ML20005G150

Forwards Rev to Informational Rept Re Test of thermo-lag Fire Barrier Matl Transmitted on 891220.Rept Provides Info Re Ongoing Investigations of Issue & Interim Corrective Actions Taken

ML20005G150
Person / Time
Site:	River Bend
Issue date:	01/09/1990
From:	Booker J GULF STATES UTILITIES CO.
To:	NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM)
References
RBG-32073, NUDOCS 9001180202
Download: ML20005G150 (4)

Text

u .; 4;- + ,

-)

F

[th;;. s '

"I + i h?$$ f's f i'jf h; 1

(! p ,

'.o*. ,

GULF. STATES ' UTELETIES' DOMPANY ,

U A_ ;

wta eno sieoN eesi omer nox 2:0' " $f f MANCISVftif. LOUIS 6ANA 70776 .

t s

AftE A CODE 603 f 36 6094 346-E6L i

.f.F W '

January 9 ,1990

,e

~ RBG- 32073 .

.y File Nos. G9.5, G9.25.1.4-R '

y ' ,

20.S.: Nuclear Regulatory Commiision

< 4 Document Control. Desk r Washington, D.C.- 20555

~

.) :

' ^

, - Gentlemen: -

River Bend Station' -Unit 1 i Docket No. 50-458 I <

Pleasefind'encloseda'revisiontoithe5 Informational" Report regarding a'recent test of Thermo-Lag fire barrier material which is us'ed at River Bend Station. The initial report was originally

. transmitted on December 20,-1989. ;This report is being submitted

to provide information regarding our ongoing investigation of this matter and interim actions taken.

Sincerely, , ,

hk

• refJ.E.. Booker -

A Manager-River Bend Oversight

- River; Bend Nuclear Group JEB/TFP/RGW/ CMC /pg

~

cc: U.S. Nuclear Regulatory Commission  ?

611 Ryan Plaza Drive, Suite-1000 g Arlington, .TX 76011 NRC Resident Inspector P.O.. Box 1051

,' St. Francisville, LA'70775 Mr. Walt Paulson .

U.S.. Nuclear Regulatory Commission

% 11555 Rockville Pike Rockville, MD 20852

,. pO[ Ob '

F g

, \

n -

4

C-l V

, REPORTED CONDITION O At approximately 1100 hours0.0127 days <br />0.306 hours <br />0.00182 weeks <br />4.1855e-4 months <br /> on 10/28/89, with the_ unit in Operational b Condition 1,.a problem was reported in a test of plant-specific Appendix R g fire barriers as described herein. Since the results of this. test placed-into question the qualification of Thermo-Lag fire barrier material, Condition Report (CR) 89-1144 was initiated which detailed the areas in the plant protected by these fire barriers. All of these areas, with the exception of the piping tunnels and the upper elevations of the reactor building, were being covered by preexisting firewatch patrols. As a I conservative precaution, the other areas were added-to the firewatch routes satisfying the action statement of section 3/4.7.7 of the plant Technical y Specifications.- I

.This condition is- currently determined not to be reportable pursuant to

~10CFR50.73 because the test results are indeterminate and thus the. impact on installed River Bend Station (RBS) equipment is unknown. If the  ;

evaluation determines that RBS equipment has been inoperable due to  !

inadequate fire barriers appropriate reporting requirements will be  !

evaluated and satisfied. J INVESTIGATION The fire barrier test was conducted to verify barrier performance and to  !

compare the three hour rated fire barrier products of two competing m = manufacturers. 0ne material used in the test, Thermo-Lag, produced by Thermal Science, Inc. is the material typically used at River Bend Station

- for one and three hour Appendix R fire barriers. The other material was 1 undergoing initial qualification testing and is not currently in use at  !

.RBS.: Standard site installation procedures were used to install the '

Thermo-Lag material on the test apparatus. Both materials were applied and 3 y inspected by Gulf States Utilities (GSU) personnel on identical 30 inch j wide aluminum cable trays. Both barrier materials were also used to i protect the tube steel support underneath the trays, coming into contact near the midpoint of-the support. Testing was performed in accordance with the American Nuclear Insurers test standard, including monitoring of circuit' integrity.

-During the performance of the test, it was noticed that thermocouples i inside the Thermo-Lag tray enclosure were experiencing abnormally high temperatures in one area. At approximately 41 minutes into the test, the

-Thermo-Lag covering the bottom of the support fell off, exposing the steel support. As the test continued, temperatures inside the cable tray i enclosure continued to increase, with a loss of circuit integrity at 47 minutes.

As a result of GSU's ongoing investigation some generic issues have been

- revealed during a recent conference with the vendor, Thermal Science, Inc.

as follows:

-1. Use of Aluminum Conduit - No testing has been performed to evaluate the effect of aluminum conduit penetrating the protective envelope. The typical "18 inch rule" where Thermo-Lag covering limits heat transfer due to penetrants may or may not be sufficient.

Page 1 of 3 1

y ~ , m

4; -

F  %. .M 12.. . Joints in Protective Envelopes - Two methods for ,ealing joints are' r r shown in the TSI~ installation manual. One method involves coating U board butt edges prior to installation, called "prebuttering". The

[ second method allows all board material to be dry fitted and the joints W covered with trowel grade material once in the final position, called h " skin coating". This second method, although sanctioned in the TSI "

manual, has not been tested.

3; Size of Tested Configurations - Until the most recent test series conducted with GSU, the maximum size of barrier tested is apparently a 12 inch wide cable tray. However, enclosures of much larger sizes are i

+

typically used at River. Bend Station, based on extrapolation of data.

r from tests'on-12_ inch _ wide trays. The effect on barrier performance P

due.to this-larger size in unknown.

-i

,The results of the testing performed on the Thermo-Lag aarrier continue to

~

be studied by GSU and the vendor, Thermal Science, Inc. The influence of

^ '

the. dissimilar material joint on the overall results of the fire test is in

-question. This application is not typical at River Bend Station.

Evaluation of.the test will continue until the results are attributed to either test article construction or test performance. At that time, the applicability of the results to the barriers installed at RBS will be determined.

CORRECTIVE-ACTION

-In addition to conservatively satisfying the firewatches specified by the action statement of Technical Specification 3/4.7.7, GSU Engineering has

'specifiert when stationary or 30 minute roving firewatches will be posted for inoperative fire suppression systems or fire detection zones in safety

related areas of the plant which utilize Thermo-Lag fire barriers. Use of either a stationary or 30 minute roving firewatch is justified as the inspection-frequency exceeds the worst case barrier rating. This action

-conservatively exceeds the current Technical Specification requirements for fire suppression and detection systems. The procedure governing control of transient combustibles has been revised to require that all combustible <

liquids brought into the power block remain in the possession of the worker until they-are removed from the building. As an alternative to removal from the building, small amounts of liquids may be stored in the flammable storage lockers in the plant.

GSU will~ continue to evaluate the test results and will provide an updated report by March 31, 1990.

SAFETY ASSESSMENT Fire safety was and is an integral part of the design of RBS. This process begins in the selection of the cable used, which is IEEE rated and fire resistive. (Electrical cable insulation forms the majority of the fixed fire load in the plant). Fire detection systems cover the entire power block, giving early warning of fires. Fire suppression systems are provided in areas such as cable chases and diesel rooms with large concentrations of combustibles. The various buildings are subdivided into discrete fire areas, usually by concrete walls and floors. This ' defense f in depth' philosophy is crafted to detect fires in the early stages, I contain fires in one area and provide control and extinguishment.  ;

Page 2 of 3 l 1

j

• !y e .;' '

J,,;. ,

.= . . = . .

' ,

/ Themo-Lagibarriers form a small fraction of -the total number of fire-barriers, y 2. . -

Based on.the. interim actions implemented, the fire protection design

features:at RBS:and the generally low combustible loadings .in the fire

r . areas.1-GSU concludes that' continued operation is justified, c

pp .

l

-l t

1 i

. E 'l l

, s 55i i

)

= .

. i Page 3 of 3