Simulation of the loss of the residual heat removal of an integral test facility using computer code Cathare

Assessment of the ability of thermal hydraulic codes to correctly predict complex phenomena during reactor transients can be performed only by means of comparison against experimental data. The thermal hydraulic CATHARE V1.3U computer program bas been used to simulate International Standard Problem (ISP38) experiment conducted at BETHSY Integral Test Facility located in Grenoble, France. BETHSY is an integral test facility modeling the three-loop, 2775 thermal MW Framatome Pressurized Water Reactor (PWR). Its main objectives are to contribute to the assessment of CATHARE code and of the physical basis of PWR Emergency Operating Procedures. This investigation dealt with experimental simulation of the loss of the Residual Heat Removal System (RHRS) during mid-loop operation. It involved opening of the pressurizer manway and the steam generator outlet plenum manway simultaneously with switching on the power of heating rods, thus simulating loss of the RHRS. The total power was kept unchanged with the level 138 kW throughout the test. Mass discharge through both manways led to core boiling and uncovery. A gravity feed injection of cold water was actuated in the cold leg when cladding temperature at the top of heating rods reached 250 C. The test was stopped when the primary cooling system was filled back to mid-loop level. The transient was executed on the HP6400 workstation at Texas A and M University. The code predictions underestimated the time of the core uncovery and the actuation of the gravity feed injection due to the over-predicted mass discharge through steam generator manway during the initial stage of the transient. This was probably caused by CATHARE's miscalculation of the phase separation effect at the hot leg/surge tine tee junction and significant water entrainment into the surge line in the beginning of the test. It was found that the model of the upward tee junction needs to be refined for the low pressure range. Overall, the code's predictions were in a qualitative agreement with the experimental data. (author)