ML20090J352
| ML20090J352 | |
| Person / Time | |
|---|---|
| Site: | Ginna  |
| Issue date: | 08/21/1984 |
| From: | Bernero R Office of Nuclear Reactor Regulation |
| To: | Eisenhut D Office of Nuclear Reactor Regulation |
| Shared Package | |
| ML082380335 | List:
|
| References | |
| CON-FIN-A-2311, FOIA-91-106 NUDOCS 8408290357 | |
| Download: ML20090J352 (2) | |
Text
{{#Wiki_filter:~ l 6 {, pn' tn-t(ga p 2 0,g gg y {% gjQ21 3 MEMORANDUM FOR: Darrell G. Eisenhut, Director Division of 1.icensing Daniel R. Mul'ler, Acting Director FROM: Division of Sytems Integration GlN*tA STEAM GENERATOR TUBE RUPTURE
SUBJECT:
EVEHT: CONF 1RMATORY A::ALYSIS C Letter to Ms. Ruth Caplan (Chairpirson, Sierra Club
- +ee) f ron H. Denton, NRR, "Direc-
Reference:
isationel Enargy Cot.: tor's Decision Under 10 CFR 2.206," October 8,1982 The reference letter cxxiitted the staf f to perform a detaileo 25, 1982, Ginna steam thernal-hydraulic analysis of tne JanuaryThis memorandum fulfills that generator tube rupture event. comitment. With cooperation of !NPO and Rochester Gas and Electric Com by IMP 0 to model the tube rupture event). We applied this r,cdel to confirc our understanding of the ther-In addition, mal-hydraulic conditions which existed during the event.we confirm adequately direct the operator to mitigate similar events. During the event, the reactor operators responded in a manner consistent However, with the lessons with the Westinghouse Generic Guideline E-3. of TH1 in mind, the operators hesitated to throttle ECCS until af ter the The RETRAN analysis confirmed af fected steam generator filled solid. 4 that had the E-3 guidelines been implemented, filling of the secondary systco would have been prevanted. Additional studies confirmed that for a postulated steam generator tube rupture event, compounded with the f ailure of the PORY and block valve to close, core uncovery would not occur. C0itTA JQu? ,./' _ CIA J.-Go t t.ur.n, R \\\\ ~) 8400290357 aggy e t ~ F AteeK orn;; .;,4 r / ,L,I r q _...._...-'i}s} f '/ NY r 7, [ j f + om. _ m._ smg L--_--________-__-_-_-_____-___________
~ D. Eisenhut We have assessed the consequences of nut terninating ECCS. Our anelysis e showed a primary system cooldown rate of 70*F/hr. The primary systen achieves a steady-state pressure of approximately 1300 psia with a constant primary-to-secondary pressure dif ferential of 300 psi. At this condition, the primry system is liquid solid and the atmospheric dump valves discharge 600 gpm (ECCS capacity). Should this condition remain unchecked, the RWST would be depleted outside contsiamn. However, the opernor guidelines adequately direct the reactor operator when ta throttle ano terminate llPI and to depressurize the primary system betaw the secwdary systen pressure. 2 Euclosed is ANL's docur.unt of the above nentiem.d 3,altses, as i contracted under FIM A-2311. These aralyses canhrTwd the staff po-sitions outlined in NRR's letter to lis. Ruth Caplan (is it relates to the thenul-hsdrou)1c und.'rs tnoding of the 6 tom SCTR event). I request thn OL forwsed this to the inwiv-d p..r ties ss appropri it. Original Signed By: Robert M. Bernero Robert l'.. Baro?ra, Director Division of Systas Integration Office of Nuc1%r Pector Regulatory
Enclosure:
As stated cc: f OISTRIBUTION E ~l1. Tocopson Docket file D 7.iemano RSB R/F L. Shotkin, RES RSB P/F: Ginna G. Dick (Gintia P!)) JGuttmann R/F NLauben BSheron RHouston RBernero JGuttmann
- PREVIOUS CONCURRENCE SHEET ON 'ILE W/RSB
<V
- DSI:RSB
- DSI:RSB
- DSI:RSS iDI T 05I:D
- ,,,cr y j
.y G'u't'tWa nn ij f ' 'NL'a'ub'en' Mh'e r'o'n' ' 'R 4gys to n' ' 'RBsfrisrd dan >' 08[01/.8d..../.01/.dddd[d1/84 Od/fj/54 d8/U/84 ~ ' ~ 08 E7onu aia nosomnc""" OFFICIAL RECORD COPY
..... _... ~ -._. - -. - - ~. - e-ANL/ LWR /NRC 84-3 June 1984 GINNA SGTR EVENT - RETRAN CALCULATIONS by J. H. Tessier and T. Y. C. Wei Light Water Reactor Systems Analysis Section Reactor Analysis and Safe *y Division -ARGONNE NATIONAL LABORATORY 9700 South Cass Avenue Argonne, Illinois 60439 Prepared for: Division of Systems Integration Office of Nuclear Reactor Regulation V. S. Nuclear Regulatory Commission Washington, D.C. 20555 s. +# 44 8 i O ? ~%,, / g p & @ T N '- a' ,o //
4 -1 TABLE OF CONTENTS Page . ii List of Tab 1es......................................................... L i s t o f F i g u re s........................................................... i i i 1 EYECUTIVE
SUMMARY
3 1.0 I N T R O D U CT I O N........................................................ 4 2.0 P L ANT M0 0 E L......................................................... 3.0 I N P O CO MP AR I SO N.................................................... 10 3.1 t= 0 D e c k..................................................... 10 3.2 t= 4 2 m i n s D e c k............................................... 13 4.0 P A R AM E T R I CS........................................................ 4 2 4.1 We s ti no house Goera tor Gui deli ne s............................. 42 4.2 S tu c k O p e n P 0 R V.............................................. 54 4.3 Fai l ure to Te rmi n a te HP I...........................,......... 55 5.0
SUMMARY
AND CO N CLU S I O N S............................................ 5 0 A c k n o d e d g e me n t s......................................................... 61 References..................................,............................ 62 O e a 1 l 1
m - \\,- i List of Tables Tabl.e Page - 4. ' - 1 O p e r a t o r A c t i o n s.................,.................................. 44 i e B W f f M 3 ~ p ) 8 i S 0 .t. = 1 i ~,. .4
x l J -111-List of Figures, Figure
- Page, 2-1 RETR AN Model Vol ume s and J unc ti on s................................ 6 2-2 RETRAN Model Heat $1 abs...........................................
7 3-1 R CS P r e s s u r e v s. T i me............................................ 16 3-2 P r e s s u r i : e r l e v e l v s. T i me....................................... 17 3-3 P re s suri zer Tempe ra tu re s vs. Ti me................................ 18 3-4 Core E xi t Tempe ra tures v s. T i me.................................. 19 3-5 Reactor Ves sel Uppe r Head Temnera ture s vs. Time.................. 20 3-6 RCS Av e ra ge Tempe ra ture v s. Time................................. 21 3-7 Loop " A" Col d Leg Tempera ture vs. Time........................... 22 3-8 L oop "B" Col d L eg Tempe ra ture v s. T ime........................... 23 3-9 " A" S te am L i ne P res su re vs. Ti me................................. 24 3-10 " B" S te am L i ne P r e s s u re v s. T i me................................ 2 5 3-11 "A" S team Genera tor Nar: ow Range Water Level vs. Time............ 26 3-12 ' "B" S teaai Genera tor Narrow Range Level vs. Time.................. 27 3-13 Steam Generator Tube Rupture, Safety injection, and Charging Flow vs. Time.................................................... 28 3-14 Reactor Ye ssel Upper Head Steam Vol ume vs. Time.................. 29 3-15 Reac tor Cool ant Lonp Fl ow Ra tes v s. Time......................... 30 3-16 Loop "B" Reactor Yessel Inle t Flow Ra te vs. Time................. 31 3-17 RCS H ot Leg Tempera tures v s. Ti me................................ 32 3-18 Reac tor Yessel Downcomer Tempera ture vs. Time.................... 33 3-19 "B" S team Li ra Wa ter V ol ume v s. Ti me............................. 34 3-20 "B" Steam Generator Safety Valve Flow Rate and Flow Area vs. Time......................................................... 35 3-21 " A" S team L i ne P re s sure v s. Time (0-20 Mi nute s ).................. 36 3-22 "B" S team Line P re s sure v s. T ime (0-20 Mi nu te s ).................. 37 [ {
-iv-List of Figures (cont'd) Figure Page 3-23 " A" and "B" Stean. Generator Level s vs. Time (0-20 Minutes)........ 38 3-24 R CS P re s sure v s. T i me (0 - 20 Mi nu te s )........................... 3 9 3-25 Pres suri zer Pres sure vs. Ti me (0 - 4 Mi nutes )................... 40 3-26 Core Power and RCS Average Temperature vs. Tice (0 - 4 Minutes).. 41 4.1-1 RCS Vol. 61 Pre s sure/SGB Vol. 63 P r? s sure........................ 45 4.1-2 Upper Head Vol. 19 Vapor Volume.................................. 45 4.1-3 H a rrow Range P res suri ze r Leve1................................... 46 4.1-4 SI Flow / Break F10w............................................... 46 4.1-5 SGB Leve1........................................................ 47 4.1-6 C ore Exi t Vol. 13 T empe ra tu re..................................... 4 7 4.1 SG A V ol. 58 P re s s u re............................................. 48 4.2-1 R CS Vol. 61 P re s sure/SGS Yol. 63 P res sure........................ 48
- 4. 2-2 -- SG Tube Dow nhi l l V ol. 44 Qual i ty................................. 49 4.2-3 Upper Support Vol. 20/ Active Core Vol.1/SG Tube.
V ol. 4 3 Q u a l i ty.................................................. 4 9 4.2-4 U p pe r H e a d V ol. 19 V a p or V o1 ume.................................. 51 4.2-5 Guide Tube V ol. 14/ Bundle Top Vol. 15 0uali ty.................... 51 4.2-6 G u i de T u be V o l. 16 Q u a l i ty....................................... 5 2 4.2-7 P res su ri z e r Mi x ture leve1........................................ 52 4.2-3 V e s sel B i nl e t J ua 65 Fl ow........................................ 53 4.2-9 P re s suri z e r PORY F10w............................................ 53 4.2-10 SI Flow / Break Flow.............................. ................. 54 4.2-11 Y e s s el O cw ncomer V ol. 18 Tempe ra ture........ -..................... 54 4.3-1 R CS ( V o l. 61 ) P r e s s u r e........................................... 5 7 4.3-2 " B " S team L i ne (V ol. 68 ) P re s s u re................................ 57 l l l
_. -. ~ - l: .y. 1 1 l P t.ist of Figur s, g s I Fig ure Pag i r l 4.3-3 HP !, Chargi ng and Tube Rup ture Fl ow Ra tes........................ 53 l 4.3-4 "B" S team Genera tor Sa fe ty Va l ve F10w............................ 58 i' 4.3-5 Reactor Vessel Downcomer (Vol.18) Teinperature.... ............ 59 f 4.3-6 Reactor iessel Upper Head (Vol. 19) S team Yo1ume................ 59 t i k J t k f v t 1 - 1 y n-i i i- ) i e i i e i a e {' i a
4 g v,g g, j#p,,' / IMAGE EVAL.UATION 3, O > ' s, Q,' TEST TARGET (MT-3) 4 .g 9 / t '\\\\ ' figy @gy yk h(O. 3,. la g 4 v ,w,,c ':a%g Il'
- 'a
- y 22
= -m d E.O. l l,l 2es a ts
- !d33H 1.2 5 il!
l.4 P
==.kmm b:.6u 4..__.__.__ ____. _.- 150mm - - - ~ mg es e --+==+we.---% .-_..mAwee.em+e==4- .mm e% 'Y* ,3.p ik 4 ,g,\\sh. 'h,\\ ' N [bz,'# y p g',r/.\\ p i A.,'., y% f .~ is 'e er s;>... ij / t gf G ~ g' g*.;> Op go Q@ ,y <p z . '. 41 3 % ) g j ^ ^ ^ ' ^ ^ a, l ' _ -; .g 'g,
f n? g g N g .s 3 s ~ IMAGE EVALUATION S Q g </ e j[f v 1 TEST TARGET (MT-3) p j s %'QpV 4g? w y. l.0 ,22 a i, i ? O lj % wu - m ts unu 4llL== d r+= i!!!am 1.25 li" I.4 till I.6 4.._____ _ ___ __.- 1 5 0 m m
- l. __ __ _
3- 'S & A. 4
- os s.
,sw,i, 4B$yY 4 $eN a, c. y ' R QQ Of7 y Li- / p c~-
g ,p. 3 O ~ f '8"'(. IMAGE EVAL.U ATION + s\\[O 9, / / TEST TARGET (MT-3) h f x/ t> s 'g w I i 1.0 m Cl 2 0 lJ w-> M 1 '[ I.2 5 jj!'l =l.4 gi 1.6 ir== ni += bl== _ _.. ---- 15 0 m m 4 -__ 6" 8' ,s. jf , ky 4 g m, c ,i t ~? ~ 0 y4 y '#O . ?' ~"': ^ ~
. l E A 7 ,w ~; ? f / lo e.4 IMAGE EVALUATION '(O
- $g
TEST TARGET (MT-3) / sh k f
- 9* 4 g.
IV V 4>9)> ^ q g 4 g g$6 l.0 1" ^ .. ? E t
=
i ii Na i s i l.8 li sy b,i!.bd. OiIb =
- 2. mse 4_.----
15 0 m m ----- 4-6" d>, D 44S*p+% Q[7i)Q,p spy /$'\\ m g,.,ji y qc 4p h- -
q-EXECUTIVE SUKMARY A simulation of the Ginna Steam Generator Tube Rupture (SGTE) event of January 25, 1982 was performed utilizing the latest EPR!-released version of RETRAN, RETRAN02/M0003, in conjunction with RETRAN02/M0003 A input decks ob-tained from the Institute of Nuclent tower Operations (!NPO) and modified by
- ANL, The RETRAN02/M0003 re su' ts agree well wi th the 1NPO RETRAN02/M0003A calcula tions.
A reasonable match is therefore obtained between calculations and the measured data fron, the actual event. Where dif ferences between the two calcu1 Ations have occurred, they can be explained in terms of code model dif ferences between the MOD 03A and the N0003 versions and in terms of sensi-tivities in the INPO calibration to data. In addi tion, three parametrics were performed which includei variations or, operator actions and further equipnent failure. Results of these para.netri: calculations demonstrate <!; that oppor-tune tiining in conforming to recent operator guidelines wooij prevent filling the disrupted stean generator solid and alleviate concerns about loading ques-tions; that additional failures occurring in the PORY line downstream of the PORY would not necessarily lead to significant core damase; and that suffi-cient thermal nurgin exists against pressurized thermal shock conditions even J if there was a further ' continuation of sarety injection flow. Furthermore, the ua.rametrics have contributed to the understanding of the thermal hydrualic phenomena that occurred during the actual cequence of events during tne Ginc.a SGTR incident. l 5 l 9
4$6.sm6.A ƼS 44-,-4 4,-.46.F 42-4f. 4 #J nM e J s s e J-5. e.dA A P W 4 4 4 d4h A M,' '" 4 14f DM4 64 "'4M 4 Mu'E (J 4 4 h-, crw446 4 A=Whear 5 35JAVh 9MJu F S E M.& M.ehe Ned 3 OJiiAJ#,& $s'M M-- he G. e5 m AW4 5&s #-.5 W-Jder- # M M J#M6r AM Sdtri9 J } -' (.g ' 6 i .g. 4-' 9 a ,E ? 4 e 4 w I f { - ? .j 3 p 5 4 4 1 E '; I a 6 4 I 4 I ? 7 [', + 1 - l e t + 1 M ' m ,t V! "; 6 F 1 i. n s l l t e = \\ i i= l '1 l Ia I r-l ~. ! b.' ' - = [ :; l> t I I h.,- i L er -? e t I^ i a I !] 1 '"tt y,y, rg 3 y I NMf1. y,. 3 'Y'd."M-'8t*W"a-6n,- -W8 N*-'mi'iee4e " ' " ' "="M -a-_ui.,_
mm__,
l 3 a 1.0 I_NTRODU CTION The - Institute of Nuclear Power Operations (!NPO) recently performed a simula tion (1) of the Ginna steam generata-tube rupture (SGTR) event of January 25, 1982 [2] using an interim version of tne RETRAH02/t!0002 code (3). In this report results are presented for a comparison of calculations utilizing RETRAN02/M0003, the mst recent release of the code, and the RETRAN02/ MOD 03A input deck furnished by IN?O with the actual event. In addi-ti on, parametric _ calculations were carried out varying the scenario of the actual event in terms of opera tor actions and mechanical failures. These I served to increase the understanding of the thermal hydraulic phenomena which cr. curred during the incident. The three parametrics performed were: a) A duplication of the event with a variation in the opera tor ac-tions. The latest Westinghouse operator guidelines for SGT_, E-3 R (July 5, 1982) were followed to determine thei r efficacy in pre-venting the ruptured generator from going soif d. b) The pressurizer PORY (pressure ope ated relief valve) is assumed to stick open during the operator's efforts to depressurize the prietary side and the downstream block valve is presumed to concurrently fail in the open position. This examines the ability of the SI (safety ~ i injection) to maintain the system in a stable condition. c) Finally, the cafety injection (SI) system was presumed to be left on beyond the point o' termination in the actual sequence of events. The quasi-steady cooldown rate in the aowncomer ob tained in this i parametric could be of significance to pressurized thermal shock problfas. Thi s document de tails the parametric calcula tions and dircusses the results obtained, as well as those calculations perfomed for comparison with 4 the INPO computations. Section 2.0 describes the INPO plant model _used, the various input modifications which had to be made for successful execution and the coding changes to the RETPAN02/M0003 source program required to correct for code deficiencies. The result of the comparison against the INP0 calcula-tions and concurrently the Ginna - da ta are described in Section 3.0 while the parametrics are presented in Section 4.0. Conclusions are drawn in Section g -5.0. l l L L
11 4 2.0 PLANT MODEL The Ginna SGTR event s tar tet wi th the plant at normal operating conditions with the primary side entirely in single phase, with the exception of the two-region pressurizer which was in thermodynamic equilibrium, and a secondary - side which was in two phase steaming off into the turbi ne. Upon tube rup ture, the priinary side commenced depressurization with a loss of inventory through the rupture. Flor through the rupture was choked. The secondary side of the SGB began to respond in the manner of a two region nonequilibrium pressurizer model (with a mixture level) as the steam generator began t6 fill up. Complica tions in the thermal hydrauli c response were introduced because of sy s tem feedback ef fects with turbine load reduc ti on, reactor scram and safety injection ini tia tion. However, until the primary system had depressurized such that the relatively stagnant upper head region reached the saturation temperature the entire primary loop was governed by single _ phase hydraulics. The outsurge from the two region pressurizer can be trea ted by a nonequilibrium pressurizer model. Bulk flashing such as tha t which ultimately cccurred in the upper head is also a phenomena simulated by nonequilibrium pressurizer models. Natural convection occurred on the primary side during the flashing period as the pumps coasted down and tie operators initiated manual depre ssuri za tion. The pressurizer rapidly refilled during the manual depressurization leading to possible nonequilibrium conditions. On the secondary side, the tubes did not uncover so dryout is not a concern-and the heat transfer is the _ noi mal two phase heat transfer. While pressure measurements were available on the secondary si de, data are lacking on the flow through the various valves as the ruptured B steam generator filled solid. (Data are limited in general as the Ginna plant is not instrumented as an experimental facility). With the filling of SGB,- single phase choked flow through the safety relief valves from the basically incompressible volume occurred. The primary side also tended to single phase again during this period of filling SGB as the head region steam bubble began to ollapse with - the continuation of SI flow. Heat losses and thermodynamic nonequilibrium had ~ to be considered during the bubble collapse. Finally, the termination of the SI did:not lead to the introduction of additional thermal hydraulic phenomena not previously discussed. i l l l l i
5 Plant models developed to simulate the Ginna SGTR event have to envelope - i all these varied thermal hydraulic phenomena. The Institute of Nuclear Power Opera tions developed such a plant model using engineering judgement where necessary to compensate for the limitations of the data availabili ty,- dis-cussed earlier, and has obtained reasonable agreement with the event af ter calibrations. 4 INPO provided RETRAN02/M0003A decks (t=0 and t=42.5 minute decks) and restart information which used the volume / junction nodalization shown in Fig. 2-1. Figure 2-2 presents the heat slab nodaliza tion used. The Fig. 2-1 nodalization was used from time = 0 to 42.5 minutes at which point renodali-zation was performed. The renodalization minly took the form of moving the l nonequilibrium pressurizer model from the disrupted steam generator (SGB) dome to the corresponding steam line. This apparently was done, at least in part, in order to avoid numerical problems with the nonequilibrium pressurizer model when complete filling occurrad. There was also a change in the volume split between volumes 19 and 70 when complete draining or filling took place, in sumary, at transient time equal to 42.5 minutes, the RETRAN model was revised by INPO to enable treating the SGB (steam generator 8) steam line as a non-equilibrium pressurizer volume during the period that it filled with liquid. This was considered the most realistic modeling available and permi ts selection of a spray option (with condensation calculated) and variation in the inter-region heat transfer coefficient threugh appropria*a input changes made in restart decks, it was also necessary to change the geometric model of the steam line to ensure that the liquid flowing from the steam generator, into the line, always entered the liquid region in order tha t excessive condensation due to homogeneous mixing of liquid and steam did not _ occur when the spray - option was turned off. This required placement of the junction connecting the steam generator and line at an artificially low elevation, but the effect of this on the calculation is judged to be acceptably small. The impetus for incorporating the non-equilibrium model of the steam line was - to provide ' added means for controlling the calculated pressure levels to match i measurements. For further details regarding the nodalization reference should be mde to the INPO draf t report (1). I l l 1 I e t s
l l = .-g 3 -( 1 ( 157 167 L., r r c [',,h. s g 4 s. .2 c u a,, T 'i' t C di O 't1 rv uw uw N Avs. To y s[e ~~ u / \\, f., h g s e e. ,j i F.edester 4J 44 74 33 s2 q l u 9 2* ar - -'f 'f~ f -.- es ,,._ j.., f f 'f f ,,r, r,) 1 42 45 .4 '*-'-.25 . 37 d r-g f ,N, /jj f T,, at ya / j j N yg. ",8 C ' f Tube er.ek.e.J/ i ,{, 'er e2 .j (,-- l.-g..
- l. - 2 T Leap "A* Loop a
'l 'l + i j e ,e s ~v ) .m e o (- j m( Q} q " ~ > k d' " c%,,, j ~ .....,i-1 t "" e.m s.1.a, -- tn 4 l RETR AN Model Votumes and Junctions rigure 2-1 a. -2 ~
e 7 m 6 m ~),f 4 .) [- _ _ 3 -
- c 3
I ,k I .'b' g ~ ~ E g 4, r--%- 4 g .R . m n g-, f w ~ j' 4 = N ~ u 2 L 3 w_- L Q Q ---.K = i g h ^ e. = .q ; .x..m y id m h e ~ ~ __~ +M, y .g. g = ,fy -m. x t T3 2 L i - =-- m _u ~ ts-e 4 Y Nd ?. e. o o ~ m T. -. c_.,. J E g 2 I ! - :$2 - l ,e
- m ea *
- . V4.
'm ~ 0 _ '- M -q t u .ef a
- lW v 3
== K : A. 0 5,i o'
- .= *
%n .m., <(g \\w, *. ' d j Ch ,/ L A s_ }.% gg..y 4- [.- 2Q ~, j T a g ,A x n ~L D Li 'J -- m n j s ..)' - - , +, C' ' _,.- __L { t \\ T r s l T $;s M -N 1 N =- i )r/ a r e m =_
8 The a f oremen ti oned renodali z a ti on, and other INPO revisions, cannot be ^ accomplished in a restart deck. Consequently, an entire new input deck (t= 42.5 mins deck) was required, i ni ti ali zed to non-s teady condi ti ons. These ini ti al conditions were the end-point values of plant parameters calcula ted b for the initial portion of transient (t=0 deck). It is noted that RETRAN does not permi t supplying inputs for all required variables, e.g., heat conductor temperatures, surf ace heat fluxes and values of slip in junctions having two- )hase ficw. Such quantitles are normally computed for steady-state conditions by i tt -a ti on. However, for ini tialization in a dynamic condi tion, the s teady- [ state ini tialization option is bypassed (word number 25, JSST, on card number 7 F 01000) is set equal to 1) and quantities tha-e are not input are compu ted without a stea dy s ta te search. This approach is reasonable if time deriva-tives in the relevant equations are acceptably small. This apparently is true j for tris application since there is no evidence of significant discontinuities in calculated results at the time of transition to the new input deck. 3 E In order to m tch the masured da ta with the limi ted ins truaen ta ti on ca pabili ty available INPO had to undertake f airly extensive calibration ef-r forts. Among the more significant adjust;nents mde are renormali za tion s of upper head " valve" areas, steam genera tor 8 S/R valve areas, MSIV flows and 9 PORV areas. The INPO draf t document provides addi tional information [1]. In = general the RETRAN02/ MOD 03A decks obtained f rom INPS, once modi fications were ~ nade for the RETRAN02/ MOD 03 version, were used "as is" to produce the results 1 documen ted in this report. Minor al tera ti ons had to be made in order to 5 af fect the changes required for the three parametrics and these are discussed in the respecti ve sections. More extensive alterations had to be made to r vampensa te for code mdifications to ob tai n the comparison resul ts wi th the INPO compu ta ti ons for the actual Ginna event. These Input deck a l te ra-g i tions/ code mdel mcdifications are listed below. L 5 For the t=0 deck: T. a) the input for the upper head heat slabs nad to be altered to accom-g moda te for the changes to the non-equilibriun pressurizer heat loss model; E B-E
9 9 b) the steam generator done to bundle junction had to be forced into a zero slip calculation to avoid non-physical results. This w&s the original M0003A option; c) while it proved infeasible to compensate for the change in bubble rise model i t should be no ted that M0003 normalizes the Wilson i bubble rite correlation to the rise velocity computed in the steady i state initializer. This was not done in M0003A. fer the t=42.5 minutes deck: d) the Wilson bubble rise model yielded a bubble velocity of zero ) except for the initial value; e) job failures occurred when the calculation called for closure of a previously opened SRY; f) non-physical results were calculated at the time of SI termina tion ( two phase flow wi th slip in a juncti on connec ted to a non-equiltbrium volume). Each of these problems were resolved af ter consultation with El person-nel. Scrae were circumvented by changes to the input deck whereas others involved FORTRAN source changes. In addi tion to these problems, M0003 stability required smaller integration time steps than those used in the !NPO calculations (M0003A). This increased the demand on ccmputer time and caused a need for more restarts to complete these calculations. I b t 4 g--,.- n v ., --., ~. - -.. -,, -. -.n,,--n- .-n, ,,n,. -..,--..n,. a.
_______.~___.m_ 10 3.0 1NPO COMP ARISON in order to simulate the neasured system response during the Ginna SGTR event, INPO [1] divided the transient into two parts: one from initiation of tube rupture to conmencement of mnual depressurization (t=42.5 minutes); the second from t=42.5 minutes to shortly af ter Si termination at which point the INPO computation es terminated. Discussion of the ANL comparison with INPO is therefore similarly divided into two parts. Section 3.1 details the comparison from t=0 to t= 4 2. 5 minutes and Section 3.2 concludes wi th the latter part of the transient from t=42.5 minutes to SI termination. 3.1 t=0 Deck During the period from t=0. the point at sich tube rupture occurs, to t=42.5 minutes when operator-initiated depressurization using the pressur-izer PORY commences, the reactor system goes through a turbire load reduction phase, reactor scram and a tripping in of the safety injection (SI). Concur-rently the condenser-dump valves are cycled, the various feedwater pumps are manually controlled and the isolation procedures are followed by the operators in grder to minimize dose rates and to reestablish control. Reference should be made to the INPO draft report for a detailed chronicle of the actual Ginna SGTR-event. The t=0 deck takes the event frcm the initiation of tube rupture to the time when operator initiated depressurization through the pressurizer PORY at - t=42.5 minutes occurs. At this point the problem is renodalized and calcu-lation continued with the t=42.5 minutes deck, in addition to the splicing of results made necessary by this renodalization procedure at 42.5 minutes there is an additional splice necessitated by INP0's further recalibrattut during the time period 112-180 seconds. With the dif ferences in code models between RETRAN02/M0003 A, the code version used by INPO, and REThM02/M0003, - the version utilized by ANL to produce the results presented in this report, it can _ be seen that there are possibilities for discontinuities at the times of 112 seconds, 180 seconds and 42.5 minutes. While some smoothing could be rationalized and was indeed used at these junctures, the result do show some discontinuities at these points. These will be discussed in perspective. The
11 i perspective of this document is to concentrate on the dif ferences between the - INPO calculations and the ANL-computations and to understand them in terms of modifications in code models between the two code versions and also of sensi- -tivities in the INP0 calibration to the data from the original event. Figures 3-1 to 3-26* present the comparison between the INPO and ANL i results. This set of graphs represents the entire set presented in the main text of the INPO draf t report. The ANL curves have been traced onto the INPO fi g ure s. Reasonable agreement has, in general, been achieved between the INPO RETRAN02/M0003A calculaticn and the ANL RETRAN02/M0003 compu ta tion. Where differences have arisen they can be attributed to three or four modifications as discussed in subsequent paragraphs. In order to compute the narrow range pressurizer level adjusted for . instrument error at nonsaturation conditions the input deck had to be altered to incorporate a stand alone control block madei. An error was discovered in the transcription process which affected the adjusted level. As the control block is a stand alone model utilizing input thermal / hydraulic (T/H) conditions from the main RETRAN T/H calculation it is completely ignored by the main computation. However, Fig. 3-2 for the pressurizer level shows that the erTor leads to an underprediction of the instrument adjusted level. In RETRAN02/ MOD 03A the nonequilibrium pressurizer model had no heat loss associa ted wi th the volume boundary. Two-si ded heat slabs could be a ttached to the nonequilibrium volume by the user but the volume does not "see" the heat slabs. The heat slabs however do "see" the volume. This model was altered in M0003 with the slabs and volume interchanging heat in an energy conservative manner. However, the use of an adiabatic boundary condition is now necessary on the non-pressurizer _ volume side of the heat slabs with Mod 03. These alterations imply that the heat transfer in the upper head region, which is modelled by INPO using the non-equilibrium pressurizer model, particularly that incurred by the "fictiticus" conductor (slab 20) cannot be. dup 1'i ca ted by the ANL M0003 calculation. Consequences of thi s model 'These Figures are grouped at the end of Chapter 3. s ,_y r.m
4 12 \\ l modification can be seen in Figs. 3-1 (RCS pressure), Fig. 3-5 (vessel upper head temperatures) and Fig. 3-14 (vessel upper head steam volume). There are certainly effects on other parameters but those should be of lesser importance particularly for the parameters on the secondary side. Heat losse1 from the pressurizer per se are treated through the use of a control block and there I are no heat slabs associated with the steam generator domes. Thus even though the non-equilibrium pressurizer model is used for the pressurizer and the steam generator domes this model alteration should not directly affect those volumes. A modi fication was rude to the bubble rise model between the M0003A and the M0003 code versions. The M0003 model now normalizes the bubble rise velocity computed by the Wilson bubble rise correlation to the veloci ty com-puted by the initializer to obtain steady state. This may sound inconsequen-tia as the Wilson correlation is only used in the nonequilibrium pressurizer model and pressurizers are normally initialized wi th no bubbles.
- However, INPO chose to use the non-equilibrium pressurizer model in the steam generator domes.
Upon switching out of the automatic initializer and proceeding into the transient M0003A would use a different bubble rise velocity from M0003. The imglications of this difference is an alteration in steam generator level behavior. Figures 3-11 (A SG water level), Fig. 3-12 (B SG water level) and the corresponding figure for the first 20 minutes of the transient, Fig. 3-23, show the difference in ini tial swell which can be attributed to this nodel modi fica tion. The INPO calibra tion of steam fl ow, e tc., has to be recali-bra ted to da ta to account for the alteration in code model. This change should also affect steam generator pressures i Finally, the ef fect of the splicing discussed earlier can be seen in Figs, 3-14, and 3-24 to 3-26. To reiterate however, in general the ANL compu-ta tions using RETRAN02/M0003 compare reasonably wi th the INP0 calculation using RETRAN02/M0003A. Where dif ferer.ces do occur at'cribution can be made to model alterations and calibration sensitivities. I m-m
13 3.2 t=42 mins Deck This portion of the calculation spans the transient time from 42.5 to 80.0 minutes af ter tube rupture. These results are also depicted in Figs. 1 through 3-20 with the ini tial portion discussed earlier. This latter portion of the Ginna calculation begins with initial opening of the pressur-izer PORY to accelerate RCS depressurization and concludes shortly af ter 51 termina tion. Throughout this period, decay heat removal and RCS cooldown occurred by continued injection of the relatively cold SI water and feed and bleed operation of the intact steam generator (SGA) as modelled in the input decks. Since there was no reasure of the SG steam and feedwater flows, they were adjusted in-the model to provide correspondence with measured pressure and water level data. As the aforemen tioned figures show, there is excellent agreement be tween ANL's calculated results and those obtained by INP0. The plots for primary side parameters overlay nearly exactly, except for some of the fine structure in the oscillatory behavior of certain parameters. In addi tion to the agreement shown for RCS tempera tures, pressures and flows, there is also essential overlap in the curves of tube rupture flow and upper head s team formation for the two sets of calculations (Figs. 3-13 and 3-14). The only differences of note, albeit small, are those in parameters calcula ted for the faul ted steam genera tor (SGB). INPO, in thei r cal cu-l a ti on s, used a complicated program for the area of tho' SGB sa fe ty relief valve (SRV) to control calculated pressure to ma tch crasured da ta. To do this, several res tarts were made wherein both open and close pressure dependent trip setpoints and open and close valve area table entries were changed. E f fecti vely, this process verted the valve area with time, opening or closing it contingent on calcu!ated pressure levels. The trip setpoints were not the normal plant values, but were selected to be close to the running pressure levels as indicated by data.- Thus, small differences in calculated l pressure levels caused the valve behavior to be di f ferent in the ANL calcu-l lations. I 8 l l l l r m ~
14 The first departure from INP0's results is evident in Fig. 3-2 which shows-three openings and closings of the SGB SRY in the time period of ~ 51 to 5$ minutes whereas only a single cycle was calculated by INPO. This is also manifested in the sawtooth nature of the ANL curve of SGB pressure (Fig. 3-10) during this time period. These additional openings of the relief valve released more liquid from the system to the atmosphere causing a slight delay (~1.4 minutes) in the calculated time to completely fill the B steam line with liquid as shown in Fig. 3-19. After the secon!ary side became solid, the calculated SGB pressure levels becare very sensitive to the flow resistance out of the system at the SRV. In the INPO calculations, the valve area, and hence its resistance, was varied of ten using pressure trips and area tables in the manner described above, attempting to ma tch pressure data. Using INP0's trip setpoints and valve area tables in the ANL calculations proved unworkable because the timing for switching between valve opening and closing redes is critical to obtaining the proper areas versus time. This timing of actuating the op,en/close trips is not deducible from the reported results and even small calculated pressure differences soon resulted in erroneous trip times and attendant valve area values, and subsequent large deviations in pressure levels. It was necessary, therefore, to change the method of programing the valve area with time in the ANL calculations. The INPO curve of valve area shown in Fig. 3 20 was used to derive numerical values which were entered in an appropriate RETRAN table and 55.42 minutes all areas were the trips were changed such tha t af ter t = obtained from that table, Also, the time at which the sudden large - area increase occurs was delayed to be coir.cident with the later time to fill the steam line wi tn liquid as calcula ted by ANL. This approach, while not precise, yields levels of agreement with INP0's results considered adequate as examplified by the comparisons of SGB pressure levels and SRV-flow rates shown in Figs. 3-10 and 3-20 respectively. In sumary, the results obtained by ANL for this phase of the Ginna event using RETRAN02 Mod 03.also show excellent agreement with and confirm INP0's earlier calculations. Although some dif ferences were encountered, they are not sufficient to negate this conclusion. Also, the available measured da ta is represented qui te well by the calcula ted sy s tem responses lending i -,.,n----,, e-, ., - -.. ~ <, .,,-. +. , - - - ~
15 credence to-INPO's overall conclusions regarding the transient plant status following the actual steam generator tube rupture event. e e
e -e a g a y .a 5 5 5. 3 . g 5 5 5 5. 5 Reactor Trip (003:12); Safety injection (0.03:19.6) 3rd PORV Opening (0 43 44)- j', ~ i 4th PORV Opening (0.44:10.1) 2000 POAV Block Valve Fully Closed (0:45.00) Calculation (RETRAfJ Volume 61) ~ ' Data (Frorn~A* Holleg) _..- A N L ~ 1500 i f, c.. 5 - ",T.I. ~ O .)1 v-I l i ct gg g. C: h1000 4 .i, ~ o mw I c ~ M O C
- B"SG Overfills (0:43:10) 500 Seiety injection Stopped (1:1200) g o.J_
0 '10 20 30 40 50 60 70 80 ELAPSED TIME, MINUTES RCS Pressure vs. Time Figure 3 - ---- - =
p too -. ,= _ _ - l8 / 90-l / i g a I 83 - Z i .I ~
- l W
l 3 70 Rapidincreasein Lecel 1 Due To Steam FormaD>n in T I i ~ ReactorVesselHead AndTo l ~ . 6 y 60 Increasing Safetyinjection Flow - 1: 3 l[ Catculation (RETilAN Volume 22) - Calculation Cortected For y 53 l . Instrument Error At. j Non-S.sturation Conditions (See l AppenditD Section45) e 0 C f 40 Data c 3: o _ _ _ ANL e 30 - / c i f 1 ~ 20 \\. ~ s'. to [,/j, ~ / /.,). ~ \\ '.... 0 0 10 20 30 ~ 40 50 60 70 80 ELAPSED TIME, MINUTES 4 Pressurider Levelvs. Time Mgure 3-2 m Yr ayr e -
'm a 1 4 . Rapid in-Surge Of Continued Risein Pressurizer 750 Subcooled Water initially, Water Level Cate.es Compression t Cools And CondensesSteam AndHeatingOf SteamRegion - Region f ~ N h ti. \\ TTI II* ua sessonsocism 7 - :- he ~ ' Th% 600 f w., W - i 4 = tu ~ 1 450 a-g ,t b F = i u N__ m Catet, lated Pressurizer Water Temperature ' ] g RETRAN Volume 22 rn v) --- Calculated Pressurizn Steam Temperature W ~ 4 8 n ~ = E Pressurizer Water Temperature Data a' Pressurizer SteamTemperature Data 150 -. _ ANL i t 4, o 0 10 20 30 40 50 60 70 80 l 1 ELAPSED TIME. MINUTES r [ r 1 'PressurizerTemperatures vs. Time Figure 3-3 4 i
750 g 3 g a g g g g g i ,y I l I ReactorTrip 003:12) t E00 '91 I g f---~~~ _m \\_ _,- ~
- '~\\
/ \\ \\ \\./ \\ / ssa ...W. 's / \\ F s. s,' s, s m i N_ w f C \\ D k . N. % e Wc. 500 sy .*N 'N. p ~ e g E d s <se ~ O Calculated Core Exit Temperature (RETRAf 4 Volume 13) Core Exit Saturallon Temperature Catculated From RCS Pressure Data Core Exit Temperature Data (4verage Of Five Cora Exit Thermocouples) 400
A N L L
f t f f f ,f f 353 i f 0 10 20 30 40 50 EO 70 80 ELAPSED TIME, Mir4UTES Core Exit Temperatures vs. Time Figure 34
e ~ mo s a .\\ ~ e r ~ u g 0 i
- w/*
,\\ 8 F .*1:*.* N },.
- i: 5 1:-
'0 7, 'w N e e N.,* m h i 0 T 6 s ~ v s e r ~ ~ u t 0 a 5 S r E e T p. \\. IN m U l
- * * :. :f i i.-
I:: : :-
- . 5
e.
- s..
M T. 3 E d io T e 0 M a 4 n I g D H e E r I R r r t S e r e e e o P p p p fpp A p p p up ) L U U 0 cu n n n cS 0 E U 2e or 3 i i i e e em me e l m~ r r ) r p u u9 ulu rep o s t t t a a1 a hU s r r e rV e e Ta ep pm pN dv e m) mlu mA ao V e9 eo eR eb ~ T1 TV TT HA r y4 x te lo LT Lo E r s o e idN d ee 0 m m iR t uA u( ph 2 c a u q pc S iqR in Un s V 6 i I d d dg mB e eN eR ee o1 R _v N t ( t taA luo lu e aR rd a n F e R r ( t ) u g^s_ la aae cT lae a o aol N ci c w ct L E g t l a R C( CR CL DtP A 0 1 9.y.., ~ a \\ ~ 0 m H 0 a O 0 0 0 e E 4 2 s 6 s S 3 5 S u C dWcD t<cwc JN o<m A cu1c.D gWgu> ChO6c r l
- l;tj,fjll.
i: 600 I s = = = = r I 9, i ? I 500 - s -l .m [ \\ l T,,, Data Not Available I 1:- Af ter 044.00 Since I b l%; I i Instrumentationisinaccurate l lb l \\ l y400 ForTemperatures Less Than 540*F- \\, " gh I. ~ -) g l w H g 3 0,, - I n E 300 F to U t l-0 i g Calculation (Average Of Temperatures ~ tu> in RETRAN Volumes i 51,61.54. And 64) [ < 200 b m --- A N L e. l 4 1 100 d 0 0 10: 20 30 40 50 60 70 40 i i . ELAPSED TIME,MINOTES i RCS AverageTemperaturevs. Time Figure 34 1
.600 = Safetyinjection Started (0.03:19.6) t lf .. N s i safetyinlection stopped (1:12.do) i 500 'g. t ~~ sm u. t_ -~ w u I 400 C O i p - 3 c w it ~ m 300 F 0-i w Calcuta!!on(RETRAN Volunv.4) t O f d Data (Upstream Of Safety [ O 200 Injection Nonle) l 4 --- A N L l c. .I O O l .J 100 i ? i ) i 0 l 0 10 23 30 40 50 60 70 80 f i l ELAPSEDTIME, MINUTES j Loop "A" Cold Leg Temperature vs. Time Figure 3-7 j i
23 g 4 3 t i e EE gn de 3 W w . e c. a -- n eEm2 m " O 4 2 O.4 5 O B. >o Em. o >o o. s !? e i,e,v&.=8:g X e a, H W-W- _ _ _C...- .~..m% ary 8 9._,c: gs g am-w_-.. ._.gv, A_w C g a m?~m o a, m-*- ,.w. u~x.,.a,: u ~ a n a M 3 a u -== -n - n - g = a: b^ _..*"'s Q1sm. o o a ~ o
- m..-- _-- s un.-
Q t l .c m~~.~%..~e. -.,7 2 c. S l see m ~("% -m I %m4 g e b**5 % ~ ~-- ~... Q .4 m. w- ~ e e o f~io. w o. w ~-: m. -. __. D M ~ -- ufR-: s W m m -,...m.._ C h 2., C2 ^ - - - Q. -. _-- - _ %_.. %. ~.__--,, - - 2: c: u, &m_ as ~q ---m?% % o. W ~~- 2 h _N%~~ d. [," U fa C) h .. h, -r [a :-A' % O d w " - - = * *
- r.
M
- C f *% ~~, =* *"
~ a ,,5 # O o, 3 U f' ', W j/ 2 CC a ,/ C ..~ C m g C 8 \\ L 8 I m a i 1 n 3 .l /1 . // 05 C f /. ~ y( 8 o ? y . t !l ! l ;i[ M ^ ~ 2 1-2 r 0 u I 4 8 s T 5 P ) 0 I i 8 7 8e mo d r c e V it .a N e t t m m! a A s a D R i nu p T I 0 l I T. i at e e !. o E 6 c s vt rT A. i e S T R R l0 w s ( le0or0o n v v0 Te 0 e t l b l e 0 t 3esa3B l e0uW0t A ev g nAu 6 a L t nt nt F a a N e aAi t tn e n Rseor o2 C D A L c 0 S oW I I wtaCei1 5 E e t Pat or u T g l c ue0i t t t v bf U n a a e.0 dnAO a 1 NSL J ( I j f i f. R t E w 1 0 L o I I 4 I r T r D a E N S r 1 P o A t L a E r 0 4, I 3 en e G V [:l m tuf d oO v b a p ef Ao e 4. LO! 0 iT~ 2 I t r entoSe S t v ai t o Wtanb nicAs B ce / int 1 dr e u aee b r t ml Pe u a fT ia1 8G f UD1 3S 0 / t I 1 / s s. ~ ** g r g,'* a
- t g,- l-L 0
m 0 0 o 0 0 0 0 0 o 0 9 8-y 6 5 A - 3 2 t F2WOCW1 JW>3 ca,Qg uo5C 3OCC< ' cop <CW $ * <fv) ya s .:!}i l1li !lj;:, .il,:l!! ii!
L t 1000 a a i e a e a a i a a a a a PORV Block Valve Closed (0.45 00) Safety injectba (0.03.19.6) 200 i Safety injection Stopped - I I PORV Openings Start (0.4230) ~ 2c. O f,00 -K ui l gMso-c>"C' p.::: O.. v.. w g >~ NNL { ~ { J.g'.. S L1 T . f-. N' l O 400 q .J 14 a I U w I H< J DO 200 J
- l-
~ a1 J 0.0 Safety injection Pumps (RETRAN Junctions 154 and 254) dt ANL e Steam Generator Tube Leakage (RETR AN Junction 142) T Charging Pumps (RETRAN Junct!cn 264) -200 ? O 10 20 30 40 50 60 70 80 ELAPSED TifdE, MirJuTES ~ Steam G eneratorTube Rupture, Safety Injection, And Charging Flow es. Time Figure 3-13
..m _. _.... _... _ _ _.. _. -... _ __ _. _ _ _ _ _ _ _. -. _. _. _ _. _ _ _ _ _ _. _. _. _ _... _... 4 to Y ' ' - ~ E I 0-g n 5 v , ' ^. ~ ~ f -- g 1 n.$ s 3 / y wm i - t / 4 a U: / R 8 E IC / \\ t t g / e - l-D d g az o 8 . il / m. / e / g E ?. Q ,, 2 S or y c. E I m e -e e g-c.* W i F il 1 2 t t y p: e o 8
- x m
i m o } g-D. o a l g
- O 1
+ 3 i m 0 m E* \\ oU i u ) S o 8 \\ E 1 4 1 a h E.' N L c L S e. 7 'k r. 1 u e p l -.- 1 - _ _ I,,, 8 G S- ?,. R 8 3 S U 8 c e i. I r.Ld '3WO10A WVals OV3H U3ddn 13SSB A UO10V38 l-s r W 'r d e u -v<- o e-r-w + e w et w oe, e--.* - e e rs= m e--or,m-r--*-----E.*-6-.-,---e -+e--- r w +-n-m-- -.,-ri rw-a,-e.-e,---e .-e-g -,nn-e-----r,er .--*v-e-r nme ev. r
30 go'1d 11nd dO ltJUDU3d 'S31Vtf Mold dOO7 It4V lOOD UO10Vau O O O g s d = y g ,-__7 7,7 7, I n f e i g t; %'d','>'*a i. r3 G u II. \\.{;, 4 N, -- n (+% ~--f As7 -r.d*2^* *~ ~~uu O.p. a ,a..q c a --rwc~J.,., ea Db j >. WTir%w mt - < d. -p %w:. 5."~ xx %.&, ws 5'* Wu ?,.?.h 48 -a - w . g3 v$. Y.,, y*'* s w_,p.<,. Ob M<v~ .s., c., o ~ ~ w N.: D ; n .5 a i 's E 1 ~ %._n < - p.
- e. a.
=m4 l, .= %..,n#viw-O 8 .J .gg - f M = s.s p_, --x;.rg;-r. ;.> w m l ll ,g3_ ' g e N._G +se-y S I ( I s ~~ 5 we... .ar-,-a::ca.%r.-l ci.% mw yg s p O N o M .J. = :: =.=t: "" Lt. 0
- 3 a,
m.u. b?' ~~~'* M. uf O ,.dC.'"'"*;Fgm..m j O
==" 7 (_ . M. Pd o c m m v1 o n. O 3.o ~ o w L.o / ON I / O / o CC a / ~ / </ /. lg I f 1 I 8 i i i f 8 8., 8 y. 0 3 8 o 4 c,4 [ ODG/WO1'U31VU Mold dOO1 DJV7000 UO10VDU z 1 l
31 2 g m-y th 2 5,1 e 8
- xl
~ %== 4 6 i 3 _ r_w%m .m- -%we-a e - . C W-~~.4 w *m-t.ce w~~ l s R
- "AiTIGEC l
g -a.:..- =. tem. N w-=_7? C e-5 .9 g m u.:-- e F. ' O "C h" c> S p-wass ~ -a=. h,-.-- g - GE.e" b.t== n r~ g j N M -- :;vr CC --- wm r-8 Q g l m h.. -. %,s w 2sy~ ~ H ($ W-= ....-..a-p, u. _~ 2
- w. w y
e.t_r%.. 2 ~wur ;. _ . i g C' ~ q 3 a - p w - x....~. ~::: y..' " p e to i O en w e; ~ m y 'Q. h. O W 8 o me l CC z Oc a O-or< o OJ' / .o I i f f f g o o o o o o 3 o, b3 o a 6 w e ODS/W0l '31VU Mold 197NI 13SSBA UO10V3U 0 dOO7 Y e --ew-w. r e w mr -*-x -, ar ,.e,* n .-rrc-%,e -,--,4 -w r .w- ~~.-e-a s--i+, -m -m -.-m-*---.r---+sm-+ +r -4 e-s si+ - -
32 p i T-- ---. th 4 5 2 c3 4 [C + l' f ^ R ) S e / E F. 2 y m e o u g ) s
u e cs F' O C2. .V E 4 / za p W O H r. w yy n u> cn / a:. m a o ~ / e pe 3J + a n m o 8 8 ,/ Z W=j ~d m 8o o / kb 4 H LL l 8 8. s. a / t<l / l a / 4 f Y I i t t t t 1 i t i O E 8 8 8.n 3 3, w a d 'S3Hn1VU3dW3103'110H SOU w y ,-~ - w w-y. --a- ,.e-,.- y m - - - -r v -+,cs- ,-m,.-e y.e w .-,--.me, --e .-w,ww -e ,-*e---* =< -- --- ----.
i f 33 9 e i i i i i i i i i is 6 s j 2 a: ,f m a: e.. 3{ 2 e I E 8 fl l c.. l a .7 I a g be g a E g g. c. gH ui (N c.". a ~ s E WJ 5 O. CN o o w c j) M o. -O U-Q g w -/ op m /
- e s
I O l a y ~ u / c3 / CC .l 5! 4 s ~ t ' i i i g 3-8 8, f,,' S S g i o n y n n do 'UUn1VU3dVl]1 U3VJOONMOG lussa A UOloyau i I-i .u- .w -
34 m i e e ~ 3 I w s-w%' i L e i 3 0 u F b C I i', cc /r - i .f 19 4 W j 3' 3E O-i 9f a (_' j3 ~ ^ a d3% F ~ .t 3 N%,%*% C, _6 w N%,%. N p of A('% 3 w c) __N W s ~ D D ~ ~:: C d> W d. c, o M m c3 g t-O L' 4 w g E ) un g a 3: 'j J J o ~6 F n w b d m., c1 E 45 "g u. CD ~ j u., 'a g Q., 2 M a = G. g o a m 5 m 9O _e n O v.> n 96 O 4 l r .__1 o q._ P.3 8 h b e h eld '3'.'Jn10 A U31V/A Df 31'l kWI?1S Ga + E l 2 i L
+ a 4 s a a sgL 4 e a a a a 4 e 4 L gy m 1. e h b 100 m 100 .; y a m j Calctdationitu:icates'B'S! cam- - Safety l C Lis.oNisCcnT:pW.alyWah [ g l$chon A Watet(258.38) gl Stopped W f5, ii:1 M ) i .i '[ 80 $ < B0 - I [ k{ i 7 S (a i r 1 O o 3 ~ j u. i g g a I a W e fi ~. 60 b ~ '. O I $ 6C l g l l b I {. ft 3 T } il 2 + l w e w i i-e l 40 c C 40 I O E 9 . I 5 P l Q i n 1 1 C ~. ,3 Cattaa!&1 FP 9 Flate (RETRAN Jur.ction 167)[- w j Ej - Calculated Row A ea (RETRAN JuncUcc116Mf k ]- l 0 I I h 3 h' }t J. } %\\s (. $ 23 AML I I .li 20 i 1 5 ( g N ?* \\ U f ~; l ,j f f.! C J h.a ( i 0 f e _.__t t* f _1 F i e a m g f i G 10 20 33 4G 50 6C 70 EQ i L i, ELAPSED TIME, MINUTES l "B" Steam Generator Safety Valve Flow Rate And Flow Area vs. Time agure >20 l r i 2 r ~. _.. =,.
r 36 p i A M, y@ h b FI / ./ U / Ei, ,o 3 E ./ a a,e ,a / 0 0 4 /
- p'
.p .Q. .6 l- - c g f O w / yo l s - g. ,/ ~ = } w. D p: m .ai > l .w e/ O Y g n ~ e m ri. g 3 as s. W C / I 3 g N m tU ~ V) e ,~ t / 4 / k;1 .- w L } f t sp'j { A [ %A4 it N / l I l. 'I,' \\. 1 i _. o 1 n 1 e a S 8 8 8 8 8 8 o = ~ za m f g 010tf '3110GS3thi DNn VJV] LG N.
r-Tusbino Load Rcdocien (00121.7) { - 8 Corxlenser Dump vat <es Open (0 02 00) 4 Condenser Dumo V.stves Closed (002 45) Reactor Trip and TostJne Trip (0 03.12 and 003.13.1) i 1100 l - Valves To Steam T ur bine e =x. Feedwater Pump Open (0 04:*5) 2 Ccodenser Dump Valves C'ose (0.05 00) 2 Condenser DumpVatves Cicse (00530) 2 Condenser DumpValves ManuanyOpened and Closed 1000 (0 13.00-0 14.00) l -B ustVClosed(0:15Es .,f~ %,N'% w. a ,\\, e% 500 ) N L C-N. y ( b 'N % ~ ~ Lo sco y e'. 'c. U. r y J 103 2 ~ ~ QrC a c.2 Eco Os% d2tM(RETRANVolume 68) sco Data 'ANL ~ to 15 5 0 ~ EdSSED TIME,MfNUTES "B" Steam Line Pressure vs. Time (0-20 Minutes) agw.s-22 I
g s s g,\\ g s \\. Torbene Load Rcdoction'0 01273) e/ e { B Condenser DumpV wes Open (00200) /~] j Reactor Trep (00312). Turbine Trip (003:13.1) j' { Mam Feaduater Isolatsort (00322) f' ' / j ~ 1 A* and'C'Motm Driven Aun. Feedwates Pumps Started (Om494G3.51) / 7C [ h/ f ~ --: /,. - Steam Turbine Aum. Feedwater Pump Started (004:15) \\ (L ~B Motor Driven f
- f 1,j h./
4 : Aunittary Feedwater /.- EO t-Pump Stopped i G j (0.07.00) / a o 5 L.~. ' l I ~ lI c =- 16 7 s I a g 7
- a
% d'r 1 I ). o gl f a e / ~ S 5 40 5 5 / a l /. / e / f ~ lfi 5 / G f f o 30 I / /=
- i<
-0 u .. / A*SG Calcuhtlon (RETRAN Votume 76)
- .a t}
../ r-g / [ ~B'SG Calculation (RETRAN Volume 'i6) ~~ 23 / ~A* SG Data f jj! a ,,,y j / , s SG Data ' *438 / 10 "l ~ I / l W '. I O 5 10 15 20 ELAPSED TIME,7.11NUTES "A" And"B" Steam GeneratorLevelsvs. Time (0-20 Minutes) Figure J-23 l J L_._____
' ~~ 0 39 Nb g e 5 1 B [ m 3 ll l a e t d C S i C 5.f g e = 2 n w 3 3 5 .J aa = l .{ ~~ Q) 4 e '2 ~ I 4 m O l w N l H e a o ~s l u u a 3 d E ~ H P. I o m w a 9 9 D o a. u ~ E:- M' ~ b*j W E L Q w I 9 o, - c $ {f ~ m [ U kJ i W e t 4 g ;,-- e i e ,e O ^ l O O b O+, 8 a e e ~ s ~ ~ OlSd '3UnSSJud SOU
l 3 a 3 1 - 4 5 5 6 ~ 8 Condenser Dumpva!ves-Calculation (RETRAt4 Volume 22) 4 Turbino Load Heduction Begins-Data (0 01:27.7) g ANL 2200 t i 4 O t c. ReactorTrip(0212) ' N, i ti 2100 e l 3 \\ ~ I L'h \\ u 8 C r \\ C s \\ b.
- 4 l
s. e D N ~ m e 2000 3 u k i c C-k. 1 I 4 I i i t l 33co O 1 2 3 4 i ELAPSED TIME, MINUTES [ i l Pressurizer Pressure vs. Time. (0 -4 Minutes) Figure 125 i
C i 6 a a a 6 s 580 j Turbtne Load Reduction Begins (001W.7) 8 Coodenser DumpValves Open i 100 Due Toincreasing T. (0.02.00) s a I .\\m 'i j
- C
'T P N \\ a e s SO ~ g j . x, N e N,; I 8 u. g a ^ Era e j B D3 \\ h c i H 8 I 3 j Z 11 F U S U l; i c c 70 {. m U 1 Q. d 2 W O i m F t ta l O m n t i 1 4 C-Calculated Core Power (RETRAF4 Vok:me 1) y c w h Calcu!ated RCS Average Ternperature (Average of ' SM k m I o Temperatures !n RETRAtJ Volumes 51.61,54 and 58) e 1 50 k E3 m Core Power Data ~ 1 C l a RCS Avera2eTemperature Data I I i i 43 ANL .L s 3 Y y l f k. ? ? 0 1 2, 3 4 i i ELAPSED TIME, MINUTES t
. ~ - _ 42 4.0 PARAME1RICS ) Three pa-ametric analyses were conducted in order to further clarify the understanding vf the thermal hydraulic phenoment which occurred during the actual sequence of events. As discussed earlier these are: simulation of the j most recent Westinghouse operator guidelines; further mechanical f ailure as i embodied in the ossumption of a non-responding block valve; and a variation in operator action with continuation of the $1 flow. It should be understood that for the most part the tables / control blocks, mde necessary by the INPO calibration to the data obtained from the actual event, were not altered for these parametrics. %ction 4.1 details the opera tor guidelines computation while Sections 4.2 and 4.3 describe the stuck open PORY and $1 con tinua tion parametrics, respectively. 4.1 Westinghouse Operator Guide)ines r Upon comparison of the actual event sequence of the Ginna SGTR incident to the latest Westinghouse Operator Guidelines for SGTR (E-3, July 5, 1982) it is to be concluded that the operator actions conformed with the E-3 guidelines. However, irrplementation of certain procedures were delayed enough to f[ gate actions takett to prevent the ruptured steam generator from going solid. For this parametric the tevent was resnalyzed using the INFO t=0 deck wi th operator initiated PORY depressurization roved up, by approximately 20 minutes, to 25 minutes af ter initiation of the transient. Judging from the i chronicle of the Ginna eveut, this scenario should provide sufficient time for j operator action and, in addition, all Westinghouse guideline conditions for depressurization had been met at this time. Modifications were rude to the deck to follow the latest Westinghouse operator guideline from the point of pressurizer PORY depressurization and continuing on until the calculation was termina ted at the time when the operator could energize the pressurizer heaters and reestablish pressure control. No alterations vore mde to most of the numercus tables / control blocks made necessary by INP0's calibration to the actual Ginna event. Chief among these assumptions during the relevant phase of the incident is the use of upper head " valve" area tables to simulate time-dependent form factors. As these areas are held constant during the period of interest to the parametric considered here, the utilization of the table s _... _ _,._ -~ _ _ _.
i 1 t 3 43 I i could be justified on the grounds of simplifiestion of a multidimensional geome tri cal problem. In addi tion, no modi fica tions were made to the tables / control blocks for the behavior of the A steam generator (in tact l generator) which implies that the procedure for utlizing the intact generator I as a heat sink follows that of the Ginn4 event exactly. Modi fica tions, l however, had to be made to the pressuriter heater Control blocks. Figures 4.1 1 to 4.1-7 111ustrate the - system response. While the transient is plotted for the period from 1000 seconds af ter tube rupture on, a ttention should be focused upon the PORY depressurization and pos t-PORY depressurization period, Mmely from 1500 seconds to 2400 seconds. The event sequence of the period prior to 1410 seconds corresponds exactly to that of the actual Ginna incident. Tabit 4.1-1 shows the timing of the various operator actions required by the guidelines. he l P t i I l 4 L M ,,,--.--,-=,.,,-,,,-n..,.,,,--,--.- ,,,-n,, _ -,-, -,~_
44 Table 4.1-1. Operator Actions Event Time (secs) Charging on 1410 01:en pressurizs PORY 1500 i Cycle PORY Same sequence as in actual Ginna incident. Just moved initiation up to 1500 seconds.. 51 fitw termination 1720-1765 (45 second ramp) Letdown established 1850 The reactor coolant system (RCS) pressure trace, Fig. 4.1-1, shows the four operator initiated pressurizer PORY openings, the PORY block valve closing and the termination of the safety injection. Figure 4.1-2 depicts the collapse of the upper head bubble with the closure of the PORY block valve. Volume 20, the volume directly below volume 19 and physically a part of the reactor vessel upper head region, also undergoes flashing and further complete steam bubbles collapse during this period. This volume was modeled as a homo-geneo,us volume by INPNO. As volume 19 does not completely empty. this is a physically consistent picture. The narrow range pressurizer level instrumen-tation (uncorrected for non-saturation conditions). Fig. 4.1-3, shows a re-filling to ~ 65% where a leveling of f takes place. At this point the operator could re-energize the heaters and reestablish pressure control. From Figs. 4.1-1 and 4.1-4 it can be seen that a quasi-steady state has been reached with the tube leakage flow now of a negligible proportion and the difference in primary to secondary pressure attributable mainly to the hydrostatic head nf a ruptured steam generator filled with substantially more liquid than at normal operating conditions. However, Fig. 4.1-5 which is the narrow range B gener-ator level, shows that there is considerable margin to filling up uw ruptured generator, in the context of filling the steam dome, a narrow-range measure-ment of ~ 210% is to be considered as full. The core exit temperature of Fig. 4.1-6 shows that the degree of subcooling has decreased to 10*F but that is due to the use of the actual onerator actions for steam generator A (intact generator) during the actual event. Figure 4.1-7 depic ts the A generator pressure which is an indicator of the generator behavior and can be compared wi th F1g. 3-9. ( g .m--2 w --e w s m
+,-,,-,,.,-,--r,--m--.-w.,-w-,-w-,.,,e,m,_
m_.y.--, -,,-.,-.-m.w.-~...s-%.----
4 45 s ct w ;4 " 4 :( 05:1 ;'.! J g. e.
- r. _
1. l ~ [.- c f I* ol - - f E +g N 5 $g N *%,- ga e EJ f ~f 1 I 1 ? mo im in me se rm am rm na fl.t iSEO) a Figure 4.1-1. RCS Vol. 61 Pressure /SGB Yol. 68 Pressure. C;kG Fiv':R ;ul L :kES i i 1 i 1 I E D s d> B h! 5; C z2' m- ~ 9 .i ' t 1 i i aem me isce is rom mo nm ma tuo I]ME l$fC) Figure 4.1-2. Upper Head Vol.19 Vapor Volume.
- -.. ~ - -. -. - 46 otW. 04te:t ; ::1.:,ti g r g
- a Wg b
~ ~ gS a h'. j' A a g J I un irso ism iv
- =
ma rsx im um TIME titc) Figure 4.1-3. Narrow Range Pressurizer Level, c!NtA cetw:; : :L. d5 15 I - 1 4 gl \\ ~ 3, h s, -== is d N . fS -9 w ?) d 6 15 _ g 3 kh h_ t .e o e ) ? ks me ix nsa ma me nce mo sa# t!*E tStti Figure 4.1-4 SI Flow / Break Flow.
47 t CINu y W s 0.:ttL:'.tt t y =,. X J l i i E6 i 1 t i ,t l !.~ 9 a wg-a. i 9 1 1 1 1 1 1 1 uv su ,a su exo nu nw rw un time 15101 r Figure 4.1-5. SGB t.evel, c:sv Ig w ;E cu:t.P,ts i E.~ t t be-E. *, I k* P-- ,_ c W W W r g., :,- s-w l i s - i t i 1 g y i i 4= in: in is me nu nm rw na i!MC ISEO) i 1 J Figure 4.1-6. Core Exit Vol. 13 Temperature.
1 i 46 Gts: lrts:':' :,t:L.oll g z i 1 i g w C., 2 ~ e k W -{ ~ o g b EI c ~ E ~ t e i I . eax izw su sm. tu - to tu tm um TIME (SEC) >igure 4.1-7. SGA Yol. Se Pressure. a~ O! w n stu u 3 5', W i - {* g, ~ _ e R R o. O t . n. eP ta ~" a n n. 8 e %.-3 W - B a. mg - y Q 5 y: . 6.$ r ~ c e y ~ [ E I I -_-M 40 - 80 13 IN 4% i'O MO IIM IlPC ISECl e Figure 4.2-1, RCS Yol. 61 Pressure /SGB Vol. 68 Preusure. ( t ,w. ---,,..-em,--m-m-,, +w-,-4. -va-r --w.ee-,--... ., se n -*w-rv.,r -r-r.
l 49 CINNA $* :( 21:N '0% 1 9 2 () d= W5 ,8' s y i a___ e b ..o e m. nn u no m no f!ME iSCO Figure 4.2-2. SG Tube Domhill Vol. 44 Oua11ty. i CINNA Stutt Orts *:R< q q r-4 q ~ 9 - =9 ~ - 15 ' d d 5 6 a ' $ =. d w 'g W-w3 = -~. t 8 Ng b.9 9 9.- y 3, ~ . 9 I I 1 1 m .o - so nro nn a no m m: IlMC iSEC) Figure 4.2-3. Upper Support Vol. 20/ Active Core Vol.1/SG Tube, Vol. 43 Ouality. h
4
- w: A?.:.
T. r:4. y ___ _.. g '- / %~ i w O. 3 l Y ? l l6 }? m_ w ~ er zS< 3 y t s !2 I t { j j 1 ao so e tto 160 _ 200 N_ - reo ng 7 11MI (SEC1 - Figure 4.2-4 Upper Head Vol.19 Vapor Volume. ctva su OrtN P:My 9 1 1 m. 1 1 w >=. + w 9 g.- -e h l h w - 'w B e, W g O* 1Q i = t + n- -_1 _. Wo,. 4-
- 0
- 40 120 160 '0G Pa0 m tim; isE:i Figure 4.2-5. Cuide Tube Vol.14/ Bundle Top Vol.15 Quality. r g y ye67 w.-+ ,r-,r v w e,w-w y e n -.yn-e www,-ey+~,4-e-_e-ywwrc--g-p,i.3 ,,rw m m e -- m -,,n y n, y .rp,,wy 4 r.p. n fe- % we,+,,,--,,..wp. w i-e-w w - w.. war + r+ r-'er--cew"*-*< '"*-em m wvr v-wr-
I i l + CINsa $tus ;P:s r;my 9 i i i 1 ( l 9 i t. g 5 ~ 8 S q q ? I i l 1 1 g 1 4 m m ao m do reo no f M I5EC) Figure 4.2-6, Guide Tube Vol.16 Ouality. 3 Civ.a si c Orts 70Mv i 1 c j p l. r b c.y / N 11 ~ ge
- 5. *
.a W : w i ', ~ m., E. E ! A a g9 4 I. i ~ I I i i t i e at 4 m m ao cao do te m TIME !$.EC.s Figure 4.2-7. Pressurizer Mixture Level, i i -m v .e-v=-e pp >.,,ew-.-wmy,,-s-nw-m.-,-nennt,.Nw,v,v,--r--,N,p.,,, -weew -,
- w w,w -e
.-e,.,.- e --- .r,,-,n-. +.~n, -,ww..e-s-- --+ -m w --e,we. ww, a s
_ _. _ ~ _ ~ _ -. -.. i gg c!su r.:s ;u v;n g 7.-,. -. v t g c$g i a-hl ~ ~ I y \\, A l i + W \\ l Y y' [ E t .o so tra m xx r,a te no time t$tJ) Figure 4.2-8. Yessel B Inlet Jun 65 Flow. ctwn sw "Es ?:5.- ~ g 1 g t 8 E -t 8 i ,a b E 't ) 1 5 y 2 l i t i 1 =0
- O to 120 40
?x 74
- W 2;t e
T IME ($r..a Figure 4.2 9. Pressurizer PORY Flow.
1 i l 53 1 s I I ctw t*,s :*ts *:% g ..-,-- m m -.,- - g g-g P ,%~ l g l '. g ~ E B b; ) gd M i d 5 B g t w h Ao No No -Y .o no nro rm 6 tiet istc) Figure 4.2-10. St Flow / Break flow, i Gl# s' :s :rts v:.n g c-_ g b
- 8. V b
= wf 1 'u g t a -a_ p 4-4
- 0 M
I?3 160 700 da ?ns 3;Q I t!*E isE; l l Figure 4.2-11. Vessel bo'a9 comer Vol. 18 Temperature. l. l - ~ l I -,-y g-----eye-------,e-ew=w--we-,m yrmwt -s+ w eg.-.,-,e w s t-y,ta,,--omgy -e,-,,ew-m -r--c.m. wv--.-- ,---1.m -- e is e-wm.ge -re m n+3r ..--.-~~,r-ww-r.,e..+-++ ime.wew ees
$4 l i 4.2 j tuck Open PORY For this parametric it is postulated that when the pressuriter PORY I failed to close as it did during the actual event, the downstream block valve i failed in the fully open position. Thus unlike the actual series of events, l and additional small break LOCA compounds the original $GTR. At t' tis point the Wes ti ng ht... SGTR operator guidelines call for the LOCA guidelines. In l this parametric the SGTR guidelines are followed to this time and the the calcuation is continued without any operator intervention until primary side recovery cumences. The computation indicates that this occurs within a few I minutes at which point the analysis is terminated with the level in the vessel slowly rising. Figures 4.2-1 to 4.211 shw the response of the system. The calculation is performed with the t=42.5 minutes INPO deck so the time origin is t*42.6 minutes of - the actual Ginna event when the operator initiated system i depressurizatten actions in order to reduce the primary / secondary pressure 4-di f ferential. The only change made to the deck were alterations to the PORY area cards. None of the numerous time tables were modifled which imples, for example, that the intact generator (SGA) is operated exactly the way that it was during the Ginna event. The RCS pressure trace (volume 61), Fig. 4.2-1, i graphs Mhe four -pressurizer PORY openings and then at - 150 seconds a pla-teauing which can be attributed to flashing in the tubes (volumes 43 and 44) of the disrupted steam generator ($GB). Figures 4.2-2 and 4.2-3 for the steam generator tube quality confirms this. With isolation the disrupted generator becomes a region of low f1w and stagnation. Reverse heat transfer across the disrupted generator tubes tends to hold the pressure up. Figures 4.2-4 to 4.2-6 show that the head (volume 19) had completely voided at ~ 160 seconds and tha t flashing in the guide tubes (volumes 14 and 16) had already com-menced. The flashing in the steam generator tubes and the vessel. head region causes the pressurizer to fill as depicted in Fig. 4.2-7 at ~ 185 seconds. Wi th the filling ' the PORY (junction 122) begins to discharge single-phase liquid. There is an initial surge out of the generator into the vessel with the flashing, as can be seen in Fig. 4.2-8, but more stable conditions occur within minutes with the-SI/ charging flow dominating over the PORY flow from a now - solid pressurizer. Figures 4.2-9 and 4.2-10 st.ow these flows. Reverse leakago occurs in the disrupted generator and the level in the vessel head.is slowly ri sing.- As can be seen in Figs. 4.2-5 to 4.2-6 the guide tubes had ~yc.m-, ,.._.,cr 7 - p. .y,wy_,..m_, ,,gr.,y%,.,.__,, ,,,,,,.,.,..,,_g.. .__,,3 .,ywy m.w.,..v,_,,,,_,_m._.mm..,_,,,....,... -. _ -,,,,..,,.,.,
I 55 reftiled at - 160 seconds when the SG tubes had cortnenced flashing. From Figs. 4.2 3 and 4.2-5 it can be concluded that no significant core uncovery occurred during this period and the calculation was terminated at this point. The minimum downcomer temperature in Fig. 4.211 appears to have decreased by approximately 10'F during this period as compared to the actual Ginna event. Further $1mulation may require renodalization to compute bubble collapse in the disrupted steam gtnerator tubes and modifications regarding the tables and control blocks for the intact steam generator behavior. 4.3 Failure _ to Terminate HP! For this case, ANL reanalyzed the Ginna event, as it occurred, but in this analysis it was assumed that t5e operator did not terminate operation of the high pressure injection system (HPI). The objective of this analysis was to determine the consequences in the primary and secondary systems when failing to terminate HP!. In the actual Ginna event, the operator secured the HPI pumps at onc hour and twelve minutes af ter tube rupture. This was considerably later than permi tted by procedures with the delay largely attributed to opera tor's concern for potential core uncovering due to upper head void formation. More-over, if RCS depressurization had begun when the governing criteria were met - it is likely that filling the 'B" steam generator and line would have been avoided (see Section 4.1). All of this is to say that there actually was a significant time delay in HP! termination during the Ginna tube rupture event which was accommodated without serious consequences to the plant. Securing the HPl pumps reduced the ongoing cooldown of the RCS and the discharge of radioactive water to the environment; some release continued because the charging pumps remained on and their flow exceeded the letdown rate, !NPO l estima ted tha t the "B" generator SRV did not completely close until three hours and two minutes af ter tube rupture; at the tine their analysis was concluded at one hour and twenty minutes, the estimated mss of water released through this SRY was 64,000 lbm. 0 ,s-y e- -y ,,.e,-..-m ,nv- --,-c-,e .-,.r-,m,,--_.2...--,-wm-----,.--,m-7-- , - ~ - - - - ,.=r- -,w -,,,ew-- y 4- --=
._____m.___.~.______. l 56 i For the purposes of this analysis, INP0's modeling of operator control of the intact steam generator (feed and bleed) was unaltered; the only l changes made in the model were to inhibit tripping the HP! and to maintain a constant flow area for $GB SRY equal to that assumed when HP! was terminated in the actual event. The calculation was continued for eight Jlnutes beyond actual HP! termination to show the response trends in the primary and second-j ary systems. The salient results of this calculation are shown in Figs. 4.3-1 through 4.3-6 These graphs begin at the time the INM deck was re-nodalized; all results up to the timo actual HPI termination occurred are Identical to the original calculation. As shown by comparing Tigs. 4.3-1 and 4.3 2, primary system (RCS) pressure remains.bove that of SGB by approximately 300 psi. This pressure l differential mintains flow throug h the ruptured tuve into $GB and attendant release through its $RV. These fle rates are nearly steady at appro.timately 600 gpm, essentially the sum of HP! and charging flow rates. Figures 4.3-3 and 4.3-4 depict these continued flows. t f Yhe sustained injection of relatively cold HPI water '- the RC5 c causes t.ts moderate rate of cooldown to continue. For exampic, ehmination of the calculated fluid temperature in the reactor vessel downcomer, as shown in Fig. 4.3 5, gives an estimated ra te of approxima tely -70*f/ hour durt rg the sad period of the calculationi the results also indicate a 710w reduction in cooldown rate as anticipated. Based upon a limiting acceptabla rate of -100
- F/ hour, these results show that a certain thermal' mrgin s ti 'i. exists even for continued operation of the HP1 system.
,i t is also noted thct the continued injection _ of cold water causes eventual elimination of tho upper head void at the time the calculation is ended as shown in Fig, 4.3-6. t d l l-r I e i l - _~___ _., _ _.- _._._._ . ~... _ _ _. _. ~ _.... _ _.. _. _ -. _ _ _... _ _ _ _ _ _, _ _ _,. _.
.a ...... - _....... ~. - ~. _ _ _ _ _. -. ~. -. _ _... _. _. _. _ ~ v N O!MA R($PCN$& 'wMEN MP1 15 N?,7 TERMINATEQ {; i i i g i 4 e J ?, ' u.t e - ~ 2,.
- 1 4
t f'"^ 1 7.fb w -3 60i d I f f l g, ;. "l s i la f .I 41.5 t! 44 m 40 (LWMD TIMt. WIWff t t pasung a.bl. RCs (vo6 61) Perssuaa. CINNA RESPONSE VMEN rd ! 15 NOT TERMINATED g, -r T l -I ~ -) / F% /[V9/ g ,y $j - 1 i 1-3 7 1 . E.' E.- cl l \\ s1 4. 2ii k, -e T-G d '. a; 71 l ) 4 F. 'i t I t i A.... 47 5 $0 60 70 80 - (LAP 50 f!M, N!*JTIS j e flouet 4. b2
- 8* $ttag litt (VOL.. 68) Petstunt, i
l I {'
I ~ l i l i N I t. CIMM RE*.20N$( VMEN hFl l$ NOT f(RMINATED l l, ' -i i*- e,_ 8 g-g i _ < g {#, j - ~ 3 i u--_ m-2- g' )< f-a -~- g. l l gw ntl g a v i U l I, b" e m i I l k $. 5 ' ) 4 } { CMAAG43 l ( i 1 5 i of 1 o h .i T 1 a2.5 $3 60 70 00 (LMitD TIM misuT*$ Ftsong 4.3 3 NPt, C=avoine Go Tust tverwat Ftow tarts. O!NNA R&$PONSE VMEN k'l 15 NCT TERMINMEO ( -. r _. 's m n' s I 4 N asb.L k ~0; lt # p si it i N, _ l; 4 l s 3 l [ l']>1 6; i --e i - !! I N l!. i! i i.l !!. l. --,-~.1 / t _ _ L. 42,5 to M 70 FC EL AF1(D flat. MIwuft1 a F vat 4.3-4.
- 5' Sitaa Gtattatoa lastry % vt FLev.
_ - ~... _ =... -.. a 4.. 59 .clNNA RE$PCNSE VMEN NP! !$ NOT TERMINATE 0 g 1 L: e b. E. $ -,, t -c: }M. g. E i n A f -- 5 a.i t- $I - I L
- Swww,
%s 4 8 t t 4 s t.1 50 64 ro so (1,4#50 t!Mt. atuutil 5:eues 4 3 5, stactos visset Downcoats (Vos.10 f ta+taatuet. GllM RESPONSE VMEN NPI !$ NOT TERMINATED ~ L g E. : W
- d. 3 W. l 1
) h. f ~ BG 1 1 b 5-E l". E l
- 1 y
~ 9 1 *' ~ n i / 5 if if '? h._ i l 1 47.$ 10 60 70 (1 IL AP$TB T!Mr. nit:0.T1 s flGuet 4.3 6 teActee-VesstL 11 ente unas (vot. 14) $tt u votuas. i n -n ~.-,
4 60 5.0
SUMMARY
AND CONCLUSION A simulation of the Ginna SGTR event of January 25, 1982 was performed utili zing the latest EPRI release version of RETRAN (3), RETRAN04/M0003 in conjunction with RETRAN02/M0003A input decks obtained from INP0 and modified by ANL, The RETRAN02/M0003 results agree well with tise INPO RETRAN02/M0003A calculations (1). A rea sonable ma tch is there fore obtained betwoen calculations an<i measured data from the actual event [2]. Where differences have occurred, they can be explained in terms of code model di f ferences between the M000lA and the M0003 versioni; and in terms of sensitivities in the INPO calibration to da ta. In order to match the limi ted thermal hydraulic i data available, INP0 had to resort to <a number of calibration adjustments, based on engineering judgement, in its application of the thermal hydraulic models availatle in the RETRAN02 code. To further the understanding of the thermal hydraulic phenomena which occurred during the actual event, three additional parametric calculations were performed which included variations on operator actions and further equipment failwe. The three parametrics performed demonstrated; that opportune timing in conforming to recent operator guidelines would prevent the filling the disrupted stea:n generator solid and alleviate concerns about loading questions; that f ailures in the PORY line downstTeam of the PORY would not necessarily lead to significant core damage and; that saf ficient thermal margin exists for pressurized thermal shock si tua tions even if there was a furtrer continuation of sa fety injection flow. While there is a significant out-of-containment loss of ECCS inventory in the third parametric where $1 flow was :not termina ted, timely (.onfor' nance ? to. the opers tor guidelines, as tvidenced by the first paran:etrie:, would prevent such.e condi tion from occurring. The parametrics have contributed to the understanding of the t.hermal hydrualic phen anNna thut occurred during the actual sequence of events _ during.the G!nna SGTR incident. e i i
.._. -.. _ _ ~,_.~. _. ~. _ _ _. _ __... 61 e-Acknowledgements 1 We are indebted to R. Eliasz of Rochester Gos and Electric (RGAE) and to the staff at the Institute of Nuclear Powcr Operations (INPO), E. Winkler and R.. Wyri ck, in particular, for their cooperation and efforts in trtnsferring the INPO decks to ANL and in providir3 additional c'erifications. M. Paulsen and J. McFadden of Energy incorporated (EI) are also to be thanked for their suggestions on RETRAN questions, j This report was prepared by ANL staff in partial fulfillment of a project under the direction of the U. 5. NkC Division o' Systems Integration, R. J. Mattson, Director: B. Sheron, Branch Cnief for Reactor Syttems; N.
- Lauben, Section Leader; J. Guttmann, Project Manager.
ANL staff who provided input to this report were J. H. Tessier and T. Y. C. Wel, authors; and K. Rank and M. Meha f fey, report preparation. 4 t 4 I 8 i a 6 I' l' Ir
_. _ _.. _.. _ ~ _... _ _. _. _ _ _... _ l 62 s References 1. INPO staff, " Thermal-Hydraulic Analysis of Glnna Stean Generator Tube Rupture Event," institute of Nuclear Powr Operations draft report (September 1983). 2. NRC staff, "NRC Report on the January 25, 19e2 steam cenerator Tube Rupture at R. E. G1nna Nuclear Power Plant," Nuclear Regulatory Connission Report, NUREG-0909 ( April 1982). 3. J. H. McFadden et al., "RE'RAN-02 A Program for Transient The rma l-Hydraulic Analysis of Complex Fluid Flow Systems," Electric Powr Research Institute Report, EPRI NP-1850-CCM (May 1981). 3 i r-i o -- ,.,}}