1、September IO, 1999 NSF Protocol PI71 Chlorine Resistance of Plastic Piping Materials September 1 O, 1999 NSF International, an independent, not-for-profit organization, is dedicated to public health safety and protection of the environment by developing standards, by providing education and by provi
2、ding superior third-party conformity assessment services while representing the interest of all stakeholders. This Protocol is subject to revision. Contact NSF to confirm this revision is current. Users of this Protocol may request clarifications and interpretations, or propose revisions by contacti
3、ng: NSF International Engineering T =temperature in Rankin (Kelvin) a = a constant b = a constant Pressure may be used in place of hoop stress for products of complex geometries that do not permit the calculation of hoop stress. The data shall have a R2 value greater than 0.8 to be considered accept
4、able for use with this protocol. 5.7.3 Pressure Linearity Confirmation For the data generated at the same temperature at three different hoop stresses (pressures), perform regression analysis based on hoop stress according to the equation: log (ft) = a + b log (S) where: ft S a = a constant b = a co
5、nstant = time to failure in hours; = hoop stress, MPa (psi) as given by the IS0 equation below Pressure may be used in place of hoop stress for products of complex geometries that do not permit the calculation of hoop stress. The data shall have a R2 value greater than 0.8 to be considered acceptabl
6、e for use with this protocol. 5.7.4 Regression Analysis Perform a multiple variable linear regression on the failure data in accordance with the three coefficient rate process model (or model Q of IS0 9080 (see Appendix A). Data shall be analyzed based on hoop stress as calculated by the IS0 equatio
7、n: S = P (D - t)/2t where: S P D t = hoop stress, MPa (psi) = internal pressure, MPag (psig), and = average outside diameter, mm (inches) = minimum wall thickness, mm (inches) Pressure may be used in place of hoop stress for products of complex geometries that to not permit the calculation of hoop s
8、tress. September 1 O, 1999 Copyright O 1999 NSF International Page 8 of 12 The data shall have a R2 value greater than 0.8 to be considered acceptable for use with this protocol. Product Type Domestic Continuous Re-circulation Traditional Domestic Note: The three coefficient rate process model and m
9、odel Q of IS0 9080 are identical. Any statistical analysis program capable of multiple linear regression analysis is suitable for analysis. Extrapolation Extrapolation Service Pressure Time Temperature 100% 60C (1 40 OF) 25% 60C (140 OF) 75% 23C (73 OF) 0.55 MPa (80 psi) 0.55 MPa (80 psi) 5.7.5 Test
10、 Lifetime Extrapolation Extrapolate the test lifetime based on the relevant conditions in Table 2. For estimation of test lifetimes under variable temperatures apply Miners Rule (ISO/DIS 13760) (see Appendix B). The resulting value is the Extrapolated Test Lifetime. Apply a 0.5 design factor to the
11、Extrapolated Test Lifetime. The resulting value is the Estimated Minimum Chlorine Resistance Service Lifetime of the tested material exposed to typical chlorinated potable water. The Estimated Minimum Chlorine Resistance Service Lifetime shall be greater than 40 years. The Extrapolated Test Lifetime
12、 and Estimated Minimum Chlorine Resistance Service Lifetime provided by this protocol are for conditions equivalent to those under which the test data were obtained and do not take into account other environmental factors that may lead to a reduction in lifetime for the evaluated material. Table 2.
13、Product End-Use Conditions 5.8 Report The final report shall include at a minimum, the following information: 5.8.1 Complete identification of the sample, including material type, source, method of manufacture, manufacturing process details, and manufacturers name. 5.8.2 Specimen dimensions, includi
14、ng nominal size and, when applicable, average and minimum wall thickness, average outside diameter and length to diameter ratio for before and after testing. When testing configurations other than pipe, include all critical dimensions based on the specific configuration being tested. September 1 O,
15、1999 Copyright O 1999 NSF International Page 9 of 12 5.8.3 A description/drawing/photograph, including sufficient information to identify all components of the test specimen shall be included in the report. 5.8.4 All process conditions, including pressure, water temperature, air temperature, tempera
16、ture drop across samples, chlorine level, pH, flow-rate and ORP. 5.8.5 A table of temperatures, pressures and time to failure in hours for all the specimens tested. 5.8.6 A description of the nature of the failures and location of the failures 5.8.7 Any unusual behavior observed in the tests. Any ch
17、ange in color, surface texture, or other change in appearance that may be the result of a physical, chemical, or environmental effect must be reported, whether or not such change played a role in the failure of the part. 5.8.8 Dates of tests 5.8.9 A summary of the regression analysis and the extrapo
18、lated test lifetime. 5.8.10 Name of laboratory and signature of the supervisor of the tests. September 1 O, 1999 Copyright O 1999 NSF International Page 1 O of 12 Appendix A Regression Analysis The multiple linear regression analysis shall be based on the three coefficient rate process extrapolation
19、 model (or the equivalent three parameter standard extrapolation model (SEM) of IS0 9080): where: ft = time to failure in hours; T =temperature in Rankin (Kelvin); S = hoop stress, in MPa (psi); A,B,C = constant; The four parameter model of IS0 9080 may be used if can be demonstrated to provide bett
20、er statistical fit to the data: logloft = c1 + c2/T + csIogl0S + c4(10g10S)/T + e (2) where: Cl = a parameter e = the error variable, having a Laplace-Gauss distribution, with zero mean and constant variance. The errors are assumed to be independent September 1 O, 1999 Copyright O 1999 NSF Internati
21、onal Page 11 of 12 Appendix B Application of Miners Rule 1) Perform regression analysis per Appendix A. 2) Based on the regression analysis equation, obtain an extrapolated test lifetime at the following end use conditions: A) A pressure of 0.55 MPa (80 psi) and a temperature of 60C (140 OF). B) A p
22、ressure of 0.55 MPa (80 psi) and a temperature of 23C (73 OF) Convert the extrapolated test lifetime to years. 3) Calculate the lifetime at 25% service at 60C (140 OF) and 75% service at 23C (73 OF) based on Miners Rule (DIN EN IS0 13760 - 1998). Example Calculation: 1) Condition A extrapolated test
23、 lifetime = tA Condition B extrapolated test lifetime = tB 2) The total damage per year (TDY) is; 3) The extrapolated test lifetime for 25% service at condition A and 75% service at condition B is; t, = lOO/TDY years Note: Miners rule provides an estimate of performance under multiple exposure envir
24、onments. It currently provides what is believed to be the best available method for making this type of estimation. The accuracy of this method may vary with material or product. The percentage of time at higher temperature (60 OC, 140 OF) has been set higher than is reported in actual practice in an attempt to minimize any over-prediction of lifetime that may result due to the possible inaccuracies of Miners Rule. September 1 O, 1999 Copyright O 1999 NSF International Page 12 of 12
