Pressure Tube Ballooning Experiments Analysis

Abstract

Integrity of pressure tubes is one of the milestones in achieving safe operation of nuclear power plants with CANDU reactors. The aim of a safety analysis is to demonstrate pressure tube integrity by showing that the pressure tube does not rupture both: (1) when it balloons (because of local strain), and (2) after it contacts the calandria tube. Therefore, the ability to model pressure-tube ballooning is a key step in the licensing analysis of a CANDU reactor during a postulated loss-of-coolant accident. AECB sponsored an experimental research project at the Stem Laboratory [1] to address the repeatability of pressure tube ballooning data and to demonstrate the effect of bearing pad fretting on rupture. This research project is part of a broader study on the effect of in-service degradation on the ballooning behaviour of pressure tubes. Newly performed experiments, in the Phase 5 of the research project [2], comprised the testing of nine pressure tube specimens to study the effects of hydride blisters and uniformly distributed hydrogen, and the composition of the internal pressurizing gas, including argon, steam, a steam-iodine mixture and hydrogen. A phase to test irradiated tubes has been put on hold pending results of COG tests. As a part of continuing support of the research project on pressure tube ballooning, the present work concentrated on resimulation of the newly performed experiments by using three computer codes, e.g., PTDFORM, PTSTRAIN and AECBALL. Intercode comparisons show that code calculated results for strain rates and time of failure are close to each other, for nonuniform ballooning under specified conditions. However, it can be seen that both codes largely under predict the creep rate and experimental failure strains. The factor, among others, that may be contributing significantly to the poor agreement between the codes is clearly nonlinear character of the straining. One of the sources of nonlinearity may certainly be attributed to the localized irreversible heat generation due to the deformation work done on the pressure tube, particularly around the location of the failure. This effect is not considered in either one of the codes used for pressure tube integrity analysis. Obviously, ballooning and straining in the plastic deformation region cannot be easily addressed with simple mechanistic models used in these codes.