An American National Standard

Designation: D 2992 – 01

Standard Practice for

Obtaining Hydrostatic or Pressure Design Basis for
“Fiberglass” (Glass-Fiber-Reinforced Thermosetting-Resin)
Pipe and Fittings1, 2
This standard is issued under the fixed designation D 2992; 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* practice may not be applicable for evaluating stresses induced

1.1 This practice establishes two procedures, Procedure A by loadings where the longitudinal stress exceeds 50 % of the
(cyclic) and Procedure B (static), for obtaining a hydrostatic HDS.
design basis (HDB) or a pressure design basis (PDB) for 1.5 The values stated in inch-pound units are to be regarded
fiberglass piping products, by evaluating strength-regression as the standard. The values in parentheses are given for
data derived from testing pipe or fittings, or both, of the same information purposes only.
materials and construction, either separately or in assemblies. NOTE 3—There is no similar or equivalent ISO standard.
Both glass-fiber-reinforced thermosetting-resin pipe (RTRP)
1.6 This standard does not purport to address all of the
and glass-fiber-reinforced polymer mortar pipe (RPMP) are
safety concerns, if any, associated with its use. It is the
fiberglass pipe.
responsibility of the user of this standard to establish appro-
NOTE 1—For the purposes of this standard, polymer does not include priate safety and health practices and determine the applica-
natural polymers. bility of regulatory limitations prior to use.
1.2 This practice can be used for the HDB determination for
fiberglass pipe where the ratio of outside diameter to wall 2. Referenced Documents
thickness is 10:1 or more. 2.1 ASTM Standards:
D 618 Practice for Conditioning Plastics for Testing3
NOTE 2—This limitation, based on thin-wall pipe design theory, serves
D 883 Terminology Relating to Plastics3
further to limit the application of this practice to internal pressures which,
by the hoop-stress equation, are approximately 20 % of the derived D 1598 Test Method for Time-to-Failure of Plastic Pipe
hydrostatic design stress (HDS). For example, if HDS is 5000 psi (34 500 Under Constant Internal Pressure4
kPa), the pipe is limited to about 1000-psig (6900-kPa) internal pressure, D 1599 Test Method for Short-Time Hydraulic Failure Pres-
regardless of diameter. sure of Plastic Pipe, Tubing, and Fittings4
1.3 This practice provides a PDB for complex-shaped prod- D 1600 Terminology for Abbreviated Terms Relating to
ucts or systems where complex stress fields seriously inhibit Plastics3
the use of hoop stress. D 2143 Test Method for Cyclic Pressure Strength of Rein-
1.4 Specimen end closures in the underlying test methods forced Thermosetting Plastic Pipe4
may be either restrained or free, leading to certain limitations. D 3567 Practice for Determining Dimensions of “Fiber-
1.4.1 Restrained Ends—Specimens are stressed by internal glass” (Glass–Fiber–Reinforced Thermosetting Resin)
pressure only in the hoop direction, and the HDB is applicable Pipe and Fittings4
for stresses developed only in the hoop direction. F 412 Terminology Relating to Plastic Piping Systems4
1.4.2 Free Ends—Specimens are stressed by internal pres- F 948 Test Method for Time-to-Failure of Plastic Piping
sure in both hoop and longitudinal directions, such that the Systems and Components Under Constant Internal Pres-
hoop stress is twice as large as the longitudinal stress. This sure with Flow4
2.2 ISO Standard:
1
This practice is under the jurisdiction of ASTM Committee D20 on Plastics and
is the direct responsibility of Subcommittee D20.23 on Reinforced Plastic Piping 3
Annual Book of ASTM Standards, Vol 08.01.
Systems and Chemical Equipment. 4
Annual Book of ASTM Standards, Vol 08.04.
Current edition approved June 10, 2001. Published August 10, 2001. Originally
published as D 2992 – 71. Last previous edition D 2992 – 96.
2
This revision incorporates a change in the data-analysis procedure.

*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
Copyright by ASTM Int'l (all rights reserved); Thu Sep 14 08:44:05 EDT 2006
Reproduction authorized per License Agreement with Jorge E. Araujo-Osorio (I.P.N.-E.S.I.Q.I.E. IPN811229H26);
 D 2992 – 01
3 Preferred Numbers—Series of Preferred Numbers5 piping component with a high degree of certainty that failure of
the component will not occur.
3. Terminology 3.1.11 hydrostatic design stress (HDS)—the estimated
3.1 Definitions: maximum tensile stress in the wall of the pipe in the hoop
3.1.1 General—Definitions are in accordance with Termi- direction due to internal hydrostatic pressure that can be
nologies D 883 and F 412, and abbreviations are in accordance applied cyclically (Procedure A) or continuously (Procedure B)
with Terminology D 1600, unless otherwise indicated. with a high degree of certainty that failure of the pipe will not
3.1.2 closure, free-end—a sealing device or mechanism occur.
fastened to the end of the test specimen so that internal pressure 3.1.12 long-term hydrostatic strength (LTHS)— the esti-
produces longitudinal tensile stresses in addition to hoop and
mated tensile stress in the wall of the pipe in the hoop direction
radial stresses in the test specimen.
due to internal hydrostatic pressure that, when applied cycli-
3.1.3 closure, restrained-end—a sealing device or mecha-
cally, will cause failure of the pipe after a specified number of
nism which relies on a rod through the test specimen or an
cycles by Procedure A or a specified number of hours by
external structure to resist the end thrust produced by internal
Procedure B.
pressure, thereby limiting the stresses in (straight) specimens to
the hoop and radial directions only. NOTE 6—The time for determination of LTHS or LTHP is specified by
3.1.4 failure—the transmission of the test fluid through the the product standard. Typically, the time is 150 3 106 or 657 3 106 cycles
body of the specimen in any manner, whether it be a wall for Procedure A and 100 000 or 438 000 h for Procedure B.
fracture, localized leaking, or weeping at a distance greater
3.1.13 long-term hydrostatic pressure (LTHP)—the esti-
than one diameter from the end closure.
mated internal pressure of the piping product that, when
NOTE 4—For this practice, specimens which have not failed may be applied cyclically, will cause failure of the product after a
included as failures under the specific conditions given in 6.3, 9.3, and specified number of cycles by Procedure A or a specified
12.2.
number of hours by Procedure B.
3.1.5 fiberglass pipe—a tubular product containing glass 3.1.14 pressure design basis (PDB)—an internal pressure
fiber reinforcement embedded in or surrounded by cured developed for fiberglass piping product by this practice and
thermosetting-resin; the composite structure may contain ag- multiplied by a service design factor to obtain an HDP.
gregate, granular or platelet fillers, thixotropic agents, pig-
3.1.15 pressure rating (PR)—the estimated maximum pres-
ments, or dyes; thermoplastic or thermosetting liners or coat-
ings may be included. sure in the pipe or fitting that can be exerted continuously with
3.1.6 reinforced polymer mortar pipe—a fiberglass pipe a high degree of certainty that failure of the piping component
with aggregate. will not occur.
3.1.7 reinforced thermosetting resin pipe—a fiberglass pipe 3.1.16 service design factor—a number equal to 1.00 or less
without aggregate. that takes into consideration all the variables and degree of
3.1.8 hoop stress—the tensile stress in the wall of the piping safety involved in a fiberglass piping installation so that when
product in the circumferential direction due to internal pres- it is multiplied by the HDB, an HDS and corresponding
sure; hoop stress will be calculated by the ISO equation, as pressure rating is obtained, or when it is multiplied by the PDB,
follows: a pressure rating is obtained directly, such that in either case a
S 5 P~D 2 tr!/2tr (1) satisfactory and safe piping installation results when good
quality components are used and the installation is made
where: properly.
S = hoop stress, psi (kPa), 3.2 Definitions of Terms Specific to This Standard:
D = average reinforced outside diameter, in. (mm),
P = internal pressure, psig (kPa), and 3.2.1 average outside diameter—a measurement obtained
tr = minimum reinforced wall thickness, in. (mm). in accordance with Practice D 3567 less any veil-reinforced
and nonreinforced exterior coating thicknesses.
3.2.2 minimum reinforced wall thickness—a measurement
NOTE 5—Hoop stress should only be determined on straight hollow
cylindrical specimens. Product evaluation of more complex shapes may be
obtained in accordance with Practice D 3567, excluding veil-
based on pressure. reinforced and nonreinforced coating and lining thicknesses;
wall thickness of fittings is determined at the thinnest section of
3.1.9 hydrostatic design basis (HDB)—a hoop stress devel-
the fitting body.
oped for fiberglass pipe by this practice and multiplied by a
service design factor to obtain an HDS.
4. Summary of Practice
3.1.10 hydrostatic design pressure (HDP)—the estimated
maximum internal hydrostatic pressure that can be applied 4.1 Procedure A consists of exposing a minimum of 18
cyclically (Procedure A) or continuously (Procedure B) to a specimens of pipe or fittings, or both to cyclic internal
pressures at a cycle rate of 25 cycles/min and at several
different pressures. Elevated test temperatures are obtained by
5
Available from American National Standards Institute, 11 W. 42nd St., 13th circulating a hot liquid through the specimens or by testing in
Floor, New York, NY 10036. an air environment where the temperature is controlled.

2
Copyright by ASTM Int'l (all rights reserved); Thu Sep 14 08:44:05 EDT 2006
Reproduction authorized per License Agreement with Jorge E. Araujo-Osorio (I.P.N.-E.S.I.Q.I.E. IPN811229H26);
 D 2992 – 01
4.1.1 The cyclic LTHS or cyclic LTHP of a pipe or fitting is strain basis HDB is most often used for buried pipe designs
obtained by an extrapolation of a log-log plot of the linear with internal pressures ranging from 0 to 250 psig (1.72 MPa).
regression line for hoop stress or internal pressure versus 5.2 To characterize fiberglass piping products, it is neces-
cycles to failure. sary to establish the stress versus cycles or time to failure, or
4.1.2 The experimental basis for Procedure A shall be in pressure versus cycles or time to failure relationships over
accordance with Test Method D 2143, which forms a part of three or more logarithmic decades of time (cycles or hours)
this practice. When any part of the procedure is not in within controlled environmental parameters. Because of the
agreement with Test Method D 2143, the provisions of this nature of the test and specimens employed, no single line can
practice shall be used. adequately represent the data. Therefore, the confidence limits
4.1.3 Joints between pipe and fitting specimens shall be should be established.
typical of those normally used for the kind of piping being 5.3 Pressure ratings for piping of various dimensions at each
tested. temperature may be calculated using the HDS determined by
4.2 Procedure B consists of exposing a minimum of 18 testing one size of piping provided that the same specific
specimens of pipe or fittings, or both, to constant internal process and material are used both for test specimens and the
hydrostatic pressures at differing pressure levels in a controlled piping in question.
environment and measuring the time to failure for each 5.4 Pressure ratings at each temperature for components
pressure level. Test temperatures are obtained by immersing other than straight hollow shapes may be calculated using the
the specimens in a controlled-temperature water bath, by HDP determined by testing one size of piping provided that (1)
testing in an air environment where the temperature is con- the specific materials and manufacturing process used for the
trolled, or by circulating a temperature-controlled fluid through test specimens are used for the components, (2) for joints, the
the specimen. joining materials and procedures used to prepare the test
4.2.1 The static LTHS or static LTHP of a pipe or fitting is specimens are used for field joining, and (3) scaling of critical
obtained by an extrapolation of a log-log linear regression line dimensions is related to diameter and pressure rating of the
for hoop stress or internal pressure versus time to failure. component.
4.2.2 The experimental basis for Procedure B shall be in NOTE 7—Scaling of fittings and joints should be further verified by
accordance with either Test Method D 1598 or Test Method short-time testing in accordance with Test Method D 1599.
F 948, or both, which form a part of this practice. When any
part of this practice is not in agreement with the selected 5.5 Results obtained at one set of environmental conditions
method, the provisions of this practice shall be used. should not be used for other conditions, except that higher
4.2.3 Joints between pipe and fitting specimens shall be temperature data can be used for design basis assignment for
typical of those normally used for the kind of piping being lower application temperatures. The design basis should be
tested. determined for each specific piping product. Design and
4.3 The HDB category is obtained by categorizing the processing can significantly affect the long-term performance
LTHS in accordance with Section 7 or Section 10. of piping products, and therefore should be taken into consid-
4.4 The PDB category is obtained by categorizing the LTHP eration during any evaluation.
in accordance with Section 8 or Section 11. 5.6 This practice is valid for a given pipe or fitting only so
4.5 Hydrostatic design stresses for pipe are obtained by long as the specimens are truly representative of that material
multiplying the HDB values by a service design factor. and manufacturing process.
4.6 Reconfirmation of HDB or PDB for Altered 5.6.1 Changes in materials or manufacturing processes will
Constructions—When a product already has an HDB or PDB necessitate a reevaluation as described in Section 12.
determined in accordance with this practice and a change of PROCEDURE A
process or material is made, a reconfirmation of the original
HDB or PDB may be attempted in accordance with Section 12. 6. Long-Term Cyclic Hydrostatic Strength or Long-Term
At least six specimens must be tested and meet the specified Cyclic Hydrostatic Pressure
criteria. 6.1 Select either free-end or restrained-end closures based
5. Significance and Use on the tensile stresses induced by internal pressure and the type
of joint in the intended piping system (see 1.4).
5.1 This practice is useful for establishing the hoop stress or
6.2 Obtain a minimum of 18 failure stress-cycle points for
internal pressure versus time-to-failure relationships, under
each selected temperature in accordance with Test Method
selected internal and external environments which simulate
D 2143 except as follows:
actual anticipated product end-use conditions, from which a
6.2.1 Determine the average outside diameter and the mini-
design basis for specific piping products and materials can be
mum reinforced wall thickness in accordance with Practice
obtained. This practice defines an HDB for material in straight,
D 3567.
hollow cylindrical shapes where hoop stress can be easily
calculated, and a PDB for fittings and joints where stresses are NOTE 8—Because of the need to cut the specimen, this determination
more complex. may be made on the failed test specimen. A corrected hoop stress is then
5.1.1 An alternative design practice based on initial strain calculated for use in the analysis.
versus time-to-failure relationships employs a strain basis HDB 6.2.2 Elevated test temperatures are obtained by circulating
instead of the stress basis HDB defined by this practice. The a heated test liquid through the specimens or by testing in a hot

3
Copyright by ASTM Int'l (all rights reserved); Thu Sep 14 08:44:05 EDT 2006
Reproduction authorized per License Agreement with Jorge E. Araujo-Osorio (I.P.N.-E.S.I.Q.I.E. IPN811229H26);
 D 2992 – 01
air environment. In either case the test liquid shall be main- 7.2 If Sxy > 0 (see A1.4) consider the data unsuitable.
tained within 65°F (3°C) of the selected temperature. 7.3 Calculate r in accordance with A1.4.3. If r is less than
NOTE 9—Where elevated test temperatures are maintained by applying
the applicable minimum value given in Table A1.1, consider
heat to the circulating test liquid, work to date indicates that the ambient the data unsuitable.
air temperature need not be controlled. 7.4 If required, determine the cyclic HDB category in
accordance with Table 1.
6.2.3 The stress or pressure values for test shall be selected
to obtain a distribution of failure points as follows: 8. Cyclic Pressure Design Basis
Cycles to Failure Failure Points
1 000 to 10 000 at least 3
8.1 Use the procedures in 7.1, 7.2, and 7.3, using pressure in
10 000 to 100 000 at least 3 place of stress.
100 000 to 1 000 000 at least 3 8.2 If required, determine the cyclic PDB category in
1 000 000 to 10 000 000 at least 3
After 15 000 000 at least 1
accordance with Table 2.

Total at least 18 PROCEDURE B

6.3 Analyze the test results by using, for each specimen, the 9. Long-Term Static Hydrostatic Strength
logarithm of the stress or pressure in Section 6 and the
logarithm of the cycles to failure, as described in Annex A1. 9.1 Select either free-end or restrained-end closures based
on the tensile stresses induced by internal pressure and the type
NOTE 10—It is the custom of those testing fiberglass pipe to plot stress of joint in the intended piping system (see 1.4).
or pressure on the vertical (y) axis and time or cycles on the horizontal (x)
9.2 Obtain a minimum of 18 failure points for each selected
axis.
temperature in accordance with Test Method D 1598 or Test
6.3.1 A specimen which leaks within one diameter of an end Method F 948 except as follows:
closure may be: (1) included as a failure point if it lies above 9.2.1 Determine the average outside diameter and the mini-
the 95 % lower confidence limit curve; ( 2) repaired and testing mum reinforced wall thickness in accordance with Practice
resumed provided the new leak is more than one diameter from D 3567 (Note 8).
a test joint, or (3) discarded and no data point recorded. 9.2.2 The inside environment for the pipe or fitting, test
6.3.2 Those specimens that have not failed after more than specimens, or both, shall be water. The outside environment
15 000 000 cycles may be included as failures in establishing shall be air. Other media may be used, but the environment
the regression line. Use of such data points may result in a shall be given in the test report. The test liquid shall be
lower or higher cyclic LTHS or cyclic LTHP. In either case, the maintained within 65°F (3°C) of the test temperature (Note 9).
lower confidence value requirements of Section 6 must be 9.2.3 The stress or pressure values for test shall be selected
satisfied. to obtain a distribution of failure points as follows:
NOTE 11—Non-failed specimens may be left under test and the regres- Hours to Failure Failure Points
10 to 1 000 at least 4
sion line recalculated as failures are obtained.
1 000 to 6 000 at least 3
6.3.3 Determine the final line for extrapolation by the After 6 000 at least 3
After 10 000 at least 1
method of least squares using the failure points along with
those nonfailure points selected by the method described in Total at least 18
6.3.1 and 6.3.2. Do not use failure points for stresses or 9.2.4 Maintain the internal test pressure in each specimen
pressures that cause failure in less than 500 cycles on the within 61 % of this pressure. Measure the time to failure to
average; determine these points by averaging the number of within 62 % or 40 h, whichever is smaller.
cycles-to-failure of tests made at the same stress or pressure
level, that is, a stress within 6200 psi (1380 kPa) or a pressure
within 620 psig (138 kPa). Include in the report all failure TABLE 1 Hydrostatic Design Basis Categories by Procedure A
or Procedure B
points excluded from the calculation by this operation and
identify them as being in this category. Hydrostatic Design Basis
Category Range of Calculated Values
NOTE 12—Since this procedure is for pipe or fittings, or both, it is
A
recommended that the pipe specimen and fitting be tested at the same time psi (kPa) psi (kPa)
as one specimen, using the normal joining procedures to join them 2 500 (17 200) 2 400 to 3 010 (16 500 to 20 700)
together, with the fitting being at one end of the specimen. If the fitting 3 150 (21 700) 3 020 to 3 820 (20 800 to 26 300)
fails first, it can be cut off, and the test can be continued using the unfailed 4 000 (27 600) 3 830 to 4 790 (26 400 to 33 000)
5 000 (34 500) 4 800 to 5 990 (33 100 to 40 900)
pipe with a mechanical end closure replacing the fitting. Should the pipe
6 300 (43 400) 6 000 to 7 590 (41 000 to 52 900)
fail first, it can be recorded and repaired and the test continued until the 8 000 (55 200) 7 600 to 9 590 (53 000 to 65 900)
fitting fails. If this recommendation is followed, it may enable the tester to 10 000 (68 900) 9 600 to 11 990 (66 000 to 82 900)
obtain failure points for both the pipe and the fitting while testing only one 12 500 (86 200) 12 000 to 15 290 (83 000 to 105 900)
specimen. 16 000 (110 000) 15 300 to 18 990 (106 000 to 130 900)
20 000 (138 000) 19 000 to 23 990 (131 000 to 169 900)
25 000 (172 000) 24 000 to 29 990 (170 000 to 209 900)
7. Cyclic Hydrostatic Design Basis 31 500 (217 000) 30 000 to 37 990 (210 000 to 259 900)
7.1 Calculate the cyclic LTHS at the specified time (1503 40 000 (276 000) 38 000 to 47 000 (260 000 to 320 000)
106 or 657 3 106 cycles) as described in Annex A1. A
Standard stress levels chosen in accordance with ISO 3, Series R10

4
Copyright by ASTM Int'l (all rights reserved); Thu Sep 14 08:44:05 EDT 2006
Reproduction authorized per License Agreement with Jorge E. Araujo-Osorio (I.P.N.-E.S.I.Q.I.E. IPN811229H26);
 D 2992 – 01
TABLE 2 Pressure Design Basis Categories by Procedure A or 11. Static Pressure Design Basis
Procedure B
11.1 Use the procedures in 7.1, 7.2, and 7.3, using pressure
Pressure Design Basis Category Range of Calculated Values in place of stress.
psi (bar)A (kPa) psi (kPa) 11.2 If required, determine the static PDB category in
91 (6.3) (630) 87 to 110 (605 to 760) accordance with Table 2.
116 (8) (800) 111 to 143 (765 to 990)
150 (10) (1 000) 144 to 172 (995 to 1 180)
180 (12.5) (1 250) 173 to 220 (1 190 to 1 510)
12. Reconfirmation of HDB or PDB
230 (16) (1 600) 221 to 287 (1 520 to 1 980) 12.1 When a piping product has an existing HDB or PDB
300 (20) (2 000) 288 to 345 (1 990 to 2 380)
360 (25) (2 500) 346 to 438 (2 390 to 3 020) determined in accordance with Procedure A or Procedure B,
460 (31.5) (3 150) 439 to 556 (3 030 to 3 830) any change in material, manufacturing process, construction,
580 (40) (4 000) 557 to 695 (3 840 to 4 790) or liner thickness will necessitate a screening evaluation as
725 (50) (5 000) 696 to 876 (4 800 to 6 040)
910 (63) (6 300) 877 to 1 110 (6 050 to 7 680) described in 12.2, 12.3, 12.4, 12.5, and 12.6.
1 160 (80) (8 000) 1 115 to 1 380 (7 690 to 9 580) 12.2 Obtain failure points for at least two sets of specimens,
1 450 (100) (10 000) 1 390 to 1 720 (9 590 to 11 800)
1 800 (125) (12 500) 1 730 to 2 220 (11 900 to 15 300)
each set consisting of 3 or more specimens tested at the same
A
stress or pressure level, that is, a stress within 6200 psi (1380
Standard pressures chosen in accordance with ISO 3, Series R10.
kPa) or a pressure within 620 psi (138 kPa), as follows:
12.2.1 For Procedure A:
Cycles to Failure
9.3 Analyze the test results by using, for each failure point, (Average of Set) Failure Points
the logarithm of the stress or pressure in pound-force per
square inch or pound-force per square inch gage (kilopascals) 15 000 to 300 000 at least 3
More than 1 500 000 at least 3
and the logarithm of the time-to-failure in hours as described in
Annex A1 (Note 8). Total at least 6
9.3.1 A specimen which leaks within one diameter of an end Include as failures those specimens which have not failed
closure may be: (1) included as a failure point if it lies above after 4 500 000 cycles provided they exceed the existing HDB
the 95 % lower confidence limit curve; ( 2) repaired and testing or PDB regression line.
resumed provided the new leak is more than one diameter from 12.2.2 For Procedure B:
a test joint, or (3) discarded and no failure point recorded.
Hours to Failure
9.3.2 Those specimens that have not failed after more than (Average of Set) Failure Points
10 000 h may be included as failures in establishing the
10 to 200 at least 3
regression line. Use of such data points may result in a lower More than 1000 at least 3
or higher static LTHS or static LTHP. In either case, the lower
confidence value requirements of 9.4.2 must be satisfied. Total at least 6

NOTE 13—Non-failed specimens may be left under test and the regres- Include as failures those specimens which have not failed
sion line recalculated as failures are obtained. after 3000 h provided they exceed the existing HDB or PDB
regression line.
9.3.3 Determine the final line for extrapolation by the
12.3 Calculate and plot the 95 % confidence limits and the
method of least squares using the failure points along with
95 % prediction limits of the original regression line in
those nonfailure points selected by the method described in
accordance with Annex A4 using only data obtained prior to
9.3.1 and 9.3.2. Do not use failure points for stresses or
the change.
pressures that cause failure in less than 0.3 h on the average;
determine these points by averaging the times-to-failure of NOTE 14—Prediction limits define the bounds for single observations,
tests made at the same stress or pressure level, that is, a stress whereas confidence limits define the bounds for the regression line.
within 6200 psi (1380 kPa) or a pressure within 620 psi (138 NOTE 15—For 95 % confidence limits, there is a 2.5 % probability that
kPa). Include in the report all failure points excluded from the the mean value for the regression line may fall above the UCL and a 2.5 %
calculation by this operation and identify them as being in this probability that the mean value for the regression line may fall below the
LCL. For 95 % prediction limits, there is a 2.5 % probability that
category (Note 11). individual data points may fall above the UPL and a 2.5 % probability that
individual data points may fall below the LPL.
10. Static Hydrostatic Design Basis
12.4 Consider any changes in the material or manufacturing
10.1 Calculate the static LTHS at the specified time process minor and permissible if the results of 12.2 meet the
(100 000 or 438 000 h) as described in Annex A1. following criteria.
10.2 If Sxy > 0 (see A1.4), consider the data unsuitable. 12.4.1 The average failure point for each stress or pressure
10.3 Calculate r in accordance with A1.4.3. If r is less than level falls on or above the 95 % lower confidence limit of the
the applicable minimum value given in Table A1.1, consider original regression line.
the data unsuitable. 12.4.2 The earliest individual failure point at each stress or
10.4 If required, determine the static HDB category in pressure level falls on or above the 95 % lower prediction limit
accordance with Table 1. of the original regression line.

5
Copyright by ASTM Int'l (all rights reserved); Thu Sep 14 08:44:05 EDT 2006
Reproduction authorized per License Agreement with Jorge E. Araujo-Osorio (I.P.N.-E.S.I.Q.I.E. IPN811229H26);
 D 2992 – 01
12.4.3 The failure points are distributed about the originally each diameter and wall thickness of pipe made from the
determined regression line. No more than two thirds of the specific materials and constructions tested.
individual failure points may fall below the original regression 14.2 For data based on internal pressure, establish the
line. pressure rating directly from the HDP for products made from
12.5 Alternatively to 12.4, consider any changes in the the specific materials and constructions tested.
material or manufacturing process permissible if the results of
12.2 meet the following: 15. Report
12.5.1 All data points fall above the 95 % lower confidence 15.1 Report the following information:
limit of the original regression line, and 15.1.1 Complete identification of the specimen including
12.5.2 At least two points exceed 4.5 3 10 6 cycles or material type, source, manufacturer’s name and code number,
3000-h failure time. and previous significant history, if any.
12.6 Data meeting the criteria of 12.4 or 12.5 may be 15.1.2 Specimen dimensions including nominal size, aver-
assumed to be part of the original data set and a new regression age and minimum reinforced wall thickness, and average
line and HDB or PDB determined using all failure points. outside diameter, and liner material and liner thickness if
12.7 If the data fails to satisfy the criteria of 12.4 or 12.5, the product is lined.
changes are considered major and a new regression line must 15.1.3 Fitting dimensions, including all items listed in
be established. While the new test program is being conducted, 15.1.2 and the type of fitting.
an interim HDB or PDB for the material or process change may 15.1.4 Procedure used, (Procedure A or Procedure B), and
be taken as the lower of the following: the ASTM designation of the underlying test method.
12.7.1 The 95 % lower confidence limit of the value ob- 15.1.5 End closure type, free-end, or restrained-end.
tained by extrapolating the failure points of 12.2.1 to 15.1.6 Test temperature.
657 000 000 cycles (50 years) by the procedure in 7.2, or the 15.1.7 Test environment inside and outside of the pipe.
failure points of 12.2.2 to 438 000 h (50 years) by the 15.1.8 A table of stresses or pressures in pound-force per
procedure in Annex A1. square inch or pound-force per square inch gage (kilopascals)
12.7.2 The 95 % lower confidence limit of the original and the number of cycles to failure (Procedure A) or time-to-
regression line at 50 years. failure in hours (Procedure B) of all the specimens tested; the
nature of the failures, and the part that failed, that is, fitting or
13. Hydrostatic Design Stress or Hydrostatic Design pipe. Specimens that are included as failures after they have
Pressure been under stress or pressure for more than 15 000 000 cycles
13.1 Obtain the HDS or HDP by multiplying the HDB or or more than 10 000 h shall be indicated.
PDB as determined by Procedure A or Procedure B by a service 15.1.9 The estimated LTHS or LTHP.
design factor selected for the application on the basis of two 15.1.10 The value for r.
general groups of conditions. The first group considers the 15.1.11 The HDB or HDP.
manufacturing and testing variables, specifically normal varia- 15.1.12 The source of the HDB or PDB (7.1 or 7.2 for
tions in the material, manufacture, dimensions, good handling Procedure A or 10.1 or 10.2 for Procedure B), and the
techniques, and in the evaluation procedures in this method. categorized value in accordance with Table 1 or Table 2.
The second group considers the application or use, specifically 15.1.13 Any unusual behavior observed in the tests.
installation, environment, temperature, hazard involved, life 15.1.14 Dates of tests.
expectancy desired, and the degree of reliability selected. 15.1.15 Name of laboratory and supervisor of tests.
NOTE 16—It is not the intent of this practice to give service design 16. Precision and Bias
factors. The service design factor should be selected by the design
engineer after evaluating fully the service conditions and the engineering
16.1 The precision and bias of this practice for obtaining the
properties of the specific plastic pipe material under consideration. HDB or PDB are as specified in Test Methods D 1598, D 2143,
Recommended service design factors will not be developed or issued by and F 948. This practice includes a statistical basis for evalu-
ASTM. ating the suitability of the data in Sections 6 and 9.
14. Pressure Rating 17. Keywords
14.1 For data based on hoop stress calculate the pressure 17.1 closure; cyclic pressure; design basis; fiberglass pipe;
rating from the HDS by means of the ISO equation in 3.1.8 for reconfirmation; static pressure

6
Copyright by ASTM Int'l (all rights reserved); Thu Sep 14 08:44:05 EDT 2006
Reproduction authorized per License Agreement with Jorge E. Araujo-Osorio (I.P.N.-E.S.I.Q.I.E. IPN811229H26);
 D 2992 – 01
ANNEX

(Mandatory Information)

A1. LEAST SQUARES CALCULATIONS FOR LONG-TERM HYDROSTATIC STRENGTH OR LONG-TERM HYDROSTATIC
PRESSURE

A1.1 General x̄ = arithmetic mean of all xivalues:

A1.1.1 The analysis is based on the following relationship: 1
5 n (x i (A1.3)
y 5 a 1 bx (A1.1)
A1.4.2 Relevant Sums-of-Squares:
where:
y = one variable, A1.4.2.1 Calculate the following sums-of-squares and
x = other variable, cross-products:
b = slope of the line, and 1
Sxy 5 n (~xi 2 x̄!~yi 2 ȳ! (A1.4)
a = intercept on the y axis.
A1.1.2 A linear functional relationship analysis (sometimes A1.4.2.2 If Sxy> 0, consider the data unsuitable for evalu-
called “covariance analysis”) is used, subject to tests for the ating the material; otherwise calculate also:
sign (that is, “+” or “−”) of the slope and the coefficient of
1
correlation for the quantity of data available. The relevant Sxx 5 n (~xi 2 x̄! 2 (A1.5)
equations are given together with example data and results, on
the basis of which any other statistical computing package may 1
Syy 5 n (~y i 2 ȳ! 2 (A1.6)
be used subject to validation by agreement with the example
results to within the indicated limits. A1.4.3 Correlation of Data:
A1.1.3 For the purposes of this annex, a design service life A1.4.3.1 Calculate the coefficient of correlation, r, from the
of 50 years has been assumed. following relationship:
~Sxy! 2
A1.2 Procedure for Analysis of Data r2 5 (A1.7)
~Sxx 3 Syy!
A1.2.1 Use a linear functional relationship analysis to
analyze n pairs of data values (as y and x) to obtain the r 5 =r 2
following information: A1.4.3.2 If the value of r is less than the applicable
A1.2.1.1 The slope of line, b, minimum value given in Table A1.1 as a function of n, reject
A1.2.1.2 The intercept on the y axis, a, the data; otherwise, proceed to A1.4.4.
A1.2.1.3 The correlation coefficient, r, and A1.4.4 Functional Relationships:
A1.2.1.4 The predicted mean and the lower 95 % confi- A1.4.4.1 To find a and b for the functional relationship line,
dence and prediction intervals on the mean value. y = a + bx (Eq A1.1), first set:
A1.3 Assignment of Variables l5 SS D
Syy
xx
(A1.8)
A1.3.1 Let x be log10t, where t is the time, in hours (or
and then let:
cycles), and let y be log10V, where V is the stress (or pressure)
value. b5=l (A1.9)
and then:
A1.4 Functional Relationship Equations and Method of
Calculation
A1.4.1 Basic Statistics and Symbols: TABLE A1.1 Minimum Values for the Coefficient of Correlation,
r, for Acceptable Data from n Pairs of Data
A1.4.1.1 The following basic statistics and symbols are
used: (n − 2) r minimum (n − 2) r minimum
11 0.6835 25 0.4869
12 0.6614 30 0.4487
n = number of pairs of observed data values (Vi, t i), 13 0.6411 35 0.4182
yi = log10 of Vi, where Vi is the stress (or pressure) at failure 14 0.6226 40 0.3932
of Observation i; i = 1, . . . n, 15 0.6055 45 0.3721
16 0.5897 50 0.3541
xi = log 10 of ti, where ti is the time to failure in hours of 17 0.5751 60 0.3248
Observation i; i = 1, . . .n, 18 0.5614 70 0.3017
ȳ = arithmetic mean of all yi values: 19 0.5487 80 0.2830
20 0.5386 90 0.2673
1 21 0.5252 100 0.2540
5 n (yi (A1.2) 22 0.5145 ... ...
23 0.5043 ... ...
24 0.4952 ... ...

7
Copyright by ASTM Int'l (all rights reserved); Thu Sep 14 08:44:05 EDT 2006
Reproduction authorized per License Agreement with Jorge E. Araujo-Osorio (I.P.N.-E.S.I.Q.I.E. IPN811229H26);
 D 2992 – 01
TABLE A1.2 Student’s “t ” Value (Two-Sided 0.05 Level of Significance)

Degrees of Freedom Student’s“ t” Value, Degrees of Freedom Student’s “t” Value, Degrees of Freedom Student’s “t” Value,
(n − 2) tv (n − 2) tv (n − 2) tv
1 12.7062 46 2.0129 91 1.9864
2 4.3027 47 2.0117 92 1.9861
3 3.1824 48 2.0106 93 1.9858
4 2.7764 49 2.0096 94 1.9855
5 2.5706 50 2.0086 95 1.9853

6 2.4469 51 2.0076 96 1.9850

7 2.3646 52 2.0066 97 1.9847
8 2.3060 53 2.0057 98 1.9845
9 2.2622 54 2.0049 99 1.9842
10 2.2281 55 2.0040 100 1.9840

11 2.2010 56 2.0032 102 1.9835

12 2.1788 57 2.0025 104 1.9830
13 2.1604 58 2.0017 106 1.9826
14 2.1448 59 2.0010 108 1.9822
15 2.1315 60 2.0003 110 1.9818

16 2.1199 61 1.9996 112 1.9814

17 2.1098 62 1.9990 114 1.9810
18 2.1009 63 1.9983 116 1.9806
19 2.0930 64 1.9977 118 1.9803
20 2.0860 65 1.9971 120 1.9799

21 2.0796 66 1.9966 122 1.9796

22 2.0739 67 1.9960 124 1.9793
23 2.0687 68 1.9955 126 1.9790
24 2.0639 69 1.9949 128 1.9787
25 2.0595 70 1.9944 130 1.9784

26 2.0555 71 1.9939 132 1.9781

27 2.0518 72 1.9935 134 1.9778
28 2.0484 73 1.9930 136 1.9776
29 2.0452 74 1.9925 138 1.9773
30 2.0423 75 1.9921 140 1.9771

31 2.0395 76 1.9917 142 1.9768

32 2.0369 77 1.9913 144 1.9766
33 2.0345 78 1.9908 146 1.9763
34 2.0322 79 1.9905 148 1.9761
35 2.0301 80 1.9901 150 1.9759

36 2.0281 81 1.9897 200 1.9719

37 2.0262 82 1.9893 300 1.9679
38 2.0244 83 1.9890 400 1.9659
39 2.0227 84 1.9886 500 1.9647
40 2.0211 85 1.9883 600 1.9639

41 2.0195 86 1.9879 700 1.9634

42 2.0181 87 1.9876 800 1.9629
43 2.0167 88 1.9873 900 1.9626
44 2.0154 89 1.9870 1000 1.9623
45 2.0141 90 1.9867 ... 1.9600

a 5 ȳ 2 bx̄ (A1.10) ji5 $lxi 1 ~yi 2 a!b%/2l (A1.12)

NOTE A1.1—In general, b takes the sign of Sxy. Y i 5 a 1 bji (A1.13)
NOTE A1.2—Since y = log10 V and x = log10 t, hence V = 10y, t = 10xand 2 2 2
the implied relationship for V in terms of t is therefore: sd 5 $(~yi2 Yi! 1 l(~x i2 ji! %/$l~n 2 2!% (A1.14)

V 5 10~a1b3log 10 t! A1.4.5.3 Calculate the following quantities:

A1.4.5 Calculation of Variances: t 5 bsd2/2S xy (A1.15)
A1.4.5.1 If tLis the applicable time to failure, then set: D5 2lbs d2/nSxy (A1.16)
x L5 log10tL (A1.11)
B 5 2Dx̄~1 1 t! (A1.17)
A1.4.5.2 Calculate, in turn, the following sequence of
statistics. For i = 1 to i = n, the best fit, ji, for true x, the best A1.4.5.4 Calculate the following variances: the variance,
fit, Yi, for true y and the error variance, sd2, for x using Eq C, of b using the formula:
A1.12, Eq A1.13, and Eq A1.14, respectively: C 5 D~1 1 t! (A1.18)

8
Copyright by ASTM Int'l (all rights reserved); Thu Sep 14 08:44:05 EDT 2006
Reproduction authorized per License Agreement with Jorge E. Araujo-Osorio (I.P.N.-E.S.I.Q.I.E. IPN811229H26);
 D 2992 – 01
the variance, A, of a using the formula: where:

H Sxy
A 5 D x̄ 2 ~1 1 t! 1 b J (A1.19)
yL = value obtained in accordance with Eq A1.24 when xL
is, as applicable, the value in accordance with Eq
A1.11 appropriate to a design life of, for example, 50
the variance, sn2, of the fitted line at xLusing the formula:
years (that is, xL = 5.6415 (h)) or to a time at which it
sn 2 5 A 1 2BxL 1 CxL 2 (A1.20) is desired to predict with 95 % confidence the mini-
2
the error variance, se , for y using the formula: mum value for the next observation of V,
sy = value obtained in accordance with Eq A1.23, and
se 2 5 lsd 2 (A1.21) tv = applicable value for Student’s t for v = n − 2 df, as
2
the total variance, sy , for future values, yL, for y at xLusing given in Table A1.2 for a two-sided 0.05 level of
the formula: significance (that is, mean 62.5 %).
sy 2 5 s n 2 1 s e 2 (A1.22)
A1.4.5.5 Calculate the estimated standard deviation, sy, A1.4.6.2 Calculate the corresponding lower 95 % prediction
for yLusing the equation: limit for V using the relationship:
sy 5 ~s n 2 1 se 2 !0.5 (A1.23) VL 0.95 5 10YL 0.95 (A1.26)
and the predicted value, yL, for y at xLusing the relationship: A1.4.6.3 The predicted mean value of V at time tL, that is,
yL 5 a 1 bxL (A1.24) VL, is given by the relationship:
where a and b have the values obtained in accordance with VL 5 10YL (A1.27)
Eq A1.9 and Eq A1.10
where:
. YL = value obtained in accordance with Eq A1.24.
A1.4.6 Calculation and Confidence Intervals:
A1.4.6.1 Calculate the lower 95 % prediction interval, A1.4.6.4 Setting sy 2= sn2 in Eq A1.22 will produce a
yL 0.95, predicted for yLusing the equation: confidence interval for the line rather than a prediction interval
yL 0.95 5 yL 2 tvsy (A1.25) for a future observation.

APPENDIXES

(Nonmandatory Information)

X1. DATA ANALYSIS

X1.1 Hoop Stress versus Cycles-to-Failure or Time-to- X1.1.2 The main limitation of the use of hoop stress is that
Failure: it can only be applied to simple tubular-shaped specimens.
X1.1.1 Hoop stress is a more convenient parameter to use Therefore, its application has been mainly limited to materials
when attempting to predict long-term hydrostatic strength of a and a few products such as pipe and simple fittings like
material. Its use reduces scatter in the data by compensating for couplings.
varying dimensions in the test specimens. It effectively nor-
malizes pressure for variations in specimen geometry, and X1.2 Internal Pressure versus Cycles-to-Failure or Time-
reduces the variable to a material parameter. For this particular to-Failure—The use of internal pressure rather than stress
reason it has been widely used for evaluating the long-term extends the application of this practice to the prediction of
hydrostatic properties of plastic materials. Essentially, once a service life for many products of complex geometries which do
value for HDS has been determined for a particular material not permit the calculation of hoop stress. The logarithm of
and construction, that value can be used to effectively predict internal pressure is used in place of the logarithm of hoop stress
the long-term working pressure of tubular products by com- in the calculations.
pensating for the various product geometries.

9
Copyright by ASTM Int'l (all rights reserved); Thu Sep 14 08:44:05 EDT 2006
Reproduction authorized per License Agreement with Jorge E. Araujo-Osorio (I.P.N.-E.S.I.Q.I.E. IPN811229H26);
 D 2992 – 01

X2. EXAMPLE CALCULATION

X2.1 Basic Data—The example data given in Table X2.1, l = 1.100457 x 10–3
together with the example analysis given in this appendix, can b = –3.31731 x 10–2
be used to validate statistical packages procedures. Because of a = 3.782188
rounding errors, it is unlikely that there will be exact agree-
ment, but acceptable procedures should agree within 60.1 % X2.5 Calculated Variances:
of the results given in X2.5. D = 4.84225 x 10 –6
B = –1.46896 x 10 –5
X2.2 Sums of Squares: C (variance of b)= 5.01271 x 10–6
Sxx = 0.798109 A (variance of a)= 4.66730 x 10–5
Syy = 8.78285 x 10–4 s n2(error variance for) x = 4.046696 x 10–5
Sxy = –0.024836 s e2(error variance for) y = 5.80057 x 10–5

X2.3 Coeffıcient of Correlation: X2.6 Confidence Limits—For N = 32 and Student’s t of

r = 0.938083 2.0423, the estimated mean and confidence and prediction
intervals are given in Table X2.2.
X2.4 Functional Relationships:

TABLE X2.1 Example Data for Example Calculation

Data Time, Stress, Log Time, Log Stress, Data Stress, Log Time, Log Stress,
Point h psi h f Point Time, h psi h f
1 9. 5500. 0.95424 3.74036 17 1301. 4700. 3.11428 3.67210
2 13. 5500. 1.11394 3.74036 18 1430. 4800. 3.15534 3.68124
3 17. 5500. 1.23045 3.74036 19 1710. 4800. 3.23300 3.68124
4 17. 5500. 3.74036 20 2103. 4800. 3.32284 3.68124

5 104. 5200. 2.01703 3.71600 21 2220. 4500. 3.34635 3.65321

6 142. 5200. 2.15229 3.71600 22 2230. 4400. 3.34830 3.64345
7 204. 5200. 2.30963 3.71600 23 3816. 4700. 3.58161 3.67210
8 209. 5200. 2.32015 3.71600 24 4110. 4700. 3.61384 3.67210

9 272. 5000. 2.43457 3.69897 25 4173. 4600. 3.62043 3.66276

10 446. 5000. 2.64933 3.69897 26 5184. 4400. 3.71466 3.64345
11 466. 5000. 2.66839 3.69897 27 8900. 4600. 3.94939 3.66276
12 589. 4800. 2.77012 3.68124 28 8900. 4600. 3.94939 3.66276

13 669. 4700. 2.82543 3.67210 29 10900. 4500. 4.03743 3.65321

14 684. 5000. 2.83506 3.69897 30 10920. 4500. 4.03822 3.65321
15 878. 4600. 2.94349 3.66276 31 12340. 4500. 4.09132 3.65321
16 1299. 4800. 3.11361 3.68124 32 12340. 4500. 4.09132 3.65321

10
Copyright by ASTM Int'l (all rights reserved); Thu Sep 14 08:44:05 EDT 2006
Reproduction authorized per License Agreement with Jorge E. Araujo-Osorio (I.P.N.-E.S.I.Q.I.E. IPN811229H26);
 D 2992 – 01
TABLE X2.2 Confidence Limits

Lower Lower
Confidence Prediction
Time, h Mean Interval Interval
1 6056 5864 5771
10 5611 5487 5379
100 5198 5129 5003
1000 4816 4772 4641
10 000 4462 4398 4293
100 000 4133 4037 3960
438 000 3936 3820 3756

SUMMARY OF CHANGES

Committee D20 has identified the location of selected changes to this standard since the last issue D 2992–96
that may impact the use of this standard.

(1) Changes acronym, RPMP, definition from reinforced plas-

tic mortar pipe to reinforced polymer mortar pipe.

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 service@astm.org (e-mail); or through the ASTM website
(www.astm.org).

11
Copyright by ASTM Int'l (all rights reserved); Thu Sep 14 08:44:05 EDT 2006
Reproduction authorized per License Agreement with Jorge E. Araujo-Osorio (I.P.N.-E.S.I.Q.I.E. IPN811229H26);