Bending Fatigue of Pipe and Piping Components Fabricated From High Density Polyethylene Materials Available to Purchase

Degradation of service water systems is a major issue facing nuclear power plant owners, and many plants will require repair or replacement of existing carbon steel piping components. High Density Polyethylene pipe has been used in non-safety service water systems for over nine years and found to perform well, but it is not currently permitted in the ASME Section III Boiler and Pressure Vessel Code, Division 1 for use in nuclear safety-related systems. To assist in the implementation of High Density Polyethylene pipe in the ASME Boiler and Pressure Vessel Code, Section III, Division 1 for Safety Class 3 applications, EPRI initiated a testing program that includes tensile and fatigue testing of HDPE piping and components and the development of slow crack growth data. Straight cantilever bending fatigue tests on PE 4710 pipe with a minimum cell classification of 445474C were conducted. The tests were designed to comply with the requirements for fatigue testing given in Appendix II of the ASME Boiler and Pressure Vessel Code, Section III, Division 1. They were also designed to achieve failure at the fusion butt welds near the cantilever support. S-N curves developed from both sets of data were found to fit well to power formulas of the type S = C/N^b required by mandatory Appendix II. The tests were conducted at various temperatures from 50° F to 160° F and in addition the effects of cyclic rate and aging were evaluated. Based on the straight pipe tests, stress intensification factors were calculated for 5-segment miter bends in both the in-plane and out-of-plane directions. The test elbows were fabricated from PE 4710 material with cell classification 445474C. Two sizes of 5-segment miter bends were tested, 4” and 12” diameter. The fatigue testing results showed one of the unique characteristics of High Density Polyethylene pipe: a significant decrease in material stiffness from the first few test cycles to a lower value that remains almost constant until failure. Thus, S-N curves and SIFs were determined twice: first based on the initial cycle results and again at the midlife of the fatigue tests. This paper provides a description and overview of the test program, testing methods and materials tested. In addition, an overview and summary of the test program results are provided.