ML20151H129

Responds to NRC 880202 Ltr Re Unresolved Item Concerning Heat Shrinkable Tubing,Identified in Insp Rept 50-482/87-37. Instrumentation Cable Splices,Control Cable Splices & Power Cable Splices & Motor Connections Addressed

ML20151H129
Person / Time
Site:	Wolf Creek
Issue date:	04/12/1988
From:	Withers B WOLF CREEK NUCLEAR OPERATING CORP.
To:	NRC OFFICE OF ADMINISTRATION & RESOURCES MANAGEMENT (ARM)
References
WM-88-0077, WM-88-77, NUDOCS 8804200268
Download: ML20151H129 (8)

Text

~ .

O W6) LF NUCLEARCREEKOPERATING Bart D. Withers Pms<s.nl and ca v e..euw.omcw April 12, 1988 WM 88-0077 U. S. Nuclear Regulatory Commission ATTN: Document Control Desk Washington, D. C. 20555

Reference:

Letter dated February 2, 1988 from L. J. Callan, NRC, to B. D. Withers, WCNOC

Subject:

Docket No. 50-482: Response to Unresolved Item in Inspection Report 50-482/87-37 Gentlemen Attached is a detailed response to unresolv ed item (482/8737-01) as requasted in the reference. The unresolved item concerns heat shrinkable tubing.

If you have any questions please contact me or Mr. O. L. Maynard of my staff.

Very truly yours, Bart D. Withers President and Chief Executive Officer BDW/jad Attachmen; cc B. L. Partlett (NRC), w/a R. D. Martin (rTRC), w/a P. W. O'Connor (NRC), 2 w/a O

880400006G B00410 ,

PDR ADOCK O'JOO I O \

(

P O. Box 411/ Burbngton KS 66839 / Phone. (316) 364-8831 l

An Es al Omyturwy E%w M F NCVET

t o-i Attachment =to WM 38-0077 ,

Page 1 of 7 April 12, 1988 ,

h RESIONSE TO UNRESOLVED ITEM 482/8737-01 CLNCERMING HEAT SHRINKABLE TUBING

, n PARAGRAPH _2.b.(6)(a): INSTRLMENTATION CABLE SPLICES __

ITEM: Generic -

Raychem ri commends that cut ends of shrink tubing be smooth and free from jarged edges or nicks. Nume rous tubing was identified with jagged edges.

RESPONSI The Raychem Corporation was'contaited concerning the guidelinc included in the product installation pamphlet thit the cut be smooth with no "jagged" edges or nicks. The theory is that additional stresses may be placed on the material during temperature excursions.above the splice material transition level which could occur during a lo3s-of-coolant accident (LOCA) or main steam line break (MSLB). The temperatute excursion may lead to a failure at high stress areas.

Based on the discussions with Raychem, Wolf Creek Nuclear Operating Corporation (WCN00) does not consider thi jagged edge a qualification concern for the reasons provided below.

1. The ststement is a recommendatitn based on theory rather than actual test results. While theory sells us that failures will originate at high stress locations (i .e. at jagged eJges), tests performed by Raychem and the industry reveal that the concern is not backed by test data. No test faliures of Raychem heat shrink tubing have been attributed to jagged edges.

2. During installation kaychem splices are snbjected to temperatures in excess of either a LOCA or MSLB tevironment at Wolf Creek Generating Station (WCGS) . If high temptrature would cause a failure to occur in a splice with severe "jagged" edges, it would most probably occut during the installation process rather than under accident conditions. The fact that the spilces at WCCS survived the high temperature excursion Curing installation without failure provides assurance that the splices would also survive accident conditions.

j Attachment to WM 88-0077 Page 2 of 7 April 12, 1988 1

3.- The smallest diameter of the splice is generally at the edge of the itaychem splice. Hence, the material is closest to its "recavered" diameter at this location and therefore the internal stresses on the material are minimum. This further explains why no failures have been observed on edges which were not perfectly smooth.

ITEM: Penetration ZNI-279

- Seal lengths of 1 1/2 inch were identified on Scheme 3BBIl5EA

- WCSF-200 tubing appears too small for the 0.5 inch X 0.6 bolted connection on Scheme 3BBIl5EA

- A bare conductor was exposed at the cable breakout on Scheme 3BBI15DA

- A shim appears to be needed on one end of a splice in Scheme 3BBIl5DA

RESPONSE

Wyle Laboratory Qualification Test Report No. 17859-02P ( Re f e rence 1) satisfactorily demonstrates the environmental qualification for Raychem WCSF-N splices with seal lengths as short as 0.5 inch for instrumentation circuits. Therefore, seal lengths of 1.5 inch identified on Scheme 3BBIl5EA are not a qualification concern. Test report 17859-02P is reflected in Equipment Qualification Vork Package (EQWP) E-01013 and is summarized in a Ictter dated November 20, 1987 ( Refere1ce 2) . However, the splices have been replaced in accordance with Raychew recommendations.

The splice identified which appeared to have a bolted connection of approximately 0.5 inch by 0.6 inch with a iCSF-200-N sleeve was determined to have been sized in accordance with Raycc am recommendations.

_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ . _ _ _ .]

Attachment t o WM 88-007 7 Page 3 of 7 April 12, 1988 Scheme 3BBil5DA is for BB-FT-446, reactor coolant system (RCS) flow transmitter loop 4. At documented in the Updated Safety Analysis Report (USAR) Table 3.ll(B)-3, this transnitter is classified as NUREG-0588 Appendix E Category C for LOCA and MSLB and Category D for high energy line break (HELB), therefore it is not required to be environmentally qualified.

Additionally, the bare conductor has been identified as a shield wire and is

~

of no nafety concern, a

Notes taken during the NRC inspection indicate the scheme that had the missing shim was 3BBil5AA (not 3BBI15DA) for BB-FT-416, RCS flow transmitter loop 1. As documented in USAR Table 3.ll(B)-3, this transmttter is classified as NUREG-0588 Appendix E Category C for LOCA and MSLB and

?

Category D for HELB. Since failure of this transmitter during an accident is not a Qualif.cacion concern, rework of the splice is not required.

x PARAGRAPH 2.b.( 6)( b): CONTROL CABLE SPLICES

/

ITEM: J-Box S J-HV-000 5 - Seal lengths of 0.25 and 0.30 were identified.

Also, one splice had a split approximately 1/2 inch long through the outer sleeve.

RESPONSE

p

= The WCGS acceptance criteria for seal length in control circuits utilized during the control circuit inspection in late 1987 was 0.5 inch. This figure was canservatively developed based on Wyle tests 17859-02P ( Reference

1) and 17859-02B (Reference 3) after consideration of a SC sample inspection performed on control circuit splices at WCGS. Note the referenced tests

, included splices with seal lengths down to 0.125 inch with acceptable results for WCGS control circuits. EQWP E-1013 currently re flect s the acceptance at 0.125 inch seal length for control circuits.

I The control circuit splice with the 1/2 inch split has been replaced in accordance with Raychem recommendations. The split outer sleeve is considered an isolated case since it was found in one out of over 600 splices examined. In addition, Wyle report 17859-02P (Reference 1) documents the qualification testing of a splice of similar cor. figuration; a severe split in the Raychem sleeva which exposed the conductor. The test g results satisfactorily demonstrates the qualification of the previous splice configuration since the leakage current of 35 ma measured during the testing would not affect the operation of valve SJ-HV-0005 under accident conditions.

v .. .

Attachment to WM 88-0077 Page 4 of 7 April 12, 1988 ITEM: J-Box SJ-HV-0012 - WCSF-115 shrink tubing, without a shim, was installed over a 0.06 inch Conax. This is beyond the use range of the shrink tubing.

RESPONSE

Per vendor drawing, the Conax wire has a 0.064 inch diameter conductor with 0.009 inch thick Kapton insulation, giving an outside diameter of 0.082 inch. The Raychem recommended minimum diameter for WCSF-Il5-N sleeving is 0.110 inch.

Wyle test report 17859-02B (Reference 3) documents testing of WCSF-ll5-N material on Kapton insulated wire with an outside diameter of 0.066 inch. A leakage current of 10 ma was measured during testing. As indicated in EQWP E-01013, control circuits at WCGS can with stand leakage currents of at least 50 to 100 ma. Therefore, test results satisfactorily demonstrate the qualification of the control circuit splice configuration. Wyle test report 17859-02B is re flected in EQWP E-01013.

PARAGRAPH _2.b.( 6)( c} : POWER CABLE _ SPLICES AND MOTOR _ CONNECTIONS ITEM: Hydrogen Mixing Fan Motor CGN03D

- it could not be determined visually whether or not the motcr connections included a molded kit to seal around individ ua conductoro, and the work psckage which documented installation of the connections was not ava.'able for review.

- Shrink tubing was installed over braided insulttion.

- Shrink tubing over one connector had been nicked.

RESPONSE

All motor leads for the hydrogen mixing fans have a braided nomex insulation with a varnish impregnated braided sleeve. During the environmental testing of the hydrogen mixing f r.n s , the motor connections were directly exposed to the accident environmont. The re fore , the three items noted above which address nuclear grade Raychem aplices are not qualification concerns at WCCS.

Although Raychem splices on CGNO3D connections are not required for qualification, Wyle test report 1/859-02B ( Re ference 3) documents the testing of Raychem material applied to both impregnated and non-impregnated braided wires. The maximum leakage current measured during the testing was 157 ca which is acceptable for the power circuits of interest.

__i

Attachment to WM 88-007 7 Page 5 of 7 April 12, 1988 The work request which documents the installation of the hydrogen mixing fan motor CGN03 D connections utilizing NMCK-3V-35-00 motor kits in accordance with design documents is available for review at WCCS. In addition, although not a qualification concern, the nick in the shrink tubing has been repaired.

ITEM: Penetration ZNE-268 (Containment Cooling Fan)

WCSF-500 shrink tubing used in this penetration was too small for the lug and bolt size and for the 1.15 inch cable.

- The seal length over all spilces was approximately 1 inch rather than 2 inches as recommended by Raychem.

- Bolt pads were not used as required even though bolt length and tang widths for all connections were greater than 0.87 inch.

RESPONSZ:

Wyle test report 17859-02B (Reference 3) documents the testing of' Raychem WC3F-N material with hold-outs of 4.14. The hold-out is defined to be the ratio of the e f fective diameter of the splice configuration to the fully recovered diameter of tne Raychem material used in the splice. No failures were observed during the test even though the 0.5 inch long No.10 bolt used in the test presented a sharp edge to the shrink material. Leakage currents, although st critical to these pcaer circuits, were a maximum of 15 ma at 132 VAC. While this test demonstrates the adequacy of using Raychem material with large hold-outs, because the leakage currents were measured at 132 VAC versus the nominal 277 VAC to ground at WCGS, the minimum wall thickness of 0.065 inch for power cables allowed by ICEA Standard S-66-524 has been used to calculate a more conservative allowable hold-out for the low voltage power splices. By solving the equation in Section 2.4.5 of the Raychem WCSF-N Application Guide (Reference 4), an acceptable hold-out of 3.43 has been calculated. Thia hold-out of 3.43 yields 1.949 inches for the maximum effective outside diameter for power circuit applications for WCSF-500-N.

With the maximum tang width of 1.76 inches used on the 500 MCM cable, a maximum equivalent diameter of 1.949 inches yields the maximum acceptable bolt length of 2.055 inches. Using the minimum wall thickness calculated from ICEA Standard S-66-S24, the maximum outside dimension is 2.185 inches.

This acceptance criteria is considerably greater than the Raychem criteria used during the NRC inspection. Also, the specific penetration examined, 7NE-268, is located outside containment in room 1410. The only harsh environment parameter for this toom is radiation, and e en then the accident dosage is only 1 6E6 rads. Since this radiation level is orders of magnitude below the tested level of 2.0E8 rads aad since the splices are not exposed to high temperatures during accident conditions, it is WCNOC's position that the splices in penetration ZNE-268 will not be adversely affected by the LOCA environment.

Attachment to WM 88-0077 Page 6 of 7 April 12, 1988 Vyle test report 17859-02B (Reference 3) documents the testing of Raychem VCSF-N splices with seal lengths as short as 0.125 inch for low voltage power circuits. The test results adequately demonstrates the qualification of the splices in low voltage power circuits with seal lengths of approximately 1 inch at WCCS.

Raychem recommends the use of a bolt pad when the bolt length or tang width exceeds 0.87 inch. The purpose of the pad is twofold 1) to keep the adhesive out of the threads in the bolted connections and 2) to provide additional assurance that the splice material will not fail at high stress areas during accident conditions.

Adhesive in the threads of the connection is not a safety concern and need .

not be addressed here. The following is justification for the WCCS installation in regards to item 2:

1. Using a long bolt (or wide tang) without a bolt pad may increase the stresses in the splice material leading to failure during accident conditions with elevated temperatures. However, it should be noted that the 0.87 inch dimension recommended by Raychem was not established based on testing, and was essentially arbitrary. As mentioned above, Wyle has successfully tested Raychem WCSF-ll5-N material on a 0.50 inch long No.10 bolt. This configuration, with its eniculated hold-out of 1.414 over the untrimmed bolt is at least as severe as using WCSF-500-N material an . 0.87 inch long 3/8 inch bolt, which has a calculated hold-out of 1.74.

2. During installation, the Raychem splices are subjected to temperatures in excess of what will be seen during either a LOCA or MSLB environment at WCCS. If high temperatures cause a failure of splices to occur in those situations with bolts longer than 0.87 inch and without bolt pads, it follows that WCGS should have had installation problems in those splices where the Raychem criteria were not strictly followed. Lack of installation ,

problems provides additional assurance that the WCCS installations are adequate.

3. At WCGS, low voltage power splices in e?ictrical penctrations are generally separated physically f rom each other and from ground. A failure of the splices, even to the point of exposing the c o nd u c t.o rs , would not be a concern since expected leakage currents have been shown to be acceptable for WCCS applications. Note the containment cooling fan motors were tested with splices directly exposed to the harsh LOCA environment and, although measured IR's were low, the motors performed their safety functions acceptably.

Attachment to WM 88-0077 Page 7 of 7 April 12, 1988

References:

1. "Qualification Test Program on Raychem Nuclear Cable Splices, Okonite Tape Splices, and Kerite Splices as Installed on Various Wire I Insulations at Commonwealth Edison Company's Zion, Byron and Braidwood Nuclea* Generating Stations", Wyle Laboratories Test Report No. 17859-02P, Revision A

2. Letter VM 87-0309 dated November 20, 1987 from B. D. Withers, WCNOC, to NRC

3. Qualification Test Program on Raychem Nuclear Cable Splices, Okonite Tape Splices, Kerite Tape Splices, Scotch Tape Spilces and Amp Butt Splices as Installed on Various Wire Insulations at Commonwealth Edison Company's LaSalle, Dresden, and Quad Cities Generat!ng Stations", Wyle Laboratories Test Report No. 17859-02B, Revision A, August, 1987

4. Raychem WCSF-N Application Guide, Revision 1, August 1983 s

l i

l

{

w