ML17054C232

License Amendment Request for the Transition to Westinghouse Core Design and Safety Analyses - OG-92-25, Steam Generator Tube Uncovery Issue

ML17054C232
Person / Time
Site:	Wolf Creek
Issue date:	01/17/2017
From:	Wolf Creek
To:	Office of Nuclear Reactor Regulation
Shared Package
ML17054C103	List: ET 17-0001, License Amendment Request for the Transition to Westinghouse Core Design and Safety Analyses, Enclosure 10 - CAW-16-4500 ML17053B393 ML17054C219 ML17054C220 ML17054C221 ML17054C222 ML17054C223 ML17054C224 ML17054C226 ML17054C227 ML17054C229 ML17054C230 ML17054C232 ML17054C233 ML17054C235 ML17054C236
References
ET 17-0001
Download: ML17054C232 (8)

Text

7~

!W*D l,I'\IH'\ER~IIIPS'

'\ fur the

~/-

Westinghouse Owners Group Oomestt<: utilities International utilities Alabama Power Georgie Power Portiano General Electric Union ElectriC Belg1an Utilities American Electric Power Roridio Power & Light Public Service ElectriC & Gas Virginia Power Kansa1 Electnc Power Carolina Power & Light Houston Lighting & Power Rochester Gas & ElectriC Wisconsin ElectriC Power Korea Electric Power Cammon\Y601th Edison New York Power Authority South Carolina Electrtc & Gas Wisconsin PubliC Service Nuclear Electric pic Consolidated Edison Northeast Utilities Tennessee Valle'( Authority Wolf Creek Nuclear Nukleama Elei<trama Duquesne Light Norlt1em States Power ru Electric Yankee Atomic ElectriC Spanish Utilihes Duke Power Pacific Gas & ElectriC Swedish State Power Board Ta1won Power OG-92-25 March 31, 1992 Mr. Robert C. Jones, Chief Reactor Systems Branch Division of Systems Technology Office of Nuclear Reactor Regulation U. S. Nuclear Regulatory Commission Washington, D. C. 20555

Subject:

Westinghouse Owners Group Steam Generator Tube Uncovery Issue

Dear Mr. Jones:

The purpose of this letter is to document the information on the steam generator tube uncovery issue which was discussed previously with the NRC staff. As your are aware, the Westinghouse Owners Group (WOG) has been actively pursuing a program since 1988 to evaluate the potential for steam generator tube uncovery during certain accidents and to determine the significance of any uncovery on the accident analyses. Westinghouse and utility personnel have met with the NRC staff several times to review the status of the program, to present program results, and to solicit NRC staff comments. The planned program to resolve the steam generator tube uncovery issue has been completed and a summary of the program results were presented to you and other NRC staff personnel in a meeting on January 31, 1991.

Based on the results presented, it was concluded that steam generator tube uncovery does not have a significant impact on the accident analyses provided in plant safety analysis reports, and thus that the steam generator tube uncovery issue could be closed.

This letter presents a summary of the WOG steam generator tube uncovery program and requests NRC action to close the steam generator tube uncovery issue. The information presented includes a history of the WOG program, a summary of the program results as presented to the NRC staff at the January 31, 1991 meeting, and the WOG conclusions from the program. Also attached for your information is one(l) copy of WCAP-13247 (Proprietary) which contain the overheads which were presented at the January 31, 1991 meeting.

1556G

HISTORY OF WOG PROGRAM The possibility that the steam generator tubes could become uncovered during accident conditions was initially identified in late 1987 as a result of evaluations of the July 1987 North Anna Unit 1 steam generator tube rupture (SGTR) event.

Because the North Anna Unit 1 tube rupture occurred at the top steam generator support plate, questions were raised concerning the possibility that a leak or rupture at the top of the tube bundle could become uncovered under certain accident conditions. Preliminary evaluations performed, based on collapsed water level, indicated that the top of the tube bundle could potentially become uncovered for some types of steam generators following reactor trip from normal operating conditions. If the tube bundle is effectively uncovered following reactor trip, the radiological releases for accidents involving steam release from the secondary side of the steam generators could potentially be increased over that stated in the accident analyses. This is the reason that the original concern on this issue was raised.

Westinghouse issued a generic letter in December 1987 notifying all Westinghouse utilities of the steam generator tube uncovery issue. Since the steam generator tube uncovery issue could potentially affect all Westinghouse plants, the Westinghouse Owners Group (WOG) initiated a program in 1988 to address the issue.

The NRC endorsed the action of the WOG to investigate and resolve the issue on a generic basis. WOG and Westinghouse personnel met with the NRC staff on July 27, 1988 to discuss the proposed WOG program. The proposed program was based oo )

utilizing the results of the Model Boiler No. 2 (MB-2) Phase 2 test series l 1 which were performed to obtain data on SGTR events. Because MB-2 is a scaled ~est facility, the program also included a scaling analysis to demonstrate the -

applicability of the test results for full scale steam generators. Based on the initial assessment of the MB-2 test results, it was postulated that the program would show that any steam generator tube uncovery would have little effect on the calculated radioactivity release for postulated accidents.

WOG personnel met again with the NRC staff on March 29, 1989. It was concluded that the MB-2 test conditions are applicable to operating plant steam generators and that the scaling analysis demonstrated applicability of the results to full size steam generators. However, a more detailed review indicated that the radioactivity release is sensitive to the behavior of the two-phase mixture level in the steam generator, as well as differences in geometry. It was concluded that scaling of the MB-2 test results alone would not be sufficient to assess the impact of tube uncovery, and therefore a more detailed analysis would be required to further quantify the effect of tube uncovery on the radioactivity release in order to establish the safety significance of the issue. It was determined that analyses of some representative SGTR transients should be performed for the purpose of evaluating the safety significance of the issue. The WOG and the NRC agreed that the purpose of the program was to determine if the effect of partial steam generator tube uncovery on the iodine releases for an SGTR or other limiting event has a significant safety impact.

(1) K. Garbett, 0. J. Hendler. G. C. Gardner. R. Garnsey, and M. Y. Young, "Coincident Steam Generator Tube Rupture and Stuck-Open Safety Relief Valve Carryover Tests (MB-2 Steam Generator Transient Response Test Program)", NUREG/CR-4752, EPRI NP-4787, TPRD/L/3009/886, WCAP-11226, (1987).

1556G

SGTR analyses were performed for two types of representative plants to determine the extent of any tube uncovery and to determine the effect of tube uncovery on the radiological releases calculated for SGTR events. The SGTR analyses were performed for the presently assumed limiting single failure of a stuck-open steam generator power operated relief valve (PORV). Two additional scenarios, where relief valves function as designed, were also analyzed to confir. that the single failure scenario remained the limiting case. The potential effect of tube uncovery on other non-SGTR accidents involving steam release from the stea. generators was also investigated utilizing probabilistic risk assessment methodology to determine the risk significance for these events. The evaluation of the tube uncovery issue was completed and the evaluation results were presented to the NRC staff in a meeting on January 31, 1991.

SUMHARY OF WQG PRQGBAM RESULTS A summary of the results of the WOG program for the steam generator tube uncovery issue, which was presented to the NRC staff at the January 31, 1991 meeting, is presented below. The overheads which were utilized for the presentation at the meeting are also attached for your information, WCAP-13247.

Introduction In previous radiological analyses of accidents involving steam release from the steam generators, it has been assumed that elemental iodine will be transferred from the primary side via tube leaks or tube rupture and mix with the secondary-side water. The iodine is assumed to be released to the environment at a rate which is directly proportional to the steam release rate and the iodine concentration in the steam (usually defined as a fraction of the iodine concentration in the steam generator water). These analyses assumed that the tube leak or rupture site was below the steam generator water level. If the leakage site is uncovered, some of the primary coolant may bypass the secondary-side water without experiencing mixing, thus potentially creating a more direct path for the release of primary water to the environment. Since the primary water can have a higher concentration of iodine, uncovery of the leakage site has the potential to increase the release of radioactivity to the environment. All events involving steam release from the secondary side may be affected by stea. generator tube uncovery. These events include: Steam Generator Tube Rupture, Locked Rotor, Loss of Non-emergency AC Power, Control Rod Ejection, Main Steam Line Break, Feed Line Break, and Small Break LOCA.

MB-2 Test Proqr11 Reyiew The first part of the WOG program to address the steam generator tube uncovery issue was based on the application of the MB-2 test results and the performance of representative plant analyses to deterMine the impact of tube uncovery on SGTR accidents. The test conditions for MB-2 were based on typical calculated SGTR transients for PWRs, and thus the MB-2 test data are applicable for use in performing SGTR analyses. The main purpose of the Phase 2 test series performed on the MB-2 steam generator model was to obtain data on primary water bypass for an 1556G

~*

equivalent SGTR at various steam generator water levels with a stuck-open safety relief valve. The results of the tests performed with the tube rupture simulated at the top of the bundle and with water levels such that steam generator recirculation existed showed that only a small fraction (0.001~) of the primary break flow bypassed the steam generator water and was released through the open relief valve.

For these tests, the break site remained well covered with a two-phase mixture which helps to prevent the escape of primary water droplets from the boiler.

When the MB-2 water level was lowered such that the collapsed level was approximately at mid-bundle elevation, boiler recirculation ceased indicating that the mixture level had fallen below the top of the riser. When the collapsed water level was lowered well below the mid-bundle elevation, the mixture level uncovers the break site siMUlated at top of the bundle, causing the break flow to be injected directly into the steam space. This direct injection results in greater likelihood of the break flow water droplets escaping fro. the generator when compared to the situation where the break flow is discharged into the steam generator water. The results obtained at collapsed water levels $.Ub~tantially below the mid-bundle elevation indicated that the primary water bypass fraction increased, but still remained relatively low ( 0.741 average value in MB-2). The fact that only a relatively small fraction of the break flow escaped, even when the steam generator water level is low, is attributed to the deposition of the primary water droplets on the internal surfaces of the model steam generator.

From the MB-2 test results, it can be concluded that the mixture level can be significantly higher than the collapsed water level, and that the presence of~

mixture layer covering the break site is very effective in reducing the primar~

water bypass.

Steam Generator Mixture level Model In order to determine if the tube rupture site becomes effectively uncovered and the duration of any uncovery, a model was developed to calculate the mixture level in the steam generator during an SGTR transient. For the mixture level model, steam is assumed to be generated uniforMly within the entire mixture volume (tube bundle and riser), and the downcomer is assumed to be saturated liquid. The two-phase conditions in the tube bundle region are evaluated based on the continuity equations for the vapor and liquid phases. For computing the void fraction, the drift velocity is assumed to be given by that for the churn-turbulent flow regime. The basic approach is to first calculate the void fractions and liquid and vapor flows in the bundle region for a given a stea. generation rate .. Downcomer water level and collapsed liquid level in the bundle region are calculated, and the boiler recirculation flow is calculated using a simple momentum balance. Mass balances are then used to calculate new liquid volumes in the bundle region and in the downcomer, and the calculation is continued. Starting with equilibriu* full power conditions and using ti .. -dependent thermal-hydraulic analysis results for an SGTR transient as boundary conditions, the MOdel predicts the change in steam generator mass, the mixture level in the tube bundle/primary separator riser region, and the recirculation flow from the downcomer into the tube bundle. The thermal-hydraulic analysis boundary conditions used as input to the mixture level model include the main steam flow, steam generator PORV/safety valve steam flow, main feedwater flow, auxiliary feedwater flow (if used), steam pressure, SGTR break flow, and break flow flashing fraction, which are obtained from the LOFTTR2 program analysis of the SGTR transient.

1556G

Verification of Mixture Level Model A series of five tests were performed on MB-2 in which differential pressure measurements were made. From these measurements, the density distribution in the test model and, hence, the liquid and steam masses can be calculated. The mixture level code was used to simulate these five MB-2 tests in order to verify the calculational model. The model mixture level predictions were in good agreement with the MB-2 test data, with the predictions lower than the test for the cases where the mixture level was below the top of the riser. This indicates that the model conservatively overpredicted the extent of steam generator tube uncovery.

Primary Water Byoass Scaling Since MB-2 was an 0.~-size power-scaled model, the primary water bypass results cannot be applied directly to a full sized PWR steam generator. A method was therefore developed to calculate the deposition of water droplets on steam generator internal surfaces for the case in which the mixture level is below the rupture site and the break flow is discharged directly into the steam space. Based on the available data, it was concluded that the drop size distribution created in a steam generator tube rupture or leak is approximately log-normal, with a number mean diameter of about 20 microns and a geometric standard deviation of 0.7. This distribution is representative of small leaks, and is considered to underpredict the drop size for larger tube ruptures, since the latter have a greater probability of producing larger drops. This drop size distribution is conservative since smaller drop sizes would result in the prediction of larger releases. A simple droplet deposition model was developed for use in the analysis to account for the differences between the internals of MB-2 and those of full-scale steam generators.

The droplet deposition model was developed using deposition mechanisms which include gravity, turbulent diffusion, and Brownian diffusion. Scaling of the MB-2 test results to PWR steam generator geometry was accomplished by benchmarking the deposition model to the MB-2 test data and then using this model in the analyses of SGTR transients for full size PWR steam generators. The benchmark calculation performed for the MB-2 model boiler using the droplet deposition model predicted that 0.791 of the injected SGTR flow was bypassed, which compares well with the 0.741 average of the four measured values.

The model predicts that the primary water bypass fraction is a function of the steam velocity and the stea. generator internal geometry and surface area. The Westinghouse steam generators were categorized into three groups according to size and number of pri.ary separators for the purpose of evaluating the droplet deposition effects. The three groups are large separators (40.5 to 56 inches inside diameter), mediu. separators (19.5 inches inside diameter), and small separators (7 inches inside d1a.eter). The droplet deposition calculation model was then applied to predict the primary water bypass characteristics as a function of the steam velocity for the geometry of these three different groups of PWR steam generators.

RePresentative SGTR Transients Analyses were performed using the LOFTTR2 code to characterize the thermal and hydraulic response to various SGTR scenarios for representative plants and to provide the boundary conditions for the mixture level IOdel. The representative plants used for the analysis were a four-loop plant with Model 51 stea* generators and a two-loop plant with Model 44 steam generators. The SGTR analyses were performed for plants with Model 44 and 51 steam generators since these types of 1556G

l steam generators are expected to be more susceptible to tube uncovery. These generator types also result in the highest scaled primary water bypass fractions using the droplet deposition model. The LOFTTR2 analyses were performed for the following accident scenarios: (a) SGTR with stuck-open PORV, (b) SGTR with modulating PORV, and (c) SGTR with cycling safety valve. The stuck-open PORV case was analyzed since it was expected to result in the highest stea* release and consequently the greatest offsite dose consequences. Although the modulating PORV and cycling safety valve cases are expected to result in lower steam releases, these cases could potentially be more limiting due to depressed steam generator mixture level resulting from low steam flow conditions. The depressed mixture level could result in longer periods of effective break uncovery and the increase in primary water bypass might result in higher offsite dose consequences. The results of these LOFTTR2 analyses (steam flow, steam pressure, break flow, and break flow flashing fraction) were then utilized as boundary conditions to compute the mixture level in the stea. generator throughout the transients. Knowing the *ixture level in the faulted stea. generator permits the computation of the net liquid height above the apex of the tube bundle (i.e., the collapsed liquid height computed in the mixture region above the apex of the bundle), which is the assumed location of the tube rupture.

Iodine Transoort Quantification for SGTR Events A method was developed to quantify the iodine release due to each of the release mechanisms, including the iodine in the steam from the steam generator water, the iodine in the flashed primary break flow, the iodine in the primary water bypa~s, and the iodine in the secondary water carryover, for the SGTR analyses. The primary water bypass fractions, calculated for the full scale steam generators utilizing the droplet deposition model, were used to determine the primary bypass contribution when the rupture site is uncovered. The rupture site was considered to be uncovered when the net liquid height above the apex of the tube bundle is less than two feet.

The iodine release was calculated for each of the SGTR scenarios utilizing standard iodine release methodology and applying the scaled primary water bypass fractions when the break is uncovered. The duration of break uncovery and therefore the amount of primary water bypass was determined using the relevant LOFTTR2 and mixture level computations. The results of the iodine release calculations indicated that the SGTR analysis with a stuck-open PORV results in the highest iodine releases, and that the primary water bypass iodine release due to steam generator tube uncovery is only approxi*ately 11 of the total iodine release.

Although the results for a SGTR with a modulating PORV and a SGTR with a cycling safety valve resulted in a higher relative primary water bypass contribution, the total iodine release for these cases were still bounded by the results for a stuck-open PORV. Thus, it was concluded that the effect of tube uncovery is essentially negligible for the limiting transient, and the current design basis SGTR analysis methodology is adequate and remains valid.

Non-SGTR Events The WOG program also included an assessment of the impact of tube uncovery on non-SGTR events involving stea* release fro. the steam generators. A probabilistic risk assessment approach was used to address the effect of steam generator tube uncovery on non-SGTR events. Although the relative effect of tube bundle uncovery on the calculated thyroid dose for non-SGTR events with coincident tube leakage 1556G

..

-*~

'

above the mixture level could be significant, potentially exceeding acceptance levels, the probability of a significant iodine release is very low for such events. The offsite iodine release risks and consequences of non-SGTR events were categorized based on consideration of the following factors: frequency of the accident requiring considerable steam relief; frequency of loss of steam dump and condenser; frequency of steam generator tube apex leakage; degree of fuel failure, if any; primary release path (steam generator leakage or leakage to containment);

and anticipated depth of effective tube uncovery. For accidents for which there is no calculated clad damage (such as loss of offsite power, steamline break, or feedline break), the calculated offsite doses have been demonstrated to be well below the acceptance criteria. With consideration of tube bundle uncovery effects, the resulting offsite doses for these accidents are still expected to be well below the acceptance criteria. For those accidents for which some fuel rods are calculated to be in DNB, the probability of the event and the set of conditions leading to significant offsite releases is considered to be very low (less than 5.0x10- 8 per reactor-year) for a representative plant. It*is possible that plant specific considerations may modify the transient behavior and change the probability value. However, for the representative plants, this probability ~~uld indicate that this issue is in the exclusion category as defined in NUREG-0933.\ J CONCLUSIONS OF THE WOG PRQGBAM The WOG program is considered to represent a comprehensive effort to determine ~he effect of partial steam generator tube uncovery on the iodine releases for SGTR and non-SGTR events. The results of the WOG progra. indicate that the impact of steam generator tube uncovery on the limiting SGTR event (i.e.,with a stuck-open PORV) is negligible. For non-SGTR events, the probability that an event could result in offsite radiological consequences that exceed the acceptance limits is estimated to be sufficiently low so as to place this issue in the exclusion category as defined in NUREG-0933. Therefore, it is concluded that the current design basis accident analysis methodologies are adequate and remain valid. On this basis it is concluded that the steam generator tube uncovery issue could be closed without any further investigation or generic restrictions.

REQUESTED ACTION Based on the above infor.ation and the previous discussions with you and other NRC staff personnel, we request that no modifications be made to the existing analytical models or results to address the stea. generator tube uncovery issue. Because several of our .albers have this issue as an open ita. on their licensing dockets, a letter fro. you to this effect, allowing the. to remove this issue from their licensing dockets would be greatly appreciated.

If you have any questions concerning this infor.ation, please contact Anthony Engel at TU Electric (817-897-8609) or Augustine Cheung of Westinghouse Electric Corporation (412-374-5992).

(2) NUREG-0933, "A Prioritization of Generic Safety Issues", (1983) 1556G

LETTER ATTACHMENT Attached for your inforMation is one(1) copy of WCAP-13247, *Report on the Methodology for the Resolution of the Stea. Generator Tube Uncovery Issue*

(Proprietary). This WCAP only includes the overheads presented at the January 31, 1991 meeting.

Also enclosed is a Westinghouse authorization letter, CAW-92-287 (non-Proprietary) accompanying Affidavit CAW-92-287 (Non-Proprietary) Proprietary Infonaation Notice, and Copyright Notice. Since WCAP-13247 contains 1nfonaation proprietar.y to Westinghouse Electric Corporation, it is supported by an affidavit signed by Westinghouse, the owners of the information. The affidavit sets forth the basis on which the infor.ation may be withheld fro. public disclosure by the Commission and addresses with specificity the considerations listed in paragraph (b)(4) of Section 2.790 of the Com.ission's regulations.

Accordingly, it is respectfully requested that the information which is proprietary to Westinghouse be withheld from public disclosure in accordance with 10CFR Section 2.790 of the Commission's regulations.

Correspondence with respect to the proprietary aspects of the Application for Withholding or the supporting Westinghouse Affidavit should reference CAW-92-287 and should be addressed to.~.J. liparulo, Manager of Nuclear Safety and Regulatory Activities, Westinghouse Electric Corporation, P.O. Box 355, Pittsburgh, Pennsuylvania 15230-0355.

Very truly yours, lawrence A. Walsh, Chairman Anthony W. Engel, Vice Chairman Westinghouse Owners Group Westinghouse Owners Group RNL/AWE:dc attachments cc: Steering Co..ittee (ll)

Westinghouse Owners Group Primary Representatives (1l)

C.K. McCoy, Georgia Power (ll)

J.A. Bailey, Wolf Creek (ll)

K.D. Desai, NRC (1l)

N.J. liparulo, H (1l, lA)

K.J. Voytell, H (1l) 1556G