ML20065H677

From kanterella
Jump to navigation Jump to search

Forwards Nonproprietary Versions of Addl Info Re Application to Facility TRs YAEC-1856P, Sys Transient Analysis Methodology Using Retran... & YAEC-1849P, Thermal Hydraulic Analysis Methodology Using VIPRE-01..
ML20065H677
Person / Time
Site: Seabrook 
Issue date: 04/04/1994
From: Feigenbaum T
NORTH ATLANTIC ENERGY SERVICE CORP. (NAESCO)
To:
NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM)
References
NYN-94035, TAC-M86957, TAC-M86958, NUDOCS 9404150111
Download: ML20065H677 (93)
v • d • e


Text

{{#Wiki_filter:en . 's . ggh, l N i Nortli North Atlantic Energy Service Corporation y.o. ne, 300 Atletic Seabrook, NII O3874 g (603) 474-9521, Fax (603) 474 2987 The Northeast Utilities System I Ted C. Feigenbaurn NYN-94035 Senior Vice President 8. Chief Nuclear Officer April 4,1994 United States Nuclear Regulatory Commission Washington, D.C. 20555 Attention: Document Control Desk

References:

(a) Facility Operating Licere,e No. NPF-86, Docket No. 50-443 (b) North Atlantic letter NYN-93020, dated February 2,1993, Request ihr NRC Review and Approval of Analysis Methodologies to be Applied to Scabrook Station," T. C. Feigenbaum to USNRC (c) USNRC Letter dated December 16,1993, " Request for Additional Information - (TAC M86957)" (d) USNRC Letter dated December 16,1993, " Request for Additional Infonnation (TAC M86958)" (e) North Atlantic letter NY'N-94024, dated March 9,1994, " Response to Request (br Additional Infonnation (TAC M86957 and TAC M86958)," T. C. Feigenbaum to USNRC

Subject:

Response to Request for Additional information (TAC M86957 and TAC M86958) Gentlemen: North Atlantie Energy Service Corporation (North Atlantic) has provided in the Enclosures. non-proprietary s ersions of additional infbrmation regarding the application to Seabrook Station of Yankee .J Atomic Electric Company Topical Reports YAEC-1856P, " System Transient Analysis Methodology Using Retran for PWR Applications", and YAEC-1849P, "Thennal liydraulic Analysis Methodology Using ' VIPRE-01 for PWR Applications" This infonnation was requested by the NRC staff [ References (c) and ) (d)]. The proprietary responses were previously submitted by Reference (e). The pages with proprietary. information have been deleted from this submittal. North Atlantic believes that this infonnation will support the completion of the review of YAEC-1856P and YAEC-1849P [ Reference (b)]. Should you have any questions regarding this letter, please contact Mr. Terry L liarpste:, Director of Licensing Services, at (603) 474-9521, extension 2765. Very truly yours, M jet 2.C.: 5 3 ~\\ l Ted C. Feig >aum s t

  • "I

I 9404150111 940404 PDR ADOCK 05000443 P PDR

y.. _ _. ~.. jV , a ct

.r United States Nuclear Regulatory Commission.

April 4,1994 Attentioni Document Control Desk Page two'- a cc with enclosures: Mr. Thomas T. Martin e Regional ~ Administrator U.S. Nuclear Regulatory Commission ~ Region I 475 Allendale Road King of Prussia, PA 19406 Mr. Albert W. De Agazio, Sr. Project Manager Project Directorate 1-4 Division of Reactor Projects U.S. Nuclear Regulatory Commission Washington, DC 20555 Mr. Antone C. Cerne NRC Senior Resident inspector .i : P.O.130x 1149 Seabrook, Nil 03874 i-cc without enclosures: i. Ms. lleidi Komoriya l International Technical Services, Inc. Suite 2545 420 Lexington Avenue New York, NY-10170 N 1 N bi ,u-

< k,

' h,' .se / i North Atlantic April 4,1994 1 I ENCLOSURE I TO NYN-94035 RESPONSE TO REQUEST FOR ADDITIONAL INFORMATION J REVIEW OF YAEC-1849P J NON-PitOPRIETARY VERSION 1 l l

a y. 4 .o FIGURE 3 SEABROOK UPDATED FSAR-INTENTIONALLY LEFT BLANK c. Mixine Technolocy The rate of heat exchange by mixing between flow channels is proportional to the difference in the local mean fluid enthalpy of the respective channels of the local fluid density and flow velocity. The proportionality is expressed by the dimensionless thermal diffusion coefficient (TDC) which is defined as: TDC - v' pVa (4.4-9)- where: v' - flow exchange rate per unit length, (1bdft-sec) 3 fluid density, lbgft p fluid velocity, ft/sec V 4.4-9 ) U i_

FIGURE 3 (Continued) SEABROOK UPDATED FSAR - lateral flow area between channels per unit length, a z ft jft The application of the TDC in the THINC analysis for determining the overall mixing effect or heat exchange rate is presented in 4 Reference 7. As a part of an on6oing research and development program, Westinghouse has sponsored and directed mixing tests at Columbia University (Reference 12). These series of tests, using the *R' mixing vane grid design on 13, 26 and 32 inch grid spacing, were conducted in pressurized water loops at Reynolds numbers similar to that of PWR core under the following single and two phase (subcooled boiling) flow conditions: 1 Pressure 1500 to 2400 psia Inlet temperature 332 to 642*F 8 2 Mass velocity 1.0 to 3.5x10 lb./hr-ft 8 Reynolds number 1.34 to 7.45x10 Bulk outlet quality -52.1 to 13.5% TDC is determined by comparing the THINC Code predictions with the measured subchannel exit temperatures. Data for 26 inch axial grid spacing are presented in Figure 4.4-4 where the thermal. diffusion coefficient is plotted versus the Reynolds number. TDC is found to be independent of Reynolds number, mass velocity, pressure and ) quality over the ranges tested. The two phase data (local, . subcooled boiling) fell within the scatter of the sing e p ase data. l'h The effect of two-phase flow on' the value of TDC has been demonstrated by Cadek (Reference 12), Rowe and Angle (References 13 and 14), and Gonzalez-Santalo and Griffith (Reference 15). In the subcooled boiling region, the values of TDC were indistinguishable from the single phase values. In the quality region, Rowe and Angle show that in the case with rod spacing similar to that in PWR reactor core geometry, the value of TDC increased with quality to a ~ point and then decreased, but never below the single phase value, Gonzalez-Santalo and Griffith showed that the mixing toefficient increased as the void fraction increased. The data from these tests on the 'R' grid showed that a design TDC value of 0.038 (for 26 inch grid spacing) can be used in determining the effect of coolant mixing in the THINC analysis. h A mixing test program similar to the one described above was conducted at Columbia University for the 17x17 geometry and mixing vane grids on 26 inch spacing (Reference 16). The mean value of TDC obtained from these tests was 0.059, and all data was well above the current design value of 0.038. ( 4.4-10 t

FIGURE 3 (Continued) SEABROOK UPDATED FSAR Since the actual reactor grid spacing is approximately 20 inches, i additional margin is available for this design, as the value of TDC increases as grid spacing decreases (Reference 12). 1 4.4.7 References 7. Chelemer, H., Weisman, J. and Tong, L. S., "Subchannel Thermal Analysis of Rod Bundle Cores," WCAP-7015, Revision 1, January 1969. 12. Cadek, F. F., Motley, F. E. and Dominicis, D. P., "Effect of Axial q 1 Spacing on Interchannel Thermal Mixing with the R Mixing Vane Grid," WCAP-7941-P-A (Proprietary), January 1975 and WCAP-7959-A, January 1975. f Rowe, D. S., Angle, C. W., " Crossflow Mixing Between Parallel Flow 13. Channels During Boiling, Part II Measurements of Flow and Enthalpy in Two Parallel Channels," BNWL-371, part 2, December 1967. 14 Rowe, D..S., Angle, C. W., " Crossflow Mixing Between Parallel Flow Channels During Boiling, Part III Effect of Spacers on Mixing Between Two Channels," BNWL-371, part 3, January 1969. 15. Gonzalez-Santalo, J. M. and Griffith, P., "Two-Phase Flow Mixing in 1 i Rod' Bundle Subchannels," ASME Paper 72 WA/NE-19. 16. Motley, F. E., Wenzel A. H., Cadek, F. F., "The Effect of 17x17 Fuel Assembly Geometry on Interchannel Thermal Mixing," WCAP-8298-P-A. (Proprietary), January 1975 and WCAP-8299 A, January 1975. \\ /

y 1 l North Atlantic .j April 4,1994 l b IE 'l ENCLOSIIRE 2 TO NYN-94035 - ) 4 rl RESPONSE TO REQUEST FOR ADDITIONAL INFORMATION REVIEW OF YAEC-1856P NON-PROPRIETARY VERSION j a A s d l S 4

)

RESPONSE TO REQUEST FOR ADDITIONAL INFORMATION REVIEW OF YAEC-1856P 1. Identify clearly and accurately transient analyses performed with RETRAN for which YAEC received NRC approval. YAEC should identify:(1) which version ofRETRAN was used, (2) the scope of application approved, such as reload and transient specificicity, (3) for which plant and (4) references to NRC appmvals. RESPONSE TO QUESTION 1: YAEC has performed transient analysis using RETRAN since the early 1980's. An overview of the history of YAEC's NRC approved applications is pmvided in the attached Table. As noted in the Table, YAEC has approved applications of RETRAN for Main Steam Line Break 4 1 and Loss ofload events for PWR's and the full range of transients for reload applications on BWR's. For the most part these applications have been in conjunction with reload analyses. The Loss of Load analysis application on Seabrook was directed at revising the safety valve setpoint tolerance from 1% to 3%. With the approval of this application this analysis is now the analysis of record for Seabrook Station. YAEC's reload analysis experience was. summarized in Table 2.1 of YAEC-1856P. It should also be noted that the reference numbers given in the attached Table refer to references given in YAEC-1856P. f The Main Steam Line Break (MSLB) applications employed on Maine Yankee and Yankee. Rowe in the 1981,1983, and 1984-85 time frames used RETRAN and a criteria of no-return-to-power to establish shutdown margin requirements. This appmved technology was extended in 1990-1991 in conjunction with the STAR code to treat cases with the potential for a return- - to power. STAR is a 3 D space-time treatment of the reactor kinetics with 3-D thermal-hydraulics (VIPRE). STAR is used in conjunction with RETRAN to predict the overall response to MSLB including the return to power following the extended.cooldown. The thermal hydraulic approach with RETRAN was consistent with previously approved YAEC applications approved in the early to mid 80's. The modeling of the reactor vessel was modified to capturo the effects related to the faulted and intact loops as they impact the core neutronic feedback. The approved Star application also included rod ejection. Both the MSLB and rod ejection application was approved on a generic basis applicable to PWRs. l j As noted in the attached Table, YAEC has applied a number of versions of RETRAN. These applications have coincided with the release / NRC approval of the corresponding version of-the code. YAEC is currently using RETRAN02 MOD 05 in all ofits applications. The options selected for each application are consistent with the original NRC approval. The upgraded - versions have been tested through our quality assurance process to assure that results obtnined with the later versions are consistent with the approved applications.

OVERVIEW OF YAEC LICENSING HISTORY RETRAN NRC PLANT TRANSIENT SCOPE VERSION SUBMITTAL APPROVAL 1981 Maine Yankee MSLB Reload 01 MOD 3 Ref.9 Ref.11 1981 Vermont Yankee BWR reload Reload 15F YAEC-1233 NRC SER Nov.1981 transients 1983 Yankee Rowe MSLB Reload 02 MOD 2 Ref.8 Ref.10 1984-85 Maine Yankee MSLB Reload 02 MOD 2 YAEC-1447 NRC SER Oct.1985 1989 Vermont Yankee BWR transients Generic 02 MOD 04 Ref.12 Ref.13 1-D kinetics 1990-91 Maine Yankee MSLB/ rod Generic 02 MOD 05 Ref.15 Ref.14 ejection STAR / CHIC-KIN 1991 Seabrook SGTR Address 02 rdOD 02 Ref.16 licensing issue 1992 Seabrook Loss of Load Safety valva 02 MOD 05 Ref.17 Ref.18 RETRAN.OHDI2 - ----- a a.

t 50 7 With local Conditions Heat Transfer - Without Local Conditions Heat Transfer \\ 40 E 30 ~o is a 1 8 20 1 10 . l . " " " = 0 0 250 500 750 1000 1250 1500 1750 2000 TIME (SECONDS) I I Loss of Feedwater-SEABROOK STATION Sensitivity of RETRAN's SG Mixture Level to RETRAN SENSITIVITY STUDY the Use of Local Conditions Heat Transfer FIGURE 3.1

) 4. On page A.10, YAEC briefly described its special technique to preserve the pressure drop when specifying geometric data. Discuss how this method also preserves the volume so that the liquid inventory is accurately predicted. \\ J RESPONSE TO QUESTION 4: The special technique is applied to nodes that are not simple pipes, so that the pressure drop i due to wall friction within the node is correctly represented. Since the wall friction pressure drop is calculated from FLOWA, FLOWL, and DIAMV, the total volume, specified by V does not affect the pressure drop calculation. The actual value of the fluid volume is independently specified. 5. Specify the gap conductance assumed for transient analysis. If values vary for different transients, discuss how they are determined. RESPONSE TO QUESTION 5: The gap conductance assumed for transient analysis is determined from an approved fuel performance code. The gap conductance is developed to pmvide a consistent thermal resistance through the fuel pellet gap and clad to establish fuel average temperatures in RETRAN for calculating Doppler feedback during the transient. For the RETRAN plant response analyses, the gap conductance is selected to provide conservative Doppler reactivity feedback for events where Doppler reactivity plays an important role. 6. Explain andjustify the upper head circulation path modeling and demonstrate that predicted flows are realistic andfor conservative on a transient by tmnsient. RESPONSE TO QUESTION 6: The upper head bypass flow in the RETRAN model is conservatively low based on NSSS vendor data. The RETRAN upper head volume is, therefore, assumed to be relatively stagnant mducing its availability to attenuate the rate of heatup and cooldown during Chapter 15 type events. For Seabrook Station the Reactor Vessel (RV) internals have flow passages which by design allow cold coolant from the inlet nozzles to bypass the core and pass directly to the upper head. Therefore, initial coolant temperature in the upper head is equal to Ta. The upper head modeling has an insignificantimpact on Chapter 15 transients except for some primary side depressurization events, e.g., MSLB. Refer to Appendix A of these RAI responses for additional discussion. ] 4 l i i i

7. YAEC should provide FSAR benchmark analysis for the full spectrum of transients before its methodology can be reviewed for generic application. Comparison should be made for all 5 categories of transients with focus upon three categories (loss of heat sink, overcooling, reactivity anomalies) including loss of normal feedwater, loss ofACpower and feedline break for Category 1, locked rotor and loss of flow for Category 2, rod ejection, rod withdrawal at different powers for Category 3. All such comparison must be thoroughly discussed and differences explained andjustified. RESPONSE TO QUESTION 7: The YAEC RETRAN analysis methodology described in YAEC-1856P has been applied to Seabrook Station and results are presented in YAEC-1871 which has been submitted to the NRC. In Appendix A to these responses to RAIs, comparisons of the YAEC RETRAN analysis methodology results to the Seabrook Station UFSAR results are presented. For some cases, results from YAEC. 1871 are compared directly to the UFSAR. Differences are explained andjustified. For other cases, YAEC has performed additional RETRAN analyses not shown in YAEC-1871 to facilitate the comparison by eliminating input parameter differences inherent in the YAEC-1871 analyses versus the UFSAR. l 8. Provide tables containing YAEC's approach toperformance ofChapter 15 non-LOCA tmnsient analysis for each plant. One table should contain ranges of key plant parameters and the other should contain (1) SG nodalization, (2) core nodalization, (3) reactivity coefficients, (4) key plant parameters and their status (highilow or min / max), (5) transient assumptions and (6) operable and inoperable components. RESPONSE TO QUESTION 8: This information is provided in Appendix B to these responses to RAIs on a transient by transient basis. The illustration provided in Appendix B is specifically for Seabrook Station. In general, the key plant parameters and their status would be similar for any PWR, e.g., Maine Yankee. Specific design features or lack thereof may yield minor differences plant to plant, e.g., automatic rod control. e

TABLE 5.0-3 TRIP POINTS AND TIME DELAYS TO TRIP ASSUMED IN ACCIDENT ANALYSES Limiting Trip Trip Point Assumed Time Delays Function _ In Analysis (Seconds) Power Range High Neutron Flux, g-High Setting 118 % 0.5 Power Range High Neutron Flux, Low Setting 35% 0.5 mj. High Neutron Flux, P-8 50% 0.5 l Overtemperature AT Variable 6.0 Overpower AT Variable 6.0 High pressurizer pressure 2425 psia 2.0 Low pressurizer pressure 1935 psia 2.0 Low reactor coolant flow (from loop flow detectors) 87% loop flow 1,0 Undervoltage Tdp 70% nominal 1.5 Turbine Trip Not applicable 1.0 Low low steam generator level 0% of narrow range level span ** 2.0 High steam generator level 94% of narrow range trip of the feedwater pumps level span 2.0 and closure of feedwater system valves, and turbine trip SafetyInjection Actuation 1665 psia 30.0 Total time delay (including RTD time response and trip circuit channel electronics delay) from the time the temperature difference in the coolant loops exceeds the trip setpoint until the rods are free to fall,

    • Zero percent of the narrow range level span is the limiting trip point based on the feedwater system pipe break analysis.

All other analyses assume a trip point l corresponding to 10 percent of narrow range span. 5.0-8 I

-c. TABLE 9.1 I ' ENGINEERED SAFETY FEATURES AND CODE SAFETY VALVE SETPOINTS j Actuation Setooint Assumed in the Time Delay j Engineered Safety Feature Safety Analyses (seconds) j Safety Injection Pressurizer Pressure = 1665 psia 30' OR Main Steam Pressure = 430 psia Emergency Feedwater SG level = 0%* 75 Main Steam Line Isolation Main Steam Pressure = 430 psia. 2

  • Zero percent of the narrow range level span is the setpoint assumed in the feedwater system pipe break analysis. All other analyses assume a setpoint corresponding to 10 pezrent of narrow rango span.

i l Opening Fully Open Fully Closed Safety Valve Pressure (osia) Pressure (osia) Pressure (osia) Pressurizer Safety Valves 2575 2652.3 2475 MSSV #1 1236 1273.1 1161 l MSSV #2 1254.5 1292.1 1179.5 MSSV #3 1272.1 1310.3 1197.1 MSSV #4 1290.6 1329.3 1215.6 i MSSV #5 1308.1 1847.3 1233.1-l 4 ' ) 4

1 n o 11. Explain why RTN initial conditions were not matched with the data for natural. 1 circulation test for Seabmok. Justify the statement that the difference in hot leg temperatures between 100 and 300 seconds while the cold leg temperature were matched with an offeet in the initial conditions was caused by the modeling of steam dump control system. Discuss the source (s)of the hot leg temperatures. Does YAEC have similar analysis for other plantsiin onier to assure that the steam dump contml system modeling is the source of mismatch (as YAECpostulates) in some benchmark analyses, provide the description of the simple steam dump model. RESPONSE TO QUESTION 11: Our primary focus and concern in the benchmark of the Seabrook natural circulation test was the final conditions reached following the Pipping of reactor coolant pumps (RCPs). The matching of final conditions is a strong iadicatien that the hydraulic modeling of the plant is an accurate representation of specific plant conditions for transients where power to the RCPs is lost. The model used in the test benchmark was initialized based on data (primary to secondary) at power. The test were run at zero power. The difference of approximately 2 degrees in initial average temperature between the RETRAN simulation and the data is attributed to this difference. Your observation that the difference in hot leg response could not be attributed to the steam dump controller because the cold leg tesponse agreed is correct. YAEC was misled in our postulation because of our focus on the final conditions and because of our understanding that the plant had some difficulty with the steam dump controller during the test. After evaluating the differences in hot leg responses further, and looking at similar analysis for Yankee Rowe, we have concluded that the differences are attributed to differences in RTD response times and the specific location of the hot leg RTD used in the Seabrook test. The attached Figure shows the results of a similar analysis performed for Yankee Rowe. This analysis was of an actual event where one of two offsite power lines were lost and the plant was manually tripped by the operators. The operators also subsequently tripped the remaining RCPs placing the plant in natural circulation. The hot and cold leg responses, as shown in the attached Figure, are in good agreement with RETRAN predictions. The RTDs used to measure the responses at Yankee Rowe are mounted in thermo-wells in the loops and have relatively fast response times. At Seabrook the hot and cold leg RTDs consist of both narrow range and wide range. The narrow range RTDs, which feed into the reactor protection system, have a relatively fast response time similar to thore at Yankee Rowe. The narrow range RTDs, at the time of the natural circulation test, were located in the loop bypass manifold and were unreliable during natural circulation conditions. The temperature measurements for the hot and cold leg responses for the natural circulation test were thus taken from the wide range channels. The wide range RTDs are located in the loops as opposed to the loop bypass manifold. Although the response times for the wide range channels have not been measured they are known to be slower than the response time of the narrow range channels (when the narmw range RTDs were located in the loop bypass manifold the response time was about 6 seconds with the RCPs operating). Under natural circulation conditions it is plausible for the wida range RTDs to have a response time on the order of 60 seconds. In addition, the location of the hot leg wide range RTD is closer to the steam generator than assumed in the RETRAN simulation accounting for part of the difference noted. This contributed approximately 40 seconds to the delay in the hot leg

4 temperature response. Therefore, the difference noted in hot leg response in Figure 4.1.4 of 'I YAEC-1856P is attributed in part to the slow response time of the wide range RTDs and in i part to the assumed location of the hot leg RTD in the RETRAN simulation. ) l I i I I 1 l ) H i \\ 1 a

i l .e-YR RETRAN SIMULATION LOSS OF Z-126 HIGH LINE EVENT (LOSS OF OFF-SITE POWER) 570 C l ,[g i d pata Tg '~ 'O - ~ s \\ l l O T-c g j O T-h tu @ 530 - l

  • -^5'* '

"s-fiarrt 'y dotgT T 510 n_o 9 490 470 0 120 240 360 ~480 600 720 840 960 TIME (SECONDS) 4 m h m. m m. m 2 m..- a W

W' A

12. What is the timing of the operator action during the reactor trip test? For the Maine Yankee reactor trip test benchmark analysis, the pressurizerpressure was predicted to be about 300 psi higher than the plant data by the end ofsimulation. Explain in depth the cause of this' difference since the pressurizer level was well matched. RESPONSE TO QUESTION 12: The reactor trip data used in the Maine Yankee benchmark was based on an event that occurred on April 29,1991 involving a switchyard failure resulting in a turbine trip reactor-trip sequence. There were t.o speciSc known operator actions taken in the time frame of interest. The higher pressurizer pressure predicted by RETRAN at the end of the simulation is attributed to the response of pressurizer heaters and interfacial heat transfer in the pressurizer. As indicated in Section 3.1.1 ofYAEC-1856P YAEC assumes a conservatively low interfacial heat transfer coefficient for preasurizer insurge transients. Assuming a low heat transfer coeflicient results in an overprediction of the pressure response. A conservatively low interfacial heat transfer cocilicient was assumed in the reactor trip test comparison. We believe that the sensitivity to interfacial heat transfer in this test was weak and not the mqjor contribution to pressure differences noted. Based on sensitivity analysis the difference noted is primarily due to difference in assumptions regarding pressurizer heaters. For the RETRAN analysis, the pressurizer backup heaters mmained on throughout the simulation. In the actual event, the backup heaters were de-energized when the pressurizer level dropped below 28% This occurs within seconda of the reactor trip. Terminating the heaters on low level in the RETRAN model results in a consistent pressurizer pressure response. 13. Provide a loss ofload analysis for Maine Yankee. YAEC demonstrated that for Yankee Rowe, GEMINI predicted higher pressure than RETRAN at slightly earlier time. Explain the reasons for the divergent results ak? the pressur: peaked. RESPONSE TO QUESTION 13: The intent of the Yankee Rowe loss ofload comparison presented in Figure 4.2.2 of YAEC-1856P was to provide an indirect comparison of YAEC's pressurizer modeling technique (treatment ofinterfacial heat transfer etc.) for overpressure transients against data. Figure 4.2.1 of YAEC 1856P presented results of a comparison of GEMINI predictions to data from Shippingport for a rapid insurge transient. GEMINI is YAEC's currently approved transient analysis model for these types of events. The comparison shows that the GEMINI prediction is conservative relative to the data. Comparison of the RETRAN response to GEMINI provides an indirect comparison of the RETRAN response to data. Peak pressures agree well but diverge following reactor trip. The divergence is attributed to differences in the post-trip

.i heat input to the Reactor Coolant System (RCS) from the following sources: 'l 1. Decay Heat - The decay heat in GEMINI is greater than that in RETRAN. RETRAN uses the 1973 ANS standard. The attached figure compares the GEMINI decay heat model to the 1971 ANS standard. 2. Delayed Fissions The heat from delayed fissions following the trip is greater in GEMINI than in RETRAN due to the different delayed neutron fractions used. 3. Stored Enercy - The initial stored energy in the fuel used in GEMINI is conservatively larger than that in the RETRAN simulation. 4 j

- GEMINI-II DECAY !! EAT FU."CTION GEMINI-Il

ANS Standard 12 4..;JJ L.._: g._... !

i .1 _:.. L } g.t p. -.. t _ y.. _.-.;. 7 q_J' .. H. ' . _._.l _- f..-+g ._ y _,,..;_'p ._ a r - r I1 ... _p. 4_. ' L.a. 47 ...w q. q g.. i - b, '_ r-y. ..q. _i ,gy. ,,.,_ w_,6 q1

1 i i i

- i _,,;, J _ ','.i g I 8 i I ii!3' e 10 -.i _M. ._._i i. i i 1 .i J. i.. -.,. i, i ei. .. _i. . _,.p i i i i

  • i:

.._..a___ t i 6 i i e g g s. .T e e i 6 e !6 y _ _ _1_. l; .a. L _.. _. a _.p_._.

. i e i i a

.s... l. ~_. o, .L .h -h'.j.. ! J_. 4 ."a 8 .. J _N .4 - : _; l,: . g .j._.K_;...-g. i. L ...u ! = g :.p;.._. _7 3 s ,t , i, .g a. a y _ _ _.. 4. +, g ,- ; : g 4 g-g' -.4 -r *h _-._-- _ a.,. ;,

; p~

a 'r t g l4 Q._,_--- r..__ __ a... 1 r4

2..'..,1

;i -9 ii-

a..

w 7

- 4'j

.y 4 1.. l-Q.J_LLLL 6 c- _.i_y q .1 4 4 ..,a. _.= i:. '4 p I 'I i e 5 .'N T L... a W Z l 14a. ' h.s S..h..J._) , U,. i . My, -a h rt x -F * * - ! I l 7_. c_ -. -+,- -.. i L -- -hn[. i m I*5 J.. irti. _p. :_ L_.,._ o i 1 .ts__ i .r,i-H,, --*-I-4l, 'x ' N_' _.- .-.- H. - +,. '. ' _. ". ' '. ' ' i!: q _____..._..s -d' . ^I' '- -. - ' ~ . _ _ + _ _.. 3 i .N _4 - - -7p.... J_ p._ .a..1,.. s . _y ____ r y 2 _. i N i i_._._> 4_

.. 2.

3 _. .q_-.... 1 4 .__. _ i.. _4 -) .,. 3. 2 _Q.. i .. _..y...p y, _. y 3 s.h w_L.,. __ _1 ,.C.... L.....,. s _. +, a- ..,..a i., ,.._ r 3 _.;_._L_

i ' i i p.

m ~.. k _~ .. 7 .i_-_4>.. .__ t.,.y . + _~_ h7_.- i -..;,; -_L. _,. q _,-....,_..... : - P.... l - L{ y, ~ g 7, '~t+ ~t -" -t , i i i;i -T m"* i riii,, O 100 101 102 103 104 - 105 TIME, SECONDS + a TW-

~ RETRAN Boron Transport vs..BIRP -50 1 4 A u%+.,...%.. by - -+ ' 4% .e-m...~we e-+-~.e.-.- -..~~4..-......r4,-_.-.,. -E vot ee i o_ Igru.r ca a. ~ C .O 30- -R t- - - as .:= / ce j O i i c RETAAN --- - - - --- r o 20-O - # g ig,p + VOL 11-c - c en o CoR.E.. '/ o c m 10- --~; - -, ~ 1 . ( / 1 l x/ - s

  1. x,

VOL46 '1 0 = = =! i 100 110 120 130 140 150 N N c'- 4 j Time (sec) ] i i y I =

APPENDIX A COMPARISONS TO SEABROOK STATION UFSAR INTRODUCTION In this appendix, comparisons of the YAEC RETRAN analysis methodology results to the Seabrook Station UFSAR results are presented. For some cases, results from YAEC-1871 are compared directly to the UFSAR. Differences are explained and justified. For other cases, YAEC has performed additional RETRAN analyses not shown in YAEC-1871 to facilitate the comparison by eliminating input parameter differences inherent in the YAEC-1871 analyses versus the UFSAR. Table A 0.1 summarizes initial conditions and uncertainties used in the RETRAN and UFSAR analyses. The transient analyses presented in the Seatrook Station UFSAR and those performed by Yankee and documented in YAEC-1871 used diffet '.it scram reactivity curves. Figure A-0.1 shows the normalized reactivity worth versus rod position scram data used in both sets of analyses. The figure shows that the Yankee scram data is more conservative than the UFSAR data. The Yankee data was chosen to bound more adverse power distributions which may result from transients when initiated from within Yankee's Wide Band Operation Strategy. Figure A-0.2 shows the comparison of the normalized worth versus time data.The UFSAR analvses assumed a scram time (from release to dashpot) of 3.3 seconds for all transients except the LOFA which assumed a 2.2 second time. The Yankee analyses assumed a scram time of 2.4 seconds for all transients and is consistent with our proposed Technical Specifications.

APPENDIX A (Continued) TABLE A-0.1 l INITIAL CONDITIONS AND UNCERTAINTIES USED IN THE RETRAN AND UFSAR ANALYSES Parametern RETRAN UFSAR l Core thermal power: Rated, MWT 3411 3411 Engineered Safety Features design rating, MWT Not used 3565-Uncertainty, % 2.0") 2.0 i Nominal RCS loop average temperature @ 100% RTP, 'F 588.5 588.5 - Uncertainty, "F 5.8") 5.8 Pressurizer Pressure, psia 2250 2250 - Uncertainty, psi 50") 30 Main Steam Pressure @ 100% RTP, psia 1000 1000 Main Steam Flow Rate @ 100% RTP,10 lb./hr 15.14 15.14 8 RCS Thermal Design Flow, gpm 375,100 ) 382,800 5 RCS Minimum Measured Flow"), gpm 392,800") N1unl . Notes: (1) 1.8"F is assumed to be a bias on T, to allow for the effects of SG tube fouling and flow streaming. This bias is applied in all RETRAN analyses. The remaining 4*F is a measurement uncertainty and is not applied when RETRAN is used for the analysis of DNB related events. (2) These values of RCS flow include a 2% penalty to allow for SG tube plugging. (3) The minimum measured flow is used in the analysis of DNB related events. (4) As discussed in the Introduction of Appendix B, these uncertainties are not applied when - RETRAN is used for the analysis of DNB related events. 1 I o

FIGURE A 0.1 SEABROOK STATION SCRAM CURVE COMPARISON Normalized Worth Versus Position 1.0 l 0.9 - .I y 0.8 - l S> 0.7 - l w / 0.6-U [ - c-- YAEC 0.o_ - / --a-- UFS AR 9 d = 0.4 - l W N l 3 03 - l 2 0.2 - / cZ 0.1 T - ~ 7 ~~ Y~.c' O.0 m: r 1 i i 5 0.0 0.1 0.2 03 0.4 0.5 0.6 0.7 0.8 0.9 1.0 ROD POSITION (FRACTIONINSERTED)

i-a FIGURE A-0.2 SEABROOK STATION-SCRAM CURVE COMPARISON Normalized Worth Versus Time 1.0 ,i 0.9 - p' E / 0.8 - / j W / A 0.7- .F F ,/ 3 0.6-HO d -o -YAEC # 02 - - *-- UFS AR $ W M l a C 0.4 - d w s3 03 - l d S 0.2 - l o 0.1-g Z 0.0 r *PC--i= - F" i - i i i i o 0.0 0.5. 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 TIME FROM RELEASE (SECONDS) q

  1. = 2.4 Seconds to Dashpot

$ = 3.3 Seconds to Dashpot t

APPENDIX A (Continued) 1. MAIN STEAM LINE BREAK (MSLB) YAEC's approved methodology for MSLB analysis is documented in YAEC-1752A. A comparison of RETRAN results with the Seabrook Station UFSAR results was pmvided in that report. The comparison is also reproduced here in Figures A 1.1 through Al.5. The sequence of events is shown in Table A 1. It should be noted that the RETRAN analysis dccumented in YAEC 1752A has differences in input parameters and assumptions from the analysis documented in YAEC-1871. For example,in YAEC-1871 the Chapter 15 RETRAN analysis of Seabmok Station is performed using a statistical combination of uncertainties while a deterministic treatment was used in YAEC-1752A. Core Power The RETRAN power response is more conservative than the UFSAR. The RETRAN results are based on the approved methodology using STAR. The reactivity weighting coefficients developed in YAEC-1752A and used in the benchmarks to the UFSAR were derived to produce a conservative power response relative to STAR. The overprediction of power as compared to the UFSAR is consistent with the conservative derivation of the reactivity weighting. Figure A 1.5 shows the various components of reactivity. As shown in the figure, the total reactivity dmps to zero when baron reaches the core, causing the core power to decrease. Dryout of the faulted SG occurs beyond this point. As noted above, the MSLB response from the YAEC-1752A UFSAR comparison presented here differs from the response presented in YAEC 1871. In the YAEC-1752A UFSAR comparison the steam piping resistance from the intact SGs to the break was neglected to be consistent with UFSAR assumptions. In the YAEC-1871 analysis the steam piping is explicitly modelled consistent with YAEC's approved MSLB methods. This changes the relative timing of when boron reaches the core and SG dryout. RV Inlet Temperatures The RETRAN predictions for both the faulted and intact reactor vessel inlet temperatures are in good agreement with the UFSAR. Small differences are attributed to differences in the break flow. l RCS Pressure The RETRAN modelincludes a non-equilibrium volume in the upper head region. Non-equilibrium effects in the upper head will result in the upper head behaving as a second pmssurizer yielding higher RCS pressure as RCS inventory recovers by the action of safety injection. The pressure l response from the UFSAR appears to follow the saturation curve corresponding to the intact loop temperatures. It appears that the UFSAR analysis neglected the non-equilibrium effects in the upper head. The overall effect of higher RCS pressure is to cause a reduced injection of boron and a higher core power. Break Flow The RETRAN and UFSAR results for break flow show good agreement. In the RETRAN analysis, the choked flow model based on Moody was used to compute the break flow rate through the SG outlet nozzle flow restrictor (1.4 ft*). In addition, the break flow was assumed to be dry steam.

APPENDIX A (Continu d) TABLE A-1 SEQUENCE OF EVENTS FOR THE MAIN STEAM LINE BREAK (MSLB) i Time (Seconds) Event RETRAN UFSAR l-Steam line ruptures 0.0 0.0 Pressurizer empty 13.5 13.6 Criticality attained 14. 12.4 Boron reaches core - 120. -125. l 4 1

e 60 o FSAA - nETnAu <0 30 Ze f' a o o o o o o 10 0 o 0 2 0 M 50 M W IM N THE (sea.W40S) = E W N o FSAR g550- _ nEvam 5 o N "" g e s N, W o o gm o o ti $.m B oEm }. a0 0 25 50 M 100 125 150 TME (SECONOS) Steam Line Rupture SEABROOK STATION (Offsite Power Available) COMPARISON OF RETRAN vs. UFSAR FIGURE A-1,1

i = i g 5m 's o FSAR .- nEmAN a o o o I am o o 5 O o 450 d ay 4m g ti 6 'E sso U M N sm a eso 0 25 50 75 100 12$ tilo TIME (SECONDS) 25m i o rsAR - REMAN emo -- o [ ism - I E \\m 1000 - g o O C o o sm a so n im 125 im T1WE (SECONDS) Steam Line Rupture SEABROOK STATION (Offsite Power Available) COMPARISON OF RETRAN vs. UFSAR FIGURE A-1.2

1200 0 FSAA 1000 - AETRAN l', E.g a-N Wu O y o ag* s Y' o o o o 0 0 25 60 75 100 125 16o TW4 (SECONOS) i2ao .I O FSAR ioco - neinAn i. k ,heco ~ L tr O <l o w ^ ^ 't 1 soo - h S 1 too f 6 0 25 50 75 100 125 150 TMi (SECONOS) Steam Line Rupture SEABROOK STATION (Offsite Power Available) COMPARISON OF RETRAN vs. UFSAR FIGURE A-1.3

e M00 ') o FSAR g ~ HETDAN t esco : F $ 20m o j 1sco - E o icoo.. t_ O d' O <l O i, 0 0 25 60 M 100 125 180 TIME (SEN) Steam Line Rupture SEABROOK STATION (Offsite Power Available) COMPARISON OF RETRAN vs. UFSAR FIGURE A-1.4

4 3 / 2 o Moderator 0 Dog:per 1 a mmn g ~ - Shutdown Fkid / "[0' ~ j x ~ r, 7 b l a: 1 ll l -2 l l " 4 0 25 50 75 100 125 150 TIME (SECONOS) 13000 12000 ^ ^ p. "*o j 10000 0000 o FautedloopVesselWet 8000 0 Csoes Fbw @ Lower Pwnum g a Csees flow @ Upper Punum uf* 7000 $ 6000 2 x 50* Oda. 3000 2000 1000-f O a L 1000 0 25 50 75 100 125 150 TIME (SCCONOS) 1 Steam Une Rupture l SEABROOK STATION (Offsite Power Available) j COMPARISON OF RETRAN vs. UFSAR i FIGURE A 1.5 i

APPENDLX A (Continued) 2. TURBINE TRIP (WITHOUT PRESSURIZER PRESSURE CONTROL, MINIMUM REACTIVITY FEEDBACK) A comparison of the RETRAN results versus the Seabrook Station UFSAR results is shown in Figures A 2.1 and A.2.2. The sequence of events is shown in Tablo A 2. The case presented (without pressurizer pressure control, minimum reactivity feedback) results in the greatest primary and secondary system pressures of the four cases presented in the UFSAR. The RETRAN results are taken from a UFSAR benchmark that was performed to verify the Seabrook Station RETRAN model(relative to the prediction of eak pressures) prior to evaluating an increase in the pressurizer p and main steam safety valve setpoint tolerances from *1% to 3E Normalized Core Power The core powers match very well. RETRAN results in a slightly later (about 1 second) reactor trip on high pressurizer pressure that is attributed to small differences in the pressurizer modelling between the RETRAN and UFSAR analyses and the initial pressure assumed. Pressurizer Pressure The peak pressurizer pressures match almost exactly between RETRAN and the UFSAR. Due to the slightly later trip in the RETRAN analysis, the time of the peak pressure is delayed about I second from the UFSAR analysis. The decrease in the RETRAN pressure below the UFSAR pressure is due to the blowdown that is included in the RETRAN MSSV model. The MSSV blowdown results in a lower SG pressure, which causes a lower RCS temperature and a lower pressurizer pressuro. Pressurizer Water Volume The peak pressurizer water volume in the RETRAN analysis is greater than that in the UFSAR analysis due to the slight delay in the reactor trip and possibly to small differences in the pressurizer modelling. The decrease in the RETRAN water volume below that in the UFSAR is due to the blowdown that is included in the RETRAN MSSV model. The MSSV blowdown results in a lower SG pressure, which causes a lower RCS temperature and a lower pressurizer water volume. Core Average Temcorature Accounting for the slight delay in the RETRAN analysis reactor trip, the core average temperature matches well with the UFSAR analysis. The decrease in the RETRAN temperature below the UFSAR temperature is due to the blowdown that is included in the RETRAN MSSV model. The MSSV blowdown results in a lower SG pressure which causes a lower RCS temperature.

APPENDIX A (Continued) TABLE A 2 SEQUENCE OF EVENTS FOR THE TURBINE TRIP -) Time (Seconds) Event RETRAN UFSAR i Turbine trip, loss of main feedwater flow 0.0 0.0 Initiation of steam release from steam generator safety valves 7.5 5.5 High pressurizer pressure reactor trip point reached 6.2 5.1 Rods begin to drop 8.2 7.1 Peak pressurizer pressure occurs 10.0 9.0

1.2 .. RETRAN UFSAR w i i

n. 0.8 -

o ,} e g E .6 0 a h0.4 Z e .'y 0.2 0.0 2600 A 2500 - neinAN .. uFsAs 3 2400 a / it 2 2300 i / \\

- 2200 4

5 \\ E 2100 3 U 2000 w 1900 1800 0 20 40 60 80 100 TIME (SECONOS) Turbine Trip SEABROOK STATION Without Pressurizer Pressure Control COMPARISON OF RETRAN vs UFSAR Minimum Reactivity Feedback FIGURE A-2.1

1300 's 's - RETRAN ,,1200 .. ursAR M 6 's 1100 E ,,N j t 1000 eu !!ll 900 g \\ \\ S 800 U iC 700 m N 000 620 - RETRAN - - UF SAR jSoo s e h 5,, i 4 580 N b o i 0) 3560 a 540 0 20 40 60 80 '100 ,l TIME (SECONDS) Turbine Trip SEABROOK STATION Without Pressurizer Pressure Control COMPARISON OF RETRAN vs. UFSAR Minimum Reactivity Feedback FIGURE A-2.2

APPENDIX A (Continued) Normalized Core Flow Except for the earlier trip of the RCPs, the RETRAN core flow coastdown agrees very well with the UFSAR. The flow coastdown begins earlier due to the earlier reactor trip in the RETRAN analysis since the RCPs are assumed to trip at the time of the reactor trip. Core Average Temperature The peak temperatures match very well. The final temperature is lower m the RETRAN analysis due to the blowdown included in the RETRAN MSSV model. The MSSV blowdown results in a lower SG pressure which causes a lower RCS temperature. SG Pressure The initial SG pressure is lower in the RETRAN analysis. The lower initial pressure conservatively delays the opening of the MSSVs and the corresponding RCS heat removal, minimizing the DNBR and maximizing the peak RCS pressure. The peak SG pressure is greater in the RETRAN analysis due to the greater setpoint tolerance assumed (3% in the RETRAN analysis vs.1% in the UFSAR analysis). The SG pressure in the RETRAN analysis falls below that in the UFSAR analysis due to the MSSV blowdown included in the RETRAN MSSV model. SG Water Volume The initial decrease in the SG water volume matches very well. The RETRAN analysis has a greater minimum volume and a faster increase in volume for several reasons: The initial RETRAN power level is lower than assumed in the UFSAR (3411 vs. 3565 MWT), with a correspondingly lower decay heat. The reactor trip occurs approximately 5 seconds sooner in the RETRAN analysis. The RETRAN analysis uses the EFW pump head flow curve adjusted to give the minimum flow required by Technien1 Specifications, while the UFSAR analysis uses a constant 650 GPM, the minimum EFW flow at 1221 psig. The RETRAN analysis uses an EFW temperature of 100 'F, as opposed to the 120 'F EFW temperature assumed in the UFSAR analysis. SG Mixture Level Figure A 3.5 shows the RETRAN mixture levelin the tube bundle region of the steam generators. The dotted line in the figure represents the top of the SG U tubes. The SG tubes are uncovered at approximately 65 seconds. Direct comparison to the UFSAR mixture levelis not possible since the 'UFSAR does not include a plot of SG level. Total EFW Flow The RETRAN analysis has a greater Emergency Feedwater (EFW) flow to the SGs since the flow varies with SG pressure, while the UFSAR analysis assumes a constant flow of 650 GPM (the minimum flow at 1221 psig). EFW flow in the RETRAN analysis is based on the actual head flow curve of the EFW pumps, which has been adjusted to give a flow of 650 GPM at 1221 psig. The difference in integrated EFW flow is approximately 300 ft over the first 2000 seconds of the 8 transient. i I

APPENDIX A (Continued) TABLE A-3 SEQUENCE OF EVENTS FOR TIIE LOSS OF NON EMERGENCY AC POWER Time (Seconds) Event RETRAN UFSAR AC power is lost 0.0 0.0 Main feedwater flow stops 0.0 0.0 Low low steam generator water level trip 36.3 41.1 Rods begin to drop 38.3 43.1 Reactor coolant pumps begin to coast down 38.3 43.1 Peak water level in pressurizer occurs 43 46.5 Four steam generators begin to receive emergency feed from one motor-driven emergency feed water pump 120 112.0 Minimum steam generator inventory 490 1892 l J i

/ s l 1.0 ~ _ nonan . - UrsAn i j0.8 o 8 (L e 8 8 3 0.6 u s. g0.4 e c i, 0.2 Q w 0.0 0 10 20 30 40 50 60 70 80 90 100 TIME (SECONDS) 2600 g 2500 - FV.ifnAN ~ .. UrsAn a 1 8 {2400 2 2300 [ S e ~ [2200 / ( ,/ \\, b -]2100 m

y m

n,,

  • 2000 ct i

1900 1800 0 500 1000 1500 2000 2500 3000 3500 4000 4500 TIME (SECONDS) 1 Loss of Non-Emergency AC Power SEABROOK STATION COMPARISON OF RETRAN vs. UFSAR FIGURE A 3.1

i 2000 l - RETRAN - - UF SAR 9_ 1750 $ 1500 m3 g 1250 .g ..----~----.. g ( g 1000 750 0 500 1000 1500 2000 2500 3000 3500 4000 4500 TIME (SECONDS) 1.25 - RETRAN - - UFSAR 1.0 \\ s 'S 1 R i. u. l 9 0.75 8 E E b .5 0 a ~.', 0'25 0.0 0 50 100 150 200 TIME (SECONDS) ) .l +, I Loss of Non-Emergency AC Power SEABROOK STATIOi COMPARISON OF RL,

1 vs. UFSAR FIGURE A-3.2 i

l

640 l - REntAN 030 -- ursAR c O B 620 E hG10 ( 3 1 i 1 I 8 000 i g ~~. R 590 5 ~ ~~,- y [, w <c 580 570 0 500 1000 1500 2000 2500 3000 3500 4000 4500 TIME (SECONDS) 1300 1250 \\ 1200 N .e g ! 1150

  • /

- nEntAN

  • - UFSAR 1100

) 5

  • 1050

/ / 1000 950 0 10 20 30 40 50 60 70 80 90 100 l TIME (SECONDS) i 1 Loss of Non-Emergency AC Power SEABROOK STATION l COMPARISON OF RETRAN vs. UFSAR FIGURE A 3.3 '{

s 2000 - FIETRAN - - UFSAR 1500 g 0 b S 1000 b N-O W 500 's 4 ~~ ~ " ~ " ~ ~ " ~ ~ ' '"""~"~- " ~ " ~ ~ " ~ ~ ~ 0 0 1000 2000 3000 4000 5000 TIME (SECONDS) 9 Y^ U" SEABROOK STATION COMPARISON OF RETRAN vs. UFSAR FIGURE A 3.4

4' 6 50 l l - RETRAN Mape W Top or so Tue g E -e> 30 3 eau 3 20 e u) ~ 10 0 800 g 600 - RETRAN h UFSAR it S

u. 400 b

w ~m 200 0 0 250 500 750 1000 1250 1500 1750 2000 TIME (SECONDS)- Loss of Non Emergency AC Power SEABROOK STATION COMPARISON OF RETRAN vs. UFSAR FIGURE A-3.5

7 ~. 0, w5CE,^9 a 8 E E S S AA HH P P o 1 2 7 I GG NN I I CC AA o P P

D S S N

6 oo '6 6' I E 2 2 G E OO L 5 I ) 5 -0 1 ( eD P 4 I e = e R 3 I o P 2 I iJ i 1 O 2 0 8 s 4 2 o 1 1 0 o o 0 o 0 0 0 a a 0 a 8r gn; f3vFf, ,EE9-@o t Mh]xU"d9 $55 ggE ~ 2s " m * "sG Ah t 3S m P# a

r 99 i 14 Justify the use of the gap conductance indicated on Table B.1.3 RESPONSE TO QUESTION 14: The gap conductance values used with our approved and VIPRE based methodologies are determined by the approved FROSSTEY 2 fuel performance code. The FROSSTEY 2 code provides gap conductanco values for a rango of KW/FT and fuel burnup. The gap conductance chosen for a particular transient is selected to bound the FROSSTEY 2 results over the rango of intorest. In gonn.s!, when the transient criteria is DN!3R a high gap conductanco is selected to maximizo the surface heat flux. When the criteria is fuel temperature a low gap conductance is selected to retain the heat generated in the fuel pellet. 15 Explain and justify how the RTDP is determined for Seabrook with the old methodology and how the proposed new limit differs. Is the methodology described in the topical report the same as that used to generate the current value and only the code is differenti RESPONSE TO QUESTION 15: The RTDP methodology has not been implemented for Seabrook's safety analysis prior to the YAEC 1849P submittal, The current DNI3R methodology uses a deterministic treatment of uncertainties with the W 3 correlation. The RTDP methodology implementation for Seabrook as described in YAEC-1849P uses WRB 1. RTDP is applied consistent with the approved method used by Westinghouse. 16 Explain why the uncertainty associated with the effective Bow fraction is assumed to be uniformly distributed rather than in accordance with a normal distribution function. RESPONSE TO QUESTION 16: Because the amount of measured data available for Seabrook was insufficient to characterize the distribution, it was difficult to justify the use of a normal distribution. As a result, the conservative assumption of a uniform distribution was made. This assumption is consistent with the older (moro conservative) ITDP approach described in WCAP-8567 (ITDP).

1 ,p. ,5 50 I I- .i i - HETMN umbre Level . Top of SG Tubee 2 Tg 30 E 3 3 20 0M 10 A 0 800 s 600 F .. FETMN CL. IESAR b ' 400 k w 76 200 1 O 0 250 500 750 1000 1250 1500 1750 2000 TIME (SECONDS) 4 I- -l H Lossof Feodwater 1 SEABROOK STATION COMPARISON OF RETRAN vs. UFSAR l FIGURE A 4.4 'l 4

J

,2 =. =m_

APPENDIX A (Continu:d) l Core Avernsto Temperature The peak temperatures match very well. The final temperature is lower in the RETRAN analysis due to the blowdown included in the RETRAN MSSV model. The MSSV blowdown results in a lower SG pressure which causes a lower RCS temperature. SG Pressure The initial SG pressure is lower in the RETRAN analysis. The lower initial pressure conservatively delays the opening of the MSSVs and the corresponding RCS heat removal, minimizing the DNBR and maximizing the peak RCS pressure. The peak SG pressure is greater in the RETRAN amdysis due to the greater setpoint tolerance assumed (3% in the RETRAN analysis vs.1% in the UFSAR analysis). The SG pressure in the RETRAN analysis falls below that in the UFSAR analysis due to the MSSV blowdown included in the RETRAN MSSV model. SG Water Volume The initial decrease in the SG water volume matches very well. The RETRAN analysis has a greater minimum volume and a faster increase in volume for several reasons: The initial RETRAN power level is lower than assumed in the UFSAR (3411 vs. 3565 MWT), with a correspondingly lower decay heat. The reactor trip occurs approximately 5 seconds sooner in the RETRAN analysis. The RETRAN analysis uses the EFW pump head-flow curve adjusted to give the minimum flow required by Technical Specifications, while the UFSAR analysis uses a constant 650 GPM, the minimum EFW flow at 1221 paig. The RETRAN analysis uses an EFW temperature of 100 *F, as opposed to the 120 *F EFW temperature assumed in the UFSAR analysis. SG Mixture Level Figure A 4.4 shows the RETRAN mixture levelin the tube bundle region of the steam generators. The dotted line in the figure represents.the top of the SG U tubes. The SG tubes are uncovered at approximately 65 aeconds. Direct comparison to the UFSAR mixture level is not possible since the UFSAR does not include a plot of SG level. Total EFW Flow The RETRAN analysis has a greater Emergency Feedwater (EFW) flow to the SGs since the flow varies with SG pressure, while the UFSAR analysis assumes a constant flow of 650 GPM (the minimum flow at 1221 psig). EFW flow in the RETRAN analysis is based on the actual head flow curve of the EFW pumps, which has been adjusted to give a flow of 650 GPM at 1221 psig. The j difference in integrated EFW flow is approximately 300 ft over the Grst 2000 seconds of the 8 i transient. l

APPENDIX A (Continued) Tf.81.E A 4 SEQUENCE OF EVENTS FOR TIIE LOSS OF FEEDWATER FI,0W Time (Soconds) Event RETRAN UFSAR Main feedycater flow stops 0.0 0.0 Low low steam generator water level trip 36.3 41.1 Rods begin to drop 38.3 43.1 Peak water level in pressurizer occurs 42 40.0 1

l 1.0 _ nc7n3n i .. OFSAR I' y0.0 E l P l j 0.6 Ea l g0.4 E \\\\ W-0.0 0 10 20 30 40 50 60 70 80 90 100 TIME (SECONDS) 2600 g - BEmAN 2500 l . UFSAR ~ $2400 3 E 2300 22200 1 2100 e 2000 a. s...*, 1900 1800 0 500 1000 1500 2000 2500 3000 3500 4000 4500 TIME (SECONDS) Loss of Feedwater SEABROOK STATION COMPARISON OF RETRAN vs. UFSAR FIGURE A 4.1

/ J 2000 I 1800 - RETRAN - - UFSAR RE 1600 0 E -31400 4 h1200 3 l/t c 1000 ( .g 8 800 600 ~ 400 l 625 __ ne1RAu -- UFSAR ^u. I w B 600 f,; e E }575 ( g ( 2 OO 550 &e O E 525 500 0 500 1000 1500 2000 2500 3000 3500 4000 4500 TIME (SECONDS) Loss of Feedwater SEABROOK STATION COMPARISON OF RETRAN vs. UFSAR FIGURE A-4.2 'l

1300 1250 3 1200 p 5 1150 ~' - RETRAN h - - UF SAR 9 1100 e

  • 1050

/ 1000 950 0 10 20 30 40 50 60 70 80 90 100 TIME (SECONDS) I 2000 _ ng, - - UFSAR. FT E 1500 o bo d 5 1000-i E3 i 1 500 \\, . ~ ~. '~..... 0 0 500 1000 1500 2000 2500 3000 3500 4000 4500 TIME (SECONDS) Loss of Feedwater SEABROOK STATION COMPARISON OF RETRAN vs. UFSAR FIGURE A-4.3

APPENDIX A (Continued) TABLE A-5 SEQUENCE OF EVENTS FOR THE FEEDWATER LINE BREAK (WITH OFFSITE POWER AVAILABLE) Time (Seconds) Event RETRAN UFSAR Main feedwater line rupture occurs 10.0 10.0 Pressurizer safety valve setpoint reached prior to reactor trip 16.0 MSSV setpoint reached prior to reactor trip 18.0 Low-low steam generator level reactor trip setpoint reached in ruptured steam generator 29.0 13.6 Rods begin to drop 31.0 15.6 Emergency feedwater is started 106.0 73.6 Low 6 team line pressure setpoint reached in ruptured steam generator 238.3 118.0 All main steam line isolation valves close 245.3 125.0 Safety injection flow started 268.3 Ruptured steam generator inventory completely discharged through break 334.0 Feedwater lines are purged and emergency feedwater is delivered to intact steam generators 342.0 456.0 Safety injection flow terminated 1158.0 Pressurizer safety valve setpoint reached 1232.0 416.0 Steam generator safety valve setpoint reached in intact steam generators 1852.0 468.0 Core decay heat plus pump heat decrease to emergency feedwater heat removal capacity 4494.0 3820.0

i 1.2 4 I I l l l Ill 1 ........ r.!. . u..,[ _ nerniu -- UFSAR 1.0 .i j V g 0.8 e l 0 .6 O 3 l a i m l @ 0.4 .l' o Z ll ',l i 0.2 g p 0.0 10' 10' 10' 10' 10' O.01 1 I I ll - RETHAN 0.0 -- UrsAa d -0.01 a i \\. i

-0.02 o

I l / 13 / i EI ' \\ 5 -0.03 y l ,,' /r ./ K i, -0.04 o -0.05 10' 10' 10' 10' 10' TIME (SECONDS) Feedwater Line Break SEABROOK STATION (Offsite Power Available) COMPARISON OF RETRAN vs. UFSAR FIGURE'A 5.1 i a.

l 2700 '~ 2500 lI il l ) i / I I Ill 2300 . RETRAN ~~ l f l .. k, 9 c 2100 - f UrsAR -o 1 E ll l l\\' l l l1700 N. / / l j f ll a' .g 1500 g 7 a l g ( 8 1300 8 l ll I l [ 1 ' 1100 I I I I I A / II l ll l Ill %'l 700 10' 10' 10' 10' 10' 2000 / [-.)T - I I 1750 l 1500 g / - RETRAN ~~ 3 1250 / 3 $1000 N;) ,fij k '\\ /l/ 3 .y f y ds = $ 500 r x\\ ...' / 250 's. ) 0 10' 10' 10' 10* 10' TIME (SECONDS) Feedwater Line Break SEABROOK STATION (Offsite Power Available) COMPARISON OF RETRAN vs. UFSAR ~ FIGURE A-5.2

3. b l 1.0 Z; 'd m _ [ N , gr,c t, 2 l-u. - RETRAN -- ursAn j 0.75 8> 2 0.5 y h 0.25 0.0 10' 10' 10' 10' 10' I l l Il - RETRAN -- ursAn 0.5 l' ll 0'0 -/ 3

  1. l l

x 2 r

r- ?T s

g 0.5 .u a 3 i E 1.0 l ^- o Z lM e e 1.5 l -2.0 l 10' 10' 10' 10' ~ 10' TIME (SECONDS) Feedwater Line Break 1 SEABROOK STATION (Offsite Power Available)

-(

COMPARISON OF RETRAN vs. UFSAR FIGURE A-5.3 ....i................._.__

l l l l llI - RETRAN g - - UFSAR o g200 8C y 150 9 is~ 100 I esw Ei E $ 50 ob g l' L. 0 10' 10' 10' 10' 10' 700 l l llI - RETRAN --oraR S 650 ll 0 l ll ....4 l 9 'l E l ll \\ I r 600 / I I W a ,,1me 8 550 a v,

l,

B '. < l s 500 450 10' 10' 10' 10' 10' TIME (SECONDS) Feedwater Line Break SEABROOK STATION (Offsite Power Available) COMPARISON OF RETRAN vs. UFSAR FIGURE A-5.4

700 ..... Jl.!,,.. l _ l ,k, ' ' ~ ~ ' ' ' ' ' ~ ' ' 650 / - RETRAN C - - UFSAR 600 S E 550 / m '3 ( v 4 500 450 10' 10' 10' 10' 10' 700 - lll l l - neinAN -- UFSAR C 650 I /- ~.i.. I ~ - / E ~. I';! E 600 h J- / .s

r x

c.550 , h l ,} b 5 \\," I / E~ 500 f Rf l l 450 10' 10' 10' 10* 10' TIME (SECONDS) Feedwater Line Break SEABROOK STATION (Offsite Power Available) COMPARISON OF RETRAN vs. UFSAR FIGURE A-5.5

700 ]"TYl YT I fQ c' 650 l ( __ pg g n (" -- ursAn / g l l s' 8 600 A / i / / 550 a/ -e 500 450 10' 10' 10' 10' 10' l l il{ - RETRAN ^ - - UF SAR 1250 j, s\\ H / 1000 s m! \\ [ 750 s s,N 3 500 5 i 2 ( 250 0 10' 10' 10' 10' 10' TIME (SECONDS) Feedwater Line Break SEABROOK STATION (Offsite Power Available) COMPARISON OF RETRAN vs. UFSAR FIGURE A-5.6

1500 1250 O, p,,,, q 7...... \\ 's' .,s' v \\ .G s g .........w...... 1,' / - - UFSAR e 5 s' ( b l' / 750 E '%' \\ l / 8 V", u 500 250 0 10' 10' 10' 10' 10' TIME (SECONDS) Feedwater Line Break SEABROOK STATION (Offsite Power Available) COMPARISON OF RETRAN vs. UFSAR ~ FIGURE A-5.7

50 llll 1 I I ill FETRAN - Fauled Loop i..' - FETRAN-irtact Loopa o Top os saiub '.,N-40 g a > 30 .. <\\f.n.......y.. > u..... 3 y..... .o o ) E h' .E "w 20

E s.

O i 0 s- / 10 / 0 10' 10' 10' 10' 10' I I t il b .FETRAN s UFSAR 800 s S c. O ~ 600 b m E 400 w ~n o H 200 4 0 10' 10' 10' 10' 10' TIME (SECONDS) Feedwater Line Break SEABROOK STATION (Offsite Power Avaltable) COMPARISON OF RETRAN vs. UFSAR FIGURE A-5.8 A J

APPENDIX A (Continued) TABL,E A-6 SEQUENCE OF EVENTS FOR THE LOSS OF RCS FLOW Time (Seconds) Event RETRAN UFSAR All operating pumps lose power and begin coasting down 0 0 Reactor coolant pump undervoltage trip point reached 0.24 0.24 Rods begin to drop 1.5 1.5 Minimum DNBR occurs 3.1 3.1

l i e 1.2 1.0 e0.8 _.8 mQ > 0.6 2a l0.4 02 0.2 -- UFSAR

  • RETRAN I

0.0 i 0 1 2 3 4 5 Time (seconds) 2800 i i I I i i i -g l l l l ' - newAn i i i i - nemAn [ l l l -- UFSAR i i i w i i i i 2600 -~~---------r-----------I-----------S-----------A------------ C i i t i 6 .i i i w i i i c i e i i c. i e i i c l l l l w i i ( N e i 4 E 2400 ------------ --------- ~~,--------~~-*-----------*------------ i ....e 1 D i i i. i i i l i w i i e i g i ',l ........----F... 'l l l a- ~ i i w, 2200 i - i i 0 2 4 6 8 10 TIME (SECONDS) SEABROOK STATION FOUR LOOPS IN OPERATION COMPARISON OF RETRAN Vs UFSAR FOUR PUMPS COASTING DOWN FIGURE A-6.1

APPENDIX A (Continutd) 7. LOCKED ROTOR Comparisons of the RETRAN results from YAEC-1871 versus the Seabrook Station UFSAR results for the cases with offsite power available and with a loss of offsite power are shown in Figures A 7.1 and A 7.2, respectively. The sequence of events are shown in Tables A 7.1 and A-7.2. For the locked rotor analysis, RETRAN is used only to determine the normalized core flow vs. time, the timing of the low RCS flow reactor trip, and the peak pressures. Normalized Core Flow For both cases (with offsite power available and with a loss of offsite power) the core flow in the RETRAN analysis initially decreases more slowly than in the UFSAR analysis. However, at the most limiting time in the transient (MDNBR and maximum fuel failures occur at 2.2 seconds) the RETRAN core flows agree well with the UFSAR. The RETRAN core flows following the limiting time in the transient fall slightly below those from the UFSAR analysis. The difference in core flow is attributed to minor differences in fluid inertia and loss coefficients between the RETRAN and ' UFSAR analyses. Pressurizer Pressure For both cases (with offsite power available and with a loss of offsite power) the initial pressurizer pressure is slightly higher (20 psi) in the RETRAN analysis due to a difference in the assumed pressure uncertainty and pressure control deadband (50 psi in RETRAN vs. 30 psiin the UFSAR). A higher initial pressure is used to conservatively increase the peak RCS pressure. Note that this RETRAN pressure response is not used in determining the MDNBR. For both cases, the peak pressure is greater than the UFSAR analysis due to the positive MTC assumed in the RETRAN analysis. This causes an increase in core power prior to the trip, depositing more heat in the RCS which increases the RCS temperature and pressure. l

APPENDIX A (Continued) i TABLE A-7,1 SEQUENCE OF EVENTS FOR THE LOCKED ROTOR (WITH OFFSITE POWER AVAILABLE) i Time (Seconds) Event RETRAN UFSA.R Rotor on one pump locks 0.0 0.0 Low flow trip point reached 0.15 0.04 Rods begin to drop 1.15 1.04 Maximum RCS pressure occurs 4.00 3.60

APPFEDLX A (Continued). TABLE A-7.2 SEQUENCE OF EVENTS FOR THE LOCKED ROTOR (WITH A LOSS OF OFFSITE POWER) Time (Seconds) Event RETRAN UFSAR Rotor on one pump locks 0.0 0.0 Low RCS flow trip setpoint reached 0.15 0.08 Rods begin to drop 1.15 1.08 Remaining reactor coolant pumps begin to coast down 2.15 2.08 Maximum RCS pressure occurs 4.65 4.0

i l i i 1.2 - ReinAn - - UF SAR ') 1.0 8 i', 8 ~ g 0.8 .y ...... w................ --- e E OZ 0.6 0.4 2600 2500 - REWN s UFSAR -a 'o s'~ n .G g4no b N, y2300 N........ ~~- m

a. 2200 E

E 2100 8w 8 2000 c. 1900 1800 0 2 4 6 8 10 TIME (SECONDS) Locked Rotor SEABROOK STATION (With Offsite Power Available) COMPARISON OF RETRAN vs. UFSAR FIGURE A-7.1

t 1.2 - RETRAN - - ur pa 1.0 's Bw 0.8 ~... 2 '6 N 0 ~.. 2 b0.4 Oz 0.2 0.0 1 2600 ["- -..., mN - RETRAN - - UFSAR 2500 ,+' ,' N g N g2400 8 2300 ..e 8 e 2200 0 E 2100 8 f2000 1900 1800 0 2 4 6 8 10 TIME (SECONDSS) Locked Rotor SEABROOK STATION (With a Loss of Offsite Power) COMPARISON OF RETRAN vs. UFSAR FIGURE A-7.2

APPENDIX A (Continuod) 8. ROD WITHDRAWAL Comparisons of RETRAN results are shown in Figures A-8.1 and A.8 2. The sequence of events is shown in Table A 8. The RETRAN case presented mimics the 80 pcm/see withdrawal rate presented in the UFSAR. Only system response parameters are presented for comparison. Core Power The core power undergoes a similar increase as the UFSAR although the UFSAR includes a 2% calorimetric uncertainty consistent with the deterministic approach. After the high neutron flux trip is initiated there is a faster decrease in core power from RETRAN caused by the use of a less conservative scram curve than that which is used in the UFSAR. The scram curve in RETRAN is still very conservative when compared to the axial flux profiles being used in the UFSAR analysis. Heat Flux The core heat flux undergoes a similar increase as the UFSAR. The UFSAR includes a 2% calorimetric uncertainty consistent with the deterministic approach. After the high neutron flux trip, the core heat flux is conservative compared to the UFSAR results. Pressurizer Pressure The initial pressure in the UFSAR is based on the deterministic approach where pressure is the nominal value of 2250 psia minus the uncertainty value of 30 psia for a final pressure of 2220 psia. The pressure goes through a similar increase as the UFSAR until the high neutron flux trip is initiated. The RETRAN pressure decreases at a smaller rate after the trip consistent with a smaller rate of decrease in the core heat flux.

APPENDIX A (Continued) 9. DROPPED ROD CLUSTER CONTROL ASSEMBLY A " typical case with rod controlin automatic is presented in the UFSAR. Specifics of the case are l not discussed. However, a reasonable comparison is possible by selecting a RETRAN case with 180 i pcm dropped rod worth, and an early in life MTC of-3pcmfF. Comparison of RETRAN results versus the Seabrook Station UFSAR results are shown in Figures A-9.1 through A-9.2. Normalized Power RETRAN results show good agreement with the UFSAR results. Normalized Core Heat Flux Results show good agreement. The lag in the RETRAN heat flux is explained by the smaller prompt increase in power following the control rod drop. Averace Coolant Temr>erature The difference in temperatures is attributed to the difference in initial conditions. The UFSAR initial temperature = nominal + uncertainty. RETRAN initial temperature = nominal. Pressurizer Pressure The difference in pressures is attributed to the smaller prompt power increase following the control rod drop in RETRAN. 1 I

1.2 t l l l - RerRAs

t.--------------j...--UFSAR 1.1 u

3,o ,......,,......i,....... c. i i i f0.9 I.,8-/-----------I,--------------$,--------------k,--------------- _.O I i i E i i i b i i i y 0.8 --------------->------------- :--------------*--------------- f I i t i t 0.7 ---------------h,--------------I,--------------4,--------------- e i i i t 0 0 f 0.6 1.2 .e i i i i l l l - RETRAN - - UFSAR 1,3 .............,.,.t..............;..............j... s s a. i i

e--.....-"-'.1...........

1.0 r-----,/--

r---------

x e i i 2 l l l u-O i y 0.9 ~------------{--------------t--------------{--------------- e i 8' 0.6 i i i i

*--------------*--------------<--~~~~~--------

i .i i l 8 i 0.7

h--------------I--------------4---------------

e i e i e i 0.6 O 50 100 150 200 TIME (SECONDS) Dropped Rod Cluster Control Assembly SEABROOK STATION AUTOMATIC Rod Control COMPARISON OF RETRAN vs. UFSAR 180 pcm Dropped Worth,-3 pctn/F MTC FIGURE A-9.1

1 600 i e 1 i i i l,.............l...............lc... - UFSAR RETRAN 590 ,;-------,,8

i- -- -

l c. '~.....- l l-E i e i i H i i i E 580

*------------- ;--------------j---------------

m i i i in 570 ---------------b--------------?-~~----------A----~~--------- e e a i e 560

h--------------i--------------4-------~~------

e i i e i i. 550 2400 \\ l l ~ %WN i i e - - UF SAR ' .) 2300

t *-

t......

4,---

i e i 8" 2200 i e i .i O e e i e a i N 2100

r--------------,.--------------,i E

e i i .e i i i e i 2000 -----------~--b--------------f--------------4--------------- i. i i i e i. i i 1900 O 50 100 150 200 TIME (SECONDS) Dropped Rod Cluster Control Assembly SEABROOK STATION AUTOMATIC Rod Control COMPARISON OF RETRAN vs. UFSAR 180 pcm Dropped Worth, -3 pcm/F MTC FIGURE A-9.2

APPENDIX A (Continued) 10. ACCIDENTAL RCS DEPRESSURIZATION A comparison of the RETRAN results from YAEC-1871 versus the Seabrook Station UFSAR results is shown in Figures A 10.1 and A-10.2. The sequence of events is shown in Table A-10. Normalized Core Power The initial core power is lower in the RETRAN analysis since it uses nominal initial conditions, without the 2% uncertainty in power. The nominal power is used to provide the appropriate input to the MDNBR evaluation, since the uncertainties on power, pressure, temperature, and flow are all included in the MDNBR limit for the WRB 1 DNB correlation that is used. Due to the positive MTC assumed in the RETRAN analysis (implemented as a table of reactivity vs. coolant density), the core power rises slightly prior to the trip as the coolant density decreases with the pressure decrease. The earlier reactor trip in the RETRAN analysis is due to an Over-Temperature AT (OToT) trip which occurs prior to the low pressurizer pressure trip reflected in the UFSAR analysis. The UFSAR indicates that auto rod controlis assumed to avoid the OTAT trip. RETRAN cases with auto and manual rod control were evaluated, both of which resulted in an OTAT trip at the same point in time. The difference in reactor trip signals may be due to a disabled OTAT trip in the UFSAR analysis. If the reactor were assumed to trip on low pressurizer pressure in the RETRAN analysis, the trip time would be almost identical to the UFSAR analysis due to the similarity in the pressurizer pressure responses. Pressurizer Pressure The initial pressurizer pressure is slightly lower (20 psi)in the RETRAN analysis due to a difference in the assumed pressure uncertainty and pressure control deadband (50 psiin RETRAN vs. 30 psi in the UFSAR). A lower initial pressure is used to conservatively minimize the MDNBR. Except for the difference in initial pressures and trip times, the pressurizer pressure response matches very well. i Averate RCS Temperaturee The initial average RCS temperature is slightly lower (4 'F) than in the UFSAR analysis since the RETRAN analysis uses nominalinitial conditions to provide the appropriate input to the MDNBR evaluation, as discussed above. The difference in reactor trip times causes a difference in the timing of the post-trip decrease in the average temperature. The difference in temperature is attributed to differences in MSSV modelling between the RETRAN and UFSAR analyses. The SG pressure response in WCAP-7769 (" Overpressure Protection for Westinghouse Pressurized Water Reactors") indicates that Westinghouse may be modelling the MSSVs as open/ closed valves. The RETRAN MSSV model ramps the MSSVs open between the setpoint pressure and the accumulation pressure (3% above the setpoint). The slight increase in the average temperature at the end of the RETRAN analysis is due to the decrease in mass flow through the RCPs as the pressure drops and the steam quality increases in the loops upstream of the RCPs. This results in an increasing AT across the RCPs since the flow decreases more than the pump heat. The increasing RCP AT causes an increasing cold leg temperature which increases the average temperature. l 1 l

APPENDIX A (Centinued) TABLE A 10 SEQUENCE OF EVENTS FOR THE ACCIDENTAL RCS DEPRESSURIZATION Time (Seconds) Event RETRAN UFSAR One pressurizer safety valve failed open 0.0 0.0 Overtemperature AT reactor trip setpoint reached 19.1 Low pressurizer pressure reactor trip setpoint 31.3 reached Rods begin to drop 21.1 33.3 4

4 4' 1.2 l - REMAN D'II'. M - - UFSAR 1.0 }' s O [0.8 3.6 i 0 3 a e @ 0.4 i 0.2 0.0 2250 l - RETRAN ~, - - UFSAA 2000 g g 's e 8 1750 >3 c" m $ 1500 m M e A 0-1250 1000 0 25 50 75 100 125 150 TIME (SECONDS) Accidental RCS Depressurization SEABROOK STATION COMPARISON OF RETRAN vs. UFSAR FIGURE A-10.1

I - RETRAN i -- UFSAR C {575 k E 4 550 v)o CC $525 ?c 500 0 25 50 75 100 125 150 ) TIME (SECONOS) ' l Accidental RCS Depressurization SEABROOK STATION COMPARISON OF RETRAN vs. UFSAR FIGURE A-10.2

APPENDIX B. ILLUSTRATION OF TIIE YAEC APPROACH TO CIIAPTER 15 ACCIDENT ANALYSIS USING RETRAN APPLIED TO SEABROOK STATION (YAEC 1871) f f 6 1 s . t D 'l . j - i ' 36' q - w a v. i +w,-

y YANKEE METHODOLOGY FOR PWR TRANSIENT ANALYSIS USING RETRAN 4 Key Features of the Base RETRAN Modelin YAEC-1856-P .S Illustration of Application to Specific Non-LOCA Transients For each transient: a) Goal (output) of the RETRAN analysis b) Initial conditions c) Model changes (if any) d) Transient specific input e) Sensitivity studies 4 Transient-Specific illustrations Available in YAEC-1871 (Submitted on the Seabrook Docket) e b v

RETRAN Nodalization lGURE 3-I <= a4 2 ---H i i ,f S s < ) j '"I 5 I ~ L E C J H 3 aj f l

  • fnt 7,

t n a _ =l a r-- f 8 04-i x ~ 5,Ef- "

  • +-

5 Ei l' g 8 q = i s-s/ 1 w l 3 .L4 + a 5 3 t Q ~I i -4 ] I I a i 1 = l = m g I__ a T =~T = b o rf' n4 O .I a l W l b G a I l E 4-- gg E =

  • $~~"

Gi"-- n ( m i m a N / ni a]a+ o a , i si sia A / g ] l ) 3 g a h i ai ai: a e 5 3 3 k I m R l l in Ri ) 1 I

LIST OF NON-LOCA TRANSIENTS 9 MSLB 4 Loss of Feedwater ~ 9 Feedwater Line Break 9 Turbine Trip 9 Dropped Rod 9 Bank & Single Rod Withdrawal 9 Loss of Flow G RCP Locked Rotor and Shaft Break 9 10% Step Load increase

  1. Feedwater Flow increase O Accidental Depressurization

LOSS OF FEEDWATER (Continued) Transient-Specific input: Maximum EFW delay and limiting single fallure (1 out of 2 EFW pumps operate) Main FW flow is terminated at the start Pressurizer PORVs, spray, ASDVs, and Steam Dump are disabled Rods in manual (no response) Trip on high pressurizer pressure trip disabled (occurs prior to low SG level) Hot FW must be swept before cold EFW reaches the SGs 9 Sensitivity Studies: Loss of offsite power (both reported in YAEC-1871) Passive heat conductors RCS pipe walls, RV walls and internals, SG walls and internals (slightly more limiting results without) Sensitivity of primary to secondary heat transfer to tube uncovery (local fluid conditions heat transfer option): a) No effect on peak RCS and MSS pressures b) SG minimum inventory was lower without local conditions heat transfer (difference - 1.5% of initial mass)

FEEDWATER SYSTEM PIPE BREAK Goals: Show that the EFW capacity is adequate to: al remove decay heat, b) prevent RCS overpressurization c) prevent fuel damage (DNB) Maximum RCS heatup 9 Initial Conditions: RCS temperatures, pressure, core power = maximum + uncertainties Minimum steam volume in pressurizer Most positive MTC Faulted SG level = nominal + uncertainty (delays trip & EFW actuation) Intact SG levels = nominal - uncertainty (maximizes heatup) 9 Model Changes: DEG break of FW line at SG. Choked flow (extended Henry & Moody) SG tow level trip setpoint conservatively calculated using minimum SG mass at the low level setpoint (see 1.oss of FW Flow) 9 Transient-Specific input: Offsite power available (RCP heat addition maximizes heatup) Pressurizer pressure and level control disabled Turbine trip at time of break ~ ASDV's and Steam Dump disabled

FEEDWATER SYSTEM PIPE BREAK (Continued) e Sensitivity Studies: Local conditions heat transfer option versus standard heat transfer modelin SG boiler region. Peak RCS and MSS pressures not sensitive. Timing of peak RCS pressure affected. (Local conditions model = more limiting) Limiting single failure in the EFW system (1 of 2 pumps versus a failed closed branch line control valve to one of the intact steam generators)(1 EFW train is limiting) Containment backpressure effect on break flow (no sensitivity) L

LOSS OF LOAD / TURBINE TRIP e Goal: Determine peak RCS and MSS pressures e initial Conditions: l RCS temps and core power = maximum + uncertainties RCS pressure = nominal-uncertainty (delays high pressure trip and maximizes core heat input) RCS flow = TDF Most positive MTC Least negative doppler e Transient-Specific input: EFW, ASDV's, Steam Dump disabled Main FW flow terminated at time of trip Rod control system disabled to maxim!ze hoat input prior to trip i e Sensitivity Studies: Comparison to UFSAR show in YAEC-1856P Initial SG pressure: a) Low is conservative for RCS peak pressure l b) High is conservative for MSS peak pressure ) RCS pressure control: al off = maximum RCS pressure b) on = maximum MSS pressure MTC (most positive !!miting) Pressurizer level (no sensitivity from YAEC-1847)

44 s.--. .4 4 4 A .t DROPPED RCCA 4 Goal: Calculate Core T&H conditions for DNB analysis during power overshoot 9 Initial Conditions: Nominal RCS temperatures, pressures, and flow-(uncertainties in DNBR limit - value) Core power = variable Doppler = least negative MTC = variable Pressurizer pressure control oj 9 Model Changes: Turbine inletjunction modeled as negative fill with constant flow (i.e., constant turbine load) i l G Transient-Specific Input: ] Dropped rod worth (up to full bank) Limiting single failure in rod control system causes power overshoot Rods in automatic 9 Sensitivity Studies: Dropped rod worth MTC Doppler (least negativd Initial power level (limiting at high powers but affected by Al bandwidth) = a.

DROPPED RCCA (Continued) e Sensitivity Studies (Continued): Rods in manual (rods in automatic) Control bank worth (high worth) Excore nuclear power " tilt" factor (smaller factor conservative) J

BANK WITHDRAWAL ANALYSIS 9 Goals: Plant T&H responso for DNB analysis Demonstrato RTS coveragoloffectivonoss e initial Conditions: 10 %,70 %,100% Nominal RCS temperature, pressures, & power lovel Model Changes: Nono Transient-Specific input: Pressurizer pressure + level controls systems functional (to minimize pressure) Rod control disabled 9 Sensitivity Studies: Reactivity insortion Rato (power level dependent) Minimum and maximum reactivity foodback (power level dependent) D,y (weak sensitivity, conservativo = largo) a

SINGLE RCCA WITHDRAWAL 9 Goal: Plant T&H response for DNB analysis e Initial Conditions: RCS temperatures, pressure, and core power = nominal MTC = most positive Doppler = least negative G Model Changes: None Transient-Specific Input: Reactivity addition for withdrawal of a single rod at maximum rate Control banks frozen 9 Sensitivity Studies: Core power = variable Minimum / Maximum feedback (least negative feedback) Minimum / Maximum B,n (no sensitivity) Pressure + level control operable / disabled (operable more limiting) Withdrawn RCCA location (limiting = bank D diagonal) j

LOSS OF FLOW G Goals: Calculate Core T&H conditions for DNB analysis (need flow coastdown only) Flow coastdown shown in YAEC-1856P Calculate peak RCS pressure response e initial Conditions: RCS pressure, temperature, core power = maximum + uncertainty Most positive MTC

i. east negative Doppler Model Changes:

None l l l 9 Transient-Specific input: i RCP's tripped Pressure and level control disabled l 1 ASDVs and Steam Dump disabled Main FW flow terminated at time of trip S Sensitivity Studies: None

RCP LOCKED ROTOR AND SHAFT BREAK 9 Goals: Plant T&H response Peak RCS and MSS pressures Normalized core flow to DNB (fuel failure) analysis Reactor trip time G Initial Conditions: RCS temps, pressure, and core power = maximum + uncertainties Most positive MTC Least negative Doppler G Model Changes: None e Transient-Specific input: Pressurizer PORVs, spray, ASDVs, and Steam Dump disabled Main feed flow terminated at time of trip Shatt break simulated by forcing pump torque to zero, allowing reserve rotation, and simulating pump impeller inertia only Locked rotor simulated by forcing pump speed to zero

RCP LOCKED ROTOR AND SHAFT BREAK (Continued)- 9 Sensitivity Studies: Looked at Locked Rotor / Shaft Break / LOOP (most conservative = Locked Rotor with LOOP) Looked at normalized coastdown from TDF and Minimum Measure Flow (no-difference) ^ Looked at most negative Doppler (least negative Doppler is most limiting) 1 'l 1 l i 1 -l

10% STEP LOAD INCREASE G Goal: Calculate core T&H conditions for DNB analysis G-Initial Conditions: RCS temperatures, core power = maximum + uncertainty (conservative) RCS pressure = nominal-uncertainty (conservative) 9 Model Change: Turbine throttle valve junction is changed to a negative fill Junction.to model 10% step increase in steam flow e Transient-Specific input: 1 Pressurizer heaters and charging / letdown are disabled (yleids minimum '. a pressure) G Sensitivity Studies: I Most positive & most negative MTC Rods in automatic and manual

Ld y FEEDWATER FLOW INCREASE 9 Goals: Calculate plant T&H response Verify that reactivity insertion rates are bounded by the RCCA bank withdrawal analysis Perform DNBR check 9 Initial Conditions: RCS temperature, core power = maximum + uncertainties (conservative) RCS pressure = nominal-uncertainties (conservative) MTC = most negative Doppler = least negative e Model Changes: Split RV model from MSLB analysis (no mixing between affected loop and other loops) Spilt core model from MSLB analysis (with reactivity weighting) S Transient-Specific input: Stop increase in FW flow to one SG disabled Pressurizer heaters and charging / letdown 4 Sensitivity Studies: HZP & HFP With and without reactor trip on turbine trip (turbine trip is from high SG level) I l

2.(

) 4 ACCIDENTAL-RCS DEPRESSURIZATION - O Goal: Calculate Core T&H conditions for DNB analysis G Initial Conditions: RCS temperatures and core power = maximum + uncertainties RCS pressure = nominal-uncertainty ? Most positive MTC Least negative Doppler G Model Changes: Pressurizer safety valve: a) Single valve stuck open b) Critical flow (extended Henry & Moody) c) Area adjusted to provide rated flow e Transient-Specific input: . Main FW flow terminated at time of trip ASDVs, Steam' Dump and pressurizer heaters disabled Level control disabled Rod control system in manual G Sensitivity Studies: None

1...L

,}}

Retrieved from "https://kanterella.com/w/index.php?title=ML20065H677&oldid=4438460"