ML20207H825

Forwards Staff Response to to J Keppler Re Comments on TVA Proposed Cable Test Program

ML20207H825
Person / Time
Site:	Sequoyah
Issue date:	08/25/1988
From:	Richardson S NRC OFFICE OF SPECIAL PROJECTS
To:	Hicks D AFFILIATION NOT ASSIGNED
References
NUDOCS 8808300047
Download: ML20207H825 (8)

Text

. ,

AUG 2 51988 Docket Nos. 50 327/328 Mr. Dallas R. Hicks 6621 Wachese Lane '

Knoxville, Tennessee 37912

Reference:

Your letter to James Keppler dated August- 21,1987, "Corinents -

on TVA's Proposed Cable Test Program dated July 31, 1987" Enclosed is the reply of our staff to your letter referenced above. If you have any further cuestions on this subject, please contact Paul Cortland at(301)492-0754 or myself at (301) 492-3288.

Sincerely.

Original Signed by Steven D. Richardson Steven D. Richardson, Director TVA Projects Division Office of Special Projects

Enclosure:

As stated D . DISTRIBUTION:,

omb 6 : Central file

@8' NRC & Local PDR

$O TVA Reading eb TVA TP Reading a JPartlow

$8 SRichardson oc BDLiaw

@" FMcCoy gg SBlack gg EMarinos HGarg r PCortland f,4 OGC.Rutberg pT JDonohew

• See previous concurrence if OFC :TVA:TP:EB :TVA:TP: ROB :TVA: ROB /Br :TVA:AD/TP :05P:* A:D :  :

NAME :PCortlandiam:HGarg* :EMarinos* :BDLiaw*  : sri son :  :

DATE :8/ 25/88 :8/ '25/88 :8/25 /88 :8/ 25/88 :8/t</88  :  :

OFFICIAL RECORD COPY

RESPONSE TO MR. DALLAS R. HICKS' LETTER DATED AUGUST 21, 1987 "COMMENTS ON TVA'S PROPOSED CABLE TEST PROGRAM, DATED JULY 31, 1988"_

Mr. Hicks comented on TVA letter, R. Gridley to NRC, "Sequoyah Nuclear Plant Units 1 and 2 - Revised Test Program," dated July 31, 1987.

MR. HICKS' "COMMENT NO. 1" TVA stated in page 2. paragraph 2, "In particular, it was apparent that the test voltages utilized were higher than necessary to verify the integrity of the cable. In addition, this testing, which was performed without the introduction of water, was able to detect damage that was insignificant when compared with the type of damage postulated in the technical evaluation report (TER) for pullbys and jaming. Previously, it has been inferred that such damage could be detected by DC high potential testing only in the presence of water."

Mr. Hicks stated that he does not agree "that the test voltages were igher than necessary to verify the integrity of the cable. It just happt +at the <

insulation was already degraded to a level such that lower level of tages ,

could have possibly detected the damage. TVA's statements concerning the '

voltage level and the lack of use of water are misleading and are apparent attempts to justify a position not to use sufficiently high voltages and water imersion of cables as has been recognized as a need for qualification and required by the industry. The fact that a cable insulation fails at a lower voltage and without water is absolutely no reason or justification to lower the voltage and to eliminate water i mersion. Other cable insulations may be unacceptable in integrity of insulation capability but at higher integrity than 4

i

• that of the subject failed cables. They would be detected by the conditions of higher voltage and water imersion."

STAFF RESP 0t!SE TO COMMENT NO. 1 The staff has determined that the proposed test voltage of 240 volts DC/ mil based on the minimum qualified thickness of cable is a reasonable test voltage for determining the gross damage to the cable. If a cable has been environmentally qualified for a certain insulation thickness and TVA is using  :

a cable with higher insulation thickness, it is unnecessary to require a Hi-Pot test based on the higher thickness. Applying the test voltage based on the minimum qualified thickness will provide assurar.ce that the cable has I not degraded beyond a certain threshold. TVA has tested cable under both wet l and dry conditions.

ftR. HICKS' "COMMENT NO. 2".

TVA stated in page 2, para' graph 3. "As a result of the above, TVA concluded j that continued testing, conducted at the original parameters, coulo potentially impact plant safety and, furtherr:cre, could result in the replacement of acceptable cable with no increase in plant safety."

j .

1 l

t i

Mr. Hicks said, "This statement is erroneous and misleading. Cables in the 600

' volt class that are in good condition should withstand 20,000 volts or higher. .

i This is recognized by the industry and is included in IEEE Standards 141 and 241, both being reconenended practice manuals produced by IEEE. TVA tried to make a case in the July 13, 1987, meeting in Bethesda that degradation of cable  :

insvlation after r.anufacture may occur in shipping, handling, and in the  !

installation process, as well as in aging. While this is true, the degradation should not be so great that slightly more than half (10,800 volts) the above l level (20,000+ volts) could not be withstood. It is well understood that the  ;

TVA process for handling and installing cabling was extremely deficient in many areas such as follows: receipt and inspection, storage, hanoling, ,

installation, testing, and quality assurance. Engineering was deficient and there was a general lack of engineering oversight for the installation, check-out, and handling and storage of cabling. One could write extensively .

about the abuses and mishandling of cabling at TVA. For brevity, one can conclude that th* ;rocess was flawed in virtually every area and any subsequent claims by TVA tv the contrary are misleading." ,

STAFF RESPONSE TO COMMENT NO. 2 The staff agrees with the test voltages proposed by TVA as stated in the Response to Comment No. 1.

4 MR. HICKS' "COMMENT NO. 3" TVA stated in the last paragraph of the letter, "In addition to the program j

]

described herein, a cable test program will be established that will be l

consistent with Inspection cnd Enforcement (IE) Notice 86-49. The program will ,

also address low-voltage cables. The purpose will be to demonstrate margins t above the system design basis. This program will be described to NRC within

• six months after restart of SON Unit 2 and will be implemented by the end of the SQN Unit 2 cycle 4 refueling outage."

I J

Mr. Hicks said, "!E Notice 86-49 does not state much and does not define specific requirements, its intended compliance by TVA means little to the i

j j resolution of the cable problems, even though TVA has rade a big deal here and ,

i in the Juij 13 meeting that they will comply with this notice."

Mr. Hicks said. "There is a r-jor problem included in TVA's assumptiens for j getting NRC's approval of a , omise to develop a test program in the future.

The NRC should not be approving a program that is not yet developed by TVA.

4 Restart should not be allowed on such a basis " ]

j STAFF RESPONSE TO COMMENT NO. 3 i

TVA has performed the environmental qualification (E0) testing for silicon j rubber insulated cables,gt Wyle Laboratories ano established their margin of

safety. This test progrant estabilshed the qualificatien life of these cables -

I for 10 years. TVA intends to perform additional E0 testing to establish the 40 l 4 year qualified life. The other cables did not experience any failures during l in-plant Hi-Pot testing. Hence, NRC approval is not based on any future test i

l

,---c - . , - - -

. - - - . . , - - - - - ,_m __r, _ _ _ - , - , - - - - . .- ,

i I

l program, but is based on the testing performed by TVA to date. The periodic testing program as required by EQ and IE-IN 86-49 is not a restart issue, but could be developed by TVA as a post-restart for Sequoyah Unit 2.

MR. HICKS' "COMMENT N0. 4" Mr. Hicks sa- "TVA's statements in Item 1 of Enclosure 1 to the letter are misleading , e do not show conservatism. One cannot make a case for conservatism on an isolated item when the overall process was flawed as discussed in item 2 above '

STAFF RESPONSE TO COMMENT N0. 4 The staff's evaluation is not based on the TVA's claim of conservatism in the conduit installation. It is based on the deficiencies and problems in the cable installation practices.

MR. HICKS' "COMMENT NO. 5" TVA stated in enclosure 1, item 2, "Recent interviews of electricians who had actually performed the cable installation at Sequoyah, conducted by the NRC/ Consultants, indicated a thorough understanding of the cable installation process and the relevant concerns. These interviews substantiated conformance to proper construction practices used throughout the industry and utilization of the engineering approved installation procedures and specification and confirmed the presence of a Quality Control Inspector on all Class 1E cable pulls. In addition, the close working relationship between construction and engineering was outlined including the evaluation made before each Class lE pull to determine the best method of installation, which included determining the direction of the pull and the need for manual assistance at pull points to o reduce tensions. The testimony in these interviews was contrary to Watts Bar Nuclear Plant employee concerns in this area."

Mr. Hicks said. "These statements raise more questions. It is recent. ended that l the NRC obtain the documentation on these interviews and investigate the degree 1 of accuracy therein. Corsidering the overwhelming problems resulting from poor l practices discussed in Item 2 above, one can only conclude that the above I statements are misleading. People can have a thorough understandina during or after the fact of performing tasks, but still do sloppy and unacceptable work.

The measure of acceptability is the end product, and in this case, it is  ;

unacceptable." l STAFF RESPONSE TO COMMENT N0. 5 The staff's evaluation is based on the premise that cable installation practices were deficient and the test program was developed by TVA to demonstrate the acceptability of cables. In addition, the staff ccnducted walkdowns and interviewed the licensee's craf t personnel and does not see any need to obtain TVA's documentation and investigate the issue further.

l l

MR. HICKS' "COMMENT N0. 6" Mr. Hicks said, "In Item 3 of the Enclosure to the letter, TVA begins to state their proposed position to sue a test voltage based on "specific minimum environmentally qualified insulation thickness for the voltage rating and in',ulation type and manufacturer." This position is weak and it does nothing to support TVA's seeking to establish an adequate margin. In fact, when one uses the minimum of anything, there is no existence of margin. This position does r.ot provide confidence that adequate margins will exist and it is not one that should be the foundation for setting an industry precedent. Further, the NRC staff assumes, from TVA's claims, that "minimum environmentally qualified voltage" is acceptable because it was the basis for Sequoyah's Operating 1.icense. I disagree, and I further challenge the NRC and TVA to prove this."

STAFF RESPONSE TO COMMENT N0. 6 The phrase, "Minimum environmentally qualified insulation thickness" means the thickness of the cable previously qualified. Sequoyah used cable insulation of greater thickness to provide additional mechanical strength and dielectric strength. Since the objective of the test program was to determine gross damage to the cable, the staff believes that testing at the voltage level corresponding to the minimum EQ insulation thickness was reasonable and has demonstrated that the insulation has not been damaged beyond a certain level.

TVA's EQ program has been reviewed by the staff and staff's evaluation is contained in NUREG-1232, Volume 2, Part 1, Section 3.2.

MR. HICKS' "COMMENT N0. 7" TVA stated in enclosure 1, item 3, "As such, use of the factory test voltage at this stage in the life of SQN cables may damage cables that are otherwise

• acceptable and inhibit a root cause determination of any test failures."

Mr. Hicks said, "This position has no basis. It entirely misses the reason for i I

testing. Known damage exists at TVA and has been verified by testing, and the very reason for testing is to find the extent of problems and replace the i I

defective cabling, thus obtaining a high confidence level for the quality of installed cabling and its ability to permit the plant to operate and shutdown safely. Any cabling that cannot meet the tests should not be left in the plant, but should be replaced."

STAFF RESPONSE TO COMMENT NO. 7 See response to comment no. 1.

l MR. HICKS' "COMMENT N0. 8",

Mr. Hicks said, "In Enc)gsure 1 of the letter and during the meeting specified in Reference 2, TVA and li's consultants quoted a number of IEEE standards.

Although a senior member of IEEE and several of its societies, I do not believe

f i

l that all of these standards are sufficient by themselves; for example, they I inadequately address 600 volt cabling. Referencing these standards and stating I an intent in meeting them means little and does not solve the problems at Sequoyah. The NRC should carefully review the IEEE and ICEA standards for cabling and should not the voltages in ICEA S-68-516, Revision 7, Table 3-1 j

(IPCEA S-68-516, Enclosed). I STAFF RESPONSE TO COMMENT N0. 8 The NRC staff reviewed many available standards, including TVA's letter of July 31, 1987, and determined that these standards apply to the design and installation of electrical cables. There is no single standard with complete guidance for ascertaining cable damage during installation. Also, in gdnoral, ,

l there 13 no guidance provided for the maintenance test voltage for the low voltage cables. Most of these standards apply to medium and high voltage cables. Therefore, the staff has agreed with the Hi-Pot testing based on 240V DC/ mil on the minimum EQ cable thicknesses as indicated in the staff response to comment no. I above.

MR. HICKS' "COMMENT N0. 9" Mr. Hicks said, "In Enclosure 2 of the letter, TVA refers to the use of PVC cable insulation. Why has the NRC not required TVA to replace the PVC at SQN, especially in view of the disastrous results from the Browns Ferry fire and all of the subsequent PVC restrictions? How can the NRC continue to allow plants 1 to contain PVC without having replacement programs and schedules in place to replace all PVC eventually?"

i STAFF RESPONSE TO COMMENT NO. 9

• As long as the plant meets Appendix R requirements and the functional require-ments of the cable are met, the staff does not require removal of the PVC cables.

MR. HICKS' "COMMENT N0. 10" i

Mr. Hicks said, "In Enclosure 2 of the letter, TVA explains its cable l selection / sampling criteria. More time has been spent on this than would have l oeen taken to test all of the cabling that is subject to damage."

STAFF RESPONSE TO COMMENT NO. 10 Since the cable selection criteria is based on worst case conduits, the sampling process is not random. The selection criteria as5 2 red that the worst case conduits were selected. Based on this the TVA's cable selection / sampling technique was acceptable to'the staff.

MR. HICKS' "COMMENT N0. II"'

Mr. Hicks said, "In Enclosure 2 of the letter and in the neeting specified in Reference 2, TVA tries to make a case for not testing all cables in water. l a

9 L

Thls is contrary to commitments described to me by Mr. Raughley and Dr. Fox in a meeting in Knoxville on June 23, 1987. TVA now states that they will wet the cables on a case-by-case basis, but "It is not intended or expected that the cables will remain submersed throughout the test." To wet cables without keeping them immersed nakes no sense and will have no intended immersion effect. Either TVA tests the cables wet (immersed) or dry. There is no middle ground that could be remotely justified on this issue. Yes, it was a ridiculcus suggestion to wet test cables in all areas, but this latter proposal is even more ridiculour. The NRC should require wet immersion testing to the One degree necessary to resolve the concerns and /lsallow the TVA approach.

solution would be to remove section(s) of cabling and conduit (all intact after having been cut on both ends). Then take it outside the plant, seal one end, fill with water, and high-pot test it. Compare this with the results of dry testing."

STAFF RESPONSE TO COMMENT NO. 11 Although all the conduits could not be wetted, enough cables were wet tested to provide a reasonable assurance that there is no gross damage to the cable because of the cable installation practices at Sequoyah.

MR. HICKS' "COMMENT NO. 12" TVA states in enclosure 2, page 3, item 2, bottom, "If the cables fail the high-voltage test, the NRC will be notified promptly."

Mr. Hicks said, "The NRC should witness all tests to assure that testing is done in accord with TVA's commitment."

STAFF RESPONSE TO COMMENT NO. 12 .

The staff witnessed high voltage tests at Sequoyah. The staff was also briefed regularly by TVA on the results of each test. The staff considers this sufficient to verify TVA's compliance with the test procedures.

MR. HICKS' "COMMENT NO. 13" l Mr. Hicks said, "In Enclosure 2 of the letter, TVA provides cable acceptance criteria and what TVA will do if they have failures. The NRC must reject this plan. If failures cccur, no options should be given. All items must be tested pPior to restart and all failed items must be replaced."

Mr. Hicks said. "TVA's sampling is unacceptable and must be disallowed by the i NRC. The referenced sampling basis, hypergeometric likelihood density function  !

f method, as well as any number of other statistical sampling bases, is not applicable to the subject et hand. Sampling theory, including the referenced I one, assumes that a controlled activity is occurring such as manufacturing of l like items on an assembly'line. Cable installation by different crews, in different configurations of conduit arrangements and sizes, with different l mixes and sizes of cables, and with different plant environments does not allow

classification in the same category as an assembly line type of operation. For this reason, small samples cannot be used to make conclusions about large populations of cabling."

STAFF RESPONSE TO COMMENT NO. 13 The staff's response for sampling is found in the response ts comment no. 10.

Concerning acceptance criteria, the staff will review the specific corrective action based on the root cause of the failure. This was applied to the silicon rubber insulated cables. The other cables did not have any failures.

MR HICKS' "COMMENT NO. 14" Mr. Hicks said, "In Enclosure 3 of the letter, TVA references construction specification G-38, which has been deficient from the beginning. Further, too much detail is spent on selection of conduits rather than in testing all of them. Also, TVA states that cable selection criteria will be documented in a calculation. When will it be done? It would not have to ba done if TVA would eliminate the sampling philosophy."

STAFF RESPONSE TO COMMENT NO. 14 Construction Specification G-38, Revision 8, dated March 17, 1986 was reviewed by the staff and found acceptable. This is documented in the staff's evaluation of Employee Concern Element C0-10900. TVA has also documented the ,

selection of conduits by their letter of November 20, 1987, "Cable Test Program 1 Procedures, Data and Results." The staff's response to the concerns about '

sampling are found in the staff's response to comment no. 10.

MR. HICKS' "COMMENT N0. 15" -

Mr. Hicks said, "TVA has stated that IEEE 383-1974 was not the code of record l with the implication that it did not have to follow it then or now. This is l ridiculous for a plant that had such a long construction time with delay after 1 delay, ope.*ating for a while, now shutdown for serious problems, and not wanting to agree that IEEE 383-1974 is required to solve problems. Who is kidding whom? It would be irresponsible for the NRC to allow TVA to proceed on less than IEEE 383-1974." l STAFF RESrCSSE TO COMMENT N0. 15 It is not necessary for a licensee tu comply with more recent standards than those to which the licensee committed in the FSAR, unless required by the NRC.

No such determination has been made by the staff for IEEE 383-1974.

STAFF RESPONSE TO

GENERAL COMMENT

S The letter from Mr. Hicks cdntained several paragraphs under the heading, "General Comments." The staff felt that they were supplements to the 15 comments discussed earlier and were addressed in the staff's response.

e

, . - - - - -