Numerical simulations with RELAP5-3D of the first experimental campaign on In-box LOCA transient for HCLL TBS

• Published in: Fusion engineering and design 2021-02, Vol.163, p.112160, Article 112160
• Main Authors: Venturini, A., Utili, M., Forgione, N.
• Format: Article
• Language: English
• Subjects: Bismuth; Elastic waves; HCLL TBS; In-box LOCA; Lithium; Lithium-Lead Eutectic; RELAP5-3D; Simulation; THALLIUM; Thermophysical properties; Wave propagation
• ISSN: 0920-3796; 1873-7196
• DOI: 10.1016/j.fusengdes.2020.112160

•Results of RELAP5-3D simulations of an experimental campaign in a mock-up of the HCLL TBS LiPb loop.•RELAP5-3D is able to accurately reproduce the pressure wave propagation in the facility.•Two main general differences between numerical and experimental results have been highlighted.•Average discrepancy of about 3.5 bar between experimental data and numerical results on 5 tests.•The code accurately predicts the first pressure peak, the most relevant input for the designers. In this work the capabilities of RELAP5-3D are tested against the experimental data gathered in the first experimental campaign on THALLIUM (Test HAmmer in Lead LIthiUM). THALLIUM, a LiPb facility, was built in 2015 at ENEA R.C. Brasimone with the aim to study the In-box LOCA by reproducing the LiPb loop of the HCLL TBS and the pipe forest at 1:1 scale. HCLL TBS (Helium Cooled Lithium-Lead Test Blanket System) was one of the Breeding Blanket concepts selected for ITER. The ITER classification system identified the In-box LOCA as a critical accidental transient for this type of TBS. For this reason, codes able to reproduce the consequences of this transient are needed to support the design of the LiPb loop. The simulations with RELAP5-3D used LBE (Lead-Bismuth Eutectic) as system fluid, as the thermophysical properties of LiPb are tabulated only up to a maximum value of 40 bar. The results of the simulations suggest that RELAP5-3D is able to reproduce the pressure wave propagation in the facility, with an average discrepancy of about 3.5 bar (i.e. an average percentage discrepancy of about 9.7 %), but also highlight two general discrepancies between computations and experiments.