Answers are grounded in this paper and cite their sources.
Experimental investigation demonstrates 3D flow dynamics and particle clogging in reactor filters, highlighting benchmark data for computational fluid dynamics models.
The reliability of the primary cooling circuit in a pressurized water reactor (PWR) is crucial for safe operation. Debris fretting, caused by solid particles in the coolant, can damage fuel rods and lead to the leakage of fission products into the primary circuit coolant. Optimized filters in the fuel assembly bottom nozzle (BNO) can capture debris while minimizing pressure loss and reducing clogging risk. To investigate the cooling flow through the bottom nozzle and filter, Magnetic Resonance Velocimetry (MRV) was employed using a 3 Tesla magnetic resonance imaging (MRI) scanner. The study focused on a simplified bottom nozzle, filter, and the leading edge of a 5 × 5 fuel rod bundle. Without requiring optical access to the complex filter structure, MRV provided high-resolution flow measurements. It obtained 3D velocity fields and, for one unclogged case, 3D Reynolds stress tensors. The pressure drop across the filter was measured separately. Wire-like particles were introduced sequentially, enabling reconstruction of their positions and analysis of their impact on flow. At a Reynolds number (Re) of 50,250, and with up to 100 particles, the filter test resembled standard conditions. A clogging scenario was created by introducing 240 additional particles at Re = 20,000. Using MRI data, the clogged filter's geometry was reconstructed for computational fluid dynamics (CFD) implementation. These CFD-grade measurements provide unique experimental data for validating particle motion and clogging models. Time-averaged velocity and Reynolds stress tensor data provide insights into how particles and agglomerations influence flow through the filter and around fuel rods, informing design improvements for enhanced reactor safety and efficiency.
No takes yet. Share an insight, caveat, or question.
Klempau et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: