IE has reviewed the roort and believes it accurately ...

...

FEB C 19M

MEMORANDUM FOR: Robert L. Baer, ChiefEngineering and Generic

Communications BranchDivision of Emergency Preparedness

and Engineering ResponseOffice of Inspection and Enforcement

THRU: Alexarider W. Dromerick, Chief, Section AEngineering and Generic

Communications BranchDivision of Emergency Preparedness

and Engineering ResponseOffice of Inspection and Enforcement

FROM: Vincent D. Thomas, Sr. Instrumentation & Controls EngineerEngineering and Generic

Communications BranchDivision of Emergency Preparedness

and Engineering ResponseOffice of Inspection and Enforcement

SUBJECT: COMMENTS ON FRC MEETING REPORT ON ROCKBESTOS CABLE ISSUEIN IE OFFICES - DATED JANUARY 5, 1984

Cognizant people from RIV, NRR, Franklin Research Center (FRC), and IE meton Thursday, January 5,1984 in the East West Towers West Building in Bethesda,Maryland to discuss outstanding issues related to the Rockbestos Cable Companyproduction program.

The enclosed report from FRC is a compilation of the minutes of the January 5,1984 meeting. IE has reviewed the roort and believes it accurately describesthe results of discussions on all items of concern addressed during the meeting.

With respect to the recommendations (items of concern) provided on page 5of the FRC report, our views on each item of concern follow.

Item of Concern No. 1 - Retesting:

Rockbestos has agreed to revise its current retesting and sampling methodologyto verify that cable qualified more closely reflects the requirements specifiedin the latest applicable ICEA (formerly IPCEA) standard. Rockbestos hasalso agreed to submit its mcdified retest program to the Region IV VendorInspection Branch for review and comment. Uldis Potapovs of Region IV hasagreed to followup on this item to assure that Rockbestos fulfills theircommitment.

8402210483 840206CF SUBJ

CF

*.

Robert L. Baer -2- FEB 6 1984

Item of Concern No. 2 - Conductor Insulation Nonconformances:

Followup action by IE of the two cable reels in question, identified asG-34789 and G-32923 and shipped to Beaver Valley Unit 2 and Nine Mile PointUnit 2, respectively, revealed that five (5) additional cables reels wereproduced from the sane production run as cable reel G-34789 and were alsoshipped to Beaver Valley Unit 2. We found that eight (8) additional cablereels were produced from the same production run as cable reel G-32923 andshipped to Nine Mile Point Unit 2. According to Stone & Webster, approximately13,000 feet of cable comprise the six reels at Beaver Valley Unit 2, andapproximately 19,000 feet of cable are involved in the nine reels of cableat Nine Ilile Point Unit 2.

With respect to findirgs resulting from followup action on the reels in questionat Beaver Valley Unit 2, only ten (10) cable cuts out of 986 cable cuts fromthe (six) reels in question, are being used in safety-related electricalcircuits. However, all of the 10 cable cuts are used in areas considered as

mild environnent. According to both Duquesne Light Company and Stone & Webster,cable reel G-34789 does in fact meet all requirements as specified in applicableICEA Standa rd S-66-524. Additionally, the licensee stated that since thisICEA standard is the document applied to determine adequacy of this specificcable reel, Beaver Valley in turn, accepted the cable despite the minor deviationfrom the Rockbestos standard value which is more stringent than the ICEA standard.To close cut this specific item of concern, Region IV is preparing a formalrequest to Stone & Webster to confirm this acceptance action to Rockbestos.Accordingly, we believe that upon receipt of this confirmation from Stone &Webster, the issue in question concerning the six cable reels at the BeaverValley Unit 2 can be resolved.

With respect to cable reel G-32923 and the eight sister reels shipped to NineMile Point Unit 2, Stone & Webster and the licensee stated (as in the case atBeaver Valley) that the parameter values for cable production as specified inthe applicable ICEA S-66-524 Standard were the governing values for acceptingcables and not those values specified by the manufacturer (Rockbestos). As aconsequence, all electrical and physical parameter values obtained in each ofthe three retests of samples from cable reel G-32923 do in fact meet allapplicable ICEA requirements. On this basis, Stone & Webster and the licenseehave accepted cable reel G-32923 and the eight related cable reels as suitablefor safety-related applications at Nine Mile Point Unit 2 station. As in thecase at Beaver Valley, Region IV will also request a formal response from Stone& Webster to confirm this position.

Additionally, subsequent to receiving the above letters of acceptance fromStone & Webster, FRC will include these findings in an addendum to subjectreport identified as FRC-FC 5569-306.

Briefly, Report FC 5569-306 provides results of the FRC review of documentationand retest data referenced by the Rockbestos Company in response to the NRCinvestigation that began in May 1982.

.

FEB 6 1984Robert L. Baer -3_

Item of Concern No. 3 - Insulation Resistance Test Voltages:

We concur in the FRC recommendation, " acceptable," in that Rockbestos indicatedthat low voltage insulation test equipment is being used only on certaincoaxial types of cables produced. We believe that this item of concern cannow be closed.

Item of Concern No. 4:

Rejacketed Cable: On the basis of the judgment provided by the NRC/FRC cableconsultant, E. Eich, that says: apparent deviations f rom specification values,would not "substantially change the ability to perform the intended function,that is, if cables with conforming jackets are suitable for the service; cableswith these jackets are substantially equally suitable," we agree that rejacketingof cable, as discussed, is acceptable for the service intended.

Environmental Qualification: Finally, with regard to environmental quali-fication of reworked or rejackced cable, we agree with the FRC suggestionthat a feasibility study be implemented by the NRC to detennine if these typesof cable need to be individually requalified.

We reconmend that NRR/EQB pursue the feasibility study effort of environmentallyqualifying rejacketed cable.

Vincent D. ThomasSr. Instrumentation & Controls EngineerEngineering and Generic

Communications BranchDivision of Emergency Preparedness

and Engineering Response, IE

cc: A. W. Dromerick, IE DISTRIBUTION ~U. Potapovs, RIV ctCSI. Barnes, RIY DEPER R/FT. Le, IE EGCB R/FA. Masciantonio, NRR VDThomas R/FV. S. Noonan, NRR ELJordanG. Toman JGPartlowFranklin Research Center SASchwartz20th and Ben Franklin Parkway VDThomasPhiladelphia, PA 19103

*SEE PREVIOUS CONCU NCES*EGCB:DEPER:IE EPER:IE*

VDThomas Vg g erick1/31/84:mkm /84

Robert L. Baer -3-

Item of Concern No. 3 - Insulation Resistance Test Voltages:

We concur in the FRC recommendation, " acceptable," in that Rockbestos indicatedthat low voltage insulation test equipment is being used only on certaincoaxial types of cables produced. We believe that this item of concern cannow be closed.

Item of Concern No. 4:

Rejacketed Cable: On the basis of the judgment provided by the NRC/FRC cableconsultant, E. Eich, that say.s: apparent deviations from specification values,would not "substantially change the ability to perform the intended function,that is, if cables with conforming jackets are suitable for the service; cableswith thesii jackets are substantially equally suitable," we agree that rejacketingof cable, as discussed, is acceptable for the service intended.

Environmental Qualification: Finally, with regard to environmental quali-fication of reworked or rejacketed cable, we agree with the FRC suggestionthat a feasibility study be implemented by the NRC to determine if these typesof cable need to be individually requalified.

We recommend that NRR/EQB pursue the feasibility study effort of environmentallyqualifying rejacketed cable.

Vincent D. ThomasSr. Instrumentation & Controls EngineerEngineering and Generic

Communications BranchDivision of Emergency Preparedness

and Engineering Response, IE

cc: A. W. Dromerick, IE DISTRIBUTION-U. Potapovs, RIV I)LsI. Barnes, RIV DEPER R/FT. Le, IE EGCB R/FA. Masciantonio, NRR VDThomas R/F

ELJordanC. Toman JGPartlowFranklin Research Center SASchwartz20th and Ben Franklin Parkway VDThomasPhiladelphia, PA 19103

EGCB:DEPER:IE Wy :IE)

VDThomas V#C AWDP merick1/31/84:mkm q /84

.

..'

MEETING REPORT

FRC Project 5569-306 Task 3006

Subject: Rockbestos Cable Company Manufacturing Testing

Location: Room 550, East West / West Towers Bldg.,Bethesda, Md.

Date and Time: January 5, 1984, 9 am

Attendees: RockbestosRobert J. GehmEugene J. D' AquannoGeorge G. Littlehales

NRC: s

Robert BaerA. W. DromerickVincent Thomas ,'Uldis PotapovsIan Barnes*Thomas Le*Armond Masciantonio*

FRC:Gary Toman s'Edward Eich (Consultant)

*Part-time attendance

Purpose of Meeting

To discuss with Rockbestos personnel recommendations contained in FRCReport F-C5569-306, " Evaluation of Rockbestos Cable Manufacturing Testing" andopen questions concerning information contained in the letter dated November9, 1983 from George Littlehales of Rockbestos to Gary Toman of FRC (attachedfor reference) .

Discussion

Since the questions concerning the Rockbestos letter of November 9, 1983overlapped the recommendations from FRC Report F-C5569-306, the meeting beganwith a discussion of the FRC/NRC concerns in regard to the letter. The letterwas in response to questions contained in a letter from G. Toman to G.Littlehales dated October 27, 1983 (also attached for reference) .

-1-L&

_ _ Frankhn R, es,earch Center

.

.'

FRC Request 1 from October 27, 1983 letter

"For those reels of cables experiencing XLPE insulation testnonconformances (i.e., Stone and Webster Engineering Corp. [SWEC), BeaverValley Unit 2, one reel; SWEC, Nine Mile Point Unit 2, two reels; andSWEC, Shoreham Nuclear Power Station, one reel), provide a description ofthe cables involved and a list by reel number and purchaser of all reelsmanufactured in the same manufacturing runs that were purchased fornuclear safety-related service."

The Rockbestos response to this request described the results of the

retests on the subject cable and stated that the results met all requirementstypically specified for nuclear safety-related cable and that Rockbestos didnot believe that identification of nuclear cable employing single conductorsfrom the same run was necessary.

/

Mr. Toman discussed his review of this response and his comparison of theresults contained in the response with the data sheets for the Rockbestosretests which were obtained by FRC during a meeting at the Rockbestos plant onAugust 9 and 10, 1983. Mr. Toman noted that for one of the cables (BeaverValley Reel G34789) , the stability factor exceeded the Rockbestos " guaranteed"value (0.6 vs. 0.5) and that for all but one of the other five reels describedin the response, multiple retests were required to obtain passing values forEM-60 electrical water absorption or tensile strength or elongation tests. Itwas also noted that one reel (Nine Mile Point Unit 2 Reel E024B-774) hadf ailed the UL vertical VW-1 (UL-83) flame test and then subsequently passed anICEA S-19-81 flame test and was deemed acceptable.

Rockbestos responded that they believed that the problems with theinitial retests were related to anomalies in the performance of the tests orpreparation of the test sample rather than to a problem with the materialsalthough, even at the time of testing, it would have been difficult to provethat either the test rig or the test sample was the cause. They also notedthat at no time did a test result exceed an IPCEA S-66-524 limit.

With regard to the change in flame test procedure, Rockbestos stated thatthe more stringent UL vertical VW-1 (UL-83) flame test was inadvertentlyplaced on the test form as a requirement and that this was not required forthis type of cable by standard or purchase order and that the ICEA S-19-81 wasthe correct test to perform.

Mr. Eich stated that variations from the values normally recorded for thecables would be cause for concern. He believed that the Rockbestos" guaranteed" values were more representative of the cable parameters than theless restrictive 1PCEA S-66-524 values. Rockbestos stated that the Rockbestos" guaranteed" values were values that the cable would generally be able to meetand were based on Rockbestos' knowledge of the cable rather than on firmstatistical data. Mr. Eich reviewed the data for the cables of concern (one

-2-AM_ _ Frenkhn, Research Center.- , n r._ -

.

'

.

reel of cable, G33955, eas rejected by the customer and is no longer ofconcern). He determined that the problems with elongation and tensilestrength were of little long-term concern. However, Mr. Eich maintains thatthe test f ailures in the EM-60 tests are significant in that the EM-60 testdoes give some indication of the ability of the cable to function with timeand that the deviating values appeared to be beyond the range that would beexpected for this type of cable.

FRC Recuest 2 from October 27, 1983

"For those cables that did not pass EM-60 accelerated water absorptiontests, provide an engineering justification that these cables areacceptable for nuclear safety-related service. This justification shouldinclude those cables that passed retests as well as those that did not."

The Rockbestos' respense to this request was that the test method was atf ault rather than the samples being tested and that the samples that failedthe test w2re those that had been allowed to intermittently touch the side ortop of the water bath chamber while readings were being taken.

Mr. Toman asked if Rockbestos referred to the submerged portion of thecable when indicating that the samples touched the tank. Rockbestos statedyes. Mr. Toman then asked why this should make a difference since the waterin the test was meant to provide a conducting path between the conductorinsulation and the tank. Rockbestos stated that the water bath had resistanceand dielectric strength and that results could change when the tank wastouched by the cable. Rockbestos cited a test in which some cables wereallowed to touch the tank during some tests and others were not. All thecables that did not touch the tank passed, whereas some, but not all, of thecables that touched the tank had scme problems in the EM-60 test results. Thereason why some passed and others did not when touching the tank could not bedetermined.

Mr. Eich did not believe that the results of this test conclusivelyproved that improper test methodology was the cause of the EM-60 failure.Rockbestos stated that since instituting procedures that assure that no cabletouches the tank during testing, only two instances of an EM-60 failure haveoccurred, which is a major reduction in failure rates.

FRC Recuest 3 f rom October 27, 1983 lettet

"For cables that are to be reworked or rejacketed (i.e., those for SWEC,Nine Mile Point Unit 2) , provide a description of the testing performedprior to rejacketing that provides assurance that no damage has occurredto the conductors or conductor insulation during jacket removal andhandling."

Rockbestos stated in their written response that no testing of theconductor insulation is performed prior to rejacketing. After rejacketing,dielectric strength, insulation resistance, and conductor resistance tests areperformed.

-3-/=. . FreN:n Research Center

.- .r., .

v... .

Mr. Toman asked what proof was available that the conductor insulationwas not damaged or significantly stressed during rejacketing such thatoriginal qualification data were not invalidated.

Rockbestos stated that the rejacketing of the cable was less stressful tothe conductors in that fewer operations were performed on the conductor andthat the twisting from the cabling procedure was not repeated and that theydid not believe that a significant increase in stresses was added. Withregard to testing, Rockbestos stated that no production test existed thatwould expose minor damage done to conductor insulation either in initialjacketing or rejacketing. Significant damage would be detected by theproduction tasts.

FRC Request 4 from October 27, 1983 letter

"The description of the resolutica status for cable provided to SWEC,Nine Mile Point Unit 2 (ten reels total: six reels, Hypalon jackettesting nonconformances; two r eels Hypalon testing and XPLE insulationtesting nonconformanceL; and two reels, XLPE insulation testingnonconformances) is not clear. Please provide a listing by reel of thecable description, the initial problem, and the resolution of theproblem. Please provide this information for the six additional reels ofcable identified as needing rework."

Rockbestos responded with the status (see full respcnse in Lttachment) ofthe cables.

Mr. Toman asked if Rockbestos had any updates on the status of thecables. Mr. Littlehales responded that for the six additional reels of cableassociated with reels rejected by the Stone and Webster EngineeringCorporation for Nine Mile Point Unit 2, five had been rejacketed and were inthe process of being retested and one (Reel G34350) had been rejected and wasto be replaced with new cable.

Further Concerns from FRC Report F-C5569-306

1. Retesting - Mr. Toman asked if a formal policy existed on retestingof cable when a test failurc was recognized. Mr. Littlehales statedthat generally two retests would be performed but that formalinstructions for retesting and rejecting cable did not exist.

2. Use of Low Voltage Insulation Resistance Measuring Devices -Rockbestos stated that confusion existed with regard to use of lowvoltage insulation test equipment. This equipment was used only fora particular type of coaxial cable used with sensitive electronicequipment. For all other cable, high voltage insulation testequipment is used. The coaxial cable is tested at low voltagebecause previous use of high voltage equipment had lef t residualcharges on the cable which had resulted in damage to the electronicdevices when connected to the cable.

-4-& ==. .. ,Frankhn ,Research Center

4*

. .4

Recommendations Resulting from the Meeting

1. Retesting - FRC recommends that formal policy be adopted byRockbestos for retesting and rejecticn of conductors and cable basedon the IPCEA S-66-524 requirements for sampling, retesting, andrejection.

\ 2. Conductor Insulation Nonconformances - With regard to those cables,] having slight deficiencies during retesting in tensile strength andelongation, there is little concern for long-term operability becausethe tensile strength and elongation required of the insulation cannever approach the as-tested levels without first breaking theconductors.

With regard to the two cables with conductor insulation that did notpass EM-60 accelerated water absorption tests, concern remains thatthe insulation could deteriorate with time. The EM-60 test gives anindication of the ability of the insulation to withstand moisture by-testing insulation properties when submergei for two weeks.Therefore, Reels G-34789 (Beaver Valley) and G-32923 (Nine Mile Point2) and cable produced in the same manufacturing runs should either besubjected to a program of in-situ testing or replaced unless thecable is not used in safety-related circuits or it is not subjectedto harsh environments while in service.

3. Insulation Resistance Test Voltages - Rockbestos' response that lowvoltage insulation test equipment is used only on certain coaxialcables is acceptable.

4. Rejacketed Cable and Environmental Qualification - As a result of the

meeting, it was determined that environmental qualification tests onthe rejacketed cables would not conclusively prove that occasionaldamage had occurred to the conductor insulation during rejacketingbecause results would be highly dependent on the sample chosen. Itwas also concluded that tne chance of conductor insulation damage wasprobably smaller during rework than during original cabling andjacketing although rejacketing causes an additional increment of

stress to the conductors. FRC does not recommend that LOCA tests beperformed on the specific cables involved in this inspection;however, FRC recommends that the NRC give consideration to LOCAtesting of rejacketed cable since the rejacketing process entails anincremental increase in stress to the conductors that has not beenincluded in LOCA test results to date. It is recognized that suchtests are out of the scope of this investigation.

-5-A=-

. F*enkhn Res,eerth Center,