ML20197F942
| ML20197F942 | |
| Person / Time | |
|---|---|
| Issue date: | 12/18/1997 |
| From: | Black S NRC (Affiliation Not Assigned) |
| To: | Reindl R ACCU-TECH EVALUATION SERVICES, INC. |
| References | |
| REF-QA-99901307 99901307-96-01, 99901307-96-1, 99909076-94-01, 99909076-94-1, NUDOCS 9712300331 | |
| Download: ML20197F942 (4) | |
Text
_
&l,((gl Y ht$
f ym T~500 i
y+
t UNITED STATES p
[
NUCLEAR REGULATOF1Y COMMISSION t
WASHINoToN, D.C. 20665 0001 4
4
,, g Decenber 18, 1997 Mr. Roy P. Reindl, Branch Manager ACCUTECH' A Division of B&G Manufacturing Co. Inc.
3873 W. Oquendo Road Las Vegas, NV 8P118 Char Mr. Reindi:
Thank you for yout 'atter dated May 2,1997, in response to our Notices of Violation and Nonconformance that were issued by the nuclear Regulatory Commission (NRC) on March 4,1997, as a result of an NRC inspection of ACCUTECH (Inspection Report 99901307/96-01).
We have reviewoo your reply and found it generally responsive to the concems raised in the Notice of Violation and the Notice of Nonconformance. However, there are several points and requests made in your letter which we would like ta address.
Violations 99901307/96-01-01 and 99901307/96-0102.
Your rcp!y to Violations 99901307/96-01-01 and 49901307/96-01-02 documented acceptable corrective act'ons to address the identified violat:ons to 10 CFR Part 21 (Part 21). However, your letter stated that ACCUTECH, a division of B&G Manufacturing Co. Inc. (B&G) was not bound to take formal action under Pcrt 21 for items manbfactured and sold by the previous owner, CardinalIndustrial Products (CIP). Your position on ACCUTECH's Part 21 responsibility has been previously addressed in obr May 29,1997, letter to you, which was issued subsequent to the NRC's receipt of the May 2,1997, ACCUTECH letter. The letter indicated that the NRC staff concluded tnat ACCUTECH was incorrect in tha intarpretation of its responsibilities under Part 21 by characterizing them as " courtesy notifications." Part 21 regulations require that the pertinent provisions of the regulations be implemented in the event of the discovery of a defect in a basic component, and ACCUTECH is bound to take formal action pursuant to Part 21, and must do so because of the rquiatory requirements it assumed due to its percha:,* ci the CIP nssets.
Nonconformance 99901307/96-01-03 Your reply to Nonconformance 99901307/96-01-03 documented corrective actions to address the identified nonconformance which concemed providing subcontracted supplier's certifications as attachments to Certified Material Test Reporte (CMTR) provided to the customer. Your reply stated that *QA personnel have been instructed to adnere to requirements imposed by the NRC...." The NRC does not impose technical or quality requirements on vendors. Such requirements are imposed contractually by the purchaser of the material. ACCUTECH certified compliance with the purchaser requirements by issuing a Certificate of Compliance or a CMTR fjg and therefore the Nonconformance identifies examples of failuro to comply with the purchase document requirements, pnD hf
& n Q >N h Q d M q y y - w,r R9 %
~~
9712300331 971218 J
PDR GA999 EPfV1 ACCU i' 4 " { n 99901307 PDR
% Hi,
i i I t
! d Mr Reindi 2-Nonconformance 99909076/94-01 03 Your reply to Nonconformance 99909076/94-0103 documented corrective actions to address the identified nonconformance which concemed the sampling methods used by ACCUTECH during the commercial grade dedication process of fasteners provided as safety-related to NRC licensees. You provided a B&G document titled, "ACCUTECH Sample Plan Methodology,"
(ASPM) dated April 15,1997, which described a proposed ACCUTECH sampling plan and requested that the NRC provide a review and acceptance.
The NRC staff has not performed a technical and acceptance review of the ASPM. The acceptability of the ASPM is dependent upon the end use of the dedicated commercial grade item (CGI), its acceptability needs to be determined by the end user of the dedicated CGI, and it is therefore the responsibility of the end user to accept the sampling phin. Because the required confidence level of a sample plan is dependent on the end use of the materir.1 being sampled, the end user of the dedicated CGI, not B&G, should specify (and accept) ths minimum confidence level to be used during CGI dedication. For example, the cust' amer may specify end uses that are highly-safety-significant applications and as such may regt ire: 1) 95 percant confidence that there are only 5 percent defective items in a lot (95/5),2) 100 percent verification of all critical characteristics, or 3) that the material be supplied in accoidance with ASME Section lil, NA-3800, material upgrade requirements. In addition, for dedicated CG!s used in low safety-significant applications, the customer may specify or accept the use of a graded approach, as discussed in the NRC draft regulatory guide DG-1064, "An Approach for Plant-Specific, Risk-Informed Decision Making: Graded Quality Assurance," dated March 24,1997.
The following comments are provided with respect to the ASPM as it addresses the NRC technical report, " Sampling Plans for Dedicating Simple, Metallic Commercial Grade items at Nuclear Power Plants," dated February 1997, Page 1? of the ASPM discusses ACCUTECH's use of visual and dimensional ch.da :.s e
indication of product homogeneity and the relationship of the information obtained by these checks to the " partial chemistry" addressed in the NRC technical report. The
" partial chemistry" analyses, addressed in the NRC technical repor1, are required to provide confidence of heat / lot homogeneity and are integral to ensuring a high confidence level that the items being sampled are homogenous. Also, the comparison of the results of the partial chemistry analyses, the full chemistry analyses, and the mechanical tests to the CMTR provided with the CGls provides additional confidence that the items being sampled are from a homogenous lot and is critical to the dedication and its sampling process. The performance of acceptable dimensional checks does not provide any information regarding the chemical and mechanical properties of the dedicated items.
Page 15 of the ASPM, which refers to the tolerances provided in the Product Analyses Talerances Table on Page 8 of the NRC technical report, contains an inaccurate interpretation of the tolerances prodded in the table. The tolerance provided in the NRC technical report is the percent tolerance from the CMTR value. The tolerance given in the figures and tables bracket the values listed on the CMTR and are a subset of the material
i e
Mr, Reindl specification range and its tolerances. Therefore, if the tolerances given in the mt.terial specification are more limiting, the tolerance given in the material specification should be used. Additionally, if the table or figures are not applicable to the element or materialin question, the tolerances contained in the material specification should be used.
The example contained on Page 15 of the ASPM, which addresses the tolerances listed in the NRC technical report for chromium in austenitic stainless steel, incorrectly specifies the tolerance as 1.25% and incorrectly calculates a 16.75-19.25 range weight percent (1811.25). As specified by the NRC technical report, the correct iglerance for 18.00 percent chrome would be 0.225 (18.00 x 1.25%), which would provide a 17.775 18.225 range weight percent. The assumption presented on Page 15 of the ASPM,'nat the tolerances contained in the NRC technical report are based on dividing the tolerance by the inw value for the range, is incorrect. The tolerances contained in the NRC technical report are the result of experience obtained from current mill melting practices.
The comparison of the chemistry analyses and mechanical and other test results to the values on the CMTR is a critical element of the dedication process and is intended to be used in conjunction with the other provisions of the NRC technical report, and should not be looked at as an individual element of the dedication and sampling process. Further, for items that do not require dedication, such as NCA-3800 meterial apgrades, the provisions contained in the NRC technical report would not be applicable.
Correct use of the tolerances, provided in the Product Analyses Tolerances Table on Page 8 of the NRC technical report, would indicate that all four product analyses provided on Page 15 of the ASPM would be within the tolerances of the heat analyses contained in the NRC technical report and would be acceptable. For exampM, the ASPM shows a heat analysis of 0.41 for carbon for Product Analysis No. 2. Using Figure 1 of the NRC technical report, the resulting tolerance would be t 0.045, and a maximum carbon of 0.455 would be permitted. Ther3 fore, correct use of the tolerances provided in the NRC technical report for Product Analysis No. 2 would result in the product being acceptable, not rejectable as stated.
Further infomiation on the subject of sampling is contained in the enclosed draft NRC Regulatory Guide, Draft DG-1070, " Sampling Plans Used for Dedicating Simple Metallic Commercial Grade items for Use in Nuclear Power Plants," dated September 1997, which provides guidance on the use of sampling when dedicating simple, metallic products. The draft NRC regulatory guide is available for public comment until January 30,1998.
..... _ _.... ~ _. ~.
_._. _ _ _.. ~
_ +'
.:e -
Mr. Reindi 4 --
h We will review the implem9ntation of your corrective actions during a future inspection to determine that full compliance has been achieved and will be maintained. -In accordance with 10 CFR 2.790 of the NRC's " Rules of Practice," a copy af this letter will be placed in the NRC
- Public Document Room (PDR).
Sincerely, c
(.
&Q h DL
.e Suzanne C. Black, Chief Quality Assurance, Vendor inspection, and Maintenance Branch Docket No: 99901307
Enclosure:
NRC Regulatory Guide, Draft DG-1070 cc:
Bill Edmonds, President, B&G Manufacturing Co. Inc.
f 1
d N
i a:
4 a
4 d
e v
---rm,y e
s
-m-A*e w
re-
--ve---y r-
- +
--+-4h-m:
^*=igiv--*
mf-rb
--r----r-omewe p
tgg
?q W-9-fg s
.