Reduced-scale water test of natural circulation for decay heat removal in loop-type sodium-cooled fast reactor

Abstract

Water tests of a loop-type sodium-cooled fast reactor have been conducted to physically evaluate the natural circulation characteristics. The water test apparatus was manufactured as a 1/10-scale mock-up of the Japan Sodium-Cooled Fast Reactor, which adopts a decay heat removal system (DHRS) utilizing natural circulation. Tests simulating a variety of events and operation conditions clarified the thermal hydraulic characteristics and core-cooling performance of the natural circulation in the primary loop. Operation conditions such as the duration of the pump flow coast-down and the activation time of the DHRS affect the natural circulation characteristics. A long pump flow coast-down cools the upper plenum of the reactor vessel (RV). This causes the loss of the buoyant force in the RV. The test result indicates that a long pump flow coast-down tends to result in a rapid increase in the core temperature because of the loss of the buoyant force. The delayed activation of the DHRS causes a decrease in the natural circulation flow rate and a temperature rise in the RV. Flow rate deviation and a reverse flow appear in the parallel cold-leg piping in some events, which cause thermal stratification in the cold-leg piping. The DHRS prevents the core temperature from fatally rise even for the most severe design-basis event, in which sodium leakage in a secondary loop of the DHRS and the opening failure of a single damper of the air cooler occur simultaneously. In the water test for the case of siphon break in the primary loop, which is one of the design extension conditions, a circulation flow consisting of ascendant and descendant flows in the core channels suppresses the rise in the core temperature instead of the primary coolant loop. However, the test result suggests that the present DHRS has insufficient performance to remove the decay heat in the case of siphon break. Repeated tests have shown that the results of the water test have sufficiently high reproducibility in terms of the flow rate of the primary loop to be used for the verification of numerical codes.