ML20199F117

Notice of Nonconformance from Insp on 970210-0513. Nonconformance Noted:Vendor Failed to Verify Adequacy of Anfb Critical Power Correlation & Adequacy of Application to Atrium 10

ML20199F117
Person / Time
Issue date:	10/27/1997
From:	NRC (Affiliation Not Assigned)
To:
Shared Package
ML20199F104	List: ML20199F101 ML20199F117 ML20199F159
References
REF-QA-99900081 NUDOCS 9711240115
Download: ML20199F117 (6)

Text

o l

4 NOTICE OF NONCONFORMANCE Siemens Power Corporation Docket No.: 99900081 Nuclear Division.

Pichland, Washington l

On February 10 through May 13,1997, the staff of the U.S. Nuclear Regulatory Commission (NRC) performed an inspection of activities conducted by the Siemens Power

-l Corporation - Nuclear Division (SPC), at the company's facilities in Richland, Washington.

j The results of that inspection found that SPC failed to perfonn certain activities in accordance with applicable NRC requirements, as detailed in this notice of nonconforma're.

A.

Criterio !!!, "DcJign Control," of Appendix B to Title 10, Part 50, of the bde of Dderal Regulations (10 CFR Part 50), requires that design control measures shall provide for verifying or checking the =d~;a y of design, and shall be applied to i

reactor physics, stress, thennai, hydrauls, and accident analyses.

Section 4, " Design Control," of SPC's Quality Assurance (QA) Program topical

+

report, EMF-1, " Quality Assurance Pregam for Nuclear Fuels and Services,"

Revision 28, dated Febmary 10,1995, requires that design verirmation activities shall be perfonned in accordance with sub-tier QA and Engineering procedures. (The NRC approved EMF-1 on January 16,1996, as meeting the reqeirements of Appendix B to 10 CFR Part 50.)

SPC adopted the NRC approved methodology to detennine the safety limit and operating limit minimum critical power ratios (SLMCPR and OLMCPR) for boiling-water reactor (BWR) licensees. Specifically, the methodology, now the property of SPC, was developed by Advanced Nuclear Fuels Corporation (ANF), as documented y

,in ANF 1125(P)(A), "ANFB Critical Power Conclation," Supplements 1 and 2 dated April 4,1990. As stated in the conditions of approval in the NRC's safety evaluation (SER) for ANF 1125(P)(A), this NRC-approved and SPC adopted methodology limited the local peaking factor to F:. s; 1.3, and did not permit a flow-dependent bias.

Contrary to the above, the following examples demonstrate SPC's failure to comply j

with established methodologies and, therefore, constitute Nonconformance 99900081/97-01-01.

(1) - IIn the following instances, SPC failed to verify the adequacy of the ANFB i

" critical power correlation and the adequacy of its application to the '

+

' ATRIUM *-10 fuel assemblies designed for the Pennsylvania Power and Light.

Company, Susquehanna Unit 2 Cycle 9 relot,d:

[

JM71agvgg t.

m m

i

- (a)

The appikation of the ANFB correlation to the ATRIUM *-10 fuel assemblies for Susquehanna Unit 2

- Cycle 9 showed that the local p-=Hng factor (F,) was t

greater than 1.3, which placed it outside the NRC-approved range for the ANFB correlation of less than i

1.3.

l (b)

As used to determine the SLMCPR and OLMCPR for the ATRIUM"-10 fuel assemblies for Susquehanna Unit 2 Cycle 9, the ANFB corr 1=' ion included a l

nonconservative flow bias and, therefore, was outside the l

NRC approved SPC methodology, i

(2)

SPC failed to develop a adequate number of test points, rnd failed to test an adequate range of conditions to justify the uncertainty values for the additive constants" used in determining the SLMCPR for the ATRIUM"-9 fuel design.

This implies that SPC should have used larger uncertainty values in the.

SLMCPR determination,, in order to reflect the full operability range of the -

ATRIUM" 9 fuel design. In addition, because the results of the ANFB correlation are used as inputs to the safety limit taethodology, this has immediate implications regarding the SLMCPR and OLMCPR of the following plants with ATRIUM"-9 fuel:

Commonwer'th Edison Company Quad Cities Unit 2 Cycle 15 Dresden Unit 3 Cycle 15 12Salle County Unit 2 Cycle 8 Washington Public Power Supply System Washington Nuclear Unit 2 Cycle 13 B.

Criterion III, " Design Control," of Appendix B to 10 CFR Prat 50, requires that design control measures shall provide for verifying or checking the adequacy of design, and shall be applied to reactor physics, stress, thermal, hydraulic, and acc!: lent analyses.

Section 4, " Design Control," of SPC's Quality Assurance (QA) Program topical report, EMF-1,

• Quality Assurance Program for Nuclear Faels and Services,"

Revision 28, dated Febnsary 10,'1995, requires that design verification activities shall be performed in accordance with sub-tier QA and Engineering procedures. (The NRC approved EMF-1 on_ January 16,1996, as meeting the requirements of Appendix B to 10 CFR Pan 50.)

2 r

4.-

, -.., -.- -,,--~.-.,w.,

,n-.,...,r w-,

~,,-r-.-y.-4

.w,

-wm.,,,,w,,w.r-.

.s QAP4 requires that Research and Technology (R&T) shall be responsible for j

supporting the design process by developing and maintaining computer codes and methodologies. QAP-4 also states that design verification is the process of reviewing, confirming, or substantiating (validating) the design by one or more methods to l

provide assurance that the design meets specifad requirements.

i Contrary to the above, the following examples demonstrate SPC's failure to comply with established procedures and/or methodologies and, therefore, constitute l

Nonconformance 99900081/97 01-02.

j (1)

For thermal hydrau.ic (T/H) analysis, SPC used a variety of system codes, such as ANF RELAP and S RELAP. SPC derived these codes from other codes developed by the NRC and the U.S. Depanment of Energy (DOE).

SFC modified the codes after adopting them for in house use and but failed to adequately verify, develop or maintain the modifed codes.

In panicular, SPC failed to (a) perform verif% tion activities for codes that are purchased from outside sources for use by SPC, (b) document testing and verification of the code, (c) perform adequate code validation, and in many instances, perform independent validation.

'(2)

Software development record (SDR) 10315 documents a change to the COTRANSA code, which is used to perform plant specific anticipated operational occurrence (AOO) analyses. After implementing this change, SPC did perform the required verification and validation (V&V); however one of the cases used in the V&V of this case was not technically adequate. SPC -

subsequently identified and corrected an input error, and reanalyzed the case with satisfactory results. However, SPC failed to record this input error or its effect on the analysis.

(3)

SDR 10610 documents several small discrepancies in the results obtained using the UNIX and Cray versions of the FLEX code, which could lead to a reduction in the calculated peak cladding temperature (PCT). SPC stated that these discrepancies were caused by differences in the floating point precision of the two computer systems. However, SPC failed to confirm and record in SDR-106-10 its. fmdings concerning this assumed causal factor.

(4)

As reponed in ANF-91-048(P)(A), published in 1993, comparison between the RELAX and FLEX codes and their respective primary documentation (XN-

.NF 98019(P)(A) published in 1982) verified that (a) the code solution

. procedure had been modified (b) important T/H models had been changed substantially, and (c) a number of new input options had been added.

Nevenheless, the reported verification of these changes corr.isted of a single

. systems calculation. No comparisons with experimental data were included, 3

= = -

~. -

e 4

nor were there any test casse to show that models unaffected by the modification gave the same results as the previous versions of the code.

Therefore, SPC apparently implemented modifications to the RELAX and FLEX codes without adequate verifier. ion of changes.

C.

Criterion !!!, " Design Control " of Appendix B to 10 CFR Part 50, requires that design control measures shall provide for verifying or checking the adequacy of j

design and shall be applied to reactor physics, stress, thermal, hydraulic, and accident analyses.

- 10 CFR 50.46(a)(3)(i) states that a sigel%

.c change or error is one which results in a calculated PCT which differs by more u, n 50'F from the temperature calculated for the limiting transient using the last acceptable model.

i Contrary to the, above, SDR-1241, whl;h addresses changes in TOODEE2, usug95 code, documented a change in PCT of 63'F, resulting from a change in RELAPS code boundary conditions. Even though the change in PCT was more than 50'F, SPC failed to consider it a significant change and provide sufficient information to NRC licensees so that licensees could report the nature of the ciange to the NRC as required by 10 CFR 50.46. This constitutes Nonconformance 99900081/97-01-03.

D.

Criterion 111, " Design Control," of Appendix B to 10 CFR Part 50, requires that design control incasures shall include provisions to ensure that appropriate quality

- standa.-ds are specified and included in design documents and that deviations from such standards are controlled.

Criterion V, " Instructions, Procedures, and Drawings," of Appendix B to 10 CFR Part 50, requires that activities affecting quality shall be prescribed by and accomplished in accordance with documented instructions, procedures, or drawings.

Section 4, " Design Control," of SPC's Quality Assurance (QA) Program topical report, EMF 1, "Qu:.lity Assurance Program for Nuclear Fuels emi Services,"

Revision 28, dated Februry 10,1995, requires thet design verification activities shall be perfonned in accordance with sub-tier QA and Engineering procedures. (The NRC approved EMF 1 on January 16,1996, as meeting the requirements of Appendix B to '9 CFR Part 50,)

QAP-4 requires that R&T shall support the design process by developing and maintaining computer codes and methodologies. In addition, QAP-4 states that errors sre to be reported according to the requirements of EMF-P00,066, QAP-13, " Control of Nonconformances " Revision 0, dated August 22,1996, and that corrective actions and their disposition are to be performed according to the requirements in EMF-P00,067, QAP 14, " Corrective Action," Revision 0, dated August 22,1996.

F 4

g.

p,

.---.-.w--

,wn.,

,rn r~

~

Contrary to the above, the following e.tamples demonstrate SFC's failure to established adequate procedures and/or methodologies and, therefore, constitute Nunconformance 99900081/97-01-04.

(1)

SPC failed to establish an acceptable verification and validation (V&V) procedure for code development and modification. SPC also failed to establish an adequate code design verification method, in tu icular, SPC failed to clearly define minor changes or errors, and it did nci establish an adequate code assessment process that included assessment of code changes against apr griate phenomenological tests, separate effects tests, and/or integral system tests.

(2)

EMF-608, " Computer Code Control Requirements - Engineering,"

(EMF-608), Revision 12, dated December 27,1995, and Revision 13, issued during the inspection on February 7,1997, failed to address notification concerning code errors and evaluation of the effects of such errors on the end-user. Revkion 14 to EMF-608, issued on March 12, 1997, also did not adequately address these issues.

(3)

SPC failed to provide adequate instructions to its analysts for use in selecting the appropriate entrainment fraction to establish the time of reflood for LOCA analyses. SPC then changed the process for determining the time of reflood, establishing not only a relative entrainment fraction, but also an " absolute" criterion dt.;ived on the bases of the mass velocity of entrained liquid, as documented in SPC memorandum, " Response Packages for Issues from NRC Inspection Week 1," dated March 14, 1997.

Howevn the rellood mass velocity is subject to significant oscillations, spikes, and other unsteady behavior (that is, it is not a smooth monotonically increasing function of time). Consequently, SPC failed to establish criteria rqnrding the steadiness of the entrainment mass velocity and the minimum ra% ion over which the mass velocity exceeds the " absolute" value used to 4mtms.e the onset of reflood.

E.

Criterion II, " Quality Assurance Program," of Appendix B to 10 CFR Part 50, requires that the QA program shall provide for indoctrination and training of personnel performing activities affecting quality as necessary to ensure that they achieve and maintain suitable proficiency.

Sectior 18, " Training," of QA topical report EMF-1, Revision 28 requires the respective supervisor / manager be responsible for defining, implementing, and documenting appropriate training to ensure that employees achieve and maintain the proficiency needed to carry out their assignments.

5

To funher define the applicable procedures, SPC developed EMF-P00,071, QAP-18,

" Training," Revision 0, dated August 22,1996. In panicular, that procedure requires that the applicable tranager snall use assessment of education and experience to determine the specific type (s) and extent of tra'.dng necessary to corcpetently perform a specific job. Highty technical jobs (for ext mple, it' Engineering) may require extensive education and prior experience befc o an ir divi 6al is qualified to perform the related responsibilities without oversight or w,, stance.

Contrary to the above, SPC's current training program fails to provide the necessary assessment of education and experience. SPC failed to establish the method or procedure for determining the knowledge and skills needed to perform the tasks associated with the functions of a specific engineering-related job or job category, as required by EMF-1560, " Nuclear Engineering and Product Mechanical Enginering Training and Proficiency Assessment Guideline," Revision 1. In addition, SPC failed to establish a method or procedure for evaluating the effectiveness of the training program. This constitutes Nonconformance 99900081/97-01-05.

Pinse provide a written statement or explanation to the U.S. Nuclear Regulatory Cuanission, A'ITN: Document Control Desk, Washington, D.C. 20555, with a copy to the Chief, Quality Assurance, Vendor Inspection, and Maintenance Branch, Division of Reactor Controls and Human Factors, Office of Nuclear Reactor Regulation, within 30 days of the date of the letter transmitting this notice of nonconformance. This reply should be clearly marked as a " Reply to a Notice of Nonconformance," and should include for each nonconfonnance (1) the reason for the nonconformance, or if contested, the basis for disputing the nonconformance, (2) the conective steps that have been taken and the results achieved, (3) the corrective steps that will be taken to avoid funher noncompliances, and (4) the date when your corrective action will be completed. When good cause is shown, consideration will be given to extending the response time.

Because your response will be placed in the NRC Public Document Room (PDR), to the extent possible, it should not include any personal privacy, proprietary, or safeguards information so that it can be placed in the PDR without redaction. If personal privacy or proprietary information is necessary to provide an acceptable response, then please provide a bracketed copy of your response that identifies the information that should be protected and a redacted copy of your response that deletes such information. If you request withholding of such material, you mM11 specifically identify the portions of your response that you seek to have withheld and provide in detail the basis for your claim of withholding (e.g., explain why the disclosure of information will create an unwarranted invasion of personal privacy or provide the infonnation required by 10 CFR 2.790(b) to support a request for withholding confidential commercial or financial information). If safeguards information is necessary to provide an acceptable response, please provide the level of protection described in 10 CFR 73.21.

Dated a3 Rockville, Maryland thisMPday of October,1997 6

l l

_