Fluid flow in a diametrally expanded CANDU fuel channel – Part 1: Experimental study

• Published in: Nuclear engineering and design Vol. 357; p. 110371
• Main Authors: Bruschewski, M., Piro, M.H.A., Tropea, C., Grundmann, S.
• Format: Journal Article
• Language: English
• Published: Amsterdam Elsevier B.V 01.02.2020; Elsevier BV
• Subjects: 3D velocity measurements; Acceleration; Aging; Bundling; CANDU; Computational fluid dynamics; Computer applications; Computer simulation; Diameters; Flow rates; Flow system; Flow velocity; Fluid dynamics; Fluid flow; Fuel channel flow; Heavy water reactors; Hydrodynamics; Large eddy simulation; Laser doppler velocimeters; Magnetic resonance; Magnetic Resonance Velocimetry; Measurement techniques; Nuclear fuel elements; Nuclear fuels; Nuclear reactors; Pressure; Reactors; Velocimetry; Velocity; Velocity distribution; Fuel channel flow; 3D velocity measurements; Magnetic Resonance Velocimetry; CANDU
• ISSN: 0029-5493; 1872-759X
• DOI: 10.1016/j.nucengdes.2019.110371

•Fluid flow measurements akin to end-of-life CANDU fuel channel conditions.•Flow bypass effect measured for the first time in a diametrally crept CANDU fuel channel.•Magnetic Resonance Velocimetry measurements provide CFD grade validation data. The overall aim of this study is to develop and validate experimental and computational capabilities to provide high-resolution isothermal fluid velocity data for nuclear reactor investigations. The current work presents full-field velocity data in a 1:1 replica of a 37-element CANDU nuclear fuel bundle using the Magnetic Resonance Velocimetry (MRV) measurement technique. The diameter of the pressure tube is 6% larger than at beginning-of-life to simulate the long-term aging process in an actual CANDU nuclear reactor due to diametral creep at end-of-life conditions. In Part 1, the MRV technique is presented with a focus on measurement errors and the estimation of these errors. The three-dimensional, three-component mean velocity field inside the fuel channel, including the subchannel flow between the fuel elements, is provided at a resolution of 0.78 × 0.78 × 0.78 mm3. The MRV data is validated through bulk flow rate calculations and Laser Doppler Velocimetry measurements conducted at selected locations inside the flow system. Overall, the deviations in the full-field MRV data are low except for local differences in regions with high convective acceleration and high velocity. The measurement accuracy can be improved by applying newly developed MRV sequences that are insensitive to these effects. In this specific experiment, it is found that up to 35% of the coolant bypasses the fuel bundle in a 6% diametrally expanded pressure tube containing a single bundle. The data is used as a reference for a companion paper (Part 2), which presents a computational fluid dynamics approach using an implicit large eddy simulation. In conclusion, the presented experimental methodology provides accurate full-field mean velocity data in complex channel geometries. More realistic flow experiments in replicas of entire fuel channels can be investigated in detail with relatively little effort and in a short time.