New knowledge and experiences of flow induced fretting in PWR fuel assemblies

One of the most important demands for a fretting free nuclear fuel assembly design is to verify that the fuel assembly is able to resist any fretting damages during its lifetime. The reactor experience and laboratory investigations have made clear to distinguish between 2 types of flow induced excitation: Turbulent Excitation (TE) and Self Induced Excitation (SIE). The former is the 'Conventional' flow excitation which is traditionally verified through integral endurance test; the latter, for which the rod (or assembly) vibration influences significantly the flow excitation and which may lead to a resonance for given flow conditions, requires a specific test protocol which allows to identify this phenomenon safely. For this purpose, an improved experimental approach to investigate the behavior of a fuel assembly design towards SIE has been developed: an integral fuel assembly flow test is performed using an assembly 'optimized' in terms of vibratory characteristics and a specially designed loop (PETER loop), able to simulate different flow boundary conditions, such as different inlet flow conditions, neighboring fuel assembly influence (different grid or bottom nozzle pressure loss coefficients, intermediate flow mixing grids), fuel assembly bow, baffle shaking or vibrations of neighboring fuel assembly. During this flow test, the resulting fuel rod and fuel assembly vibration behavior are measured under many parameterized flow conditions and the verifications of 'non resonance' or instability are done. Moreover, the maximum rod vibration amplitudes are recorded for the conditions of a flow bounding those expected for the reactor. In a second step, the design can be verified towards TE through two different approaches which are considered as complementary or alternative: 1)The conventional approach which consists of introducing a fuel assembly prototype into a flow loop for 1000 hours under bounding flow conditions and then to inspect the fuel rods for potential wear marks. 2)The alternative approach consists of applying the maximum vibrations measured in PETER loop on a single test rod installed into a specially designed fretting test facility or 'Autoclave'. The response of the fuel rod is monitored in respect of different aspects (vibration amplitudes,excitation spectrum) and the result is evaluated in term of wear depth. Further, this approach makes it possible to perform comparative evaluations of different spacer designs under quantified and controlled conditions and to derive guidelines for the design of future products. Both approaches must use rod supporting conditions representative of what is expected in the reactor at EOL. (authors)