A Review of Service Life Prediction Models for High Density Polyethylene Piping for Nuclear Safety-Related Applications

Session 9A
12:00 pm

Prabhat Krishnaswamy, Engineering Mechanics Corporation of Columbus

The ASME Boiler and Pressure Vessel Codes Committee (BPVC) has published Code Case N-755 that describes the requirements for the use of PE pipe for the construction of Section III, Division 1 Class 3 buried piping systems for service water applications in nuclear power plants. The code case was developed by Special Working Group–PE Pipe (SWG-PP) within Section III (Design) of the BPVC. The US Nuclear Regulatory Commission (USNRC) has not as yet approved this Code Case for use in regulatory decisions. However, two Relief Requests for installation of PE Piping in safety-related applications at US Nuclear Power plants have been approved by the USNRC.

The three major failure modes for PE Piping are (i) ‘ductile’ failure due to overload, (ii) ‘brittle’ or slow crack growth (SCG) failure due to the effect of long term sustained loads, and elevated temperatures and stress concentrations, and (iii) rapid crack propagation.

The paper focuses on the susceptibility of PE pipe to premature failure due to SCG – specifically, the various forecasting models and accelerated testing protocols that are used to analyze short term experimental results to predict long term (50 years and beyond) service life. The models most commonly used for this purpose include is based on the rate process method (RPM). Also used in recent years is the bi-directional shift approach. Two other models less commonly used include a correlation between PENT failure time (a laboratory coupon specimen used to determine SCG resistance of various resins) and service life, and a procedure to integrate SCG initiation and propagation processes to predict failure times.

A critical review of these models with regard to the acceleration factors for temperature, stress, and stress intensity factors is provided in the paper. As expected, elevated service temperature has the most significant impact on the predicted service life using the various models. Areas where improvements are needed in these forecasting models for newer generation PE100 or PE4710 bimodal resins are identified.

Prabhat Krishnaswamy, Vice-President
Do Jun Shim, Principal Engineer
Engineering Mechanics Corporation of Columbus (Emc2)