LD-97-024, Part 21 Rept Re Error in Energy Redistribution Factor Used in LOCA Analysis for Listed Plants for Which ABB-CE Performed Analysis.Issued Recommendations Info Bulletin 97-004
| ML20210K246 | |
| Person / Time | |
|---|---|
| Site: | Calvert Cliffs, Palo Verde, Saint Lucie, Arkansas Nuclear, Waterford, San Onofre  |
| Issue date: | 08/14/1997 |
| From: | Rickard I ABB COMBUSTION ENGINEERING NUCLEAR FUEL (FORMERLY, ASEA BROWN BOVERI, INC. |
| To: | NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM) |
| References | |
| REF-PT21-97 LD-97-024, LD-97-24, NUDOCS 9708190196 | |
| Download: ML20210K246 (5) | |
Text
- __
August 14,1997 LD-97-024 l
1 Document Control Desk U.S. Nuclear Regulatory Commission Washington, DC 20555
Subject:
Report pursuant to 10 CFR 21 Regarding Error in the Energy Redistribution Factor Used in LOCA Analysis
References:
)
April 1,1971.
1
(
Gentlemen:
The purpose of this letter is to notify the Nuclear Regulatory Commission of a defect pursuant to 10 CFR 21, " Reporting of Defects and Noncompliance." The identified
" defect" is that the cae gy redistribution factors (ERF) used by ABB-CE in LOCA analyses did not directly reflect the efTects of moderator voiding Juring a LOCA and such effects have recently been calculated to be somewhat higher than previously thought. This involves all plants for which ABB CE performed the LOCA analysis of record (AOR).
This defect afTects only the Large Dreak LOCA analysis significantly, since the Small Break LOCA analysis msensitive to the ERF.
j The variation of ERFs with moderator voids was documented in Reference 1 in response to NRC questions. All calculations were performed using the SIIADRAC code. These results were used, in part, to derive the ERF value of 0.94 cited in the NRC approved large break LOCA evaluation model topical report. The reported ERF was based on a hot rod pin / box ratio of 1.07 with the hot rod located adjacent to a CEA guide tube location, and was calculated for the Mark V (14x14) assembly type. These assumptions were i
representative of the core designs existing at the time the ERF was derived. The ERF values reported in the LOCA topical reports were correctly calculated for assembly ig designs which were typical at that time and included the efTects of voiding.
h[
P Subsequently, with the implementation ofimproved assembly designs, lower hot rod pin / box ratios (~1.04) were observed. Lower values of the pirt box ratio yield higher
/
values of the ERP since there is less redistribution of the hot rod power among ABB Combustion Engineering Nuclear Systems
\\{llllll)$IE!!'Il!!
l 19 Q()5i}
Cmommo Eng.oecnng loc M1J Day HJi IDM Teiegf&e (S@) UsS-1911 i
ro tw sm r3. <swms 5m 9708190196 970814' wnosacnum e PDR ADOCK 05000317 S
PDR;
. _ _ _ _g
LD 97-024 Page 2 neighboring fuel rods. In addition, the implementation of enrichment zoning causes the hot rod location for some core designs to be in locations that are not adjacent to a CEA guide tube. For these reasons, the variation of ERF with pin / box factor for a fuel rod not located adjacent to a CEA guide tube was calculated in 1975 using the hiORSE hionte Carlo computer code for both the 14x14 and 16x16 assembly designs.
The 1975 calculation included benchmarking to the Reference 1 (SIIADRAC ) results for the ERF using the same geometrical configuration, a pin / box ratio of 1.07 and no voids, and was found to be in essential agreement. The MORSE calculation, however, was found to yield somewhat more adverse ERFs for actual lattice geometries. On the basis of l
the agreement for the benchmark geometry and the more adverse results for actual I
geometries, MORSE was used to determine ERFs as a function of pin / box ratio for both 14X14 and 16X16 lattice types. Ilowever, review of the 1975 calculation indicates that the calculation did not incorporate the effects of voiding during a LOCA.
The results of Reference 1 indicated that the effects of voiding (as calculated by SHADRAC) were relatively small. Since the 1975 calculation contained a number of conservatisms, it was believed that new calculations would confirm the continued applicability of the ERFs incorporated in the ABB-CE ECCS performance evaluation models, even in the presence of voiding. Calculations were consequently initiated in early July,1997 using ABB-CE's current Monte Carlo computer code for photon transport, i
MCNP. The McNP calculations indicate that the effect of voids is larger than had been calculated in Reference 1 using SHADRAC. As a result, the calculations have failed to confirm the continued applicability of the ERFs used in ABB-CE's ECCS evaluations.
Specifically, the MCNP calculations indicate that the energy deposition in the hot rod is underpredicted by 0.5 to 1.5%.
A 0,5 to 1.5% underprediction of the ERF corresponds to an underestimate of approximately 20 to 60 F in PCT in the large break ECCS evaluation. Depending on the value of PCT calculated and the magnitude of compensating conservatisms present in the AOR, the corrected PCT could exceed the ECCS acceptance criterion of 2200 F.
ABB-CE recommended actions (described in the Attachment) to assure the validity of the current LOCA AOR. Those recommendations continue to be applicable.
Very truly yours, COMBUST 10N ENGINITE
,Inc.
MAyW lan
, Director Operations Licensing cc:
M. F. Barnoski (ABB-CE)
1 Attachment to LD.97-024
- ABB Combustion Engineering Nuclear Operations 10 CFR 21 Report of a Defect or Failure to Comniv The following information is provided pursuant to 10 CFR 21.21 (c)(4):
(I)
Name and address of the Individualinforming the Commission:
l lan C. Rikard, Director Operations Licensing i
l Combustion Engineering i
2000 Day 11ill Road Windsor, CT 06095-0500 (11) _
idestification of thefacility, the activity, or the basic component suppliedfor suchfacility or such activity within the United States whichfalls to comply or contains a defect:
The energy redistribution factors (ERF) used in the LOCA analysis for all plants for which ABB CE performed the analysis of recc:d (AOR) did not properly consider the effects of significant moderator voiding which may occur at various times during a LOCA.
(ill) identification of thefirm constructing thefacility or supplying the basic component whichfalls to comply or contains a defect:
Combustion Engineering, Inc.
2000 Day liill Road Windsor, CT 06095 0500 (iv)
Nature ofdefect orfailure to comply and the safety hat,ard which is created or could be created by such defect orfailure to comply:
The defect identified is that the energy redistribution factors (ERF) used in the LOCA analysis for all plants for which ABB-CE performed the AOR did not
- properly consider the effects of significant moderator voiding which may occur at various times during a LOCA. Specifically, calculations indicate that the energy deposition in the hot rod is underpredicted by 0.5 to 1,5%.
(v)
The date on which the information ofsuch defect orfallare to comply was obtained:
Information indicating that the ERF's may not have considered the effect of voids was obtained on July 9,1997.
2 Attachment to LD-97-024 (vi) in the case of a basic component which contains a defect orfalls to congoly, the number and location of all such conysonents in use at, suppliedfor, or being suppliedfor one or morefacilities or activities subject to the regulations of this part:
This issue affects the following plants for which ABB CE has provided the LOCA AOR:
- Palo Verde Units 1,2, and 3 San Onofre Nuclear Generating Station Units 2 and 3 e
Waterford Unit 3 e
Arkansas Nuclear One Unit 2 e
l e
Calvert Cliffs Units 1 and 2 Saint Lucie Unit 2 e
(vil)
The corrective action which has treen, is being, or will be taken; the name of -
the individual responsiblefor the action; and the length of time that has been or will be taken to complete the action:
For plants for which ABB CE has performed the LOCA AOR, ABB-CE has provided the following recommendations via Infobulletin 97-04:
For the ABB CE Digital Plants, assure that one of the following conditions is met:
Assure that there is at least 0.2 kw/A margin between the COLSS PLHGR and the PLIIGR LCO value Reduce the COLSS PLilGR LCO by 0.2 kw/R
- Assure that there is at least 2% margin between the measured e
all rods-out Fxy peaking value and the ARO Fxy value installed into COLSS For the ABB-CE Analog Plants, assure that one of the following condNons is met:
Assure that t!
-- is at least 2% margin to the Tech Spec Fxy limit, if monitoring linear heat rate using ex-core detectors, or Assure that there is at least 0.2 kw/R margin between the measured PLHGR and the PLHGR LCO, if monitoring linear -
heat rate using in core detector signals.
J
3 Attachment to LD 97-024 (vil) Any advice related to the defect orfailure to comply about thefacility, activity, or basic component that has been, is being, or will be given topurchasers or licensees:
ABB CE issued Infobulletin 97-04 on July 11,1997 to notify licensees of this issue. The Infobulletin provided the recommendations identified in item (vii) above.
The above recommendations constituted interim advice. Licensees are further advised to either; Confirm that suflicient PCT margin or other conservatisms exist in the AOR to accommodate the defect in the ERF, or Reduce the PLifR LCO by 0.2 kw/ft to compensate for the defect in e
the ERF.
s
-.-.