A Semi-Passive Containment Cooling System Conceptual Design

Abstract

The objective of this project was to investigate a passive containment cooling system (PCCS) for the double concrete containment of the Korean Next Generation Reactor (KNGR). Two conceptual PCCS designs: the thermosyphon loop and the internal evaporator-only (IEO) were studied. Based on their requirements, a number of full-scale single-tub experiments have been conducted to investigate the performance of the evaporator, the key component in both PCCS designs. The thermosyphon loop design consists of an evaporator (with integrated exit steam separator) and a condenser heat exchanger. The evaporator heat exchanger is located in the containment atmosphere; on its outside tube surfaces steam condensation in presence of noncondensable gases takes place. The condenser heat exchanger is placed in a large water pool located exterior to the containment building; its storage capability serves as the final heat sink. The numerical simulation in GOTHIC of this design shows that, depending on the water pool initial temperature, ten to fourteen thermosyphon loops are needed in order to keep the containment temperature and the total pressure below the design values for the design basis accident (60 psia) and three-to-five loops for the severe accident (120 psia). The IEO design is similar to the PCCS concept using internal condensers discussed earlier by KAIST. The difference between the IEO and the thermosyphon loop is that the steam exciting the evaporator is directly vented to atmosphere in the IEO rather than the exterior condenser in the thermosyphon loop design. The target of this system is to keep containment pressure below 8.3 bar (12 psia) in the absence of any other containment cooling following a severe accident, which postulates core damage and hydrogen combustion. A DBA scenario (LB LOCA, ECCS flow and no spray) and a severe accident scenario (LB LOCA without ECCS and containment spray flow, 100% Zr oxidation and complete hydrogen combustion), as used in KNGR safety analyses (similar to those in the standard safety analysis report for SYSTEM 80+) were modeled using the GOTHIC computer code. GOTHIC performance analysis of the IEO for the DBA condition shows that this concept can likely meet the design peak pressure of 60 psia, if 10 IEOs are used assuming that the separator water level is sufficiently low. However it is inherently difficult to meet the second design criterion of half of peak design pressure within 24 hours because the temperature difference between the containment and the IEO wall is low in the long term. Fir the severe accident, even with two IEOs, there is no problem in meeting the design criteria of 120 psia during the long-term period, with a generous margin. In addition the peak pressure is just 110 psia even assuming 100% zirconium oxidation and the complete burning of hydrogen. The fouling effect by aerosols on the IEO performance was calculated to be negligible. Judging from the above findings for the performance analysis involving DBAs and severe accidents, it is concluded that the IEO has considerable merit for severe accident mitigation and is worthy of further evaluation. A smooth tube, an axial-finned tube and a radial-finned tube have been tested to experimentally estimate the performance of the reference smooth evaporator tube and the enhancement factors, which may be achieved by finning smooth tubes. An empirical correlation has been developed for numerical analysis use. The Diffusion Layer Model (DLM) has been recommended for use beyond the range of this empirical correlation. Condensation in the presence of helium and tube bundle effects were also studied. Theoretical analysis and experimental results of the two finned tubes suggested an enhancement factor of 4 be used in GOTHIC simulation of the PCCS concept based on smooth tube modeling.