ML20213D752

Rev 0 to S-C-A910-EEE-0087-R0, Environ Qualification Testing of Rockbestos Cable

ML20213D752
Person / Time
Site:	Salem
Issue date:	08/13/1985
From:	Michael Levine, Stanley R Public Service Enterprise Group
To:
Shared Package
ML20213D757	List: ML20213D752 ML20213D785
References
IEIN-84-44, S-C-A910-EEE, S-C-A910-EEE-00, NUDOCS 8611120218
Download: ML20213D752 (13)

Text

• - .

S-C- Ae t >-e EE-e e e >-Ra  ;;;;;,1;;es PSEG l CABLE l

Ec ear Oeca tment -

ITLE: ENVIRONMENTAL QUALIFICATION TESTING OF ROCKBESTOS 1.0 _

PURPOSE: l 84-44 outlined potential problems regardingtesting of Class lE IE Notice No. (EQ)

Rockbestos E.1vironmental QualificationThe purpose of this evaluation is electrical c ables.

address the Salem Generating Station, thi's information and to determine w

2.0 REFERENCES

d Vendor 2.1 Information N'JREG -0040Notice Vol. No. 84-44 9, No.1 " Licensee Contractor an 2.2 inspection Status Report" "Rockbestos 2.3 Salem Equipment Qualification Report No. 08,

• 32, "Rockbestos Cerro Report" 2.4 Salem Equipment Qua'.ification Report No.

2.5 Coaxial Cable" Salem Generating Station Cable Designation List E203652 to 2.6 Purchase Order No. Littlehales, Manager - OA Rockbestos, 2.7 Letter from G. G.

B. G. Horst, SAG, 7/20/85 3.3 .pISCUSSION:

the Rockbestos Company which The NRC performed an inspection at test procedures involved reviewing qualification instruments, test test plans, log books, related raw calibration of test Several inconsistencies and ce of data, and QA documents.nonconformances were discovered with the Quality Assu certain testing and documentation practices.

impose quality For example, the Rockbestos Company did not i assurance / test control requirements on outside test ng i facilities that performed LOCA/HELB A total of 8 such tests on the were concerns cable dur ng the period from 1969-1979.

Because these inconsistancies b t s listed in IE Notice 84-44.

raised cable, questions as to the qualification of the Rock es oa lis include in the Notice.

861112O218 PDR 861107ADOCK PDR 05000272 =.

G f

10 SEPT 81 EDD-7 FORM 1 REV O -_

O

1. -.__.___m _ _ _ . . . - __ . _ _ _ _ _ w ___..-_ _ _ . _ _ . . .- %_ -

-_____. Page 2 3- 2 f i e Date: 3/12/35 jl910-EEE-0067-R0 A

0 EVALUATION: l in safety related The type uses 2 types of Rockbestos cabified eide by containment.

the test reports Salem cpplicatrons,-both inside and outsare Firewall-EP, which was names in Reference 2.3. and Firewall-III cois used in qualified in Reference 2.4.

rn because ntainment it (xer. 2.3). One Firewall-EP-is of particular conce erformed by an outside test carecy-relateo of the Qualification Test applications Reports p insioethecoduring NRC as the years of organization (Franklin)This report2.2). was also Thisidentified report isby not, F-C73798.

possibly being problematic (Ref.

f l quilification Re f . 2of. 7this

, item 2 nt, in ~~

however, cable. required for the success uOR No. 1804 t states that with their QA Departmen , .

conjunction The program fully complies withOfg fu f

for qualification testing.10CFR50 Appendix B criteria. identif '

Re f . 2. 2 nor Re f . 2. 7 ,

concern. ble used in safety-related This The second type of Rockbestos Ca i l cable (RSS6-ll3).

applications is Firewall III coax a thus the concernis of not is not used inside containment, tion (Re f.it, 2.2)the existing cable inadequate applicable.

test report, LOCA simulation instrumentaOn Furthermore, t

Constructions for Class lE Serviceis acceptable.

f supplementary l

Stations' ,1979 (Re f. 2.4)Rockbestos the validity of earlier tests.

isJanuary, performing a ser ill be completed in qualification tests to supportThe testing of the coaxial ca 1986 (Re f. 2.7) .

5.0 CONCLUSION

i t.e problem involving the Rockbestos IE Notice No. 84-44 described a poss individually and inadequate Quality Assurance practices atReference Company.

discusses the method Based in which h on this Rockbestos reference and thisR correcting the concern. evaluation, the qualification of t e Salem is not affected. >

l I

i

)

l l

o 10 SEPT 81 .

EDD-7 FORM 1 REV 0

.%,. . - _ . . _ _ _ - . . _ _ _ . t - . . - ...,.--+,__,..m- .P.+

Page 3 of 3 Date: 8/12/85 3-C-A910-EEE- 00 a 7-R0 6.0 REVIEW AND APPROVAL:

$ < rkob $- 0 /. h

" verifler or rginator

~

[ /5T fS&f Group Head Wf * /V originator's Group Head w

i.

t OkT Manager - Nucipar Systems Engineering .

BGH: jab GH-BGH/l 1-3 m

EDD-7 FORM 1 REV 0 10 SEPT 81

%_ --. --^- _._

.-w_._,,N s

, $ TTA ltil'U,T R l

ORGANIZATION: THE ROCKBESTOS COMPANY eif '

18 A MEMBER OF THE MARMON GROUP NEW HAVEN._ CONNECTICUT REPORT INSPECTION NSPECTION N0.: 99900277/84-03 DATE(S): 11/13-15/84 UN-SITE HOURS: 38 CORRESPONDENCE ADDRESS: The Rockbestos Company A Member of the Marmon Group ATTN: Mr. N. A. Lorimer, Vice President and l

General Manager Post Office Drawer 1102 New Haven, Connecticut 06504 ORGANIZATIONAL CONTACT: Mr. G. G. Littlehales, Manager, Quality Assurance TELEPHONE NUMBER: (203) 772-2250 l

PRINCIPAL PRODUCT: Insulated wire and cable. 8 NUCLEAR INDUSTRY ACTIVITY: Currently, the testing laboratory located at the New Haven plant conducts all loss-of-coolant 2ccident (LOCA) simulations, thermal aging, and flame tests performad for safety-related electrical equip-ment oualificatione(EQ) for Rockbestos pr oduced wire and cable. Nuclear i related manufacturing effort at the New Haven plant new accounts for approx-  !

imately 5 percent of the plant's total output.

ASSIGNED INSPECTOR: . [- ar# /-Jo f.S G. T. Hubbard, Equip. Qual. Inspec. Section (EQIS) Date OTHER INSPECTOR (S): V. D. Thomas, Engineering & Generic Communications Branch L. D. Bustard, Sandia National Laboratories APPROVED BY: kl.dtA ,

~

pp [ - 3c-S C U. Fotapevs, Chief,'ESIS, VPb Date INSPECTION BASES AND SCOPE:

A. F, AttS : 10 CFR Part 50, Appendix G.

B. SCOPE: This inspection consisted of. (1) a review of the status of the requalification test program (RP) heir 4 conducted on currently manufactured Class 1E csble types and (2) an inspcction cf the upgraded LOCA facility which will be used during the PD.

PLANT SITE APPLICABILITY: Docket Numbers: 50-206, 50-208, 50-213, 50-220, 50-237, 50-245, 50-249, 50-250, 50-251, 50-255, 50-263, 50-265, 50-266, 50-277, 50-278, 50-282, 50-286, 50-293, 50-295, 50 301, 50-304, 50-305, 50-305, 50-309, (continued on next page)

'-f g"I '

1

.M L . ..i

r I, .

y ORGANIZATION: THE ROCKBESTOS COMPANY A MEMBER OF THE MARMON GROUP NEW HAVEN, CONNECTICUT l m REPORT INSPECTION N0.: 99900277/84-03 RESULTS: PAGE 2 of 8 PLANT SITE APPLICABILITY: (continued) 50-313, 50-317, 50-318, 50-333, 50-335, 50-336, 50-361, 50-362, 50-368, 50-369, 50-373, 50-374, 50-377, 50-389, 50-409, 50-410, 50-413, 50-414, and 50-508.

.. VIOLATIONS

None.

B. NONCONFORMANCES:

None.

C. UNRESOLVED ITEMS:

None.

D. STATUS OF PREVIOUS INSPECTION FINDINGS:

1. (0 pen) Nonconformance (84-02, Item B.1): Traceability could not be demonstrated for a reel of single conductor insulated wire, (shop crder 90205-01). The NRC inspector took no action on this noncon-formance during this inspection.

2. (0peni Nonconformance (84-02, Item B.2): Six timer units used for rework processes and one unit used for hot aodulus testing had not been calibrated as required. The NRC inspector took no action on this nonconformance during this inspection.

3. (0 pen) Nonconformance (83-04, Item B.1): Rockbestos had not main-tained sufficient records to furnish evidence that activities affecting quality were performed. The NRC inspector took no action on this nonconformance during this inspecticn. This noncanformance remains open pending evaluation of RP results relative to the validity of past test results affected by this nonconformance.

4. (0 pen) Nonconformance (83-04, Item B.2): Rockbestos did not evaluate or document the evaluation of 4 cable failures which occurred during the F-C3798 EQ test and 11 cable failures which occurred during the F-C3859-2 EQ test. The NPC inspector took no action on this noncon-formance during this inspection. This nonconformance remains open until the results of the RP can be evaluated and the validity of cable qualification tests affected by this nonconformance can be determined.

104

.P

[.'

ORGANIZATION: THE ROCKBESTOS COMPANY A MEMBER OF THE MARMON GROUP NEW HAVEN, CONNECTICUT REPORT INSPECTION NO.: 99900277/84-03 RESULTS: PAGE 3 of 8

5. (0 pen) Unresolved Item (83-04, Item C): Rockbestos used single conductor or single twisted pair test results to establish qualification for a vastly different cable, such as an 80 conductor (80/C) cable. This is a questionable use of similarity analysis.

The NRC inspector took no action on this nonconformance during this inspection. This item remains open pending evaluation of RP test results.

6. (0 pen) Nonconformance (83-03, Item B.1, B.2, B.3, B.4, and B.5):

(B.1) lack of calibration stickers on irradiation units; (B.2) lack of evidence of test data sheet review; (B.3) lack of evidence of elongation requirement revision; (B.4) lack of evidence of review and approval of quality control instructions; and (B.5) lack of records to substantiate thermocouple monitoring performance. The NRC inspector took no action on these five nonconformances during this inspection.

7. (0 pen) Unresolved Item (83-03, Item C): Incomplete data entries for "EGTL" worksheet number 4, oil aged section. The NRC inspector took no action on this unresolved item during this inspection.

8. (0 pen) Nonconformance [83-02, Item D.1, Example (2)]: Use of inadequate test instrumentation was identified. The NRC inspectors and Sandia consultant reviewed information concerning the additional instrumentation that Rockbestos had received since the July 23-27, 1984, NRC inspection, 99900277/84-01. Speci'ically, a pressure transducer, a clamp-on ultrasonic flowmeter and a strip chart recorder are now available for RP test instrumentation.

The consultant checked the calibration status of several test monitoring instruments during the inspection. Rockbestos has engaged the services of an outside calibration laboratory to calibrate its test instrumentation. Calibration services were provided on April 17-24, 1984, and November 8, 1984. The Scndia consultant verified by examination of calibration records that the November 8, 1984, calibration included the pressure transducer, six dielectric test sets employed to voltage load cables daring testing, and ammeters to monitor test load currents. The calibra-tion of the insulation resistance test set was also verified.

The consultant determined that two instruments planned for use '

during t.0CA simulation. tests were not calibrated. These instruments are the ionapalyzer which monitors the chemical spray pH and the ultrasonic flowmeter which monitors the chemical spray flowrate.

Rockbestos plans on calibrating each item prior to use but had not l

\

105 A

. \

ORGANIZATION: THE ROCKBESTOS COMPANY A MEMBER OF THE MARMON GROUP NEW HAVEN, CONNECTICUT i REPORT INSPECTION  :

NO.: 99900277/84-03 RESULTS: PAGE 4 of 8  ;

yet developed written procedures describing such calibration techniques. The NRC will verify the implementation of written procedures for the above items during a future inspection.

Based on discussions with Rockbestos personnel, the NRC inspection ,

team determined that Rockbestos has modified some instrumentation I plans for its upcoming LOCA simulations. l Specifically, (a) Rockbestos has decided against employing a separate test vessel for post-transient humidity exposures. The steam chamber is now planned as the only test vessel for the l entire 100-day test exposure. Test environmental conditions have been modified to reflect this test equipment change. 1 It is currently planned that a 220*F, saturated steam exposure will be applied from day 4 through 100, rather than a 100 percent RH, 200*F exposure.

(b) Rockbestos has decided that buffering its chemical spray during the LOCA simulation is not feasible. Hence no effort to main-tain chemical spray composition will be attempted. The chemical spray pH will be monitored throughout the 24-hour spray exposure using an ionanalyzer and drip sampling techniques.

The NRC inspection team monitored a " mock run" at the Rockbestos test facility of a LOCA transient simulation. Test instrumentation was sufficient to monitor cable load voltages and currents and test chamber temperatures and pressures. During the " mock run," a steam ramp from 140 F to 340 F was accomplished. One minute was required to achieve a 280*F temperature. Approximately four minutes were i

necessary to achieve 340*F.

The consultant's review of the " Qualification Test Procedure Manual" (QTPM) determined that t$e manual was last revised on July 16, 1984, but it did not reflect recent instrumentation and test procedure modifications. The NRC will review the QTPM during a future inspection l to verify that the above modifications have been included in the QTPM.

This nonconformance will remain open pending completion and evaluation of the RP relative to the effect of this nonconformance on past EQ tests.

106 t

l

~

y .

l ORGANIZATION: THE ROCKBESTOS COMPANY A MEMBER OF THE MARMON GROUP f NEW HAVEN, CONNECTICUT l REPORT INSPECTION NO.: 99900277/84-03 RESULTS: PAGE 5 of 8

9. (0 pen) Nonconformance (83-02, B.2): Rockbestos did not maintain the records required to provide documentary evidence of activities effecting their EQ testing effort. The NRC inspector took no action on this nonconformance during this inspection. This item remains open until the results of the RP can be evaluated and the validity of past test results affected by this nonconformance can be

! determined.

I i 10. (0 pen) Nonconformance (83-01), B.8): Rockbestos test instrumentation was not adequate to demonstrate that the required LOCA parameters were achieved. See the discussion of paragraph 0.8 for actions taken by the NRC inspectors and the Sandia consultant relative to this noncon-formance.

11. (0 pen) No'nconformanca (83-01, B.9): There was no documented evidence available to indicate that the pH'was monitored during testing support-

ing report QR 3803. See the discussion of paragraph D.8 for actions I

taken by the NRC inspectors and Sandia consultant relative to this

, nonconformance.

12. (0 pen) Nonconformance (83-01, B.10): There was no documented evidence

- available to indicate that the cables were continuously energized I with a voltage potential of 600 VAC during EQ testing. See the

discussion of paragraph D.8 for actions taken by the NRC inspectors i and the Sandia consultant relative to this nonconformance, j 13. (0 pen) Unresolved Items (82-02, C.2): It was not apparent that; (1) brazed connection of conductors had been measured, or (2) a procedure existed for accomplishing the task. The validity of retests for; (1) accelerated water absorption and (2) flammability i properties is questionable. During NRC inspection 83-03 the NRC inspector deferred action on this item until staff technical evaluation was complete. The NRC inspector took no action on this item during this inspection.

14. (0 pen) Followup Item (81-01): Evaluation of the original qualifi-cation testing of the RSS-6-100. series coaxial cable. In May 1981, General Atomic Company submitted a 10 CFR Part 21 report reporting a failure of Rockbestos RSS-6-104 coaxial cable during main-steam-line-break (MSLB) and oven tests. Rockbestos responded to this failure by (a) redesigning the construction of the coaxial and triaxial cable series; (b) performing qualification type tests on the " newer" second and third generation coaxial and triaxial cables; (c) recognizing that the redesign causes an additional technical concern for the larger diameter members of the RSS-6-100 series 107 UC __ _ _ _ _ _ _ ___ _ _ _ __ __

I ORGANIZATION: THE ROCKBESTOS COMPANY A MEMBER OF THE MARMON GROUP NEW HAVEN, CONNECTICUT REPORT INSPECTION N0.: 99900277/84-03 RESULTS: PAGE 6 of 8

- 4 cables (for example RSS-6-109); and (d) p rforming specialized qualification tests for a customer on the RSS-6-109 cables to alleviate qualification concerns. This item was examined extensively during the inspection of June 6-10, 1983, and a number of questions concerning whether qualification had been demonstrated by the Rockbestos effort were raised. These concerns were outlined in the 83-01 inspection report and this item will remain open pending the NRC evaluation of the Rockbestos requalification test program being conducted in response to inspection reports 83-01, 83-02, and 83-04.

E. OTHER FINDINGS AND COMMENTS: ,

1 Test Plins: The NRC inspectors and the Sandia consultant reviewed test plaE number 4804, 4805, and GS83. The 4804 and 4805 test plans are for the qualification of both chemically and irradiation cross-1.nked polythylene (XLPE), respectively. The GS83 test plan is for the qualification of shielded instrumentation cable for a plant specific application. Based on review of the above test plans, discussions were held with Rockbestos QA and engineering personnel concerning the following items:

a. The test plan lists minimum documentation requirements that will be maintained to support completed qualification activities. The test plan documentation requirements, however, do not include several IEEE Std 323-1974 documen-tation requirements, specifically: (1) a description of the test facility; (2) a description of the instrumentation used; (3) reference to calibration records; (4) reference to applicable test procedures; and (5) approval signature and date.

b. The test plan appendix currently includes a discussion of Beta radiation effects. This discussion is based on an analysis of Beta radiation calculations for the Limerick generating station, a BWR reactor. The appendix should be specific as to whether the conclusions provided are generically applicable to other reactor designs, especially PWRs, or are limited to Limerick.

c. The test plan anticipates ramp times of approximately 30 seconds; however, in light of the observed " mock test," the test plan should be modified to reflect actual expected ramp times.

108

[

i I

ORGANIZATION: THE ROCKBESTOS COMPANY l A MEMBER OF THE MARMON GROUP NEW HAVEN, CONNECTICUT REPORT INSPECTION l PAGE 7 of 8 N0.: 99900277/84-03 RESULTS:

l l

2. Test Documentation: The NRC inspector and Sandia consultant reviewed test documentation for chemically XLPE cable test samples and deter-f mined that the samples had completed sample production tests, RP base-line testing, and thermal aging prior to the inspection. Irradiation aging cable samples had been delivered to Isomedix, Inc. for the I irradiation aging, at the time of the inspection. The documentation review included technician data notes (signed and dated), instrumenta-tion used including calibration status, evidence of QA particioation during test activities, and notes on visual inspection results. The consultant noted that Rockbestos was not using pre-prepared data sheets for various stages of test documentation; however, no problems were identified with the technician and QA documentation generated up to the time of the inspection. Rockbestos should consider using pre-prepared data sheets as a means to assure all required data is obtained during the testing.

3. Technical Issues: The NRC inspector and Sandia consultant discussed with Rockbestos QA and engineering personnel three technical issues which were identified during the July 23-27, 1984, inspection.

Rockbestos is planning to provide additional information on these issues as part of the qualification analysis of the RP. These issues were discussed to assure that Rockbestos understood the NRC's concerns relative to these items. The items discussed were:

a. Correlation of currently produced and previously produced

! chemically XLPE Firewall III products: In support of this correlation it is expected that Rockbestos will provide an analysis that will support similarity between 760-5 and 760-D insulation types used in chemically XLPE cable. The analysis should include a discussion of (1) thermal

• endurance data, (2) wet electrical data, and (3) radiation tolerance data for the two insulation types.

b. Correlation of coaxial LD cable to coaxial LE cable: Since Rockbestos is only testing RSS-6-100 series /LE cable during the RP and will use analysis of test results to justify qualification of RSS-6-100 series /LD cable, Rockbestos should assure that adequate acceptance test data is obtained during the RP to justify the qualification of LD cable.

I i 109 A

ORGANIZATION: THE ROCKBESTOS COMPANY A MEMBER OF THE MARMON GROUP NEW MAVEN, CONNECTICUT REPORT INSPECTION NO.: 99900277/84-03 RESULTS: PAGE 8 of 8

c. Rationale supporting thermal aging times for tett specimens with neoprene and hypalon jackets, including specimens where the hypalon jacket is bonded to ethylene propylene rubber (EPR) insulation: It is expected that the Rockbestos ar:alysis will include a discussion of practical aging techniques for EPR/

hypalon composite insulation constructions.

4. Additional Observations: During the inspection two additional observations were made to help Rockbestos obtain maximum benefit from their RP tests. The observations were that:

l a. Test procedures should require the auditability of the data logger and strip chart output records. Dates and times should be recorded; sensitivity for each channel should be noted; chart speeds should be documented; and notes should describe the location and type of each channel signal input.

b. Insulation resistance data for engineering information should be obtained on test samples aged for 954 hours0.011 days <br />0.265 hours <br />0.00158 weeks <br />3.62997e-4 months <br /> at 150 C as well as on samples aged for 168 hours0.00194 days <br />0.0467 hours <br />2.777778e-4 weeks <br />6.3924e-5 months <br /> and 121 C.

110

Y.

ORGANIZATION: THE ROCKBESTOS COMPANY A MEMBER OF THE MARMON GROUP NEW HAVEN, CONNECTICUT REPORT INSPECTION INSPECTION N0.: 99900277/85-01 DATE(S): 1/7-10/85 ON-SITE HOURS: 84 CORRESPONDENCE ADDRESS: The Rockbestos Company

, A Member of the Marmon Group ATTN: Mr. N. A. Lorimer, Vice President and General Manager Post Office Drawer 1102 New Haven, Connecticut 06504 ORGANIZATIONAL CONTACT: Mr. G. G. Littlehales, Manager, Quality Assurance TELEPHONE NUMBER: (203) 772-2250 PRINCIPAL PRODUCT: Insulated wire and cable.

NUCLEAR INDUSTRY ACTIVITY: Currently, the testing laboratory located at the New Haven plant conducts all loss-of-coolant accident (LOCA) simulaticas, thermal aging, and flame tests performed for safety-related electrical equip-ment qualification (EQ) for Rockbestos produced wire and cable. Nuclear related manufacturing effort at the New Haven plant now accounts for approx-imately 5 percent of the plant's total output.

ASSIGNED INSPECTOR: .2k #//f/M G. T..Hubbard, Equipment Qual. Inspec. Section (EQIS) Date OTHER INSPECTOR (S): S. D. Alexander, EQIS  :

R. N. Moist, EQIS L. D. Bustar h National Laboratories (SNL)  !

APPROVED BY: b) b W 4 -i f-T lildis Potapovs, Chief. E(iIS, Vendor Program Branch Date INSPECTION BASES AND SCOPE:

A. BASES: 10 CFR Part 50, Appendix B.

B. SCOPE: This inspection consisted of a review of the status of the requal-ification test program (RP) being conducted on currently manufactured Class 1E cable types with emphasis on observation of LOCA simulation testing of chemically cross-linked polyethylene (CXLPE) insulated cable.

PLANT SITE APPLICABILITY: Docket Numbers: 50-206, 50-208, 50-213, 50-220, 50-237, 50-245, 50-249, 50-250, 50-251, 50-255, 50-263, 50-265, 50-256, 50-277, 50-278, 50-282, 50-286, 50-293, 50-295, 50-301, 50-304, 50-305, 50-306, 50-309, (continued on next page)

.g Ak A

ORGANIZATION: THE ROCKBESTOS COMPANY

, A MEMBER OF THE MARMON GROUP NEW HAVEN. CONNECTICUT REPORT INSPECTION N0.: 99900277/85-01 RESULTS: PAGE 2 of 11 PLANT SITE APPLICABILITY: (continued) l 50-313, 50-317, 50-318, 50-333, 50-335, 50-336, 50-361, 50-362, 50-368,  ;

50-369, 50-373, 50-374, 50-377, 50-389, 50-409, 50-410, 50-413, 50-414, and l 50-508. j A. VIOLATIONS:

l l

None.

B. NONCONFORMANCES:

1. Contrary to the requirements of Criterion V of Appendix B to 10 CFR Part 50, there were instances in which activities affecting quality '

were either not prescribed by documented procedures or procedures did not provide for determining if the activity had been satisfac-torily accomplished.

a. Monitoring loss-of-coolant accident (LOCA) simulation chemical spray pH using batch sampling and measurement with an alternate ,

type of pH meter was accomplished without documented .

instructions or procedures.

b. The instructions for preparation of LOCA. chemical spray )

solution did not provide for determining satisfactory )

l accomplishment in that they did not provide for checking the I pH prior to use to determine if the reagents had been mixed ,

in the correct proportions to achieve the initial pH of 10.5 '

prescribed by the test plan. During the test, anomalously l

high pH (11.0 to 11.73) was observed.

2. Contrary to the requirements of Criterion V of Appendix B to 10 CFR Part 50, and the requirements of step 14A of Autoclave Operating Procedure AOP No. 6. The LOCA autoclave chamber vent valve was not  !

shut when reaching 140F to 160F as specified in the procedure. l 1

3. Contrary to the requirements of Criterion V of Appendix B to 10 CFR j Part 50 and the requirements of step 3 and step 5 of Calibration l Procedure No. 48, (1)the LOCA chemical spray sample solutions were '

not stirred continuously using a magnetic stirrer during the normal, )

continuous pH measurement nor were the buffer solutions stirred during .

standardization of the normal pH meter. (2)The pH probe of the in-  !

strument in use was not rinsed following removal from each imersion i ,

in sample or buffer' solution.

112 l l

Y ORGANIZATION:

THE ROCKBESTOS COMPANY A MEMBER OF THE MARMON GROUP 1

' NEW HAVEN. CONNECTICUT [(

REPORT INSPECTION NO.: 99900277/85-01 RESULTS:

. PAGE 3 of 11 l

i

4. t Contrary to the requirements of Criterion V of Appendix B to 10 CFR l Part 50, and the requirements of the procedure entitled " Requirements for Purchase Orders Covering Outside Calibration of Equipment", dated 4/18/78, of the Rockbestos Calibration Procedures Manual (CPM),

Rockbestos Purchase Order (P0) No.13944, dated 11/12/84, for "In-Plant Calibration of Instrumentation for Qualification Testing",

did not contain any of the information specified in the procedure.

5.

Contrary to the requirements of Criterior. V of Appendix B to 10 CFR Part 50 and the requirements for changes made to document entries in paragraph 7.3.2 of the Rockbestos Technical Manual for Class 1E Qualification Tests, (1) a test value entry on the strip chart recorder calibration data sheet for test 4804 was written-over such that it was not clear which was the original entry and which was the corrected entry.

by the person making the correction.(2) The new entry was not initialed and date C. UNRESOLVED ITEMS:

None.

D.

STATUS OF PREVIOUS INSPECTION FINDINGS:

1.

(0 pen) Nonconformance (84-02, Items B.1 and 8.2):

(B.1) Traceability could not be demonstrated for a reel of single conductor insulated wire, (shop order 90205-01).

(B.2) Six timer units used for rework processes and one unit used for hot modulus testing had not been calibrated as required. The NRC inspector this inspection.took no action on these two production nonconformances during 2.

(0 pen) Nonconformance (83-04, Item B.1):

Rockbestos had not maintained sufficient records to furnish evidence that activities affecting quality were performed under suitably con-trolled conditions. The NRC inspectors reviewed the latest revisions of quality assurance documents, procedures, inspection plans, etc. for compliance with regulatory requirements and verified by observation of test and QA personnel activities and review of gg associated records that the controls were being implemented except ,

as noted in the nonconformances of paragraph B. above. This noncon-formance remains open pending evaluation of RP results relative to

! the validity of past test results affected by this nonconformance.

b -

113

l ORGANIZATION: THE ROCKBESTOS COMPANY A MEMBER OF THE MARMON GROUP NEW HAVEN CONNECTICUT REPORT INSPECTION RESULTS: PAGE 4 of 11 N0.: 99900277/85-01

3. (0 pen) Nonconformance (83-04, Item B.2):

Rockbestos did not evaluate or document the evaluation of 4 cable failures which occurred during the F-C3798 EQ test and 11 cable failures which occurred during the F-C3859-2 EQ test. The NRC inspector took no action on this nonconformance during this inspection. This nonconformance remains open until the results of the RP can be evaluated and the validity of cable qualification tests affected by this nonconformance can be determined.

4. (0 pen) Unresolved Item (83-04, Item C):

Rockbestos had used single conductor or single twisted pair test results to establish qualification for a vastly different cable, such as an 80 conductor (80/C) cable. This was a questionable use of similarity analysis. The NRC inspector observed that RP specimens now included 7/C and 2/C completed cable section samples as representatives of multiconductor (M/C) types. This item remains open pending evaluation of RP test results.

5. (0 pen) Nonconformances (83-03, Item B.1, B.2, B.3, B.4, and B.5):

(B.1) Lack of calibration stickers on irradiation units; (B.2) lack of evidence of test data sheet review; (B.3) lack of evidence of elongation requirement revision; (8.4) lack of

, evidence of review and approval of quality control instructions; I l and (B.5) lack of records to substantiate thermocouple monitoring performance. The NRC inspector took no action on these five production nonconformances during this inspection.

6. (0 pen) Unresolved Item (83-03, Item C):

Incomplete data entries for "EGTL" worksheet number 4, oil aged section. The NRC inspector took no action on this unresolved item during this inspection.

l

7. (Opea) Nonconformance (83-02, Item B.1, Example (2)):

Use of inadequate test instrumentation had been identified. The NRC inspectors examined instrumentation and reviewed related documentation. Instrumentation has been improved including equipment for recording test. parameters and calibrations were current. In reviewing this aonconformance, the following was noted:

114 k -

-Y ORGANIZATION: THE ROCKBESTOS COMPANY A MEMBER OF THE MARMON GROUP

, NEW HAVE.N, CONNECTICUT REPORT INSPECTION 1

N0.: 99900277/85-01 RESULTS:

~ PAGE 5 of 11 l

a.

Calibration report ECL 84008A for the LOCA chamber pressure transducer, certified it to be calibrated over the range of 20 to 200 psig. However during a significant portion of the LOCA exposure, pressu,re was expected to be less than 20 psig. Rockbestos test personnel contended that this was not a problem with saturated steam conditions in the chamber ,

since the pressure could be determined by the temperature, and that temperature was the primary means of indicating the state of the system in this phase of the LOCA exposure.

b. l The LOCA chemical spray flowmeter was calibrated by the manufacturer for the specific application, however, the calibration certificate was not on file at Rockbestos. The spray piping is stainless steel, but the manufacturer's calibration label incorrectly identified it as carbon steel.

The effect of this on accuracy could not be determined during the inspection. After contacting the manufacturer, Rockbestos stated that this concern would be resolved in the calibration certificate which was being forwarded to Rockbestos.

This information will be reviewed during a future inspection.

c.

Rockbestos has developed written procedures for before-use calibration of the LOCA chemical spray flow and pH meters.

l Review of these procedures and related records resulted in the following comments:

(1) Standardization procedures neglected buffer solution pH temperature correction, however there was no documented evidence that this source of error had been evaluated and determined to be neglectable.

(2) The maximum error recorded during before-use calibration of the flowmeter was 6.3%, exceeding the 5% allowed by paragraph 13.2.5 A staff memo, of the Rockbestos Quality Manual (RQM).

dated 12/11/84, indicated that the flow-meter accuracy specification should be changed to 220%,

however procedures had not been revised accordingly.

pgY 115 SI --

U j

ORGANIZATION: THE ROCKBESTOS COMPANY >

A MEMBER OF THE MARMON GROUP NEW HAVEN. CONNECTICUT INSPECTION REPORT PAGE 6 of 11 RESULTS:

NO.: 99900277/85-01

d. The thermocouples now in use had a sufficiently fast response to be able to track the transients achievable with the Rockbestos autoclave system.

This nonconformance (D.7) will remain open pending evaluation of final RP results relative to the effect of this nonconformance on past EQ tests.

8. (Closed) Nonconformance (83-02, B.2):

Rockbestos had not maintained records required to provide documentary evidence of activities affecting their EQ testing effort. The NRC inspectors review confirmed that appropriate records are now being maintained.

9. (Closed) Nonconformances (83-01, B.8, B.9, and 8.10):

(B.8) Rockbestos test instrumentation was not adequate to demon-strate that required LOCA parameters were achieved. (B.9) There was no documentary evidence available to indicate that pH was monitored during testing described in report QR 3803. (B.10) There was no documentary evidence available to indicate that appropriate cables were continuously energized with 600 VAC during EQ testing.

The NRC inspectors found that new instrumentation is capable of sensing the LOCA parameters achievable with Rockbestos' autoclave.

Test procedures and equipment now provide for monitoring and record-ing pH and sample cable current and voltage. The NRC inspectors observed the continuous monitoring and automatic recording of these parameters during the inspection.

10. (0 pen) Unresolved Items (82-02, C.2):

It was not apparent that (1) brazed connection of conductors had I been measured, or (2) a procedure existed for accomplishing the task. The validity of retests is questionable for (1) accelerated water absorption and (2) flammability properties is questionable.

During NRC inspection 83-03, the NRC inspector deferred action on this item until staff technical evaluation was complete. The NRC inspector took no action on this item during this inspection.

I 116 ,

w .. -- - . . . . . . - -

(

ORGANIZATION:

THE ROCKBESTOS COMPANY A MEMBER OF THE MARMON GROUP NEW HAVEN. CONNECTICUT

. REPORT

! INSPECTION NO.: 99900277/85-01 RESULTS:

i PAGE 7 of 11

11. (0 pen) Followup Item (81-01):

Evaluation of the ori series coaxial cable.ginal qualification testing of the RSS-6-100 submitted a 10 CFR Part 21 report reporting a failure ofIn may 198 Rockbestos RSS-6-104 coaxial cable during main-steam-line-break (MSLB) and oven tests. Rockbestos responded to this failure by (a) redesignin cable series; g the construction of the coaxial and triaxial (b) performing qualification type tests on the

" newer" second and third generation coaxial and triaxial cables; I (c) recognizing that the redesign causes an additional technical concern for the larger diameter members of the RSS-6-100 series cables (for example RSS-6-109); and (d) performing specialized qualification tests for a customer on the RSS-6-109 cables to alleviate qualification concerns.

sively during the inspection of JuneThis6-10, item1983, was examined exten-and a number of questions concerning whether qualification had been demonstrated by the Rockbestos effort were raised.

These concerns were outlined in the 83-01 inspection report and this item will remain open pending the NRC evaluation of the Rockbestos requalification test 83-02, and 83-04. program being conducted in response to inspection reports 83 E.

OTHER FINDINGS OR COMMENTS:

1.

During this inspection, the NRC inspectors observed the first four days of LOCA exposure testing of CXLPE insulated cables and the examination of partially thermally aged RXLPE specimens and examined related documentation. Implementation of test controls and involvement of QA personnel were also observed.

2.

Rockbestos revised test plan TP4804 (CXLPE) since the November, 1984 NRC inspection as follows:

a.

Post conditioning sample evaluation now includes insulation resistance (IR) measurements on samples thermally aged at both 150*C and 121 C.

Y 117 1

3 ORGANIZATION: THE ROCKBESTOS COMPANY A MEMBER OF THE MARMON GROUP i

NEW HAVEN. CONNECTICUT REPORT INSPECTION r NO.: 99900277/85-01 RESULTS: PAGE 8 of 11 l

b. Instrumentation requirements were modified to specify continuous thermocouple readings and recording during the LOCA transients.

c. The test plan now requires IR measurements to be made on all i circuits rather than "on selected circuits."

d. Environmental ramp estimation was changed from 30 seconds to two minutes due to the limitations of the autoclave system.

3. Test Plan 4805 (November 1984 revision) for qualification of FWIII Irradiation Cross-linked Polyethylene was observed to have incorrect product codes. Rockbestos personnel agreed that Test Plan 4805 will need to be modified to correct this and reflect the December,1984 revision to TP4804 as well as to reflect test experiences for the 4804 test.

I 4. Test Plan GS83 (dated 10/16/84) for Instrumentation Cable (Cable Mark Nos. NGP-67, NGP-68) is also expected to be revised to reflect l TP 4804 and 4805 modifications and test experiences. '

5. Procedures did not include requirements for auditability of data logger and strip chart recorder outputs in terms of anotation with dates, times, charts, speeds, channel ids, scales, sensitivities, input ID and descriptions although these annotations are being made in practice.

6. The calibration statement in Appendix I (Equipment List) of Rockbestos Test Plan 4804 states, in part: " ...The calibration program is conducted to meet the requirements as outline in ASTM

, D-2865 and MIL-C-45662A...," but the procedure in the Calibration Procedures Manual for preparing calibration services l P0's does not include these requirements.

7. Test Observations:

a. Four minutes were required to achieve a 280 F steam temperature in the first transient. Five minutes after the start of the first transient, Rockbestos personnel observed that a chamber vent had been inadvertently left open. It was closed and a steam temperature of 340 F was achieved seven minutes after the spray was turned on. See noncon-formance B.2.

118 i

3I o -

ORGANIZATION: THE ROCKBESTOS COMPANY A MEMBER OF THE MARMON GROUP NEW HAVEN. CONNECTICUT REPORT INSPECTION NO.: 99900277/85-01 RESULTS: PAGE 9 of 11

b. Prior to the start of the chemical spray exposure, the NRC inspectors noted that at least two cable load supplies

" tripped off." They were reset after a quick check and adjustment of cable connections to the power supplies. This anomaly did not occur during the second transient exposure.

c. Throughout the first transient, Rockbestos had problems with its chemical spray pH monitoring, including:

(1) trouble stabilizing the pH reading (2) pH readings anomalously higher than the expected 10.5.

See nonconformance B.1.b.

(3) no monitoring of chemical spray reservoir pH prior to the start of the test. See nonconformance B.1.b.

(4) failure of pH monitoring equipment during the first 340 F peak steam exposure. Prior to the second transient, back up pH monitoring technique was adopted.

Rockbestos did not have an approved, written procedure for the alternate method of measuring pH and buffering the alternate meter. (See nonconformance B.1.a.) Some instructions for use were given on the instrument and in its technical manual but were not observed to be consulted.

d. The test cable schematic, number of power supplies, and loading of one cable deviated from the test plan. These changes were documented and justified by notes in the qualification file.

e. Approximately 2-1/2 minutes were required to achieve 280*F steam temperatures in the second transient. Five and one-half minutes were required to achieve 340*F steam temperatures. Chemical spray, which had been turned off for the transient, was restarted at completion of the transient.

119 l

.md hm .

o ,

ORGANIZATION: THE ROCKBESTOS COMPANY A MEMBER OF THE MARMON GROUP NEW HAVEN, CONNECTICUT REPORT INSPECTION N0.: 99900277/85-01 RESULTS: PAGE 10 of 11

f. There was an anomalous reduction and fluctuation in load on one current set. A loose test cable connection was tightened and the current regained its nominal value.

g. Approximately 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> into the LOCA, some of the test cables were observed through the chamber sight glass to be immersed in accumulating condensate. The chamber was drained and no attributable effects on performance were subsequently observed.

h. Insulation resistance (IR) measurements consistently varied as expected with temperature, specimen exposed length, number of conductors, shielding and ground connections.

i. Observation of the cables in the chamber revealed that a multiconductor cable's jacket had split prior to the 250*F IR measurements which were consistent with this condition but not unsatisfactory.

8. Thermal Exposure for Radiation Cross-Linked Polyethylene Products (RXLPE): Several RXLPE single conductor specimens had been subjected to a 150*C thermal exposure for approximately 6 days. This exposure was prematurely stopped because of indentation of cable insulation by the mandrel cable supports.

Subsequently, Rockbestos reported they were restarting the aging with new RXLPE specimens and had fitted the mandrel cable supports with sleeving to prevent the indentations.

9. Test Documentation: The test plan had been modified to provide for leaving the cable samples in the LOCA chamber under low temperature and pressure saturated steam conditions instead of moving them to a humidity chamber, however, the test data sheets (July 9, 1984 revision) included in the December 1984 revision of test Plan 4804 had not been revised to reflect this change.

120 L __ . J

II ORGANIZATION: THE ROCKBESTOS COMPANY A MEMBER OF THE MARMON GROUP NEW HAVEN, CONNECTICUT REPORT INSPECTION NO.: 99900277/85-01 RESULTS: PAGE 11 of 11

10. With respect to nonconformance B.2.b, the prescribed, continuous sampling technique used initially in monitoring LOCA chemical spray pH provided some mixing of the sample solution, but there was no documentation authorizing or justifying the deviation from established procedures.

11. Some elements listed on the personnel QA/QC qualification record sheets of two test personnel were marked with arrows proceeding from elements above in the blocks for instructors' initials and date of completion and were not indicated as completed by entering the instructor's initials and date for each element completed as required by step 2 of Personnel Qualification Procedure Q-25B, rev. O, dated 4/26/83. This area will be reexamined in a future inspection.

12. Rockbestos QA personnel stated that the'y had done a vendor survey on the calibration laboratory which calibrated much of their EQ test equipment but the record of the survey was not available for review during the inspection. The pror.ess of approving and procuring calibration services will be reexamined in a future inspection.

121 l l!

A L. ...

1

a "

A rrac a n gur *3 P IM ORGANIZATION: THE ROCX8ESTOS COMPANY NEW HAVEN, CONNECTICUT REPORT INSPECTION 4/22-24,5/22-24, INSPECTION NO.: 99900277/85-02 DATE(S): 9/16-17/85 ON-SITE HOURS: 104 CORRESPONDENCE ADDRESS: The Rockbestos Company A Member of the Marmon Group ATTN: Mr. R. S. Thayer General Manager Post Office Drawer 1102 New Haven,' Connecticut 06504 ORGANIZATIONAL CONTACT: Mr. George Littlehales, QA Manager TELEPHONE NUMBER: (203) 772-2250 PRINCIPAL PRODUCT: Insulated Wire and Cable.

NUCLEAR INDUSTRY ACTIVITY: Currently the testing laboratory at the New Haven plant conducts all thermal aging, loss-of-coolant accident (LOCA) simulations, post-LOCA sample evaluation testing, and flame testing for Class IE (safety-related) electrical equipment qualification (EQ) on Rock 3estos cable. Nuclear related product manufacturing at the New Haven plant comprises approximately 5 percent of the plants total output.

ASSIGNED INSPECTOR: h m /0[J y S. D. Alexander, Equip. Qual. Inspec. Section (EQIS)'Date OTHER INSPECTOR (S): M. Jacobus, Sandia National Laboratories (SNL)

APPROVED BY: _

M 2,qll ,

[0 -87 U. Potapovs, Chief. EQI3, Vendor Program Branch Date INSPECTION BASES AND SCOPE:

A. BASES: 10 CFR Part 50, Appendix 8.

B. SCOPE: This inspection consisted of three parts in which the status of the requalification program (RP) on currently manufactured Class IE cable types was reviewed. Emphasis in the first part was on observing post-LOCA sample evaluation testing of chemically cross-linked polyethylene (CXLPE) insulated Firewall III, in the second part, on observing LOCA simulation for radiation cross-linked polyethylene (RXLPE), and in the third part on observing post-LOCA sample evaluation testing of the RXLPE.

PLANT SITE APPLICA8ILITY: Sew 4nofer'1*(50-206); Haddam Neck (50-213); Nine Mile Point 1 (50-220); Dresden 2 (50-237); Millstone 1 (50-245); Dresden 3 (50-249); Turkey Point 3 and 4 (50-250/251); Palisades (50-255)

(continued on next page)

((

cD

7 f

, ORGANIZATION: THE ROCKBESTOS COMPANY NEW HAVEN, CONNECTICUT  ;

REPORT INSPECTION NO.:' 99900277/85-02 RESULTS: PAGE 2 of 13 PLANT SITE APPLICABILITY: (continued)

Monticello (50-263); Quad Cities 2 (50-265); Point Beach 1 (50-266); Peach Bottom 2 and 3 (50-277/278); Prairie Island 1 (50-282); Indian Point 3 (50-286); Pilgrim 1 (50-293); Zion 1 (50-295); Point Beach 2 (50-301);

Zion 2 (50-304); Kewaunee (50-305); Prairie Island 2 (50-306); Maine Yankee (50-309); Arkansas 1 (50-313); Calvert Cliffs 1 and 2 (50-317/318);

Fitzpatrick (50-333); St. Lucie 1 (50-335); Millstone 2 (50-336); San Onofre i 2 and 3 (50-361/362); Arkansas 2 (50-368); McGuire 1 (50-369); LaSalle 1 and 2 (50-373/374); Med. Coll. Hanover (50-377); St. Lucie 2 (50-389); Lacrosse (50-409); Nine Mile Point 2 (50-410); Catawba 1 and 2 (50-413/414); and WPPSS-3 (50-508).

A. VIOLATIONS:

None.

B. NONCONFORMANCES:

None.

C. UNRESOLVED ITEMS:

None.

D. STATUS OF PREVIOUS INSPECTION FINDINGS:

1. (Closed) Nonconformance (85-01, Item B.1):

Rockbestos had conducted chemical spray pH monitoring using an alternate technique and instrument without documented procedures and the procedures for preparation of chemical spray solution did not include criteria for determining if the solution had been correctly prepared. The NRC inspector verified that procedures covering the alternate pH monitoring technique and instrument were being prepared and procedures for preparing spray solution had been revised to include appropriate acceptance criteria.

The NRC inspector also observed preparation of the spray solution and verified compliance with the procedure.

2. (Closed) Nonconformance (85-01, Item B.2):

Rockbestos had failed to follow their LOCA chamber operating pro-c'edures by not shutting the vent valve when required. During this inspection, Rockbestos stated that the operators had been retrained

5q3 8 ORGANIZATION: THE ROCXBESTOS COMPANY NEW HAVEN, CONNECTICUT REPORT INSPECTION NO : 99900277/85-02 RESULTS:

PAGE 13 of 13 c.

The NRC inspectors reviewed qualification test plan TP 5802 3 for Rockbestos adverse service coaxial and triaxial cable l

- (including-RSS-6-100--series coaxial cables), reviewed associ-ated data generated thus far and examined the specimens which had recently been returned from the irradiation facility. The samples on the mandrel upon which they had been i rradiated were identified as prescribed in the documentation and there was no apparent degradation noted upon cursory visual examination.

The irradiation report was not yet available for review, but pro-curement documents were reviewed and found to be in compliance with regulatory requirements. The following comments pertain to the coaxial cable documents examined:

(1) TP p.6: TP 5802 specifies IR values of 106 megohms /1000 ft.

With18f)to10sampigs, in the 10 it will be necessary to measure values megohm range to demonstrate conformance to this specification.

specification.

Rockbestos is reevaluating the (2) TP p.7: References to particular Sandia National Laboratories (SNL) tests a're not given to support the assertion that synergistic effects of simultaneous radiation and thermal aging have been considered and determined not to be present. The next report revision is to include those references.

(3) TP p.11: Soak time specifications omitted from the currut revision of the TP will be included in the next revision.

(4) TP p.24: The specification for LOCA chemical spray flowr.ste is given as 15 gpm instead of the correct value of 1.5 gpm.

This typographical error is to be corrected in the next re-vision to TP 5802.

(5) The initial functional check data sheet did not include the timer in the list of equipment used. The Rockbestos Technician corrected.this and added the timer's identi-fying information to the TP 4805 test log book as well.

(6) The documentation included aging calculations which used an activation energy of 2.75 eV for the coaxial and tri-axial cable insulation (XLPE). The inspector verified that the aging calculations were correct for the values used, but the testing used as a basis for the calculations will be reviewed in a future inspection to verify its applicability to the cable currently being tested.

A yy ENT f (<n f 1

fT has been demonstrated to be qualified, similarity of tested equipment to that installed in the plant (e.g., insulation class, materials of components of the equipment, test c$nfigurauon compared to installed configuration, and documentation of both),

evaluation of adequacy of test conditions, aging calculations for qualified life and replacement interval determination, effects of decreases in insulation resistance on eauioment na iormance, adequacy or uemonstrated accuracy, evaluation of test anomalies, and applicability of EQ problems reported in IE ins / Bulletins and their resolution.

During its_ review of the component files the inspection team identified one concern, described below. Otherwise, the files were found to be readily auditable and complete.

The inspectors were unable to conclude that the files adequately

! documented qualification of two types of Rockbestos (Cerro) cable, Firewall SR and Firewall III, although there is no reason to believe that either would fail during a LOCA.

In each case the concern

.specimens.

centers on establishing similarity between plant cable and test The qualification of Firewall SR was based on a Rockbestos test report, dated March" Qualification 2, 1978. of Firewall SR Class 1E Electrical Cables,"

3 This report is subject to IN 84-44, which reported quality assurance qualification test programs. deficiencies in the Rockbestos equipment The SMUD qualification file identifie this problem with the Rockbestos test report and stated that the Firewall SR qualification is based on the Rockbestos test report...

"with the understanding that the results are subject to verificati n by the Rockbestos Supplemental Test Program." The Supplemental Te t Program is scheduled to be completed in July 1986. On May 24, 19 ,

SMUD presented additional data to support qualification of Firewa 1 SR using the Franklin report F-C2857, dated September 1970 but ,

similarity between the plant cable and test specimens was not adequately established. (

For Firewall III cable, the file deficiency involves similarity f between the Rancho Seco cable and specimens tested by Sandia i Laboratory.

Rancho Seco's cable is insulated with chemically cross-linked polyolefin formulation 760-5, obtained by Rockbestos from a different supplier than the formulation 760-D that was tested. The licensee argued that all cross-linked polyolefins meeting IPCEA standards would be satisfactory, as evidenced by successful Sandia test results for cables from three different vendors. The licensee obtained thermal aging and dry electrical i data for the plant cable, but did not have corresponding data for the tested specimens. Other infomation such as radiation and wet electrical characteristics was not provided for either type of cable. .

The inspectors concluded that similarity between the plant cable and test specimens was not adequately demonstrated.

Rockbestos Firewall SR and Firewall III Cable is an Open Item, 50-312/85-14-01. -

t

._ _ _ _ _ _ _ _ - - 13_ _ _. _ _ _ _ . - - - -

. , , , y- -3 INTERNAL MEMORANDUM cara Arp e 28 /88T~

,0

% Aa-, A d ,,0, A LO LOCATION LUBJECT /#  %

72Eed.CL &R fc "z feo32) ws nu n L ,e / L M m J E / c u n de -

Al #~ E 205744 -f:- A ,, . G , G 7 C A A A d wA ~ 1% . 7L e z//n ne-c u.se/4' %4

'bn,- n- #c.a L w k h aac xr.p. .a' fa&

l

} $

=

k g.

A - 4 nETunN REPLY REQUESTED C YE5 O NO O PSEG _ _.

.<TE,

. o. . ffHA<&*L6 W #

$ p .L f 5 6

THE ROCKBESTOS COMPANY July 26, 1985 O 285 NICOLL STREET. PO. OAAWER h 02. NEW HAVEN. CT 06504 TELEPHONE (203)772 2250 TLX:956042 Systems mgineering P.O. Box 236 Hancocks Bridge, New Jersey 08038 Attention: Ms. Barbara Horst Subject : NRC lE Notice No. 84-44 Environmental Qualification Testing of Rockbestos cables

Dear Ms. Horst:

his letter will sumarize action being taken by Rockbestos in response to 1E Notice 84-44 and the preceding NRC inspections of our Qualification Testing Program. Our program to address the NRC concerns consists of the following:

1. Purchase of newly available instrumentation with improved resolution and capability for calibration.

2. Performance of a series of supplementary qualification tests covering each cable type currently produced by Rockbestos, including the various size groupings within each type. Results of this testing will support the validity of prior tests on the same cables and on similar cables. This special test program is underway and is scheduled to be completed for all cable types in August, 1986. Testing on chemically cured Firewall III was successfully completed in June,1985, testing on irradiation cured Firewall III will be completed in September,1985 and testing on Coaxial cables will be completed in January,1986.

A technical manual and program to control all aspects of qualification testing has been in place since late 1982. m is program has been updated as necessary and neets all requirements applicable to qualification testing activity. The program is audited periodically under our quality assurance system, and there will be ongoing QA/QC inspection and verification of all future testing.

We have responded in detail to the NRC reports and Will continue to work cooperatively with the Commission to resolve any furtner issues. All findings and comments have been addressed and we have had no adverse reaction to our resconse during many subsequent e.eetintia sich the (GC. The :30 verified action completed to date during inspections in Jur/, and Nove'.oer, 1984, and in January, Apell and Miy, lh5.

4xbestoa and 7 tan <ltn naarch Center in ene last fitteen yeara nave perrormed over thirt/ Jeparate teats of six oaste caole types produced b/

Rockbestos. The collective resulta of this repeated tasting provide confidence in the ability of the designated insulation materials to perrorm under postulated adverse conditiona.

Rockbestos Firewall III was successfully tested by Sandia Lanoratories for the NRC as outlined in the report entitled "Eqdipment Qualification Researcn Test of Electrical Cable with Factory Splices and Insulation Resork", dated April, 1982 and the report entitled " Die Ef fect of IDCA Simulation Procedures on Cross-Linked Polyolefin Caole's Performance", printed April,1984.

4 Div$0N OF CEROCs WRE 4 CA8LE GROUP sNC

r r .

l Ms. Barbara Horst (2) July 26, 1985 I

b We remain fully confident that our cables are able to meet all applicable environmental qualification requirements, and we feel certain that the above action when cogleted will enable the NRC to verify the environmental qualification of all Rockbestos Class IE cables supplied to the nuclear industry.

% e following will address the eight items of concern listed in IE Notice 84-44:

1. Rockbestos did not ig ose QA/ test control requirements on an outside test organization which performed testing during the period of 1969-1979.

oximents:

Rockbestos engineering personnel worked closely with Franklin Research Institute during the referenced testing which took place between 1969 and 1974. They were fully acquainted with the methods and procedures employed by FRI during this period.

Franklin Research Institute was and is widely recognized as a qualified testing organization, and its services have been utilized for qualification testing by many companies involved with the nuclear industry.

Except for early flame retardant XLPE cables, qualifications established by FRI testing were lai.cr re-established by Rockbestos testing.

2 e Rockbestos % chnical Manual now documents the QA procedures for qualification testing. All future qualification testing will be accomplished in accordance with this manual. It provides for QA controls on service and material suppliers, as well as for 2A inspection and auditing of the entire program.

2. The Rockbestos equipment qualification program was controlled by an engineering organization which was not under a QA program until 1983.

Comments:

Rockbestos qualification tests, except in special cases, were performed as generic tests associated with product development and prior to marketi.y the products. As such, they were conducted as an engineering activity not within the scope of the GA system .it tnat tias.

S. 1982, the Rockbestoa engineertnj organiation ..tca B t at c.Tr/-

u.atituted a TecnnicM *Tnuett

. and uccwnted progrt) tur ialtdt? -

t utinJ. we art;tna purpcse af tr.is aceton was to i . :r 0.3 ::ucu.

critetti f or craou.: n :46S teervattation at organta ts..ta con Lc.. .

<icalification testtnJ of aafety systems equipment. cneae crit m t . :: .

based on AMdI/AfiE 'KA-1-1979. The IEEE program was not put into eff2ct, and the URC instead undertook a series of inspections to 10CFa50, Appuc.Jix 3 criteria. the Rochovatos program has been updated so enat it corplie; fully with 10CFR50, Appendix B. It provides the necessary QA controla and procedures and will govern all future qualification teating of acekcestou cables.

Our planned supplenental test program will support ene results of prior qualification tests snien were not conducted uncer i QA program.

• " *Ms. Barbara Horct (3) July 26, 1985 l *

3. Testing and the required documentation were not properly controlled.

Several discrepancies between final qualificgtion reports and suworting data were found.

(bments:

Discrepancies noted during the mid-1983 NRC inspections of Rockbestos qualification testing were addressed where possible with supplementing notes to the data files. All supporting data have been reviewed and reorganized. Several qualification reports will be revised to incorporate corrections. None of these report revisions, however, will affect test results or conclusions. In order to address items such as lack of formal documented test plans or lack of monitoring data, Rockbestos has comitted to a two year program of supplemental generic tests on all currently produced cable types, including the separate use classifications, i.e.

control, instrumentation, power and thermocouple cables. Results of these tests will support the validity of prior testing on the same and similar cables. Reports of new testing will also be available for future submittals as nee &d. 'Ihis program has comenced and will be cogleted for all cable types in August, 1986. Testing for Firewall III flame retardant utility cables will be cogleted in September,1985, and testing for Firewall III Coaxial Cables will be completed in January, 1986.

The Rockbestos Technical Manual and program provides the necessary testing.

and documentation controls for future qualification testing.

i 4. Rockbestos QA and engineering organizations did not impose QA requirements i and acceptance criteria on organizations that performed qualification testing for Rockbestos between 1969 and 1979. No supporting data for these tests were available.

(bments_:

Our coments applicable to Item 1 apply also to this item.

FRI disposes of testing records after five years. Prior to disposal, FRI '

states that these records are offered to the client. Rockbestos files, however, contained supporting data for only two of nine test reports. FRI confirmed that no supporting data for any of these reports were in their j own files.

Since cables tested by FRI were closely related and in some cases identical to caoles produced today, the results of the planned supplemental testinJ and data associated with these tests will support the earlior :'R! reportJ.

( ACorementioned gaalif t'nian praaru controla will as Les Mat ceca.rv/ 3 regairenanta and '- ~ mce cri:eria are imtxued tar my w.ure or n a l testing to qwliff naclear Jatety r21ated caala.

5. rest equipment and instrumentation were obcerved to nave inadequate resolution to record [DCA test parameter 3 and tunctioning of test specimens during testing.

I

• Ms. Barbara Horst (4) July 26, 1985

5. comments: g The supplemental test program is expected to confirm our position that the problems described above had no effect on test results. Rockbestos instrumentation for instance was not capable of recording accurately a 10 second rise in tenperature or a 15 minute duration as required by the specified tenperature profile. Instrumentation also did not permit measurement of spray chemistry during testing. We conclusion of our engineering personnel is that possible minor inaccuracies in measuring these parameters would not have a significant effect on test results for the types of materials which were tested by Rockbestos.

In order to update and improve the resolution and capability for calibration of our instrumentation, a multi-channel data logging and trend line recorder has been installed to monitor the following qualification test functions:

a. Autoclave temperature - resolution sufficient to differentiate changes in time periods on the order of seconds. Continuous recording.

b. Autoclave pressure - same as above.

c. Spray flow rate - transducer with 0-2 gpm range. Continuous recording.

d. Spray PH - PH meter with recorder output. Continuous " drip" sanpling of spray solution.

e. Voltage - new 0-2000 volt power supplies with analog output of voltage and leakage current to recorder.

f. Current 100 aap supplies with analog output to recorder.

W e recorder provides alpha numeric recording with time and date of recording, as well as trend line continuous monitoring of the abovo functions. Hard paper copies are capable of accepting QA/QC witness notations. In addition, the equipment will provide reproducible copies of raw data upon request and will enable inclusion of data with qualification test reports.

Future testing will employ this newly purenased instrumentation.

6 P st equipment was not properly callorar?d or under control of the calibration system.

% mta:

A .s;. , c;a tuttr u. A m .: sa :ay, b ;] . mt i. : t a ; u tore wa p: M r.:

JL th d*NC AnatrJCOHC 3.  %%C CGr a f1Jd .hdter WnLCd Gcts 130apelO'e . Of ning effectively caltorated, all prooltua aere satictactorily resolvec enaultntj close out of ena accxtasstoa audic canding. rne clas .eter nu oeen replaced with a unit which fully meets out requirements.

y

,ss',

Ms. Barbara Horst (5) July 26, 1985

6. Coments: (continued) of major concern to the NRC was the autoclave temperature recorder which appeared from the data to be providing readings outside of accuracy limitations. This problem, however, affected only testing performed during a limited time period. Adilevement of required testing tenperatures was verified for affected tests by recorded saturated steam pressures and by tenperatures obtained using a Imeds & Northrop potentiometer.

'1he supplemental test program is expected to remove any remaining doubts regarding the effect of these calibration anomalies on prior test results.

Future problems will be prevented by new instrumentation, established program controls and by periodic m auditing of the program.

7. '14st plans, acceptance criteria and test procedures for certain qualification tests were not made available during the NRC audits.

Cbments:

We supplemental test program will verify results of earlier tests where detailed plans and procedures were not part of the supporting records.

We Technical Manual now provides an auditable basis for assuring that these documents are made part of the qualification program.

8. A nunber of test deficiencies, deviations and other anomalies were not documented and evaluated in the test reports.

Coments:

Oar coments under Item 3 also apply to this item.

As indicated above, we presently have reports of testing by Sandia Laboratories which support the qualification of Firewall III cables, both chemically cured XLPE insulation and irradiation cured XLPE. In addition, qualification of Rockbestos Firewall III and coaxial caoles is supported by a number of other tests performed by outside concerns for their own purposes.

M believe that enese reports, supplementst witn our wn analpes, will provide intetim support of our Firavail 1:I x.d Cmnal ch :alifiettions to th?

e tafiction of all consta l. A t e r > ;n ' w. . a . . Penrha r a:cordtq r

-1 p Mdule, h t; ** r t. *,

UiE dVCE.3552.'4 *r UN W Pjn J, p.

/4 /,%.%

4. G. LLtti.naleJ 41 nap!r, 3idlit/ \SaurMca GGL:.Tra

ATTACHMENT B - 2 A potential enforcement item was identified by the NRC related to the absence of gear box grease reliefs on Limitorque actuators located in containment. The contention is that the gear box grease reliefs must be provided in the plant installation since they were installed in the Limitorque actuators subjected to the LOCA/MSLB test simulation. It is stated by Limitorque that the relief valve provides for telief of grease and the pressure build-up which occurs due to thermal expansion at prolonged elevated temperatures.

However, Limitorque actuators for Nuclear Plant applications are designed to operate in normal and accident conditions without depending on absolute sealing. In fact, the ambient is not absolutely restricted from entering the actuator. The seals, as such, are of no importance for qualification and, therefore, need not be considered for the purpose of qualification.

Mr. J. Drab of Limitorque stated in a telephone conversation with Ms. B. Horst of PSE&G on April 9, 1986, as documented in the " Record of Verbal Discussions" (See attached)..."They are not an EQ requirement and their absence does not adversely affect the qualification status of Limitorques", Mr. Drab and Limitorque have subsequently refused to issue this statement in a signed letter.

In FIRL test report F-C3441 prepared for Limitorque, it was identified that an escape of grease (type not identified) through the relief valve was due to flooding of the chamber from steam condensate completely submerging the Limitorque. As this was the mechanism by which the grease escaped, rather than a pressure buildup due to extreme temperatures, there was no correlation to accident conditions. When the water level was reduced, the drains became clogged. It should also be noted that a partial breakdown of grease was observed.

s This flooding and escape of the grease occurred on the fifth day of the test. Test conditions were established at 250*F/15 psig and decreased to 200*F/10 psig. This condition had boon stabilized for at least four days and therefore had no direct offect on the greaso oscaping. It should be noted that 250*F reflects the same ambient test conditions exhibited in the out-of-containment test reports No. B0003 and F-C3271 where grease reliefs were not installed. Those tests supported the qualification of the Limitorque actuators without utilizing grease reliefs.

The Salem plant temperature profile for a Main Steamline Break shows a peak temperature of 347'F for 3 1/2 minutes dropping to 264*F for 30 minutes. For a LOCA, peak temperature is 264*F for 30 minutes. Each of these accidents produces a peak pressure of 43.2 psig for 170 minutes. Therefore, an external positive pressure is maintainod for the duration of the temperature transient minimizing the pressure differential due to internal pressure buildup

Test report #B0027 was performed to determine the internal temperatures of components in the Limitorque valve actuators due to superheat ambient temperatures. A peak temperature of 385'F at 66 psig was maintained for 3 minutes. The internal temperature never exceeded the saturated steam temperature of 315'F during the test. It was further shown that an ambient temperature of 492*F in excess of 17 minutes resulted in a maximum internal temperature of 315*F.

It can therefore be concluded that the ef fects of ambient temperatures will not result in internal temperatures greater than saturation. As such, excessive internal pressure will not result. Additionally, since there is no airtight / waterproof seal, the positive external pressure differential will be maintained containing the expansion of the lubricant.

The grease utilized for lubrication of the Limitorques at Salem is Mobil Grease 28. This is a synthetic lubricant exhibiting high pressure characteristics at temperatures in excess of 350'F. This grease has a dropping point of about 500*F, which precludes its changing to a liquid state even under superheat conditions. Appreciable expansion of the grease would not occur until a change to the liquid state. Therefore, within the profile of a LOCA/MSLB, pressure buildup would not occur within the Limitorque housicg. Additionally, even should an internal pressure result, the external ambient accident pressure would minimize any not positive pressure differential. Therefore, an expansion of the grease due to increasing temperature causing a breach of the housing seals is not possible.

The greases shown in the Limitorque " Inspection Procedure and Data" sheets have an operating temperature range up to 250*F whereas Mobil Grease 28 is recommended for applications "wherever extreme temperature conditions...are factors. Typical appilcationn where it is has offered advantages are... temperatures near 350*F. Mobil Grease 28 is capable of retaining its consistency under high temperatures and shear conditions over extended periods of operation."

The characteristics of the grease would be a factor in determining the necessity of grease reliefs on the gear case housing. Since the greases utilized during the qualification tests were not identified, a correlation of actual parameters cannot be made. However, based on the Limitorque Data sheet, a lower temperature resistant grease than Mobil Grease 28 appears to have been used.

In summary, as proven by test, the internal temperature of the operator, even under superheated ambient conditions, would not cause the internal temperature to exceed saturated conditions.

This temperature is well within the design limits of the specified grease and would not be cause for it to exhibit appreciable expansion. A precsure buildup sufficient to breach the switch compartment is not possible under postulated accident conditions and therefore grease relief plugs are not required to maintain operability of the valvo.

r_ -

For-your information, the grease reliefs have been installed on all remaining in-containment EQ actuators at Salem, 1.1 accordance with the original Limitorque recommendation to

, utilize the grease reliefs as good engineering practice.

• ['

. RECORD OF VERBAL DISCUSSIONS Subject Discussed:

LIMITORQUE GREASE RELIEFS PSE&G Facility Involved:

ENGINEERING DEPARTMENT Type of

Contact:

Date, Time and Place Personal Conversation of

Contact:

b3TelephoneConversation 4/9/86 2:30 p.m.

Initiated By:

Conference / Meeting Barbara G. Horst Scheduled or Requested By:

Barbara G. Horst N me and Company of Persons Involved:

Joe Drab of Limitorque (804) 528-4400 i

Summary of Discussions: There are some questions about the requirement for grease reliefs on the Limitorque motor operated valves. Joe Drab (their EQ expert) was questioned and his response was that the grease roliefs are solely a good engineering practise. They are not an EQ rcquirement and their absence does not adversely affect the qualification Ctatus of Limitorques.

Future Conference / Meeting Scheduled:

Dates h/A Time:

Places R ported By: Location: Extension:

Title:

&A b! _

l 13 - 7_ 43S O Depa r tme nt:

6AJLiOG4.1 '

( pq EO19

e 9 ATTACHMENT B - 3

, Friday, August 15, immediately prior to the exit meeting, the NRC identified a potential enforcement item regarding the documentation on file not supporting the Salem 10 year maintenance interval of the pressure transducers and RTDs for the Exo Sensor hydrogen analyzers. -This was based on the statement in the Exo Sensor test report, EXO-QTR-101, on Rev. 2 which suggests a qualified thermal age for the platform components of two years (Attachment 1).

However, it is also stated that " specific thermal aging of platform components was not extended beyond two years at this time since the individual components are designated for environmental levels significantly in excess of the levels required for this application."

Exo Sensor-supplied information as noted in Attachment 1 and documented in Table 5-1 of their Maintenance Manual (Attachment

2) indicated replacement of this equipment every 120 months.

The Salem Maintenance Field Directive ( Attachment 2) reflected this 10 year cycle.

Section 3.0 of the test plan (Attachment 4) states that."The basis for in-containment equipment is that routine replacement of the. hydrogen sensor be accomplished at 18 months, at which time the support equipment may be inspected and/or maintained."

Test results indicate that at worst case temperatures (which are lower than indicated in the test plan), the hydrogen sensors at Salem would have a qualified life of at least 6 years. The Salem Maintenance Field Directive has conservatively provided for replacement of the hydrogen sensor every 3 years.

This, along with the attached supplemental test data which was available during the inspection ( Attachment 5) provided sufficient documentation to support the qualified status of the RTDs and pressure transducers utilizing a 10 year maintenance cycle.

In addition, a data package recently supplied (Attachment 6) by Exo Sensor provides further back-up information to support the extension of the qualified life.

All back-up documentation has been incorporated into Environmental Qualification Report, EQ-38, Exo Sensor Hydrogen Analyzer.

Ena sem.a w ,w ,,.~ rer, we , _

e t t

. A7rscM8cai7 /

5.0 TEST RESE25 MyEar SystmR has beEl F_-- M Ol du.M i

Q- 2_% of the SMc;--- cepleted to date (ICCR 9 t= 1 90 days). All i all phases of t==*4 j

fac+4= =1 tests have been sati=0 w.ily perfezmad verifying the j

ability of the analyzer to meet the perfonnance requirements stated

m.  !

TNEZ 5-1 DEMESTRATED PEREGt9N2 I

Envia. d.el M l

Temperature 220-300*F (Short Term) 185'? (L:mg Term)

Pressure 0-60psig m=4dity 5-1004- ,

l See 7 i s=i==4a Radiation (TID) 4x10 to 6.9x10 N e h

Conratimal- '

l 0-64 (Test facility limitation) l H2 Concentraticas Accuracy 2 .2%inHg_0 l-Response 904 12 Seccmds '-

S*=hi14 ty Drift (post ICCA) 10.2% N2 in 30 days i

i Die final evaluatitzt of perfozmance is the *4'=d shiHty of the .

i T m_y the hy& e;=1 u u = a .t.icrt 1

hyd v analyzer to measure and d4In ordar for this to occur each of I within the W *ied toler-ram.

the other active e-;=- ts in the systaa umst also be fenc*4=Nr j

within =:=-4 74ention. Esis incitztes the pressure W ah -ar, RID, .

and mim.e: =:--/ relay assembly. Additionally, the :=M4e auto-c _--T software matic a=14hmtions performed through the =4. -r _

i .ted to be perforning as required to maintain azitrol has been d-systen accuracy. .

! -5.1 PERIUt9H 2 St1WNT -

1 The final qn=14 fie=ticri tmit has been assembled fran ecsyments i

which in sans instances were " aged" and ==4==4-=11y tested in two separate groupings.~ Each however was always tested in the1 funy ==e1=d, as instanedJsomditim. required For eample, indiv4d"=14 zed clevels.

aging of the various u-;= k l

The evaluaticrt of overall systam performance t.at, is based upm omponent performance until Jttly 15, 1981, when the final assembly was ocspletal. Fran this point cn, overall system perfonnance data is available.

i 5-2 .

t t t

Aging evaluations are based upon the relationships for equivalency in EPR1 NP-1558 (page 4-15) which

{ suggests use of the Arrhenius equation relating activation energy to time at a given temperature.

i Utilizing this relationship, the qualification unit hardware was thermally aged to the following:

Component Ecuivalent Ace 9 120*F H 2 Sensors -

l 824 6. years 830 . years a

i Platform Components

• T.~5 years -.+

'-~~

3o

+d, f#ll-c

+. ~p $r.r

\

• See Drawing fl05D006, H2 Sensor c=1ihraticm Assembly " 7/s/tV

] '

I Specific thezmal aging of platform wwwts was not extended l ,

beyond two years at this time since the individual ocuponents ,

l are designed for envire mental levels significantly in excess.of the levels required for this application (see Tables 2-1 and 4-3 -

4 for =p i & data) .  :

The equivalent age is the state of the equipment prior to LOCA simulation with respect to natural aging not -

- .-- - - exposure to hydrogen. Note that no factors were used

! to account for the additional radiation and seismic

- stresses imposed on these assemblies.

On this basis, analyzer performance has been demonstrated i to be as follows:

i i

Pre-LOCA Performance, drift of less than 1.2 percent of reading between required 90 day calibrations. ,

I

' Post-LOCA Performance, deviation from nominal of plus i

0.2 percent hydrogen over the specified 30 day cali-bration period.

J In the pre-LOCA performance while aging, there is a safety factor approaching 67 percent on the accuracy.

For the post-LOCA performance, operation appears nominal.

This is a conservative evaluation since it also includes the errors: inherent in the: gas: chromatograph reference ~

j l j . system. . ,_ .

Upon automatic calibration of the analyzei, performance l

was nominal and accurate to within +.07% H2 for three '

l successive samples.

i f I

' 5-3 I- l

. l

T I d) Cw?= the hy&%.i ccncentraticn in voluan % using the data frca a, b, & c.

e) Perfom a c=14km*4m of the sensor ireind4 .g autcmatic '

sequencing of c=14kh manifold soleid valves, and electreic gain adjustment.

f) Provida alarm contact closures for high hy&.w.. and sewer loss.

g) Provide e=14kmticn gas supply cy14ndam and transport manifold for sensor r=1ikation (available on scme medals) .

h) Provida display of hydrogen partial r ==sm a, hyi.%=1 ccmem-traticm, total pressure and +mmture as de*==4nad by each .

individual detector in the systs.

2.3 PESICAL DESCRIPTICN .

'Ihe Core =4==nt Hydrogen Analyzer System consists of hydrogen sensor -

calikaticm assemblies to be aconted in con *=4-t and a micro-processor /elec*r=1 mntrol asseubly for locaticn in a control roca f envim:mmant. Each microprocessor is connected to the in-sita hydrogen sensor assenblies via elecMr=1 cables - no smaanatic or other mechan- '

ical interfaces are required.

2.3.1 HYDROGEN SENSOR CALIBRATEN ASSDEY

'Ihe imtainment or in-situ portion of the systan provides a self con *=4nad seismically and envia.um.-it 11y qualified monitoring l staticn with the folicwing major m :-=sts Tigure 2-1):

HP  ;= Sensor Pressure Traparhv-me RID N14km*4m_ Gas Supplies (High & Icw Span)

Calibraticn Gas Manifold Exo Sensors drawing 105D006 shows the details of the assembly' which was subjected to the qn=1474raticm test envia.wts.

a) Ee; s sesor: This self-powered ele.L_ - '-- ic=1 cell reacts directly to exposure-to hy&.g . and g= =mtes an elec-M r=1 signal which is directly *r# im=1 to hy&. w .

partial pressure. This measurment is acccuplished without IM to the diluent gases in the sample and without av*===1 elec M e=1 s+4=1=ticn or heating. No reference or reagent gases are required for operaticn. 'Jha elecMc=1 signal is transmittai to the centrol roca via a twistad sb4=1 dad pair ir-L- cable. The sensors were *= =11y pre-aged to a min-imiza equivalent age of 6 years prior to envircrmiental testing.

2-3

._..,,v--..m.__. _

_ _ , , . . - . ._-,_.,_% . _ . . , . . , _ _ , . . , . _ . . _ _.___m._. . . _ _

f I It is necessary to know the ccntaimnent b) Pressure Tran d 'e=*:

pressure to ccavert hydrogen partial pressure to hyd:rgen cen-centraticn. A spuu .i strain gage type pressure trarsducer This m n d wer oper-is ia.w. gated into tk m enclosure.

ates frcan a 10VDC activaticzi signal prended by the m4 w cessor. The retum signal in :nillivolts is a direct measure-ment of absoluta pressure.

22e pressure trar a ver has a range of 0-100 psia, a ccupensated ta perature range of 75'F to +400*F and an cperable temgerature range of -65'? to +450*F. Design proof pressure is 200 psia, manufacturer has qmli*ied design to a shock spa.i a of 100 g, 11 msec, half she wave with:ut damage.

====% t. is made with this 3-wire .

c) luD: A local *menture RID in order to increase the accuracy of the The*menture signal frcm ecm- the pensatial of the hydrogen sensor signal. RIDi is frcm utilized by the m e ; ==ation of the partially upsated signal emanat ng ,

the hy%=. sensor.

c=14hra*4m Gas Su=cly: Two pressurized supplies of certified f

d)

' hydrogen sn.xtures are r.a.co.ded to pe.cnit pariM4e c=14kr=ticn ,

of the hydrogen sensors. Each supply bottle has a pressure re- i ducing regulator and solenoid valve connecting it to a ecum:n g manifold. The regulator drops the supply pressure to 20 psig.

Die solenoid valve is actuated only when a -=1ihmticxi is called -

This routinely occurs every 90 days c

for by the muc -:==er.

at which time the sensor performance is verified at a low span

(^ 2.0%) and high span (- 6.0%) ccmdition. Tha total e=14kra-

' tica sequence takes 17 minutes.

Sr -44'4r- design details of functim=1 e - --=.ts of the Gas t

I Supply man 4 41d are shown below: ,

TABLE 2-1

( < --- amnt Design Parasneters T-- , ;5sture Pressure OCsr l C - - -P- t Pressure 3000psig -

Pogulator -40*F to +500'F Solenoid 300 psi C - r4r q'=14#4ex-Valves 356*F tion to IEEE-323,1974 N/A I RED _.500*F

~

Manual 6000 psi Valves 600*F l

I 346*F N/A Data available to i

Cal Cy14M ar 5360 cycles i

I I

I 2-5

, , A 774cMMEN 7 2 PSEG

~ " " " ~ "

August 13, 1985 To'the General Manager - Salem Operations NUCLEAR DEPARTMENT FIELD DIRECTIVE

' NO. S-C-A910-C FD- 0 3 2 9-80, STATUS CODE 56 MAINTENANCE PROCEDURE BOK MAINTAINING THE ENVIRONMENTAL QUALIFICATION OF THE f EXO-SEMSORS, INC. HYDROGEN ANALYZER Engineering has evaluated the maintenance requirements of the Exo-Sensors Hydrogen Analyzers and determined that specific maintenance is necessary to preserve their qualified status. The specifics of the maintenance

~

requirements and the procedures are contained in this Field Directives.

The Field Directive shall be implemented upon issuance.

Please complete Part B and C of the attached receipt form, then sign and return this form to the Manager - Nuclear Engineering Control.' This Field Directive was discussed with R. Watson of the Salem Chemistry department on July 23,

1985.

i .

D. J.

Assis l General Manager -

Nuclear Engineering BGH: jab Attachment .

C 1 General Manager - Nuclear Engineering 1 General Manager - Nuclear Quality Assurance 1 General Manager - Nuclear Services 1 Assistant General Manager - Nuclear Engineering

. 1 Manager - Nuclear Systems Engineering 1 Manager - Nuclear Plant Engineering 1 Manager - Nuclear

~

Engineering Design

' 3 Manager - Nuclear Engineering Control

! 1 Manager - On-Site Sa fety

1 Site Engineering l 1 Salem Operations Response Coordinator

. _ , _.1 B. G. Horst ____ _ . _ _

1 SAG File ~ ~ ' .

1 TDR t

GR4/1

._. - _ _ _ _ _ _ . _ -_