ML20197B483

From kanterella
Jump to navigation Jump to search
Responds to Questions Raised at 780830 Meeting W/Nrc Re Qualification of Elec Penetration Assemblies for Subj Facil. Addresses Justification of Qual by Similarity,Verification of Repaired Assemblies & GE Knowledge of Repair
ML20197B483
Person / Time
Site: Diablo Canyon  Pacific Gas & Electric icon.png
Issue date: 10/16/1978
From: Cleveland J
GENERAL ELECTRIC CO.
To: Allison D
Office of Nuclear Reactor Regulation
Shared Package
ML20197B467 List:
References
NUDOCS 7811080008
Download: ML20197B483 (14)
v • d • e


Text

_

D A 2/hs w ECUGH cupr O oENEa AL @ ELEcTaic sucte.e ~ ov PROJECTS DIVISION GENERAL ELECTRIC COMPANY,17s CURTNER AVE., SAN JOSE, CALIFCRNIA 9s12s MC 682, Ext. 52996 October 16, 1978 U. S. Nuclear Regulatory Cocnission Division of Project Management Office of Nuclear Reactor Regulation Washington, D.C. 20555 /*

Attention: Mr. D. P. Allison, Project Manager Light Water Reactors Branch No. 1 Gentlemen:

This letter is in response to questions raised during a meeting August 30, 1978 between the General Electric Company and the NRC's Light Water Reactor Branch No.1, to discuss the qualification of electrical penetration assemblies supplied for the Diablo Canyon Project. Sub-sequent to the aforementioned meeting, additional infomation was requested by Division of Operating Reactors regarding penetration assembly repair and the acceptability of repaired assemblies.

The following sumarizes the NRC interests addressed herein:

1. Justification of qualification by similarity of the Diablo Canyon penetration assemblies,  ;

i

2. Verification that repaired assemblies are qualified for their intended application, and
3. General Electric kncwledge of penetration assemblies which  ;

have been repaired.

RESPONSE TO OUALIFICATION BY SIMILARITY Attachment 1 to this letter provides the rationale for qualification by  :

i similarity. Diablo Canyon utilizes the canister design assembly, currently in use in a numcer of operating reactors, e.g., Millstone 1, Oyster Creek, Monticelle, Dresden 2 & 3, Quad Cities 1 & 2, etc. The canister design is environmentally qualified due to the similarities in its design to the 100 series electrical penetration assemblies which have successfully completed several simulated LOCA environmental cual-ification tests. Attachment 1 concludes that the results of the 100 7811080o08

' ~

GENER AL Q ELECTRIC Mr. D. P. Allison U.S. NRC Page 2 October 16, 1978 series environmental qualification are directly applicable to the canister assembly due to distinct design similarities, i.e., epoxy, steel and conductors. Also, relative to environmental considerations the canister assembly is a more conservative design than the 100 series.

(The 100 series LOCA environment qualification test report was submitted to the Power Systems Branch during the August 30, 1978 mee ting. ) ,,

EN'/TRONMENTAL OUALIFICATION -OF ELECTRICAL PENETRATION ASSEMBLIES EXHIBITING GAS LEAKAGE Attachments 2 and 3 are reports of LOCA and post-LOCA environment

ualification tests, performed on two cenetration assemblies received fr:m Millstone.2. Both assemblies were found to be deficient, i.e. ,

exhibited gas leakage in excess of specification. The penetrations sere returned to General Electric C mpany and subsequently subjected to the year-long environmental qualification test described by Attachments 2 and 3.1 The test demonstrates that at<<emperatures(well below tt.nh

^

stulated to occur during a LCCA)(p6st-LOCAbscm a ci), the stresses which created the fissures were alleviated such that a leak tight boundary (per the requirements of IEEE 317-1972) was formed.

.. .w.

penetration assemblies exhibiting excessive gas leakage.g.the field are field repaired. The re: air consists of pulling, by O~'a'enemically compati;le resin into tne fissure. The repaired assemoly is then gas-leak tested pursuant to IEEE-317. Periodic surveillance of the seals verify the acceptability of the repair.

A::!TIO'#.L INFC M T!CN

  • n no case has General Electric repaired a penetration assembly where b:th redundant seals exhibitad leakage and wnere the fissure was so large that repair was im;cssible. Only General Electric Ccmpany autnor-i:ed ersennel have made tnese repairs. No cenetration assembly ex-nibiting gas leaks prior to final release to a customer has ever :een shipped.

Ecth canistar and 100 series assemblies e .phy an epoxy system u the electrical istlat:r and part f tne :ressure barrier. In both cas,ges tnere is : circum #erence of steel with epcxy embeded therein. E:0, exnioits ona-::-roo er : ore orders of magni:uce grer.er tharial c .nt - -

l tnan steel. Any fissure in the epoxy is -he result of shrinkace :.

l mechanical fcact.'e.

i i Evidenza - ' :. -  ;* - 5 el ec ; : ' - -

4an assemolir -

were cefici .:4 -

& as.ir; is - . Or . W tre test re:: .

'; rthee .' .. M 2 al E% ir; 'an-i / sear:-'-

:ts: y .

r.

..c,-a. .' --

au"de- a

  • 1t

. ,1 .. - '. - I' . t' i~ s

. A AKdL

4

'G E N E R A L $ E LECTRIC Mr. D. P.- Allison U.S. NRC Page 3 j October 16, 1973 ,

During any postulated incident where containment temperature rises, the epoxy will necessarily expand and soften, causing the small fissures to i seal. The aforementioned repair procedure ensures that at any tem- I perature elevation a gas tight seal will be provided.

General Electric's recommendation for periodic monitoring of the seals assures that annoyance leakage will be remedied, and that any remote postulated event will not violate the containment barrier through the assurance that two reliable seals are in place; M Z.- p 411 qualifica- l tions have demonstrated that only one seal is necessary to maintain containment integrity.

I Each epoxy system currently utilized in General Electric's electrical penetration assemblies has demonstrated sufficient margin to maintainThe l pressure integrity (tested at 102 psig saturated steam conditions).

Millstone 2 qualification test demonstrates the relative mechanical properties of epoxy and steel. Therefore, the Diablo Canyon 1 penetration assemblies will maintain containment integrity in the unlikely event of a LOCA.

Very truly yours, James W. Cleveland, Sr. Engineer BWR Product Standards JWC:bp/1048-50 Attachments cc: Mr. G. C. Lainas, Chief Plant Systems Branch U. S. Nuclear Regulatory Commission Division of Operating Reactors Mr. R. A. Young Pacific Gas & Electric Co.

77 Beale Street, Rm. 1805 San Francisco, CA 94106 bec: G. wherwocd Barton A. Smitn Mr. L. Davidson W. R. Morgan Northeast Utilities Service Co. W. H. D'Ardenne P. C. Box 270 N. G. Luria Hartford, CT 06101 R. Schuster R. O. Brugge Mr. J. F. Kilty Operating Plant Services bp/1048-50 General Electric Co.

ATTACHMENT 1 COMPARISON.0F CANNISTER AND 100 SERIES PENETRATION DESIGNS

- v l

The 100 Series Electrical penetration design can be used as the I

basis for qualification of the cannister design because of the

~

direct correlations between the two penetration designs. The ,,*

100 Series modular design was developed by reducing the 12 inch l cannister concept to a 2-1/2 inch module. The materials that would l

be exposed to the LOCA environment are the same in both configurations, These materials included steel, XR5126 epoxy, electrical conductors, f

)

l and petting beards.

l 3cth the cannister and the modular design used an epexy system as the electrical isolator and part of the pressure barrier. The cannister design included a steel header plate embedded in the epoxy l

and the conductors were routed through this plate bef:re being i

spliced together and held in place by the petting boards. The 100 Series modular design has only the epoxy to provide a barrier between the environ =ent and the pin type connection. The pin connections in the module are less than 1-3/4 inches from the environment as compared with a minimum of 3 inches to the splices in the cannister design.

l The spacing between cenductors in the 100 Series design is closer than in the cannister design. The conductor center-to-center ,

distance in the modular design is .140 inches as ce= pared to

.306 inches in the cannister design for the sa=e sized conductors.

The distortion of the splices during crimping may result in a ,

i reduced distance between conductcrs. However, the reduced distance is still further apart than the spacing between conductors in the 100 Series design.

The following should be noted in summarising the two penetration designs:

1) Overall design configuratiens are basically the sane
2) Materials subject to the LOCA environment are the same
3) Ecth designs were built to the same overall quality standards and design parameters Additional conservative features in the cannister design are:
1) the separation between ccnductors is greater than in the modular design, 2) the splice connections are a minimum of 3 inches frem the exposed surf aces as ccmpared to 1-3/4 inches on the =cdular design, and 3) the cannister has the additional protection of a steel header plate embedded in the epoxy.

Based on the design similarities and the more conservative features in the cannister design, the qualification testing perfemred en the 100 Series medules demonstrates the acceptability of the cannister design.

- - - - - - - - - - _ - - ~ _ _ - - _ , . _ _ _ _ _ . _ _ _ _ _ _ _ _ _ _ _ _ _ . _ _ _ _ . _ _ _ _ -_

CANNISTER AND MODULE CONFIGURATIONS

. CANNISTER DESIGN Y Epoxy Steel Header Plate Petting Board Splice

~; ; .. -

l l

d N..!

a .. ' .

. N

{ ;.

. I

, c _.

I j

[ .

__ s I

ll U

c. *I

'j mw I

- 3 in6hes' minimum SERIES 100 MODULAR DESIGN Epoxy Steel Sleeve Pin Connection -- s Potting Soard _. y s

. su

.s. .3 c..-> -- -

, _ _. _ _ _J 1 --__ .

n- y a .,

Q........' . .. .. ;.. .i.li

- r 1/8 inch

1-3/4 inches '

w

- - - - - - - _ _ - _ - _ _ _ _ ~~

941-51610-5 ATTACHMdiT 2 ,

GE -

fl E D A LQE LE CT BiC **""'""' .. ._ 2 '" "* 1 fift.CFROTOTYPE TEST REFORM

[ LOW VOLTAGL c.Li.cTRIC FEMETRATIONS SIMULTN; ECUS LOCA, CURPINT & VOLTAGE TEST 941-51610-3 FIRST H A0g g(REFF.F15CE BEliTEL SPECI.FICATEO. REvis:c .

]

sa ao.

5., e. s-L 2 1 -

bl O*2.10 1.0 _ INTRODUCTION This report is applicable to the low voltage elictric penetrations i which we required to cartf a specified current / voltage during the LOCA serv ce environ =ent.

  1. 4/0 Tuo generic design penetration modules were selected for this test:The 4/0 module was co (4 conductor) and #16 (140 conductor) .the #16 module was only irradiated prior to the aged test.

prior to the test;In addition, only icu voltage pcuer For the wire#16sizes AWG (#8 thru 4/0) have l cted by specified current values tested in the specification.:odule, assumed oe .

General Electric Co:pany. '

2.0 REFESINCES

[

" Eechtel Specification 7604-E-34 para. 12.3.2b7 .

Ceneral Electric Cocpany, Table I, Data Sheet I 3.0 TEST REOUIR N *S the 4/0 conductors ,

! Simultaneeusly with the below listed LOCA environ =ent, shall carr/ 148 a:ps G460 Vac.

289 260 220 w Temperature 'F 40 20 Pressure psig 54 100 100 100 RH 15 min. 45 min. 23 hours2.662037e-4 days <br />0.00639 hours <br />3.80291e-5 weeks <br />8.7515e-6 months <br /> Dura tion The assumed electrical loading for the #16AWG todule is 3 a=ps G120 vac.

4.0 ~EST MARO'JA?2 Dwe. No.

Serial No.

Name T7257-312 4/0 157C4837G001 157C48360003 Y7257-259

  1. 16 5.0 TEST SIT U?

The two codules were installed The autoclave in an actual header is partially plate filled with which wate r is counted to self centained autoclave. A CalredIM '

containing H 30 /NaOH to =ake the pH of the solution 10.2.is used to bo 3 3

~

heater elecent In additien, nitrogen gas is added to the vessel -

in the autoclave. Figures 1 and 2 define -

( 10 psig) to achieve the specified environ =cnt.

up during the LOCA environ =ent.

the electrical test set

) . PRINT l 941-51610-5

. m ,, 7 ~ [.a,,3'.3,5,D , ,,, , , ,, ,,, , ,, ,'Z '

N C L'2ria u:ree ee , e s-ue 2 .. .. 1 f* [ hN San Jose. Ca.

i l

" " +

~ --- - - _ __ __ _

j --

i 941-S1610-5

} G E!! ER Al. @ ELE CTRIC

,.... .. .. , 0 .. . 2

=* - TITLE PROTOTYPE TEST REPORT

' SIML'LTANEOUS LOCA, CL*hREh7 & VOLTAGE TEST co., e= s-u, 3 s ao. 2 FIRST M ACC FOR REFERENCE BECHTEL SPECIFICATION 7604-E34 Revisto:

6.0 TEST 7.FSUL,T_S, S The tvo =edules were subjected to the envirce:ent and elec::ical leading defined in para. 3.0.?:ior to :his tes:ing, the =edules received the following aging.

66 ther=al cycles , 50-150-50*F,

  1. 4/0 3 to 4 cycles per day @l00% RH ,'

5x107 Rad, ga=na exposure

  1. 16 No ther=al aging ,

3x107 Rad, gan.a exposure A.f ter co=ple:1on of all testing, both modules were helium leak tight to less than 1x10-6 cc/sec 020 psig. An exact value is not provided since

- the recorders on the helius mass spectro =e:er showed no sign of change

~ ~ '

from the calibratica of 1x10-9 cc/sec. .

The ef fec:s of irradiation on :he epoxy inche =odule is to increage the electrical insulation resistance. As the specified value of 1x10 Rad is no: at:ained until :he end of Fost LOCA, a tes t en the =odules which were

> previously 1::adiated to this value vould no: cons ti:ute realistic condi-tions. The value selected (3x107 R) increases the hardness of the epop bu: has negligible af fec: on electrical properties. In addi:1on, 5x10/ R is the exposure tha: the seccndary seal experiences due :o protec:fon fro: the headerplate. This :es t, therefore, as cenducted, af fords the

=edule with :he grea:es chance of electrical breakdown during LCCA servic a enviren ents.

l.

J

/

PRINTS *

** m os W

N. C. Luria . ... B'...------.......

ASD m g,lN i. . 2

/ -

Sen Jose. Ca.

1 l C1 uu,, - ee.,

l

-u, 7 .

c...n.-

cent en sheet 4 sh #3 PROTO's..E TEST REPORT k-516io-5 LOW VOLTAGE ELECTRIC PENETRATIONS SIMULTANEOUS LOCA, CLARENT & VOLTACE TEST REFERE:;CE BECHTEL SPECIFICATION 7604-E34 13002715 l

en w.

60

  • LIN(

+

o o V ,

g...

I ,

__ _ _ _J v

Meeutt SERI($ -

CIRC 1JIT e

AUTCCLAVE 0

0

'^ ~---- - - - - _ - _ _ _ _ _ _ _ _ _ _ _ _ _ - - - - - - - - - - - - - - ..__ ___ ____ _ _ _ _ _ _ _

- - ~ - - - - - - - - - - - - - - - - _ - _ _ _ _ _ _

I IEST REPORT nc on sheet T sh #4 PROTOTY1 q,-51510-5 LO VOLTAGE ELECTRIC PE iETRATICNS SDfULTM:EOUS LOCA, CURRENT & VOLTACE TEST '

REFERENCE BECllTEL SPECIFICATION 7604-E34 ILi '471G 460 V AC 40 ~

~ +

WILO MACHINE DEVE LOPS CURRENT.

M I __

I I L _

I ISCLATED CONCUCTOR 1

SERits  !

CIRCutT .

AUTCCLAVE

' ~ - - - - - - _ _ ~ _ - - _ _ _ _ _ _ _ _ _ __ _ __

ATTACHMENT 3 GENER AL @ El.ECTnlC 941-51610-6 um 2 . 1

' n.r 7,1gg PROTOTYPE TEST REPORT HIL). STONE 2 ELECTRIC PENETRATION 941-51610-6 POSI-LOCA PROTOTYPE TEST - FINAL REPORT '

l 'ca v e. s-ic, 2 s o. I riast Mact rCR REFERENC{ SECHIEL $p{CIF{CATIC,l 7604-E34 REYl51CN PROTOTYPE TEST REPORT l

Mll.LSTONE 2 ELECTRIC PENETRATION POST-LOCA PROTOTYPE TEST FINAL REPORT Prepa..d by: H. G. Luria Test Conductor: B. J. Nalezinski k

PRtHTS T~

. .a . .

Wynnng.h# ,,,_ _ _ . L'25D _ . **= I 941-51610-6 af e !mt a

//(/

[.M /d / . , .

L OCaf rom C981 Om Saet f7 g., og

- - - - - _ _ ', , .----_--_-___[______ _ _ _ _ _

GEN ER AL @ ELECTillC 941-516b6 2

cut o met 3 m."

' 7,rtg PROTOTYPE TEST REPORT MILLSTONE 2 ELECTRIC PENETRATION 941-51610-6 POST-LOCA PROTOTYPE TEST - FINAL REPORT em c. : m 3 s . ... 2 rinsi s4ct reR REFEDENCE BECHTEL SPECIFICATION 7604-E34 REVlstCN:

1.0 SU.YMARY For the peried frem 3 March 1974 thru 10 March 1975, including a one week shugdown,theicwvoltagepowerandcontrolmodulesweresubjectedto 165 F, 5 psig 100% RH environment. Upon completien of the test, both units exhibited a leak rate of less than 1 x 10' cc He/sec at 1 atm. 0 )

pressure. The high voltage, over the same period was subjected to 220 F, 20 psig6 100'.' RH environment and was also helium leak tignt to less than ,.

1 x 10 cc/sec 01 atm pressure, 2.0 Reference

1) Bechtel Specificatien f7604 E 34
2) GE Test Report - 1. w Voltage Pcwer 1941-51610-5
3) GE Test Report - High Voltage (74-502-1 l 3.0 Test Recuirement '

i Bechtel Specification #7604-E-34 para 5.5(2).

" Items B, F, G (low voltage control and Ocwer)* shall remain electrically operable under the conditiens statec belcw ..."

54 40 20 5 pressure, psig 260 220 160 temperature, F 289 duration 23 hrsj 1 year d5 min 45' min Y see references 2 & 3 that the assembly shall remain helium leak tight to less (It is assume!:

than 1 x 10'* cc/sec af ter completien of the test.)*

.

  • Brackets are GE clarification.

4.0 Hardware Identification Thefo11cwinghardwarewassubjectedtotherequired(orgreater) environment. l Cwe No. S/N Name 174B9302GC01 6574036 High Voltage 157C4837GC01 Y7257-312 f4/0 Module Asm 157C4836G003 Y7257-259

  1. 15 AWG Module Asm PRINTS T;

_ a ,, 941-$1610-6

't c 1"ota /(f#.3- .. .. ._ .-.. _I -. _ D31D _ _ ___%'4

.... 2

/W, san Jost. Cs, toe.ne. em ..m 3

...t.....

4

l/

i AO GENER AL @ ELECTRIC N PROTOTYPE TEST M. PORT 941-S1610-6 MILLSTONE 2 ELECTRIC PE4ETRATION 941-S1610-6 POST-LOCA PROTOTYPE TEST - FINAL REPORT een .. s.m 4 s - . .. 3 rmsi mot rom REFERENCE BECHTEL SPECIFICATION 7604-E34 REYlslo N.

5.0 Test Setuo The above hardware was installed in the follcwing autoclaves, in a vertical position to simulate a worst case cendition.

Penetratien Name Autociave s- Low Voltage Power S/N 330 Low Voltage Control S/N 330 i

High Voltage S/N 5641 Tests were conducted at GE, San Iose, Ca. facility, 31dg. W l

6.0 Test Results The tests for low Voltage were conducted frem 3/3/74 to 3/10/75 with a one week shutdown for a periodic test ecuipment maintenance. The tests for High Voltage were conducted from 3/3/74 to 3/3/75.

The test units were subjected continuously to the following envircnment:

Lcw Voltage High Power & Centrol Voltace temperature, OF 165 -

220 pressure, psig 5 20 duration 1 year 1 year E100% 4100%

% RH During months 6, 7, 10 & 12, the icw voltage units were brought to ambient conditions and the leak rates were measured at less than 1 x 10~g cc He/sec @ atm pressure. After 1 year, the high voltage 4 was measured to have a leak rate less than 1 x 10~ cc He/see @ amb conditiens.

+

  • PRINTS it a as e, weswa WP4 941-51610-6
  • - ' N;. - - - - ~r- , Gn ?.,Gn,...

% -m *: :

m a. -> .y .. . ,

n . , - . . 3

- - - - - - - _ _ _ ~ _ - _ _ _ _ - _ _ _ _ _

GENERAL h ELECTRIC 941.s161o-6 com on saart F i= =e. 4 "g

.n un,g .

PROTOTYPE TEST REPORT 941-51610-6 MILLSTCNE 2 ELECTRIC PE!:ETRATION POST-LOCA PROTOTYPE TEST - FINAL REPORT c2wf en sair, F .. A riRsT Mact ren oerece..cr n ecutes ceertetesttrus vena.n a REVi$loNs 6.0 (cont.) -  :

The following electrical measurements were recorded en the test units:

Date of Insulation Continuity Test Resistance

  • Hioot

.. 1 Low Voltage Power 50 VDC n No Change

(#4/0) 3/12/75 9 (S/NY7257-312) .5 x 100 0 100 VAC l j 1/15/75 1 x 107 a 100 VAC 2.8 x 105 n 10/2/74 700 VAC 3 x 105 a 9/3/74 2.3XVAC i

" Low Voltage Control

(#16) 3/12/75 .5 x 106 0 100 VAC No Change (S/NY7156-259)- 1/15/75 1 x 107, n 150 VAC 4 " "

10/2/74 5 x 10 a 450 VAC 5 " "

9/3/74 1 x 10 a 800 VAC 4

3 0

PRINTS T.

~" O , . ., . i I/y7/2E/;f (_ _ _ _ _ _ E"35_0_. _ _ _ _ _*;a', 941-51610-6 F

s,w JOSE, CA. . . .. . e. - .... 4

,'""" ' / MJ

. .m A g g

- _-__ - _____ _ _ - _ - _ - _ _

Retrieved from "https://kanterella.com/w/index.php?title=ML20197B483&oldid=3013005"