ML20092P669

Potential Part 21 Rept Re Bug in Detector Computer Code for in-core Detectors.Bug Incorrectly Calculated Enthalpy Rise Peaking Factor.Bug Corrected

ML20092P669
Person / Time
Site:	Mcguire, Cook, McGuire, 05000000
Issue date:	05/24/1984
From:	Shanstrom R SHANSTROM NUCLEAR ASSOCIATES
To:	Murley T NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION I)
References
REF-PT21-84-263-000 NR84051, PT21-84-263, PT21-84-263-000, NUDOCS 8407090239
Download: ML20092P669 (8)

Text

- -

e 4 SHANSTRDM NtJcLEAR ASSOCIATES P. a BOX 1122 DARIEN. CONNECTICUT O&e2O 4203) 655-9400 Ref: NR84051 May 24, 1984 Dr. Thomas E. Murley Regional Administrator USNRC Office of Inspection and Enforcement 651 Park Avenue King of Prussia, Penn. 19406

Dear Dr. Murley:

This letter with attachments constitutes a written notification of a potential 10CFR21 item, requiring " Reporting of Defects and Noncompliance."

The item involves a bug introduced in an August 1983 modification to the DETECTOR computer code. The result of the bug is that technical specification limits for the enthalpy rise peaking factor, FAH, may be incorrectly calculated.

DETECTOR is a component code of the CORE computer package (Codes for Operating Reactor Evaluation). The purpose of DETECTOR is to reduce measured results from incore detectors (miniature fission chambers) in Westinghouse PWR reactors, and combine these data with calculated values for fuel pin power distribution and detector response distribution, resulting in the best estimate for the actual power distribution in the operating reactor core.

W1.ile there is no regulatory requirement that codes such as DETECTOR perform a technical specification compliance analysis, this feature has been incorporated into DETECTOR, and is of obvious benefit to the utility company users.

In August '83 the Tech. Spec. compliance analysis was extended to allow fuel technical specifications which can vary with fuel type. In particular to cover a specification for a mixed core with both Westinghouse and Exxon Nuclear Co. fuel. Prior versions of DETECTOR, as well as use of the Aug '83 version with a single set of Tech. Sp'ec. parameters, are not affected by the bug discussed in this notification.

8407090239 840524 PDR ADOCK 05000315 S PDR fb

_ - - - - - AL-

3 Dr.iThomas E. Murley May 24, 1984 l

In practice the effect of the bug was trivial. All flux  !

maps in which the Aug '83 version of DETECTOR was utilized I have been reanalyzed. No violations of technical specification l parameters occurred, and no hazards to personnel health or safety were involved.

The bug has been corrected and responsible parties have been notified. The purpose of this notification is solely to conform to the regulatory requirements of 10CFR21.

The'Aug '83 version of DETECTOR was utilized for only one nuclear unit, the-Donald C. Cook Unit 1 Nuclear Power Plant, during Cycle 8 operation.. This plant is operated by the American Electric Power Company and their engineering support group is the American Electric Power Service Corpor-ation (AEPSC). For D. C. Cook Unit 2 AEPSC utilized an earlier. version of DETECTOR.

The only other utility company which has a DETECTOR code with the Aug '83 bug is Duke Power Company (DUKE). DUKE utilizes an earlier version of DETECTOR for current operational analysis of their McGuire Unit 1 and Unit 2 plants.

Earlier versions of DETECTOR have been provided to the Exxon-Nuclear Company, Northern States Power Company, and the Union Electric Company. . DETECTOR may also be accessed on the UCC and CDC computer service networks.

The CORE including DETECTOR code was written,'is maintained, j and is supplied-by Shanstrom Nuclear-Associates (SNA).

Pursuant to the requirements of 10CFR21, the following l notifications have been made:

i

! (1) . Telephone call from AEPSC.to SNA informing SNA of the apparent bug as discovered by AEPSC.

j Confirmation of the bug by SNA and proposed i-correction by SNA in the same call, apx. ll AM i May'22, 1984; i

j (2) Correction of the SNA version of DETECTOR and

! completion of sample run to verify.the proposed

! correction, apx. 1 PM, May.22; I

! (3) Telephone call to DUKE to notify them of the bug i and required correction, apx. 3 PM, May 23; I

.(4) Telephone-call to Dan Fieno,' Richard Lobel, and I Marv Dunenfeld of thelNRC, Core Performance Branch-to discuss the issue, apx. 9 AM, May 24 p. 3 i

i lJ

. _ _ _ . . _ _ . _ - _ . --. - a __, , _u _ . _ _ . - _ _ _ . . - . _ . - -

Dr. Thomas E. Murley May 24, 1984 (5) Telephone call for initial notification to Dr. Shanbaky, NIK: Office of Inspection and Enforcement, Region 1, apx. 10 AM, May 24; (6) Return telephone call from Dr. Shanbaky giving address for written notification, apx. 2 PM May 24; (7) This written notification dated May 24, 1984 and sent at apx. 9 AM May 25 to:

i a) Dr. Thomas E. Murley, NRC Region 1, Attn:

l' Dr. Shanbaky b) Director, Division of Licensing, NRC, l Attn:_ Mr . Richard Lobel, Core Performance Branch c) Mr. Milton Alexich,-AEPSC, Attn: Mr. George

? John l-

'; d) Mr. K. S. Canady, DUKE, Attn: Mr. Raymond P.

Wood.

! In summary, no violations of technical specification requirements

{ have occurred. AEPSC utilized the Aug '83 version of DETECTOR 2

for the analysis of forty-seven flux maps for Cycle 8 of the Donald C. Cook Unit 1 Nuclear Power Plant. Except for two low

power flux maps the code-identified the proper limiting fuel

. pin. Reanalysis of all flux maps, including the two taken at i low powers, showed that no technical ~ specification violations

~

cccurred. Responsible parties have been informed and corrections to the DETECTOR code have been effected.

Should you wish more information please call Dr. Raymond T.

Shanstrom at (203) 655-9400.

4 i

i Very truly yours, i

G

/M N - -~

l lhW Raymond T. Shanstrom hb i

4 i

-Attachments si '-

. . . , _ ._- , ~ _._ --._,._ ,~.-. ~ . . , _ .- ....-... - -,

3

, , Dr. Thomas E. Murley May 24, 1984 Att. 1. Modification to August 1983 Version of DETECTOR to Correct Bug in Calculation of Limiting F a ues.

AH Subroutine S1234 (Bug is in location of Statement No. 21370)

Aug '83 Modification RBFCT = 1. - RBFCT FSHTS = FSHTSX(K) 21370 TSFSH = FSHTS

• RBFCT TSDIF = TSFSH - FSUBH Correction RBFCT = 1. - RBFCT 21370 FSHTS = FSHTSX(K)

TSFSH = FSHTS

• RBFCT TSDIF = TSFSH - FSUBH Discussion:

Statement Number 21370 is an entry point if calculation

, of the rod-bow penalty (RBFCT) is bypassed. FSHTSX(K) is the constant multiplier (see "CONST. MULT." in Att. 2.) in the technical specification limit for F For example, AH.

l FSHTSX (K) =a k *( * +

k *( * ~} }

where, ak ""d Dk are constants for technical spevification parameter set k, and P is the core power relative to rated power. The error caused FSHTS to always be set to FSHTSX(KMAX),

the values for the last technical specification parameter set.

This error only occurs if the calculation of the rod-bow

(

penalty is bypassed.

(

l i

i I

i

Att. 2. Sample Edit of F AH Technical Specification Parameters 105029 975 POWEp. CIDAEDE . SouE PHONY TECH SRCCS FOR SET 2 AEP - THIMRLE DATA CONSTANT FACTOR INCLUDED IN THE CALCULATION OF ACTUAL ENTHALPY RISE. FSURH MEASUREMENT UNCERTAINTY FACTOR. FStlRHU = 1 0400 9653 h O

RATIO OF ACTUAL POWER TO RATED POWER. P =

9 7EMP FLT CONST. $

T.S. TLOW FACTOR FACTOR MULT.

SET FACTOR t1 0.0000 0.0000 1.0000 1.5205 1

0.0000 1 0000 1.5114 3 2 0.0000 ~"

c N

H

• INDICATES vt0Laff0N OF TECH SPEC $~ (D N

20 LOWEST FRACTIONAL MARGINS FOR ENTHALPY RISE FACTORS ROD BOW TECH SP. ACTUAL TECH SPEC MARGIN FUEL F.A.

FSURHN FACTOR FSURH FSURH DIFF. FDACT. VIOL. I ORD if SET ASM. LOC.

1 37,4 9998 1.5111 a.4325 .0786 0549 m 1 7 2 232 3 -N 1.4325 .0786 .0549 2 7 2 250 3 -C 1.3774 9998 1.5111 8 9958 1.5111 1.4325 .0786 0549 3 7 2 5)) ]3-N 1.3774 0549

~

13-C 1 3774 9998 1.5111 1.4325 .0786 4 7 2 529 .09R6 0694 6 309 6 -L 1 3672 1.0000 1.5205 1.4219 5 1 1.5205 1.4219 .0986 0694 6 6 321 6 -E 1 3672 1.0000 1

1.3672 1.0000 1.5205 1.4219 .0986 .0694 7 6 1 441 10-L 1.4219 .0986 0694 8 6 453 10-E 1.3672 1.0000 1.5205 3

1.0000 1.5205 1.4217 .09R8 0695 9 6 308 6 -L 1.3670 0695 1

320 6 -E I.3670 1.0000 1.5205 1.4217 .0988 10 6 1

.0988 .0695 6 440 10-L 1.3670 1.0000 1.5205 1.4217 11 1 1 3670 1.0000 1.5205 1.4217 .098R .0695 12 6 1 452 10-E .0990 0696 6 280 5 -4 1.3668 1.0000 1.5205 1.4215 13 1 1.4215 .0990 .0696 14 6 2R9 5 =F 1.3668 1.0000 1.5205 1

1 3668 1.0000 1.5205 1.4215 0990 .0696 15 6 1 471 Il-K 0696 3 4RO 11-F 1 3668 1.0000 1.5205 6.4215 .0990 W 16 6 1

.1007 0710 6 281 5 -K 1 3651 1.0000 1.5205 1.4197 M 17 3 1.4197 .1007 0710 18 6 1 290 5 -F 1 3651 1.0000 1.5205 y 1.5205 1.4197 .1007 0710 19 6 1 472 ll-K 1 3651 1.0000 0710 45 6 4R1 11-F 1.3651 1.0000 1.5205 1.4197 .1007 ,

20 1 H

Discussion: $

.o.

This is a copy of a sample run output used to test the Aug '83 Version of DETECTOR.

Phony Tech. Spec. parameters were used for Fuel Type 7, TS Set 2, to test the rod-bow penalty calculation and to cause FT 7 fuel pins to appear in the edit values. The "CONST. MULT." is FSHTSX(K), see Att. 1. This edit is always correct even in the Aug '83 l

'.* . Dr. Thom:s E. Murlev May 24, 1984 Att. 2. Discussion (cont'd.)

version. The " TECH SP. FSUBH" is TSFSH of Att. 1, ie FSHTS times the " ROD BOW FACTOR." The values in the edit of Att. 2 are correct since the calculation of rod bow factor was not bypassed (even if unity). If this factor had been bypassed values for " TECH SP. FSUBH" for Fuel Type 6, Tech. Spec. Set 1, would have incorrectly been edited as 1.5114. (This bypass was utilized by AEPSC). In fact the technical specification for Unit 1 does not require a " ROD BOW FACTOR" nor does it require a " FLOW FACTOR" or a " TEMP FACTOR."

The values "FSUBHN", F H, ae e est estimates from the DETECTOR code for the enthalpy rise peaking factors. These values.are correct even in the Aug '83 modification. The

" ACTUAL FSUBH" values are "FSUBHN" times the measurement uncertainty factor "FSUPHU", F. The key technical specification requirement is that " ACTUAL FSNBH" values not exceed " TECH SP.

FSUBH." This requirement was fulfilled in all the maps that utilized the Aug '83 version of DETECTOR.

l 4

l i

r l

i

Dr. Thoman E. Murley May 24, 1984

-Att. 3. n< commended Additional Surveillance Current Surveillance:

-(1, SNA verification.

(2) DETECTOR Training & QA Course Both AEPSC and Duke personnel have completed this course. In this training hand-calculational

• verifications are performed for all the DETECTOR results, starting with raw detector data and progressing to technical specification compliance, for the limiting fuel pin locations for Fg, FAH' and F PDC (Q surveillance).

Recommended Additional Surveillance:

(1) Increase the size of the edits for F AH "" O technical specification edits (eg from 20 to the maximum code allowance of 100). This would have clearly identified this particular bug since the

" TECH SP. FSUBH" for TS Set 1 would have incorrectly been listed as the "CONST. MULT." for TS Set 2.

(2) For each change in DETECTOR versions and for any change in input values for calculational options, the user should verify, via hand calculations, that the DETECTOR results for limiting technical specification are valid for each fuel type.

(The SNA verification and the DETECTOR training include hand-calculation verification of results for all expected options).

y

-Dr. Thomas E. Murley May 24, 1984 Att. 4. List of Addresses for Responsible Parties I

official Notification: '

Dr. Thomas E. Murley Regional Administrator (Region 1)

USNRC Office of Inspection and Enforcement 651 Park Avenue King of Prussia, Penn. 19406 (215) 337-5000 Vendor:

Dr. Rrymond T. Shanstrom Shanstrom Nuclear Associates PO Box 1122 Darien, CT 06820 (203) 655-9400 Copies:

Mr. Milton P. Alexich Ass't. Vice President American Electric Power Service Corporation PO Box 16631 Columbus, Ohio 43216 (614) 223-1000 Mr. George John, (614) 223-2055 Mr. K. S. Canady, Manager Nuclear Engineering Services Duke Power Company Nuclear Production Department PO Box 33189 Charlotte, NC 28242 (704) 373-4011 Mr. Raymond P. Wood, (704) 373-2373 Mr. Darrell G. Eisenhut, Director Division of Licensing Office of Nuclear Reactor Regulation t'SNRC Washington DC, 20555 Mr. Richard Lobel Core Performance Branch Division of Systems Integration Office of Nuclear Reactor Regulation USNRC Washington DC, 20555 (301) 492-9475

.