PREDICTION OF SUBCOOLED WATER FLOW BOILING CRITICAL HEAT FLUX (CHF) AT LOW PRESSURE

In vessel retention (IVR) is one of the key severe accident mitigation strategies to maintain reactor pressure vessel (RPV) integrity. IVR designs utilize the reactor pressure vessel lower head to contain molten fuel and rely on external reactor vessel cooling (ERVC) to remove decay heat. The capacity of ERVC is limited by the critical heat flux (CHF) of flow boiling on the outside of the reactor vessel surface. Therefore, the determination of critical heat flux (CHF) is crucial to predict whether the adoption of IVR would be successful in mitigating severe accidents. In 1999, Celeta et.al proposed a superheated layer vapor replenishment model. In this model they postulated that CHF would occur when the superheated layer was occupied by the vapor blanket coming into contact with the heated wall and they successfully predicted the critical heat flux in subcooled water flow boiling under high mass flux, high liquid subcooling and low/medium pressure conditions. To evaluate the practicability of this model in predicting CHF under IVR conditions, CHF experiments were performed under natural circulation conditions on the experiment facility "Test of External Vessel Surface with Enhanced Cooling" (TESEC). Experiments are conducted in a 30 mm wide, 61mm high rectangular flow channel with a 200 mm long heated surface along the flow direction. Two quartz windows are installed at the sidewalls of the flow channel for visualization. In order to simulate various positions of the reactor lower head, experiments at different inclination angles of the test section were conducted. The high speed visualization data at CHF point at various inclination angles were processed and analyzed by a MATLAB code developed by the author. The vapor blanket thickness at various inclination angles was measured from the visualization data and was also predicted by the Celeta model. By using geometry data from high speed images, CHF values were calculated by Celeta model and compared with the experimental results at various inclination angles. Limitations of the Celeta model in adaptation of predicting CHF under IVR conditions were further discussed.