ML20056C092
| ML20056C092 | |
| Person / Time | |
|---|---|
| Site: | Diablo Canyon  |
| Issue date: | 03/16/1993 |
| From: | Culver N SAN LUIS OBISPO MOTHERS FOR PEACE |
| To: | Atomic Safety and Licensing Board Panel |
| References | |
| CON-#193-13755 92-669-03-OLA-2, 92-669-3-OLA-2, OLA, OLA-2, NUDOCS 9303300056 | |
| Download: ML20056C092 (53) | |
Text
. .. '
/%155
>~ '
SAN LJIS OBISPO MOTHERS FOR PEACE 3 3 I"'S I 9 before the ATOMIC SAFETY AND LICENSING BOARD '
+
In the matter of Pacific Gas and Electric Co. Docket No. S0-275-OLA - 7__ !
Diablo Congon Nuclear Power Plant 50-323-OLA !
Unit Nos. I and 2 ASLBP No. 92-SSS-03-OLA-2 i March 16, 1993 Son Luis Obispo Mothers for Peace !
Second Late-Filed Contention In accordunce with 10 CFR 2.714Co][i], this document supplements the San Luis Obispo Mothers for Peace Supplement to Petition to Intervene. On January 21, 1993, the Son Luis Obispo Mothers for Peace C"SLOMFP"] was granted a hearing and petition for leave to intervene in the proceeding
?
involving the proposed omendment of the operating licenses for the Diablo Canyon Nuclear Pc"wer Plant C"DCNPP"J, Units 1 and 2. This amendment would extend the life of those licenses by more than 13 years for Unit 1 and almost 15 years for Unit 2. For the reasons set forth below, the SLOMFP is submitting the following to supplement its original Supplement to Petition ;
to Intervene, j
U. The Son Luis Obispo Mothers for Peace contends that the interim i fire protuction measures in place at DCNPP to compensate for the faulty fire barrier materiol, Thermo-Log, are inadequate because the material ,
itself creates o fire hozord. The proposed license extension request should therefore be denied until this situation is resolved. l l
Basis: In their original contention, the SLOMFP noted that the Thermo-Log material is combustible and therefore itself a fire hozord. Son Luis Obispo Mothers for Peace Supplement to Petition to Intervene COctober ;
I 9303300056 930316 PDR
- ADOCK 05000275 i eon 3psg
i 25, ISS2] at 29. Additional information reinforces and extends that .
contention. !
- 1. Fire watch Tests conducted at National Institute of Standards and Tests on July {
i 15, 1992 document Thermo-Log failed the NRC cold side temperature limit in 22 minutes and burned through in 95 minutes. NRC Information Notice S2-55 ,
CJuly 27, 19923 ot 1, 2. CAttochment 13 SLOMFP contends that these time t fromes document o Thermo-Log failure rote where on hourly fire watch does !
not reasonablu ossure public safety at DCNPP.
f
- 2. Combustibility .
I l
Thermo-Log 330-1 hos been documented by the NRC to be combustible.
Results of Thermo-Log 330-1 Combustibility NRC Information Notice 92-82: ,
Furthermore, the total Testing CDecember 15, 19923 at 1. CAttachment 23 amount of heat released from the Thermo-Log exceeded that released from I
Gypsum board and was about equal to the heat released from fire-retordant 4 plywood." Jda at 2.
SLOMFP contends that the continued presence of Thermo- l the Log ct DENPP represents fuel for fire that connot be compensated for by 1
r i
intermittent observations of fire watch personnel. I
- 3. Seismic !
Professor of Civil Engineering and consultant to
- Dr. Philip L. Gould, ,
Thermal Science, Inc. CTSI), the manufacturer of Thermo-Log, hos performed I a seismic onolysis of the material. The NRC stoff reviewed Dr. Gould's ;
onelysis and determined that the material may crock or crumble into powdery i materic1 or fropments in the event of on earthquake. Office of Nuclear Reactor Regulation Portial Director's Decision under 10 CFR 2.205, Docket ,
i
- Nos. 50-995, 50-495, 50-329, 50-325, 50-900, 50-391, 50-397, 50-958 i r
2
o CFebruary 1, 1993]C"Portial Director's Decision") at 22, 23. The SLOMFP contends that this crumbled fire barrier material provides the potential scenario of loss of fire protection and fuel for the fire. PGBE's compensatory measures are not sufficient to prevent this scencrio.
- 4. Ampocity derating In NRC Inspection Notice 92-95, the stoff reported that TSI mode o calculation error on the ampacity derating factor for Thermo-Log. Because of the insulating effect of the fire barrier materiol, ampocity derating is needed to lower the current carrying capacity of cables enclosed in the electrical raceways. This insulation limits the ability of the cable insulation to shed heat and, if not accounted for, con create a fire. J d, at 29. NRC Inspector General David Williams submitted testimony before the House Subcommittee on Oversight and Investigation, chaired by Congressmon John Dingell CD-MI] on March 3, 1993 regarding Thermo-Log ampocity derating figures. The Office of Inspector General reports:
In 1986 ond 1987, the NRC received information through official reporting chonnels which indicated that the cmpocity derating figures for Thermo-Log were much higher than initially reported. One source of this information was TSI and the other source was the Comanche Peck Nuclear Power Plant. The Comanche Peck report explained that failure to consider the higher figures could cause power cables to exceed their design criteric and, if left uncorrected, could odversely of fect the sofety of plant operations. Transcript at B. CAttachment 33 SLOMFP contends that to date DCNPP hos yet to take steps to oddress r
this issue other than compensotory measures. PG&E continues operation with the realization that Thermo-Log con initiate fires or seriously degrade function of ccbles os to jeopardize public hecith and safety, t
- 5. Voids
)h
e a
In some locations where Thermo-Log must be wrapped or bent, voids or "delominotaans" have been documented to form. NRC Memorandum: Telecon with Rubin Feldman, TSI CDetober 21, 1592] CAttochment 9]. These voids provide fast burn-through ovenues that con potentially increase combustibility and accelerate the spread of fire.
- 5. Hose stream SLDMFP documents NRC ocknowledgement of "the failure of the barrier to pass o fire hose stream test" under the current standard fire test as defined by the Americon Society for Testing and Materials in ASTM E-119,
" Standards for Fire Resistance of Building Materials." Fortial Director's This consists of directing a stream of Decision CFebruary 1, 1993] at 18.
water onto the fire barrier immediately following the fire endurance test.
Thu success criterio is determined by demonstrating that no opening in the barrier developed which would allow water penetration to power cables and NRC tests potentially disable the safe shutdown capability of DENPP.
demonstrated that Thermo-Log 330-1 disintegrated under the required full bore hose stream nozzle test with fragments being projected across the test creo ptrking lot. '
SLDMFP contends that because Thermo-Log 330-1 is a realized fire load installed at DCNPP, fire watch personnel and other compensatory measures could not prevent this scenario.
SLDMFP further objects to the NRC ,
(wide proposed testing change from o solid hose stream to a fog nozzle ongle sproy] because it reduces the severity of the environmental condition E which this conservative test was designed to opproximate. ,
- 7. Irevestigations s
g .
V^ .
Based on NUREG-1150, " Severe Accident Risks: An Assessment for Five U.S. Nuclecr Power Plants," Fire Analysis, Appendix C, at C-128 CAttachment 5] NRC has postulated that a typical nuclear power plant will have three or four significant fires over its operating life. The document reports that
" Fires are o significont contributor to the overoll core domoge frequency, contributing anywhere from 7 percent to 50 percent of the total, considering contributions from internal, seismic, flood, fire, and other events." Yet PG&E continues to operate the DCNPP in clear violation of ;
i minimum fire protection standards and requirements as contained in 10 CFR Port 50.48 and 10 CFR 50, Appendix R. This situation hos come to the attention of officials and investigations are currently underway. On March i
3, 1993, NRC Chairman Ivon Selin was sworn in to testify before the House- !
Subcommittee on Oversight and Investigation, chaired by the Honorable John Dingell CD-MI] CAttochment 63 cn the admitted failure by NRC oversight of the Thermo-Log issue that has resulted in substonderd fire protection at '[
DCNPP. The SLOMFP further understands that the NRC Office of Inspector i General hos initiated a Grand Jury investigation into Thermo-Log problems I
and hos issued subpoenos from the utilities for all documents relating to the purchase and installation of Thermo-Log. United States District Court !
for the district of Maryland, Subpoeno to Testify before Grand Jury CFebroory 1, 19933. ;
Conclusion:
The NRC has odmitted that " fire watches are not a final i i
solution." Fortial Director's Decision CFebruary 1, 1993] ot 18. The presence of Thermo-Log in the plant is on obvious hozord. SLOMFP contends-that PG&E's interim measures are not adequate to compensate for this '
e potentially dangerous material. A 13 to 15 year extension to the 4
f i
.{
t operating licenses of DCNPP is certcinly not in order under these i
circumstances. !'
By order of the Atomic Justification for Late-Filed Contention: {
Screty and Licensing Board EJonuary 21, 1993], the SLOMFP was granted o !
f hearing on a portion of Contention V involving the implementation of the ,e interim fire prntaction measuruu put in picce at DCNPP to compers;te for l t
the faulty fare barrier, Thermo-Log I
1 t
- 1. SLOMFP has good cause tu file Contention U ct this time beccuse i' l
additianol and pertinent information has become avoilchle to SLOMFP since- f SLOMFP 1'
tha preparation of thuir originci contention COctober 25, 1992].
has proceeded as quickly as possible to evolucte this information cnd to 4
ossemble enough evidence in support of Contention U to satisfy the.
l l
s Y
Commission's stonderd for cdmissibility of the contention. s t
- 2. SLOMFP's participction in the litigation of this contention will l j
lead to the development of a sound record. SLOMFP hos obtained technical css 2 stance in preparing its cose on this issue. !
- 3. There is no other pcrty to this cose which con represent SLOMFP's l s
l '
t interests. I Admission of this contention at this time con be expected to j
9.
broaden and delay this proceeding.
However, any such delcy would not be )
the fault of SLOMFP. Moreover, the litigation of this issue would not l prevent or delcy the operation of DCNPP.
I 1
Respectfully submitted, V bt!
Noney Cu ver, President Son Luis Obispo Mothers for-Peace
Certificate of Service 93 W.R 19 P 2 :48 1 hereby certify that copies of the foregoing Son LuisIObispo-nothers for Peace Second Late-filed Contention have been served upon the fcllowing-persons by U.S. mail, first class.
Office of Commission Appellote Administrative Judge Adjudicotion Charles Bechhcefer, Choirman U.S. Nuclear Regulatory Commission Atomic Safety and Licensing Board Washington, DC 20555 U.S. Nuclear Regulatory Commission Washington, DC 20555 Administrative Judge Jerry Kline Administrative Judge Atomic Sofety and Licensing Board Frederick J. Shan U.S. Nuclear Regulatory Commission Atomic Sofety and Licensing Board Wcshington, DC 20555 U.S. Nuclear Regulatory Commission Washington, DC 20555 Edward O'Neill Ann P. Hodgdon, Esq. Peter Arth, Jr.
Office of the General Counsel Trumon Burns U.S. Nuclear Regulatory Commission Robert Kinosion Washington, DC 20555 Peter G. Fairchild, Isq.
Californio Public bsalities Commission Joseph B. Knotts, Jr., Esq 505 Von Ness Avenue Winston & Strown Son Francisco, CA S9102 1900 L Street, N.W.
Washington, DC 20005 Adjudicatory File Secretary of the Commission U.S. Nuclear Regulatory Commission Dccketing and Service Bronch Washington, DC 20555 U.S. Nuclear Regulatory Commission Washington, DC 20555 Robert R. Wellington, Esq.
Dichlo Canyon Independent Sofety Committee 857 Cass Street, Suite D Monterey, CA S3S90 Christopher Worner, Esq.
Richard Locke, Esq.
Pocific Gas and Electric Co.
77 Beale Street Son Francisco, CA 94106 Dated March 16, 1553, San Luis Obispo County, CA Jill ZomEk 1h_ ,,
200 20W 23; nu 7g:g 73, g7 mt
. UNITED STATES NUCLEAR REGULATORY C0KMISSION OFFICE OF NUCLEAR REACTOR REGULATION .
WASHINGTON, D.C. 20555 July 27,1992 NRC INFORMATION NOTICE 92-55: CURRENT FIRE ENDURANCE TEST RESULTS FOR THERM 0-LAG FIRE BARRIER MATERIAL Addressees All holders of operating licenses or construction permits for nuclear power reactors.
purDose The U.S. Nuclear Regulatory Comission (NRC) is issuing this information notice to inform addressees of the results of current Thermo-Lag 330 fire endurance tests conducted for the NRC at the National Institute of Standards and Technology (NIST). It is expected that recipients will review the information for applicability to their facilities and consider actions as appropriate. However, suggestions contained in this information notice are not NRC requirements; therefore, no specific action or written response is required. >
Discussion t
The NRC has been reviewing Thermo-Lag 330 fire barrier systems to determine their ability to adequately perform their 1-hour or 3-hour fire resistive functions. The NRC has issued three information notices and a bulletin on this subject:
- 1. Information Notice 91-47, " Failure of Thermo-Lag Fire Barrier Material to Pass Fire Endurance Test," August 6, 1991
- 2. Information Notice 91-79, " Deficiencies in the Procedures for Installing Thermo-Lag Fire Barrier Materials," December 6, 1991
- 3. Information Notice 92-46, "Thermo-Lag Fire Barrier Material Special Review Team Final Report Findings, Current Fire Endurance Tests, and Ampacity calculation Errors," June 23, 1992
- 4. Bulletin 92-01, " Failure of Thermo-Lag 330 Fire Barrier System to Maintain Cabling in Wide cable Trays and Small Conduits Free from Fire Damage," June 24, 1992 NIST conducted small scale 1-hour and 3-hour fire endurance tests to determine the fire resistive properties of Thermo-Lag pre-formed panels.
- On July 15, 1992, NIST conducted the 1-hour fire endurance test. The average thermocouple reading on the unexposed surface exceeded 162.7'C (325*F) (NRC ,
cold side temperature limit) in approximately 22 minutes and the unexposed '
surface of the material reached an average temperature of 652*C (1206*F) at
ret 3 m IIc s rc:s z e. . g nr IN 92-55 July 27, 1992 Page 2 of 3 45 minutes.
'4 35 minutes.The unexposed surface of the material exhibited visible browning During of reached a peak reading the935*C test, one thermocouple on the unexposed surface temperature of 923*C (1694*F), as(1716*F), exceeding the corresponding furnace baseline furnace temperature. the material burned and added heat to the The panels burned through at two locations in 46 minutes, as cold resulting air entered in a corresponding drop in surface thermocouple readings the furnace.
85 percent of the unexposed surface was blackened.At the end of 1-hour, approximatel The 3-hour test was conducted on July 17, 1992. The average thermocouple reading exceeded 162.7'C (325'F) in 2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br /> and 20 minutes, the average temperature at the reading was 222*C end of 3 hours3.472222e-5 days <br />8.333333e-4 hours <br />4.960317e-6 weeks <br />1.1415e-6 months <br /> was 206*C (403*F), and the peak thermocouple (432*F).
At the conclusion of the test, the material was '
soft and exhibited crumbled plastic deformation, and the fire-exposed stress skin upon contact.
side were surface. limited to off-gassing, slight browning, and crystallization a The furnace used to conduct these tests was a natural gas-fired small scale type 37 bywith 42.9internal inches).dimensions of 0.94 m by 0.94 m by 1.09 m (37 inches by The top of the furnace was equipped with a frame for by 31.75 inches). supporting horizontal test specimens of up to 0.81 m by 0.81 m (3 lower edge of the frame supported the sample.An 86.5-m (3.375-inch) wide ste This lip was insulated along its bottom, edge, and top with a nominal 13-m (0.5-inch) thick ceramic-fiber blanket.
Thus, the actual area of the sample exposed to the furnace was approximately 0.584 m by 0.584 m (23 inches by 23 inches).
To conduct this series of tests, nominally square samples were cut with dimensions After from being placed 0.794 in the m to sample horizontal 0.800 frame m (31.25 inches with and centered, to 31.5 ribs inches) on a facing upward (i.e., the ribbed face being the unexposed face), the gaps of ceramic-fiber blanket.between the edge of the sample and the frame were lo side-b The test configuration used bricks placed place.y-side along the perimeter of the sample to hold the test samples in The stress skin on the 3-hour material was thereby restrained in compression at the edges of the panel around the lip of the furnace and restricted from separating from the panel.
- The 1-hour fire endurance test was conducted on a Thermo-Lag 330 fi.e-barrier panel, ' nominal" thickness 13 m (0.5 inch).
sample ranged from 13.7 to 18.3 m (0.540 to 0.720 inches).The actual thicknes This material had stress skin on only the ribbed (unexposed) surface.
The three-hour fire endurance test was conducted on a Thermo-Lag 330 fire-barrier panel, " nominal
- thickness 25 m (1 inch).
27.7 to 39.6 m (1.09 to 1.56 inches).The actual thickness of the test sample ranged surfaces. This caterial had stress skin on both The ribbed surface was on the unexposed side during the test.
The furnace temperature was measured with three slow-response chrome 1 alume Materials (ASTM) Standard E-119.thermocouples, which met the requiremen followed the ASTM E-119 Standard time-temperature curve.The furnace to mantu i
~
. :5 zg, ;--
IN 92-55 July 27, 1992 l Page 3 of 3
~
The unexposed material surface temperatures were monitored at five points, placing one thermocouple approximately at the center of the specimen, and one at the approximate center of each of its quarter sections. The temperature acceptance criterion was that the temperature rise on the unexposed surface not exceed 138.8*C (250*F) above its initial temperature of 23.9'C (75'F) as specified in the National Fire Protection Association (NFPA) Standard 251.
The NRC will provide additional information on fire endurance testing as it becomes available.
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. 1 1
- [
harles E. Rossi, Director l
Division of Operational Events Assessment !
Office of Nuclear Reactor Regulation 2
l Technical contacts: Ralph Architzel, NRR
, (301) 504-2804 '
Patrick Madden, NRR '
(301) 504-2854
Attachment:
List of Recently Issued NRC Information Notices c
l i
n b
i
\ s 1
i
UNITED STATES MlM(AI [ .
NUCLEAR REGULATORY COMMISSION l OFFICE OF NUCLEAR REACTOR REGULATION WASHINGTON, D.C. 20555 December 15, 1992 "
'NkC INf0RMATION NOTICE 92-82: RESULTS OF THERMO-LAG 330-1 COMBUSTIBILITY '
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 (IN) to alert addressees to the results of Thermo-Lag 330-1 combustibility tests conducted for the NRC by the National Institute of Standards and Technology (NIST). It is expected that recipients will review the information for applicability to their facilities and consider actions as appropriate. However, suggestions contained in this information notice are -
not new NRC requirements; therefore, no specific action or written response is
. required.
Description of Circumstances As part of a small-scale testing program of Thermo-Lag 330-1 fire barrier material, NRC staff had NIST perform combustibility tests using the following standards: (1) American Society for Testing and Materials (ASTM) E-136,
" Standard Test Method For Behavior of Material in a Vertical Tube Furnace at 750 *C;" and (2) ASTM E-1354, " Standard Test Method for Heat and Visible Smoke Release Rates for Materials and Products using an Oxygen Consumption Calorimeter." NIST documented the results of these tests in Attachment 1, '
" Report on Test FR 3989, Analysis Of Barrier Material For Noncombustibility,"
of August 31, 1992.
Based on the ASTM E-136 testing standard, the NIST tests revealed that Thermo-Lag 330-1 fire barrier material is combustible. This testing standard '
prescribes the material as combustible if three out of four samples exceed any of the following criteria: (1) the recorded temperature of the specimen's
, surface and interior thermocouples rise 30 *C [54 *F] above the initial furnace temperature; (2) there is flaming from the specimen after the first 30 seconds of irradiance; or (3) the weight loss of the specimen during .
testing exceeds 50 percent and either (a) the recorded temperature of the '
surface and interior thermocouples at any time during the test rise above the furnace air temperature at the beginning of the test or (b) there is flaming of the specimen. Each of the four Thermo-Lag specimens tested exhibited a weight loss of greater than 50 percent and exhibited flaming beyond 30 seconds.
I
IN 92-82 December 15, 1992 :
Page 2 of 3 !
NISTperformedtheASTME-1354 calorimeter}estonfourThermo-Lagspecimens, subjecting them to an irradiance of 75 kW/m [6.6 BTU /ft2 -s]. The total !
amount of heat released from the Thermo-Lag exceeded that released from Gypsum ;
board zand was about equal to the heat released from fire-retardant plywood.
Discussion l Section 50.48(a) of Title 10 of the Code of Federal Reculations requires that each operating nuclear power plant have a fire protection plan that satisfies i General Design Criterion (GDC) 3, " Fire protection," in Appendix A to 10 CFR '
Part 50. GDC 3 requires that " structures, systems, and components important ;
to safety shall be designed and located to minimize, in a manner consistent '
with other safety requirements, the probability and effects of fires and explosions. Noncombustible and heat resistant materials shall be used wherever practical throughout the unit, particularly in locations such as the .
containment and control room."
NRC-approved plant fire protection programs referenced by the plant operating license and Section III.G, " Fire protection of safe shutdown capability," of Appendix R to 10 CFR Part 50, require one train of systems necessary to -
}
achieve and maintain hot shutdown conditions from either the control room or i emergency control stations to be free from fire damage.
Section III.G.2 of Appendix R permits separation by a horizontal distance of more than 6.1 meters [20 feet] with no intervening combustibles or fire j hazards as one of several methods to ensure that cables and equipment and ,
associated circuits of redundant safe shutdown trains located outside !
containment are maintained free from fire damage. If this method is used, fire detection and automatic fire suppression must be installed in the area of ,
the redundant safe shutdown trains. Some licensees use Thermo-Lag fire barrier material to enclose intervening combustibles to achieve 6.1 meters -
3 [20 feet] of separation free of intervening combustibles between the redundant safe shutdown trains. i Section III.G.2.f of Appendix R allows licensees to separate cables and ;
equipment and associated circuits of redundant trains inside noninerted '
containments by installing a noncombustible radiant energy shield as one of ;
several methods to achieve required fire protection for these circuits. Some -
licensees use Thermo-Lag to construct radiant energy heat shields inside containment.
Related Generic Comunications Attachment 2 is a list or recently issued generic communications concerning Thermo-tag 330-1 fire barrier systems. .
- l i
I I
r
- I
~
IN 92-82 December 15, 1992 Page 3 of 3 This information notice requires no specific action or written response. If i 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.
~ A
~
Brian K. Grimes, Director Division of Operating Reactor Support Office of Nuclear Reactor Regulation Technical contacts: Ralph Architzel, NRR ,
(301) 504-2804 -
Patrick Madden, NRR (301) 504-2854 ,
Attachments: -
- 1. Report of Test FR 3989, Analysis Of -
Barrier Material For Noncombustibility
- 2. List of Recently Issued Generic Communications Concerning Thermo-Lag 330-1 Fire Barrier Systems ;
- 3. List of Recently Issued NRC Information Notices i
f t
f f
a T
I i
I
Att:chnent 1 IN 92-E2 Decerber 15, 1992 Page 1 of 6 U.S. DEPARTMENT OF COMMERCE NATIONAL INSTITLITE OF STANDARDS AND TECHNOLOGY Gaithersburg, MD 20899 REPORT OF TEST FR 3989 on i
ANALYSIS OF BARRIER MATERIAL FOR NONCOMBUST1BILTTY !
by Vytenis Babrauskas 5
August 31,1992 ,
t e
Submined to:
C5ce of Nuclear Reactor Regulation ,
United States Nuclear Regulatory Commission Washington, DC 20555 a
l J
Attacinent 1 IN 92-82 l Dece:aber 15, 1992 Purpose of test Page 2 of 6 1 1
Tests were conducted to determine if material submitted qualifies as noncombustible.
I Material tested l l
ne test material comprised board stock supplied by the Nuclear Regulatory Commiuion (NRC) to .
the National Institute of Standards and Technology (NIST). Two variations of board stock were {
supplied: nominal 1/2* and nominal 1* thick boards. De thickness of the boards is intended by the !
i manufacturer to have a certain 'plus' tolerance, but a zero 'minus' tolerance. Thus, the boards were generally thicker than the nominal 12.5 and 25 mm values, but with substantial variation from point i to point on the board. De nominal 1* stock contained a stainless steel wire mesh on both sides of !
the board, while the nominal 1/2" stock contained only wire mesh on one side of the board. !
,l With the exception of removing the wire mesh (in those instances where stated, below) and cutting l specimens to size, NIST did not alter or modify the specimens. j Definition of noncombustibility
, ne Uniform Building Code (International Conference of Building Omcials, Whittier, CA,1991) l deGnes concombustible as follows (pp. 29 30): ,
t ,
i "Sec. 415. NONCOMBUSTIBLE as applied to building construction materials means a l l
material which,in the form in which it is used, is either one of the following: !
(a) Material of which no part will ignite and burn when subjected to fire. Any material j conforming to U.B.C. Standard No. 4-1 shall be considered noncombustible within the !
] meaning of this section. ;
i j (b) Material having a structural base of noncombustible material as defined in Item !
l (a) above, with a surfacing material not over 1/8 inch thick which has a Dame-spread rating l J of 50 or less." l 4
.I We note that U.B.C. Standard No. 4-1 is functionally identical to ASBi E 136 [1]. De other two - -
l U.S. model building codes, BOCA National and Southern Standard, use essentially identical l j definitions for noncombustibility.
! To explain in more detail, U.S. practice provides for two different ways by which a product can be i i qual fied as concombustible. Part (b) was originally developed to make certain that conventional ,
DTum wallboard would be allowed as noncombustible. To qualify under part (b), two tests must bc :
]
. run: an ASTM E 136 test on the substrate material and the ASTM E 84 Steiner Tunnel on the l complete product, including its thin surface layer. Part (b) cannot be successfully applied to a product j l
a unless the material comprising all of its thickness, save the topmost 1/8*, can pass the ASBf E 136 i test, while the top layer cannot. In the present case, the steel mesh layer is accepted to be t l noncombustible. Rus, for the test specimens submitted, part (b) is not applicabk. Only testing of the :
bulk board stock material needs to be done, and this must be done using the ASTM E 136 test.
For reference, a specimen is recorded as passing the ASTM E 136 test if:
j " at least three of the four specimens meet the following conditions:
1 1 2 T i
j '
i s
$ l f
Attachment 1-IN 92-82 '
Dacember 15, 1992 ;
Page 3 Of 6 .
He recorded temperature at the surface and interior thermocouples do not at any time i during the test rise more than 54*F (30*C) above the furnace temperature at the beginmng [
of the test, 4
There is no Daming from the specimen after the first 30 s, and 4-
~ When the weight loss of the specimen during the testing ~M SWe, the recorded !
' temperature of the surface and interior thermocouples do not at any time during the test rise i
above the furnace air temperature at the bernning of the test, and there is no flaming of the '
specimen."
i To examine the submitted specimens for noncombustibility, the standard procedure as described above was conducted. To obtain additional information, some supplementary tests were also i conducted to better assay the behavior of the material. The standard procedure was conducted for NIST under contract by United States Testing Company, Inc. ncir results are described below, while t their report submitted to NIST is appended to this Report. The supplementary tests were conducted in our own laboratories.
Standard tests Specimens of 6e nominal l' board stock were submitted to the United States Testing Company, Inc.
for testing according to ASTM E 136 procedures. -
- i
' he material FAILED the test, since, for all 4 specimens tested, the percent weight loss was greater than SWc and flaming continued in excess of 30 s.
j ;
he report received from the test:ng laboratory is attached as the Appendix to this report.
Supplementary tests he report from an ASTM E 136 test is recorded simply in pass / fail terms. To derive additional insight into the behavior of the material, it is possible to conduct tests which provide a quantitative .
3 measurement scale. Such a test is available in the form of the ASTM E 1354 test [2]. His release rate (HRR) test and, as such, it provides time-rc. solved information on the combustion of a specimen. In engineering terms, specimens which are ' noncombustible' are those which show ic u heat '
release. This concept has not yet been introduced into the U.S. building codes, but it is in the proce of being approved for use in Canada [3] and U.S. usage is expected to come shortly thereafter. He 4
tentative decision in Canada is to use a test irradiance of 50 kWm . Work has also been NIST on the development of this concept, with most of the studies being conducted at a 75 4 kWm '
t irradiance [4). His heat-release-based approach conceptually simpli5es the treatment of combustibility, since a special two-part testing approach, necessary for the conventional method to properly characterize Dpsum wallboard, is no longer necessary. With this heat-release based approach, a single method is adequate to characterize all materials, including gypsum boards. ,
For the supplementary examination of the subject material, tests were conducted at both 50 and 75 kW m4 irradiance levels. The specimens were cut from the nominal 1/2" board stock to a size of 103 mm by 100 mm. The actual thickness of the board stock was found to be approximately 18 mm. In all cases, the prescribed edge frame was used, since the material substantially intumesces.
3 1
l
Ittachment 1 ,
IN 92-82 ;
December 15, 1992 he test conditions and the test results are given in Table 1. Page 4 of 6 s
i l Table 1. Results obtained in supplementary tests l
4 I
I Test Test Irradiance . Peak Total beat in Total heat m - ;
no. -' ' condition . kWm4 HRR - '600 s: 900 s . # '
kWm4~ MJm4 MJm' 4 l
J t 5489 grid used; 50 74 -
34.1 .
whe mesh up 5490 grid used; 50 83 -
44.1 ;
wie mesh up 5491 grid used; 50 74 -
28.2 whe mesh down f
! 5492 grid used: 50 76 -
25.9 !
I wie mesh down
, 5466 grid not used: 75 120 28.5 -
wire mesh up ^
i 5486 grid used; 75 107 46.9 - ;
wire mesh up
- 5487 grid used; 75 110 38.9 -
l wire mesh down ,
5467 grid used; 75 100 35.3 -
wire mesh down i r
i It is intended in the ASTM E 1354 test that the specimen surface condition be approximately planar. f Hus, with intumescing or deforming spectmens, the testing laboratory needs to use a restraint grid l
" described in the standard or,if necessary, arfatternate restraint technique to ascertain that specimen !
deformations are not excessive, ne notation " grid used/not used" refers to the restraint grid specified I in ASTM E 1354. In the present case, the test board stock had a wire mesh which can be considered ;
as also serving some restraint function. Thus!various combinations of standard grid and wire mesh !
topside /bottomside to ation were tried to determine if there was a systematic effect of the restraint condition used. He notation " wire mesh" refers to the wire mesh which comes as part of the subject test specimen. !
I interpreta: ion ofresuhs !
' The above findings must be mterpreted in the light of the performance of materials accepted for use i as noncombustible. By examining the published data it can be seen that the following performance ;
d for gypsum wallboard (unpainted, both regular and Type.X grades) is obtained. 1 t
a
}
i 1 ,
4 ;
e i
t
.= - - .. . .
Attachment 1 i p
IN 92-82 :
December 15, 1992 !
Table 2. Published results on opsum waUboard Page 5 Of 6 ;
I i r l
Irradiance . Ref. Peak HRR Total beat in ' Total beat in -
kWm4 kWu-2 600 s : 900 s l
. . hU m4 <
hum 4 .
50 [3] 70 - 103 -
4 75 [4] 97 4.2 - '
- l ""
j The results of Table 1 can now be compared to those of Table 2 to determine more quantitatively
' the differences between the tested product and opsum wallboard. There are two bases of comparison: the value of the peak HRR and the total beat released in the control period (600 s when I 4
testing at 75 kW m ; 900 s when testing at 50 kWei ). The value of the peak HRR for the tested I board product is seen to be roughly similar to Dpsum wallboard. The value of the total beat releaW ,
bowever, is more than 8 times higher. This quantitative measure is consistent with the qualitative ;
observation in the E 136 test that " flaming continued for the remainder of the test."
i !
1 As a result of the Canadian study, Richardson and Brooks [3] made a proposed recommendation that i
l
' combustibility of materials can be grouped into categories, based on performance in the ASTM E - ;
1354 test. In the highest category (#1), limits of 10 kWm 4 on peak HRR and 5 hU-m 4 on the total i
- heat released would be set. This is intended to correspond to specimens which pass solely by means
{
of the ASTM E 136 test. In the next category (#2), would be Dpsum wallboard and other products i
- which are currently qualified on the basis of the ASTM E 136 test for the core material, plus the !
ASTM E 84 test on the finished product. For category #2, the limits are taken at 100 kW m4 peak i
.l HRR and 25 AUm4 total beat released. For the present test specimens, the total beat relewd j exceeds 25 hum 4 in all cases; thus, by the proposed Canadian criteria, the test material would not qualify to be rated in a class comparable to apsum wallboard. '
4 n
i References !
i
)
3
[1] Standard Test Method For Behavior of Materials in a Vertical Tube Furnace at 750 *C ,
(E 136). American Society for Testing and Materials, Philadelphia. '
{ [2] Standard Test Method for Heat and Visible Smoke Release Rates for Materials and ;
) Products using an Oxygen Consumption Calorimeter (E 1354). American Society for !
Testing and Materials, Philadelphia.
i 1
[3] Richardson, LR., and Brooks, M.E., Combustibility of Building Materials, Fire and ;
Materials. 15, 131-136 (1991).
[4] Babrauskas, V., North American Experiences in the Use of Cone Calorimeter Data for Classification of Products, pp. 89103 in Proc. of the Intl. EUREFIC Seminar 1991, .
- interscience Communications Ltd, London (1991).
l l
s 5
e
-r- ,
m- - m
- ,%. -e., , y
Attach ent 1 1 Di 92-D2 December 15, 1992 Page 6 of 6 w,
Prepared try: l APPROVED ;
1 ?
lC ! :
o I 4
m c/ !
Vytenis Babrauskas Andrew J. Fowell Chief, Fire Science and Engineering Dhsion ,
I i
O I
e I
e s
E
?
b I
i d i 4
6 s
I e
i Attachment 2 IN 92-B2 December 15, 1992 '
Page 1 of I y
List of Recent1v Issued Generic Communications Concernino Thermo-taa 330-1 Fire Barrier Systems 1
- 1. Information Notice 91-47, " Failure of Thermo-Lag Fire Barrier Material to Pass Fire Endurance Test," August 6, 1991 i' 2.~
Information Notice 91-79, " Deficiencies in the Procedures for Installing Thermo-Lag Fire Barrier Materials," December 6, 1991
- 3. Information Notice 92-46, "Thermo-Lag Fire Barrier Material Special Review Team Final Report Findings, Current Fire Endurance Tests, and Ampacity Calculation Errors," June 23, 1992 i
- 4. Bulletin 92-01, " Failure of Thermo-Lag 330 Fire Barrier System to i
' Maintain Cabling in Wide Cable Trays and Small Conduits Free from Fire Damage," June 24, 1992 i
- 5. Information Notice 92-55, " Current Fire Endurance Test Results for ~
i Thermo-Lag Fire Barrier Material," July 27, 1992 '
- 6. Bulletin 92-01, Supplement 1, " Failure of Thermo-Lag 330 Fire Barrier
! System to Perform Its Specified Fire Endurance function," !
August 2B, 1992 l i
a A
a i
\
i i
Attachment 3 IN 92-82 ,
December 15, 1992 ;
Page 1 of I ;
LIST OF RECENTLY ISSUED i NRC INFORMATION NOTICES !
Information Date of l
. Notice No. Subject Issuance Issued to l I
- 92-81 Potential Deficiency 12/11/92 All holders of OLs or cps I of Electrical Cables 'for nuclear power reactors. l with Bonded Hypalon !
Jackets 92-80 Results of Thermo-Lag 12/07/92 All holders of OLs or cps 330-1 Combustibility for nuclear power reactors.
! Testing 92-79 Non-Power Reactor 12/01/92 All holders of Ols or cps Emergency Event Response for test and research l
- i reactors.
92-78 Piston to Cylinder 11/30/92 All holders of OLs or cps .
Liner Tin Smearing on for nuclear power reactors. !
Cooper-Bessemer KSV i j Diesel Engines l 92-77 Questionable Selection 11/17/92 All holders of OLs or cps I and Review to Deter- for nuclear power reactors. !
mine Suitability of !
-; Electropneumatic Relays l l for Certain Applications !
- 92-76 Issuance of Supple- 11/13/92 All holders of OLs or cps [
~
ment I to NUREG-1358, for nuclear power reactors, j
" Lessons Learned from -
the Special Inspection :
Program for Emergency t i Operating Procedures !
(Conducted October 1988 - I
- September 1991)"
92-75 Unplanned Intakes of 11/12/92 All holders of OLs or cps Airborne Radioactive for nuclear power reactors.
Material by Individuals
. at Nuclear Power Plants '
92-74 Power Oscillations at 11/10/92 All holders of OLs or cps
[
Washington Nuclear for nuclear power reactors. ;
l Power Unit 2 i
r OL - Operating License [
~
CP - Construction Permit -i
=
i [
< t i
b
0(,AhElik b .
I i
STATEMENT -
SUBMITTED BY !
!k THE OFFICE OF THE INSPECTOR GENERAL !
THE UNITED STATES NUCLEAR REGULATORY COMMISSION :
t TO THE SUBCOMMITTEE ON OVERSIGHT AND INVESTIGATIONS f COMMITTEE ON ENERGY AND COMMERCE i UNITED STATES HOUSE OF REPRESENTATIVES L
2 t CONCERNING '
THE ADEQUACY OF NRC STAFF'S ACCEPTANCE AND REVIEW OF THERMO-LAG FIRE BARRIER MATERIAL
- i
- - DAVID C. WILLIAMS s
INSPECTOR GENERAL i 4 I t
h i
h I
i
i
?
INTRODUCTION '
r MR. CHAIRMAN AND MEMBERS OF THE SUBCOMMITTEE, I AM PLEASED FOR THIS OPPORTUNITY TO DISCUSS OUR AUGUST 12, 1992, OFFICE OF THE INSPECTOR
+
l:
GENERAL (OIG) INSPECTION REPORT, ENTITLED, " ADEQUACY OF NRC STAFF'S ACCEPTANCE AND REVIEW OF THERMO-LAG 330-1 FIRE BARRIER MATERIAL."
IN OUR REPORT, WE CONCLUDED THAT THE NRC STAFF DID NOT CONDUCT AN :
l ADEQUATE REVIEW OF FIRE ENDURANCE AND AMPACITY DERATING .f l i INFORMATION CONCERNING THE ABILITY OF THE FIRE BARRIER MATERIAL. i THERMO-LAG, TO PERFORM ITS REQUIRED FIRE PROTECTION FUNCTION. WE-l ALSO CONCLUDED THAT ALTHOUGH THE NRC STAFF RECEIVED REPORTS OF }
PROBLEMS WITH THERMO-LAG OVER NEARLY A 10-YEAR PERIOD, THEY FAILED !
TO TAKE ANY SIGNIFICANT ACTIONS TO ADDRESS THESE PROBLEMS. THERMO-l l LAG WAS ACCEPTED BY THE NRC FOR INSTALLATION IN NUCLEAR POWER i
PLANTS TO PROTECT ELECTRICAL CABLES REQUIRED TO SHUT DOWN THE i REACTOR DURING AN EMERGENCY. l l
OUR INSPECTION OF THE NRC STAFF'S PERFORMANCE WAS IN1TIATED BASED ON AN ALLEGATION THAT QUESTIONED THE ADEQUACY OF THERMO-LAG. IN j ADDITION TO THE INSPECTION, AN INVESTIGATION OF THE ~THERMC-LAG MANUFACTURER WAS ALSO INITIATED. THE INVESTIGATION IS ONGOING AND !
IS BEING CONDUCTED JOINTLY WITH THE NRC OFFICE 0F INVESTIGATIONS UNDER THE DIRECTION OF THE U.S.- ATTORNEY'S OFFICE FOR THE DISTRICT OF MARYLAND.
2 .
l ;
L
BACKGROUND .t 1
i BY WAY OF BACKGROUND. THE CRITICAL NEED FOR CABLE FIRE PROTECTION ,
WAS RECOGNIZED FOLLOWING A FIRE ATTHE BROWNS FERRY NUCLEAR POWER PLANT IN 1975. THE FIRE BURNED FOR NEARLY 7 HOURS AND DAMAGED BOTH 4 !
THE PRIMARY AND EMERGENCY BACKUP CABLES. THE DAMAGE _ TO THE l
' CABLES JEOPARDIZED THE ABILITY OF THE CONTROL ROOM TO MONITOR THE -!
STATUS OF THE REACTORS. ALTHOUGH MUCH OF THE EQUIPMENT WAS i
RENDERED INOPERABLE, ADEQUATE EQUIPMENT REMAINED OPERATIONAL TO I SHUT DOWN THE TWO REACTORS. FOLLOWING THE FIRE, A SPECIAL REVIEW t i
GROUP IDENTIFIED PROBLEMS AND RECOMMENDED IMPROVEMENTS IN THE l FIRE PROTECTION AREA. IN" 1981, THE COMMISSION ISSUED A NEW FIRE PROTECTION RULE. 10 CFR PART 50, APPENDIX R, WHICH INCLUDED !
REQUIREMENTS FOR THE PROTECTION OF EMERGENCY BACKUP CABLING l
KNOWN AS REDUNDANT SAFE SHUTDOWN CABLES.
l
?
I ONE OF THE APPENDIX R OPTIONS FOR PROTECTING THE REDUNDANT CABLES WAS ENCLOSING OR WRAPPING THE CABLE TRAYS AND CONDUITS WITH A FIRE {
RETARDANT MATERIAL. THE MATERIAL WAS REQUIRED TO HAVE A FIRE !
RESISTANCE RATING OF 1 HOUR IF A SPRINKLER SYSTEM WAS INSTALLED, OR !
3 HOURS WITH NO SPRINKLER SYSTEM. INSTALLATION OF THE FIRE RETARDANT l MATERIAL WAS THE OPTION THAT MOST POWER PLANTS SELECTED.
EVENTU ALLY. S4 PLA NTS INSTALLED THERM O-LAG FIRE RETARDANT M ATERI AL. i l
3 c
l J
STAFF REVIEW AND ACCEPTANCE OF TilERMO-LAG i
i NUCLEAR POWER PLANTS WERE REQUIRED TO SUBMIT PLANS TO EXPLAIN HOW ,
THEY WOULD COMPLY WITH THE NEW REQUIREMENTS, AND THE NRC WAS l RESPONSIBLE FOR REVIEWING THE PLANS. THE NRC MANAGERS OF THE FIRE f t
PROTECTION STAFF ADVISED OIG THAT THE NRC REVIEWS WERE LIMITED TO
, AUDITING WHAT WAS SUBMITTED. THIS APPROACH WAS BASED ON THE f i
ASSUMPTION THAT THE UTILITIES WOULD PROVIDE ACCURATE AND TRUTHFUL <
1
] REPRESENTATIONS REGARDING FIRE RETARDANT PERFORMANCE TESTS.
j VERIFICATION OF THIS INFORMATION WAS NOT . BELIEVED NECESSARY SINCE \
THE UTILITY PROPOSALS WERE SUBMITTED UNDER OATH. OIG FOUND THAT f BETWEEN 1981 AND 1991, THE NRC STAFF DID NOT OBSERVE ANY TESTS OF.
1
{
THERMO-LAG. FURTHER, - THE NRC STAFF DID NOT INVESTIGATE THE QUALIFICATIONS OF OR VISIT THE LABORATORY WHICH PURPORTEDLY :
t SUPERVISED MOST OF THE THERMO-LAG TESTS. ADDITIONALLY, THE NRC DID i
, 1
- NOT CONDUCT AN INSPECTION OF THERMAL SCIENCE INC.(TSI), THE MANUFACTURER OF THERMO-LAG, LOCATED IN ST. LOUIS, MISSOURI. !
a i
NRC REQUIREMENTS STATE THAT FIRE RETARDANT PERFORMANCE TESTS BE CONDUCTED BY A NATIONALLY RECOGNIZED FIRE TESTING LABORATORY.
BEGINNING IN 1981, THE NRC BEGAN RECEIVING UTILITY FIRE PROTECTION i
PROPOSALS THAT INCLUDED FIRE TESTS OF THERMO-LAG THAT WERE !
CONDUCTED BY TSI AND WITNESSED BY INDUSTRIAL TESTING LABORATORIES, !
a INC. (ITL), ALSO OF ST. LOUIS. NEITHER TSI NOR ITL IS A NATIONALLY i
f
'e k
i
RECOGNIZED FIRETESTING LABORATORY. ALTHOUGH THETEST RESULTS WERE l PUBLISHED AS ITL REPORTS, OIG LEARNED THAT ITL HAD NO FIRE TESTING !
i EXPERTISE. ;
i i l
i NEVERTHELESS, AN OIG REVIEW OF A NUMBER OF ITL REPORTS OF FIRE TESTS f CONDUCTED BY TSI AND WITNESSED BY ITL DISCLOSED THAT THE TSI TESTS l HAD NOT BEEN PERFORMED IN ACCORDANCE WITH THE REQUIRED FIRE l TESTING STANDARDS. FOR EXAMPLE, THE TEST FURNACE WAS NOT LARGE ENOUGH TO CONDUCT THE REQUIRED FULL SCALE TESTS. REPORTS i
! L DOCUMENTING SMALL SCALE TESTS WERE SUBMITTED. DESPITE THESE AND [
OTHER DEFICIENCIES, THE NRC STAFF ACCEPTED THE TSI TESTS AND ITL i i
- REPORTS. i l'
4 THE TSI FIRE ENDURANCE TESTS WERE REPORTEDLY VALIDATED BY THE PRESENCE OF ITL REPRESENTATIVES, UTILITY OFFILIALS, AND INSPECTORS !
1 l
FROM THE AMERICAN NUCLEAR INSURERS (ANI). OIG FOUND THAT ITL f
REPRESENTATIVES, UTILITY OFFICIALS, AND ANIINSPECTORS ONLY WITNESSED' THE ACTUAL CONDUCT OF THE FIRE TESTS. THEY DID NOT INSPECT THE TEST i
- ARTICLES BEING ASSEMBLED TO ENSURE THAT THEY WERE CONSTRUCTED AS. !
j i DESCRIBED IN THE TEST REPORTS. FURTHERMORE, THE ANI AND UTILITY !
WITNESSES WERE OFTEN ABSENT DURING SIGNIFICANT PORTIONS OF THE FIRE !
TESTS.
l; 1
ALTHOUGH THE ITL TEST REPORTS STATE THAT THE FIRE TESTS WERE C n
f 5
c
. I SUPERVISED AND. CONTROLLED ENTIRELY BY ITL, THE ITL REPRESENTATIVE
-l WAS PRESENT ONLY AS A WITNESS TO VERIFY THAT A TEST WAS CONDUCTED.
ACCORDING TO ITL, THE TESTS WERE CONDUCTED BY' TSI, AND THE TEST t i
REPORTS WERE WRITTEN BY TSI AND THEN SIGNED BY THE PRESIDENT OF ITL l i
_ WITH NO SUBSTANTIVE VERIFICATION THAT THE DATA IN 'THE REPORTS !
i REFLECTED THE DATA GATHERED DURING THE ACTUAL TESTS. IN SOME !
INSTANCES, THE ITL PRESIDENT SIMPLY SIGNED TEST REPORT COVER SHEETS FOR TSI WITHOUT ACTUALLY SEEING THE REPORTS.
i l
l ;
DURING THE SUMMER OF 1982, THE NRC STAFF RECEIVED TWO CONFLICTING FIRE ENDURANCE TEST REPORTS ON THERMO-LAG. THE FIRST TEST WAS -l PERFORMED AT A NATIONALLY RECOGNIZED LABORATORY AT THE REQUEST OF SUSQUEHANNA NUCLEAR POWER PLANT. THIS I-HOUR TEST OF THERMO-LAG FAILED TO PASS THE ACCEPTANCE CRITERIA. THE SECOND TEST WAS !
CONDUCTED BY TSI, WITH ITL- AS A WITNESS, AT THE REQUEST OF THE WASHINGTON NUCLEAR POWER PROJECT. THIS TEST PASSED THE ACCEPTANCE CRITERIA. THE NRC STAFF ACCEPTED THE SECOND TEST AND REIECTED THE FIRST. HOWEVER, THE STAFF DID NOT INVESTIGATE WHY ONE TEST PASSED I
- AND THE OTHER FAILED OR WHATTHE FAILUREIMPLIED ABOUTTHERMO-LAG'S 5 ABILITY TO PERFORM.
ANOTHER FACTOR THE UTILITIES HAD TO CONSIDER WHEN DECIDING TO i INSTALL A PARTICULAR FIRE BARRIER MATERIAL WAS ITS AMPACITY !
I DERATING. WHEN ELECTRICAL CURRENT PASSES THROUGH A CABLE. HEAT IS - !
6 I
__ __ i
~
GENERATED. NORMALLY, THIS HEAT IS ABLE TO RADIATE AWAY FROM THE CABLE, AND THE CABLE TEMPERATURE REMAINS SUFFICIENTLY LOW.
1 HOWEVER, IF THE HEAT BECOMES TRAPPED, THE CABLE OPERATES AT TOO l
1 HIGH OF A TEMPERATURE AND MAY DETERIORATE PREMATURELY. IF CABLES ARE ENCLOSED IN A FIRE BARRIER MATERIAL. THE HEAT CANNOT EASILY i a f
- ESCAPE. THE NEGATIVE EFFECT OF THE HIGHER CABLE OPERATING 4
TEMPERATURES CAN BE COMPENSATED FOR BY DERATING (LOWERING) THE i j
AMOUNT OF ELECTRICAL CURRENT PASSING THROUGH THE CABLES, THEREBY l LOWERING THE AMOUNT OF HEAT GENERATED. FIRE BARRIER _ MATERIALS :
4 REQUIRING THE LEAST AMPACITY DERATING WOULD BE MOST ATTRACTIVE TO !
. A UTILITY. THE NRC REQUIRES THAT AMPACITY DERATING BE CONSIDERED WHEN THE UTILITIES DESIGN AND MODIFY ELECTRICAL SYSTEMS. !
t AS WITH THE FIRE ENDURANCE TESTS OF THERMO-LAG, TSI ALSO CONDUCTED !
ITS OWN AMPACITY DERATING TESTS WITH ITL AS A WITNESS. THE TSI REPORTED AMPACITY DERATING DATA WERE BY FAR THE LOWEST AMONG THE MANUFACTURERS. IN FACT, IT WAS LOWER BY A FACTOR OF FOUR WHEN '
COMPARED WITH THE FIRE BARRIER MATERIAL MANUFACTURED BY !
l t
MINNESOTA MINING AND MANUFACTURING. THE THERMO-LAG AMPACITY !
i I DERATING TEST REPORTS WERE USED BY THE UTILITIES IN SUPPORT OF THEIR PROPOSALS TO USE THERMO-LAG. OIG FOUND NO EVIDENCE TO INDICATE THAT -
i l
THE NRC STAFF HAD REVIEWED THE AMPACITY DERATING TESTS ON THERMO- I LAG. THE STAFF EXPLAINED THAT IT WAS THE RESPONSIBILITY OF THE UTILITIES TO CONSIDER AMPACITY DERATING REQUIREMENTS, AND IT WAS f 1
7 i t
4 i
ALSO THE RESPONSIBILITY OF THE UTILITIES TO ENSURE THE VALIDITY OF ANY AMPACITY DERATING TESTS THEY MIGHT USE. !
DURING THE OIG INSPECTION, WE LEARNED OF SEVERAL INSTANCES OVER THE !
i PAST 10 YEARS THAT WERE REPORTED TO THE NRC AND WHICH QUESTIONED ;
THE ABILITY OF THERMO-LAG TO PERFORM .AS CLAIMED. OUR REVIEW OF i THESE INSTANCES DISCLOSED, HOWEVER, THAT THE NRC DID NOT TAKE ANY [
SIGNIFICANT ACTION TO ADDRESS THESE INDICATIONS OF THERMO-LAG !
PROBLEMS. FOUR OF THESE INSTANCES ARE SUMMARIZED BELOW:
l j IN 1986 AND 1987, THE NRC RECEIVED INFORMATION THROUGH OFFICIAL i
.h REPORTING CHANNELS. WHICH INDICATED THAT THE AMPACITY DERATING !
FIGURES FOR THERMO-LAG WERE MUCH HIGHER THAN INITIALLY REPORTED.
ONE SOURCE OF THIS INFORMATION WAS TSI AND THE OTHER SOURCE WAS THC COMANCHE PEAK NUCLEAR POWER PLANT. THE COMANCHE PEAK REPORT EXPLAINED THAT FAILURE TO CONSIDER THE HIGHER FIGURES COULD CAUSE j POWER CABLES TO EXCEED THEIR DESIGN CRITERIA AND, IF LEFT UNCORRECTED, COULD ADVERSELY AFFECT THE SAFETY OF PLANT l OPERATIONS. THE NRC DID NOT FOLLOW UP WITH TSI TO GET AN t
EXPLANATION OF THE HIGHER FIGURES, NOR DID THE NRC TAKE STEPS TO ENSURE THAT OTHER UTILITIES WERE NOTIFIED OF THE HIGHER FIGURES.
THERE IS NO INDICATION TH AT THESE REPORTS WERE REVIEWED WITH GENERIC IMPLICATIONS IN MIND. IN FACT. IT APPEARS THAT THE ONLY ACTION TAKEN BY THE NRC WAS TO RECORD THE RECEIPT OF THE REPORTS.
S
~ -, ,
i t
IN 1989. COMANCHE PEAK REPORTED TO THE NRC THAT THERMO-LAG PANELS MEASURING LESS THAN THE REQUIRED THICKNESS OF I/2 INCH WERE BEING RECEIVED AT THE POWER PLANT. COMANCHE PEAK OBTAINED FROM TSI AN EXPLANATION FOR THE VARIATION IN THICKNESS. COMANCHE PEAK ACCEPTED I
\
l- THE TSI EXPLANATION AND PROVIDED IT TO THE NRC. THE NRC THEN fe ACCErito COMANCHE PEAK'S EXPLANATION WITHOUT FURTHER QUESTION OR REVIEW. HOWEVER, DURING THE OIG INSPECTION, WE LEARNED THAT THE COMANCHE PEAK EXPLANATION PROVIDED TO THE NRC WAS TECHNICALLY '
I INADEQUATE AND WOULD NOT PROVIDE AN EFFECTIVE INSPECTION METHOD FOR THERMO-LAG PANELS.
THE RIVER BEND NUCLEAR POWER PLANT SUBMITTED A NOTICE TO THE NRC l l
SPECIFYING SEVERAL GENERIC ISSUES INVOLVING - THERMO-LAG IN .1989. !
ALTHOUGH RIVER BEND CONSIDERED THESE CONCERNS TO BE OF A GENERIC !
l NATURE, THE NRC VIEWED THEM AS APPLICABLE TO ONLY RIVER BEND. THE !
OIG INSPECTION COULD IDENTIFY NO ACTION TAKEN BY THE NRC TO ADDRESS l THESE MATTERS FOR OVER A YEAR AND A HALF. Al-1ER THE OIG INITIATED ITS INSPECTION INVOLVING THERMO-LAG ALLEGATIONS, THE NRC STAFF FINALLY ,
i CONDUCTED AN INSPECTION AND IMMEDIATELY FOUND PROBLEMS WITH 1
THERMO-LAG WHEN USED TO PROTECT LARGE CABLE TRAYS. THE INSPECTION l
TEAM ALSO DETERMINED THAT THE RIVER BEND PROBLEMS WITH THERMO-LAG l WERE APPLICABLE THROUGHOUT THE INDUSTRY.
4 IN CONCLUSION. WE BELIEVE THAT THE NRC DID NOT CONDUCT AN ADEQUATE 9
- _ . . . ~ . ... . - . - .- ... .
. *I
.l l
REVIEW INVOLVING THE APPROPRIATE LEVEL OF VERIFICATION AND INSPECTION OF UTILITY SUBMISSIONS INCLUDING THE THERMO-LAG TEST l 1
REPORTS. FURTHERMORE, THE SITUATION THAT THE NRC AND INDUSTRY ARE CURRENTLY STRUGGLING WITH RELATED TO THERMO-LAG IS A DIRECT RESULT' l OF THIS INADEQUATE REVIEW. THIS FAILURE WAS '
EVEN- LESS UNDERSTANDABLE GIVEN THAT THIS REVIEW WAS- CONDUCTED WHEN ,
APPENDIX R WAS A NEW REQUIREMENT AND THE INDUSTRY WAS LIMITED'IN- 5 ITS KNOWLEDGE OF HOW TO COMPLY WITH THESE STANDARDS. IN A ' j SITUATION SUCH AS THIS,IT WOULD SEEM APPROPRIATE FOR THE NRC TO HAVE BEEN MORE ACTIVELY INVOLVED IN THE INITIAL ACCEPTANCE PROCESS. ;
t MR. CHAIRMAN, THIS CONCLUDES MY STATEMENT. :
l i
f.
I i
i
?
l
?
.f 10 ;
t
-i
. _ , , _ . - - ~ _ . . . . ,
October 21, 1952 M
MEMORANDUM FOR: File FROM: Frank J. Miraglia, Deputy Director Office of Nuclear Reactor Regulation
SUBJECT:
TELECON WITH RUBIN FELDMAN CN OCTOBER 20, 1992 At 3:55 p.m., October 20, 1992, I received a phone call from Rubin Feldman, Thermal Science, Inc. (TSI). Mr. Feldman made a report of potential concern.
On October 18, TSI was notified by Texas Utilities (TU) that less than ten of 2,000 conduit sections provided by TSI showed signs of delamination and some had occluded porosity in the cross sections of the conduit sections.
Mr. Feldman indicated that he met with TU on October 18 to fully understand ,
this potential concern and that TSI is evaluating the issue to determine cause and corrective action. He indicated that TSI's evaluation would be completed in 30-45 days.
Mr. Feldman indicated that TU does audit and conduct QA/QC receipt inspections of material at TSI's facility. TV will provide an additional QA inspector to follevup TSI's manuf acturing process. TSI will provide TV and the NRC of its evaluation when available.
t M
Frank J. Miraglia, Deputy Director .
Office of Nuclear Reactor Regulation DISTRIBUTION:
SPLB TSI File >
Central File /TSI X005 NRC PDR JPartlow :
WRussell AThadani GHolchan i
/
CMcCracken/
RArchitzel l
PHadden JFouchard, PA GMulley, IG EPawlik, RIII SWest JLee AMasciantonio l
D J!
FM ia:jw ,
10,d/92 f d
0FFICIAL RECORD CDPY
- - -- - -- o=u r.= r =g,en
Appenix C 41 D@l C.12 Analysis of Fire Issue Based on plant cperaung expenence over the last 20 years, it has been observed that ty power plams will have three to four sigmhcant hres over the:r cperaung Meume. Prev.ous p assessmems (PRAs) have shown that Dres are a significant hequency, contributing anywhere from 7 percent to 50 percent of the total, consice from frr theseinternal, findings sel-ic. flood, hre, and other events (Refs. C.12.1 and C.12.2). There ar as an midator but can also compromise mitgating systems beca .
the necessary background for the subsequent discus m i of the fire iss C.12.1 Analysis Procedure for NUREG-1150 Fire Analysis as (1) screening and (2) quantification of the remaining unscreened e The screening analysis is comprised oh 1.
Identifcation of relevant fire zones. Those Appendix R (Ref. C.12.4) identified fire non either safety-related analysis. equipment or cabling for such equipmem were cetermined to req 2.
Screening of fire zones based on probable fire-induced mitiating events. Determinatio frequency for all "off-normal" plant plant states locations were made. and determination of the resu!dng fire-induced inniadn 3.
Screening of fire zones based on both order and frequency of cut sets.
4.
Numerical evaluation and culling of remaining fire zones based on 'requency.
4 Aher of dominantthe screening analysis cut rets is completed has eliminated all but the probabilistically si;nincant as follows: . ,
n fire rone 1.
Determine temperature response in each fire :cne.
2.
Compute component fire fragilides. The latest version of the fire growth code, COMP C.12.5). with some modifications is used to calculate fire propagation and equ:pmem fire calculations a+ only performed for the hre areas that sunived the screening analys .
3.
Assess the probability of barrier failure for all remaining cornbinations of fire areas A barri analysis is conducted for those combinations of two adjacent fire areas which, with or w addidonal random failures, remain ther the screemng analysis.
4 non-fire-related random failures is addressed. Appropriate addressed as necessary.
5.
frequencies. As in the internal-event analysis, the TEMAC uncertzinty analysis.
Addidonal detail on the fire analysis methods used in NUREG-1150 may be found in Refe .
C.12. 2 PRA Results Tables C.12.1 and C.12.2 proe:de the core damage frequency results for the Surry and risk assessmems, respectively (Refs. C.12.7 and C.12.E). All fi:e areas that sur ived t are listed. When companng fire-induced ccre damage frequency with the total from a!! oth (including seismic, using the LLNL se:smic hazard curves), fire is 7 percent rry and of the total for NUREG-1150 C-128
Appendix C- L i
I Table C.12.1 frequency /yr) (Ref. C.12.7). Dominant Surry fire area core damage frequencyl i Fire Area 5th 'l Mean Percentile 95th ;
Median Percentile Emergency .
Switchgear Room 6.1 E-6 3.9E-9 3.1E-6 i Control Room 1.6E-6 2.0E-5 1.2E-10 4.7E-7 l Cable Wult/ Tunnel 1.5E-6 6.9E-6 6.5E-10 i 7.0E-7 5.8E-6 Auxi!!ary Building 2.2E-6 5.3 E-7 g 1.6E-6 5.6E-6 Charging Pump. i Semce Water e Pump Room 3.9 E-8 l 1.4 E-10 5.7E-9 :
.- L 6E-7 !
Total 1.1E-5 5.4E-7 8.3E-6 3.3 E-5 *
'I i
Table C.12.2 ;
Dominant damage Pench /yr) frequency Bottom (Ref. fire area core damage frequency contributors (core C.12.S). i t
Fire Area 5th Mean Percentile 95th - i Median Percentile Emergency )
Switchgear Room 2A 7.4 E-7 4.6E-10 1.6E-7 Emergency 3.0E-6 :
Switchgear Room 2B 3.6E-6 I 3.5E-9 2.0E-6 Emergency 1.3E-5 ;
Suitchgear Room 2C 4.7E-6 4.2E-9 2.2E-6 ,
Emergency 1.7E-5 t Switchgear Room 2D 7,4 E-7 .
4.6" 10 1.6E-7 (
Emergency 3.0E-6 l Switchgear Room 3A 7.4 E-7 1
4.6E-10 1.6E-7 i
Emergency 3.0E-6 Switchgear Room 33 (
7.4 E-7 4.6E-10 1.6E-7 3.0E-6 Emergency Switchgear Room 3C 7.4 E-7 4.6E-10 1.6E-7 !
Emergency 3.0E-6 Suitchgear Room 3D S.1E-7 5.3E-10 1.7E-7 j Control Room 3.3E-6 6.2E-6 4.2E-10 i 1.4 E-6 E.0E-6 Cable Spreading !
Room 6.7E-7 9.1 E-9 1.7E-7 i 2.3E-6 ,
Total 2.0E-5 1.1E-6 1.2E-5 6.4 E-5 !
C-129 '
e Appenda C and C.I'.4 for Surry and Peach Bottom, respecusely.19 ven in Tables percent C '2.3 o The overal!
reactor fire-induced year. The mean core damage dommant contributing frequency for Surry Umi 1 was fo plant und to be 1.1E-5 per buildmg. control room, and cable vauk/ tunnel Fires in these four areas comprise percent of theroom total cab fire core damage frequency. In the cases of the emergency switchgear ,
tunnel, ar'd the auxiliary buildmg. a reactor coolant pump seal loss-of-coolant accident (LOCA The fire itself fails cabhng for both t! e high-pressure injection and componen)t co g water systems resulting in a seal LOCA. For the control room, a general transient shutdown panel resuhs m core damage. power-operated reliefthevahe auxi!!ary leads t The overall fire-induced mean core damage frequency for Peach Ecttom Umt 2 wa ound to be 1.9E-5 2C, and emergency switchgear room 2B. Fires ,
in these w tchgear cent of the total fire room th abandonment of the area. Failure to control the plant from e-induced esults in core damage. For the two emergency sw:tchgear rooms, a fire-induced loss of offs:te tram of the emergency semce water (ESW) system occurs Random fail wer and failure of one resuits in stadon blackout and core damage. ure of the other two ESW trains Detailed tracing of control and power cabling revealed that fires can damage th funcdons in cable "pinchpoint" areas. Because of this added edetail most plant (cablesafety locado compared w:th past fue PRAs, many more fire scenarios arose in the inidal pnases of sc most of these scenarios could be screened from further studyreening. by allowing However, for ope failures. The final number of fire areas that sunived screening and ythe of random total fire i d frequency for both Peach Bottom and Surry is similar to that - n ucedfound core damage in previous f reasons invohtng plant-specific cabling configurations. These cable s utpinchpoint for dissimilarareas w e also found in conclusion wid respect to containment systems failure.nost cases a similar to fail a
. C.12. 3 1ssue Definition and Discussion The endcal fire nsk issues and their effect on core damage frequency estimates w of most of the fo!!ow:n; issues.secdon. Uncenainty analysise resuhs discussed in this (documente ponance A fire analysis must rely on a panitioning scheme (for fire frequency) to d t frequencies for any given plant area. Addmonal pardtioning e ermine wiiinfiremost plant areas occurrence cable "pinchpoint" areas have typica!!y been found to occur in se will 10critical occur percent because or less plant fire acnes. This additional panidoning. based on fire propagation .
code calculat
, s used to determine ons i components or their cabling. In older fire PRAs mage (from the entical the ear longer time-to-damage esumates were calculated .
rger area ratics and in the N
- s. As was discussed picpagaden predictions. For NUREG-1150n most Therefore, fue frequencies were decreased to take credit for the fact small ottom fue fires that most fi cases.
would not have yielded damage. For plant areas where hct gas layer res for in the data base ratios of 10 percent or less riso led to reduced fue frequenciesato reflect the s:tuadon a postulated f.re esumate) teduced core damage frequencysaously (using COMPEPS). However, a competing effect also occurred Shoner oesdmates and fire si:e ti predicted fro the effect of allowing less credit for manual suppression of these me f to damage estimates had overshadowed by the o&er impacts noted above). Fire tendng and code upens walk Peach Bonom and Surry fire areas to determine the reasonableness was their ceinion. for th: 5:tuations that own each ofof 1 predicuens. It thethe COMP occurred in NUREG-1150, that reasonable predictions. An analysis of the potential effect of COMPERN resuhsCOMPBPS II cn fae-induced core NUREG-1150 C-130
l i
I Appendix C
-i i
Table C.12.3 ;,
Dominant C.12. 7) . Surry accident sequence core damage frequency contributors (Ref.
t Sequence hiran Core Damage Fire Area Frequency /yr Emergency Switchgear T3D3 WD i Room i
~
6.1E-6 Auxiliary Building !
Cable Vault / Tunnel 2.2 E-6 !
1.5 E-6 T3OD i Control Room i
1.6E-6 Charging Pump. Service ,
Water Pump Room 3.9E-B ,
i Symbol KEY h Definition Di Failure of the charging pump system in the high-pressure injection mode. '
D3 Failure of the charging pump system in the sealinjection flow mode.
O Failure of the SRVs/PORVs to close after a transient. '
T3 Transient consisting of a turbine trip with main feedwater available. i W ,
Failure of component cooling water to thermal barriers of all reactor coolant pump Table C.12.4 Dominant Peach Bottom accident sequence core damage frequency contributors (Ref. C.12.S). -;
i Sequence Fire Area hican Core Damage i
~
t Frequency /yr T BU iU2 i
Emergency Switchgear Room 2A ,
7.4 E-7 Emergency Switchgear Room 2B >
3.6E-6 Emergency Switchgear Room 2C 3.6E-6 Emergency Switchgear Room 2D
' 7.4 E-7 '
Emergency Switchgear Room 3A l 7.4 E-7 Emergency Switchgear Room 3B i 7.4 E-7 Emergency Switchgear Room 3C 7.4 E-7 I Emergency Switchgear Room 3D S.1E-7 T 3 U,U X iU3 Centrol Room Cable Spreading Room 6.2E-6 TiBU,W,X2 W2 6.7E-7 W3 U.V V3Y Emergency Switchgear Room 2C S.1E-7 T BUiW,XzW i
W3 U.V:V3Y Emergency Switchgear Room 2C 2.7E-7 l
l C-131 NUREG-1150
Appendix C Table C.12.4 ' t con'inued)
KEY Symbol Definition B
Fadure of all ac power (station blackout).
T3 Loss of offsite power transient.
T3 U 3 Transient consisting of a turbine trip with main feedwater available.
U2 Failure of the high-pressure coolant injection (HPCI)' system.
U3 Failure of the reactor core isolation cooling (RCIC) system.
U. Failure of the control rod drive ( ;RD) system (2 pump mode).
Failure of the CRD system (1 pump mode).
V2 V3 Failure of the low-pressure core spray (LPCS) system.
W3 Failure of the low-pressure coolant injection (LPCI) system.
W2 Failure of the suppression pool cooling mode of the residual em. heat rem W3 Failure of the shutdown cooling mode of the RHR system.
X3 Failure of the containment spray mede of the RHR system.
Failure to depressurize the reactor coolant system via SRVsaor system.
on the autom X2 Failure RHR system totodepressurize operate. the reactor coolant system to allow thee shutdown Y
Failure of primary containment venting (including makeup to the poo! as .
C.12.11). This analysis concluded that up eto a factor
- s. C.12.10 ome scenarios.
and of 20 A second critical issue was the probability of operator recovery own from theInremote panel. the shutd NUREG-1150 fire analysis, this recovery action was quantified on a consistent bi probabilities calculated for the internal-event analyses. However, as s with recovery since no deta and actuation circu:ts was performed complex control system control interactions w types of interactions exist for either Surry or Peach Bottom the effect t ese recovery from the remote shutdown panel and to increase the fire-induced requency.
core da In two studies where detailed models were developed, interactions between the rem ote shutdownthoughpanel found the and to becontrol electricallyroom were offound.
independent These the control room.interactions existed even own panel was the rem The control systems interaction and fire code prediction issues are two of six issues Risk Scoping Study. Four other issues that will not be covered in detail ressed here but also in the Fire ve the potential to have a significant effect on fire-induced core damage frequency are:
1.
Manual fire brigade effectiveness.
2.
Total environment survival.
3.
Fire barrier effectiveness. and 4
Fixed fire suppression system damage effects.
these issues were not addressed in the Surry andirePeach -
analyses. Bottom the potential to increase the fire core damage frequency when incl d d ies. However, any cf thes ue n an assessment.
NUREG-1150 C-132
bf%\niertt b l
STATEMENT SUBMITTED BY ;
UNITED STATES NUCLEAR REGULATORY COMMISSION f 1
TO THE ,
SUBCOMMITTEE ON OVERSIGHT AND INVESTIGATIONS i
COMMITTEE ON ENERGY AND COMMERCE j UNITED STATES HOUSE OF REPRESENTATIVES ;
i CONCERNING i FIRE PROTECTION FOR U.S. NUCLEAR POWER PLANTS .i l
PRESENTED BY i IVAN SELIN CHAIRMAN !
f MARCH 3, 1993 '
I t
b i
I t
INTRODUCTION Mr. Chairman and Members of the Subcommittee, the Commission is l pleased to appear before you today to address serious questions i about certain aspects of the fire protection measures used in nuclear power plants. Specifically, concerns have been raised regarding the use in many of the nation's nuclear power plants of '
Thermo-Lag 330-1 fire barrier systems to meet the NRC's requirements for the protection of redundant shutdown systems.
In brief, four principal issues have been raised concerning these matters. i First, did the Thermo-Lag material conform to the NRC's standards l as certified by the vendor, Thermal Sciences Inc. (TSI), the l third party testing laboratory, Industrial Testing Laboratories {
(ITL), and the individual licensees? That question is part of an l 4 ongoing investigation that we cannot discuss further here today, !
l
- except to say that in addition to recognizing the efforts of the Inspector General (IG) and the Office of Investigations (OI) in i pursuing this matter, we wish to acknowledge the significant role which allegers have played in bringing specific aspects of this -
matter to our attention. j The second issue, and clearly the most immediate one, is the impact-of deficiencies in Thermo-Lag on the safe operation of the l 79 operating nuclear power plants (see Attachment 1) in which the f material is used. At the outset we want to assure the Subcommittee that the NRC staff has assessed the safety l significance of this issue to the affected plants and has taken
! steps.to assure plant safety. The licensees that use Thermo-Lag
' fire barriers have implemented measures, such as fire watches, to compensate for potentially deficient barriers. We have j confidence that the compensatory actions taken at the affected j
- nuclear power plants provide reasonable assurance of the protection of the public health and safety chile this issue is i being fully resolved.
While our inquiries to date indicate that repairs or upgrading may be needed, the specific plant by plant changes to be undertaken have not yet been determined. ecfore we require !
licensees to undertake additional repairs and upgrades we want to !
make sure that we have adequately identified what criteria are appropriate to decide when standards have been met. Once we are confident that our criteria are adequately defined, the NRC will assess the current capabilities of fire barriers, particularly Thermo-Lag, both on a generic basis and in plant-specific installations, to determine what improvements in fire barrier systems will be needed in order to meet NRC requirements. {
With respect to Comanche Peak Unit 2, a low-power license for .
testing purposes was issued on February 2, 1993. The NRC staff ,
reviewed important design, test and installation aspects of fire f
barriers used at Comanche Peak Unit 2 before they considered ,
issuing the low-power license. Although the utility has taken ,
m compensatory measures which the NRC staff has concluded are adequate, the Commission has reached no conclusions on the subject. The Com=ission currently has scheduled a discussion on a full-power operating license for Unit 2 on March 15. We will receive a public briefing from the staff and the applicant on the !
use of Thermo-Lag barriers in the plant and will closely examine
, the status of fire protection measures at the facility before I deciding on issuance of a full-power operating license, j A third issue is the deficiencies in our regulatory process which ,
contributed to questionable acceptance of the Thermo-Lag material in the first place and to numerous missed opportunities to '
identify and correct the problem since that time. There were serious deficiencies on the NRC's part, as well as on the part of the utilities involved. Our testimony will indicate what we are doing to improve our regulatory process to ensure that problems at nuclear facilities are recognized, evaluated properly and ,
remedied effectively. We have endeavored to share with the '
Subcommittee all the relevant information we can find, so that j'
insofar as is possible, we are all working from the same information. For example, we have asked the Executive Director for Operations (EDO) and the IG to outline the circumstances surrounding the missed opportunities to identify the Thermo-Lag problem (Attachment 2). We are also providing preliminary
, observations by knowledgeable staff on insights and lessons learned from this experience. (Attachment 3).
The final issue to be addressed by the Commission is whether there was any misconduct or dereliction of duty on the part of :
NRC employees that contributed to the mishandling of this matter. !
When the IG's investigation into aspects of this matter is !
finished the Commission will take a hard look at what, if any, !
disciplinary action may be appropriate. When we have finished ;
this phase of our work (in about six months), we will report the
- results and any follow-up actions to the Subcommittee. !
II. HISTORY AND BACKGROUND 3
A major fire that damaged safe shutdown equipment occurred at the Browns Ferry Nuclear Station in March 1975. The fire damaged over 1600 electrical cables and caused the temporary unavailability of some emergency core cooling systems. Because the fire did substantial damage, the NRC established a Special Review Group to evaluate the NRC's requirements for nuclear power l plant fire protection programs. The review group found significant inadequacies in fire protection design and procedures l at Browns Ferry. However, in spite of the damage to the plant as I a result of the fire and the inoperable safety equipment, the !
reactors were shut down and cooled down successfully. The review group concluded that avoidance of further damage was due to the defense-in-depth design used to provide safety in nuclear power
, l l
c 4
plants. They recommended specific improveaents aimed at achieving safety through a combination of: i I
- 1. Preventing fires from getting started.
- 2. Detecting and extinguishing quickly such fires as do ;
get started and limiting their damage;
- 3. Designing the plant to minimize the effect of fires on i essential functions.
In 1976 the NRC developed technical guidance for fire protection !
on the basis of these recommendations. Each licensee was asked to use that guidance to analyze the consequences of a fire in !
each plant area, and to provide a fire hazards analysis to demonstrate that redundant systems required to achieve and ,
maintain cold shutdown were adequately protected against damage '
by a fire. Although significant progress was made plant by plant in addressing fire protection issues, by early 1980 a number of implementation problems remained, including disagreements with :
several licensees over alternate and dedicated shutdown capability. To establish a definitive resolution of these contested issues, in 1980 the Commission issued a fire protection rule, 10 CFR 50.48 and Appendix R to Part 50, which set out minimum fire protection requirements for the unresolved issues. !
The Commission determined that protection of safe shutdown capability in case of fire was one of the three items of such '
significance that it must apply to all operating and new plants.
^
Even after the rule was issued, questions of interpretation arose. Generic letter 86-10 (GL 86-10) was issued in 1986, with the intent of answering these questions. Unfortunately GL 86-10 did not resolve important uncertainties in testing criteria, l which the staff is trying to resolve now.
III. STANDARDS AND TESTING CRITERIA NRC's fire protection requirements prescribe a defense-in-depth approach to protect safe shutdown functions, through (1) fire prevention activities (limits on combustibles through design, construction and administrative controls); (2) the ability to detect, control, and suppress a fire rapidly (trained fire l brigades); and (3) physical separation of redundant safe shutdown functions and emergency lighting for safe shutdown actions and access. Appendix R,Section III.G. of 10 CFR Part 50 specifies three approved methods, any one of which is an acceptable method, to ensure that at least one means of achieving and maintaining j safe shutdown conditions will remain available during and after any postulated fire in the plant:
-3 -
- 1. separation of the redundant system by a passive barrier able to withstand fire for at least three hours; or
- 2. separation of the redundant system by a distance of twenty feet containing no intervening combustible ;
material, together with fire detectors and an automatic fire suppression system; or-
- 3. separation of the redundant systems by a passive ,
barrier able to withstand fire for one hour, coupled with fire detectors and an automatic fire suppression y system.
One objective of this requirement is to protect the electrical !
cables that provide power and control signals for equipment ,
needed to shut down the reactor safely. Many licensees have used cable raceway protective envelopes, such as Thermo-Lag fire barriers, to satisfy these separation requirements. Over the years the NRC has provided guidance on acceptance criteria for ,
use in determining whether the rassive barrier standard is met.
Much of the guidance was adopted without modification from standard fire endurance tests and specifications for fire walls, >
as developed by such groups as the National Fire Protection- '
Association (NFPA). Rather than developing standards more specifically tailored to structures such as electrical cable raceways, these tests and specifications were judged to be acceptable surrogates. ,
As I mentioned earlier, we have concluded that the acceptance criteria embedded in GL 86-10 require clarification with respect to the process for determining whether an electrical cable will continue to function in the event of a fire. Refinement of the criteria is important even now because we recognize that repairs ,
and upgrades on a large scale may be required. However, before ;
we require any actions by licensees, we3also need to know the ,
actual characteristics of the existing fire barriers. That is why we have been pursuing an aggressive program, which includes independent small scale testing by the NRC, industry testing programs and requiring licensees to complete their assessment of the fire barriers installed in their facilities.
The NRC's fundamental reculatory recuirement to provide one hour or three hour fire rated barriers to separate redundant safe shutdown functions within the same fire areas has not been chanced. Nor, from an enforcement persoective, will the revised criteria be applied retroactive 1v to TSI and ITL certifications.
Each licensee's obligation was to meet the standards in effect at the time of approval of their fire program. No subsequent reexamination of the acceptance criteria will excuse them from their responsibility to comply with the criteria in effect at the time of their certifications.
r
i one of the proposed modifications to NRC's acceptance criteria li concerns the hose stream test. The proposed NRC position states that the hose streem test may be performed with fog nozzles where ,
licensees' fire fighting plans call for the use of fog nozzles.
The NRC considers the use of a fog nozzle for hose stream testing to be consistent with the intent underlying the acceptance i criteria in NFPA Standard 251 and with typical fire fighting ;
practice for electrical equipment.
The revised guidance is currently undergoing NRC's internal '
review process and will be issued for public comment before it is finalized. We exnect that the staff's revised criteria should l' not be less ricorous than the GL 86-10 criteria, except in broadenina the rance of fire hose nozzles that may be used. ,
A more complete description of NRC's fire protection standards is contained in Attachment 4.
IV. WEAT WENT WRONG e
A. What is Thermo-Laa? l Thermo-Lag is a proprietary fire barrier material manufactured and supplied by Thermal Science, Inc. (TSI), St. Louis, Missouri.
Thermo-Lag is supplied as prefabricated panels, preshaped sections that fit around conduits, and mastic (grout). Thermo-Lag fire barriers are constructed from the panels and conduit sections. The joints are sealed with the mastic material.
1 The function of the Thermo-Lag is to protect the cables located .
within the Thermo-Lag enclosure from the heat of a fire until the fire can be detected and suppressed. Thermo-Lag provides this protection by absorbing the heat and vaporizing in the process.
Thermo-Lag is normally used to enclose one train of redundant ,
saft shutdown cables when both trains of cables required to achieve safe shutdown are located in the same fire area.
B. What Hannened?
Prompted by an allegation, in June 1991 the Director of NRR established a special team to perform a comprehensive review of [
Thermo-Lag fire barrier system fire endurance and ampacity derating test reports, installation procedures, and as-built j configurations.
The special review team presented the results of its review on February 11, 1992. They concluded, concerning the current safety situation, that:
e The fire resistance ratings and the ampacity derating factors for the Thermo-Lag 330-1 fire barrier system are indeterminate. (That is, the team could not make a general f
r r
statement on Thermo-Lag performance because of inconsistencies in the TSI/ITL test results).
m 7 e Some licensees have not adequately reviewed and evaluated the test results, used as the licensing basis for their !
Thermo-Lag barriers, to determine the validity of the tests and the applicability of the test results to their plant >
designs.
- Some licensees have not adequately reviewed the Thermo-Lag i fire barrier configurations installed in their plants to ensure that they are supported by tests or appropriate engineering evaluations.
e Some licensees used inadequate or incomplete installation procedures during the construction of their Thermo-Lag ;
barriers.
- The NRC staff's licensing reviews and inspections of the barriers may not have been commensurate with either the problems or the importance of the barriers.
At the same time as the NRR special review team was conducting its study, the Inspector General began his inspection covering the Commission's handling of Thermo-Lag matters. On August 12, 1992 his report concluded that the NRC had not conducted an adequate review of fire endurance and ampacity derating information concerning the ability of the fire barrier material, Thermo-Lag 330-1; if it had, a number of problems with the test program and Thermo-Lag could have been discovered. In addition, the IG identified a number of missed opportunities between 1982 and 1991 when the staff had information that, if pursued, could have identified the generic issues earlier than 1991. The facts surrounding these missed opportunities are enclosed as attachment 2. In summary, the seven occasions identified by the OIG were:
(1) In 1982 the staff found Thermo-Lag tests conducted by the vendor unacceptable for application to the Susquehanna Nuclear Power Plant and then accepted another test conducted by the vendor for the Washington Nuclear Project 2 (WNP2).
(2) In 1985, an NRC inspection at Fort Calhoun Station found that the licensee had not verified ampacity derating figures provided by its fire barrier vendor, 3-M. The inspector did realize that analogous questions should also be put to TSI and obtained the TSI test data, but did not follow-up on the review of the TSI test data.
(3) In 1986 TSI reported new test data for the ampacity derating factor for cables protected by Thermo-Lag.
?
(4) In 1987 Comanche Peak provided a report on Thermo-Lag ampacity derating errors.
(5) In 1989 an allegation was received that Thermo-Lag gave off ;
lethal gases when burned. The allegation also mentioned that Thermo-Lag had " disintegrated" and had been " consumed in a fire test".
(6) In 1989 NRC Region IV was informed that Thermo-Lag panels '
arriving at Comanche Peak measured less than the required thickness. The NRC staff closed the issue without a sufficient technical basis and without pursuing the generic implications with the manufacturer, TSI.
(7) In 1989 the River Bend nuclear power plant submitted an ,
Informational Report stating that a Thermo-Lag test had failed. It was not until May 1991 that the NRC staff followed up on the generic aspects of this issue and only then because of additional allegations. ,
The NRR special review team found additional instances of specific inspection issues that had been competently handled as .
specific cases, but where the generic applicability continued to i be missed.
4
- c. Evaluation of Safety Incact !
Without a doubt, in the event of a fire, Thermo-Lag's failure to meet the original specifications could have had a negative impact on the ability of the plants to resist fire, but the extent of this impact will not be known until the staff completes their ,
action plan. As I noted earlier, the NRC staff has assessed the safety significance of this issue to the affected plants and has taken steps to assure plant safety.
It is also important to consider the question of passive barriers within the broader question of fire protection in general.
Several observations can be made. Each is based on fragmentary data, but when taken together they provide a consistent picture.
Based on a detailed probabilistic risk assessment (PRA) of two i plants and a partial assessment of half a dozen more, the over- l all effect of implementation of the fire protection rule (Appendix R) appears to be a reduction of at least an order of i magnitude in core damage risk due to fire compared to what existed before the adoption of the rule. l Based on the compendium of licensee event reports (LERs) prepared by NRC's Office for Analysis and Evaluation of Operational Data (AEOD), the number of significant fire events.has decreased steadily from .13 per plant per year (rolling average, 1980-1982) l l
to .04 per plant year (1990-1992) since passage of Appendix R. '
(Attachment 5)
Based on discussions with Electric Power Research Institute and ,
some licensees which use Thermo-Lag, fire protection defense-in-depth has prevented any fire from reaching the stage where any Thermo-Lag barrier has been challenged in a real-life situation.
Real progress has been made in reducing fire-induced risk in '
nuclear power plants. Nonetheless, additional work must be done to ensure the adequacy of fire barrier material as an element of fire safety defense-in-depth. We also see opportunities for improvement in fire prevention (not just in passive barriers) identified by the staff study (Attachment 3); we also see this in the current PRAs of eight plants which suggest, in spite of methodological differences, that fire risk still contributes approximately 20% to overall nuclear power plant core damage frequency.
- We expect the potential generic safety significance of underestimates of ampacity derating to be less important. First, the effect of thermal insulation on ampacity is an aging phenomenon, not an immediate one. More importantly, cables enveloped by Thermo-Lag insulation either (1) carry very low ,
currents with negligible heating (control cables), or (2) may carry high currents, but do so intermittently, e.c., valve motor operators are only energized for short periods of time, usually one to two minutes. Emergency core cooling pump motors are powered during short periods for periodic surveillance testing or during reactor shutdown for residual heat removal. Very little thermal embrittlement of cable insulation or jacket material would be expected in these applications. Some systems, such as chemical and volume control, service water, and component cooling water systems are needed both for normal plant operations and for safe shutdown decay heat removal. Cables which supply power to ,
these pump motors normally are sized with sufficient design '
margin to accommodate additional ampacity derating. Also, many plant designs provide extra pumps to anticipate equipment '
breakdown and maintenance.
D. Follev-up to the Two Reports In early July 1992, after reviewing the special review team's final report, the NRC staff developed a comprehensive action plan to resolve the issues associated with testing, design and installation of Thermo-Lag fire barrier systems.
Following receipt of the IG's report in August 1992, on behalf of the Commission I directed the staff on August 17, 1992 to determine whether additional measures were necessary to evaluate i the adequacy of Thermo-Lag and to address the programmatic l
i l
i
i
[
findings of the IG. (Attachment 6) Specifically, I directed the !
staff to address: ,
- 1. The reasons the initial review process did not identify the i problems with Thermo-Lag 330-1 and the cause of deficiencies i in our response to later indications of problems that were ;
brought to the agency's attention;
- 2. Whether the problems identified with respect to the initial review and the lack of follow-up to later indications of problems represented a systematic weakness in our review and i response programs; and
- 3. What corrective actions were necessary to rectify the deficiencies identified with respect to the review and ,
response processes. ;
I also directed the staff to submit quarterly reports on its progress toward completion and to report expeditiously any
- significant findings or obstacles to completion of their action.
i The staff has been implementing an aggressive action plan to ;
review and resolve the technical deficiencies of Thermo-Lag. ~
(Attachment 7). Completed actions include: issuance of Bulletin !
92-01 and its supplement requiring compensatory measures for [
inoperable fire barriers; issuance of Generic Letter 92-08 i requiring licensees to describe their assessment of the acceptability of Thermo-Lag fire barriers; sponsorship of toxicity, combustibility, and small-scale fire endurance tests of !
Thermo Lag barrier materials; and monitoring the testing of i upgraded Thermo-Lag fire barriers installed at TU Electric's l Comanche Peak Unit 2. Significant activities in progress include l the development of fire barrier acceptance criteria, review of TVA's full-scale test program and that planned by NUMARC, and review of industry's plans for an ampacity test program that will adequately determine appropriate derating factors. Future :
activities called for by this action plan include consideration of issuence of a supplement to Generic Letter 92-08 to address '
fire bairiers manufactured by other vendors; combustibility of i fire barrier material; and additional fire barrier and ampacity testing. In addition, there will be continued inspections of fire barrier installations at the plants.
But the action plan addressed only part of the problem. To go further and fully understand all of the regulatory lessons to be ,
learned requires a thorough examination of the events of the 1982-1991 tine period.
V. LESSONS LEARNED The staff's programmatic reassessment in response to the Commission's directive is well underway. We have recently i
Y received a review and analysis by technical staff in response to l our request for a programmatic reassessment of the NRC fire '
protection program. While final review of this report has not been completed by senior management or by the Commission, I believe this report is a constructive and insightful effort which <
should provide a sound point of departure for identification and ,
implementation of programmatic refinements. A copy of this l report is provided as Attachment 3.
Included below arc a few quotations from the staff's report which give some indication of the frank nature of their findings:
1 "The underlying cause of the problems appears to be poor staff communications and poor work coordinction.
In too many cases, staff reviewers were unaware that the issue they were reviewing was also being reviewed by other staff members at headquarters or in the regional offices; or that it was related to an issue previously reviewed. Many of the tracking systems
...available to assist staff members are incomplete and
... difficult to use. Efforts by both IRM and NRR, Events Analysis Branch to coordinate event-related information have been helpful in addressing the need for follow-up information, but those efforts are not widely known ...
"It has been and will continue to be difficult for (NRC] managers to make informed decisions in the fire protection area since the subject is removed from the experience and expertise of most managers and the complex history of NRC fire protection requirements may not be fully appreciated. It is incumbent on the managers to take a greater interest in the subject on a day-to-day basis and to make themselves available for training or other learning experiences. In the past, it appears that too many important decisions were made at the reviewer or inspector level without the advice and consent of management. It is incumbent on the NRC fire protection engineers to take the extra effort to involve management in important fire protection issues.
"In the area of information management, as it related to the follow-up of problem indications, management involvement and effectiveness have been mixed.
Management has always insisted that information tracking systems be put in place and management has been supportive of the related hardware needs...
However, management has not been sufficiently involved or sensitive to the practical problems of information 10 -
t i
[
management. .... The staff and management have yet to ;
adopt a full commitment to accurate, timely and useful j information management...
i
~
l "In the area of follow-up of indications of proble=s, l the staff's performance has not been adequate. ... In l some cases there appeared to be a reluctance on the part of the staff to aggressively pursue issues because of a general view that fire protection concerns were rarely serious safety concerns. These failures have resulted in at least a three to five year delay in addressing the issues of fire endurance and ampacity '
derating."
l >
i A self-critical assessment of these missed opportunities shows that under the circumstances revealed thus far, the NRC was too i passive in its review and acceptance of the test results reported by TSI and ITL. Where information was submitted indicating a potential problem, reviewers appear not to have aggressively pursued issues in the fire protection area, and failed to .
recognize the generic implications of the matters before them. I i The reports presented themselves as isolated events, and none of l our tracking methods provided a mechanism for identifying the i pattern that was developing. l I
The generic regulatory implications were significant. First of all, it is the utilities' responsibility to determine the characteristics of materials that they procure. They provide their results to NRC based on their own quality assurance .
programs. There appears to have been a widespread failure of i these licensee QA programs with respect to fire protection.
NRC's slowness in recognizing the situation has effectively >
delayed us in holding the utilities to their responsibilities or from applying timely enforcement. Second, by not carrying out i our responsibilities in a timely fashion, we missed our .
opportunity to apply timely oversight and enforce =ent which would !
have assured that utilities were fully meeting the objectives of the fire protection rule.
Once the senior management has had an opportunity to complete its review of this report, the Executive Director for Operations will come to the Commission with his proposals for a program to implement the recommendations in the report or others which he t may make.
It is already clear, however, that at a minimum, actions such as the following will have to be taken:
- 1. An evaluation of the current role of the Office for Analysis
. d Evaluation of Operational Data (AEOD) to determine l
l
whether it could aug=ent its responsibilities with respect to tracking and evaluating safety information to assure that relevant safety information, irrespective of source, is promptly and properly acted upon.
- 2. A reevaluation of information management systems to maximize the availability of all relevant safety information to staff reviewers and inspectors and the coordination of related activities.
- 3. A reevaluation of the criteria by which staff reviewers, inspectors and investigators identify potential generic issues.
- 4. An evaluation of the adequacy of the number and utilization of our fire protection engineers.
We vill keep the Subcommittee apprised of the recommendations received and the status of their implementation on a quarterly report cycle.
VI. ARE THERE OTHER ANALOGOUS SITUATIONS WHICH REMAIN UNDISCOVERED?
The generic characteristics of the Thermo-Lag issue have been analyzed to create a template which we can apply to other aspects '
of our regulatory program in order to identify other safety reviews where shortcomings might have occurred. We have identified three generic characteristics of the Thermo-lag review.
The first is the adoption of a surrogate standard for a new use without fully analyzing whether the standard was sufficiently suited for application in nuclear power plants. When the Commission adopted its regulation requiring fire barriers between redundant trains of the safe shutdown systems, the generally understood concept of a fire barrier was a wall. When licensees proposed to wrap cable trays with a fire retardant material (a new use of these products) the Commission accepted use of the !
material without adequately examining the proposed use or whether the general industry fire barrier standard was appropriate for this application.
A second characteristic of the Thermo-Lag experience was that the fire barrier review was conducted as part of a much larger review of a number of fire protection measures and it did not receive the scrutiny that an important component should have received.
The third characteristic was the establishment by the NRC of a ;
certification program without adequately defining the !
qualification criteria which the certifying agent should meet and inspecting the certifying agent against those criteria. As a general practice, much of NRC's ongoing review of safety matters is predicated on review and audits of licensee or third party such reviews and certifications. Appropriately impi a.o ~ed, programs provide reasonable assurance of adequau m Nty and are an efficient use of resources. When we or our in." :sceu do accept third party reviews, however, the obligation of each of us is to assure that the third party is reliable and technically competent. In this instance, as the IG points out in his report, .
the staff accepted licensee provided certifications from TSI and i ITL without assuring that the licensee had verified that ITL had '
the characteristics of an approved laboratory, and without obtaining satisfactory assurance from the licensee or on our own that it was conducting its tests in accordance with appropriate testing procedures. .
The staff is still in the process of its review to determine whether this template applies to other agency activities. If programs with the potential for similar problems are identified, they will be reevaluated at least on a sa=ple basis to determine !
if a full re-evaluation of those programs is warranted.
VII. WHAT'S TO PREVENT THIS FROM EAPPENING AGAIN?
In addition to the specific measures mentioned earlier, several general developments at the Commission in the last few years provide-confidence that a recurrence of a Thermo-Lag type problem is unlikely.
One is the efforts of AEOD. AEOD is responsible for the review f and evaluation of operating experience in order to identify significant events and their root causes, the generic l implications of these events and concerns, and the adequacy of corrective actions. AEOD serves as an independent source within the agency for empirically evaluating NRC's response and assessment capabilities. .
Several new communications techniques have been developed within the agency. For instance, event reports are distributed electronically within the agency on a daily basis. Individual -
plant issues are also discussed daily between the NRC resident office inspectors and their supervisors in the regional offices and between the regional offices and their counterparts at NRC ;
headquarters. The significant issues are discussed in a weekly events briefing with all five regional offices. Activities such '
as these allow for a greater number of the staff to become involved with developing issues and with issues generally regarded as of lesser safety significance. i The Commission also has in place an Allegations Management .
Tracking System. It is designed to capture all allegations received from any source about any NRC licensed activity and to i provide for a disciplined process of evaluation and close-out.
i l
i
, ~ ,
The staff's reassessment efforts may indicate some opportunities
- to improve this system, but our understanding of the IG's report i is that, for the most part, once the NRC received allegations I about Thermo-Lag, our program worked as designed. l The sensitivity on the part of the staff to the importance of !
fire risk on a generic basis has certainly been greatly increased '
by this experience. In other areas where this sensitivity already existed (e.g., motor-operated valves, pressure vessel ;
embrittlement) the staff has been much quicker to appreciate ;
generic implications when a specific problem arose. l Finally, the IG's organization itself has been of significant j help in the assessment of NRC performance. In addition to the reviews which they undertake on their own initiative, they have been responsive to our requests for programmatic reviews. In the past year, at my request, they have been conducting a review of our inspection program. They have participated in several Joint l Task Forces with the Office of Investigations, where their ,
presence has provided an added dimension to the investigation.
We have been successful at devising a protocol for these joint efforts which preserves the IG's independence and ability to .
direct efforts when he believes necessary, but at the same time give his office and OI investigators access to a full array of information. Representatives of the IG's office have also been -
welcomed as observers on several formal Incident Investigation i Teams, where their expertise has proven valuable. !
That completes my statement. We are prepared to answer any ;
questions you may have. ;
i P
A l
1 l
l l
I
I ATTACHMENTS TO MARCH 3, 1993 STATEMENT )
SUBMITTED BY THE U.S. NUCLEAR REGULATORY COMMISSION TO THE !
SUBCOMMITTEE ON OVERSIGHT AND INVESTIGATIONS t COMMITTEE ON ENERGY AND COMMERCE t U.S. HOUSE OF REPRESENTATIVES {
CONCERNING FIRE PROTECTION FOR U.S. NUCLEAR POWER PLANTS !
i TABLE OF CONTENTS !
> : Operating Reactors That Use Thermo-lag Fire Barriers; Operating Plants That Do Not Use Thermo-Lag; Plants Deferred, Under Construction, or !
Shutdown. : Summary of Missed Opportunities. j
! : Preliminary NRC Staff Re-assessment of the NRC ,
! : NRC Fire Protection Regulations and Guidance. !
I : Fire Event Data Review. {
.! : Memorandum from I. Selin to J. M. Taylo*-
August 17, 1992;
Subject:
Inspector General's !
Inspection of the NRC Staff's Acceptance and i Review oi Thermo-lag 330-1 Fire Barrier Material, ft
! : NRC Action Plan - Resolution of the Thermo-Lag Fire Barrier Issues. l i
i
[
i I
e f
J 5
- _.