ML20041F445
| ML20041F445 | |
| Person / Time | |
|---|---|
| Site: | Yankee Rowe |
| Issue date: | 03/09/1982 |
| From: | Kay J YANKEE ATOMIC ELECTRIC CO. |
| To: | Crutchfield D Office of Nuclear Reactor Regulation |
| References | |
| TASK-15-07, TASK-15-12, TASK-15-7, TASK-RR FYR-82-30, NUDOCS 8203160509 | |
| Download: ML20041F445 (14) | |
Text
- - - - - - - - - - - - - -
YANKEE ATO 10 ELECTRIC C0~~PANY
" * ^ * " * ' ' ' " ' - * ' *
.~.
__ ff 1671 Worcester Road, Framingham. Massachusetts 01701
. YANKEE g,_30
~
j'y y
March 9,1982
' C ehe,3
,3 il Msg United States Nuclear Regulatory Commission g
- 8 tlashington, D. C. 20555 are 3
At t e r.t l on :
fir. Dennis M. Crutchfield, Chief Operating Reactors Branch #5 Division of Licensing Re f e re nce s :
(a) License No. DPR-3 (Docket No. 50-29)
(b)
USNRC Letter to YAEC, SEP Topic XV-7-Request for Additional Information, December 14, 1981.
(c) USNRC Letter to YAEC, SEP Topic XV Request for Additional Information, December 30, 1981.
(d) YACC Letter to USNRC, SEP Integrated Assessmant Program, Ja nua ry 20, 1982 (FYR 82-7).
(e) Letter f rom USNRC to Public Service Company of New Hampshire, F. J. Miraglia to 11. C. Ta l ba n, Ja nua ry 20, 1982.
Su bjec t :
NRC Request for Additional Information on SEP Topics XV-7 and XV-12.
Dear Sir:
The Attachment provides discussions cf SEP Topics XV-7 and XV-12 in response to the NRC's requests for additional information, References (b) and (c).
The evaluations submitted for these topics were performed within the scope of SEP objectives, based upon NRC-approved methodologies.
In a recent letter to the NRC (Reference [d ]), these overall objectives for SEP were stated as follows:
1.
Provide a comparison to present day criteria; 2.
Produce documentatic.: of safety margins; 3.
Provide a means to delineate potential backfits; and db 4.
Provide a mechanism to make integrated and balanced decisions with respect to the implementation of backfits.
5 Yankee believes that the information previously submitted concerning these SEP topics is sufficient to enable the NRC to fulfill these stated objectives.
![
0203160509 g % 9 POR ADOCK PDR r3
~
U. S. Nuclear Regulatory Commission Ea rc h 9, 1982 Attention:
Mr. Dennis M. Crutchfield Page 2 De further believe that the NRC's information requests go beyond the SEP objectives, in that Foth requests asserted that compliance with the Standard Review Plan's acceptance criteria is a requirement and that the NRC staff could not complete their safety evaluation of the SEP topics unless such compliance was demonst ra ted.
Fu rt he rmo re, in one of these requests, the fuel enthalpy content criterion for rod election accident assessments was discredited without explanation or justification by the NRC.
The discredited value currently appears in both Section 15.4.8 of the Standard Review Plan and in Pegulatory Culde 1.77, which provide the NRC's informal guidance for licensee evaluation of this event.
No quantitative response by YAEC is possible, however, when neither substitute guidance nor further explanation is furnished for the NRC's abolishing the previously acceptable value of this criterion for fuel enthalpy content.
A recent letter received f rom the NRC's Division of Licensing, Reference (e), expressed that a standard of " practicability" is to be used by licensees when applying the informal gcidelines f rom the Standard Review Plan. YAEC believes that the original information provided for SEP Topics XV-7 and XV-12 falls within this standard of practicability, and will permit the NRC staff's assessment to meet the objectives of SEP.
The additional information in the Attachment, however, is provided in order to clarify technical issues raised in Ref erences (b) and (c) regarding YAEC's approved licensing methodologies.
If you have any questions, please contact us.
Very truly yours, YANKEE ATOMIC ELECTRIC COMPANY 6
J. A. Kay Senior Engineer - Licensing JAK/fst Attachnent
ATTACHMENT: Additional Technical Information for SEP Topics XV-7 and XV-12 I.
INTRODUCTION Evaluations were submitted to the NRC in Reference 1 for SEP Topic XV-7,
" Reactor Coolant Pump Rotor Seizure and Reactor Coolant Pump Shaft Break", and Topic XV-12, " Spectrum of Rod Ejection Accidents". The same information provided to the NRC for these SEP topics was submitted in Reference 2 in support of licensing Core XV for operation. In Reference 3, the NRC issued its amendment allowing operation with Core XV and provided the staff's safety evaluation for the reload core.
Regarding loss-of-coolant flow events (Topic XV-7), the staff concluded in its Core XV safety evaluation that:
"The results of this reanalysis confirmed that the reference cycle analysis continues to bound Cycle XV plant operation and therefore it is acceptable."
(Reference 3, p. 10, emphasis supplied.)
Regarding rod ejection events (Topic XV-12), the staff concluded that:
" Based upon the above discussion, we find the accident analyses for Core XV to be acceptable for these events."
(Reference 3, page 4, emphasis supplied.)
Although the SEP evaluations for Topics XV-7 and XV-12 submitted in Reference 1 contained results previously appr ved by the NRC, most recently via Reference 3, additional information was requested by the NRC for these two SEP topics. The following paragraphs are provided in response to these recent information requests, References 4 and 5.
2 II.
DISCUSSION: SEP TOPIC XV-7 II.A Loss of Pumping Power Events (Complete Loss-of-Flow)
F The methodology currently employed for complete loss-of-flow analysis was submitted and approved in support of Core XI operation, References 6 and 7.
This same methodology was most recently used and approved in the Core XV analysis to confirm that Core XI results were more limiting, References 2 and 3.
Results from the Core XI analysis, which the NRC found acceptable for Core XV, were that a complete loss-of-flow event could not result in more than 1.25% fuel failures. Thus, the radiological evaluation for Topic XV-12 rod ejection events, which assumes 10% fuel failures, bound these consequences for loss-of-flow events.
Several conservatisms are used for a design basis evaluation of complete loss-of-flow, a Condition IV event for Yankee Nuclear Power Station (YNPS), namely:
1.
No credit is assumed for the coastdown delay associated with the two main coolant pumps powered from the station turbine generator -
instead, electrical power to all four pumps was assumed to be simultaneously lost.
2.
Heat conduction parameters of thermal conductivity, specific heat capacity, and gap conductance were selected to maximize cladding temperatures - instead of allowing them to vary according to their j
temperature dependencies.
3.
All clad temperatures in excess of 1100 F were assumed to indicate fuel failures for radiological evaluation purposes.
Each of these conservatisms is further discussed in turn below.
ANSI N18.2 claselfies Condition IV events as limiting fault events that are not expected to occur during plant lifetime, but are postulated because of the patential for radioactivity release.
First, the simultaneous loss of all reactor coolant pumps was determined to be an incredible event, as discussed in the submittal of Topic XVe4, " Loss of Non-Emergency Power to the Station Auxiliaries", Reference 9.
Upon loss of all non-emergency power, the main generator will disconnect from the 115 kV off-site transmission system. The two reactor coolant pumps powered from the main generator will continue to operate for 30 to 60 seconds during generator coastdown.
In the design basis analysis, however, conservative results are obtained by assuming a simultaneous loss of all pumps without crediting this staggered loss of pump pairs.
Second, the use of specified constant values for thermal properties is known to result in conservative solutions to the heat conduction equation because their temperature dependencies are very non-linear. Thus, these values for computer code calculations are appropriately selected to purposely result in a conservative cladding temperature response.
Third, the 1100 F clad failure criterion used in this methodology is significantly conservative, compared to the current-day criterion for fue.1 coolability during Condition IV events.
In contrast, the NRC's recent safety evaluation of Westinghouse Corporation's Topical Report WCAP-9500, Reference 8, concluded that a criterion of 2700 F was an acceptable limit for coolability.
In comparison, clad temperatures for the Core XI reference analysis remained below 1200 F.
Approval of this limit for Westinghouse was based upon the relatively short period of time that fuel is typically calculated to experience DNB conditions for locked-rotor type events.
If this more realistic 2700 F criterion were applied to Core XV to determine fuel failure fractions, instead of the 1100 F criterion used for Core XI I
methodology, fuel failures of less than 1.25% would be predicted.
Furthermore, design power distributions are assumed rather than the more realistic, less peaked values typical of normal operations.
If realistic power distributions are credited for loss-of-flow events, there exists reasonable assurance that no fuel failures would be predicted for YNPS.
a The evaluation of Topic XV-7 events submitted in Reference 1 explains that because the main coolant pumps for YNPS were not designed with flywheels, their low inertial mass results in a very rapid flow coastdown. Hydraulic performance for loss of power to a single pump is thus similar to;a rotor-seizure or shaf t-break event. This similarity, as emphasized in Reference 1, means that loss of power to a single main coolant pump can produce conditions of system pressure and fuel thermal performance that are as limiting as either rotor-seizure or shaf t-bre ak events.
Despite these applied conservatisms discussed above, the main coolant system pressure response is provided in Figure 1 to demonstrate that 110% of system design pressure is not exceeded for this event. Design pressure for YNPS is 2500 psia. This result is based upon calculations performed using the RETRAN computer code, assuming conservative values for initial conditions of plant opertion. This same explicit model of the main coolant system with RETRAN was recently described in Reference 13, which discussed an analysis of natural circulation cooling at YNPS submitted in response to NRC Generic Letter 81-21.
The peak pressure obtained for a complete loss-of-flow event is shown in Figure 1 to be less than 2200 psia.
II.B Seized-Rotor /Shaf t-Break Events (Partial Loss-of-Flow)
The limiting variety of partial loss-of-flow events is loss of a single pump at rated power level without direct reactor scram. During four-loop operation, a low main coolant flow signal from at least two loops must occur for direct reactor scram. Reference 1 explains that a loss of power to all four main coolant pumps, with reactor trip, is a more limiting event than a single pump seized-rotor /shaf t-break event without direct reactor trip. This was confirmed for the reference core analysis for Core XI, which was submitted and approved in References 6 and 7.
The minimum DNB ratio for either a seized-rotor or shaf t-break event is in excess of 2.54, as reported for Topic XV-7 in Reference 1.
This DNB result was conservatively obtained, on a steady-state rather than a transient basis, by assuming continuous reactor operation at full rated power plus 3%
uncertainty (e.g., 618 MWt, at a main coolant system flow rate value for three-loop operation). Although this condition is prohibited by procedure (e.g., three-loop operation disallowed per Technical Specifications), it permits conservative determination of DNB ratios for the scenario of sudden stoppage of a single pump from rated conditions. The resnits for both fuel thermal performance and system pressure response, however, are leis limiting than for the complete loss-of-flow event discussed in the previos section. No fuel failures occur for seized-rotor /shaf t-break of a single pump; therefore, the radiological consequences need not be considered.
II.C Summa ry : Topic XV-7 Recent NRC approval by the reload analysis for Core XV was based upon the same information supplied for SEP Topic XV-7.
A review of this information has been conducted, and additional results are now provided in response to the Reference 4 request. Reasonable assurance exists that the fuel pin failure estimate of 1.25% for complete loss-of-flow events is conservative. The radiological consequences of this event are bounded by those for rod ejections, for which a 10% assumed value for fuel failure is assumed. Design pressures are not exceeded for this event.
l l
l
o III.
DISCUSSION:
SEP TOPIC XV-12 III.A Rod Ejection Events
?
The me';hodology currently employed for rod ejection events was originally sebmitted and approved in support of Core XI operation, References 6 and 7.
The same methodology was most recently used and approved in the Core XV analysia to confirm that the Core XI results were more limiting, References 2 and 3.
This same methodology, which is known to be conservative, is commonly used by vendors for analysis of fuel thermal performance during rod ejection accidents. The submittal for SEP Topic XV-12, Reference 1, reviewed these conservatisms.
In addition, the same rod ejection methodology is used for Maine Yankee Nuclear Power Station licensing analysis. Reference 10 discusses this methodology in comprehennive fashion, and also describes the criteria employed for determining whether fuel rod failures are expected to occur.
If fuel failures are predicted to occur, radiological consequences are then determined.
Results submitted for YNPS since Core XI for reload cores are that no fuel pin failures are expected to occur. Thus, the assumption employed for assessing radiological consequences of rod ejection that 10% fuel pin failures occur is considered to be conservative. The conservatism that is judged to apply to this SEP Topic XV-12 evaluation has been confirmed for each core reload, including Core XV, pursuant to the NRC staff's review and approval.
This evaluation is based strictly upon a radial-average enthalpy criterion applied to fuel pin thermal performance, which is considered to be the most reliable indication of fuel pin damage for rapid high-energy insertions.
III.B Enthalpy Criterion This criterion for YNPS is that fuel pins whose radial-average enthalpy exceeds 200 cal /gm, at any point axicily along the pin, are considered
" failed".
Failure of a pin at any axial location is assumed to result in a release of all of the gap activity for the entire length of that pin.
In d
comparison, the enthalpy criterion suggested by the NRC in its Standard Review Plan, Section 15.4.8, and in Regulatory Guide 1.77 is 280 cal /gm.
In light of recently acquired empirical understanding of fuel damage criterion, obtained from NRC-sponsored research as discussed in Reference ll, the 280 cal /gm value appearing in both Standard Review Plan (Section 15.4.8) and Regulatory Guide 1.77 is considered subject to revision to some lower value. No guidance has been provided from the NRC, however, concerning wnat lower value is considered appropriate. The recent information request of Reference 5 asserts that, "the staff does not accept the 200 cal /gm limit for clad damage".
This contradicts the guidance provided in both the Standard Review Plan and the Regulatory Guide. No substitute value is suggested, however, and no mention is made in Reference 5 of the applicability of the NRC's enthalpy criterion of 280 cal /gm.
The enthalpy criterion of 280 cal /gm for rod ejection accidents was found to be acceptable by the NRC, however, as recently as May 1981 in the Reference 12 safety evaluation for Westinghouse Topical Report WCAP-9500. The NRC's non-acceptance of the even lower 200 cal /gm limit for YNPS is inconsistent with this May 1981 staff opinion.
Thus, no consistent basis is furnished for concluding that the assumed failure of 10% of fuel pins is not valid for evaluating radiological consequences. This value is judged to be conservative.
It is appropriate to emphasize that 10% fuel failures have been consistently assumed for each reload-analysis of this event, despite the prediction that no fuel pin failures would occur.
Prior to the Reference 5 request for additional SEP topic information, neither the 10% fuel failures assumption for radiological assessment nor the prediction that no failures could actually result based upon the enthalpy criterion of 200 cal /gm has been questioned by the NRC.
III.C DNB Criterion l
The subject of whether DNB ratio is a more adequate measure of fuel pin failure, instead of radial-average enthalpy, however, has been addressed in the NRC-sponsored research reported in Reference 11.
Notably, the conclusions
( L
e concerning fuel pin damage from this NRC research do not specify a DNB ratio criterion. Rather, the statement is made concerning experimental studies for rod ejection type events that:
?
"Both the PPP and SPERT results are presented in units of radial ave age peak fuel enthalpy because the NRC design criterion is written in terms of fuel enthalpy, and radial average peak fuel enthalpy can be closely correlated with rod damage at high-energy insertions."
(Reference 11, p. 584, footnote, emphasis supplied.)
This research further indicated that fuel pins subjected to high energy depositions during rod ejection type events "may fail before departing from nucleate boiling" (Reference 11, p. 593).
In addition to this experimental data suggesting that a DNB criterion for fuel damage may be non-conservative, analytical considerations of the nature of DNB correlations give weight to the applicability of an exclusive enthalpy-based criterion for radiological assessments. No DNB correlations are available for analysis of subchannel DNB ratio response during rod ejection accidents.
The subchannel conditions observed experimentally in the NRC research indicate that inlet flow reversal and high quality coolant conditions exist during a rod ejection. The Westinghouse W-3 DNB correlation, approved for Core XI and subsequent reload cores, is not strictly valid for these conditions (Reference 11, p. 589). Thus, a correspondence between DNB ratio and fuel pin failures is not supported experimentally. The enthalpy criterion, however, is so supported and is judged to be an appropriate and conservative measure of fuel pin damage for radiological assessments. The Reference 11 testing indicated that a principal mechanism of fuel failures was the fracture or tearing of previously irradiated cladding, but further stated j
that, " departure from nucleate boiling has no influence on this damage l
mechanism" (Reference 11, p. 601).
III.D Summary: Topic XV-12 l
Thus, use of an enthalpy - based criterion exclusively when performing j
radiological evaluations, rather than using a DNB-ratio criterion as the Standard Review Plant suggests, is considered to be technically justifiable based upon NRC-sponsored research. This departure from the informal guidance i
of the Standard Review Plan is not considered to result in any adverse safety consequences. In light of the NRC's recent approval of the Core XV reload submittal for the rod ejection event, reasonable assurance exists that no fuel pin failures are expected to occur based upon a 200 cal /gm criterion.
If a reasonable but lower enthalpy value is applied, the 10% assumed fuel failures for radiological assessments is considered to be a conservatively large amount. Experimental results of NRC-sponsored research indicate that the NRC's limit of 280 cal /gm may be subject to reduction, but no consistent guidance exists concerning a more appropriate value.
If DNB analyses are performed, results for the approved W-3 correlation for YNPS are not strictly valid, and new methods would need to be developed and approved for use. The NRC research, however, indicates that radial-average enthalpy corresponds more closely to fuel pin damage estimates. Thus, the 10% assumed fuel failures are considered to be conservative for assessing radiological consequences of rod-ejection events.
1 4
3. _-
e IV.
REFERENCES 1.
FYR 81-95, YAEC Letter to NRC, SEP Topic Assessments, June 30, 19 1.
?
2.
FYR 81-52, YAEC Letter to NRC, Core XV Refueling Proposed Change No.
173, March 26,1981 (4 Attachments).
3.
LS05-81-07-071, Docket No. 50-29, NRC Letter to YAEC (Notice of Issuance of Amendment No. 69 to Operating License DPR-3), July 22, 1981 (3 Enclosures).
4.
LS 05-81-12-038, Docket No. 50-29, NRC Letter to YAEC, SEP Topic XV Request for Additional Information, December 14, 1981.
5.
LS05-81-12-100, Docket No. 50-29, NRC Letter to YAEC, SEP Topic XV,
Request for Additional Information, December 30, 1981.
6.
YAEC Letter to NRC, Proposed Change No.115 (Core XI Refueling),
March 29,1974.
7.
DOL /AEC Letter to YAEC, Amendment No. 9, July 30, 1974.
8.
WCAP-9500, Westinghouse Topical Report, " Reference Core Report 17 x 17 Optimized Fuel Assembly".
9.
FYR 82-11, YAEC Letter to NRC, SEP Topic Assessment Completion, February 1, 1982.
10.
FMY 81-92, YAEC Letter to NRC, Rod-Ejection / Loss-of-Flow Methodology for YAEC, June 17, 1981.
11.
Vol. 21, Nuclear Safety No. 5, Article by P. E. MacDonald, et al.,
Assessment of Light Water Reactor Fuel Damage During A Reactivity Initiated Accident, September-October 1980.
e t
12.
Letter from NRC to Westinghouse, Acceptance for Referencing of Licensing Topical Report WCAP-9500, May 22, 1981.
13.
FYR 81-155, YAEC Letter to NRC, Natural Circulation Cooldown, November i
24, 1981.
3 i
i T
4 1
6 o
82/02/23 YR LOSS OF FLOW (RETRAN
)R04 h
l I
I I
I I
I N
o go.
N w
~
w t
M E
e me a_ -
wN LL.)
E Dwo
(.f) *
~
~
e Ltj N E
Q O
N e
N O
l l
l l
3 l
l v-O M) 2 4
6 8
10 12 14 16 TIME (SEC)
FIGURE 1.
Main Coolant System Pressure Response Complete Loss-of-Flow Event Design-Basis Analysis
.