ML20056A561

Safety Evaluation Re Util 860108 Submittal of Steam Generator Tube Rupture Analysis.Nrc Does Not Concur W/ Conclusion Reached as to No Steam Generator Overfill During Design Basis Steam Generator Tube Rupture Accident

ML20056A561
Person / Time
Site:	Callaway
Issue date:	08/06/1990
From:	Office of Nuclear Reactor Regulation
To:
Shared Package
ML20056A560	List: ML20056A559 ML20056A561
References
NUDOCS 9008080225
Download: ML20056A561 (8)

Text

-_

. p uc o UNITED STATES

,[, Ig

[ g NUCLEAR REGULATORY COMMISSION

'5 2 WAsHINoToN. D. C. zosss

\...*/

SAFETY EVALUATION BY THE OFFICE OF NUCLEAR REACT 0f. REGULATION

, RELATING TO STEAM GENERATOR TUBE RUPTURE UNION ELECTRIC COMPANY

,CALLAWAY PLANT DOCKET NO. 50-483 1.0 DACKGROUND On January 8,1986, the Union .lectric r Company (UE) and Kansas Gas and Electric Company (KG&E) jointly submit 9d a steam generator tube rupture (SGTR) analysis applicable to both the Calla,ay Pla $. and the Wolf Creek Generating Station (Reference 1). The submittal is covietely independent of the generic SGTR methodology submitted by Westinghouse. Although these two SNUPPS (Standardized Nuclear Unit Power Plant System) plants have Westinghouse-designed Nuclear Steam Supply Systems, they chose to analyze the SGTR accident independently, .

with the key dif ference being the use :,f the RETRAN computer code by SNUPPS as opposed to the LOFTRAN computer code as used by Westinghouse. The two plants  : ,

have virtually identical SNUPPS designs and forwarded joint original submit-

• tals; however, the proposed analyses are being reviewed separately because they  !

tended to deviate from one ancther after UE switched the Callaway plant from i Westinghouse standard (LOPAR) to Vantage 5 fuel assemblies.

2.0 INTRCDUCTION I In accordance with License Condition 2.C.(11) of Operating License NPF-30, UE )

has submitted a SGTR analysis for NRC approval. Two scenarios were analyzed as '

requested by the staf': the scenario most conducive to steam generator over-fill, and the scenario with maximized offsite dose. The main concern in the SG l overfill scenario is the resultant water release to the environment through the

( safety valves, which are only designed and qualified for steam passage. (Water  ;

l relief through the safeties may impede reseating of the valve, thereby poten-tially resulting in continuous release.) In the maximized offsite dose sce-  !

natio, the concern is that radioactive offsite doses will exceed regulatory limits.

3.0 EVALUATION OF OVERFILL SCENARIO 3.1 Assumptions '

For the SG overfill scenario, the worst-case single failure for the four-loop plant is the failure of the auxiliary feedwater (AFW) control valve in the full-open position on the faulted SG loop. AFW flow to the faulted SG is as-sumed to be maximized at 723 gpm, with the shared flow to the intact SG assumed to be minimized at 305 gpm. The tube break is assumed to occur above the tube i sheet on the tube exit for a maximum primary-to-secondary leak rate. At the time of the rupture, operation at 102% power with 1% failed fuel is assumed.

900806:0225 900806 PDR ADOCK 03000483

p. PDC 4

L J

i l ,

.g.

The effects of turbine runback were considered for the SNUPPS plants, but were found to increase SG inventory by about 25% less than an increase in inventory due to an earlier AFW initiation. Hence, turbine runback was not assumed in the design-basis analysis for overfill. The longer duration of AFW flow initiated from 100% power into the faulted SG wa. determined to be more '

conservative and, therefore, is acceptable.

3.2 Operator Action Times The staff's evaluation (Reference 5) of the Westinghouse Owners Group WCAD-10698 stipulates plant-specific criteria for assessing operator action times in the ,

event of an SGTR. Those criteria were employed to evaluate the information provided by UE regarding operator action times during an SGTR at Callaway Unit 1.

The evaluation is based on the following:

UE's letters dated February 3, 1987 (Reference 4), May 27, 1987 (Reference 6), October 21,1988 (Reference 11), January 29, 1990 (Reference 12), and September 4, 1986 (Reference 13).

Conference calls between UE and NRC staff on August 2, 1989. January 24, 1990, and March 14, 1990.

Criterion 1. Provide simulator and emergency operating procedure training .

related to a potential SGTR. ,'

The licensee documented, by letters dated October 21, 1988 and January 29, 1990, that onsite simulator and Emergency Operating Procedure (EOP) training  ;

relevant to an SGTR are provided. The staff finds that the licensee has satisfied Criterion 1.

~

Criterion 2. Utilizing typical contro1 room staff as participants in demon-stration runs, show that the operator action times assumed in the SGTR analysis are realistic and achievable.

By letter dated January 8,1986, the licensee provided the assumed operator action times for the worst case overfill scenario (a stuck-open auxiliary feed-water control valve) and the worst case dose scenario (a stuck-open atmospheric relief valve)_ Tt.. i1censee's submittals dated October 21, 1988 and January 29, 1990 reported the demonstrated operator action times for the worst case scenarios. The assumed and demonstrated response times were as follows:

6 l

. l i ..

t DOSE SCENARIO OVERFIl' SCENARIO AVERAGE AVERAGE ACTIONS ASSUMED DEMONSTRATED ASSUME] i DEMONSTRATED (CUMULATIVE RESPONSE TIMES IN MINUTES)

Isolate ruptured 28 17 16 11 -

steam generator ,.

Initiate reactor 39 26 24 13 coolant system (RCS) cooldown i

Complete RCS 55 39 35 22  ;

! depressurization Terminate safety 58 41 38 25 injection (SI)

Equalize pressure 63 53 43 -*

• Scenario ended when SI was terminated (re: conference call between staff and ,'

licensee on March 14, 1990).

The licensee noted that the assumed times are supporttd by the demonstrated

. ti nie s . The licensee also indicated that, demonstration runs included senior reactor operators, reactor operators, and training personnel. Based on this information, the staff finds that Criterion 2 is satisfied.

Criterion 3. Complete demonstration runs to show that the postulated SGTR accident can be mitigated within a period of time compatible with overfill prevention, using design basis assumptions regarding available equipment and its impact on operator response times.

As noted above, the demonstrated times for the worst case overfill scenario are o bounded by the assumed operator action times. The licensee indicates, there-l fore, that the demonstrated' times are reasonatic and conservative. Further, i the licensee points out that the demonstrated times for the overfill scenario are as close to design-basis conditions as the simulator would allow. The staff finds that the licensee has sati sfied Criterion 3.

Criterion 4. If the emergency operatin,1 pro:soures (EOPs) specify SG sampling as a means of identifying the SG with the ruptured tube, provide the expected time period for obtaining the sample results and discuss the effect on the duration of the accident.

Step 2 of E0P E-3 for Callaway specifies SG sampling as one of four methods of  ;

identifying an SG with a ruptured tube. The licensee's February 3,1987 sub-mittal indicates that manual sampling would take 15-20 minutes, allowing suf-ficient time to isolate the affected SG by 28 minutes. The staff finds that Criterion 4 is satisfied.  :

3.3 Auxiliary Feedwater (AFW) System Model The AFW System model used by UE for Callaway is based on the assumption of censtant SG pressure (1125 psig) for the duration of the transient. The SG was found not to overfill as a result. However, Callaway's sister plant, Wolf Creek, modified its AFW flow rate to vary with the fluctuation in the faulted SG pressure, which led to the conclusion that the design-basis SGTR would lead to SG overfill. Because the results of the Callaway overfill analysis left minimal margin to overfill, and because the conservativeness of the AFW flow rate used in the Callaway analysis was not verified (Ref. 7), UE was asked to analyze an SGTR assuming overfill with the SG safeties stuck 5% open, even though it indicates that overfill would not occur. Section 4.0 of this report deals with this analysis.

3.4 Impact of New Fuel l

At the time of the original analysis (Ref. 1), Callaway used Westinghouse standard fuel assemblies (LOPAR). Since then, its reloads have been based on '

the ut, of Vantage 5 fuel assemblies instead of LOPAR. Therefore, UE modified certain core parameters, such as temperatures, flow areas, etc., assumed in I

the SGTR analysis to be consistent with the recent reload analyses.

Key parameters, including reactor coolant system (RCS) pressure, faulted SG pressure, faulted SG liquid volume, and, break flow rate, were compared by UE for the two different fuel types. The analysis (Ref. 6) showed very similar behavior of those four parameters, although the endpoint values--those at the end of tae SGTR event--were a bit higher than the results of the original analysis which was based on the LOPAR fuel assembly. However, even though the Vantage 5 results show less margin to overfill, the radiological consequences of an assumed overfill for the design-basis SGTR scenario remain within regulatory limits as discussed in Section 4.0. Therefore, because the analysis shows acceptable results with SG overfill and bounds the results of the Vantage 5 analysis which shows no SG overfill, the staff finds it to be acceptable.

3.5 Steam Line Intearity The licensee addressed the effects of steam line filling due to SG overfill.

Weight and thermal stress analyses were done, applicable to both plants, i

assuming hot flooded conditions. The analyses were based on the assumption i that the main steam line filled with water. Piping stresses were calculated l; using the same Bechtel stress analysis program used and approved for use in the current FSAR. Results from the analyses demonstrated main steam line integrity to be maintained within the allswable values of ASME Code, Section 3, for Class 2 components. Because the results are reasonable and are analyzed using a previously approved methodology, the staff finds the analysis to be acceptable.

I

Water hammer has been investigated for the SG overfill case. One of three situations must exist in order for water hammer to occur--that is, (1) rapid condensation, (2) sudden interruption of flow, or (3) entrainment of water in a steam-filled line. The licensee has concluded that none of the situations necessary for water hammer to occur in the main steam line exist in the SGTR scenario. First, there is no introduction of relatively cold water into the main steam line in the SGTR scenario, therefore condensation is not a concern.

Secondly, water flow involved in the SGTR model does not approach the high velocity necessary to cause water hammer if that flow is suddenly interrupted.

Finally, the driving pressure--between the primary and secondary sides during overfill--are too small to propel a slug of water to produce a water hammer event. The licensee has determined that the SGTR scenario resulting in SG overfill will not result in water hammer in the main steam lines. The staff finds this to be acceptable.

4.0 EVALUATION OF SCENARIO FOR MAXIMUM OFFSITE RADIOLOGICAL CONSEQUENCES 4.1 Thermal Hydraulic Evaluation of Assumptions The worst case single failure for maximized offsite dose was determined by the licensee to be a stuck-open atmospheric relief value (ARV) in the faulted SG.

In order to maximize the offsite dose, the licensee modeled the RCS to be as hot as possible, maximized the flashed fraction, and minimized the faulted SG -

inventory, with no assumed iodine partitioning. This scenario, analyzed in '

the original submittal, addresses the uncovered-break prompted by the North Anna SGTR event. However, the staff raised concerns over whether,the faulted SG would overfill and, if overfill was a consequence, whether the radiological consequences would be more severe. Therefore, at the request of the staff (Ref 6), the licensee analyzed what the staff believes to be the worst-case scenario--that is, overfill with a stuck-open safety valve. (Even though UE did not conclude overfill, at the staff's request UE performed the analysis assuming a 5% stuck-open safety valve with SG overfill.)

4.2 E_ valuation of Radiological Consequences An SGTR accident releases primary coolant to the secondary side of a steam generator, thus providing a pathway for iodine and noble gases from the primary coolant to be released to the environment. Assuming a coincident loss of offsite power, the staff evaluated the radiological consequences of the release to the environment for a consequential iodine spike (i.e. , a temporary rapid increase in rate of fuel rod leakage) and a pre existing iodine spike for the SGTR scenario which results in overfilling of the affected steam generator with release of water through the safety valve to the environment. The licensee's calculated doses for this accident meet the guidelines of Standard Review Plan Section 15.6.3.

In reviewing the licensee's finding for the accident, the staff independently calculated the thyroid dose from this accident (the whole-body dose calculated

l <

6-by the licensee was very small, and the staff accepted its value). The. staff's calculated doses are summarized in Table 2, with the assumption for the analysis provided in Table 1. Case 1 assumes no pre-accident iodine spike, while Case 2 assumes a pre-accident iuS.1e spike. Both cases assumed loss of offsite power with subsequent overfilling of the affected steam generator and the release of water and steam to the atmosphere. For Case 1, the dose is based upon the plant technical specification for the equilibrium dose equivalent I-131 primary coolant activity limit of 1.0 pCi/gm. No additional fuel failure is expected to occur as a result of this accident.

The NRC staff concludes that the distances to the exclusion area and to the low population zone boundaries for the Callaway site are sufficient to provide reasonable assurance that the calculated radiological consequences of a postu-lated steam generator tube rupture accident do not exceed (1) a small fraction (510%) of the exposure limits, set forth in ?.0 CFR Part 100 paragraph 11, for the case of an iodine spike that results from the accident, and (2) the ex-posure limits set forth in 10 CFR Part 100 paragraph 11, for the case of a pre-accident spike.

The staff's finding's are based upon (1) its audit review of the licensee's radiological consequence analyses; (2) our independent dose evaluation using conservative regulatory assumptions and conservative atmospheric dispersion factors; and (3) the Callaway technical specifications for primary and second-ary coolant iodine concentrations and for the amount of unidentified primary-to-secondary leakage in the unaffected steam generators.

5. 0 RETRAN COMPUTER CODE RETRAN02/ MOD 03 was used by the licensee to analyze the SGTR event (Ref. 1) for

' Callaway. This version of RETRAN has not previously been approved for generic licensing applications. Therefore, the NRC staff had its contractor, Interna-tional Technical Services, Inc. (ITS), evaluate (Ref. 7) the appropriateness of RETRAN02/ MOD 03 for the SGTR scenario presented by Kansas Gas and Electric Company for Wolf Creek and Union Electric for Callaway. ITS compared RETRAN02/M0003 with the previous version of RETRAN, RETRAN02/ MOD 02, and found that the staff safety evaluation (Ref. 10) written on the earlier version was applicable to RETRAN02/ MOD 03. In the evaluation of the SGTR analysis by UE, ITS found nothing that violated the restrictions imposed by Reference 10. They concluded that there is reasonable assurance that RETRAN02/ MOD 03 is adequate for use in this SGTR analysis. Therefore, the staff finds that RETRAN02/ MOD 03 is acceptable as applied in this licensing application.

6.0 CONCLUSION

S Based on the staff's review and based on additional review provided by its contractors (Ref. 6), the staff does not concur with the conclusion reached by UE as to no SG overfill during a design-basis'SGTR accident. However, the staff agrees that UE has demonstrated that, for the event of a design-basis SGTR with overfill, calculated offsite radiological consequences remain within the guidance of SRP 15.6.3 ind within the limits of 10 CFR Part 100. Therefore, the staff finds that UE ct.n successfully mitigate a design-basis SGTR accident

as shown in its Accident Analysis. Additionaly, the staff has reviewed UE's responses regrading operator action times during an SGTR, concluding that the licensee has satisfactorily verified the times assumed in the SGTR analysis for Callaway Unit 1.

7.0 REFERENCES

1. Letter from N. A. Petrick, Standardized Nuclear Unit Power Plant System (SNUPPS), to H. R. Denton, NRC/NRR, dated January 8,1986.

2. Letter from N. A. Petrick, SNUPPS, to H. R. Denton, NRC/NRR, dated April 1, 1986.

3. Letter from N. A. Petrick, SNUPPS, to H. R. Denton, NRC/NRR, dated September 4, 1986.

4. Letter from D. F. Schnell, Union Electric Company (UE), to U. S. NRC, dated February 3,1987.

5. LetterfromC.E.Rossi(NRC)toA.E.Ladieu(Westinghouse),WCAP-10698, "SGTR Analysis Methodology to Determine the Margin to Steam Generator Overfill," March 30, 1987.

6. Letter from D. F. Schnell to the NRC, "Callaway Plant Steam Generator Tube -

Rupture Analysis," May 27, 1987. '

7. Letter from H. Komoriya and P. B. Abramson, International Technical Services, Inc., to A. P. Gilbert, U. S. NRC, dated May 4, 1988.

8. NUREG-0800, Standard Review Plan, Section 15.6.3, " Radiological Consequences of Steam Generator Tube Failure'(PWR)," Rev. 2. July 1981.

9. Letter from D. F. Schnell, UE, to U. S. NRC, dated July 15, 1988.

10. Letter from B. W. Sheron, NRC/NRR to I. L. Hirst, Central Electric Generating Board, dated September 12, 1984,

11. Letter from A. C. Passwater (L) to the NRC, " Steam Generator Tube Rupture -- Operator Action Times," October 21, 1988. ,

12. Letter from D. F. Schnell (UE), " Steam Generator Tube Rupture -- Operator Action Times," January 29, 1990.

13. Letter from N. A. Petrick (UE) to H. R. Denton (NRC), " Steam Generator Tube Rupture Analysis - SNUPPS," September 4, 1986.

7.1 ACKNOWLEDGEMENT Principal Contributors: G. West, LHFB

1. Spickler, PRPS A. Gilbert, SPXB A. Gody, DRSP Dated: August 6, 1990

, 1 TABLE 1 ASSUMPTIONS USED FOR THE CALCULATION OF THE RADIOLOGICAL CONSEQUENCES FOLLOWING A POSTULATED STEAM GENERATOR TUBE RUPTURE ACCIDENT

1. Power = 3565 Megawatts thermal.

2. Pre accident dose equivalent I-131 in primary coolant = 1.0 microcuries per gram and 60 microcuries per gram (two cases analyzed).

3. Initial secondary coolant activity of 0.1 microcuries per gram DE I-131,

4. Primary to secondary leak rate of 1 gpm to unaffected steam generator.

5. Iodine release rate from fuel increases by a factor of 500 at reactor trip for iodine spiking case.

6. X/Q value: ,,

0-2 hours at 1200 meters = 1.5 x 10,3 seconds per cubic meter.

0 8 hours9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br /> at 4023 meters = 2.1 x 10 seconds per cubic meter.

TABLE 2 RADIOLOGICAL CONSEQUENCE OF SGTR ACCIDENT Low Population Exclusion Area Zone Course of Accident Steam Generator 2 Hr. Oose (REM) Dose (REM)

Tube Rupture Thyroid , Whole-Body Thyroid Whole-8ody With concomitant iodine spike 8.8 < 0.1 1. 5 < 0.1 With pre-accidental '

iodine spike 26 < 0.1 3.8 < 0.1 l

i l

l

\

4 ev -- -