Experimental Measurement of Suppression Pool Void Distribution During Blowdown in Support of Generic Issue 193 (NUREG/CR-7186)

On this page:

Download complete document

Publication Information

Manuscript Completed: March 2011
Date Published: September 2014

Prepared by:
M. Ishii, T. Hibiki, S. Rassame,
M. Griffiths, D. Y. Lee, P. Ju, J. Yang,
S. L. Sharma, and S. W. Choi

Purdue University
School of Nuclear Engineering
West Lafayette, IN 47907

A. Ireland, NRC Project Manager

A. Velazquez-Lozada, NRC Technical Monitor

Office of Nuclear Regulatory Research
U.S. Nuclear Regulatory Commission
Washington DC 20555-0001

Availability Notice


The possible failure of an Emergency Core Cooling System (ECCS) train due to a large amount of entrained gas in the ECCS pump suction piping during a loss of coolant accident (LOCA) is one of the potential engineering problems faced in a Boiling Water Reactor (BWR) power plant. The void fraction distribution and void penetration in the Suppression Pool (SP) during blowdown from a LOCA are key parameters necessary to analyze potential gas intrusion into the ECCS pump suction piping. To study void fraction distribution and void penetration in the Suppression Pool (SP) during blowdown, two sets of experiments, namely steady state tests and transient tests, were conducted using the Purdue University Multi-Dimensional Integral Test Assembly for ESBWR applications (PUMA-E) facility. The design of the test apparatus used is based on a scaling analysis from a prototypical BWR containment (Mark I) with consideration of downcomer size, SP water level and downcomer water submergence depth. Several instruments were installed to obtain the required experimental data such as the gas volumetric flow, void fraction, pressure, and temperature.

For the steady state tests, the air was injected through a downcomer pipe in the SP. Sixteen tests with various air volumetric flow rates, downcomer void conditions, and air velocity ramp rates were performed. Two periods of the experiment, namely, the initial air injection period and the quasi-steady period are observed. The initial air injection period gives the maximum void penetration depth. The quasi-steady period provides less void penetration, but with oscillation in the void penetration. It was found that the air volumetric flow rate has a minor effect on the void fraction distribution and void penetration during the initial air injection period in the range of high air volumetric flow rate conditions while it strongly impacts the void fraction distribution and void penetration during the quasi-steady state for the entire range of air flow rate conditions. The initial downcomer void conditions were found to strongly affect the void fraction distribution and void penetration during the initial period. The air velocity ramp rates were found to have a minor impact on the void distribution and penetration in both periods.

For the transient tests, sequential flows of air, steam-air mixtures, and pure steam with the various flow rate conditions were injected from the Drywell (DW) through a downcomer pipe in the SP. Eight tests were conducted at various gas volumetric fluxes at the downcomer, two different downcomer sizes, and two different initial air concentration conditions in the DW. Three periods of the experiment, namely, initial period, quasi-steady period, and chugging period are observed. The void penetration depth was maximum in the initial period and reduced in the quasi-steady period. The penetration of noncondensable gases during the chugging period, which occurs at the end of the transient, reached depths similar to those observed during the initial period. It was determined that the void distribution and area of void penetration in the SP is governed by the gas volumetric flux at the downcomer and by air concentration in the downcomer. It is noted that the transient conditions were well scaled for the initial period but not necessarily well scaled to simulate the chugging phenomena. Chugging is a complex phenomenon that depends primarily on periodic sudden condensation of steam into colder water, but also depends on gas volumetric flux, noncondensable gas concentration, frequency of the phenomenon, heat transfer, and subcooling, as well as the downcomer and suppression pool geometry. The rudimentary scaling methods used here are not suitable for use with such complex phenomenon. Instead, more specific and advanced scaling techniques would be needed.

Page Last Reviewed/Updated Wednesday, March 24, 2021