Capability of RELAP5 MOD3.3 code to simulate density wave instability in parallel narrow rectangular channels

Density Wave Oscillation (DWO) is one of the most common types of instability in boiling systems. Many experiments were conducted to investigate DWO in various geometry channels during last several decades, but what we always want to know is the stability characteristic of one specific system. In this paper, DWO in narrow rectangular channels which have a cross section of 25 mm × 2 mm and a heated length of 1000 mm are investigated by means of RELAP5/MOD3.3. Then 62 DWO experimental data in two narrow rectangular channels and 60 DWO experimental data in two narrow rectangular channels with a large bypass over 1-10 MPa of pressure, 200-800 kg/m2 s of mass velocity, and 10-50 °C of inlet subcooling conditions are adopted to verify the RELAP5 results. The thermal hydraulic behavior, parametric effect study, and flow instability boundary in narrow rectangular channels are simulated and compared with the experimental results. The errors associated with the predictions are addressed. The density, flow rate and pressure drop characteristics during DWO condition are also analyzed. Compared with experimental flow instability boundary, 90% prediction data for two narrow rectangular channels are between ±20% errors and 85% data for two narrow rectangular channels with a large bypass are between ±20% errors. The error becomes higher under the low pressure with high subcooling, low flow rate condition and high pressure condition. During DWO condition, density, flow rate and pressure drop of the heated channels oscillate periodically; flow rates are different along the channel and the inlet and outlet flow rates oscillate almost in reverse and the non-boiling region pressure drop oscillates out-of-phase with respect to boiling region pressure drop. The results aim to be a contribution to the assessment of the code capability to detect the onset of DWO in narrow rectangular channels. The purpose of the paper is to highlight strengths and weaknesses of the code to simulate DWOs as well as to evaluate carefully the suitable numerical settings necessary to assure a correct prediction of the phenomenon.