ML20197C130

Affidavit of M Walsh,Advising That Statistical Sampling Being Performed & Proposed for Facility Inappropriate. Applicant Reliance on Statistical Sample Will Not Identify Problems W/Pipe Supports.Certificate of Svc Encl

ML20197C130
Person / Time
Site:	Comanche Peak
Issue date:	05/06/1986
From:	Mary Walsh Citizens Association for Sound Energy
To:
Shared Package
ML20197C088	List: ML20197C079 ML20197C105 ML20197C130
References
OL, NUDOCS 8605130237
Download: ML20197C130 (33)

Text

--

$! $ $1 v r, \\

UNITED STATES OF AMERICA /y NUCLEAR REGULATORY COMMISSION [7 c F-f Q'ot BEFORE THE ATOMIC SAFETY AND LICENSING BOARD y

1 V .

- L; In the Matter of I Docket Nos. 50 45 I and 50- m /T~~ g '

TEXAS UTILITIES ELECTRIC i COMPANY, et al.

~ - -

t t (Application for an (Comanche Peak Steam Electric i Operating License)

Station, Units 1 and 2) l AFFIDAVIT OF CASE WITNESS MARK WALSH In this affidavit, I will discuss: why I believe that the statistical sampling being performed and proposed for Comanche Peak is not appropriate ot :pplicable (based primarily on what I have learned through my review and analysis of cable tray supports and pipe supports at Comanche Peak); and why I believe that the Applicants' reliance on a statistical sample will not identify all of the problems with pipe supports, cable tray supports, conduit supports, other designs, or the plant in general, and will not resolve the Board's doubts as expressed in its 12/28/83 Memorandum and Order (Quality Assurance for Design) and later Orders.

In addition, I have done some sampling of my own - not statistical, J

but sampling which I believe is nonetheless valid for these proceedings.

This is a sampling of some of the attitudes of the reviewers of the plant I (Applicants and their numerous internal organizations and consultants, and the NRC Staff and its various special teams) -- attitudes which, I believe, j

demonstrate a pattern, and which have allowed and encouraged design problems l-to develop at Comanche Peak (which obviously has carried over to f construction and QA/QC problems as well).

i 8605130237 860506 I PDR ADOCK 05000445 G PDR

o ,"

.y V

~

From the relatively small sample of pipe supports which Jack Doyle and I worked on at Comanche Peak br which we became familiar with, one problem which appeared obvious was that there waa' a la'ck of supporting data and analyses (including statistic'al analyses [being perfoned. After we recognized problems and reported them to Texas Utilities officials or their sub-contractors, nothing was done to address and correct the problems. In

.-6 .

fact, as stated-in my testimony and as is still apparently the case at

~'

Comanche Peak (see page 2, Supplementary Testitaony of Mark A. Walsh, CASE Exhloit 659H, 7/29/82, admitted at Tr. 3198):

". . . this is the first nuclear plant at which I have worked that I was ever directly told to deliberately disregard what I-know is

~

technically right. It's the 11rst job that I've ever held'where I have been told to deliberately ignore or go against basic engineering fundamentals. .."

This was one example of the attitude of Applicants and the NRC Staff, (in addition to concerns regarding the design of the plant).

- As a matter of fatt, af ter Mr. Doyle and I testified in the l'icensing hearings in July and September 1982, both the Applicants and the NRC Staf f

~.

~

stated that there was no technical merit to our concerns. This and other o

aspects regarding pipe supports are addressed rather thoroughly by Mr. Doyle in his affldavit 'Ew ch is also attached to this pleading, and I will not

^

addrese them here except to state that I agree with his overall assessment.

'The' primary point which I wish to make here is that this attitude (of both m

the Applicants and the NRC Staff) is itself another example of attitude,

~

which is already in the record.

At the time of, and immediately following, the May 1983 hearings which were held on the NRC's SIT Report (as discuased in more detail in Mr.

Doyle's affidavit), the attitude of the Applicants, the NRC Staf f, and the NRC's Special Inspet. tion Team (SIT) was that there were basically no i 2 l

l

I s

problems at the plant, and this is the attitude with which they all approached potential problems at Comanche Peak. Now there was an additional reviewer (the SIT) to be added to one more example of the attitude of Applicants and the NRC Staff, which (in addition to concerns regarding the design of the plant) is one more example of attitude.

Then Cygna came in with their supposedly independent review and they also could not find any problems which were so significant that they could not be dismissed. I got involved with the cable tray supports as a result of my review of Cygda's Phase 1 and 2 Report in February of 1984 (although hearings on the cable tray supports was not actually held until May 1984).

Cygna's conclusion on the cable tray supports was (see Board Exhibit 1, 2/24/84 hearings, Draft Final Report, Independent Assessment Program for Comanche Peak Steam Electric Station, Prepared by Cygna Energy Services, 11/5/83, at page 4-12; see also Cygna's Final Report, 10/12/84, which stated the same conclusions, at page 4-14):

l "In conclusion, Cygna has found that the cable tray supports within the Cygna scope are adequate. The design and revision process are, however, difficult to follow; Cygna suggests that a set of standard instructions be prepared for design, revision and

, review of cable tray supports. Such a set should include a set of

! justified assumptions and would ensure uniformity across the I

design process."

When CASE finally received [he calculations on discovery for just a few of the supports Cygna had looked at, and I was given a very limited amount of time to look at them, I found errors in design (some of which were later shown to be very significant) that Cygna had not recognized, although Cygna had reviewed cable tray supports which they stated " represent 43% of the cable tray supports for the plant" (Draft Cygna Report, page 4-11; Final Cygna Report, page 4-13). Further, during the 2/21/84 hearings, Cygna Witness Mr. Ward (during cross-examination by Mr. Doyle and a follow-up i

l 3

clarification question by Judge Bloch) testified regarding the team of Cygna people working on the Cygna Report effort (Tr. 9784/10-16):

". . . They were real tigers. They really wanted to go af ter this one and did an exceptionally good job. My concern to them was that it wasn't in the best interest of anybody who would receive this report if we spent an awful long time trying to judge which method was better than the other when both methods for design might be adequate and perfectly acceptable to assure the safety of the plant."

During the May 1984 hearings, 1 called into question the adequacy of Cygna's review of the cable tray supports, both during my cross-examination of Cygna and in my testimony (see generally Transcripts of May 1, 2, and 3, 1984, hearings, 1r. 13126-13794; see also Revised Testimony of CASE Witness Mark A. Walsh, accepted and bound in following Tr. 13731), 1 questioned the attitude and bias of Cygna towards the utility (testimony pages 1 and 2), as well as the scope of their review. During, and by the end of, the May 1984 hearings, only my testimony and cross-examination raised any question as to the adequacy of design of cable tray supports at Comanche Peak (see my prefiled testimony and generally the transcripts of those three days' hearings, referenced earlier in this paragraph).

There are numerous examples of the inadequacy of Cygna's review. For example, Cygna did not look at whether or not the cable trays within the l

containment building required a decrease in the allowables; Cygna did not consider that as part of their scope (Tr. 13226-13229). It is my 1

understanding that the calculations used for the Safeguards Building were also erroneously used in the containment building; but whether or not this is correct, the point is that Cygna would not even have caught such a significant design error because it was outside the scope of their review.

1 So here was yet another reviewer of Comanche Peak (Cygna) to be added as one l more example of the attitude of Applicants and the NRC Staf f, which (in 1

i 1

4 1

s I

addition to concerns regarding the design of the plant) constitutes one more

.' example of attitude.

Another example of the Applicants' and Cygna's attitude was that Applicants changed their FSAR to accommodate design errors found in calculations regarding the flexibility vs. the rigidity of cable tray

, supports (Tr. 13372 et seq.). Cygna apparently did not know that Applicants 4

had not followed their FSAR commitment. The Applicants' changing of the FSAR because the design was not in accordance with the FSAR, and their failure to Inform Cygna of this fact, provides another example of Applicants' attitude.

It is important to note that not the NRC Staff, Cygna, nor the Applicants had realized that they had very serious problems in the design of the cable tray supports prior to my testimony and cross-examination of the Cygna witnesses in May 1984. In fact, Cygna had not recognized any potentially significant problems (in cable tray supports, in pipe supports, or in anything else in the scope of their review) at the end of the May 1984 hearings. And Cygna's counsel Mr. Pigott confirmed emphatically that Cygna had found no item that would rise to the level of a 10 CFR Part 21 or a 10 CFR 50.55(e) disclosure item as of the end of the day on May 2. 1984 (Tr.

L 13591/13-13592/1). (It should be noted that there was no indication from either Applicants or the NRC Staff that they believed any such significant problems existed either.) And here we have yet another example of attitude, l

In mid-1984, the Applicants filed 17 Motions for Summary Disposition,

regarding issues chosen by Applicants to show that the plant had been E designed properly. CASE responded to each of those 17 Motions with little time fa properly address the issues. This method (the use of Motions for 5

Summary Disposition) which was one which the Applicants themselves elected to use, did not show that Comanche Peak was properly designed; instead, it provided CASE with additional proof that there were indeed severe design problems at the plant. It also showed the Applicants' attitude about resolving technical issues.

One example is the case regarding friction. The Applicants referenced a pipe support that was in CASE Exhibit 669B (Attachment to CASE Witness Jack Doyle's Deposition / Testimony, admitted at Tr. 3630). They misinterpreted the ASME code to their advantage; in addition, they made errors in their calculations which increased the stresses significantly.

Yet this was one of the examples with which Applicants hoped to prove that their plant had been designed properly. This .fs another example of the Applicants' attitude to avoid properly addressing design problems.

Cygna has now confirmed (at least in part) most of my concerns with the design of the cable trays (although this is not saying that they have covered all of my concerns or that 1 am in complete agreement with everything Cygna has done on the cable tray supports). Finally, after some additional review by Cygna in their Phases 2 and 4 reviews, on January 25, 1985, Cygna sent a letter to the NRC Staf f's Vince Noonan in which Cygna stated, in part (see Cygna letter 84056.050, January 25, 1985, Attachment B, Sheet 2 of 6, which was supplied as Attachment A to Cygna's 9/6/85 Response to Board's Memorandum (Information Concerning Cygna Independence):

"3. Cable Tray / Conduit Supports (Phase 2 and 4)

"Cygna reviewed cable tray support design as part of the Phase 2 work scope and is currently reviewing both cable tray and conduit support designs as part of the Phase 4 work scope. As a result of the Phase 4 reviews, Cygna is withdrawing all Phase 2 conclusions for both technical adequacy and design quality assurance of cable tray support design." (Emphasis added.)

6 1

i Since then, Cygna has found more problems and raised more questions regarding a variety of design and other matters (see recent Board Notifications attaching correspondence from Cygna).

The Board also needs to be aware that, from Information which CASE has received from Cygna, there appear to be as many, if not more, problems with the conduit supports as there have turned out to be with the cable tray supports.

Another item that both Mr. Doyle and I saw while working at Comanche J

Peak was that the control room ceiling was not seismically qualified. We did not think of this at the time we were testifying initially, and it was not brought up in the context of the licensing hearings. However, it was reported to the NRC in a 3/11/83 letter from CASE President Juanita Ellis to I Mr. Richard C. DeYoung, Director of the NRC Office of Inspection and Enforcement. Af ter the NRC Region IV Senior Resident inspector -

Construction, Mr. Robert Taylor, went to Comanche Peak, he concluded that the design was acceptable, and the utility also said it was an acceptable design; this was written up in NRC Region IV Inspection Report 50-445/83-24, 50-446/83-15 (copy of which is attached hereto). So there it was again*

neither the NRC's Senior Resident inspector of Construction nor the Applicants recognized this significant design problem. We have again the same old attitude problem, and yet another example of attitude.

However, I later brought this to the attention of the NRC's Technical Review Team (TRT), and it is discussed in the Staff's NUREG-0797, Supplemental Safety Evaluation Report (SSER) No. 8 (pages K-83 through K-85). Fortunately (in this particular instance at least) the attitude of the NRC Staff was better, and when the NRC's TRT went out there, they agreed that the ceiling was not properly designed for seismic and Applicants are now redesigning (or already have redesigned) the control room ceiling so it 7

. . . __ . _ _ _ _ . _ _ . -- _ _ ~ - _-

s-can- be qualified as seismic.

It really bothers me that this appeared to be such an obvious design problem. Yet there were other people out there who should have noticed it; but why was it' that Jack Doyle and I were the only ones who saw it and were l concerned about it'and nobody else was? In addition to the attitude of the

- Applicants and NRC Region IV, this leads me to believe that this may

} Indicate the presence of yet another type of example of a related problem --

what has come to be known in these proceedings as intimidation and harassment of employees at Comanche Peak. (I will not even attempt to get too far into this area, but I am aware that there have already been several 4

i confirmed findings of such intimidation and harassment. I believe that the Board should consider this in conjunction with my examples of attitude.)

One must question why Cygna, the utility, and the NRC Staf f, af ter their own reviews did not find the problems that have been unvelled by my short review of cable tray supports and the relatively small reviews by Jack Doyle and me of pipe supports. I cannot believe that Mr. Doyle and I are the only ones that can find design problems at Comanche Peak. It must have something to do with the attitude of the people and their incentive to bring

problems forward.

From the information that is contained in and out (so far) of the record (both on pipe supports and cable tray supports), it becomes quite apparent that there is no incentive for people to discover or correct deficient designs, for if there were, the problems that Jack Doyle and I found (and finally Cygna and the NRC's TRT found) would have been identified and corrected a long time ago). Instead the contrary is true at Comanche -

1 Peak, i.

8 d

_ . - . . ,, _ . . - , . _ , . - _ _ , _ _ _ . . . . - _ . . . , ~ _ . . - - _ . . _ , . . _ . _ _ - _ . . _ _ _ _ . _ _ _ ______- ___-

i 1

It is also important to point out that Jack Doyle and I are talking primarily about the pipe supports and the cable tray supports. We have not gone in and looked in detail at the building, the piping system itself, HVAC systems, or other items, but it appears to me that those items are not being properly reviewed or that Applicants are utilizing their inapplicable statistical sampling methods for those items. It is obvious that past reviews, reinspections, independent assessment programs, etc., have failed to identify major design and construction problems. There is no reason to believe that their current review is any better than their past reviews have been. And it must be assumed that when they complete their reviews and statistical sampling, they are going to come up with favorable answers because of their past proven attitude.

The attitude of the Applicants and the consultants which they have hired (both those which are dependent and supposedly independent), and the NRC Staff and their various special teams throughout these proceedings on design issues, in and of itself constitutes a pattern of attitudes, and can now be used to predict their attitudes in any ongoing or future reinspection or statistical sampling. Without having to perform any kind of statistical sampling, 1 can predict that they will come to the same conclusions they have always come to: that the plant is safe, that what they have done proves the plant is safe -- none of which bears any resemblance to reality, and none of which can be relled upon by the Licensing Board to resolve its concerns, and none of which can he relled upon to prove that Comanche Peak has been designed and constructed properly or that it is safe.

This is our third go-around in regard to the design aspects (see 10/29/85 Board Memorandum and Order (Status of PendIng Motions) at page 6).

As I have stated earlier, it is hard for me to believe that only Jack Doyle 9

t

and I can find significant design problems at Comanche Peak (or that we have found even a fraction of the design problems which exist at the plant). But every time we go into new formal or written litigation, we find more problems. Every time we go in and look at a new area of design, we find new problems which were not identified by anybody else. Hov many times is the Licensing Board going to expect Jack Doyle and me to continue to do this?

The attitude at the plant must be changed. The overriding goal at Comanche Peak was (and apparently still 1s) to get it done, no matter what.

That's what the position was while I was there -- they were all gung ho to get that plant done by the end of 1982 so they could get their operating license. Now it's four years later, and I have not really heard of any adequate positive system of detecting and correcting design errors at Comanche Peak.

I do not believe that the present system of statistical sampling is applicable to the current situation at Comanche Peak or that it is going to work. Because of the ingrained pattern of attitudes, all of the statistical samples the Applicants or NRC Staff want to conjure up will not identify and correct the design (and construction) problems which exist at Comanche Peak.

There should be a 100% reinspection of the entire design of the plant --

cable trays, conduits, HVAC, pipe stress, pipe supports, building design, anything and everything -- under very precise and carefully controlled conditions approved in advance by the Licensing Board. But whatever is done, whatever reinspection or reanalysis is performed, it has to be done in accordance with NRC Regulations (and specifically 10 CFR Part 50, Appendices A and B -- which is not the case with the CPRT effort). And there has to be the proper motivation, attitude, and independence. This apparently does not 10

exist at Comanche Peak and cannot exist with the present personnel in control.

d 11

O .

N I have read the forejoing affidavit, which was prepared under my personal direction, and it is true and correct to the best of my knowledge and belief.

(Signed) 14 [t/ /[/t Date: # /dcu _ fl,

[

STATE OF (EX A 'i COUNTY OF O ri L ' A A On this, the 'l day of 9)9 e < - ,198/, personally appeared Ar/E //42.5A/ , known to me to be the person whose name is subscribed to the foregoing instrument, and acknowledged to me that he executed the same for the purposes therein expressed.

Subscribed and sworn before me on the # day of *2/kup ,

19816 pllto rte.v- '7b $/tol-G~

Notary Public in hnd for the State of TE*tGR SAMUEt. W. NESTOR My Commission Expires 13189 My Comission Expires:

1

- . _ - ,_.-------__....-m_. . - , , , , , .- , .

i

i, .

, e . ?te t-: Tit

~

- M.'" L E A : REGU ATORY CON *r,nssio- .

- e

• , s.  ;;* C E G'O' i n

'. ** E * , * (* ; c s !.*. N ;a D e e. ; 5,  ; * :.

..... A %IP.G T O *. T E A A i 7M' In Reply Refer it:

Docket: SC 445/E3-24 E ,4 3g33 50-446/E3-15 Texas 1Jtilities Generating Concany ATTN: R. J. Gary , Executive Vice President f General Manager 2001 Bryan Tower Callas, Texas 75201 Gentlemen:

This refers to Ine insce: tion conducted by our Senior Residerit Inspe: tor, Construction, Mr. R. G. Taylor, during tne cerio: March tnrospr. July 19E3, of activities autnerized Dy fiRC Construction Fermits CPPF-126 and C: ;-127 for Comnche Peak, Units ' and 2, and te the discussion of our fin::in:s with

!4r. R. G. Toison, anc ciner renbers of your staff durin; tne inscettice..

Areas examined during tne instection inclucec review, inscecticr., and evah 5-tier, of several allegations made tc saricus fir: cersons, inciudin: - e ;.:o-it Safet3 and Licensing Ecarc in their cro:eedings regarding the c::datir.; license for Comanene Pea! Stea- Ele:tri: Station (CDSES). Wi thin these areas , the insce: tion consisted cf selective exa-inatio of :rs:eaures anc reoresentative records , interviews wi-h as-sonnel, anc coservations by tne inscector. Tr.ese findings are oo:urertec ir tne enclosec iastectier recor .

Caring tnis inscactior., it was found that certain of year activities were in viciation witn NF.: rec .irenents. You were notifiec cf one su t viciation by car letter of May 31, I?E3, tc which you have resooncec. Details of the ite enciesed witr car May 31, 1983 letter are inclucec in tne enciesec inspection recort.

One cr. resolved iter is identified in caragracn 15 cf the enciesec inspection report.

We nave also examined a:tions you nave taken witr. regar:: c creviously identified inscection findin;5. The statu cf :nese ite s 15 icer-i'ied in para;raan 2 cf tne enciesec recert.

Ir. accorcance witn 10 CFR 2.790(a), a c ;y of t*iis letter anc ve enciesure i will be placed in tne NR: Public Eccu ent :.cc uniess jeu n:tify this of fice ,

cy tsieoncne, witnin IC days cf tne cate of tnis letter, arc su:~it written I acclication to witnnele ir.formatier containec therein v.itnir 3C cays of the l cate of tnis letter. Sucn a:Dlication rust ce consistent witn tne recuire-ments of 2.790(b)(1).

Texas Utilities Generatin; 2 Comoany AUG 2 41933 Should you have any auestions concerning this insoection, we will te cleased to discuss tner witn yeu.

Sincerely, ifj 'M7 J.> "

5. L . ".a c s en . Cr i e f Reactor Prc,4ect Erancn i Encicsure:

Appendix - t'RC Inspection P.eport 50 345/23-2 50 446/53-15 cc w/encis:

Texas Utilities Generating Consany ATTh: H. C. 5:ncid , Prcjet: :2 nager 2001 Bryan Torier Dallas, Texas 75201 .

Texas Utilities Generatinc Consane ~

ATT!.. E . E. Ci emen I, '.' ice ;resicert, f.acisar 2001 E yan Tcwer,- Saite n.

1735

..llai.

L: texa! ,Ocu.

- - - - - - - . - - - . - -c, .- - -.

.w. .

APPENDIX U. S. NUCLEAR REGULATORY C0f1 MISSION REGION IV NRC Inspection Report: 50-445/83-24 50-446/83-15 Docket: 50-445 Ca tegory: A2 50-446

. Licensee: Texas Utilities Generating Company (TUGCO) 2001 Bryan Tower Dallas, Texas, 75201 Facility Name: Comanche Peak Steam Electric Station (CPSES), Units 1 and 2 Inspection At: Comanche Peak, Units 1 and 2, Glen Rose, Texas Inspection Conducted: March through July 1983 Inspectors: h) Of R. G. Taylor, Senior Resident Inspector R// 7/83 Ce te '

Construction (SRIC)

Approved': h ,? o,l ezz f D. R Hunnicutt, Chief 8// l83 Dste Reactor Project Section A Inspection Sumary Inspection Conducted March through July 1983 (Report 50-445/83-24 and 83-446/83-15)

Areas Inspected: Special. inspections, announced and unannounced, related to allegations made to various NRC persons including the Atomic Safety and Licensing Board in their procedings regarding the operating license for Comanche

~ Peak Station. The inspections involved 449 inspector-hours by one NRC inspector.

Results: The inspection confirmed the need to issue four violations initially identified by the Construction Appraisal Team (CAT) (NRC Inspection Report 50-445/83-18;50-446/83-12). These involved the areas of HVAC, Eouipment Installation, Document Control, and Storage of Equipment.

8 g

Details

1. Persons Contacted Principal Licensee Emoloyees

• R. G. Tolson, Site QA Supervisor

• C. T. Brandt, Non-ASME QC Supervisor

• J. R. Merritt, Engineering, Construction and Startup t'anager

• J. B. George, Project General flanger

• D. N. Chapman, QA Manager

• B. R. Clements, Vice-President, Nuclear Brown & Root (B&R)

• G. R. Purdy, Project QA Manager

• D. Frankum, Construction Project f*anager The SRIC also interviewed many other licensae, B&R, and subcontractor personnel during the course of the inspection.

• Denotes those persons who attended one or more management interviews with the SRIC.

2. Licensee Action on Previous Inspection Findings (Closed) Unresolved Item (50-445/82-22-02), " Analysis of Weld Discrepancies."

This unresolved item concerned a substantial number of identified defects in a large whip restraint essentially surrounding the mainsteam and feed water lines located several feet outside of the ASf1E code boundry point.

The device was engineered by the licensee's A/E and manufactured by NPS Industries. Due to the overall size of the structure, it has been nick-named " George Washington Bridge" by the site labor and quality forces. .The licensee had reported the. finding of the defects as a potential 50.55(e) item to the SRIC on September 30, 1982, which was subsequently stated not reportable in a letter dated December 27, 1982. An NRC inspector followed r up on the matter during a visit to the offices of the A/E, as documented in NRC Inspection Report 50-445/83-12. This review pertained to all of the defects involved with the exception of two cracked welds that had not been analyzed at the time of the inspection. The engineer has recently analyzed these two defects and has detennined that had they not been detected, the structure could have fulfilled it's function. The SRIC has reviewed the location of the cracks and their length in relation to the size of the welds and the functional application of the structure. Since the structure has no continuous service application and is essentially subject to a one-time loading, the cracks would not have the potential for further propagation.

Further, the cracks are at points in the structure that would receive rela-tively low stresses in the one-time impact based on their small size in relation to the. members being welded. It appears that the cracks formed due to the stresses developed during the tightening of high strength bolting in

3 the imediate vicinity of the welds during the site assembly of the structure.

Taken in conjunction with the earlier documented review of the engineers calculations and the SRIC's review of these cracks, the SRIC has concluded that the engineer's overall analysis was adequate and that deficiency (s) were not reportable under 50.55(e). Both the licensee's initial report (CP-82-12) and the above identified unresolved item are considered closed.

It shoula be noted for the record that this closure only applies to the reportability aspects under 50.55(e) and not to the correction of the def d...

The defects, including the cracks, have been documented on a nonconformance report. The final disposition and closure of the NCR will be evaluated during future routine inspections.

3. Review of Licensee Self-Evaluation (Usino INP0 Criteria)

The SRIC has reviewed a report of the licensee's self- evaluation performed during October 1982 which was based on criteria that has been developed for the purpose by INP0. The evaluation was perfomed in behalf of the licen-see by personnel in the employment of Sargent & Lundy, an architect-engineer firm with substantial nuclear power involvement. /, copy of the report was furnished to the NRC, and subsequently, to the Atcmic Safety and Licensing Board in the matter of Comanche Peak Station operating license by letter dated May 2, 1983. The purpose of the review by the SRIC was to determine if any of the 47 findings in the report were of a type and of sufficient significance to have been reported to the NRC as required by 10 CFR 50.55(e).

The SRIC reviewed each of the 47 findings and the supporting documentation in the report pertaining to each finding. This review revealed that none of the 47 items were based upon identified deficiencies in structures, systems, or components nor were there any significant deficiencies in desigi ,

or engineering,(e).

10 CFR 50.55 testing that would constitute conditions reportable under

4. Car Wash In Containment During the limited appearance statement portion of the Atomic Safety and Licensing Board hearing on l'ay 16, 1983, a person stated at transcriot page 6152 that he understood that the containment looked something like a car wash. The person stated that it was his understanding that the situa-tion developed at about the same time that there was a meeting at the 0/FW Airoort between the NRC and any interested parties to discuss NRC decen-tralization. That meeting took place on Aoril 5,1983. For the ourposes of evaluatirg this allegation, the SRIC expanded the period of interest to include the 3 weeks prior to the meeting. Curing this entire period, the Unit I reactor systen was undergoing what is referred to as " Hot Func-tional Testing". This particular test is an accurate simulation of the operation of the reactor system and its accurtenances but without a reactor core being in place. The heat and pressure in the system is generated by the reactor coolant pumps in conjunction with the chemical and volume con-trol system charging pumps. The test could readily be construed to be a pressure test but in fact is an operational test at pressure. This parti-cular test extended overall for about 90 days beginning late in February

\

o -

4 and continuing until late l'ay. The SRIC monitored the test but was by no means continously in the containment. The SRIC interviewed personnel in the licensee's startup test group, QC inspectors who had reason to be in the building and others to obtain a picture of the events that occurred in the Unit 1 Containment Building during the period of interest. The SRIO also reviewed the licensee's control room logs for any indication of oper-ational problems indic3tive of a major leak in any of the fluid filled systems under test. ~he picture obtained was that there were several small leaks, generally at the gaskets between valve bodies and their bonnets. In addition, there was a considerable amount of condensation dripping from the reactor coolant pump motor cooling coils. This was caused by the cold water in the coils condensing the humidity from the atmosphere within the building and was not indicative of a leak in the reactor coolant system. The SRIO found from the control room logs that on March 29, a steam leak occurred during one phase of the test when a drain valve was partially open. Perhaps this valve should have remained closed. The room in which the valve was located was apparently filled with steam vapor which would have condensed out on the cooler walls as water. On t' arch 30, the reactor vessel head vent valves were partially opened, which in turn would give some amount of steam blowoff into the reactor refueling cavity area and would rise up into the building until cooled and condensed out as water. fione of these events are typical of any major leak indicative of piping or piping component (such as a valve) failure. The type of small events described above are, within the experience of the SRIC, typical of what would be expected during such a test and is one of the reasons for perfanning the test.

5. Design of the HVAC System Succorts By letters, both dated l' arch 11, 1983, Citizens Association for Sound Energy (CASE) notified the f4RC's Offices of Inspection and Enforcement and the Executive Legal Director of a concern that the HVAC system for Comanche Peak had not been properly supported, nor had it been properly considered in regard to seismic load conditions or its treatment as potential mis-siles. CASE specifically states that from their review of the FSAR, it appears that the licensee has not analyzed the HVAC supports for a seismic load condition. Specific reference is made to Sheet 21 of Table 17A.

In addition, the personal observations of l'essrs. Walsh and Doyle are relied upon to point out that there are no lateral supports en the HVAC systems within the containment. CASE also states that all HVAC components and supports inside containment should be treated as missiles under Cri-terion 4 of the General Design Criteria for fluclear Power Plants, 10 CFR 50, Appendix A.

Sheet 21 of Table 17A of the FSAR lists the containment ventilation sys-tems as being Seismic Category II. Aoparently, it has been assumed by CASE that this category excludes seismic loading in the design. This assumption is incorrect since the FSAR, Section 3.2.1.2 defines Seismic Category II as being those portions of systems or components whose

5 continued function is not required but whose failure could reduce the func-tioning of any Seismic Category I system or component required to satisfy the requirements of C.I. A through C.1.Q of Pegulatory Guide 1.29 to an unacceptable safety level or could result in incapacitating injury to occupants of the control room. These systems are designated Non-Nuclear Safety (NNS) Seismic Categury II and are designed and constructed so that a safe shutdown earthquake (SSE) will not cause such a failure.

CASE also states that if the HVAC systems within the containment failed during a SSE, this would allow the temperature within the containment to rise quickly to unacceptable levels which could over time cause compon-ents and monitoring equipment to fail and which could also mean that it might be impossible for workers to enter the containment due to the heat.

Containment heat renoval is required by Criterion 38 of the General Design Criteria for Nuclear Power Plants. The system to remove heat from the reactor containment at Comanche Peak does not rely on the HVAC system but rather is composed of two separate containment spray recirculation trains each with 100 percent capacity. Each train contains two separate pumps, one heat exhanger, and seven spray headers, and each system is fed from its individual electrical Class IE bus. The containment heat removal system is designed to ensure that the failure of any single active compon-ent, assuming the availability of either onsite or offsite power exclusively, does not prevent the system from accomplishing its planned safety ft.nction.

CASE's concern with being able to enter the containment following certain oesign basis accidents is unfounded in that it is not a requirement.

In order to assess the adequacy of the design of HVAC supports, an inspec-tion was conducted at the home office of " Corporate Consulting & Develop-ment Company, LTD. ," the support design consultant. It was determined that all permanent HVAC supports are analyzed for seismic loading. Two methods are utilized: Zero Peak Accleration (ZPA), or 1.5 Times the Peak Accelera-tion When the Fundamental Frequency Falls Below 20 Hertz. Of the latter method of design, only about 6 out of 4000 supports have been designed that way. A typical HVAC duct run is supported axially at every third support This may explain why Nessrs. Walsh and Doyle may have felt that there were no lateral supports on the HVAC systems. The NRC inspector reviewed the design of a typical HVAC duct run at elevation 852'-6" in the Auxiliary Building. Supports were designed utilizing two computer programs entitled FEASA-2D and FEASA-3D. The acronym stands for frame eigenvalue and stress analysis. The -20 version is used on the transverse supports and the -3D version is used on the axial supports. The inclusion of equivalent weights from both up and downstream transverse succorts and accesories such as vol-ute camcers and vane turns in the design of the axial supports was verified.

This inspection verified the adequacy of the siesnic design techniques being utilized for the design of HVAC supports at Comanche Peak.

The concerns expressed by CASE have been found to be without merit.

Persons contacted during the course of the inspection at Corporate Consulting

6

& Development Company, LTD. were:

J. Roland Yow, President & Chief Executive Officer Gary Hughes, Vice-President for Operations David Lindley, Principal Engineer Stephen Lehrman, Seismic Department Manager Daryl Hughes, Project Engineer

6. Heating, Ventilation, and Air Conditioning System (HVAC)

During the CAT inspection (NRC Inspection Report 50-45/83-18;50-446/83-12),

the CAT inspectors noted that a significant portion of the welds on the ducting support structures were deficient in relation to the applicable welding code requirements. The dominate deficient condition noted was that the welds were significantly undersized. Based upon this information the SRIC toured various areas of the facility with special emphasis on the ducting in the Unit 2 Containment Building since that was one of the more recent areas of installation by the HVAC contractor. In accordance with the design drawings, the bulk of the welds should have been fillet welds with hinch leg size. The SRIC noted by visual comparison to the hinch thick base metal that very few of the welds were of proper size. The CAT inspectors also found cases where the bolting and gaskets between ducting sections were loose and/or missing.

The CAT inspectors also found that some support members were not within the dimensional tolerances on the design drawings. It was noted that the centractor's inspection records did not reveal these various facts, indicating ineffectual QC by the contractor. Further, a review of the licensee's audit program indicated that the licensee was unaware of these several problems in the fabrication, installation, and inspection of the HVAC systems. Based upon the CAT inspectors' findings and his own observations, the SRIC recomended that a notice of violation be issued to the licensee pertaining collectively to these matters (Notice of Violation issued on May 31, 1983.

Reference 50-445/83-18 and 50-446/83-12, item 4).

7. Installation of Major Items of Equipment The CAT inspectors noted during their' inspections of certain major items of equipment that there were several variables in how the equipment was fastened to the building equipment pads. In some instances, tanks for example, CAT inspectors found that there were two nuts (double nuts) on the embedded bolts securing the equipment, other bolts had one nut, (single nut) and some had a combination of both single nuts and double nuts en one piece of equipment. The CAT personnel also noted that certain heat exchangers had slotted holes in one of the mounting bases to allow for thermal expansicn during operation. The holddown nuts acoeared to be installed too tightly and may have prevented freedom of movement. The SRIC obtained the design and installation drawings for two of the referenced heat exchangers identified in the CAT report. Both were found to be horizontal Utube heat exchangers whose function is nonsafety, but whose pressure boundary in the tubes is safety-related since the process fluid could be radioactive. The SRIC found that the construction drawings for the mounting pedestals had a flat steel plate on one

T 7

pedestal that would be suitable for the type of mounting detail on these heat exchangers. The SRIC then reviewed the installation travelers for each heat exchanger and found that 'these documents did not note or address the slotted details, the plate, or the fact the bolts should be left loose. The SRIC would note that the vendor manual which provides the details does not provide infonnation on how loose or tight the nuts should be nor how these nuts are to be locked at that looseness or some torque value. The SRIC with the assistance of site QC and craft labor had one of six nuts loosened on heat exchanger TCX-CSAHLD-01. On all six of the studs involved, each had only one nut (single nut). The one nut that was loosened had been very tight, as evidenced by the amount of force required to break the nut loose. On another heat exchanger of comparable design, it was found that each stud was double nuted and when the top nut was loosened, the second nut was approximately one flat (about 1/6 of a turn) from being fully tight. This degree of looseness should allcw sufficient freedom of movement. During the document review, the SRIC found that the engineer had specified that all rotating and vibrating equipment should be double nutted and that other equipment could be securad with only one nut. fio document could be located that established the identity of vibrating equipment nor were there any apparent provisions made to lock nuts where they must be deliberately left loose. This was considered overall to be a violation of Criterion V of Appendix B to 10 CFR 50 (flotice of Violation was issued on May 31, 1983.

Reference:

floti ce of Violation 50-445/83-18 and 50-446/83-12, item 1).

8. Maintenance of Eouipment In Outdoor Storaae Areas The CAT found that a considerable amount of equipment such as pipe support struts, clamps, and like items, normally stored outdoors, was not being properly maintained in accordance with procedure MCP-10,

" Storage and Storage Maintenance of Mechanical and Electrical Equipment", as evidenced by rusting bolts and adjustment screws on s truts . In addition, the strut bearings were dirty from dust and

the bearing load pins, in some instances, were rusted. By a tour of the storage areas, the SRIC confirmed the CAT inspectors find-ings. The SRIC would also note that the IftP0 Self-Evaluation Report at page 111 describes essentially the same finding. This situation was determined to be a violation of Criterion XIII of

, Appendix B to 10 CFR 50 (flotice of Violation issued on May 31, 1983.

Reference:

flotice of Violation 50-445/83-18 and 50-446/83-12, item 2). The SRIC would note for the record that there is little evidence that any items which indicated substantial deterioration from such storage conditions have in fact been installed in the nuclear power block. It would appear that the various items involved 1,

have been cleaned and restored prior to installation such that they I can perform the required function.

9. Obsolete and/or Illegible Drawinos In The Field The CAT inspectors found a group of drawings in one particular area adjacent to the control room that were found to be out of date by

, up to several issues and further, that some drawings in other areas l were incomplete in the title and revision blocks. The SRIC discussed l

b

7 8

the finding with supervisory personnel of the licensee's central document control center who indicated that they had located the drawings identified by the CAT inspectors along with many more that were obsolete in other areas. It was stated that distribution system for engineering drawings had become faulted by the simple volume and by the need for so many points of distribution and audit verification thereof. Since problems are obviously still present, it was detemined that the licensee had violated Criterion VI of Appendix B to 10 CFR 50 (flotice of Violation was issued on May 31, 1983.

Reference:

flotice of Violation 50-445/83-18 and 50-446/83-12, item 3) and that substantial steps would be required to correct the problems.

10. Allecations Relative To Improperly Supported Items In The Control Room The president of CASE in a letter dated March 11, 1983, addressed to Mr. Richard C. DeYoung, Director of the NRC Office of Inspection and Enforce-ment, indicated that CASE had received information from an unidentified source to the effect that:

a. There is field run conduit above the control room supported only by wire.

b. There is drywall (or sheet rock) that is supported by wire,

c. There may be lights that are supported by wire.

The SRIC has examined the susiended ceiling and the area above the sus-pended ceiling in the centrol room area and has examined the pertinent engineering drawings depicting both in relation to these allegations with the following findings:

a. There is a considerable amount of both safety-related and ncnsafety related conduit in the area above the suspended ceiling. The safety-related ccnduit is supported by Seismic Category I supports typical of those used in other areas of the facility. The nonsafety-related conduits are generally supported by simpler and less substantial sup-ports that are typical of those that the SRIC has observed in large open factories and are not designed to seismic standards. In each case examined, the ncn-seismic support was structurally paralleled with a small stainless steel cable that would assume the full weight of the conduit were the nomal support to fail in a seismic event.

b. The drywall materials were found to be part of the suspended ceiling above the central part of the contral room and to fom a part of the sloping wall area below the control room observation room, lhese dry-wall materials have been securely fastened to a metal frame work (metal batten) which in turn is supported by conventional and non-seismic straps and wires to the concrete primary building. The franca work is also attached to a system of stainless steel cables which in turn also attach to the primary structure such that if normal sup-ports fail during a seismic event, the weight of the framing and drywall will be assumed by the cabling thus preventing the materials from falling.

).

9 .

c. The lighting fixtures in the control room are supported fran an intermediate substructure of "unistrut" by light-weight conduit.

The substructure is likewise supported by the same type of conduit from the primary structure ceiling. The conduit used appears

.to be the typical of that supporting the light fixtures in most offices with suspended ceilings. Paralled with each conduit are two small stainless steel cables which would assume the load if the conduit or its attachment were to fail. In the case of the actual light fixtures, the cable is attached to the light fixture at the edge of the reflector assembly.

The SRIC would note for the record that above described design features appear to fully satisfy the intent of the licensee's commitment to comply with NRC Regulatory Guide 1.29, " Seismic Design Classification."

The licensee has used terminology in the classification system that is at variance with that of the regulatory guide but is explained and cefined in Section 3.2 of the FSAR. In essence, the licensee has defined all safety-related items that must remain fully functional during and after a seismic event as Seismic Category I. Items not having a safety function but whose failure could damage components which have a safety function or cause injury to the occupants of the control room during an event are

.eferred to as Seismic Category II. In the case of the items involved in this allegation, all are Seismic. Category II since their falling could cause injury to the control operators. The cabling system described can be expected to prevent such a fall even though the normal supports could possibly fail. The stainless steel cable used in this design feature, which at a short distance away looks much like bright galvanized common steel wire, is of relatively high strength. As an example, the test strength of an 1/8-inch cable is in excess of 1760 pounds. With four cables attached to a light fixture, two at each end, the total support capability of the cables is over 7000 pounds. It is apparent that the designers have elected to use conventional suspended ceiling and light fixture support techniques in order to use conventional and available materials and then provide a high strength backup support system in a seismic event.

No violations or deviations were identified during this special inspection effort.

11. Placement and Curing of Concrete During Freezing Weather During the limited public appearance portion of the Atomic Safety and Licensing Board (Board) hearing conducted on May 15, 1983, there were two references to the placing of concrete in freezing weather at the Comanche Peak Station which in turn lead to a question from the Board to the NRC staff as to whether there were any NRC personnel present with knowledge of the matter. The two references are at 6106 and 6134 of the hearing transcript while the Board question is at 6109. Also at 6109, an uni-dentified voice responded to the Board that tt n matter had been reported in IE inspection reports. Research of the NRC inspection reports revealed that there had been such a discussion in NRC Inspection Report 50-445/77-01 which was categorized as an unresolved item pending the licensee's review and action on their finding of the problem. The unresolved iten was further discussed in NRC Inspection Report 50-445/77-04 with the closure of the item by an improvement in the OA procedures.

y 10 The SRIC has reviewed the matter, particularily with a view toward deter-mining whether the practices involved actually caused damage to the concrete involved. The primary focus'of NRC Inspection Report 50-445/77-01 (Details II, paragragh 5) was directed toward two licensee " Site Surveillance Reports" which had been prepared approximately 2 weeks earlier than the inspection period covered by the inspection report. The first of the licensee's reports (C-134-77) was directed specifically to findings by a licensee inspector that the surface temperature of Concrete Placement 101-2808-001 some 6 hours6.944444e-5 days <br />0.00167 hours <br />9.920635e-6 weeks <br />2.283e-6 months <br /> after the placement was completed were well below freezing in some locations.

The other licensee report (C-135-77) was directed toward records and was not considered in this review. The SRIC obtained the necessary records to review the matter and found that placement 101-2808-001 had taken place on December 30, 1976, being completed at approximately 6:00 p.m.

Later, the same evening at approximately midnight, the licensee inspector found that some surface areas were chilled to as low as 200F. The records reflect, however, that there was _ disagreement between the B&R inspection personnel assigned to monitoring the curing of the placement and the licensee's inspector as to what the surface temperatures actually were.

The B&R personnel contended that the licensee inspector was actally mea-suring the air temperature rather than the temperature of the concrete. No resolutbn of that disagreement was reflected in the records. The SRIC interviewed the licensee inspector of record during the course of this review to gain a clearer understanding of the events which took place.

The licensee inspector stated during the interview that he was confident that his measurements were accurate and also stated that there was no phy-sical evidence that the concrete was frozen.even though the surface-temperatures were well below freezing. The records also reflect that in order to resolve the issue, swiss hammer tests were run on the suspect areas af ter the concrete had fully cured. These tests indicated that the suspect areas had attained strengths comparable to known properly cured areas, indicating that the concrete had not been damaged even though the possibility exists that it had been frozen for a period of time. The records reflect that good concrete curing temperatures, i.e., above 400F were established and maintained shortly after the licensee's inspector's observation.

For the record, the SRIC would note that Placement 101-2801-001 took olace in the Unit 1 Reactor Building. The placement became the open area floor at the lowest full floor in the building. This floor area, while suppor-ting some equipment, serves primarily as a walk area. As such, it is fully topped with an architural concrete making the structural concrete no longer accessable.

NRC Inspection Report 50-445/77-01 also discussed comparable events to that documented on Surveillance Report C-135-77. One of these events was docu-mented by Surveillanct Report C-C68-76 on January 7, 1976, and on B&R deficiency / disposition reports (now titled nonconformance reports).

These documents indicate that on January 7,1976, the surface temperature of Placement 105-2773-001, the foundation basemat for the Unit 1 Safeguards Building, were found frozen as evidenced by frozen wet burlap over certain areas that were not covered by insulating blankets. The records also

11 reveal that the reported finding took place almost 7 days after the place-ment of the concrete. Although the placement should' not have been allowed to freeze in the time frame involved in accordance with the project speci-fication, the placement was accepted "use-as-is" on the premise that the curing temperatures during the 7 days were conducive to a good cure and that after 7 days there would be little free water in the concrete to freeze even though the burlap was froze. This conclusion is considered valid by the SRIC based on his review of publications of .he Anerican Concrete Institute and the Bureau of Reclamation. Further, in responding to a. separate finding that the field cure test cylinders made for the placement tested lower than allowed by the project specifications, swiss hammer tests were perforned.

The swiss hamer tests indicated the concrete placement had full specified s trength. Relative to the low reported strengths of the field cure cylin-ders, the SRIC would note that in his experience field cure cylinders will frequently test low under cold weather conditions. The reason is that the cylinders' small mass generates little heat of hydration, thus making th0n either more vulnerable to freezing and/or curing much slower than normal due to their depressed temperature.

The final events covered by f!RC Inspection Report 50-445/77-01 included DDR-C-460 ut ich in turn discussed low temperatures during the curing per-iod of three separate placements that were made during the late Cecember time period of 1976. In each case, the records reflect that the placements were accepted "use-as-is" since the least amount of cure time was 9 days, again with good conditions until the cold weather occurred.

The NRC inspector involved in fiRC Inspection Report 50-445/77-04 which closed the unresolved issue has stated that he had visually inspected each of the placements discussed in f1RC Inspection Report 50 445/77-01 for evidence of damaged concrete and found none. NRC Inspection Report 50-445/77-04 did not reflect those inspections since the fiRC inspector was aware that the concern was for prevention of repetition rather than any specific concern about the quality of the placements involved.

The SRIC would note for the record that there are no regulatory or industry prohibitions on placing concrete in cold weather conditions. The /merican Concrete Institute and the0 Bureau of Reclamation both indicate that if the fresh concrete is above 40 F at the time of placement, the chemical prccess of hydration will generate sufficient heat to prevent the concrete from freezing provided that precautions are taken to prevent heat loss. In mass concrete applications, the greatest danger to the concrete is on the exposed surface areas, particularily at corners and other edges of the placement.

It would be exceedingly rare for the mass of the concrete to free:e and sustain damage. These publications also indicate that even if frozen, the concrete will normally cure to full design strengths if temperatures con-ducive to the hydration process are restored.

12. Allecations Relative To The As-Built Verification and Cesian Verification Activities.

During April 1983, fiRC personnel received allegations to the effect that l

12 the QA group performing as-built verifications were not measuring support member dimensions and therefore, the " Vendor Certified Drawings" of the supports would not be accurate. A second allegation from the same person indicated that the QA group charged with responsibility for verifying that design changes have been incorporated into the plant and that the inspection records for the installations accurately reflected that incorporation was being required ~with the use of a computer generated status document to make the verification of records. The allegation was that the computer list-ing was faulty and therefore, the verification effort was equally faulted.

The SRIC has examined each of these allegations as to the factualness of the allegation and as to whether the allegation has or will have an effect on the safety of the facility when operating. In regard to the first allega-tion, the SRIC found that the allegation was and is factual. The allegation, however, does not appear to have any significant impact on safety in that the as-built inspection was not developed to assure that the " Vendor Cer-tified Drawing" was an accurate representation of the support in all aspects.

The as-built program was established to assure only that the support loca-tion on the supported pipe and the direction of support is accurate for the purposes of perfoming the final pipe stress analysis. The responsibil-ity for assuring that the support members and other characteristics of the individual support reflect the design drawing requirements reside in other QA groups associatec aith the fabrication and installation efforts. To also perform these functions in the as-built verification inspection would be a redundant inspection that would not contribute significantly to the safety function of any given support.

Regarding the second allegation, the SRIC found that it too was factual but only at the specific time the allegation was made. When making the allega-tion, the alleger provided the NRC personnel with a reference to a QC inspection report which he said would fully display his concern. This report, identified as IR DCV-00421, was found to contain notation that the verification was based on a computer tabulation and that the report was being completed at the direction of the inspector's supervisor. The original report was dated April 4, 1983. The permanent file copy was found to have been marked " voided" by the originating inspector as of May 20, 1983, with a notation that the report had been superce&d by IR DCV-00423. This latter inspection report was examined by the WIC and found to document essentially the same inspection effort by the same inspector but without any notation of having n based upon a computer tabulation and without notation of apparent prctist of directions given by supervision. The SRIC interviewed the QC inspector who prepared and signed all of the reports noted above in order to ascertain what had and is transpiring in the OC design verification program effort. The inspector stated that the attempt to use the computer based data in the performance of the assigned task was in error from the beginning because of errors by persons genera-ting the computer data. The interviewee stated that only the one verifica-tion effort had been done using the computer based data and that all prior and subsequent verifications have been done by the assigned inspectors directly and personally examining the existent quality records in compli-ance with applicable QC procedures for the task. He stated that the only

13 procedural deviation was the one instance stated in the allegation. Dis-cussions between the group supervisor at the time the allegation was received and the SRIC indicated that he had attempted to use the computer tabulation to expedite the task on a trial basis by management direction t and that he had caused the original inspection report to be filed as it was to give management a picture of the faults in the computerized data. It thus appears that the design verification effort has been perfonted in accordance with procedures except for the one-time pertubation that was subsequent correctly reaccomplished in accordance with approved proce-dures.

i fio violation to flRC requirements were revealed during this special inspection effort.

13. Improperly Certified Liquid Penetrant Examination Materials The CASE informed the Atomic Safety and Licensing Board by a letter dated May 18, 1983, of a potential problem with the liquid penetrant materials in

. use at the Comanche Peak Station. The letter stated that CASE had been made aware of the potential problem during a phone conversation with Charles A.

Atchison, who in turn learned of the " problem" from a Dallas area represen-tative of the Magna-Flux Corporation, the orginal manufacturer of the material.

The letter states that the problem surfaced only 7 to 10 days earlier. Based on the date of the letter, it would seem that the problem arose between approximately May 8 to May 11, 1983. .

The situation bears close resemblance to the situation outlined beginning with fiRC Inspection Report 50-445/82-18;50-446/82-09 based upon an inspection conducted during the period of September 7-10, 1982. The f4RC inspector noted that some certified test result documents had been altered by " pen and ink" changes not immediately explainable. The matter was considered unresolved at that time. During a second inspection of the matter, conducted during flovember 1982 and documented in flRC Inspection Report 50-446/82-11, the

, inspector found that previous corrective actions were not adequate and fur-ther that the " pen and ink" changes sometimes didn't match the type of material being certified. A flotice of Violation was issued as part of the inspection report on the matter. The licensee responded to the flotice of Violation by a letter dated Decemoer 21, 1982, wherein he stated that a supplier had altered the certificates but that the original manufacturer had been able to furnish valid certificates and further, that all future purchases would be direct frcn the manufacturer rather from a " middle-man" supplier. The licensee also stated that specific receiving inspection pro-i cedures had been implemented to prevent repetition, f4RC Inspection Report 50-445/83-10;50-446/83-05 documented verification that the licensee's acticas were acceptable and the matter was closed.

It appears that the situation outlined in the CASE letter parallels the flRC findings in all details except for the dates which probably arose as a result of misunderstood or incomplete communications between the

, . . = -

l

'14 '

y

/

Magna-Flux representative and Mr. Atchison and/or with ' CASE.

CASE also posed two questions on the matter as follows: s

a. Has an NCR been written on this problem? .

Answer: The above discussed inspection reports document a total of <

five NCR's that were issued.

b. Has either TUGC0 or Texas Utilities or B&R notified the NRC'of this problem?

f Answer: The roles of reportabil'ity were effectively reversed in that the NRC identified .the problem and notified the licensee.

A need for further NRC action on this matter has not been identified and the matter is considered closed. ' ,

14. Penetration Seals f i This special inspection was undertaken to' ascertain the validity and sig-nificance of allegations received initially by an NRC Headquarters Duty Officer on or about March 22, 1983, which were confimed and added to during a telephone interview with the alleger on Parch 23, 1983, by the/SRIC and a NRC inspector assigned to NRC Region I. The allegations, as understood by the SRIC, were:

a. The overlap seal for flexible boots should be 3 inches whereas 2 inches is being used oy BISCO.

b. There maybe a problem with the' strength of the fabric used in the flexible boots since the material supplier and BISCO are involved in a lawsuit.

c. The aggregate used in a radiation seal may separate giving rise to improper personnel protection. ,

Since BISCO was and is on the Comanche Peak site. installing seals, kegion IV was selected for the purpose of? this special inspection although the com '

pany has involvement at several other nuclear power sites throughout the United States. The SRIC obtained from the BISCO site manager all of the production and quality procedures applicable to the work at CPSES as well as some that are not. The alleger specifically mentioned that the NRC should review Procedures QC-507, SP-504, SP-505, SP-505-1, and SP-505-2.in regard to the flexible boot overlap problem. Each of the above procedures was in the books offered to the SRIC for review. A brief discussion fol-laws as to the contents of these procedures:

a. QCP-507: This procedure covers the final inspection of installed

,. s s 15 flexible boots. The amount of overlap is not mentioned in the procedure, although the procedure does require that the seam be examined for evidence of poor. sealing such as " fish-mouthing" which is taken to mean that the exposed edge of the overlap is puckered and not adhering to the base fabric.

b .~ SP-504: This procedure provides instructions and a calculation sheet to initia'ily cut the fabric into a shape that would subse-quently allow the fomation of a truncated cone. The fomula on the calculation sheet requires that 1-inch be added at each edge of the fan shaped fabric which is evidently to pro-vide the overlap. The base fomula prior to adding the 1-inch-provides a dimension just equal to the circumference of the pipe and/or sleeve to which the boot will be attached.

Thus, the 1-inch at each edge will provide for 2-inches of overlap, assuming that the pipe and sleeve are concentric.

If pipe and sleeve are not concentric, the resulting cone will be skewed and the seam overlap will be something other than 2-inches.

c. SP-505: This is a generic procedure for the installation of flex-ible boots. It was noted that the procedure requires that the adhesive for the overlap seam be spread over a 3-inch depth from the fabric edge prior to fitting up the fabric where it is to be installed. Although not so stated, it appears that the 3-inch width of adhesive is to provide sufficient area of adhesive in the event the above men-tioned cone skewing occurs, d' . SP-505-1 and SP-505-2: These are additions to SP-505 having appli-cation when the boots are used as a simple pressure seal

~

only and for when the boot is used as part of a fire pro-tection seal, respectively.

The SRIC interviewed the BISCO site manager as to whether the procedures had ever required a 3-inch overlap. The site manager indicated that 3-inch seam had been used up to sometime in 1979 and that his homeoffice engin-eering had then changed the seal seam detail. The SRIC reviewed the results of a pressure differential test perfomed by BISCO in September 1979 which indicated that the fabric boot would withstand a differential pressure of 44 psig without sustaining damage. The project specification (2323-tis-38F) reouires that the pressure seal maintain its integrity only up to 2 psig.

While the BISCO test data does not specifically state what the overlap seam width was on the test boot, it would strongly appear that the strength mar-gin is so high that even a reduction of 1/3 in the area of the overlap would have the effect of changing the safety factor from 22:1 to approximately 14:1.

It is the SRIC's conclusion that while the allegation relative to the reduction in seam from 3 to 2 inches is correct, the reduction would have no significant effect on the performance of the boot in service at CPSES and that, therefore, the allegation has no technical merit.

r .

]

.' c ,

~

16 Regarding the matter of tne possibility of some undefined problem with the boot fabric, the BISCO site manager stated that his company has been engaged in a law suit with the supplier of the fabric but only in regard to the per-fonnance of the fabric in one application which is understood to involve the tearin') of the fabric after being punctured._ It is understood that the puncturing has occurred when a gel type radiation seal hardens under radia-tion. Since the specific design involved. is not scheduled for use at CPSES, f the allegation has no technical merit.

Regar, ding the matter of possible separation of the radiation seal aggregate material from the carrier material, the SRIC can only conclude that the al-legation is potentially correct but without apparent merit. The BISCO test reports indicate that the seals involved met the engineers specification.

The separation of the aggregate (powdered lead) from the carrier (a silicone material) would appear to be process sensitive in that if they are not well mixed, pockets of lead might form with resulting pockets of silicone without sufficient lead. Since the specification and the BISCO procedures require careful control and monitoring of the mixing process, the SRIC can only con-clude that these measures are effective in production operations as they wer,e in preparation of the test samples.

i

15. Electrical Cable Solicing The SRIC became aware that the Comanche Peak project electrical engineer had authorized the splicing of safety-related and auxiliary electrical cables within several control panels curing the inspection period. Since the licensee nas committed in FSAR Section 8.1 to comply with IEEE 420,

" Trial-Use Guide for Class IE Control Switchboards for fluclear Power Gener-ating Stations," whici forbids splicing of wiring in such panels, the SRIC judged that the licensee was deviating from these commitments. The licen-see engineer indicated that he interpreted the IEEE standard to prohibit such splicing only between the cabinet terminal boards and the cabinet devices and did not prohibit such solicing in the field run cables attach-ing to the terminal boards. The engineer stated that action had been initiated with the flRC Office of fluclear Reactor Regulation to clarify the issue in the FSAR. The SRIC confirmed that such action had been initiated by a telephone conversation with the flRR Licensing Program Manager for Comanche Peak. Pending action by flRR, this matter will be considered as an unresolved matter.

16. Unresolved Items Unresolved items are matters about which rrore information is required in order to ascertain whether they are acceptable items, items of non-compliance, or deviations.

One such item, disclosed during the inspection, is discussed in paragraph 15 above. This item is identified as " Splicing of Electrical Cables in Cabfnets." (8324-01)

, . ,4 ? ,

17

17. ftanacement Interviews The SRIC met with one or more of the persons identified in paragraph 1 of this report at frequent intervals during the inspection period to discuss the licensee's position and proposed actions on a significant number of issues which occurred during the period.

. . . .,N s

.j.-

V hN l UNITED STATES OF AMERICA '

iT I NUCLEAR REGULATORY COMMISSION Gl l '

3 ;,

BEFORE THE ATOMIC SAFETY AND LICENSING BOARD f//,

,\

In the Matter of }{ x' '

}{

TEXAS UTILITIES ELECTRIC }{ Docket Nos. 50-445 COMPANY, et al. }{ and 50-446 (Comanche Peak Steam Electric }{

Station, Units 1 and 2) }{

CERTIFICATE OF SERVICE By my signature below, I hereby certify that true and correct copies of CASE's Additional Response to Licensing Boards' 11/11/85 Memorandum (Statistical Inferences fran CPRT Sampling) have oeen sent to the names listed below this 6th day of May ,19 8,6_,

by: First Class Mail Administrative Judge Peter B. Bloch Nicholas S. Reynolds, Esq.

U. S. Nuclear Regulatory Commission Bishop, Liberman, Cook, Purcell Atomic Safety and Licensing Board & Reynolds Washington, D. C. 20555 1200 - 17th St., N. W.

Washington, D.C. 20036 Judge Elizabeth B. Johnson Oak Ridge National Laboratory Geary .e. Mizuno, Esq. ,

P. O. Box I, Building 3500 Office of Executive Legal Oak Ridge, Tennessee 37830 Director i U. S. Nuclear Regulatory Dr. Kenneth A. McCollom Commission 1107 West Knapp Street Washington, D. C. 20555 Stillwater, Oklahoma 74075 Dr. Walter H. Jordan Chairman, Atomic Safety and Licensing 881 W. Outer Drive Board Panel Oak Ridge, Tennessee 37830 U. S. Nuclear Regulatory Commission Washington, D. C. 20555 1

[

Chairman Renea Hicks, Esq.

Atomic Safety and Licensing Appeal Assistant Attorney General Board Panel Environmental Protection Division U. S. Nuclear Regulatory Commission Supreme Court Building Washington, D. C. 20555 Austin, Texas 78711 Mr. Robert Martin Anthony Z. Roisman, Esq.

Regional Administrator, Region IV Trial Lawyers for Public Justice U. S. Nuclear Regulatory Commission 2000 P Street, N. W., Suite 611 611 Ryan Plaza Dr., Suite 1000 Washington, D. C. 20036 Arlington, Texas 76011 Lanny A. Sinkin Mr. Owen S. Merrill Christic Institute Staff Engineer 1324 North Capitol Street Advisory Committee for Reactor Washington, D. C. 20002 Safeguards (MS H-1016)

U. S. Nuclear Regulatory Commission Dr. David H. Boltz Washington, D. C. 20555 2012 S. Polk Dallas, Texas 75224 Robert A. Wooldridge, Esq.

Worsham, Forsythe, Sampels William Counsil, Vice President & Wooldridge Texas Utilities Generating Company 2001 Bryan Tower, Suite 2500 Skyway Tower Dallas, Texas 75201 400 North Olive St., L.B. 81 Dallas, Texas 75201 Thomas C. Dignan, Jr., Esq.

Ropes & Gray Docketing and Service Section 225 Franklin Street (3 copies) Boston, Massachusetts 02110 Of fice of the Secretary U. S. Nuclear Regulatory Commission Ms. Nancy H. Williams Washington, D. C. 20555 Projtet Manager Cygna Energy Services Ms. Billie P. Garde 101 California Street, Suite 1000 Government Accountability Project San Francisco, California '

1901 Que Street, N. W. , 94111-5894 Washington, D. C. 20009 Mark D. Nozette, Counselor at Law Roy P. Lessy, Jr. Heron, Burchette, Ruckert & Rothwell  %

Morgan, Lewis & Bockius 1025 Thomas Jefferson Street, N. W., .

1800 M Street, N. W. Suite 100 i Suite 700, North Tower Washington, D. C. 20007 Washington , D. C. 20036

_ L&w s.) Juanita Ellis, President ASE (Citizens Association for Sound Energy) 1426 S. Polk Dallas, Texas 75224 214/946-9446 2

_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ __ _ /