Information Notice No. 92-20: Inadequate Local Leak Rate Testing

                                UNITED STATES
                        NUCLEAR REGULATORY COMMISSION
                    OFFICE OF NUCLEAR REACTOR REGULATION
                           WASHINGTON, D.C.  20555

                                March 3, 1992


NRC INFORMATION NOTICE 92-20:  INADEQUATE LOCAL LEAK RATE TESTING


Addressees

All holders of operating licenses or construction permits for nuclear power 
reactors.

Purpose

The U.S. Nuclear Regulatory Commission (NRC) is issuing this information 
notice to alert addressees to problems involving local leak rate testing 
(LLRT) of containment penetrations under Part 50 of Title 10 of the 
Code of Federal Regulations (10 CFR 50), Appendix J.  It is expected that 
recipients will review the information for applicability to their facilities 
and consider actions, as appropriate, to avoid similar problems.  However, 
suggestions contained in this information notice are not NRC requirements; 
therefore, no specific action or written response is required.

Description of Circumstances

Quad Cities Station, Unit 1

On February 26, 1991, the Commmonwealth Edison Company (the licensee) 
performed a Type B LLRT on the containment penetration bellows for 
penetration X-25 at Quad Cities Station, Unit 1, and found an acceptable 
measured leakage rate of 6 standard cubic feet per hour (scfh).  The 
licensee performed this LLRT by pressurizing the volume between the two 
plies of the bellows through a test connection.  On February 28, while 
performing the primary containment integrated leak rate test (ILRT) under 10 
CFR 50, Appendix J, the licensee found excessive air leakage from the 
penetration.  The licensee recognized the inconsistencies between the LLRT 
data for the penetration and the ILRT results and began a test program to 
determine the source of the error.  Using a blank flange on the containment 
side of the bellows, the licensee pressurized the bellows for a "local 
ILRT," which yielded a leak rate of 137 scfh.  The licensee also repeated 
the Type B LLRT with holes drilled in the bellows.  This LLRT result (8 
scfh) was only slightly higher than the previous LLRT result.  The results 
of this test program led the licensee to conclude that it is not possible to 
perform a valid Type B LLRT on this type of bellows assembly.  The licensee 
replaced the bellows for penetration X-25 due to cracks identified by the 
tests.  Commonwealth Edison is also investigating alternative test methods 
that would provide accurate LLRT results for bellows penetration assemblies.  
This problem was reported to the NRC under Title 10 of the 
Code of Federal Regulations, Part 21 (10 CFR 21).


9202260208 
.

                                                            IN 92-20
                                                            March 3, 1992
                                                            Page 2 of 3


Dresden Nuclear Power Station, Unit 2

On December 17, 1990, the Commonwealth Edison Company (the licensee) found a 
leakage rate significantly greater than the maximum allowed during the 
pressurization phase of its ILRT.  The licensee identified the source of the 
leak as the inboard flange of the torus purge exhaust inner isolation valve 
with an estimated leakage rate of approximately 25 weight percent per day at 
15 psig.  The licensee had last performed maintenance on this valve during 
the previous outage.  Although a LLRT had been performed on the valve 
following the maintenance, the test did not challenge the inboard flange.

Perry Nuclear Power Plant, Unit 1

On July 7, 1989, the Cleveland Electric Illuminating Company (the licensee) 
discovered a leak through the inboard flange of the seal leakoff line on a 
relief valve for the residual heat removal system during the plant's first 
periodic ILRT.  The licensee had performed maintenance on the valve three 
times from 1986 to 1989.  In each case, the licensee had performed a LLRT 
following the maintenance.  However, the LLRT did not challenge the inboard 
flange.

Clinton Power Station, Unit 1

On December 18, 1990, the Illinois Power Corporation (the licensee) found 
that lines from the residual heat removal system relief valves were not 
water-sealed under post-accident conditions as previously indicated in its 
safety analysis report.  These lines were intended to terminate below the 
suppression pool minimum drawdown level, allowing the water to maintain a 
seal on the containment isolation valves under accident conditions.  The 
problem was first identified when it was discovered that a line, considered 
to be water-sealed, included a vacuum breaker.  The vacuum breaker would 
open following an accident, bypassing the water seal.  The licensee 
investigated this condition and found that a number of other lines that 
empty into the suppression pool either contained flanges or terminated above 
the pool minimum drawdown level.  Since these lines would have been open to 
the containment atmosphere following an accident, the associated isolation 
valves should have been tested for leakage using Type C air tests.  To 
correct this problem, the licensee removed the vacuum breaker connections 
and the flanges and extended the pipes to ensure that a water seal would be 
maintained.

Discussion

Steel expansion bellows are used on piping penetrations in many plants as 
part of the containment isolation scheme.  A Type B LLRT is performed on the 
bellows periodically to verify that containment integrity is being 
maintained.  The event at Quad Cities revealed that the LLRT performed 
between the two plies could not be used to accurately measure the leakage 
rate that would occur through the bellows under accident conditions.  The 
two plies of the bellows were in contact with each other, restricting the 
flow of the test medium to the crack locations.  The NRC staff investigated 
and found that this problem is not isolated to the bellows manufactured by 
the vendor involved at Quad Cities.  Any two-ply bellows of similar 
construction may be susceptible to this problem.
.

                                                            IN 92-20
                                                            March 3, 1992
                                                            Page 3 of 3


The NRC granted an exemption from the Type B testing requirements of 
10 CFR 50, Appendix J, to Commonwealth Edison for Quad Cities and Dresden on 
February 6, 1992.  The exemption covers the testing of the two-ply bellows 
discussed in this information notice because no valid Type B LLRT can be 
performed on these bellows.  The exemption specifies an alternative program 
of bellows testing and replacement that involves testing with air at a 
reduced leakage limit, testing any leaking bellows with helium (sniffer 
testing), replacing bellows that are unacceptable, and performing an ILRT 
each refueling outage until all of the bellows have been replaced with 
testable bellows.

The two events involving leaking flanges occurred because the licensees 
failed to consider all possible leakage paths when they established their 
leak rate test programs.  Both licensees identified the valves involved in 
the events as containment isolation barriers, but they failed to consider 
the gasketed flanges as leakage paths.  Both licensees tested the isolation 
valves in the reverse direction which did not challenge the flanges 
properly.  Any containment isolation valve could have this problem, 
particularly if the valve is tested in the reverse direction or if both 
valves on a penetration are outside of containment.

Appendix J to 10 CFR Part 50 requires all licensees to perform local leak 
rate testing on containment isolation valves.  The licensees for some 
plants, including Clinton, have received credit from the NRC for maintaining 
a water seal on the valves instead of performing local leak rate testing 
using air as the test medium.  One requirement for a valid water seal is 
that the penetration have no potential air leakage paths, such as flanges or 
vacuum breakers.  For those lines that depend on the minimum water level in 
the suppression pool to prevent leakage, it is expected that the lines 
terminate below the minimum suppression pool level in the actual plant 
configuration.

This information notice requires no specific action or written response.  If 
you have any questions about the information in this notice, please contact 
one of the technical contacts listed below or the appropriate Office of 
Nuclear Reactor Regulation (NRR) project manager.




                                   Charles E. Rossi, Director
                                   Division of Operational Events Assessment
                                   Office of Nuclear Reactor Regulation

Technical contacts:  M. P. Phillips, RIII         A. J. Kugler, NRR
                     (708) 790-5530               (301) 504-2828

                     F. A. Maura, RIII            J. C. Pulsipher, NRR
                     (708) 790-5696               (301) 504-2811

Attachment:  List of Recently Issued NRC Information Notices

.