PSI - Issue 60

Keshav Mohta et al. / Procedia Structural Integrity 60 (2024) 402–410 Author name / Structural Integrity Procedia 00 (2019) 000 – 000

404

3

2. Methodology The adapted methodology encompasses employment of detailed transient computational fluid dynamics (CFD) analysis and transient dynamic finite element analysis (FEA) with contact simulation to evaluate the normal work rate, and thereafter, based on the evaluated normal work rate, the extent of fretting wear is determined in terms of average saturated fret depth at the locations in contact with bearing pads. The detailed description is as following. 2.1. Archard’s Wear Model As per Archard’s wear model, the volumetric material removal rate due to fretting is proportional to the n ormal work rate, and is given as following relationship: V̇ = K Ẇ N Where, V̇ = Material volume removal rate (m 3 / s) K = Empirical wear coefficient (Pa -1 ) Ẇ N = Pressure tube to bearing pad normal work rate (W) The work rate may be calculated as, Ẇ N = (1/ Δ t) ∫ F N dS Where, F N = normal force (N) S = sliding displacement (m) Δt = time interval (s) The wear coefficient, K, depends on material combination and the operating conditions. Yetisir and Fischer (1997) have suggested a range of 500-2500×10 -15 Pa -1 for zirconium at coolant inlet temperatures (265°C). In present work, it is taken as geometric mean of both the extremes at 265°C, i.e. 1118×10 -15 Pa -1 . 2.2. Assumptions and Considerations in Numerical Assessment Fretting in an inherently complex phenomenon. The interaction between bearing pad (BP) and pressure tube depends on many factors such as driving turbulent forces, irradiation effects and morphology of contact surfaces and material etc. The underlying complexity and uncertainties attached with these variables do not allow for completely deterministic assessment of fretting wear, and reasonable idealization and simplifications are needed. In view of this, coupled fluid- structure interaction has not been modelled and it is assumed that the change in geometry due to hydrodynamic forces is negligible and the flow domain is not affected. Similarly, the material removal due to fretting is not updated in finite element model, rather is calculated from post-processing of the analysis results. For a conservative assessment, it has been assumed that the entire material removal occurs from PT. 2.3. Estimation of Normal Work Rate The normal work rate is not available directly as an output of FEA and is computed externally using the FEA results. For its estimation, contact force and relative sliding time histories are required. A schematic representation of bearing pad-pressure tube contact interface is shown in Fig. 1. The step-by-step procedure for evaluation of contact force time histories and relative sliding time histories has been shown in Fig. 2. The transient hydrodynamic forces acing on fuel elements are evaluated first using 3D transient dynamic CFD analysis. These are then used as input to transient dynamic finite element analysis, with contact modeled between fuel bundle and pressure tube. For evaluation of the work rate, a portion of the time history from the stabilized response duration is considered. Material wear at the interface of interacting surfaces results in change of local topology/ geometry during fretting. With continuous fretting, the crests of the interacting surfaces wear out gradually, and larger areas from the