5. Apparatus
tics 5.1 General—One type of impact tester is illustrated in Fig.
D2122 Test Method for Determining Dimensions of Ther- 1.
moplastic Pipe and Fittings
5.2 Tup:
5.2.1 The tup nose shall be as shown in Fig. 2. When used
1
with the 0.50 in. (12.7 mm) radius nose, it is designated as Tup
This practice is under the jurisdiction of ASTM Committee F17 on Plastic
Piping Systems and is the direct responsibility of Subcommittee F17.40 on Test A. When used with the 2.00 in. (51 mm) radius nose, it is
Methods. designated as Tup B. When used with the 0.25 in. (6.3 mm)
Current edition approved Feb. 1, 2021. Published February 2021. Originally radius nose, it is designated as Tup C.
approved in 1965. Last previous edition approved in 2019 as D2444 – 19. DOI:
10.1520/D2444-21. NOTE 2—It is suggested that tups be made of scratch-resistant steel to
2
For referenced ASTM standards, visit the ASTM website, www.astm.org, or reduce damage to the nose. Badly scarred noses may affect test results.
contact ASTM Customer Service at [email protected]. For Annual Book of ASTM
Standards volume information, refer to the standard’s Document Summary page on 5.2.2 The mass of the tup shall be 6, 12, 20, or 30 lb (2.7,
the ASTM website. 5.4, 9.1, or 13.6 kg).

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

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

1
Licensed to FRANCISCO LEMUS. ANSI store order # X_785830. Downloaded 02/10/2022. Single user license only. Copying and networking prohibited.
 D2444 − 21
5.3 Drop Tube:
5.3.1 The drop tube shall be of sufficient length (approxi-
mately 12 ft (3.7 m)) to provide for a fall of at least 10 ft (3.0
m) and shall be mounted so that the lengthwise direction is
vertical as measured with a plumb bob or a spirit level at least
2 ft (610 mm) in length.
5.3.2 Care must be taken to ensure that the tup falls freely;
it must not “chatter” down the tube.
NOTE 3—No material for the drop tube is specified. However, a
colddrawn seamless steel tubing with an inside diameter of 2.563 6 0.003
in. (65.10 6 0.08 mm), or acrylonitrile-butadiene-styrene (ABS) or
poly(vinyl chloride) (PVC) plastic pipe, 21⁄2 in. nominal pipe size, with a
standard dimension ratio (SDR) of 21, have been found to be satisfactory.
NOTE 4—Provided equivalent results are obtained, the tup may be
dropped without a drop tube or guided by other means. The drop tube is
used to reduce the hazard to operators and property that may occur when
the tup rebounds. It also helps guide and center the tup so that it will be
more likely to strike the top of the test specimen. It may also be necessary
to provide a protective barrier around the specimen, particularly for larger
sizes of pipe and fittings, to protect the operator from flying broken pieces.
NOTE 5—The tup may not fall freely if the clearance between the tup
and tube is too large or too small, or if it is restrained by a partial vacuum
above the tup, such as can be caused by the hold and release device.
5.3.3 Means shall be provided (1) to hold the tup at steps of
2 in. (50.8 mm) for a distance of 2 to 10 ft (610 mm to 3.0 m)
above the specimen holder, (2) to release the tup in a
reproducible manner, (3) to allow the tup to fall freely, and (4)
to catch the tup on the first rebound. Refer to Fig. 1.
FIG. 1 One Type of Tup Impact Tester 5.4 Specimen Holder—Two specimen holders are described:
the V-block (Holder A) and the flat-plate (Holder B).
5.4.1 The V-block holder shall be at least as long as the
specimen being tested and shall have a 90° included angle. It
may be fabricated or of solid construction. The side supports
shall be of sufficient depth to support the specimen in the V and
not on the top edges of the V-block.
NOTE 6—Both aluminum and steel have been found suitable for the
holders.
5.4.2 The flat-plate holder shall consist of a 1 in. (25.4 mm)
minimum thick plate at least as long as the specimen being
tested, in which a groove to position the pipe specimen shall be
cut. This groove shall be about 0.12 in. (3 mm) in depth with
the edges rounded to a radius of about 0.06 in. (1.5 mm).
Fittings that do not contact the plate directly under the point of
impact shall be supported at this point by a flat steel plate or
shim. The specimen holder shall be mounted on a rigid base
fastened to a concrete slab. Means shall be provided to center
the specimens under the drop tube. A bar or rod placed inside
the specimen and retained by a light spring may be employed
if difficulty is encountered in holding the specimen in position.
6. Test Specimens
6.1 The pipe specimens shall be equal in length to the
nominal outside diameter but not less than 6 in. (152 mm) in
length. Burrs shall be removed.
6.2 Fittings shall be tested either unassembled or assembled
to pieces of pipe each 6 in. (152 mm) in length.
7. Number of Test Specimens
NOTE 1—Dimensions are in inches (millimetres).
FIG. 2 Tup Nose Detail 7.1 The number of test specimens shall be as specified in the
product standard. If the number of test specimens are not

2
Licensed to FRANCISCO LEMUS. ANSI store order # X_785830. Downloaded 02/10/2022. Single user license only. Copying and networking prohibited.
 D2444 − 21
specified in the product standard, the following shall apply; test 10.2.2.2 If all four specimens fail, test four more specimens
20 sample specimens. When 4 or more of 20 specimens tested with a lighter tup.
fail, test 20 additional specimens. 17 or more of 20 specimens 10.2.2.3 If all four specimens pass, test the same specimens
passing or 32 or more of 40 specimens passing shall indicate an with a heavier tup, at a drop height of 5 ft (1.5 m).
acceptable pass rate for the test. 10.2.2.4 Once a weight has been used that results in the
NOTE 7—See F725 Practice for Drafting Impact Test Requirements in failure of some, but not all, specimens at a height of 5 ft (1.5
Thermoplastic Pipe and Fittings Standards. m), continue the test as described in 10.2.3. If no failures occur
with the heaviest tup at a drop height of 5 ft (1.5 m), test at the
8. Conditioning
maximum drop height. If no failures occur with the maximum
8.1 Condition the test specimens at 40 h, 73 °F (23 °C), and weight at the maximum drop height, test a total of 50
50 6 5 % relative humidity for not less than 40 h prior to test specimens at this drop height, record the results, and discon-
in accordance with Procedure A of Practice D618, for those tinue testing. If one or more of the preliminary test specimens
tests where conditioning is required. passes with the maximum weight at the maximum drop height,
8.2 Test Conditions—Conduct the test in a room maintained proceed to 10.2.3.
at the test temperature or, if the test is made in an atmosphere 10.2.3 After the optimum tup weight has been determined,
or at a temperature other than that at which the specimens are divide the remaining group of preliminary test specimens into
conditioned, conduct the test as soon as possible after removal two equal lots. Each group shall contain at least eight speci-
from the conditioning atmosphere, but in any case within 15 s. mens. Test one lot at a lesser drop height so that substantially
In case of disagreement, conduct the tests in a room maintained less than 50 % failures result. Test the second lot at a greater
at the test temperature. drop height so that substantially more than 50 % failures result.
8.3 The test conditions shall be as specified in the product Plot these results on probability graph paper to serve as the
standard. basis for estimating the proper drop heights for the final test.
10.2.4 If the 6 lb (2.7 kg) tup produces 100 % failures at 5
9. Procedure—General ft (1.5 m), use lesser drop heights for the procedures outlined
9.1 Measure the dimensions of the test specimens in accor- in 10.2.3.
dance with Test Method D2122. NOTE 10—Because there must be both failures and nonfailures in this
9.2 The point of impact for all specimens shall be at the top test, conditions that result in 100 % failures or 100 % passes are of limited
of the vertical diameter. value in the development of information. When all specimens pass, the
problem can be helped by (1) changing from the flat-plate holder to the
9.3 Test unsymmetrical fittings with the specimen lying on V-block holder (in the case of pipe specimens), (2) changing from Tup B
its side. to Tup A, and finally to Tup C, and (3) lowering the test temperature.
When all specimens fail, the sequence is reversed.
9.4 Position the pipe specimens at random angular orienta-
tions. 10.3 Final Test:
10.3.1 Divide 100 test specimens into two equal sets.
9.5 Impact each specimen only once.
10.3.2 On the basis of the preliminary test, trial and error, or
9.6 Conduct the test and record the results as described in judgment, estimate the drop height at which 85 % of the
Sections 10 and 11. specimens will pass and test the first set of 50 specimens at this
10. Procedure—Details height. Record the value of the mass, the drop height, and the
number of passes.
10.1 Impact one of the sets using an energy estimated to
10.3.3 Estimate the drop height at which 85 % of the
cause 15 % of the specimens to fail; impact the second set at an
specimens will fail and, if it is 10 ft (3.0 m) or more, use this
energy estimated to cause 85 % to fail. Record the actual
height. Test 50 specimens and record the number of passes and
percentage that fail and plot on normal probability graph paper.
the drop height.
10.2 Preliminary Tests: 10.3.4 Acceptable Data:
10.2.1 The purpose of the preliminary tests is to identify the
10.3.4.1 To be acceptable, there must be at least one pass
optimum tup mass and to determine the drop heights to be used
and at least one fail in each set of data; one set of data should
for the final tests.
list less than 50 % failures; the number of passes in the two sets
NOTE 8—The use of this preliminary procedure is optional. Trial and should differ by at least 20.
error, previous experience with the material, or guesswork may serve as
10.3.4.2 If the requirements of 10.3.4.1 have not been met,
well. Also, see Appendix X4.
NOTE 9—The chief source of problems in conducting impact tests is the test 50 additional specimens at a drop height chosen to rectify
ratio of the standard deviation to the mean strength. Use of Tup C with the deficiency.
pipe specimens will reduce the size of the problem. 10.3.5 Construct a straight-line plot of the test results on
10.2.2 Select a tup estimated to cause failure of some normal probability paper. Preliminary tests which were con-
specimens when dropped from a height of 5 ft (1.5 m) and use ducted at the same drop height as the final tests shall be used
it to test four specimens. to compute the percent passing at that height. All other
10.2.2.1 If at least one but not all four specimens fail with preliminary test results are ignored for purposes of obtaining
the first tup selected, continue the test as described in 10.2.3. the final plot.

3
Licensed to FRANCISCO LEMUS. ANSI store order # X_785830. Downloaded 02/10/2022. Single user license only. Copying and networking prohibited.
 D2444 − 21
10.3.6 The intercept of the plot with the 50 % pass/fail line 12.1.3 Test temperature and conditioning procedure,
is the average impact resistance of the set as measured by the 12.1.4 Tup used,
test. 12.1.5 Holder used,
10.3.7 The intercept of the plot at the 16 % or the 84 %
pass/fail line differs from the average strength by (for practical 12.1.6 The mass of the tup, lb (kg),
purposes) one standard deviation. 12.1.7 Point of impact for fittings,
12.1.8 Types of failure and any deformation observed, each
11. Criteria for Specimen Failure
specimen tested should be recorded as “pass” or “fail” in
11.1 Failure in the test specimens shall be shattering or any accordance with the description in Section 10.
crack or split created by the impact and that can be seen by the 12.1.9 Date of test, and
naked eye. Lighting devices may be used to assist in examining
the specimens. A crease visible on the surface shall not be 13. Precision and Bias
construed as specimen failure. If criteria of failure in the test
specimens other than those cited here are used, they shall be 13.1 This practice is neither precise nor accurate within the
listed in the report. limits usually associated with those terms unless larger num-
bers of test specimens are employed. Therefore, no statement
12. Report
of bias can be made.
12.1 The report shall include the following:
12.1.1 Complete identification of pipe or fittings tested, NOTE 11—A brief treatment of the subject of precision and accuracy for
binomial tests is provided in Appendix X2.
including type of plastic, source, manufacturer’s code, size,
average dimensions and minimum wall thickness, and history,
12.1.2 In the case of fittings, the dimensions of pipe used to 14. Keywords
prepare the specimens, how joints were made, and the position 14.1 fittings; flat-plate; impact resistance; groove; pipe;
of the weld mark in relation to the point of impact, thermoplastic; thickness; tup (falling weight)

APPENDIXES

(Nonmandatory Information)

X1. DETERMINATION OF IMPACT STRENGTH

X1.1 Impact one of the sets using an energy estimated to m), continue the test as described in X1.2.3. If no failures occur
cause 15 % of the specimens to fail; impact the second set at an with the heaviest tup at a drop height of 5 ft (1.5 m), test at the
energy estimated to cause 85 % to fail. Record the actual maximum drop height. If no failures occur with the maximum
percentage that fail and plot on normal probability graph paper. weight at the maximum drop height, test a total of 50
specimens at this drop height, record the results, and discon-
X1.2 Preliminary Tests: tinue testing. If one or more of the preliminary test specimens
X1.2.1 The purpose of the preliminary tests is to identify the passes with the maximum weight at the maximum drop height,
optimum tup mass and to determine the drop heights to be used proceed to X1.2.3.
for the final tests. X1.2.3 After the optimum tup weight has been determined,
NOTE X1.1—The use of this preliminary practice is optional. Trial and divide the remaining group of preliminary test specimens into
error, previous experience with the material, or professional judgement two equal lots. Each group shall contain at least eight speci-
may serve as well. Also, see Appendix X3. mens. Test one lot at a lesser drop height so that substantially
NOTE X1.2—The chief source of problems in conducting impact tests is less than 50 % failures result. Test the second lot at a greater
the ratio of the standard deviation to the mean strength. Use of Tup C with
pipe specimens will reduce the size of the problem.
drop height so that substantially more than 50 % failures result.
Plotthese results on probability graph paper to serve as the
X1.2.2 Select a tup estimated to cause failure of some basis for estimating the proper drop heights for the final test.
specimens when dropped from a height of 5 ft (1.5 m) and use
it to test four specimens. X1.2.4 If the 6 lb (2.7 kg) tup produces 100 % failures at 5
X1.2.2.1 If at least one but not all four specimens fail with ft (1.5 m), use lesser drop heights for the practices outlined in
the first tup selected, continue the test as described in X1.2.3. X1.2.3.
X1.2.2.2 If all four specimens fail, test four more specimens NOTE X1.3—Because there must be both failures and nonfailures in this
with a lighter tup. test, conditions that result in 100 % failures or 100 % passes are of limited
X1.2.2.3 If all four specimens pass, test the same specimens value in the development of information. When all specimens pass, the
problem can be helped by (1) changing from the flat-plate holder to the
with a heavier tup, at a drop height of 5 ft (1.5 m).
V-block holder (in the case of pipe specimens), (2) changing from Tup B
X1.2.2.4 Once a weight has been used that results in the to Tup A, and finally to Tup C, and (3) lowering the test temperature.
failure of some, but not all, specimens at a height of 5 ft (1.5 When all specimens fail, the sequence is reversed.

4
Licensed to FRANCISCO LEMUS. ANSI store order # X_785830. Downloaded 02/10/2022. Single user license only. Copying and networking prohibited.
 D2444 − 21
X1.3 Final Test: X1.3.4.2 If the requirements of X1.3.4.1 have not been met,
X1.3.1 Divide 100 test specimens into two equal sets. test 50 additional specimens at a drop height chosen to rectify
the deficiency.
X1.3.2 On the basis of the preliminary test, trial and error,
or judgment, estimate the drop height at which 85 % of the X1.3.5 Construct a straight-line plot of the test results on
specimens will pass and test the first set of 50 specimens at this normal probability paper. Preliminary tests which were con-
height. Record the value of the mass, the drop height, and the ducted at the same drop height as the final tests should be used
number of passes. to compute the percent passing at that height. All other
X1.3.3 Estimate the drop height at which 85 % of the preliminary test results are ignored for purposes of obtaining
specimens will fail and, if it is 10 ft (3.0 m) or more, use this the final plot.
height. Test 50 specimens and record the number of passes and X1.3.6 The intercept of the plot with the 50 % pass ⁄fail line
the drop height. is the average impact resistance of the set as measured by the
X1.3.4 Acceptable Data: test.
X1.3.4.1 To be acceptable, there should be at least one pass
and at least one fail in each set of data; one set of data should X1.3.7 The intercept of the plot at the 16 % or the 84 %
list less than 50 % failures; the number of passes in the two sets pass/fail line differs from the average strength by (for practical
should differ by at least 20. purposes) one standard deviation.

X2. GUIDELINES FOR DETERMINING PRECISION AND ACCURACY BY APPLYING BINOMIAL PROBABILITY TECH-
NIQUES

Tests that have only two possible results (pass or fail, heads
or tails, black or white) are governed by the rules of probabil-
ity.
s5 Πp ~ 1 2 p!
N
(X2.3)

where:
X2.2 The most elementary application of the rules occurs s = standard error,
when the true probability of success is known. When a fair coin N = number of specimens tested, and
is tossed, for example, the probability for “heads” is known to p = measured rate of success.
be exactly 0.5 chance out of 1.

X2.3 The probability,rPn, for exactly r successes in n tries Example—50 specimens are tested and 32 pass. Then the
when the probability for success, p, is known for each try, is best estimate for p (true probability of success) is 0.64, and the
given by the equation standard error equals:

r Pn 5
n!
r! ~ n 2 r ! !
~ p r ! ~ 1 2 p ! n2r (X2.1) Π0.64 ~ 1 2 0.64!
50
5 0.07 (X2.4)
The confidence level is 68.27 % (note the similarity to plus
Example — The chance that one will obtain exactly 25 heads and minus one standard deviation); the true probability of
when a fair coin is tossed 50 times is equal to success is in the range 0.64 6 0.07, or from 0.57 to 0.71.
50!
25! ~ 50 2 25! !
~ 0.525! or 0.112 (X2.2) X2.7 The 90 % confidence interval is equal to the experi-
The factorial of zero (0) is one (1). mental value p plus and minus 1.65 standard errors, and the
95 % confidence interval is equal to the experimental value
X2.4 The total of the probabilities for all possible results is plus and minus 1.96 standard errors. Note that the foregoing
equal to 1. The equation is mathematically exact. deals with error due to probability only and does not take
account of error caused by equipment or experimental tech-
nique.
X2.8 Charts defining the confidence limits for binomial test
X2.5 A complication exists when the true probability of data are listed in many reference works. Dixon and Massey3
success is not known, but must be deduced on the basis of provide charts for 80, 90, 95, and 99 % confidence coefficients.
binomial test data. This is the case when thermoplastic pipe and X2.8.1 Use of Fig. X2.1—If 100 specimens are tested, and
fittings are subjected to impact testing as outlined in this 24 pass, calculate p = 0.24. Find the ordinate labeled 0.24 at the
practice. bottom of the chart and read up. At the intersection with the
first curved line labeled “100”, move to the left-hand side of
X2.6 The binomial probability equation can be employed to
define the likely limits of error for binomial test data. It will be
simpler to refer to a statistics text or to use the approximate 3
Dixon, D. W., and Massey, Jr., F. J., “Introduction to Statistical Analysis,” 3rd
equation as follows: ed., McGraw Hill Book Co., Inc. New York, NY, 1969, pp. 501–504.

5
Licensed to FRANCISCO LEMUS. ANSI store order # X_785830. Downloaded 02/10/2022. Single user license only. Copying and networking prohibited.
 D2444 − 21

NOTE 1—Reproduced by permission of the Biometrika Trust.

NOTE 2—This figure provides confidence limits for p' in binomial sampling, given a sample fraction. Confidence coefficient = 0.95. The numbers
printed along the curves indicate the sample size n. If for a given value of the abscissa, PA and PB are the ordinates read from (or interpolated between)
the appropriate lower and upper curves, then Pr {pA ≤ p' ≤ pB} ≥ 0.95.
FIG. X2.1 95 % Confidence Limits for Binomial Test Data

the chart and read p' = 0.16. Return to the ordinate labeled 0.24 (0 %) or one (100 %), then it requires more testing than when
and find the intersection with the upper curved line labeled the failure rate is somewhat removed from these points.
“100” and read “0.33” at the left-hand side of the chart. The
probability of success (p') is identified as 0.16 to 0.33, to a X2.9.3 The use of only two data points presumes that the
95 % confidence level. distribution of attributes is normal. When good information is
required at some specific point (for example, the place where
X2.9 Improving Accuracy: 90 % of all specimens should pass), then the two-point plot can
X2.9.1 This practice provides less accuracy than may be be used to define the new test level most likely to be correct,
required for some purposes. and a third series of tests run at that level to check the accuracy
of the preliminary plot at that point.
X2.9.2 Analysis of the material represented by Fig. X2.1
reveals that accuracy is related to the square root of the number X2.9.4 Finally, a series of tests can be conducted at different
of specimens subjected to test. To reduce the size of error by a energy levels when it is desired to define the true distribution
factor of 2 requires that four times as many specimens be of attributes over the range p = 0 to p = 1.
tested. Note, also, that when the failure rate is close to zero

6
Licensed to FRANCISCO LEMUS. ANSI store order # X_785830. Downloaded 02/10/2022. Single user license only. Copying and networking prohibited.
 D2444 − 21

X3. PROBABILITY GRAPH PAPER

X3.1 Probability graph paper is available in a variety of if the two data points are close to each other (for example, 40
forms. One of the most common is illustrated in Fig. X3.1. and 52 % passing), then errors in either will affect the slope of
the plot seriously. This is the reason for the requirements listed
X3.2 The vertical grid is evenly spaced, whereas the hori- in 10.3.4.1. The mean impact resistance of the lot is defined by
zontal grid spacing expands in both directions from the line
the intersection of the plot with the line labeled “50 %.” To find
labeled “50”. The spacing is such that when statistical data are
the standard deviation, locate the intersection of the plot with
plotted, and when the distribution of attributes is normal, the
the line for 16 % or for 84 %. The difference between either of
plot will be a straight line. Numbers printed on the sides of the
graph will typically read “0.01, 0.1, 0.2, 0.5, 1, 2, ... 40, 50, 60, these points and the 50 % point is equal to one standard
... 99, 99.9, 99.99.” These numbers refer to the percent passing deviation. Since there is no assurance that the properties will be
or failing at a given test level. The user labels the vertical lines. distributed completely normally (particularly true for yard-
Ordinarily, equal increments will be used to label these lines. If aged pipe exposed to differing amounts of ultraviolet light), the
it is known, for example, that all data of interest will fall in the reliability will fall off in the regions where few pass or few fail.
region encompassed by 60 ft·lbf and 160 ft·lbf, the lines would The plot suffices, nevertheless, to give a good general descrip-
be labeled “60, 70, 80, 90, ... 150, and 160.” Two data points tion of the properties of the pipe.
suffice to define the plot if the distribution is normal. Note that

7
Licensed to FRANCISCO LEMUS. ANSI store order # X_785830. Downloaded 02/10/2022. Single user license only. Copying and networking prohibited.
 D2444 − 21

FIG. X3.1 Probability Graph

X4. “UP/DOWN” TEST PROCEDURE TO DEFINE THE MEAN IMPACT RESISTANCE

X4.1 Number of Test Specimens X4.2.2 Test Conditions—Tests at lower or elevated tempera-
X4.1.1 Determination of Median Value: tures shall be made in a room maintained at the test
X4.1.1.1 Six or more specimens shall be used to determine temperature, or if the test is made in an atmosphere at a
the approximate starting point of the test. Once this is done for temperature other than that at which the specimens are
a specific size and type of pipe, it may not be necessary to conditioned, the test shall be made as soon as possible on
repeat this preliminary step. removal from the conditioning atmosphere, but in any case
X4.1.1.2 The median impact resistance shall be determined within 15 s after removal from the conditioning atmosphere. In
with 20 test specimens of each sample. The specimens shall be case of disagreement, the tests shall be conducted in a room
labeled from 1 to 20 and the order of testing determined by a maintained at the test temperature.
set of random numbers.
X4.3 Procedure—General
X4.1.2 For specification purposes, the number of specimens
X4.3.1 Measure the dimensions of the test specimens in
tested shall be in accordance with Section X4.5.
accordance with Test Method D2122. Note the position of the
X4.2 Conditioning thinnest wall of each specimen.
X4.2.1 Conditioning—Condition the test specimens at 73.4 X4.3.2 Mount the first test specimen with the thinnest wall
6 3.6 °F (23 6 2 °C) and 50 6 5 % relative humidity for not section on top so that the tup strikes this area. Mount the
less than 40 h prior to test in accordance with Procedure A of second specimen so that the point of impact is 18° from the
Practice D618, for those tests where conditioning is required. thinnest wall, the third 36°, etc.

8
Licensed to FRANCISCO LEMUS. ANSI store order # X_785830. Downloaded 02/10/2022. Single user license only. Copying and networking prohibited.
 D2444 − 21
X4.3.3 Test unsymmetrical fittings by laying the specimen X4.5.3 For unsymmetrical fittings, six specimens shall be
on its side and impacting the other side. Strike six specimens struck on one side and six on the other. When two or more of
on one side and six on the other. the specimens fail, the product does not meet the specified
X4.3.4 The point of impact for all specimens shall be at the impact requirement. Eleven nonfailures out of twelve speci-
top of the vertical diameter. mens tested shall be acceptable.
X4.3.5 Determine the impact resistance (energy) by the“ Up X4.6 Definition of Failure
and Down” (hereafter called UD) method.4 In this method the
X4.6.1 Failure in the test specimens shall be shattering or
energy required to break the specimens is bracketed by making
any crack or split on the inside or outside created by the impact
the tests at different heights of drop in a systematic manner. No
and that can be seen by the naked eye. Lighting devices shall
specimen is tested more than once in this method.
be used to assist in examining for cracks and splits in the walls
X4.4 Procedure—Details of the pipe or fittings specimens. Any permanent deformation
of the specimens shall be reported with a short description of
X4.4.1 The weight (mass) of the tup shall be 6, 12, 20, or 30
the deformation. If other criteria of failure than those cited here
lb (2.7, 5.4, 9.1, or 13.6 kg).
are used they shall be given in the report.
X4.4.2 Make preliminary tests to determine the approxi-
mate height of fall of the tup required to produce failure in the X4.7 Calculations
pipe or fittings made of the particular material and in the
X4.7.1 Calculate the median height of fall (h) and standard
particular dimensions to be tested.
deviation (s) from the test data obtained as follows:
NOTE X4.1—It has been found desirable to use increments of at least 6 h 5 h o 1d @ ~ A/N ! 6 ~ 1/2 ! # (X4.1)
in. (152 mm) for this initial testing.
X4.4.3 Test the first specimen, chosen by a set of random
numbers as indicated in X4.1.2, by dropping the tup onto it
s 5 1.620d S
NB 2 A 2
N2
10.029 D (X4.2)

from the 2-in. (50.8-mm) or 6-in. (152-mm) increment imme- where:

diately below the approximate height determined in X4.4.2. If h = median height of fall, ft (m),
this specimen ruptures or cracks, test the second specimen at d = increment in height of drop, ft or m, (1⁄6 for 2-in. or 1⁄2
the next lower increment. If the first specimen does not fail, test for 6-in. (50.8 or 152 mm) increment),
the second one at the next higher increment. In a similar s = standard deviation for h,
manner, determine the height of fall of the tup to test each N = lesser number, either failures or nonfailure,
specimen from the results observed with the specimen tested h0 = lowest height at which any one of N occurs, ft (m),
immediately prior. A = k

X4.5 Procedure—Specification Requirement

(i n
i50
i

X4.5.1 For use in determining whether or not a sample where h0, h1, h2, h3, etc. are the heights in progressive order
meets the impact requirement in pipe or fitting specifications, it of magnitude in the range of heights where N occurs: i is the
is recommended that specimens be tested on a “go, no-go” subscript of h, and ni is the number of failures or nonfailures,
basis. The tup, the weight, and the height from which the tup whichever is pertinent, at the level of i, and
is dropped shall be specified in the product specification. B = k

X4.5.2 For pipe or symmetrical fittings, ten specimens shall (i n .

i50
2
i
be tested. When two or more specimens fail, the product does
not meet the impact requirement. Nine nonfailures out of ten X4.7.2 The negative sign is used in the parentheses to
specimens tested shall be acceptable. calculate h when N refers to failures, and the positive sign
when N is the number of nonfailures. Calculate the median
4
impact resistance by multiplying h by the weight used in
This system of testing, nomenclature, and calculations is described by Dixon,
W. J., and Massey, F. J., Jr., Introduction to Statistical Analysis, Chapter 19, Third pounds. Calculate the standard deviation by multiplying s by
Edition 1969, McGraw Hill Book Co. Inc., New York, NY. the weight used in pounds.

9
Licensed to FRANCISCO LEMUS. ANSI store order # X_785830. Downloaded 02/10/2022. Single user license only. Copying and networking prohibited.
 D2444 − 21

X5. GUIDELINES FOR SELECTION OF TESTING COMBINATIONS

X5.1 It is recognized that the impact resistance of thermo- X5.1.2 Use the tup which is most representative of the
plastic pipe and fittings is a function of the material in the service concerned.
product, of the quality of the extrusion or molding, of the X5.1.3 Use the tup and weight combination which will
geometry and dimensions of the test specimen and of the produce failure within a drop height between 2 and 10 ft (0.6
testing variables used (temperature, tup nose geometry, tup and 3.0 m).
weight, specimen support, and height of drop). Because all
X5.1.4 If no combination of tup-weight-height-support is
these factors and their interrelations affect the test results,
available that will give satisfactory results because of high-
impact tests are useful mainly for quality control and compara- impact resistance, lower the test temperature of the test
tive purposes, not as simulated service tests. The combinations specimen by use of an ice-water mixture or by an air-
of test variables should be selected so that the test can be easily conditioned environment to an even lower temperature.
run with little or no hazard to personnel and the results
obtained are reasonably reproducible. To achieve this end, the X5.1.5 The impact resistance of a specimen increases when
following guidelines for selection of the testing variables are (a) the temperature is increased; (b) the tup nose geometry
changes in order to C, A, B; (c) the specimen support is
suggested:
changed from the V-block to a flat plate; (d) the tup weight is
X5.1.1 Use the tup which will give the best reproducibility. decreased; and (e) the height drop is decreased.

SUMMARY OF CHANGES

Committee F17 has identified the location of selected changes to this standard since the last issue (D2444 – 19)
that may impact the use of this standard.

(1) Section 2.1 revised. (4) Note 7 added.

(2) Section 5.4.2 revised. (5) Section 14 revised.
(3) Section 7.1 revised.

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). Permission rights to photocopy the standard may also be secured from the Copyright Clearance Center, 222
Rosewood Drive, Danvers, MA 01923, Tel: (978) 646-2600; http://www.copyright.com/

10
Licensed to FRANCISCO LEMUS. ANSI store order # X_785830. Downloaded 02/10/2022. Single user license only. Copying and networking prohibited.