ML20004C564

Final Deficiency Rept Re NSSS Radiograph Noncompliance to Code Requirements,Initially Submitted on 800522.Deficiency Is Not Reportable Per 10CFR50.55e

ML20004C564
Person / Time
Site:	Grand Gulf
Issue date:	05/29/1981
From:	Mcgaughy J MISSISSIPPI POWER & LIGHT CO.
To:	James O'Reilly NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION II)
References
10CFR-050.55E, 10CFR-50.55E, AECM-81-157, NUDOCS 8106040281
Download: ML20004C564 (9)

Text

. .' 0 i -

,/j j MISSISSIPPI POWER & LIGHT COMPANY J) Helping Build Mississippi -

adMiddidds P. O. B O X 1840. J ACK S O N, MIS SIS SIP PI 3 9 2 05 Y,UfMd,f," May 29, 1981 ,X y

-8 1

[y Office of Inspection & Enforcement c  : 2 ~, yl3 --

g U. S. Nuclear Regulatory Commission ( y, t . .a aE -

Region 11 L , e j[

101 Marietta Street, N.W. g c 'C d a S Suite 3100 '

Atlanta, Georgia 30303 -

Attention: Mr. J. P. O'Reilly, Director

SUBJECT:

Grand Gulf Nuclear Station Units 1 and 2 Docket Nos. 50-416/417 File 0260/15525/15526 PRD-80/28, Final Report, NSSS l Radiograph Noncompliance to Code Requirements AECM-81/157

References:

1) AECM-80/134, 6/20/80

2) AECM-80/261, 10/20/80 On Mr.y 22, 1980, Mississippi Power & Light Company notified Mr. M. Hunt, of your office, of a Potentially Reportable Deficiency (PRD) at the Grand Gulf Nuclear Station (GGNS) construction site. The deficiency concerns the failure of radiographs, supplied by subcontractors to our NSSS vendor, to meet ASME Code requirements. This report. was originally due on May 7, 1981, but exten-sions 'ere obtained until May 29, 1981, during telephone conversations with l your Mr. P. Taylor on May 7,1981, and Mr. "ajac on May 22, 1981.

Our investigation of th.(9 deficiency has been completed. We have deter-mined that, although certain radiographs do not meet the ASME Code require-

! ments, there would be no adverse ef fects on safety. Thus, the deficiency is not reportable within the meaning of 10CFR50.55(e). Our final report is submitted as Attachment A to this report.

Additionally, during a meeting in Atlanta, MP&L committed to provide information with respect to the NRC open item pertaining to Dikkers Radio-graphs. This information is submitted as Attachment B to this letter. '

Yours truly, yJ. P. McGaughy, Jr.

30/7 3

ATR/JMK:st Attachment A: Review of NSSS Radiographic Film except Dikkers B: Review of Dikkers Radiographic Film

!f Member Middle South Utilities System 6 m

~

w AECM-81/157  ?

. O'Reilly Page 2

. N. L. Stampley

. R. B. McGehee .

. T. B. Conner

. Victer Stello, Director fica of Inspection & Enforcement S. Nuclear Regulatory Commission . . , .

shingeca, D.C. 20555

. G. B. Taylor

.uth Mics. Electric Power Association

0. Box 1589 icticeburg, MS 39401 l

1 1 i

l l

l

, . i O  !

- bec -Dr. D. C. Gibbs Mr. D. C. Lutken Mr. J. N. Ward Mr. W. A. Braun Mr. R. Trickovie Mr. J. W. Yelverton Mr. L. F. Dale Mr. C. K. McCoy Mr. T. H. Cloninger i Mr. R. A. Ambrosino Mr. R. C. Fron Mr. G. B. Rogers Mr. M. R. Williams Mr. L. E. Ruhland Mr. D. L. Hunt Mr. A. G. Wagner Mr. P. A. Taylor PRD File File l

e i

r i

O I

l i

l l

l Attachment A to AECM-81/157 .

Page 1 of 4 FINAL REPORT FOR PRD-80/28 (Review of NSSS Radiographic Film except Dikkers)

I. Description of the Deficiency Two (2) NRC inspections (416/79-23 and 416/79-34) identified various radiographic technique discrepancies with respect to non-destructive examination of components provided by two (2) sub-suppliers (Dikkers Valves and Associated Pipe) of NSSS equipment /co'aponents. As a result, MP&L commissioned Bechtel to review, for radiographic technique, ten percent (10%) of the radiographs provided by all NSSS sub-suppliers except for the two aforementioned sub-suppliers previously cited. This review included 14 additional sub-suppliers who had provided a total of 7,961 radiographic films as of February 13, 1980. Information relative to Associated Pipe was previously submitted by AECM-80/67, April 4,1980 in response to NRC Infraction Item 79-34-01.

The review was performed to the applicable edition and addenda of the ASME Code as noted on the Code Data Report. It encompassed radiographic density, selection of penetrameters, penetrameter sensitivity, number of penetrameters and completeness of coverage.

The review resulted in the identification of nine (9) sub-suppliers whose radiographs failed to meet ASME Code requirements.

II. Analysis of Safety Impl! ations Due to the nature of the discrepancies, a review of 100% of onsite radiographic film provided by NSSS sub-suppliers for Unit 1, with the exception of Dikkers Valves, was conducted. This film represented pipe spools, pumps, valves, and associated components. The review consisted of approximately 10,276 film locations, which includes,those provided by Associated Pipe, and revealed approximately 1,424 deficiencies. These deficiencies can be categorized as follows:

Catego ry Total Deficiency Deficiency Density 821 Densities eaceed limitations specified by code.

Documentation 72 Material thickness mis-stated and/or radiographic technique I indicated on reader sheet is

! dif ferent from technique l used.

Fenetramete rs 97 Incorrect size, inadequate sensitivity, or shim in the l area of interest.

l l

t

, , - - e

1 Attachment A to AECM-81/157 Page 2 of 4 C4tegory Total Deficiency l Deficiency Penetrameter/ 434 Combination'of the Documentation deficiencies as noted above for the separate category but

! not readily discernible.

The 821 film locations with deficiencies in density have been determined readable. Even though the radiographs do not meet code requirements, by varying the lighting intensity, in and out . movement of radiographs, composite or triple viewing of film, overlapping film, performance of ultrasonic examination or viewing component end prep areas on installa-tion welds, we have verified defect free components.

Documentation errors were identified on 72 film locations. These resulted because the radiographic technique indicated on the radio-graphic reader sheet was dif ferent than the technique actually used or incorrect material thickness was recorded. Through measurement of penetrameter images on the film and ultrasonic thickness examinations, we are confident that acceptable techniques were employed and incorrect documentation related to radiographic technique will not affect material quality.

The 531 deficiencies categorized as either penetrameter (97) or pene-trameter/ documentation (434) errors have been determined to be satis-f acto ry. Information supplied by the vendor confirms that minimum wall thickness was recorded on the radiographic reader sheet in lieu of the actual material thickness and that the penetrameter selection was based on actual material thickness plus clad thickness at the time of radio-graphy. We believe that correct penetrameters were used and that the material thickness was incorrectly recorded on reader sheets. To con-firm this assumption, MP&L calculated the equivraint penetrameter sensi-tivity as specified in ASME V, Article 22. 4!niem wall thickness, and required penetrameter/"T"-hole versus actual p4aetrameter/"T"-hole were used in the computations. All discrepant film has sensitivity that exceeds code requirement. The clad pipe ends also received a 100%

ultrasonic examination and a liquid penetrant examination of each wold layer at the time of cladding. Satisfactory equivalent penetrameter sensitivity calculations were also performed on other systems where utilization of incorrect penetrameter was indicated.

The chain of events which lead to HPSL final acceptance of the above items are as follows:

o MP&L contracted Nondestructive Testing Engineering Division, Hartford Steam Boiler Inspection, to perform a 100% review of all on site G.E. supplied film for Unit I. The review was conducted by a team of four reviewers, one certified Level III and three certi-fied Level II interpreters.

1

Attachment A to AECM-81/157 Paga 3 of 4 -

o Each radiograph was reviewed and the results documented on data I sheets. This team documented every noncompliance, no matter how minct. An example of this level of documentation ist densities.

were recorded as 1.98 where the required minimum is 2.0, and densi- -

ties of 4.01, where the maximum requi ed is 4.0. MP&L believes that these radiographs have been reviewed more striagently than any radiographs at existing nuclear sites. i o On completion of this review, MP&L acquibed the services of two recognized consultants in the field of radiography: Mr. Sam Wank, Institute Engineer with Southwest Research Institute and Mr.

Charles Hellier, General Manager of the Nondestructive Test Engineering of Hartford Boiler Inspection. MP&L Quality Assurance assigned Mr. James Kelley, from our QA Staf f, to coordinate the effort. Mr. Kelley has extensive experience in radiography.

o These individuals reviewed each potential problem that had been identified by the review team and recommended a disposition to each concern. Many of the recommended dispositions were based on the premise that certain radiographs, although not meeting each requirement of the code, were interpretable and could be used to assure the integrity of the component. When the variety of the components is considered, the dif ficulty of this task is evident.

The manufacturer who originally reviewed the radiographs was familiar with the configuration of the component and also had the component in an unassembled state to visually inspect. This would have allowed the immediate visual verification of any questionable indications found by the reader. For* example, Main Steam Isolation Valve film revealed what appeared to be linear anomalies which proved to be only " mold marks" when the valves were physically inspected by our team. In the majority of cases,' the manufac-turer's interpretations were substantiated af ter reviewing drawings and visually inspecting the items. This is why " lack of adequate information on the reader sheet" was identified as a category of deficiency. In many instances, the original reviewer was satisfied as to the identification of an indication and accepted the radio-graphy, without documenting the rationale for accepting the film.

o As MP&L attempted to clarify the limits of acceptability for densi-ties, a review of alternate codes revealed quite a variance as to

, the limits industry has determined to be acceptable. For instance,

! for piping, whereas ASME Section III requires a density range between 2.0 and 4.0, ANSI B31.1 allows a range from 1.3 to 4.0, and older editions of Navships 250-1500-1 allowed a range from 1.5 to 3.3. Again, we reiterate that the review of radiographs is subjec-tive and dependent upon the experience and knowledge of the pro-cess, and of the reviewer. MP&L did not set acceptable or reject-able limits for density. The team reviewed each radiograph which was noted to be outside the code limit and judged its accept-ability. The recommended resolution to each problem was then j documented and dispositioned.

I 1

Atttchment A to AECM-81/157 Page 4 of 4 MP&L is not trying to infer that there should not be adherence to the ASME Code; rather, we are postulating that radiographs of 'c ertain com-ponents, such as pump and valve castings, are many times impractical, if not impossible, to radiograph and meet all aspects or the code due to the configuration of the component. The ASME Code, Figure NX2573.1-1, allows discretion in these areas, but for many of the components instal-led at Grand Gulf, radiographs were provided covering the total area of

~the components. In these areas some discerning judgement must be exer-cised in determining acceptability.

Detailed documentation is available to justify MP&L's acceptance of the radiographs in question.

III. Corrective Actions Taken In addition to the actions noted above pertaining to Unit 1, MP&L intends to provide an additional confidence level in the adequacy of radiography for Grand Gulf, by subjecting a sampling of Unit 2 film to a similar review. The review of Unit 2 Film will include any new sup-pliers who had not previously supplied components for Unit 1.

i 9

• e D

i l

l 1

1 I

. _ . -_ _ _ _ . , _ . _ _ _ , -, . . .,a,, ,- y.p . _,-- ,_r

Attachrant B to AECM-81/157 Page 1 of 2 Inspection Item 79-23-01 (Review Dikkers Valves Radiographic Film)

Corrective Steps Taken and Results Achieved

1. G.E. contracted the Nuclear Engineering Service Company to perform a 100 percent review of GCNS Unit I and Unit 11 safety relief valves (SRVs) manufactured by Dikkers Valve Company. The 100 percent film review con-firmed that the radiogrepsic quality level was adequate for film inter-pretation and, in many instances, exceeded code requirements. The radio-graphs did not reveal any unacceptable discontinuities and all film was readable for defect evaluation.

2. To satisfy questions relating to the dark density area of the inlet flange, re-radiography was performed on areas (3-4-5) and (10-x-y) for each Unit 1 SRV valve. This re-radiography was perfoemed for information only to provide an additional level of confidence and did not reveal any unacceptable discontinuities. This re-radiography substantiated GE's statement that the valves were sound in the heavy banded area.

3. As part of the original GE procurement documents, all valves were sub-jected to a 1.5X design pressure hydrostatic test and performance verifi-cation test under full steam and flow conditions. Each valve was deter-

' mined operable and structural integrity was sound.

4. The raised adjusting bolt pads on approximately 80. valve bodies, similar to those used at GGNS, were re-radiographed by the Dikkers Valve Company. This re-radiography did not reveal any unacceptable discontinu-ities and confirmed that the material casting process is ' sound. This leads to the conclusion that the valves supplied for use at GGNS are also sound.

5. Stress was calculated for CGNS Unit I, II and spare safety relief valves. There was no indication of any region with excessive stress or inadequate thickness. In general, the stresses are well below the pre-scribed code limits. For example, the calculated body wall thickness is 0.63 inch, whereas the actual minimum wall thickness is 1.1 inches. The primary stress at the crotch between inlet and outlet is 3,493 PSI, com-paring to a code allowable of 18,900 PSI. The primary stress at the crotch between bonnet and outlet is 4,524 PSI comparing to a code allow-able of 18,900 PSI. This demonstrates that there is significant excess margin in the valve design.

6. Where radiographs display density variations out of specification limits, the location on the casting is that of substantial extra thickness or of rapidly changing cross sectioa.

7. HP&L contracted Mr. Sam Wenk (Southwest Research Institute) to review a sampling of Dikker Valve radiographs. Mr. Wenk is a certified ASNT Level l III Radiographic examiner. Mr. Wenk and an MP&L representative previously l certified as an ASNT Level III Radiographic examiner, reviewed 100% of the l

_ ._. - _ . - _ , , . . ._.m _ .

. . ,~.-

• O Attachment B to AECM-81/157 Page 2 of 2 -

radiographs on nine SRVs. Additionally, they reviewed all the areas of the valves cited in NRC inspection report number 416/79-23. From this review, they concluded that all the reviewed radiographs were readable fuc defect evaluation and all components reviewed were defect free.

8. Two NRC Inspectors, Mr. S. A. Wenk, and an MP&L representative reviewed 100 percent of the radiographs for one Dikkers SRY. They concluded that the radiographs were readable for defect evaluation and in some instances had a quality level of 2-1T sensitivity. The samplings listed in para-graphs 7 and 8 represent more than 20 percent of all GGNS Unit I, II and spare Dikker SRV radiographs.

9. MP&L acknowledges that the Dikkers valves do not meet all code require-ments relating to radiography, particularly in areas where the casting is of substantial txtra thickness or rapidly changing cross sections.

Investigations made and actions taken clearly establish that no rejectable material discontinuities exist in the GGNS Dikkers safety relief valves.

The anomalies existing in the radiographic film will not af fect safety.

Actiona Taken to Prevent Recurrence All Dikkers radiographs for GGNS Unit 1 & 2 valves have been received on site. Should additional radiographs be received from Dikkers in the future, MP&L intends to review a sampling of the item. GE has reported the following corrective actions to MP&L:

1. Dikkers Valve Co any has upgraded its QA and radiographic progria to the satisfaction of 7

NRC vendor compliance inspector.

2. GE held a training seminar and instructed 45 GE quality assurance field representatives on the QC requirements, including the interpretation of radiographs and code requirements. Also, GE imposed densitometer scanning requirements for radiography density upon GE's vendors on new purchase o rde rs .

l l

. ..-n.

- - -c.-_ <.m,.y -- - , y- -. -c, .w- , y+- , y u-