ML20133C655
| ML20133C655 | |
| Person / Time | |
|---|---|
| Site: | Catawba, 05000000 |
| Issue date: | 01/03/1984 |
| From: | Garde B GOVERNMENT ACCOUNTABILITY PROJECT |
| To: | |
| Shared Package | |
| ML20132B649 | List:
|
| References | |
| FOIA-84-722 NUDOCS 8507200547 | |
| Download: ML20133C655 (4) | |
Text
.2 4'
- k. \D t '
,. n, .
- . f (
U.S.HRC ,
/
UNITED STATES OF AMERICA j/
NUCLEAR REGULAT0gCgl.33I0g n 55 p
b BEFORE THE ATOMIC SAFETY E tifEECIhh SING BOARD M
(
In the Matter of ) ,
)
DUKE POWER COMPANY, et al . ) Docket Nos. 50-413
) 50-414 (Catawba ~ Nuclear Station, )
Units 1 and 2) )
LIMITED APPEARANCE STATEMENT 0,F
- BILLIE PIRNER GARDE GOVERNMENT ACCOUNTABILITY PROJECT My name is Billie Garde. I am the Director of the Citizens Clinir of the Government Accountability Project. The Government Accountability Project is a project of the Institute for Policy Studies, Washington, D.C. The purpose of the program is to broaden the understanding of the vital role of the public employee, private citizen, and nuclear worker in preventing health and safety dangers, corruption, fraud, waste, and other abuses. The Project is not an
" anti-nuclear" organization. The Citizens Clinic for Accountable Government of GAP responds directly to requests for assistance from community groups, local government bodies, and individuals who are facing difficult struggles -
against the federal government, large corporations, or other entities that
- seek to repress or intimidate the individual struggles for fairness and equity.
1
~
On September 14 of the year GAP filed a request with the Nuclear Regulatory Commission for an independent audit of the as-built condition of ;
J l 8507200547 850524 PDR FOIA g7 [
f BELL 84-722 PDR qL _77 3 51-/
. ~
O*
2 the Catawba Nuclear Power Plant. This would require that och of the safety systems of.the plant would be inspected by an independent and competent con-tractor. This has been done at troubled. nuclear projects across the nation--
Diablo Canyon received an Independent Design Verification Program (IDVP), at the Midland Nuclear Plant, the NRC required a complete reinspection of all -
. accessible hardware, at the Zimmer Plant a new construction completion reinspection program is awaiting NRC approval amid much debate of its adequacy. Three Mile Island and Marble Hill both required audits by independent contractors when obvious problems with potentially serious consequences became known. -
4 At LaSalle, Clinton, and Midland evidence of a contractor quality control breakdown in the Heating, Ventilating, and Air Conditioning System (HVAC) resulted in an independent audit by an outside finn.
The time such an audit would take and the cost of such an audit would be minimal compared to the damage resulting from a nuclear accident at Catawba . The request is simple and straightforward--and I submit the request to this Board for independent consideration. -
This Plant has serious problems. For nine weeks, this Board has heard testimony from Duke's workers, NRC staff, and Duke management officials about the condition of the Plant. That condition 'has been consistently described as " problem free" by the two entities which have the most to lose without this Board's stamp of approval--Region II staff and Duke Power Company. Unfortunately, for the public, the regulatory rubber stamp of the Regional Administrator and his Fesident inspector is not good enough to answer whether omnious questions left !
l by the testimony of welding inspectors have been answered. )
i
'O '
- . ~ .
There is nothing complicated about what happened at the Catawba Nuclear Power construction project--human error.
There are procedures written to certain specifications which are supposed to be followed in the construction of a nuclear power plant. Those procedures were not followed. The procedures required that every change to the design of the plant, every inadequate weld that was identified, every bad
, piece of pipe or steel that was found was to be written down, reevaluated, and replaced, if necessary.
. That did not happen at Catawba. Instead, construction managers made decisions that saved money and time for Duke.
.h So Catawba is done, ahead of schedule, but in an indetenninate condition.
,- The Regional NRC office would have the public of Charlotte believe that there is nothing wrong with Catawba--but they don't know that for sure. All they know is that Duke claims :that Catawba's safe. Duke also told the welding inspectors that all of their concerns had been taken care of. They told the public, just last Spring, that the problems raised by the welding inspec-tors were only complaints about pay. This was not true. The Duke welding inspectors of Catawba described the same type of situation at Catawba that another regional administrator described as unacceptable at an October 28, 1983, briefing of the Commissioners on Diablo Canyon. .
The Region V Administrator described a ' company that "was not documenting non-conformance reports properly." He said they
~
~'
- found a situation where, if a'n inspector considered something a non-conformance yet his supervisor disagreed with him, it never found its way into the records. ...our conclusion was that it is not really the right way to do business. If bO('l
- p. -
'U '
4 Q
someone has a complaint, say an inspector, and his supervisor concludes he is off base, there is no reason why it shouldn't all go in the records for everyone to see. I think otherwise
'we get into this very difficult situation where people develop a frame of mind where they were being overruled and they can't bring prob,lems up.
That practice was stopped in Region V. In Region III, at least two utilities .
and several subcontractors have had to pay hundreds of thousands of dollars in enforcement fines as a result of policies which did not document or identify construction problems. -
In Charlotte, however, the public is expected to becept a lessor standard of quality and to risk a greater chance of danger. This Board has an obligation
, to the public to require the NRC inspection and enforcement staff to insure thq$3 safety of the pbulic around Catawba. Bureaucratic pride, inefficiency, and blind faith in Duke Power Company are all stumbling blocks to a safe plant--something everyone involved in this case wants. Duke's reputation in the industry, its history of building nuclear plants, and its alleged commitment to the public health are not substitutes for an inspection to the as-built condition of the plant. The residents of Charlotte have a right to knowwhat is in the ground at Lake Henry, n.ot just what is supposed to be there.
e D* 39 g O
'laly ,
r bll V '
.'% UNITE D ST ATEs 2 .,
'~
NUCLEAR REGULATORY COMMISSION ds, h. x f,f g,5, pC
" " ' ' ' ~ * * " '
U.S.NRC N
g jgg 30 W'3 ?'i J!,N 16 Fall I'3 gsiGlhN i:- g L Ff [77g,ydru E ' "isr;. 2.iD
' O .E'.T.".~.;
January 13, 1984
&; )
7 .
MEMORANDUM FOR: Ben Hayes, Director P, , .}G. l p
Office of Investigations .,
George H. Messenger, Acting Director Office of Inspector and Auditor f' s FROM: James Lieberman Director and Chief Counsel Regional Operations and Enforcement Division, OELD
SUBJECT:
CATAWBA 9 2.206 PETITION The staff has received a response from Duke Power Company to the pending petition for action under 10 CFR 2.206 submitted by Billie Garde of GAP on behalf of the Palmetto Alliance. Because both your offices have been looking at matters raised in the petition, I am providing you a copy of Duke Power's response. I have not included the lengthy attachments, but will be happy to provide them at your request.
449 - M *74 ~
James Lieberman Jirector and Chief Counsel Regional Operations and Enforcement Division, OELD cc: w/o encl.:
R. DeYoung, IE I J. O'Reilly, Region II R. Heishman, IE f/ '
r k (
R. Purple, NRR f ,
CONTACT:
Stephen G. Burns, OELD //
y x27268 h '
j yho awwacsz -
du_to/.;m s u unma s y.z
i
- DUKE POWEH C OM P.iN Y .
Powra Bertnixo. Bo.x a3180. Cn Antorrr. N. C. 2e2 3s
- CaCN
. . . . . . u ..a :ic
.. .e...:.:......:.r.
,.,,s...s
-January 5, 1984 Richard C. DeYounE, Director Office of Inspection and Enforcenent U.S. Nuclear Regulatory Co= mission Washington, D.C. 20555
Dear Mr. DeYoung:
On Septe:bar 14, 1983, the Govern:ent Accountability Preject (GAP) submitted o document to the Nuclear Regulatory Commissioners which requested (1) that the NRC modify the construction per its for the Catauba Nuclear Station, Units 1.and 2, to require a " review by an independent contracter." GAP asked that this review include: (a) a "100* reinspection of the safety-related areas of the plant"; (b) a review of "the design deficiencies and the breakdewn in the design change control systems which render the . . . FSAR design . . . inaccurate cnd incocplete"; and (c) a review of "the quality assurance / quality ' control program which has existed with zajer weaknesses in the Catawba facility since the beginning of construction" (GAP Petition, p.1). In addition, GAP seeks (2) a "canagement audit of the Catawba upper and mid-level =anagers responsible
, for both design and i=plementation cf the Catawba (QA/QC) program"; (3) an investigation by 01 of "the deliberate mishandling by Duke Power Company manage-
=ent of certain serious complaints by Catsuba welding inspectors" to determine the existence of possible violations of 10 C.F.R. Parts 19 and 21 and 29 C.F.R.
Part 24; and (4) a 'tonnission review" of the engeing OLA investigation te assure that it encompasses the allegedly improper conduct of Region II officials (GAP ,
Petition, p. 2). -
Duke Power Company subtits that the requested relief is unwarranted. To assist the NEC, Duke has prepared a detailed respense to the allegations of GAP. This response is attached. .
Sincerely, .
ll if.l % * ,
i i
k'. H. Owen Executive Vice President l Engineering and Construction l WHO:=d gu -3 () " -
8 56 " Y _{-
attachment ,
y c,[ & f
r-t "
Tr.tv :
- Ceries J. P.;0'Reilly, Regicnal Administrator U.S. Nuclear Regula tory Commissien, Regien II 101 Marietta Street, N.W., Suite 2900 Atlanta, Secrgia 30303 James Lieber:an Diretter and Chief Counsel Regional Operations and Enforcerent Dieisien, OLED T.S. Nutlear Regulatcry Ccnmissien Washingten, D.C. 20555 Billie P. Garde Gevernment Accountability Project Institute.fer Policy. Studies 1901 Que Street, N.W.
Washington, D.C. 20009 S
he b
O w
b C h YO*
- . E ::.. : ::::'. i :!! :'.II ': G': ET:TIOr, I . suppcrt of its ecsest that the construction perrits fc- Catawba Units 1 and 2 te cocifiec to recaire a review by an independent centractor and that an aucit of the QA management at Catawba be conducted,2 GAP identifies five areas c' allege: "f ailures by Duke Power Corpany which p. cc"tedly demonstrate the "ctntinuing nature of the QA breakdown at Catawba" (GAP Fetition, p. 5). These areas of alleged deficiencies are as follows:
A. Failure to ensure that the As-Built Conditier. of the Plant Reflects the Final Version of an Acceptable Design E. Failure tc Maintain an Adequate Quality Assurance Program to
~
Ider.tify and Correct Construction Deficiencies C. Failure t0 Maintain Adequate Controls to Process and Respond to Nonconforming Conditions D. Failure to Maintain Adequate Msterial Traceability to Identify anc Docume the History of all Material, Parts, Components and Special Processes -
E. Failure to Maintain an Aaequate Quality Assurance Program for VenEors DPC will address each of these five asserted areas of deficiency below.
Because GAP relies heavily upon findings from Duke's Self-Initiated Evaluation 2
Duke addresses the need for an 01 investigation anc the need for Commission review of the ongoing OIA investigation at pp. 51-53.
Yst 67
( ' 5
- E ' ), ' as t e casis for its ailegations, DDC's respcase als: fcceses largely
- c. the SIE fincings.
In this regard, it is important to note that the SIE items referenced by GAP were included in an inquiry by an Atomic Safety & Licensing Board empaneled to rule en Duke's application fcr an operating license fer Cataca. An intervenor in the licensing proceeding, Palmetto Alliance, relied upon Sections of the SIE to support a motion to extend discovery in the operating license proceeding and in essence to expand the scope of .the admitted QA contention to embrace all aspects of QA. The Licensing Board, in an effort to understand the import of tne 5:E, called for a presentaticn by 11 members of the 5:E e.aluatier. team, 5 frem Duke, 5 f rom TVA, and 1 from INPO. This presentation is set forth in Transcript Pages 20053-10276 and is attached hereto (Attachment 1) for the convenience of the decision makers. Significantly, this panel stated that the observations and findings regarding design (DC), construction (CC), testing (TC), and quality procedures (QP) set forth in the SIE Report (which includes those relied upon by GAP in its pleading) are not " reflective of a systematic breakdcan in the quality assurance program at Catawba" (Tr. 10153-5) and "do not reflect any practice which did or would have led to ur.sa'e construction or operation of the plant" (Tr. 10064-10069).
2 During September - November 1982, separate Self-Initiatec Evaluations were conducted on the design and construction act'ivities of all nuclear power
- plants under construction. The evaluations were conducted using criteria a'nd performance objectives developed under the direction of the Institute of -
Nuclear Power Operations ("INP0"). The evaluation criteria and performance objectives were established as standards of excellence which, if met, would result in a product that clearly exceeded regulatory requi reme.its . The evaluations were designed to point out potential areas of weakness for the utility inspected to evaluate and determine if corrective action was warranted. The evaluation team for the Catawba evaluation consisted of 18 utility personnel, 9 from Duke and 9 from Tennessee Valley Authority.
(Tr. 10053-71) f
- ncted, t* -
E.ans, the CatFaba SIE team leaCer (now a full-time INPO employee) stated trat tased on his involvemer.1 with 9 other evaluations of nuclear plants uncer construction, the Catawba SIE reflects that " Duke's programs clearly exceed all but one of the other plants evaluated, and I would consider Duke's programs approximately equal to those of the other plant". (Tr. 10063. See Also Tr. 10263.)
It should be noted that the statements of this panel were subject to extensive Board questions (Tr. 10071-10136 and'10260-10274) and examination by the parties, including Palmetto Alliance (Tr. 10162-10230).
Af ter considering the record on this subject and the arguments of counsel, the
~
Licensing Boarc denied Pa1metto Alliance's request. This ruling is set forth in the in camera Transcript pages 948-954 and is attached hereto (Attachment 2) for the convenience of the decision maker. In sum, reliance upon the SIE as~ a basis for commission action has been rejected by an Atomic Safety & Licensing Board and should likewise be rejected by The Commission.
A. " Failure to Ensure that the As-Built Condition of the Plant Reflects the Final Version of an Acceptable Desion" In Section A of its Petition, GAP attempts to demonstrate that:
Catawba design documer.tation does not reflect the plant as designed, and it is unclear whether it reflects the plant as-built.
Substantfll documentation from Duke Power itself, and confirmation from workers leaves no doubt that Catawba's design .and field engineers built this plant 'by the seat of their pants,' not b9 the book. (Petition, p.6.) ,
(c of 65
l E Elt: c- t * 's , G3 affe ts t*at the Cata-ca C a.t cces nct satisfy 10 C.F.R.
l-Fart SC, Appendix B, Criterior III.2 I
In support of its assertion, GAP raises issues related to the SIE, allegations of Mr. McAfee, and concerns related to variation notices. These issues are accressea belo.:
8 Criterion III reads:
Measures shall be . established to assure that applicable regulatory
' requirements and the design basis, as defined in Paragraph 50.2 and as specified in the license aDplication, for these structures, systems, and components to which this appendix applies are correctly translated into specificaticns, drawings, procedures, and instractions. These ressures shall include provisions te assure that appropriate cuality stanca-cs are specified and included -in design documents and that deviations from such standards are controlled. Measures shall also be established for the selection and review for suitability of application of materials, parts, equipment, and processes that are essential to the safety-related functions of the structures, systems and components.
Measures shall be established for the identification and control of design interfaces and for coordination among participating cesign organizations.
These measures shall include the e stabli s h,Te nt of procedures among participating design organizations for the revie=, approval, release, distribution, and revision of documents involving design interfaces.
The design control maasures shall prcvide for ve-ify ing or checking the adequacy of design, such as by the performance of desig reviews, by the use of alternate or simplified calculational methods, or ty the performance cf a
, suitable testing program. The verifying or checking process shall be performed by individuals or groups other than these who performed the original design, but who may be from the same organization. Where a test program is used to verify the adequacy of a specific design feature in lieu of other verifying or checking processes, it shall include suitable qualification testing of a prototype unit under the most adverse design conditions. Design control measures shall be applied to items such as the following:
- reactor physics, stress, thermal, hydraulic, and accident analyses; compatibility of materials; accessibility for in-service inspection, maintenance, and repair; and delineation of acceptanc6 criteria for inspections and tests. -
Design changes, including field changes, shall be subject to design control measures commensurate with those applied to the original design and be approved by the organization that performed the original design unless the applicant designates another responsible organization.
O l
~
96($[
7 l
l ~
l
.. 5:E; In su; crt of its pcsitice., GAP cites various sections of the Self-l Initiatec Evaluation related to the area of cesign, e.c. , findings DC.1-1, DC.1-2, DC.1-3, DC.1-4, DC.1-5, DC. 3-1, DC. 3-2, DC. 3-3, and DC. 3-4, and 5 observations. (Petition, pp.E-8.) While each of these sections are addressec in cetail below, it sneele be ncted that SIE team members (i.e.,
Messrs. Hendricks [ Duke), [TVA),
Clift and Belew [TVA)) who were responsible for conducting and writing the design evaluation section of the SIE upon which GAP relies, each stated before the Catawba Licensing beard that the findings and observations set forth in the design (DC) sections of the SIE did net " reflect any practice which did er wcule have led to ur.saf e constructier. cr 0;:eration cf the plant". (Tr. 10064-9.)
With regard to the specific sections o' the SIE relied upon by GAP, Duke's response is as follows:
- a. DC. 1-1: No requirement exists for identifying, tracking, and assuring that commitments icentified in the PSAR are met. Design inf or.fa tion shoulc be clearly cefined and controlled. (Petition, p.6.) ,
Af ter analyzing the bases fer the findings, DPC cetermined that adequate controls existed
- to assure SAR commitments were controlled.5 Howe $er, DPC further determined that a more formal program would be of benefit.
4 D' uke Power Company's past methcd fer control of SAR commitments has been as -
follows:
- 1. 1.istings of commitments mai'ntained by Design Engineering have been i used to identify, track, and control:
(continued)
L .
Ve( 52
4 ::( :' * ;'y , ?"'s . 'ce-EvesiCent Cesigr' Engineeri >g i .i;iated a task 1: :e in March 353 to ; resent recom endations for a'ternatives for an upgracec cci.itrent tracking ;rogram In April, 1983, the task force made recommendations to the Vice-President
,es'gr Erginee ri .; fcr the impleme-tation of a prcg-a ef design criteria specifications. This program would include compilation of all Design and 4
(continued)
- a. PSAR commitments, outside those embodied in normal engineering practice.
- b. Ccmmitments sent tc the NRC af ter F5AR approval and prior to submittai of FSAR.
- c. CommitmNtsidentifiedpriortopreparationoftheFSARtobe included in the FSAR.
- c. Deviation from PSAR commitments.
- e. Deviations from FSAR commitments. These deviations are included in licensing documents via normal SAR and license amencments.
~
- 2. Commitment and deviation listings have been distributed to Design Engineering Section Heads and updated on ne less than a quarterly basis.
- 3. All SAR and Regulatory Commitments are reviewed by appropriate levels of Duke management prior to subrittal to NRC. Quality Assurance procedures require the consideration of SAR and other regulatcry commitments in the preparation of calculations, design specifications, and related documents.
5 Signi ficantyy, the SIE was not designed to co'nduct, in all cases, in-death evaluations of a re.a s of potential weaknesses te determine if problems existed. (Tr. 10062.) Rather, the SIE took the conservative approach and pointed out to the utility all " observations which reflected a potential for weakness". (Tr. 10063.) It was up to the utility to conduct further reviews to determine if the factua'l observations had significance. Id.
Indeed, based on subsequent and more in-depth review by the members of the evaluation team preparing for the presentation to the Catawba Licensing Board, it was determined that "many" of the findings which appeared to point out weaknesses were not totally valid. Tr. 10151. See e.g., Tr. 10076-7; 10078-9; 10080-1; 10084-5; 10085-7; 10088; 10115-9; 10128-9; 10133.
%pil
- 5;r cc ritte.ts for Cata ta Nuclear Sta tc-1 4 cr This recor.encation o - C '. i C e c for a prograr to meet the rec Ch?.e n Ca t i C- Cf the Self-Initiated E.aluation. This reccmmentatier was approved by ra .a;e .ent and is scheduled for full impler.entaticn on Catawoa by May 30, 1964.
- b. DC.1-2: No control program for defining responsibility for providing Desigr. input could be found. I .put is usually providec on a request basis. (PetitiCD, c. 6.)
The thrust of this finding was that no documented procedures for providing design input were apparent. In response to 'this finding, a comprehensive rev,iew within each line division was conducted. This review determined no procedural changes were needed, based on the following policies and practices:
(1) The orga izational responsibilities in the Design Engineering Department are formally documented in responsibility statements. In addition, significant commitments to provide design input are included in an integrated schedule by responsible groups.
(2) The Design Engineering Department Manual docurents a Civil / Division interface concept with other divisions which is ve ry specific regarding required structural design inputs. Design input da'te s appear on design schedules, and information is submitted formally on marked sepias of structural drawings. This program is formalized, well contrc11ed and well understood, by responsible engineering supervisors in the Mechanical & Nuclear Division and the Electrical Division. The Civil / Environmental and Mechanical & Nuclear Divisions also use specific interface ayreements to define responsibility for various aspects of pipe stress analysis and support restraint design.
/0 e{Si
(2) T r.e orga.izatic.ai stru:ture cf t.* e Me : r.a r. i c a l 6 c reclear Divisic-pec. ides a logical flow of cesig, inputs which progress from flo.
diagrams , to ecuipment data sheets, to e:;uipre .t specifications, to piping and equipment arrangener.t designs, to formal documented syste.?
checkouts on a scale model, to final system verification analyses in a carefully centrclied p-ecess. The t'.echa nical/El e ctrical instrumentation design process provides the necessary information through flow diagrams, I&C data sheets, I&C details, I&C lists,
. electrical elementaries and wiring diagrams.
(4) The Electrical Division has nu .erous work place precedures to address administrative contrci. Additionally, this division defines responsibility for providing cesign input through its organizational structure by providing a controlled, logical flow process in performing the electrical designs. For inii.ial design, t'he Electrical Division makes use of the integrated schedule as a valuable aid in defining, controlling, and scheduling inputs to the design process. This mechanism is used for internal division inputs and schedules as wall as defining and docume. ting needed design inputs from other divisions.
(5) Once initial design is completed and released for construction, e
revisions t.o system design are authorized by Design Change Authorization Forms and later by Nuclear Station Modification Requests. These revisions are not entered into the integrated schedule but are tracked by a Data Base " Punch List" maintained by the Project Management Division. This " Punch list" is used to define l
/nfM '
e paeties inscivec in the revisic. alcng with ces'g, ingets (I&C information, vencce drawings, etc.) and serves as a netho'l te define and contr:1 inputs to the desicn process,
- c. DC.1-3: Design input information is not always provided in a controlled manner. Memeranda serve as the primary vehicle f e* cccutenting Design inpet. (Petition, p. 6.)
Upon reviewing this finding, DPC determined that no corrective action was required because the procedural controls currently in use at Catawba in this area are adequate to provide controlled design input information.
Specifically, QA procedures are in place for the control of design input informatien through the use of calculatiens, specifications, drawings and correspondence. In adcition, reviews of the as-built condition of the plant provide a final verificatbn that correct design inputs are used.
As to the finding that "[m]emoranda serve as the prima ry vehicle for documenting design input," DPC evaluated this clairr and found it to be inaccurate. While correspondence is sometimes used to transmit design
~
information from one design group to another, it is not the primary source of design input information. Moreover, it is controlled by QA procedures wh'ich establish responsibilities in issuing design correspondence.
- d. DC.1-4: System descriptions and flow diagrams do not always agree as to the current' requirements. Several system descriptions were observed to lag revisions of system flow diagrams. (Petition, p. 6.) -
The fact that system descriptions are not always kept current with flow diagrams and electrical elementary- drawings has absolutely no safety significance. System descriptions are prepared very early in DPC's design process. They are intended as preliminary and general working documents which 9
/24fQ
prctice a cuerview of the COTc; P.ent! and fun;t*C* c' a ;a" tic.'a' system, anC .
as a genera' referente in initiating the mere a ci a"> Cec de s ig". C # a system.
They are used as " supporting dccuments" or g.ites in the preparstien of the more detailed flow diagrams and electrical elementary dra.ings. These flow diagrams and electrical elementary drawings, rather than the system cescri:ticns themselves, are the controlled oe s igr. cc:; erts w-ic* ars the sources of information used in the detailed design of a particular system.
These controlled design documents are updated in a timely manner.
Because of their limited and preliminary role as a g.rideline for developing ac anced system designs, scte system descriptiens were not updatec regularly before the Self-Initiated Evaluation because this was not cer.sidere: a high priority matter. In response to the evaluation team's findings, ho.ever, DPC's Design Ergineering Department agreed to implemer.t varicus corrective rneasures to update system descriptions and to keep them current. ~
These measures have been implemented.
In short, in that system descriptions are not considered contrclied design documents, the observations of the evaluation team with respect to system descriptions have no bearing on the safety of the plant's construction or upon the quality of its design. Any lag in updating system descriptions has no impact on the final quality or accuracy of station designs, test, and cperating procedures, of safety of the constructed plant.6 8 On page 7 of its Petition, GAP cites five instances from the SIE in which it was discovered that " system descriptions and diagrams did not agree."
Similarly, on page 8, GAP quotes from finding DC.2-1, stating that " design interf aces were found to be lacking in that 'the process for controlling (continued)
/$ 1(S4
- e. DC.1-5. Nc dccc ertec pre; ai was f o ur.d for ass -ing correct a;%icatic- cf se's-ic re s;c a.s e spectra. -
(Petition, p. 6.)
It is DPC's pcsition tnat adequate precedures are in place to contrcl the use of seismic response spectra. Seismic respense spectra and building disclacements used for design or qualification of safety related structures, syste.s and cc~;cnents are co.t-clied by i s s uing t*.et. in a specification.
Included in this are general instructions for the use of these spectra,
. including specification explanations of how to ~cbtain OBE or SSE spectra where only one is supplied and information on the areas of buildings for which spectra are apclicable. Specific instructions on methods of seismic analysis or testing. anc dampir.g values are nct included because they very depending on the type of structure or compone"t. This information is included ir design specifications or procedu es for specific structures, systems or corrponents.
This finding indicates a concern related to documented procedures but does not find any indication of associatec design deficiencies. The response to this finding was developed by an interdivisional committee and reviewed by all Chief Engineers. Neither the response to this finding nor an earlier internal seismic review conducted by Design Engineering, identified any indicatien' of generic design deficiencies. .
8 (continued) ,
i design input documents does not require timely updating of system d'escriptions.'" -
Again, on pages 8 and 9, GAP cites additional statements and SIE findings (DC. 3-3 and DC. 4-3, which are quoted in the Petition) which relate to the updating of system descriptions. DPC's response to DC.1-4, above, is also applicable to all of these additional findings.
Eestc*,se s;ect'a sctcificat cns are Cistributec by paccecJres contrcliec by the Q
- prcgram. Res'.sicns to respc*:se spectra specificaticns have been generated as necced; bat, there ha.e bec nc significant revisions to spectra for the station. This further minimizes the possibility that spectra were used incorrectly. It is Duke's position that adequate procedures are and have been in piace to centrci the a se Cf seis.?ic response spectra.
- f. DC.3-1: Design records are not always being filed in a timely manner. Waiver form originals on file in a manual had not been transeitted to General Services for Corporate filing. (Petition, p. 8.)
The drawing waiver form referrec te in this finding cocuments which Divisions have waived the requirement te inspect a drawing. The purpose of the inspection by other Divisions is to assure that all design interface considerations are included. The waiver of this inspection indicates that the interfacing departments have determined that there are no design interfaces .to be considered on the waived drawing. The quality of the plant's design would not have beer impacted in any way had this isolated deficiency gone undetected.
If no copies of the particular drawing waiver could have been located, the only consequence would have been unnecessary inspection sign 6ture(s) on that category of drawings. ,
Nevertheless, despite the fact that there are ne safety implications associated with this firfing, DPC tock appropriate measures to insure that this type of incident would not recur. The QA Procedure addressing drawing waiver forms was revised to clearly define respcnsibility for filing the drawing waiver forms.
- g. DC.3-2: Changes to Design drawings do not receive the same degree of documented review as the original issue.
Originals are documented by designer, drawer, checker and
acDrcser. Etwisic s are cccume".tec t cnecker a n;:
approver. ( E s t 14 c r. , p. E.)
Tr 's fia. ding emphasi:Es that criginal issue drawings rec. ire the signatures of t'.e designe*, drawe , checker, aa,c approver, while revis4:*s tc design drcwings only require the signatures of a checker (verifier) and an approver. The point here, is i t;; rta r. h0.ever, that all dra,ci r.;s receire a design verificaticn (checker) and approval. Accordingly, the quality of design is not brought into question by this finding. Although DP"'s original response to the SIE indicated no corrective action would be taken, Design Engineering has recently decided to add a space for the originator of drawing revisions to sig., in order to provide mcre complete documentation threughout.
- h. DC.3-3: Design documents relating to the cesign of the RHR System are in disagreement creating a pctential for error in the design. This is also true for the AFW system.
(Petition, p. 8.)
This finding is addressed in response to d above. See ncte 3, supra.
- i. DC.3-4: Calculations are not being raintained in a controlled manner that support [s] issued Design documents.
(Petition, p. 8.)
Finding DC. 3-4 pertains to certain Mechanical Group De:artment calculati'ons not being in the required documentation format. 'Specifically, this finding dealt with the lack of documentation in the calculation files for the Auxiliary Feec ater System.
Although the supporting calculation for the Auxiliary Feecwater System was not in QA format in the document file, it had been prepared and was in the sponsor system engineer's file. This calculation had verified that the Auxiliary Feedwater System met its design basis.
4 d 59
Oct tg ' e s t,I s i r.; .E; "e c ' r e c ha ni c a l systens ces'; s. it is ir: practical to
,p Cccu"'E .1 PC"ticSs cf a s y s t e.'T design as i t is ce.elC~ec. InsteaC, it is s ~( ,
pruuent ,tc check pe-iccically to see that the total system meets its design
- f tils i s . 'Once _the design is firm, equipment cata, piping layout, and other
. functional,~ d ' ign teouirements are verified in a fina' system verification calculatibr.pict to reacter fuel icac, as reouirec b., ANil N45.2-11.
DPC determinec that the only corrective action needed was to formalize the final system verification procedure and to schedule completion of final design calculations for all systems. Final system verification is sufficient to assure that nc safety cer.cerns exist.
- j. Five acciticnal SIE observations. (Petitic., p. 7)
On page 7 of its Detition, GAP cites excerpts from five observations found in the SIE to suppory its assertion that tnere are " design control deficiencies" at Catawba and that "there can be no question that work at Catawba has large'ly proceeded on theJoasis cf informal drawings and procedures instead of design changes approved $$, the project engineers."
Of these five observations, four are related to' tne findings discussed above -- specifically, observation 1 relates to DC. 1-1, cbservations 2 and 3 relate to DC.1-2; and observation 4
,m -
relates to DC. 1-3. Thus, the DPC respenses set forth above apply to 't,hese
(.
observations as well as to the related fincings.
x .
V The only obsenation which has not already been discussed and resolved above is l' item F, which states that ". . .no program requirement for conducting
- t
< constructibility, maintainability, or operability reviews was found to exist. . ..
This observation relates to SIE finding DC. 4-2, which is as follows:
e
- 14 -
l7c(S8
~;.4-2; hc docu er.te:: prograr was iceitified fo" deterrining and assuring re.iew fc" ccnstruc;itility, :taintainability anc c;erability is conductec. Reviens are performed where a need is identified.
DPC reviewed this finding and determined that no corrective action was necessary because adequate reviews for constructibility, mEintainability and operatility have been conducted during the design and construction phases of Cata ta Nucisar Station. There are three primary programs for s uc r. review:
- 1) Scale Model Reviews, 2) Composite Drawing Reviews, and 3) Piping Design Criteria Reviews.
All three of these programs are well documented and there is participation in the reviews by all af fected Company organizations. The follo ing is a synepsis of each program:
(1) Scale Model Reviews - For Catawba Nuclear Station, a scale model was constructed (3/8" = 1 foot). This model consisted of 33 model tables and included the plant structure, mechanical and electrical equipment, piping, HVAC and cable tray.
As the model was being constructed, periodic checkouts were held to review each area of the plant. Representatives' from DPC's Construction Department, Nuclear Production Department and Design Engineering Department attended these reviews. All aspects of the plant's design were
, reviewed, including constructibility, maintainability and operability. ,
Any problems detected by the reviews were documented and coordinated with appropriate parties f or - resolution.' Each model table is scheduled to be reviewed four times during the design and construction phases of the project. To date, three reviews for each area have been completed.
/$CJ M
I r. acci*iC&. special enlarget i.cce's Cf c Sgeflec areas were Constructed when receired.
Tr.es e models us .;a l ly inc hcet all designed and field
-located compcnents (including instrumentaticn cempccerts, cables, tubing, support / restraints, insulation, etc.). The fcur-tier review process for these models is the same as noted above.
(2) Composite Drawino Reviews -
The composite drawings for Catawba Nuclear Station consist of a series of orthographic crawings for the reactor,.
auxiliary, turbine and service buildings. Each ccepos.ite drawing shows all the disciplines (structure, equipment, HVAC, pipe, cabletray and support - restraints) within a given area. Alsc, the composite drawings depict required maintenance access space, ecuipment maintenance space or other special space Nguirements.
As design work is released, a review is done using the composite drawings.
During this review, the design is checkec for maintainability and constructibility. Any problems detected are documented and coordinated with the appropriate parties for resciution.
The above program for reviewing composite drawings has been particol,arly beneficial in the area of support restraints. In congested areas of the plant, it is very difficult to design the support restraints due to the space l i?. tati ons . On Unit 1, each support restraint design is located and verified .on the composite drawing. On Unit 2, each support / restraint design is located, verified and the required space envelope is shown on the composite drawing. All of this is done prior to release of the
~
_ . . N f(5$
s.:::rt restraint c-a ings : Cc .structier. ar.c helps to prevent many field protlers.
(1)' Fioine Desicn Criteria Reviews - The Piping Design Criteria is a recent development (late 1081) in DPC's piping design process. There are specific criteria on Ccnstructibility (PDC-020); Maintenance Considerations (PDC-100); and Operability (System and Equipment Considerations [030) and In-Service Inspections [070]). When all of the criteria were developed, constructibility, maintainability and operability were considered and incorporated into each criterion. Also, affected groups within the Company reviewed and approved the criteria which applied to their areas.
Since the Piping Design Criteria is a recent development, these criteria were not utilized on the initial design efforts for Catawba. However, the criteria are now being applied on new designs and design revisions where appropriate. Prior to the development of the criteria, good engineering judgment was utilized in the design process. Now, the Pipiig Design Criteria prevides DPC with a documented design process which produce's a consister.t approach to piping design and problem resclution. It is,our opinion that the Piping Design criteria assure that constructibility, maintainability and operability are considered in the design process and that an' appropriate review is performed to verify compliance with the
. criteria.
In summary, DPC believes that the three programs outlined above assure appropriate review for constructibility, maintainability and operability.
~
76 0/66'
>(it prcgrars alle. fcr inpct f rcr af f ecte:: crganizaticns within the Company at sarious stage! of the project and alsc allom for continucas review of the cesign as revisions are made.
- 2. McAfee allecations On page 9 cf its Petiticn, GAC refers to the cepcs' tion cf Ron McA'ee as supporting evidence of design and construction weaknesses. Mr. McAfee's concern was that drawings were revised after inspection in order to conceal a " construction foul up" (See McAfee deposition, Tr. 41), and that this " totally backward approach to design control" is evidence of DPC's deficient approach in constructing Catawba. These assertions are misleading.
There was a period of time at Catawba (prior to 1980) when inspectors had difficulty in understanding design drawings. This dif ficulty was the result cf limited experience by the Catawba inspectors in interpreting
. . design drawings and the lack of a tolerance on some designs. When tolerances were not specifiec and craft varied from the drawing due to field conditions, this caused inspectors tc ask nute cus questiens which led to many NCIs. In each of these cases, the actual installed condition was analyzed and approved. Because of the numerous questions and NCis in this area, Design Engineering issued a tole,rance specification to define e
allowable variations from design drawings. In addition, Design Engineering reviewed all electrical hanger drawings to identify potential conditions that could lead to inspettor questions. Both the craft and the inspectors were trained in the proper interpretation of the tolerance specification and design drawings. In addition, all electrical hangers 18 -
7/o/ SS
.ere reinscectec : t r.e s e r e q. ' r e me r.t s . The N;: reviewed the action taken, fcunc it tc be complete anc suf ficiert, anc closed the item on E-2-80 (NRC Inspectier. Report 50-413/80-23).
DPC submits that, to the extent they were valid in the first place, Mr.
McAfee's concerns have been mere than acequais'y accressec by these corrective actions. GAP f ails to acknowledge any cf these actions in its Petition, instead confining itself to unsupportec insinuations about the
" quick and dirty approach used by DPC management to build Catawba."
(Petition, p. 10.)
- 3. Variation Notices GAP asserts on page 10 of its Petition that "a review of all the Design Drawing and Specification Variation Procedures used for design control affirms the worst fears of Cata-ba's critics." The basis for this unwarranted assertion is unclear. GAP makes much of the f act that in, September, 1976 QA Procedure R-3 was revised to change responsibility for control of Variation Notices from Project Engineering to Project Manager.
However, there was nc substance whatsoever tc this change. The wording in the previous revision stated that "the Project Engineer, Senior Construction Engineer, or someone of higher authority within the project management organization" had responsibility, for approving VNs. Since all
, of these individuals reported to the Project Manager, the wording change was made merely to simplify the procedure.
4 GAP further alleges that " prior to that transfer" no meaningful QA/QC review of design changes (VNs) occurred until May 1,1974 May 1, 1974
1 ccr es;;.ncs tc t it formation of the OA Depart.ent as a separate ce artment. PriO" ~ to this time, therefore, there as nc QA review of cesign changes since the QC assignments were made by the responsible field engineer. This cces not mean, however, that design changes were not reviewed before 5/1/74. The review was performed by another department.
In additicC, safety related Construction on Cataati C c nct begin until after May, 1974. Therefore, no VNs were issued on Catawba before May 1, 1974.
GAP goes on to assert that design control procedures at Catawba " remained inacequate throughout the decade" (Petition, p. 10). In supocrt of this serious allegation, it cites Revision 7 to QA procedure R-3 and suggests there is something " inadequate' about restricting use of Variation Notices (VNs) to those cases where their use is essential to maintain work in progress or work soon to begin. This restriction was placed in the procedure to-preclude V'Ns from remaining outstanding for long' periods of time. In cases where work is not imminent, drawings are revised prior to the beginning of work.
In the same vein, GAP refers to Revision 8 of QA Procedure R-3 (GAP Attachment
- 6) and implies that there is something wrong with construction and design implementing a field change prior to formal revision of the approved design document. The Variation Notice is used by DPC to get design apprcval of field e
changes prior to actual drawing revision. This process is controlled in that the Variation Notice is held open until the drawing is revised and verification is made that the final approved design , reflects the as-built condition of the plant.
e e
Or ;Egt . .
GM refere ces E.Evision 13 tc QA Procecare R-2 (catec 1/11/62) as D: . e 's first attempt at addressing reportability of Cesigr Ncnconformances.
This is inaccurate. Ir. fact, Duke had in place in August 1975, Q* procedures tc accress reportability in accordance with 10 CFR 50.55(e). Tnese procedures aoplied to all items during the design and construction of Catawba which were re:ortatie te the NRC in accordance with 10 CFR 50.55(e). By January 1918 our QA procedures also fully incorporated the reporting requirements of 10 CFR 21.
These reporting procedures required all persons detecting an item or event which they believed to be reportable to bring their concern to the attentien of management. This requirement was posted. The procedure referenced by GAP merely adds documentation of the previous requirerent. GAP further contends that this improvement was short-lived when Rev. 17 was issued, exempting
~
certain VNs from review as a design nonconformance. This too is misleading.
The VNs that were exempted were deemed to be minor in nature and not potential design nonconformances.
GAP concludes Section A of its Petition by alleging that "af ter a review of the design procedures, the INPG findings, and the experiences of Catawba workers who have talked to GAP," there can be "no reasonable assurance" that the pTant is built as designed, and that design changes satisfy 10 CFR Part 50, Appe.ndix E, Crite-ion III. (Petition, p. 12.) As DPC's response to the various allegatiens made in this section has made clear, GAP's assertions are totally
, without foundation. They appear to be based upon the petitioner's incor.plete and, in some cases, incorrect understanding of the SIE findings and their implications and of OPC's QA procedures. This irresponsible approach, coupled with GAP's vague and unsubstantiated references to "the experiences of Catawba workers," and its gratuitous and unsupported comparison of Catawba with other Map 68
cla .is
.;; e a e -ith w wicr Cl.: has been insolvec, fu-tne* undee.p ne the crecibility c' this Petitier..
Contrary to the allegations set forth in Section A, DPC maintains that we have a comprehensive and well-developed program which ensures that the as-built cor.cition of Catawca agrees ith the final appresed cesig.s.
B. " Failure to Maintain an Adecuate Quality Assurance Procram to Identify and Correct Construction Deficiencies."
On page 13, GAP begins a series of unsubstantiated allegations regarding the Quality Assurance program at Catawba, asserting that the " premise" of QA has been "fix it first, fill in the paperwork later," and that " violations of regulatory requirements were common." These claims are unsupportea by any evicence.
First, _ GAP refers to " continued low quality assurance /cuality control ratings by the NRC" as evidence that Catawba has not been constructed in accordance
~
with Appendix B, citing the 1981 SALP report and a draft of the 1983 5 ALP repcrt. In considering the 1981 SALP report, it should be recalled that the time period covered by this report coincided with a period of extremely heavy ,
. construction activity at Catawba. We believe this accounts for a greater number of construction deficiencies and violations than may have been written at other sitcf during this period, and that th'i s wa s a facter in our below average rating.
In DPC's view, this rating was not justified.' In any event, a "Below Average" rating does not in itself, indicate systematic or significant deficiencies. In explaining the rating system in the 1981 report, the NRC stated: "[a] rating of
1 l
- E := a.Erage CCe5 nCt near. that a facility was unfafe c" that its operation o" c vstructicn should be stoppeo."
As to the 1983 5 ALP Report, GAP's remarks are incorrect. The final 1983 SALP report gave Quality Assurance at Catawba a " Category 1," rather than a Category 2, rating.7 GAP also refers to "a series of critical reports by consultants." (Petition, p.13.) That 00 further attempt is made to substantiate this damaging reference is perhaps not surprising, since it lacks any basis in fact.
Tne gist of GAP's concern about DPC's QA program (described by GAP as the
" fatal flaw" in the program) is that the QA program "is not, and never has bee"., independer,t of construction."
By the time safety-related construction of Catawba began in November,1975, DPC had (in May,1974) established a separate Quality Assurance Department, which reported independent of Construction to the Senior Vice-President Engineering and Construction. At that time, it was decided that QC inspectors sh$uld remain in the Construction Department for administrative purposes; however, they were under full functional control of the QA Department. That is, the Construction Department was responsible for timekeeping, facilities, 7
" Category 1" is defined as:
Reduced NRC attention may be appropriate. Licensee management attention and involvement are aggressive and oriented toward nuclear safety; licensee resources are ample and effectively used such that a high level of performance with respect to operational safety or construction is being achieved.
?( rf V
s:*ecaling. Lay cll, Etc., while the QA Depart en was rsspensible for training a r.c certificaticn of inspectors, providinc inspection procedures, and evaluating and a;; roving inspection records. Or qualit;. matters, the Quality
. Control personnel were under the direct control o' the Quality Assurance Department.
On page fifteen, GAP states that DPC President anc Chairman of the Board, William 5. Lee, "neither respected nor implemented the NRC instructions to split ,up the QA/QC function from construction and engineering," adding that this was in " blatant disregard" of the laws geverning commercial nuclear plants. This accusation is unfcunded, as evidenced by NRC approval of Amendment 2 of the Duke Power Quality Assurance Topical Report on April 17, 1975. In its letter of approval the NRC' stated:
Based on our review and evaluation of Duke-1, we conclude that:
- 1. The organizations and persons performing QA functions within DPC have the required independence and authority to effectively carry out the QA Progrtz without reservation or undue influence from these directly responsible for ccsts and schedules, and
- 2. The DPC QA Program contains the necess ary requirements, procedures, and controls to demonstrate that quality related..
activities will be conducted in accordance with Appendix B to 10 CFR Part 50.
On page 16, GAP states that Mr. Lee " chose administrative convenience over regulatory re'aui rements for nine years," and further contends that "the original organizational structure of DPC [ presumably GAP is referring'to the QA Department] continued without accountability until 1981. . . ." In fact, Duke was in compliance with applicable NRC regulatory requirements for this entire period. Administrative convenience was a consideration only within the
cc st ai .ts cf resting the regulatcry receire c.ts.
This ptint is u.cerscorec by the fincings of the various Licensing Ecards a: A; peal Esards which resiewed and approsed the QA structure which GAP calls inte questicr. . (See, i.e. Duke Power Company (William B. McGuire Nuclear Station, Units 1 and 2),
ALAE-143, 6 AEC 623, 625 (1973.))
GAD next cites the concerns expressed by welding inspectors as evidence of pressure from construction. (Petition, p. 16) The welding inspectors'
. concerns were thoroughly investigated by DPC and were found to stem from poor channels of co .munication, primarily related to the lack of adequate feedback tc inspectors when werk rejected by them was later evaluated as being acceptable. In assessing these concerr.s it must be recalled that the job of QA inspectorsistoreportdbiationsfromspecifiedrequirements. The resolution of reported deviaticr.5 is the responsibility of cthers within the organization.
The inspectors' natural reaction to this lack of feedback was to interpret it as lack of support and "giving in" tc construction pressure. Our evaluation revealed no ur. safe conditions.8 However, we did strengthen our communication channels ano upgrade some procedures in response to their concerns.
On page 17, GAP states "even outside consultants failed to convince. DPC management of the failure of its QA Program." This assertion is misleading s
8 During the Catawba operating license evidentiary hearing held in the fall of 1983, testimony was presented by every welding inspectcr who had raised
- concerns. (The Board ruled that much of the testimony was cumulative and thus, admitted into evidence the testimony of approximately one-half of the welding inspectors.) Each welding inspector concluded that, despite whatever problems he might have experienced, he did his work correctly and that based on the work he performed, the plant was safe. Further, no welding inspector stated that construction pressures affected his ability to do his job.
(
si :E n; o.ts4 CE CCnsultant, inclucing "a ace e*st Anal s's 'cTpan) (MAC), has esE- cc.,cludec that CEC's QA Program hac failed.6 G'F further criticizes MAC for dividing the inspectors' cc cerns into technical and nontechnical areas, and for failing to address the " nontechnical prograr.T.atic fl aws" (Peti tion, p.17). In fact, the ressc- for separating the concerns into technical and nontechnical areas was to enable technical experts to address the technical concerns and to assign the nontechnical concerns to
, experts in the field of personnel relations. Contrary to GAP's claims, this separation of concerns did not leave the inspectcrs "no chcice but filing their massive complaints" (Petition, p. 17). The labeling cf the concerns as technical or ncntechnical occurred af ter the concerns we e filed and was part of the process of evaluating the concerns.
In addition, GAP's assertion (p. 17) that "because the welders (SIC) documented their specific concerns it is only their work that the public can be assured" is adequate, lacks any basis. We point out again that nc unsafe conditions were found as a result of the welding inspector cencerns. Since no deficiencies were four.d that would have caused unsafe operations, DPC concluded that there had been no QA program breakdown.10 In additicn, interviews ~ with inspectors cther thar welding inspectors revealed no concerns about the safety or quality of the plant. Furthermore, all of t,he progra enhancements which 9
9 GAP fails to identify the "outside consultants" it is referring to.
20 DPC notes that GAP has attempted to 'use the SIE as evicence of a systematic breakdown in QA. However, as noted in this response, neither the SIE itself nor the testimony of the SIE panel members support this allegation.
T a'ilow inspectors to voice CCGcerns are in place t*rc;;*; I the Various ir.specticn grcups.
GAD further claims that the NRC Atomic Safety anc Licensing Board has inappropriately " narrowed the secpe of the claims to allow intervenors to litigate only the welding concerns" anc ignored the f act that the allegee QA breakdown is " site wide" and " programmatic." This claim is similarly unsupported by any evidence other than anonymous letters cited by GAP and a vague reference to statements made by "other QC inspectors."
GAP cencluces Section E by referring to the Self-Initiated Evaluation (called INFO by GAP), pages 3-97 and Section D.3, which, it asserts, address the " lack of independence." We fo d no such statement in our review of the SIE. In fact, our conclusion based on our own review of the SIE findings, was that the SIE demonstrated seven strengths, two of which are noted in the SIE summary as:
" Quality Assurance and Quality Control functions were performed adequately and independently to support and control the quality of the' facility." (SIE at p. 2a.)
and, "A strong corporate commitment to design and construct a safe facility." Id.
We submit that the SIE, when viewed in its proper perspective, demonstrates DPC's commitment to quality work and self-improveient.
C. " Failure to Maintain Adecuate Controls to Process and Respond to Non-Conformino Conditions" GAP maintains that "the breakdown of the NCI reporting system at Catawba illustrates the breakdown of the entire QA program" (Petition, p. 18). In 1
- &rticWar, G; a t ta c i.s CPC fer hanc'ing ce"
- f en:ies t
) rea.s cthe- than a-
~
{ hCI. Eefere respencing te each cf GAC's specific i c enc e rr.s . a general ,
l ces:ription of DPC's metheds of icentifyir.; !
and correcting inspection 1
' discrepancies is set forth below:
Tne Quality Assurance Prograr in use at Duke Pe er Ccmpa during ccnstruction at Catawba provioes several means of correcting discrepancies that are discovered by inspectors. There are four basic methods available, three of
-which do not involve writing an NCI.
The first, which is sometimes referred to in seme of the pro:ecures as the "hoid point" method, consists of an inspector making the craf t aware of a deficiency, the deficiency being corrected to the satisfaction of the inspector, and the inspe: tor signing of f the item. In this method, the item is not signed off until all necessary action has been completed, and the inspector is satisfied. This "hoid point" method is common, and has
, been in use at Catawba throughout construction.
e The seconc is the " process control" method, whereby tne inspection report itself provides the means to do:ument a repair. This method is used primarily in welding where, for example, a final visual inspecticn might detect defects which would be recorded on the inspection form. The e
procedure for the inspection and for making the weld would provide instructions on how to correct that item (or that defect) and then provide .
instructions for reinspection. All of .this would be documented on the Process Control Form, which serves both as a documentation of the work and the inspection of that work.
80
(
- The thir Pethcc is a Ceficie".Cy Eecert Ferr The*e have been seve*al ciffere91 procecres availacle to inspectcrs u9eer this methoc. The f
l procedure currently i r, use is the Discrepancy Report Form, commenly referred to as an R-2A. By this method, the inspector would document the problem he identified, and that would then be sent to the Construction Technical grcup at the site. That group would cetermine necessary cor-rective action. If such action involved the craft redoing work, it would go to the craf t to be done.
The form would then be routed back to the inspector who would reinspect the work and, if satisfied, sign off on it.
Inspectors cay use QA Procedure Q-1, "Centrcl of Ncncenforming Items " and its corresponding Fcrm Q-1A, "Nonconforcing Itet Report," commenly referred to as an NCI. This method is used when the discrepancy is not handlec by one of the methods discussed above.
Our response tc specific allegations by GAP follo s:
- 1. DPC Policy tc Circumvent Nonconforming Items (NCI) (Petition, pp. 18-20)
In support of this allegation, GAP first contends that the Planning and Facilities Engineering staff, rather than QA/QC, began the NCI process.
However, GAP conveniently overlooked paragr,aph 4.1 of Revision 5, which
, describes- the responsibilities of inspectors and engineers in initiatino NCIs, and tagging the nonconformed item.21 The sole 21 Paragraph 4.1 of revision 5 states:
(continued) l .
}
ti;cnsiti'it c' the F '. a - n i r. ; a-d Facilities E ;' es ' .g s ta was one of statasing N is.
I t'
l More impcrtantly, GAP further contenes that QA as merely a " glorified file clerk for the engineering staff" when it CETe to the resolution of NCIs. Petitioner references paragraph 4.4 o f :.e . . 5 to CAP Q ', again.
overlooking paragraph 4.3 which clearly requires GA Department approval of resolutions to NCIs: "It shall also be approved and inspection requirements assigned by the Quality Assurance Department." This approval '
is performed by an engineer or an individual with a high oegree of technical expertise designated by the Project QA Ma.a;er.
In addition, it can clearly be seen on Fore Q-1A, Rev. E, block 8 that QA Approval is required. (GAP Attachment 2.)
GAP's contention that the status report described in paragraph 4.5 (QAP Q-1, Rev. 5) is a " trending list" is inaccurate. GAP apparently is trying to argue that the NCI status log was a substitute for an NCI trend 12 (continued)
Any inspector or engineer discovering a nonconforming item as defin'ed in paragraph Q-2 shall initiate Form Q-1A, Nonconformi.g Item Report.
The report shall then be reviewed for completeness and correctness by the responsible Senior Engineer or his designated representative. He shall sign the report proces' sing.Ncnconforming god submit it toitems the Planning and Facilities shall be controlled Engineer by tagging, Staff for marking, or other means of identification where physical segregation is not practical, although physical segregation and marking are preferred. The means of identification must be stated on Form Q-1A. The identification of a nonconforming item must be maintained until a resolution to the nonconformance has been approved on Form Q-1A. If tags are used to identify the nonconformance, Form Q-1B shall be used.
l g
a.a ys's.
- '.s na.E ceen trentte by QA for conciticns acve se te cuality (not scheOu'.E) since 1974 GAP further states that "if QA/QC had ever been given authority over the construction NCI files, there tright be a possibility that documentation could be ret-ieved." This statement is rnisleading, in that although Construction dees maintain the status log and initial copy of the NCI, they are held accountable by DPC's QA Program for the completeness of the log and for accounting for each NCI. '
This activity is subject to QA audit. Final copies of all N:!'s are kept in the QA vault.
- 2. In Process Inspections (Petition, pp. 20-23)
GAF contends that Revision 9 to QA Procecure Q-1, paragraph 4.1 (GAP Attachment 14), violated Criterion X of Appendix B when it was in use during 1976-1978, because the procedure directs inspections to be suspended on nonconforming activities. This is incorrect. Paragraph 4.1 requires the activity (i.e. , the work) to be stopped, not the inspection of the work. The parallel which GAP attempts to draw between'QA procedures at P.idland found to be improper and this Catawba procedure, is, therefore, entirely inappropriate.
GAP further contends (p. 21) that QA/QC inspectors had no authority to write NCIs without first getting approval from the Senior Engineer. This is not the case. Inspectors could initiate NCIs both before and after Revision 12. The reference GAP cites to support its argument (Sec. 5.1.13 of QAP Q-1, Rev. 12) does not remove the authority of inspectors to write
N:: 5. F i tr.e , i; g;ses acthc"ity ic the SeriC' Engireer te determine the va'icity cf a ; cpesed N:1.
GAP also asserts on page 21 that DPC "put the screws on" Catawba management to eliminate "the NCI problem." This statement constitutes a grcss mischaracte*izaticn cf what intentions aitt respect tc NCIs were at this time.
The following is a statement of our policy on this issue prior to 1981:
Wner an ins;ector determi ne:: that a deficiency cotid nct or should not be handled by a method other tnan the nonconforming method, he would obtain a nonconforming item forrr (form Q-1A) anc complete the top portion of the form.
If there was a question in the inspector's mind as tc whether an item was in fact nonconforming, he might talk with his supervision to make a determination.
If this dctermination was that the item was not, in fact, ncnconforming, or
. that another method would be appropriate to handle the item, then the form (Q-1L) would not be completed or would be discarced. If the form was com;1eted, the inspector would describe the item and its ccndition along with other information, such as location, on the top part of the form. The inspecter would then sign the form as originator. A revie- of the form by the inspector's first line supervisor was sometimes,concactec at this point, but 9
was not required.
Next, the NCI procedure required revied by the Senior Engineer. There was no requirement to have a serial number assigned to the NCI form at this point.
However, usually a number had been obtained by the originator at this point.
Ar ins;ector r';-t nct be sure the item was nonconf ere ng arc would therefore d
ir.tentionally nct have a serial number assigned prior tc tr.is review. This was not the ncrmal case, and the vast majority of NCIs subrittee to the Senior Engineer for review already had a serial number assigned to them. The assignment of a serial nutber was purely a clerical function and in no way involved a review of the NCI for validity. A serial number was assigned by the facilities group in construction simply because they maintained the NCI log book. -
The typical situation was for the inspector to give the Senior Engineer a completed NCI f orm that was signed and had a serial nutter. He would review the NCI for accuracy, ccmpleteness, and validity, and might request that additional infctmation be added to the form by the originator. This review was
- to determine:
- whether the item wa's clearly identified;
- whether the problem had been clearly described;
- what requiremerit has been violated and whether it had been identified;
- whether all the available information ha'd been given such that the party assigned resolution would have all they would need to understand the deficiency; ,
- whether the form was legible; and ,
- whether there was another, more appropriate way to handle the item If the NCI' form needed work in clarity or legibility, or more information was needed, the Senior Engineer would explain what was needed to the originator and 36 c(55
c' e: hir : ccia'n the 4 r.f: - .a t i c r. , o- cla-ify it. E c ine r. resubmit the .
f : r.- If the Serict Engineer nad ouestic .s ateut its valicity er thought it wss insalid, he woule discuss it witn the originaty. ' tis discussien mignt aise include the inspector's supervisor. If the Senior Engineer determined that the NCI was net valid, he would exclain this to the originater and handle it in ene cf t-c .sys, depending on whether the for: hac a serial nce er on it.
If it had a serial number, he would either explain on the form why it was invalid or g ahead and approve it, and ask the QA group to assig'n it to him for re s ol uti or. , in which case he would resolve the NCI by stating why it was invalid. In bcth cases the form would be forwarded to QA. If it did not have a serial nc-ter, the Senior Engineer would retur. it to the originator explaining wh> it was not a valid NCI. If the inspectcr expressed disagreement about the validity of the NCI, the Senior Engineer wculd usually sign it. In some cases he would direct that the discrepancy be hancied by another method, such as a Corrective Action Nctice (R-2A) or by informing the craft to correct it.
At least 17,000 ~ NCI's have been originated at Catawba. To the best of the recollection cf the then Senior Engineer, only a fe per year, perhaps as many as 20, woulc be invalidated during this kind of review. Most of these situations arcse because the inspector had a cuestion as to whether the discrepancy should be an NCI. In the vast majority of cases, the Senior 9
Engineer conciaded that these discrepancies would prctably be best handled as an NCI and would sign the NCI form.
The NCI Procedure in effect at this time stated that if an NCI was determined to be invalid, the reason should be stated in the description block. We
~
$$f {f
.5 : s*.e: ;s: EE- ; .t . ;s : e.isi:- ar;;ie: it NC:s that had been -
- e
- E ,e s e *# a'i
- e 0. 50 s'* e l'e NC3 pr0Ce:J'e mas used f0r nonsa#ety re"Etet ite .s. rany t res ar. NC2 was cetermined te :s in.alid te:suse it was n:: en a safety relate: system, and the same QA recuireme ts did net a; ply.
Some of the deficiencies identified by inspectors would not be valid r: :: ':' r.a r. E s c no s a f e . rel a:e: systems.
After review by the Senior Engineer the NCI was sent to the QA group for assignment of resolution responsibility. In June 1978, a block was added to the NCI form to include a QA review of the origination also. This change was ir,;cnsequential because the form was always routed tc QA after Senior Engineer review.
The QA group would determine who would be assigned to resolve the NCI and route it to them through the facilities group so the log could reflect the assignment. The resolution could be assigned to either the Design Engineering, Construction, or QA Departments, depending on what receirement was violated and whether or not Engineering evaluation was required.
The resolution or disposition would be determined and added to the forrtand approved by a competent individual for technical content. This approval was not required for resciutions develcped by QA because it would get the review autor.atically. .The QA group would then review and approve the resolution and indicate any action to be taken.
In summary, there was no attempt to prevent NCIs from being written. There was an effort to make sure the NCIs were clearly and completely written and that
t
.. s me*e P: :e'*; .le * *e* l'.e's ae's Cthe' 0"Og a- Elic w&ys tC Correct the -
~ ~ !:"e:a 0) Ee:a.se C # l'# 1 er;*as i s 5 0*.e i t!;e tt *s "! have felt they aere f #*g .*0.D Cur ened #*
- ;erformi,g 19e'r jt ; hCae.e *, * *e intert Of CPC Was
} :: e.s.re the OA ; reg *ae .as being preperly im;1emer.tet.
j :: s;ates e. ; age 22 tr.at the f # rst e cui re e r.: fe a *; CFR EO A;;e[diy B l Criterior. XVI evaluation was with Rev. 17 to QA Prccecure Q-1. Actually, in Revision 16 we first began using the term " Criterion XVI evaluation.
Throughout the construction of Catwaba, Duke had in place procedures designed to meet the requirements of Appendix 2, Criterion XVI. It became obvious in Februa y 1951 as a resuit of NRC ILE findings preser.ted in ILE Repert 411/El -
C2 tr.it a mere fermal structure to our NCI procedures and forms was necessary.
As a result of NRC violations, DPC conducted a thorouch audit of approximately 11,000 NCIs in 1951 with tne following results:
Techr.ically, the safety of components and systems was being assured.
. Editorially, our procedures and forms did net present an adequate oescription ne:essary for audits and long-term use as a QA record. -
~.
Therefore, procedure Q-1 was changed in April 19El anc January 1982 to address these inadequacies. The action taken was reviewed by the NRC and closed in IE Reports50-41f, 414/81.-1E (8/28/81), 50-413, 414/81-27 (12/10/E1), and 50-413, 414/82-03 (2/19/82).
I C
- . : <(- : :.:3 ;..e5 d: ' : :: ':-- . -: :te- Fer:-ts
(:st t :r. ;;. 23-25) .
G'D cc .;e .cs that R-2As are use: ty DPC as ir.ferier sces-itetes fer NCIs.
r.is is ir. accurate. R-2As are not used as substit.tes , it ary cegree, for NCIs. Rather, R-2As are simply a mechanism for the inspector to identify discre;a c'es se that they a'.' te evalcated a..: a:: ::-ia e a:1 r. t'aken. 4 In that R-2As de identify problems, a part of the evaluation is to assure that all R-2As are evaluated for possible elevation to NCI status.
CAP goes on to assert that R-2As are "de.ficient from NCIs in at least seven respects." (Petition, p. 23.) Our p; int-by ;oint response to this assertion follows.
- a. ASSEFTION: NCIs identify the cause of the problem
RESPONSE
10 CFR 50, Appendix B, Criterion XVI requires that for "Significant" conditions adverse to quality the cause be determined. The individual problems identified on R-2As that are "Significant" Will be elevated to an NCI. The repetitive problems which are, not ,
significant in themselves but which demonstrate an unfavorable trend if they recur will be identified by R-2A trend analysis. The cause will,pe determined and appropriate action will be taken based on the trend analysis. Accordingly, this assertion is misleading.
- b. ASSERTION: NCIs cannot be closed with an informal, . undocumented .
design change.
I .
- H ::'d E.-
- -245 :Ennet ts : lese: .itt a . *
- f:- .al , :. :::. s .te g ge s ig . e t ar.ge ,
i eitne". If t.e ite is r.ct ir. a:::rcance .iU ces 'gn re:;uiremer.ts ,
.it m.:st be 'ei ther re.crked a ic sig :ec off er s'avated to an NC1 fcr !
L Design evaluatibn and action. If Design evaluation determines the as-Ouilt cer.citic , is .t:ce:ta:Ie, cra. ' .;5 1 .: :Elculaticns are revised to reflect the as-built condition.
- c. ASSERTION: NCIs give inspectors the ability to stop work on a nonconforming item that needs to be isolated.
RES ONSE:
Although R-2As d6 not always step the work at the time they are written, the specific discrepancy is identified and controlled and that item will not be signed off as complete until all necessa'ry rework, justifications, and reinspections are completed, signed, and documented. Accordingly, the work is stopped at the appropriate time if all these steps are not complete. However, if an inspector thinks work should be stepped immediately, he can ir,ititte a Work Stoppage notice. .
- c. ASSERTION: NCIs are sent to the N~I (5,1C) for review.
. 9
RESPONSE
We presume that GAP intended to say "NRC" for "NCI". Based on this assumption our answer is as follows: Although each R-2A is not sent [
to the NRC, all R-2As are available for review by the NRC.
. ~
- t. ! !!!: ~ : : *- N;:s a e trer.cet 'n 1 -
?
EE5;CNSE:
R-2As are trencec by Construction. Cor.s: .: tion is in the best position to trend R-2As since R-2As are crimarily " rework" type ciscre;ar.cies As suc5, tne Cce.strectic "s:s-;re ; is cictest to the work and, therefore, can determine the root cause and take corrective action more directly. Moreover, R-2A trend reports are required by QA procedure to be sent to QA and are reviewed by QA.
- f. ASSE: TION: NCIs have contrel numbers (once iss.ed).
RESPONSE
R-2As aise have control neebers. See QA Frecedure R-2, Rev. 9, paragraph 4.1 (Duke Attachment 2). -
- g. ASSERTION: NCIs require written resolution.
RESPONSE
R-2As also require written resolution. See QA Procedure R-2, Rev. 9, paragraph 4.2 (Duke Attachment 3). .
In a.ddition tc*.he foregoing seven points, GAP invokes the SIE (referred to by GAP as the INPO report) as further evidence that R-2As are " inferior substitutes" for NCIs. (Petition, p. . 24.) Our responses to the SIE observations cited by Petitioner are set forth below: -
The (SIE) reports the following about the R-2As:
l psySF i
- e ;e 'c-rs'.:e C C: ~ s t r.;;t '. C t C C e re :t i'. e E t t : *.s k5s 's . ' e ve C .
" e s ::r.s ':
- i i ty fer t'e-d ar.alysis cf E-24s ( i ! : E : t '. :.- :is: e:ancy r e : : - 1 ! ,' *e C e *.1 1 y
- E s C P.a *.;e : fro- (A tt *0*st'. liO*. I - *. s r e h i e ve i :icate: the foiic d.; areas of .eakness:
- a. Nc trend a .E' sis .as been performed d';-ing 1.e perio: 6-1-E2 threugn E-22-E2. (Fetition, p. 24)
- Responsibility fer trending R-2As (ccrrective actien) was transferred to the C r.5 *. t'en OEpaet .e-; ' :- 0 a'ity Ass ura .:e just LE':-e tre 5:E. After the transfer, the method of perfctming the trend analysis took constro: tion a few me r.t hs to develop. It s h:uld be ncted that , subsequent trend analyses did include R-2As written during this period. Since the SIE, the Ad .inistrative Methods Section has been established at the Catawba site. One of this section's responsibilities is to trend R-2As. The trending is perfcrmed m:nthly in acccrdance witn Ccnstruction Deptrtment Procedure CDA-9. Results of this review are forwarded to appropriate management personnel for their a*.alysis and pessible action.
In addressing trending analyses, however, it should be understood that this is simply a management tool anc does not replace the normal and required QA inspections of each activity being trended. In shcrt, even if trending analysis was not conducted, the required QA inspections of each appropri1ste activity would still take place to provide assurance of quality construction,
- b. Construction has net performed any, trend analysis cf QA surg illance reports. (Petition, p. 24)-
Any deficiencies founb during QA surveillances are required to be reported in .
accordance with QAP R-2 or Q-1. R-2As and Q-1As are trended fer conditions adverse to quality. Therefore, although the surveillance reports are not ,
trended, significant deficiencies identified on surveillance reports are trended.
- ~
9y 4 59
( 1
- . 3 s .:tien tas r:t cerferrec a.,. t enc ana'iys t s c' .
nen::-f:--in; ite rec:-ts . (:stitlen, c. 24) i y :e*ferms t*ent analysis of N"!! a .C fcTwarCs tre- tc Ccnstructior for
( c e r
- e c '. '. '.'s a C l i e n .
l
- d. Statener.t of acticn cr. R-2A #5677 coes net actress all areas ef, c o r.c e "n. Ciping syste" as ;,res suri2ec c ric' to release cy.
hyc c grcup. E-2A cic nc; accress prccecure violation nor (perscnnel) safety implications. (Petition, p.'24)
R-2A 5677 documents pressurization of pipe prior to release for testing to the systems group in construction. It should be noted that the type of activity is rc tinely perfctmed to check for leaks in joints prior to turnover. However, it cces net replace the hydrcstatic tests of recuired safety systems.
Ac:crcingly, such activity does nct ccnstitute a viciation of procedures. In any event, revie. shows the system pressurized to be a non-nuclear safety related system. Testing of this ncnsafety related piping is the responsibility of the construction department. The noneritical. QA review of this R-2A is attributable in part to the nonsafety related nature of the piping involved.
- e. Acticn recuired on R-2A FM5350, although cleared by QA, has not been completed. (Petitien, p. 24)
R-2A M5350 described the observatien by QC during a hydrostatic test that twc instrument taps , while correctly installed per the design isometric, were not located as shc.n the design ficw diagram. This was a case of differing design documents. The taps were installed per the'. installation drawing. The 4
resciution was to corre'ct the flow- diagram to show the taps located as called .
for by the design drawing. The QC inspector who initialed the reinspection i block was aware that there was action. underway to revise the flow diagram and
[ .
( correct the discrepancy between the two drawings. In this case, the R-2A should not have been approved by QA prior to receiving the revised flow t------.-
L. .4 5; a-- b e ta . e 'c l a - t 'i e d g re:e :.; e :.- 2 c . ;r i s ;: - .t. 5:.:e tr.e ,ggs .ere. .
[.
l
" !;ai'Ec a*d ins;etted as desig*>e , t'e e ii . Saft ;. # ;act frt.- t *. E la:A cf -
i
- j. :r..4:a Qt. revie cr. this R-2A.
t
? n..
L I- f ., Deficiencies were noted in Forms R-2A (Inspecticn Discrepancies)
- lg '. 4 ,f/ where the Quality Assurance grcuo
- j. A i ;/p l A. Ac:E;tec the statement of action rectired when the action 6
1 did nct accress the root cause cf the problem or u ,
~-
Es Approved clearance of the R-2A when ccrrective action had 7 ', *
.not yet been taken. (Fetition, p. 24) 77h(5 statement .from the SIE report is based on items (d) and (e) abeve relating
'to R-23s #5677 and #M5350. The specifics of eac'h case are discussed above in c.
car respp.6ses te items d and e. The la:k of criticei etaitatier. by QA to these e
discrepancy reports was- not significant.
4 Voiding the Nen:cnforming Item Repcrt (Petition, pp. 2-5) .
GAP cites the.. testimony of C R Baldwin to support its claim that " a particularly oEinous method of cutting down NCIs at Catawba has been
" voiding" or " verbally overriding" the NCI by management, and, in
- particular, that there were " numerous occasior.s" when an inspector did
,/j
'(%
not write an NCI af ter talking te his supervisor. In summary, ther,e are three conditions under which " voiding" of NCIs occurred. These are~ as ip - ,
, , follows:
4
. 9
- e. There h, s another preferred procedural method available fo.r getting .
the deficiency corrected. These methods were built into the DPC QA l , Program to keep from flooding the' NCI process with insignificant ,
0' items. In this case, the inspector would be advised on the correct l way to proceed.
(. a
- 42 -
- - hh--N
g IhE c E f i C i E *- *. mas r E a '- y 'O.E!!ic!'
a Cf a::E :* at ility.
Ihe
- S uO e r '. # s 0
- WC '. I C Cistuss l'E situatiCr with t'E #~10ectCr ar.d attEr.;t -
tc answer the C;ues tion. Dr.ce the cuestion ha: :EE. resched, the NCI
, was no. longer necessary.
- . The p*ccosec NCI was n:1 va'id. ine tuservis :ul: try te exp'Ein his decision to the inspecter. If the NCI ha a serial number, it woulc be processed in accorcante with'QA procedures. If the proposed NCI did not have a serial number it was returned to the originator.
Only a few of these cases resulted in a written NCI being " voided."
In no case Was the NCI " voided" as a cover up or in response to r
Construction pressuresT In croer to provide more complete documentation, DPC has adopted two revisions to QA Procedure Q-1 to improve the process. They are:
Aiways recuiring a preposed NCI to have a serial number assigned befcre submitting it for approval. '
i When a proposed NCI is determined to be nonvalid, an explanation is written on the NCI and a copy is given to the originator. The record copy
, is kept in*.he QA vault. ,
D. Failure to Maintain Adecuate Material Traceability te Identify and Docu-ment the History of All Ma te ri a.l s . Parts. Components, and Special Processes GAP cites several SIE findings in the area of construction as evidence of a
" massive breakdown in material traceability." Significantly, three SIE team
~
//7Cf.6S
e ;e s . r.: P.a c tie - ' .C.E: #-
.4 :: s t :t '. : - a ets (: e es:n from TVA, .
lae, a.c IND:) s tatec :e't-e the Catacca Licersi g 5:t c t'.at construction .
- # r.c # rig s set T:"th i t. 1 se SIE Cid not "ef'ect a crea A
- i.*. the QE program
(~r 10153-5) or "a ny pract ce W eh dic or wouic
- ave led to unsafe .
cor.struction operation of the plant". (Tr. 20064-9.) CFC's response to SIE ciacs is set 'crin oeit.: .
- 1. CC.3-1: Site receipt inspection does not ensure that material
! and ecuipment received on site are evaluated against the recuirements of the procurement specifications. Examples of the problems icentified may potentially result in delays, waste of
- f. materials, additional time spent on disposition of deviations j from procured materials and work stoppage. (Petition, p. 26.) {
l l This finding related to cadweld sleeves and powder which had been requisitioned anc received at another Duke site (Cheroise) as safety-related material. Wnen transferred froc Cherokee.- o Catawba the material was downgraded to nonsafety I related status. Accordingly, there were no errors in the Catawba requisition or receipt inspection process.
A Catawba steelworker foreman mistaken.ly l
r allocated this material for a safety-related cadweld. The QA inspector, upon l i
j conducting the required inspection, would have determined that the material was j
not on safety-related material listing, and, accordingly, would not have let l
! him use the material. (The SIE team observation occurrec prior to the inspection poi nt. ) Therefore, we oo not consider this finding to be l_ significant. '.
l I
- 2. A consistent method for material identification was not CC.Q2:
in effect in,the warehouse. Several instances were noted where I . D. tags had fallen off, equipment was marked with ink;. and when material was being sectionalized to start fabrication, a means for maintaining the identification was not being done. '
(Petition, p. 26.) .
Safety-related material is marked or tagged with proper identificat. ion. During the SIE audit, stick-on paper tags for electrical equipment identification were x .
k . . . . . . .. ..
f l ': e : -iss'.; an: 'a',.; c c' t.e ec.': s .
In that sa:' taes were usec -
I ':r ease c' ' a e".CJs e ice .:i#icati:r (nt: - - 6. i , g s d er ;; ; r?cses), we Co not -
i cc .s'cer this te ce si; i:En . Duke Fc-e 's inspe: tier prcg a: prevents the d r.s tal ' a t i c . of imprc:erly identified sa'ety-related ite s by recuiring inscetters to verify before installation that material er equip. rent is properly 4:Er:'f#ec anc is the c:- e:: raterial fer ine jeb in cuestien Hc.ie v e r ,- in order to prevent recurrence of this finding, a uniform warehouse L marking system was developed and implemented in April,1983. This system is l
l audited periodically and will make the warehoosing function mere efficient, l 3. CC.3-3: Proper protective measures were not taking place for L i environmentally-sensitive equipment that was " robbed" for spare '
parts. Some partl were being stored in an open door instrument cabinet. (Petition, p. 26.)
The " environmentally sensitive equipment" referred to in the finoing was a nonsafety-related circuit breaker being temporarily stored at the ' Catawba Site
- by the Transmission Department. This breaker would not have been installed at i
i .
Catawba and, therefore, could not have any affect on plant operation. It should also be moted that Duke Power Company's electrical checkout .and y
functional tests would de ect any equipment with missing or cefective parts.
4 CC.3-4: Prote:ere QFP-8.002CNS, Rev. IA does not indicate the disposition of unused filler ma te ri a l,. Confusion appears to exist regarding handling of unusec filler material and adherence to Ak.? code requirements could not be determined.
This finding applies to the control of welding filler material used by' the 'HVAC contractor on-site. ]n this case, the unused filler material was being returned to Duke for rebaking at tne end of each shift. The supervisor questioned in the observation was not responsible for this activity and was not
.r a ;;sitic- it E 5 e! the 0.e sti * "ak.e nas pt*#:
- e: seVe a' &acits and J -
s .". e ' ' ' a .:e s C f E a T.".s t * ' s m e I C T ,0 CO*.1P01 p*Cgrar a-d #0 c ".C the Saterial was t e ' .; t.t.:.e: E0eOUatel . e '. e n t hC:. gP the pr0 eC.*e CiC 001 give specific l
- instrc: lions. Since tr.at time the prece:cre has bee re.ised te describe the l
i practice in use.
- 5. CC.3-5: Materials are not being maintained or stored cffectively at work site locations. Several examples were noted which reflected improper control. (Petitier., p. 26.)
This finding is based upon several specific inciderts. Each of these incidents .
is discussed below.
- a. The SIE tear. found some pitted carben steel piping in the pipe f ab shop. Tt should be noted that the pipe minimum required wail thickness remained and thus, this did not present a safety concern. In any event, as part of the weld' preparation procedure, any unacceptable ' pitting in piping is removed or
, repaired. Prior to installation of the pipe QA personnel inspect it to make sure that unacceptatle pitting is removed or repaired. Therefore, due to the acceptability of the piping *in question, and the procedures in place to prevent installation of ,
unacceptable pipe, this finding is insignificant.
s -
. b. The SIE team found rust streaks on stainless steel piping that had been placed on carbon steel rollers. When left in contact ,
with carbon steel, stainless s, teel may show some surface rust.
However, rust streaks on the surface of stainless steel is not detrimental to the pipe. In any event, the required QA
- 4E -
P c'Ea?' int!! # ~.s E 1 0T recuitec
'*4
- 1 65E c# ;5;inE in a
- sa#E*y rela 1Ec 5;. 51 E r =0ulC ha.e U" cvE*ec a9y unacCeOtable
- C lea r.li ne s s pic*.Ers. }n sncri, inis #indir.g was not sig*ifican*.
t
- c. The third incident on which this finding is based involved the discovery that, in the storage yard, sema ends of cable stored on cable reels were not taped as required. Such taping is, performed to protect the cable ends f rc.T. environmental damage prict to their installation. It shculd be nctec that prior to installation, the ends of cable relis are routinely discarded in theinstabationprocess. Thus, any damaged ends would not have an adverse impact. In any event, QA inspectcrs examine all safety-related cable end terminations. Any defective or suspect cabies are identified so that they will not be installed in the plant. These measures ensure the installation of only those cables whose ends ar'e acceptable. Further, extensive electrical checkouts and testing of safety-related systems prior' to operation would detect any problems caused by damaged ends.
Thus, plant operational safety is not compromised.
-d . 91nally,, the SIE team found uncapped pipe in the turbine building. The pipe involved was not safety-rela ~ted and, therefore, was not within 'the scope of our QA Program. Because of this, there is no safety concern involved. ,
i 1
6/ c/ 95
e ._
- t. ~ ; . 5 -1 :
Senec,.dec ; e.e twe naintema ce activities en insta'.lec
- e-ice cf ec ic ert a e net al.ays assurec t ce;nc u t th ".
Cctstructic. Department e entire cer. trol Eouipment was icentifiec fcr w.icn preventive maintenance t c 21 cc nt ~.s age, and .
ac be eceirements hac ceen estaDished.there tas no evide .ca t'.aten cancele ine (Fetiticn, p. 27.)compensatcry SIE team ncted that some plant ecuipment
- eventive mair.te nanc e was not receiving regular thrcu;*. cut the tine betweer. syster completien anc te nceer to cpe ating persennel.
significant It was determined that this problem was not because all si.fety related equipment undergoes redundant inspect-ions and tests by Construction and/or Nucl systems will function as designed. ear Production to ensure all Any problems noted during these tests and inspections are documented and corrected as ry.necessa In response tc this finding, the Ccnstruction Department updated its pre in March, 1983. ventive maintenance program This Tevision ensures that all equipment preventive maintenance receives ongoing during the period between turnover. installation and final E.
Failure to Maintain an Adecuate Oualitv Assurance am for Venders Procr In support of this allegation, GAP raises findings in the Cc Quality Procecure (QP) areas from the nstruction (CC) and SIE regarding certain problems in the HVAC contractor's program. (Petition, pp. 27-8.).
GAP asserts that these examples (Fetition, " illustrate that
- p. 27). serious weaknesses exist in the ve Significantly, members, in each of these areas, cognizant SIE team
- resppsible for the findings raised have stated to the Catawba 1.icensing Board that such findings do not reflect a systemati QA program (Tr. 10153-5) c breakdown in the -
or "any practice ,which did or would have led to unsafe construction operation of the plant". '(T r.
10064-9.) Duke Power Company's response to these SIE observations follows:
. - as - .
r
. 'e i .' E s . t . t s *a..'t
. fc'%. ; -ta. tsses -t t ccst .ec ir ccrtrci cf '.
h.-c cc .; acic 's -e'.cin; ;ro; a-- (~nt 5:E is a;;acnec as Attachme..; E) ",
bc -s'cer k .ew the weic prececere -
cer whicn he was w c t r.i n g. (p. 36)
- Ali eicers knew requirec welt size a.c iccation, but did net know how they acquired that ir.f c ema t i on. (p. 36)
No process control was available te s;ecify the welding erecedure fer p i e r..;- erec - ~~
Dra ine CN *.E E!- G-000h , Re v. C. (p. 36)
Weiter/ Supe rvisor [i.e., foreman) picts -sicing procedure from all available welding precedures. Supervisor documents welding precedure(s) used cr. a support af ter the support is complete. (p. 36)
There is no traceability of weld procedures to the finished weld. (p. 69) .
The HVAC contracter's welding program includes a list cf qualified welding precedures in ead. work ;ackage. The welding fere a . selects one of these procedures, cepending on ,the welding process to be used, the material being welded, and the type of Nid being made. The foreman instructs the welder in the welding parameters and criteria to use in makin; the weld; thus the welder himself does not need to memorize the procedure number. The drasings, list of qualified welding procedures, and instructions by the foreman constitute the process control under the HVAC contractor's program. In addition, there is no receirement fer traceability of the weid procedure used on type welds in question. In short, the SIE did not pcint out any instances where the welding technique was inappropriate and, accordingly, this finding, has no safety significance.
- .e e
- 2. The SIE states that a welder was making welds without removing galvanizing material. (p. 36)
This was an isolated instance, not a common practice. -
or this application, removal of galvanizing material was censidered to be cf no detriment to the
-4c-
e aE'c. :-EsE . S PcJ c ga's a >iW.; *:t be acetualEly rE.cved
- in Cther I
a:; ice *.':ns, scacsity sc.'.c tG 4Ery Ev' cent at finEl VisLEI ir5?ectiC0 Cau!ing
- l l
inE welc to be rEjEctic ar.c cCrTEctEC. Ihf rEfore, this finci".; had no safety
{
sicnificance.
- 3. The SIE states t *,a t the HVAC support 2-H-VC-4395 had uncercut in excess of that allowed by AWS DI.1 code." (p. 36.) Further, procedures did not meet Code requireme~nts.
(p. 69.)
In that the vendor's welding procedures did not strictly adhere to the undercut recuirements cf AWS D1.2 as referenced in Design Engineering's specification, these findings were valic In short, the undercut found did violate AWS DI.1, but not the vendor's procediTres. DPC conducted an analysis of the undercut and determined that the vendor's procedures were clearly adequate and would not adversely impart safety. As a result of this investigation, the Design specification was revised to remove overly restrictive undercut requirements.
The basis for this action was that restrictive undercut requirement in AWS D1.1 are based on fatigue concerns. The design of the HVAC duct support systems is such that fatigue is not a concern and stresses for all other loadfng conditions other than seismic are relatively low. T'herefore, HVAC systems-were erected in accordance with design basis; there was never any cuestion of the operaticn and function of HVAC systems. Accord,ingly, this is not judged to
. have,any safety significance.
To assure that no other similar misiriterpretations of Design Engineering specifications have occurred without Design Engineering approval,- a complete audit of Bahnson has been completed by QA Vendors.
.': : 'asit - .
F:- t .e actse stEtec reas; :s DCC sa:-its tnat the GAO petitien snculd be ci .iec. DOC wouic ncte that in accition it the matters accressed abcve, GAP alsc recuests the monitering of Dffice of Inspector and Auditer (01A).12 ggp 3 recess; is predsec u;;c - an alie;stien t r.a t CU nas breachec the confidentiality to be extended to individuals who come to them with information. DPC disagrees and of fers the following response regarding the lack of confidentiality allegation against the NRC.
EE F E:.E NC E : GAP, page 30:
Mr. Davisen's testimony (Duke Attachment 4) and deposition, (Duke Attachment 5) state the facts rel ative 'to conversations with NRC Resident Maxwell. Mr.
Maxwell conveyed to Mr. Davison that welding inspectors had expressed their concerns to him and that they also were writing down their concerns,in "a black book." At no time were the names of the inspector's given to Mr. Davison as alleged on page 30 of the petition to the NRC by Billie P. Garde. It is our understancing that Mr. Maxwell was advising Duke of non-nuclear safety related problems and Duke did net consider any discussions to be a breach of confidentiality. .
REFERENCE:
GAP, page 43:
s Review of the dxt to .the last paragraph at the bottom of this page does not reveal any breach of confidentiality by the NRC to Duke Power Company: No one was identified as having called the task force a " whitewash."
12 GAP's request for an DI investigetion has already been acted upon.
4 rr . M
r
- ErlEE':E:
G A . A' tis c h s ; 2 6 , r a gs E :
.c the af fica.it scbrittec ty Eiiiie P. Garde (G'F) an incident is described in wr.i c h tac ur.namec welding inspecters hac allegediy " tested" the resident NRC Inspecter tc see if he couic be trustec to nct tell Duke specific information.
Acccrding to the affidavit, they had reported a specific technical problem to its NFC and ;%at the probler welcs were "myste-icusly "'ixed" by DPC - without documentation of correction the next day."
There is no evidence to support the allegatio.n that the NRC handed the problem over to Duke. Site personnel in QA and Construction-Welding are not aware of any problem welds being identified as a concern to the NRC being improperly handed to site personnel to have corrected.
No specific evidence is presented by GAP to support this allegation. In fact, it is appropriate for NRC to point out problems and pctential problems to Duke. When identified, by whatever source, Duke is responsible for correcting deficiencies in accordance with its QA Program.
It is alsc important to note that not all weld repairs warrant documentatien depending up n the stage of fabrication and the governing QA Procedure. During fabrication, the craf tsman is allowed to correct p'roblems he finds with his workmanship and depending upon the nature of the defect that may/may not get documented. .,
We question the validity of this allegation. If the problem weld existed and was truly nonconforming, we have every ' reason to believe that the inspectors would have followed the QA procedures and identified this problem weld in accordance with the QA Program.
r
.'. s' , t*Ere "5 n: 5.;;; . 't- G 's re c..:e s i a-- ii s hei;i c al s te denied. .
i .
e 9
e O
O O
9 e
%e 1
e
. 9 4
G J
e
.c ,
, ,IST OF ATT ACHu.ENTS -
ATTACHMENT DOCUMENT 1 Atomic Safety and Licensing Ecard Transcript (pp. 10053-10276) 2 Atomic Safety and Licensing Ecard In Camera.
Transcript (pp. 54E-954) -
3 Quality Assurance Procedure R-2, Rev. 9 (pp.
CS-1,1, 2, 3 of 2, Forms R-2A, R-2B
-4 Testimony of Larry R. Davison (pp.1,1.3,14, 15) i.
5 Deposition of Larry R. Davison (pp.1,16-23) 6 The Construction Project Evaluation for Catawba Nuclear Station, Ur.its 1 and 2, September 27 - October 14, 1982 (SIE) r-
- t e
S
[,
._ _.__._ -