Overview of Fire Barriers

In response to the Browns Ferry fire, the U.S. Nuclear Regulatory Commission (NRC) created fire protection regulations and guidelines to ensure that nuclear power plants (NPPs) can be safely shut down in the event of a fire. An important requirement of these rules was to require the protection of redundant pathways of equipment and cables required to place the plant in a safe shutdown state. When these redundant pathways of cables and equipment were in the same room or fire area, licensees often installed electrical raceway fire barrier systems (ERFBSs) to achieve the required separation.

The NRC's concern with the performance of fire barriers at NPPs began with the failure of Thermo-Lag to pass performance tests in 1989. The tests were done for the Gulf States Utilities Company after visual observations of degradation of Thermo-Lag at River Bend Station. In 1992, two sets of full-scale fire endurance tests on Thermo-Lag were conducted by Texas Utilities Electric Company for Comanche Peak Steam Electric Station, with similar results. Also in 1992, the NRC sponsored a series of small-scale fire endurance tests at the National Institute of Standards and Technology (NIST). Again, 1-hour and 3-hour rated Thermo-Lag barrier material failed to consistently provide its intended protective function.

Because of questions about the ability of 1-hour and 3-hour rated Thermo-Lag fire barrier material to perform its specified function, and because of the widespread use of Thermo-Lag in the nuclear industry, the NRC issued numerous generic communications to inform licensees of the Thermo-Lag test results and to request that licensees implement appropriate compensatory measures and develop plans to resolve any noncompliances with the fire protection regulations.

In the wake of the Thermo-Lag fire barrier developments, the NRC examined licensees' uses of other fire barrier materials. The staff performed a survey of fire barrier materials and systems that licensees used to meet the NRC's fire protection requirements and guidelines and identified a number of systems including Promatec Hemyc. In 1993, the staff contacted the vendors of these fire barrier systems to obtain technical information concerning their systems, such as fire endurance characteristics, ampacity derating characteristics, and the technical bases for meeting the NRC's technical requirements. The staff's examination of this material led to the issuance of a number on Information Notices concerning the adequacy of particular non-Thermo-Lag fire barriers, but not one for Hemyc.

Also in 1993, NIST began a series of small-scale fire tests of fire barrier materials. One of these tests, FR 3994, "Pilot-Scale Fire-Endurance Tests of Fire-Barrier Mats, Blankets, and Panels," included Hemyc. These pilot-scale tests were carried out to obtain fire endurance information and did not identify the Thermo-Lag failure mode (e.g., burnthrough and other physical failures) in Hemyc, and did not detect the later-identified Hemyc shrinkage failure mode. The small-scale tests were not designed to evaluate the performance of full-size barriers, and were for screening purposes only. Additionally, the test setup could not replicate the actual full-scale physical configuration of Hemyc installations.

In March 1994, the NRC issued Generic Letter 1986-10, Supplement 1, "Fire Endurance Test Acceptance Criteria for Fire Barrier Systems Used To Separate Redundant Safe Shutdown Trains Within the Same Fire Area." This document refined and clarified the NRC's fire endurance test acceptance criteria for fire barriers. Previously, some licensees used functional criteria, for example, as published by American Nuclear Insurers. In 1999, during the pilot fire protection inspections at Shearon-Harris, inspectors discovered that the fire endurance tests the licensee used to qualify its Hemyc installations did not satisfy the Generic Letter. The NRC's evaluation of this issue resulted in Task Interface Agreement (TIA) 99-028. As a result of TIA 99-028, the NRC determined that the fire resistance of the Hemyc fire barriers was indeterminate. The NRC publicized these conclusions and began negotiations with the industry for an industry-led resolution of this issue; however, the industry declined to lead this initiative.

After concluding that corrective actions would not be required without a clear answer regarding Hemyc fire endurance ratings, the NRC initiated confirmatory fire tests in 2001. This test program was concluded in 2005 and performed based on the NRC guidance in Generic Letter 86-10, Supplement 1. The test results indicated that Hemyc did not achieve the fire endurance consistent with its rating for the configurations tested. That year, the NRC issued an Information Notice that described the results of the NRC-sponsored confirmatory testing of Hemyc. In 2006, the NRC issued Generic Letter 2006-03, "Potentially Nonconforming Hemyc and MT Fire Barrier Configurations," to aid in achieving final resolution of Hemyc issues.

The Generic Letter required licensees to address the Hemyc issues and describe how other fire barrier materials are capable of providing the appropriate fire resistance rating. By the end of calendar year 2007, all responses to the Generic Letter had been accepted by the NRC, and all of the licensing actions approved. All commitments made in the Generic Letter responses and approved licensing actions have entered the normal inspection process.

In May 2010, the NRC Office of Nuclear Regulatory Research published NUREG-1924, Electric Raceway Fire Barrier Systems in U.S. Nuclear Power Plants." NUREG-1924 documents the history of Electrical Raceway Fire Barrier Systems (ERFBS) used to protect post-fire, safe-shutdown functions. It provides a description of regulatory requirements and testing acceptance criteria, a detailed evaluation of each type of ERFBS, and a discussion of individual plants use of ERFBS. This report also documents the current state of the use of these barriers for nuclear power plants.