ML20153D171

Responds to Concerns That Environ Qualification (EQ) for Bunker Ramo Instrumentation Penetration Installed at Facility Not Demonstrated.Summary & Clarification of Past Submittals in Response to EQ Issue Also Provided

ML20153D171
Person / Time
Site:	Braidwood
Issue date:	04/29/1988
From:	Hunsader S COMMONWEALTH EDISON CO.
To:	Murley T NRC OFFICE OF ADMINISTRATION & RESOURCES MANAGEMENT (ARM)
References
4545K, NUDOCS 8805090076
Download: ML20153D171 (12)

Text

-

3

, . -

8 s - > .;.

I"

, gg 7

@; J ie* - . i/ One Frat Nabonel Plaza. CNoego, lainois V Address Reply to: Post Omco Box 767 CNcago,luinois 60600 0767 i

• April 29, 1988 I;

, f 9 ,

{

I

^

l, .I Mr. T. E. Murley Office of Nuclear Reactor Regulation  ;

U.S. Nuclear Regulatory Commission

• Washington, DC. 20555 [

Attn: Document Control Desk ,.

i

Subject:

Braidwood Unit 2 Environmental Qualification Bunker Ramo penetration NRC Docket No. 50-457 -

Reference (a): April 8, 1988 D.R. Muller letter to L.D. Butterfield.

(b): April 7, 1988 S.C. Hunsader letter to T.E. Murley (c): March 23, 1988 S.C. Hunsader letter to T.E. Murley L s F

Dear Mr. Murley:

}

l l Reference (a) provided the NRC staff's conclusion that the "

i

! environmental qualification for the Bunker Ramo Instrumentation penetration assembly installed in Braidwood Unit 2 had not been demonstrated. The NRC -

staff presented in reference (a), that during the Midland Containment penetration Environment Qualification (EQ) Test, nsulation Resistance (IR) readings had not been taken at frequences consistent with IEEE Standard  !

323-1974. Reference (a) also stated that although the NRC staff had not t received the thermal lag analysis, it was believed that such analysis would l not resolve this issue. In a teleconference on April 26, 1988 these  !

concerns were discussed with the NRC Staff. -

The purpose of this letter is to provide a response to these  !

concerns and to provide a general sununary and clarification of the past submittals made in response to the Bunker Ramo EQ issue. ,

Ceco is utilizing both 10CPR50.49 and NURBG-0588, Section 2 -

) requirements which allow the entire penetration assembly to be [

~

environmentally qualified by combining the type test data from individual .

components that comprise the assembly, f The utilization of these requirements is based on the fact that the i

, Braidwood Unit 2 Bunker-Ramo EQ Test.(123-2220) did not include any IR {

readings applicable to the installed penetration type, conductor size or i termination configuration. Since no IR measurements were taken for a [

penetration similar to those installed at Braidwood Unit 2, the 10CFR50.49 l l

i i l

8805090076 880429 I PDR ADOCK 05000457 P DCD ,

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

--_.--.--_..--...---_,,--.-.--l -

and NURBO-0580 requirements were applied. These requirements allow the summation of individual components to equal an assembly. Thus the Braidwood Unit 2 penetration assembly IR value was determined by summing the individual component IR values that comprise the penetration assembly.

(i.e., penetration Module + B1W pigtail + Raychem WCSF-N splice =

• penetration Assembly.) These individual component IR values were taken from several sources as defined herein.

t

.y Based on the penetration module construction and its location within the header plate (i.e. the module feed-through conductors between the seals are isolated from the LOCA envilonment) we concluded that there would be negligible IR degradation in the module. Therefore, the IR values for the BIW pigtail and the Raychem WCSP-N splices (available from other independent LOCA tests) were the basis for our acceptance of the Bunker-Ramo Test Report (for a more detailed discussion, please refer to Appendix A transmitted with our March 23, 1988 letter, NRC Docket No. 50-457). Even though it is believed that the above documentation adequately demonstrates environmental qualification, the NRC staff determined that additional test data were necessary to fully substantiate the acceptability of the Bunker-Ramo penetrations. We believe that the Midland supplementary II Qualification Test Report 123-2201, Rev. A, does the following:

(a) provides additional test data in that, it provides acceptable IR values for a complete assembly (reference Appendix B transmitted with our March 23, 1988 letter, NRC Docket No.

50-457) at the worst case temperature expected for Braidwood Unit 2; (b) supports our initial conclusion and use of the Bunker-Ramo Test Report, 123-2220, and; (c) if considered in conjunction with all of the documentation and analysis presented to the NRC staff, demonstrates environmental qualification in accordance with the L requirements of IEEE Standard 323-1974.

IEEE 323-1974 Section 6.3.1.4 states the following:

"The time interval between measurements shall be such as to obtain the time dependence of each variable."

IEEE 323-19'4 Section 6.54 states:

"The electric equipment type shall be considered to be qualified by demonstrating that the equipment performance will meet or exceed its specified values for the most severe environment or sequence of environments in the equipment specification during its qualified life."

i There were two IP measurements taken during the LOCA Portion of the l Midland Test rather than one indicated in the subject NRC letter. Both of l these IR Measurements were recorded at environmental conditions (i.e.

l temperature, pressure, humidity and chemical spray) that were equal to or more severe than those anticipated to be seen by the installed Braidwaod

Unit 2 penetrations (reference the Thermal Analysis transmitted with our April 7, 1988 letter, NRC Docket No. 50-457). Therefore, these IR measurements are "such as to obtain the time dependence" of IR during a Design Basis Event (DBE) and are in compliance with the above IEEE-323-1974 requirement. The two recorded IR values represent the minimum values anticipated based on the severity of the expected environment.

As previously stated, the Midland Test results were presented at the NRC's request for additional documentation demonstrating that the main source of the IR degradation identified in the Braidwood Test Report, was due to the terminal blocks and connectors utilized in that test rather than the penetration module. We believe that the acceptable IR readings from the Midland Test provide this documentation. Attachment I provides the following:

(a) shows how the Midland test results were utilized to pinpoint that the terminal blocks were the main source of leakage cur ent in the Braidwood test; (b) calculates the expected IR values for the installed penetrations based on the BIW cable and Raychem VCSF-H splice test reports. These independent test reports included monitoring of IR degradation at a frequency consistent with the intent of the above referenced IEEE-323-1974 requirement.

Thus, the qualification documents which have been presented to the NRC demonstrate collectively the adequacy of the Braidwood Unit 2 nstalled Bunker-Ramo Instrumentation penetration Assemblies for the Braidwooc DBE environment under the requirements of IEEE-323-1974, 10CFR50.49 and NUREG-0588:

Bunker-Ramo Test Report 123-1220: - Established penetration qualification for all requirements other than IR, during the LOCA:

BIW Pigtail and Raychem WCSP-N Splice Testsi Established the IR values applicable to the installed Braidwood Unit 2 penetrations:

Midland _JI Test: provided additional acceptable test data (i.e. IR Measurements) that are applicable to the installed penetrations and further established that the main source of leekage current was the terminal blocks; thereby supporting our conclusion and initial acceptance of the Braidwood Test Report.

Attachment II provides a general summary and clarification of the past submittals including a compilation of IR values from various test reports that support the qualification of the Bunker-Ramo penetration assembly.

,, In summary, it is our contention that the documentation submitted, supports the environmental qualification for the Braidwood Unit.2 Bunker-Ramo Instrtssentation Penetration Assembly and meets the requirements of IEEE-323-19*14, 10CFR50.49 and NUREG-0588.  ;

• Please address any questions concerning this matter to this '

office. We are willing to meet with you to discuss this further.

Very truly yours,

// C' w' .

S. C. Hunsader Nuclear Licensing Administrator

/k1) cc: S. Sand 0 (NRN)

W. Forney (RI11)

NRC Resident Inspector Braidwood i

4545K L

)

i I

4 l

. t Page 1

+

Attachment !

Engineeri g E3 1uation The intent of using the Midland test result is to support our engineering analysis, which was provided for the NRC audit concern number 34. The Midland test included a penetration module similar to those installed at Braidwood Unit 2 and corifirmed that the main source of leakage during a DBE is from the terminal blocks. From the IR readings reported in our response to the NRC. dated March 23. 1988, the above conclusion is confirmed and explained as follows:

The electrical equivalent networks to represent the leakage paths for the circuit 1 (epoxy seal) and circuit 2 (tenninal block) are as follows:

c, p _

q f

J J R>d g gR9FR g Rg gREP0XY gR$RRFR M $R C R TB D E # End Seal 2 8 3

~~

57cuit 1 Circuit 2 Note: The difference between circuit I and circuit 2 is the leaktge path at the termination point, Where R:

g Leakage resistance conductor to conductor (C-C) for the module.

Leakage resistance (C-C) for the RFR splice.

RRFR:

R: Leakage resistance (C-C) for the associated ;igtail conductors.

C Leakage resistance (C-C) for the epoxy end seal.

REP 0XY:

RTB: Leakage resistance (C-C) for the terminal block.

6 The IR leakage measured for circuit I at 16 hours1.851852e-4 days <br />0.00444 hours <br />2.645503e-5 weeks <br />6.088e-6 months <br /> (at > 200 F) is 1.5 x 10 a (C-C) 3 The IR leakage measured for the circuit 2 at 16 hours1.851852e-4 days <br />0.00444 hours <br />2.645503e-5 weeks <br />6.088e-6 months <br /> (at > 200"i) is 4.2 x 10 0 (C C) i (cemge , a x

. 1 . 1 . .. ;

. mR s wm a

m v

i l

x,

a. a, 1

Page 2 F ,

y, ,

from the above relation we derive the following: ,

- Rg .(Rleakage)1 R

RFR ( leakage}l

. 3C (Rieakage)1 (leakage)1 R '

EP0XY a r.d R // R RFR // R C * ( leakage)1 g

Therefore, Rg, R RFR, R , and REP 0XY ' '*

C and R M

// R RFR // R C #-

• Similarly for circuit 2 above:

I 1 "

1 TM

+

R 1

TC + 1~

(2 R(leakage)2 RFR f~B T

1 1

" +

T R TB Where 1 l 1 l_ < I

+

R _ g+R RFR C 1.5x10 6

From equation 2 above:

1 1 I 1 -

--- > 6 3 R 3 a.2x10 TB 4. 2x10 1.5x10 R =

'4'2*10 a TB i

l l

{

l l

l

Page 3 3

Since the IR of the 6terminal ' block is 4.2x10 0 and the IR of the e 14.2x10 = 357 o

' greater than 1.5x10 n, the leakage of the terminal block is 1.5x10gd sea times higher than the end seal. This confirms that the terminal block was pre-

> dominantly the main source of leakage current as stated earlier in our previous response.

4 In addition, the total leakage of circuits 1 and 2 represent the sum of leakages obtained from various components. Havino l

established the source of leakage current in the feed through module as the terminal block, the excessive leakage current exparienced in the Bunker Ramo l

Qualification Report # 123-2220 can be attributed exclusively to the tenninal l blocks and connectors used in the test. The above test results show that the module epoxy does not degrade during a DBE and the leakage from the module is negligible.

Based on the configuration of the installed Bunker Ramo penetration assemblies, the feed through modules along with the Raychem (WCSF) splices and BIW pigtail a3semblies comprise the entire leakage path of the assemblies. The leakage path of the entire assembly is based upon the sum of the leakages from the various' components addressed above. The total leakage from the assembly is shown by the following equivalent circuits:

Inside containment Outside containment _

4 h$Y Yh/

Installed configuration

' gh!

g / / [/[M:]

Penetration Raychern WCSF N pigtail Module feedthrough spIice conductor conductors i ub g d Cabletray

.7 Field cable Cn sc J '

'nsignificant

-1, h 2; E n , i r ymm u l Q :< ,.

'6ILO / ' ' G il ,i

, '+i ,} ggg, , , , , ,q g ,

..  ?

-A N

, . - . i~_.__ . L- _ - <

g, ..

v -

.1

, k A ' : lq ' ,

~

i , 'M e r

... 3 Page 4 splice x R pigtail R R penetration =

E Es m

[R splice + R pigtail) ' ' '

=

Where: R penetration = Leakage resistance (C-C) for -the . entire feed - ,

through assembly R i p gtail

= Leakage resistance (C-C) for the, pigtail ,

P.sp'e ics '

-Leakage resistance (C-C) for.the splice e g k 4 r

The pigtail in the Bunker Ramo penetratich 1ssembly is BIW and has a maximum length of.10' . The leakage from the ' pigtail is _ included in the field cable 10 . and choosing a . worst case IR .

by (0.545 increasingM n/1000'). the field cable length Therefore, eliminati by 'ng'the leakage from' the pigtail, the' leakage of the penetration.is'merely the leakage from -the splice given by 4

- equation 3.

• 4 ,

R penetration = R' splice '3 Based on the data obtained from several-'Raychem WCSF splice tests, thq worst-case Raychem WCSF-N splice for the Braidwood DBE conditions ig 1.2x10' O.

Hence, leakage' IR for the entire penetratic, module is 1.2x10' O. This 'eakage-

'IR fnr the penetration assembly has acceptable inaccuracy for the instrument loops at Braidwood Station.

e il r

L,.

t

= -

( .,

v .

ATTACHMENT II QUALIFICATION METHOD FOR BUNKER RAMO INSTRUMENTATION PENETRATIONS o 10CFR50.49 and NUREG-0588, Section 2 allow qualification of equipment based on (a) analysis in combination with partial test data that s'ipport the analytical assumptions and conclusions and (b) combination of +ype test ,

data from individual components that comprisa. an assembly. In this case the individual components are the WCSF-N splices, the Boston Insulated Wire (BIW) penetration pigtails, and the penetration module.

o Table V of Braidwood Test Report 123-2220 summarizes th'e Insulation Resistance (IR) values recorded from selected penetration circuits during the LOCA test. This table does not include any IR readings applicable to the installed penetration conductor type and size (No. 16AWG TPS), or termination configuration (WCSF-N splices). The low IR values in Table V are attributed to "shorts" at the circuit terminations used in the test (i.e., connectors and terminal blocks) rather than to the penetration module itself. Sandia National Laboratories NUREG/CR-3691, "An Assessment of Terminal Blocks in the Nuclear Power Industry," confirms that terminal

. blocks in instrumentation circuits "experience leakage currents and low IR levels when exposed to steam environments."

o IR measurements applicable to the installed penetrations were taken from other test reports that independently qualify the individual components comprising the penetration assembly; (i.e., the WCSF-N Raychem splices and the BIW penetration pigtails).

o It was determined that the penetration module would exhibit negligible IR degradation. This determination is based on (a) the module's construction (i.e.,.the module feed through conductors are insulated from one another and from the header plate via a glase reinforcad epoxy), (5) the module's location within the header plate (i.e., the module area between the seals is pressurized with dry nitrogen and is isolated from the LOCA), and (c) the statement in the Braidwood Test Report that "no significant deterioration occurred in the Amphenol module or seal material."

o The above qualification method is in compliance with the applicable NRC regulatory requirements, o Additional test data was obtained from Midland Test Report No. 123-2201.

This data (a) supports CECO's initial basis for establishing qualification, (b) demonstrates that the terminal blocks were the main path of leakage current and therefore the cause for the low IR values in Table V of the Braidwood test report, and (c) independently demonstrates two acceptable IR measurements for a Bunker Ramo instrumentation penetration assembly (No. 16AWG TPS) similar to that installed at Braidwood station.

I _

I . ..

o The two IR measurements in the Midland Test were taken 16 hours1.851852e-4 days <br />0.00444 hours <br />2.645503e-5 weeks <br />6.088e-6 months <br /> and 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br /> into the LOCA at environmental conditions (i.e., temperature, pressure, humidity, and chemical spray) that were equal to or more severe than those anticipated to be seen by the installed penetrations based on the thermai analysis performed. Therefore, these IR measurements are "such as to obtain the time dependence" of IR during the LOCA as req'lired by IEEE-323-1974. The two recorded IR values represent the minimum values anticipated based on the severity of the expected environmental parameters that affect IR.

o The following is a list of the qualification documents collectively used to demonstrate the adequacy of the installed penetrations:

1. Bunker Ramo Test Report 123-2220 Establishes penetration qualification for all requirements other than IR, during the LOCA.

2. BIW Test Report B-915 Establishes qualification (including the IR) for the BIW penetration pigtails during the LOCA.

3. Raychem Test Report 58442-1 and EDR-5040/5046 Establishes qualification (including the IR) for the Raychem WCSF-N splices during the LOCA.

4. Midland Supplementary II Test Report 123-2201 Provides additional acceptable test data (including acceptable IR measurements) that are applicable to the installed penetration and further establishes that the main pati, of leakage current was the terminal blocks; thereby supporting CECO's initial conclusions and

, acceptance of the Braidwood Test Report.

l l o The following is a list of additional Test Reports that include qualification test data which support the conclusions previously reached:

l 1. Viking Test Report - Wyle's NEQ_46880-1, 1

l Provides acceptable IR test results for a Viking penetration having I

some similarity to the Braidwood Unit 2 penetration construction and

! termination method.

2. Conax Test Report IPS-1077 Provides acceptable IR test results for a soft epoxy module with j terminations similar to those installed at Braidwood.

l l

l l _

- .n .

8

3. Calvert Cliff Test Report 123-1252-Provides acceptable IR t'est results for an Amphenol penetration and termination conf,iguration similar to that installed.at Braidwood, o Table I includes the various IR test data and associated environmental

- parameters from the'above test reports for comparison purposes.

w r

4

+

DPG:dlk 04/28/88 NBI/MBAB-EP603 1

4 l

l I

l-3-

& .,-,--,,---,wa

- , . , , , + . , , . , +-_,w-,-n,-an,,--- - - - . ,-, , - + , - , ---. ,- , , , - - , - , , , - -

, - z. : - --

_i *

- TABLET 4- ~, ,

i - -BRAIDWOOD STATION UNIT 2 ,

BUNKER-RAMO INSTRUMENTATION PENETRATION ASSEE LIES *

- f TEST C0eFARISONL . .. .

' BRAIDWOOO 2 BUNKER-RAMO Biw RAYCHEM= MIDLA@ 11 Vl%ING ~CALVERT TEST-REPORT TEST REPORT TEST REPORT TEST REPORT TEST REPORT CLIFF CONAX . , "A - .-

TEST PARAMETERS EQ REQUIREMENTS 58442-1 & 123-1201 WYLE'S NEO TEST REPORT TEST REPORT; AM) PENETRATION BASED ON THER*4AL 123-2220 B-915 ANALYSIS EDR-5040/5046 46880-1 123-1252 iPS-1077 CrdSTR'CT 80H J

346/3.2 hrs 340/6 hrs, 385/4 min' 300/190 sec 430/20 sec 276/10.5 min 226/52 min g g g Temp 224/15 min Steam /100f Steam /1005 Steam /100s Steam /1005 Steam / loof Steam /100% Steam /1005 Steam /100t.

S?eam/$ R.H.

Chemical Spray YES YES YES YES YES YES -YES Steam Test j

8 Onty Accident Pressure (PSIG) 50 113 110 66 -- 60 32 3.6' 30 days 30 days 197 hrs ' - -- 52 min

. Accidaat Duration 365 days 30 days- 161 days Epoxy Type Hard Glass Hard Glass N/A N/A Har'd Glass Polyurethane .Hard Glass Soft Reinforced Reinforced Reinforced Potting and Reinforced ~

Polysalfare Insulator N/A Raychem- Raychem RFM- Raycnw Rsychem RaycSem ,

Yermination Method Raychem Terminal Blocks WCSF-N & Connectors WCSF-N Epoxy End Seal WCSF-N M3F-N - WCSF 1

N/A N/A 12 8 2 Note 1 Note 2 4 No. of IR measurements N/A #16 AWG #12 AWG #16 AWG #16 AWG -- #16 AWG Peootration Condu-tor #16 AWG _

Size 7

Lowest IR Recorded (n)/ N/A N/A 5.5x10 / 1.2x10 / 1.5 x 106 / 1.26 x 105 / - -- 3 x 10 / .

500VDC 500VOC 50cvDC 50.47VDC 100v0C 4 voltage (Temperature) #

(340 F) (314 F) (>200" F) (375*F) (225 F) 1 IR at 250 F (N / 1x106 / N/A 1.7x10 / 8x10 / 1.5 x 106 / 2.5x1057 .,

3X10 /

voltage 400VC 500VDC 500v0C 500VDC 50.28vDC (300 F) 100VDC (225 tl s

Notes:

1. Load current was avmitored during the accident test.

2. Leakage corrent reported du.-!S .the accident test is less than imA.

3. N/A = Not Applicable 4 - - = Not Available in Test Report ..

- - - - , b , s r v- ,