ML20055F937

Submits Info Re Revised Acceptance Criteria for THERMO-LAG Fire Barrier,Per 900705 Telcon.On Exposure to Heat Flux at Surface of Barrier,Listed Mechanisms Activated.Fire Testing Demonstrates That Panels Qualified W/Variation in Thickness

ML20055F937
Person / Time
Site:	Comanche Peak
Issue date:	07/13/1990
From:	William Cahill TEXAS UTILITIES ELECTRIC CO. (TU ELECTRIC)
To:	NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM)
References
TXX-90255, NUDOCS 9007190390
Download: ML20055F937 (3)

Text

- ,.

I7 Log # TXX 90255 File # 909.5 Ref. # 10CFR50.48 filELECTRIC July 13, 1990 U. S. Nuclear Regulatory Commission Attn: Document Control Desk Washington, D.C. 20555

SUBJECT:

COMANCHE PEAK STEAM ELECTRIC STATION (CPSES)

DOCKET NO. 50-445 THERMO-LAG PREFABRICATED PANELS Gentlemen:

On Thursday, July 5, 1990, in a teleconference with NRC Region IV personnel and the Site Resident Inspector, TV Electric committed to provide the basis for implementation of THERM 0-LAG fire barrier material acceptance criteria as revised in October of 1989. The criteria were revised and implemented after

minor deviations from our originei material acceptance criteria were l- identified. The following information regarding the revised acceptance criteria is submitted.

( To understand the relative sensitivity of THERM 0 LAG to configuration

! variations, a review of the behavior of THERMO LAG under fire conditions is usei'ul . THERMO-LAG is a passive barrier system until it is exposed to the heat flux of a fire. On exposure to the heat flux at the surface of the '

barrier, the following mechanisms are activated:

- An inorganic salt is contained within an organic binder that contains glass fiber reinforcement. This salt undergoes sublimation, which occurs at a constant temperature, absorbing heat and leaving behind a char layer through which sublimed gases must transpre.

\

l

- The sublimed gases ancounter a sufficient residence time in the char laye' <

to undergo endothermic decomposition and disassociation before injection into the surface boundary layer.

- The mass transfer of the sublimed gases carries heat to the surface h boundary layer.

- The charred surface re-radiates heat energy away from the system.

Ob ggr188tn taQr ,

P.O. Bot 1002 Olen Rose.Teus 7600-10M

y

/

• p>. -t TXX-90255 Page 2 of.4 THERMO-LAG does not work solely as an insulating material where thickness

governs material effectiveness. THERMO-LAG behavior under fire conditions is

, dependent on the mass (density) of the THERMO LAG and minor configuration

.(thickness) differences in the THERMO LAG are not significant. Material that is compressed from normal fabrication thicknesses retains the mass of

• sublimation salts equivalent to normal fabricated THERMO-LAG.

Thermal Science, Inc.'s (TSI) Quality Assurance / Quality. Control program requires that each THERMO-LAG prefabricated panel be subjected to detailed QC

> thickness measurements to verify minimum 1/2 inch material. thickness prior to

• accepi' ance. TSI's QC procedures require that the entire surface be visually  ;

scanned and a minimum of 18 preselected locations on each panel be measured-to verify required panel thickness. Measurements are made with devices which are l tested to verify their accuracy. These tests are performed at a prescribed i frequency. Only panels which meet the criteria of the TSI QC procedure are '

shipped. This sanel fabrication and inspection methodology has remained essentially uncianged since TSI began production of prefabricated panels in ,

the early 1980s.

TSI randomly selected THERMO-LAG prefabricated. panels for use in fire tests >

conducted to qualify the product (including fabrication and inspection i processes)'from panels accepted by QC in accordance with TSI's QA program.

Fire tests involving 200 or more randomly selected panels manufactured and L inspected under identical requirements have resulted in no failures of the -

l THERM 0-LAG Fire Barrier System.

l .

L The' fabrication and inspection methodologies employed by TSI assure a qualified product; however, minor variations in panel . thickness not associated with the specific preselected inspection points may occur. Such minor variations would have occurred in any of the panels subjected to the fire testing performed by TSI. 'Thus, the fire testing demonstrates that the panels ,

are qualified even'with minor variations in the panel thickness. .

. Subsequent to TV Electric's initial supplier qualification audit of TSl's QA Program in 1981, a number of additional audits and surveillances have been conducted to verify TSI's continued compliance.with procurement documents, u During late 1989, TV Electric maintained source (shop) inspection at TSI's L facilities over extended pereds to monitor the performance of TSI's QC inspection efforts and performed independent inspections of completed panels.

These audits and surveillances have established that TSI has satisfactorily

, implemented the approved QA program requirements. .

t

-,-v -- - , _ _ - - - - - .

-r - - - - ------- . , --- - , , , - , . . - -- ,,.,,n .- -

1 c

n. .d d

TXX-90255 Page 3 of 4 The QC inspections by TSI provide the basis for determining that THERMO-LAG j shipped to CPSES complies with the TSI and TV Electric acceptance criteria, q including acceptance criteria on thickness. In addition to the QC inspections performed by TSI, CPSES performs receipt inspections.

Minor deviations from TU Electric's QC receipt inspection criteria contained f in Specification 2323-MS-38H, Appendix A, regarding minimum thickness of ^

1 THERMO-LAG prefabricated panels were identified during receipt of material at l CPSES in October 1989. These devictions were evaluated by TV Electric and. )

TSI. ,In many :..M- 7 d incalized thickness reductions, areas of the panels q

'were apparently compressed di: ring handling and or shipment of the material ~  :

In other instances: localized thickness reduction', apparently resulted from> '

the fabrication process.

To' address localized areas of'th'ickness . reduction of panels supplied to l TV Electric in October 1989, a tolerance for the 1/2 inch minimum panel l I thickness was developed. TSI 3rovided a quantified tolerance that allowed j deviations up to minus 1/8 inc1 from the 1/2 inch minimum for no more than 2%

l-of the entire surface' area of the panel. This tolerance is similar to Underwriters' Laboratories tolerance for sprayed on fire barriers. The basis ,,

for reduced panel thickness tolerance included : H

1) Localized areas of compressed THERMO-LAG material did not represent an -!

actual reduction in the amount of subliming material available for fire ,

response; therefore, initial fire test results were not compromised. (

2) Minor localized areas of reduced panel thickness resulting from the  !

fabrication process were not unanticipated. The random selection of . '

THERM 0-LAG panels for fire testing and acceptable fire test.results' hhve 'y,-

established that minor thickness variations of this nature do'not have an:

adverse effect on the THERM 0 LAG Fire Barrier System and do not compromise- i

, initial fire test results.

Based on the above and TSI's responsibility for t rtification of the product  ;

as a one hour fire barrier, the tolerance as defined by TSI was included in i TU Electric's receipt inspection criteria. However, in attempting to implement this criteria it was determined that the area measurement (e.g., 2%

of surface area) was not practical at receipt inspection. Therefore,  ;'

TU Electric receipt inspection criteria based on panel weight were developed

.in conjunction with TSI's recommendation. In addition, TV Electric still u 7 requires-inspection for damage due to shipping and handling. This does not change the performance criteria required of TSI to provide a one hour fire  :

rated material to TU Electric, nor does it affect TSI's responsibility to ,

inspect and accept THERM 0-LAG for the requisite thickness prior to shipment to. .

CPSES.

h i

, +- - . . . , - - , , , - . , - v-,.--... -, ,, - - - , , . ,- . w, -,4.

=

g

]  !

. r. . - s , , ,

, 1

'r ,, ,

1, '

\L y t

~ '

i TXX-90255 '6 Page 4 of 4 i

(

. TU Electric's weight measurement in conjunction with the controls and  :

manufacturer's inspections implemented by TSI and the visual receipt

' inspections.to identify any shipping damage performed at receipt of material '

provide assurance that the THERMO-LAG prefabricated panels meet applicable 4- ' fire test _ qualification requirements.  :

Sincerely . 1 e ..

William J. Ca ill, Jr.

JDS/ DEN /daj c'- Mr. R. D. Martin, Region IV' [

ResidentInspectors,CPSES(3)

Mr.-J. H. Wilson, NR3 1

' i  ;

w f I b

6-

, i

-l a

1 i

\;

i:

i L

I; ,

• i

__ := -.a__ _ . . - . -. . - . .- . - --. . .