Cable Fire Response Data for Fire Model Improvement (NUREG/CR-6931, Volume 2)
On this page:
Download complete document
This page includes links to files in non-HTML format. See Plugins, Viewers, and Other Tools for more information.
Manuscript Completed: January 2008
Date Published: April 2008
S.P. Nowlen, F.J. Wyant
Sandia National Laboratories
Risk and Reliability Analysis Department 6761
P.O. Box 5800
Albuquerque, NM 87185
H. Woods, NRC Project Manager
NRC Job Code N6125
Office of Nuclear Regulatory Research
This report documents the cable electrical performance and fire-induced failure test results from the Cable Response to Live Fire Project (CAROLFIRE). CAROLFIRE testing included a series of 78 small-scale tests, and a second series of 18 intermediate-scale open burn tests. The tests were designed to complement previous testing and to address two needs; namely, to provide data supporting (1) resolution of the ‘Bin 2’ issues as identified in Regulatory Issue Summary 2004-03 Revision 1 - Risk informed Approach for Post Fire Safe Shutdown Circuit Inspections and (2) improvements to fire modeling in the area of cable response to fires. The small-scale tests involved exposure of from one to six lengths of cable to grey-body radiant heating in a cylindrical exposure chamber called Penlight. The intermediate-scale tests involved exposure of cables in various routing conditions to open fires created by a propene (propylene) gas diffusion burner. In both test series cables were tested as individual lengths of cable, in bundles of from 3 to 12 cables, and in a limited number of tests, fully loaded electrical raceways. Cables were tested in cable trays, in conduits, and in air drop configurations. The intermediate-scale tests included exposure of cables both in the fire plume and under hot gas layer exposure conditions. A broad range of representative cable products were tested including both thermoset and thermoplastic insulated cables that are typical of the cable types and configurations currently used in U.S. nuclear power plants. All tests measured the cable thermal response using TCs placed both on the surface and embedded within the target cables, and cable electrical performance based on two different electrical monitoring systems. This volume of the three volume project report focuses on the second need area, namely, the fire modeling improvement. The test data gathered are presented and discussed in this context. The discussions focus in particular on data documenting the exposure conditions, the thermal response of the cables, and correlation of thermal response to the onset of electrical failure. A more detailed discussion of the cable electrical performance and failure data and the ‘Bin 2’ items is provided in Volume 1. Volume 3 was prepared by the National Institute of Standards and Technology (NIST) and documents the thermally-induced electrical failure (THIEF) model whose development was based on the CAROLFIRE test data. THIEF takes, as input, an estimate of the air temperature time history near a cable during a fire and predicts, as output, the temperature response of the cable. The time to electrical failure is then based on an assumed failure threshold temperature characteristic of the cable of interest.