ML20091Q832

Forwards Ul Rept on Fire Resistant Cables & Vtec Lab Test Rept on Rockbestos Cables.Util Sponsored Development & Testing of Fireproof Cable as Alternative to Installation of 1 H Fire Barriers to Meet 10CFR50,App R Requirements

ML20091Q832
Person / Time
Site:	Three Mile Island
Issue date:	06/09/1984
From:	Hukill H GENERAL PUBLIC UTILITIES CORP.
To:	Stolz J Office of Nuclear Reactor Regulation
Shared Package
ML20091Q834	List: ML20091Q832 ML20091Q844 ML20091Q847
References
5211-84-2141, NUDOCS 8406140174
Download: ML20091Q832 (6)

Text

7 GPU Nuclear Corporation Nuclear  :::,orrs:r8o Middletown, Pennsylvania 17057-0191 717 944 7621 TELEX 84 2386 Writer's Direct Dial Number:

June 9, 1984 5211-84-2141 Office of Nuclear Reactor Regulation -

Attn: John F. Stolz, Chief Operating Reactors Branch No. 4 U. S. Nuclear Regulatory Commission Washington, D.C. 20555

Dear Mr. Stolz:

Three Mile Island NJclear Station, Unit I, (TMI-1)

Operating License No. DPR-50 Dodet No. 50-289 TKI-l Fire Proof Cable Test Program - Test Reports

References:

1) Underwriters Laboratories Inc. Report File R10925-1,

" Report on Fire Resistant Cables", dated April 10, 1984

2) GPUN Letter 5211-83-324, dated November 30, 1983.

3) NRC Letter, from John F. Stolz to GPU Nuclear Corporation, "TMI-l Fireproof Cable Test Program", dated March 15, 1984.

GPU Nuclear Corporation sponsored the development and testing of a fireproof

. cable by the Ro&bestos Company. The intent of the program was to develop a fire-rated cable as an alternative to installation of one hour fire barriers to meet the requirements of 10 CFR 50, Appendix R. Fire-rated cable

- represents a practical alternative for providing levels _ of fire protection equivalent' to the methods listed in 10 CFR 50, Appendix R, Section III.G.

Enclosed is a copy of the independent test report, " Report on Fire Resistant Cables". by Underwriters Laboratories Inc., (Reference 1) for your review and

. evaluation.-

The test program was submitted to the NRC by Reference 2. The comments in Reference 3 were discussed with'GPUN prior to the initial testing on February .

21, 1984 and are described in Attachment 1. GPUN believed many of the concerns expressed in Reference 3 have been addressed by one of the following:

'0406140174 840609 PDR ADOCK 05000289 F PDR 0 h k

GPU Nuclear Corporation is a subsidiary of the General Public Utilitl'es Corporation t

l

1) The methods by which the testing was performed (e.g.,

distributed flame during the oven exposure; additional fire hose exposures).

2) The subjection of sheathed fireproof cables to mechanical stresses during the full scale testing.

3) Performance of a small scale test using 450 V AC o simulate the ,

actual voltage stress conditions.  ;

i I It should be noted that the full scale test period was extended from the i

. planned 80 hours9.259259e-4 days <br />0.0222 hours <br />1.322751e-4 weeks <br />3.044e-5 months <br /> to 94 hours0.00109 days <br />0.0261 hours <br />1.554233e-4 weeks <br />3.5767e-5 months <br />, j In addition, the Rockbestos cable was tested by VTEC Laboratories on January 16 and 17, 1984, for performance at high voltage after a one hour exposure fire and water hose stream test. A copy of this test report is also enclosed

. for your review.

Based on the testing to date, the Rockbestos Company is developing engineering data which will allow designers to properly size each application of fireproof cable to provide acceptable circuit performance during a one hour fire exposure (ASTM E-119). GoVN will provide a specific exemption request for each application of fireproof cable at TMI-1. The use of fireproof cables in lieu of one hour fire barriers or radiant energy heat shields at TMI-l will be >

subject to the following guidelines: ,

1) Fireproof cables may be used for safe shutdown circuits inside containment in lieu of radiant energy heat shields.

2) Fireproof cables may be used for safe shutdown circuits in lieu of protection of such circuits by one hour fire barriers in locations protected with fixed automatic fire suppression and detection systems.

3) Locations for the use of fireproof caoles for safe shutdown circuits in lieu of one hour fire barriers for protection of such circuits will be as follows: a) circuits which will be required only during hot shutdown and cooldown; b) circuits required for hot shutdown, cooldown and cold shutdown for which one train can be repaired within 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br />. All other circuits required during cold shutdown will oe rerouted, protected with fire Darriers or repaired within 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br />.

It is requested that the NRC review the report enclosed as GPUN is proceeding with the design nodification based an the guidelines discussed above.

Sincerely,

. D. Bukill, Director, TMI-l

.HDH/MI/mle Attachment (1)

Enclosures (2) cc: J. Van Vliet D. Kubicki e V. 2 cia Fyn

4 * .. .p

.c o

.i g ~

Attachment 1

- GPUNiresponse to IEC comments listed in NRC letter dated March 15, 1984, w- entitled;"TMI-1EFire Proof Cable Test Program".

The test program gives no consideration to the effects of l iComment fli 4 '

~

collapsing cable trays or other debris thatymight be generated A

.-during an actual fire. in the plant. This consideration is the same, of- course, as for. one hour enclosures and cable wraps. .

' Under postulated fire conditions in the- plant, temperatures -

.would rise to such a' levels that cable trays,' HVAC ducts, 211ghting fixtures and other such physical' features could be c, ' damaged .to the point of collapse. If such features were located in the proximity of.the fire. rated cable,1 collapsing debris

. ;would likely cause1significant damage and might affect continued o fcableoperability, This would have to be considered for each use on a case-by_-case basis.

• IResponse:f ;The's@ ports for tray and conduit in the' test configuration were~ i fireproofed to prevent their collapse,' as the prime purpose, of.-

~

~

the test was to_ evaluate cable performance under fire conditions

• .in; trays and in conduit.

[

However, this ' concern will be addressed in the following manner

n p -a.' Inf amas which are protected by area wide automatic fire s@pression systems, an exemption from the requirement to

~

=

E fireproof s@ ports for cable raceways containing the-.-

proposed cable as.well as any services whose collapse' could.

cause failure of. the proposed cable will be' requested. The

~

exemption will'take credit for the ability of the

~

1" aforementioned s @pression system.-ta; extinguish an'y 1 . postulated fire promptly, thus preventing temperatures-in Y the area from rising to the point where failure of any o- . s@ ports can occur. lThis should permit .use of the proposed cable in _ existing trays (if physically possible) and will

,_ not place restrictions on the routing of any new conduits

> intended to contain rerouted circuits.'

s b).. In areas which are not protected by area wide a'utomatic -

s @ pression systems, it is expected that any safe shutdown-

- ' circuits.in cable trays will'be rerouted to new. conduit.

Rockbestos cable will be run in the conduit,.or standard cables will be run and the conduit protected by. one' hour fire barriers consistent with the criteria outlined in the '

1 ,

• - , . cover letter for this attachment. New conduit will be;

[ routed as close as possible to the ceiling or wall in order

'co eliminate the potential for any collapsing' dabris to-affect'the rerouted conduit. In addition, GPUN is l  : currently developing design details for conduit s@ ports which, when analyzed, will be capable of retaining.their

.  : structural integrity withcut the benefit of fireproofing.

+

x

".. -.]

D11s ; analysis will serve as .tne basis for an exemption from

- the requirement to fireproof supports for rerouted conduit containing' the proposed cable in areas not protected by area _ wide automatic suppression system.

GPUN does not expect any exemptions will oe required for utilizing the proposed cable in the Reactor Building. A

-request will be made to qualify U1e proposed cable as L

^

equ.tvalent ;to radiant energy heat shields for use inside containment. Radiant energy heat shields are not one hour fire barriers, and as such, the cable raceways to which Uley are applied need not be additionally protected by fireproofing' their supports. Therefore, no exemption ,

requests will be required for unprotected steel supports inside containment.

Comment #2: The test program will.not establish tnat the cable is " free of fire damage" per Appendix R. In fact, significant physical

. transformation / damage is likely to occur as a result of a fire. ,

Consequently, fire rate cables will not literally comply with Sec2 ion 1III.G.l.a of Appendix R.

Also, since the cable is not enclosed in a' physical carrier which limits temperature rise on the non-fire-exposed-side to

- not more than 2500 F above ambient and otherwise protects the cable from the' harmful effects of a fire, tne tested cable will not' literally comply with Section III.G.2.c of: Appendix R, which '

pertains 1t o-fire _ barriers'.

w<

Response: LThere was no intent.to estaolish the fact that _the proposed cable was _" free from fire damage". GPUN: agrees tnat _ the fire rated cables do not -literally comply'with the requirements of Appendix R. The results of Febrimry 21, 1984, test demonstrated that' the intended functional capability of the caoles during and after fire exposure were met, thus representing an equivalent level of fire protection, and hence should ce acceptaole under

.the exemption process..

Comment #3:. .In the proposed application, the cable would oe enveloped in flame, it is_ not clear .that- the first hour of. the ASTM E-119

- time vs temperature. curve is conservative for- this cor.dition.

Response

GPUN believes-- that the qualification of fire rated cable must be against an acceptable industry standard. The test simulated

. fire conditions by utilizing the time vs. temperature curve as outlined in U1e ASTM E-119 test which is the _same. exposure to which caDie raceway envelope systems have oeen exposed. The NRC

~

recognized the ASTM E-119 Time vs Temperature curve as the basis for qualifying rated fire barriers; therefore, the test caole need=not be exposed to more stringent fire exposure over a one L hour time frame.

l-s p_ _

[It should be noted,1however,'that-during the actual test otner

^,

caoles in the tray were ignited thus' causing a more severe fire condition than' the ASTM E-119 time vs temperature curve. -The Lperformance of the Rockbestos caole was~ satisfactory.

] , ,

Comment #4

. .During the' tests the conductors are energized at 100 V.AC;.

however,-the cable will be used at higher voltage,s'. Therefore, fthe testidoes:not simulate the voltage stress, particularly tne

• voltsge. transients. of the actual: installation due to switching

and motor starting.

_ Response: Although the' full scale test on February 21,-1984, . was not'run Eat rated... voltage due -to safety considerations, two (2) tests run

. before and after the fulliscale test in ASTM ovens demonstrated

' the~ test caole's ability to operate' at well aoove. rated voltage

=at_the peak. temperature. . The results of tnese ancillary tests tu, iare. contained in test report prepared by VTEC. Laboratories and are also covered on page 32 of the U.L. test report.

Comment #5':: ,The performance characteristics of the cable that are necessary i for successful operation are not specified. The cable ratings.

are not specified.

Response:l  : Additional cable data is included in the test report. Full 5

~~ '

cable. specifications are being developed by Rockoestos and will l

.be available to perform the necessary work.

i Commenti#6: ." Wet short" post-fire conditions are not simulated; however, in X ~ 'the installation the~ damaged cables may be immersed in water for-significant periods-of time.-

E^ 1 Response:. 1The initial hose stream test'which was required to last for

~

1-1/2 minutes continued on and off for approximately 45 minutes because of the violence and persistence of tne' fire in the cable

~

J - tray caused by the fuel-loading caoles. -Seventeen days after the-test;(400 hours0.00463 days <br />0.111 hours <br />6.613757e-4 weeks <br />1.522e-4 months <br />),' the cables were subjected to another hose

~

. e ._

stream application (note that caole in conduit-is not impacted C ,

ibyJtne hose stream; this concern applies only to exposed caole x 'in tray).. The caules were then subjected to an AC hi-post test,

~

l

'during which they all supported 1000 volts:AC with a combined

' charging /leakaga- current of less . tnan 1 milliampere.

• The thermal expansion' forces.under real fire conditions are not

~

1 Comment #7::

. . simulated.

~

MResponse:- ~

Thermal expansion forces under' fire conditions will De addressed E as follows; y,

M'.

• P 1

e Is=

M ,

j

1

^

~

a .' .Where the-proposed cable is routed in cable tray or conduit in-locations protected by area wide fire s@pression

. l systems, thermal expansion.will be addressed in exemption l

requests, but considered negligiole since credit is taken I

. . for automatic's@pression preventing a significant temperature' rise as discussed.previously..

-b.. Where the proposed cable is routed in conduit in' locations

'not protected by area wide automatic s@pression (will not

> ;be run in cable trays in these areas), the design of f s@ ports not requiring fireproofing and the spacing of

=~v ' these _s@ ports as well as conduit runs, will account.for ,

thermal expansion forces. .This will also serve as the

~

. basis for an' exemption request from the-requirement to I L~ fireproof s @ ports in these areas. 1 l: .

L Comment #8: ~'The' post; fire mecha'nical forces due to firefighting and recovery

~

E . operations.are not simulated.  ;

. s,. ; - .

m .

.. Response: Laboratory testing cannot accurately emulate post fire mechanical forces. However,the _ test did demonstrate that the cable can withstand considerable physical abuse since some of the caDle trays did fail. In addition .the' caoles were subject -

n, to' repeated hose stream forces, which were required. . ProceduresE

--' will be developed for post-fire recovery based on the actual m - applications of the cable.-

~

l Comment #9: , JNo(posttestassessmentofthe'cablesoperabilityisincluded.

Response:' LThe-cable was tested for 94 hours0.00109 days <br />0.0261 hours <br />1.554233e-4 weeks <br />3.5767e-5 months <br /> during this mock up,~though

~ ~

- Appendix R allows repairs of cold shutdown equipment within 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br />. The guidelines for.the test'as stated in this. letter.-

will assure operability of equipment until .the necessary repairs .

s

of cold shutdown' equipment can be performed. GPUN plans.to v._

. replace the damaged circuitry prior to returning the plant.to:

~

~

service should a-fire occur.

L b .. ,

y hd , ['

. m 4

N ,

~. , ,

{_.