ML17252B181
| ML17252B181 | |
| Person / Time | |
|---|---|
| Site: | Dresden  |
| Issue date: | 06/20/1973 |
| From: | Commonwealth Edison Co |
| To: | US Atomic Energy Commission (AEC) |
| References | |
| 29 | |
| Download: ML17252B181 (29) | |
Text
DRESDEN UNIT 3 SPECIAL Rf'.:POHT NO. 29 TR~J.JSIEl*J'r J1..NALYSE~. FOR CYCLE 2 uSING cunm: .N"r l' T!i;THODS
./
June 11, 1973 Revis~d June 20, 1973
\ ..
TABLE OF CONTENTS I. INTRODUCTION I-1 II. CONCllJSIONS AND RECOf:D.'illATIONS II-1 III. DISCUSSION III-1 A. Basis for Changes III-1 B. Input Parameters Used in Reanalyses III-2
- c. Transients Not Reanalyzed III-3 IV. RESULTS OF TRANSIENT REANALYSES IV-1 A. Scope of Reanalyses IV-1 B. Recirculation System Transients IV-2
- 1. Flow Control Malfunction. - Full Coupling Demand IV-2
- c. Steam Flow Disturbances IV-4
- 1. Turbine Trip Without Bypass (Relief Valve Adequacy Check) IV-4
- 2. Main Steam Isolation Valve Closure Hi th Indirect Scram (Safety Valve Adequacy Check) IV-6
- 3. Nominal Turbine Trip IV-8
- 4. Loss of Generator Load IV-10
- 5. Loss of Li.ain Condenser Vacuum IV-10
- 6. Main Steam Isolation Valve'Closure With Position Scram IV-11 D. Feedwater System Transients IV-13
Ma...xinrum Demand IV-13 i
- v. PROPOSED TECHNICAL SPECIF'ICATION CHANGES V-1 A. Scope of Changes V-1 B. Specific Changes V-1
- I ii
LIST OF FIGURES Number Title Page I-1 Scram Reactivity Curves I-2 II-1 Control Rod Drive Scram Times II-2 IV-1 Flow Controller Malfunction IV-3 IV-2 '!Urbine Trip Without Bypass IV-5 IV-3 MSIV Closure - Indirect Scram IV-7 IV-4 '!Urbine Trip With Bypass IV-9 IV-5 MSIV Closure - Position Scram IV-12 IV-6 Feedwater Controller Malfunction IV-14 iii
I. INTRODUCTION scram reactivity curve could occur due to increased exposure during an*
operating cycle. (See Figure I-1). As can be seen, even though the total '.3cram reactivity has increased somewhat, the insertion rate is
...:*~;:;~::...
. slower at the beginning of the stroke where the transient analyses described in tlle F3Jill for single-event caused abnormal occurrences could be affedted. This, in turn, could affect desisn, oper~1tional, and safety provisions derived from these analyses for such things as the relief and safety valve capacities, set points, and Technical Specificat:ion requiremE_?nts. For this reason all transient analyses previously performed for .Uresclen 3 have been reviev!ed, those w1*,ich could be affected hav.e been. redone, and the adequacy of the present Technical Specification confirmed or the appropriate Technical Specification revisions _deteI'l!lined.
-:~
- .' / ...
I-1
NEW ANALYSIS FOR (Spring 1972, Curve
-30 PREVIOUS ANALYSIS
-20 (FSAR, Curve A)
-10
, 2 3 4 6 6 TIME (sec)
Fi"gure I-lSCRAM REACTIVITY CURVES - ORESOEN-2. 3
II* CONCLUSIONS .AND RECOMMENDATIONS Th"e-safety-va-J:ves--were-determined-to-be-adequa--bely-sized-in-accordance~------
wi th ASME code requirements and the adequacy of the.present Technical S~ecification was confirmed.
For licensed f\111 power, cycle 2 conditions the relief valves were determined to be inadequately sized in accordance with recommended design margins.
With operating power reduced to 97% of the licensed value, the recommended design margin of 25 psi between the peak pressure of the turbine trip withuut bypass transient (relief valve sizing transient) and the setting of the first safety valve is maintained. This power reduction will occur at the time the scrar.i reactivity curve used in the previous (FSAR curve) is no longer valid.
In conjunction with these revised input para111eters and the results of the new transient analyses, appropriate operating and Technical Specification revisions were determined and are included at the end of this document.
II-1
80 I
. 70 60 RANGE OF TYPICAL, RECENT. EXPERIENCE 50 C\I
'I:
40 H H
30 20 CURRENT TECHNICAL SPECIFICATION 10 PROPOSED NEW TECHNICAL SPECIFICATION
- ti
- o""-~"-....&..__...., ___..___.___________________.._________________....,__________________....___________________.
0 0.2 0.4 0.6 0.8 1.0 2.0 3.0 4.0 5.0 ELAPSED TIME AFTER sCRAM SIGNAL (sec)
Figure II.i.l CONTROL ROD DRIVE SCRAM TIMES - DRES!>EN-2, 3
III. DISCUSSION A. Basis for Changes It has been recognized in the past that there could be substanti~l changes in axial reactivity characteristics with increased exposure which could affect the shape of the scram reactivity curve. However, the previous thinking was that during most of the fuel cycle, enough stubbed. (partially inserted) rods were available to effect a fast scram reactivity rate at the beginning of rod stroke. On the other hand, even though all rods could be out at the end of the fuel cycle, flux would be peaked at the bottom of the core and this, combined with other exposed core characteristics was tr1ought to be sufficient to still obtain a relatively fast scram reactivity rate at the beginning of rod stroke. Thus, the old curve was previously judged to be adequate to cover the worst of ther:'e cases without being so extreme as to unduly penalize the rlant.
Later work with axially distributed Gd fuel using appropriate rod, patterns, etc. was based on the desire to follow the Halin8 principle.
This is done by establishing an axial flux distribution which peaks flux towards the bottom of the core early in the cycle in order to reduce reactivity at that location while rods are still there to control it.
This prevents limiting flux peaks from developing towards the bottom of the core when all rods are out and voids are reducing reactivity at the top of the core. With the rods all the way out (or the resulting tendency to all rods out/in patterns) and a reduced flux peak at the bottom of the core, there was concern that the early part of the scram stroke might not be as effective, i.e., the scram reactivity insertion
..J III-1
might be slower at the beginning of the rod stroke. Data from operating plants-has-coni'-i-nned-th;is-tendency-.-Eur-ther.,_if:1pr.oYLed_analy_ti_cal____=====--
capability allowed a more refined .calculation of scram reactivity characteristics for exposed cores.
B. _Inpuj~ Parameters Used in Reanalyses
- .*. S:i.ilce t!iE{r:iew scram reactivity curve represents the effects Of actual exposure for the Dresden Unit 3 core during cycle 2, the reanalyses were done using other inputs at similar core exposure conditions to ensure a realistic case. For exam9le, the void coefficient is reduced which tends to reduce the peak of *!;ressurization transients r.tt higher
- exposures. Conservative assumptions such as a multiplier on the void coefficient, and average control rod scram times equivalent to the proposed Technical Specification limit, were used in the reanalyses.
The void reactivity coefficients utilized in pressurization type transient analyses have b~en
- conservat~vely assumed to be about lt that' which corresponds to the particular core condition w'..ich would exist at* the time of the postulated transients.
A slightly improved control rod drive scram time was assumed, the same as used for later plants (Vermont Yankee and Browns Ferry).. The Dresden Unit 3 control rod drive equipment is the same as these later plants and is capable of meeting the i;-nproved scram time requirements. As illu{3trated in Figure II-1 the scrarr1 time improvement is in the early
\
part a*r:_ the scram stroke w:!ere it is most beneficial to t.'ie results of the transient analyses.
III-2
For the relief valve sizing analysis the reactor power was nominally were analized using 100% reactor power.
Consistent with current desi[71 practice, the scra.'il worth was derated to 80% of the actual scram worth of the control rods.
Consistent with current design practice 290 milliseconds was allowed between a neutron sensor reaching the scram point and the start of motion of the control. rod. Approximately the first 90 milliseconds of this !
time interval results from the neutron flux sensor and circuit delays; after this 90 milliseconds interval the pilot scram solenoid de-energizes.
It was conservatively assumed that 200 milliseconds after the pilot solenoid is deenergized control rod motion actunlly begins. This 200 milliseconds is included in the allowable scram time Technical Specification.
C. Transients Not Reanalyzed The FSAR included about 20 analyses of wor.st. case abnormal transients in six categories of events. 'l'hese categories are primary system pressure increases, moderator temperature decreases, reactivity insertions, core coolant inventory decreases, core coolant flow increase,. and core coolant flow decreases. These were all reviewed to determine those which might be significantly affected by the new exr~osed core characteristic assumptions and therefore needed reanalysis. The breakdown of categories, events and lor;ic for those in i';hich a review was deemed to be adequate, is shmm below:
III-3
Category Event Reasons Reanalysis Not Needed
--Nuclear-Sy.stem_ _ None----All-ev:en:ts __wer.e-reanaly_zed __ and_resul:ts ________===-----*___
Pressure Increase are discussed below.
Moder::itor All, The only event of significance in' this Temperature except category is the feedwater controller Feedwater failure maximum demand, >llhich was Controller reanalyzed and is discussed below.
Failure Reactivity Rod With- These transients are ter.nioated by the Insertion drawal rod block monitor and not a scram, so error they will not be affected by the scram reactivity*curve chanee. Other core characteristics will not change suffi-cfontly to sii'.:oific: 1ntly affect the out-0 come of the analysis or the*block set point.
Decrease of All These transients result in an RPV Coolant d.epressurization and, in some cases a low Inventory level scram. Pml.'er level drops due to void formation before the scram, and MCHFR effects are minimal. A mild repressuriv:tion on MSI:V closure at 850 psig occurs on some. RPV temperature transients are the only concern on so!Tie.
Core Coolant Startup of. This transi.ent is less severe than the Plow Increase Recircula- recircula tio'n syste.'11 flow controller tion pump malfunction - full coupling demand which is analyzed _in Section IV of this report.
- Core Coolant Trip of a A scram does not occur as a direct result Flow Decrease recircula- of t~-1is transient, so the void coefficient tion purn:.:i, change will be the principal effect to the and all results of these transients. The change others in void coefficient is not .sufficient to
- substantially affect the results. Further, startup\ tests resu.l ts where actual recir-culation purr!p trips 1*.rere condli cted demonstrate that these transients are of a substantially milder nature than was ori,:;inally shov.,'Tl by the conservative analysis in the F~;J\H.
Others lilly* other transient analyses conducted ancillary to the standard ones or for other s1:ecie.l purposes, such as a DC power interruption, do not include considerations pertinent to thic:i discussion and are t 11erefore not included.
III-1+
IV. RESULTS OF TRANSIENT fl.EANALYSES
A.---SCOPE-OF-HEAHAI,YSES_ -------:----:--:----~------'----------:-=====---
' .The following transients were reanalyzed in order to determine the specific changes that mi:::7J.t occur to the previous analytical results:
Turbine trip with b:;rpass
'furbj_ne trip \*.!ithout bypass - relief valve ade(:uacy Loss of condenser vacuwn ( sar:1e as turbine trip wi thoy t bypass)
Generator triD 1.*1i ti: and :.rl thout bypass - (same as 1
turbin~ tripf Main steam isolation valve closure (includes delayed scram for safety valve adequacy check)
Flow controller n:alfunction - full coupling demand Feedwater control failure - n:axirrrum flow Specific v..rrite-ups for each of these analyses, together with the transient curves obtained, are included belov.r.
1 t should be noted that the ori::;inal F~;1\R analysis used for the safety valve sizin(!: transient was tlle turbine trip without bypass and relief valves with flux scram. However, it has been deterr.1iried the main steam line isolation with flux scram could be more severe. During the reanalysis work reported herein, this possibility was checked by performing both analyses and the results showed a somewhat higher peak pressure with main steam isol::ttion valve closure (without relief valves). Hence, this analysis is used for checking safety valve adequacy in this report.
IV-1
I B. RECL'9.CULATION SYSTEM 'l'RANSIENTS B-;1--Fi-ow-Gontroi-Ma-lfunct-ion---Fu,l-l-G0upl-i-ng-1Jemand----------------::::::;===:-::----
The flow control malfunction of failure of one of the M/G set sp'eed controllers was reanalyzed at the new conditions. 1 rhe case reanalyzed was the failure of a speed controller in-the direction to demand full generator speed, starti1ig from a typical low power initial condition i;..ri. th power at 65 percent of 2527 I*'fr.!t and with core recirculation flow at app~oximately lt9 percent of deGign. The failure of the controller causes the scoop positioner to move at its maximum rate to _establish full coupling between the drj_ve motor ;md the generator. The results of -
the transient are illm>tr;!ted by Pigure IV-1.
Both speed loopn are initially operatj_ng a.t about 41 percent of rated speed. Upon tlle failure of the controller, tbe i:.;rmerator and pump speeds of the failed loop rapidly increase to about 102 percent of that speed which produces rated drive flow. The other loop remains at its initial speed conditions. Drive flow in the failed loop also increases, causing the ciet pump diffuser flows associa.ted with that loop to increase from 49~percent to approxim~tely 140 percent of rated flow. The drive flow in the other loop remains essentially constant, (decreasin8 only 2 percent); however, the ,~et pumps associated with that loop experience decreasing diffuser flows because of the greater core.6.P, with flows essentially zero at 20 seconds. 'rotal core flow increases rapidly to about 74 percent of rated flow causinr.; reactor power to increase such that a reactor scrarn from hi[.:)1 neutron flux results at about 3.8 seconds, Neutron flux peaks at 170 percent of the value at 2527 M',*it. Peak surface heat flux however, increases to only 76.5 per cent and thus thermal margins are maintained throughout the transient.
IV-2
z STH LINE PRE:! ~*SElPSIJ l YESSEl PRES Rl lPSlJ a TURBINE P!lES I IS£ lPSU 35tl.
& BYPASS Vlll. VE I aw l%l I s REJ.IEF VRLVE f aw l:Q B SAFETT VRl..VE I aw 1'%1 Cl LLJ 100. 200.
I-er a::
u..
0 I-z UJ u so. so.
er LLJ
- . .*
- I a...
- ~ .
~ I i
ob: u. . I I*' . oa&M~
I 12.
lSECl "* 8.
- TIME CSE Cl
- 12. 18.
~
VI 100. 120.
~~
21 SURF
~ FLUX HERT FLUX
/
Cl LLJ 1-cr er u..
80.
.. ~,
J 0
1-z l1J u ~-
a:
l1J ...
a..
... I
- 8. 12.. 16.
- . I
- 25. so.
\
- 75. 100.
TIME lSECl CORE FLOW l/.l r
-FIGUREIV-1 FLOW CONTROLLER MALFUNCTrnN - INCREASING FLOW - -FIXED FLUX SCRAM *.*
C. STEAM FLOW DISTURBANCES C~l-.-Turbine- -Trip-Without-Bypass---Rel--ief'-Va-1-ve-Adequacy-Transien:t The relief valve sizing transients were reanalyzed at the new conditions to verify the adequacy of the five electromatic relief valves to terminate the pressure transient when the reactor is subjected to a rapid pressurization event such that satisfactory margins are maintained between the peak pressure resultipg and the first safety valve setpoint. The event analyzed is the turbine trip with simultaneous reactor scram but with a failure of the turbine bypass system. The results for this transient are illustrated by Figure IV-2.
Initial conditions prior to the event are reactor power at 97 percent of licensed, core recirculation flow at 98 million lb per hour and reactor stea~ dome pressure at 1005 psig. 'I'he relief valve setpoints remain unchanged.
'!be sudden closure of the turbine stop valves with no initial bypass flow causel? a rapid rise in system pressure at a rate essentially double that which results when the b;,'pass system functions. Position switches on the stop valves initiate immediate re:~ctor scram. The rapid pressurization causes core void collapse and neutron flux.increases, peaking at about 143 percent of the initial v:-tlue at the start of the transient before the scram becomes effecti1re. Core average surface heat flux dips initially and then increases to 100.5 percent of its initial value at the start of the transient at about 1.36 seconds. No significant reduction in I-iCHFR occurs as the core recirculation flow increase and the heat flux increase is small di;e to the reducti011 in core voids resulting from the pressure increase and the reactor scram.
IV-4
0 100.
UJ t-a:
cc LL 0
t-z UJ 50.
u a:
UJ 0..
~
ob. 8.
H TIME
~
V1 100.
-sob~.............................~~~~.~~~~~8~.~~~~7.1.2.~~~~---:-1s~._.,,.'-'="'"""'=a..
TIME CSECl i<"'igure IV-2 DRESDEN 3 REL~AO. CYCLE 2f TURBINE TRIP WITH~T BYPASS
Reactor pressure rises to the setpoint of the first relief valve wi-th:ln-1--.-B-secends-ana-al-l-rel-ief-va+/-ves-a-re-epen-ey-2-.1-seconds.--'l'he--------
pea..~ pressure at the location of the safety valves is 1185 psig.which is 25 psi below the 1210 psig set point of the first two safety valves.
C.2. Main Steam Isolation Valve Closure With Indirect Scram - Safety Valve Adequacy Transient Safety valve sizing transients were also reanalyzed at the cycle 2 conditions to verify ti1at the eight spring safety valves will provide sufficient margin to the AS1'*1E Code limit when the reactor is subjected to main steam isolation valve closure event. Recent analyses of the turbine trip without bypass and MSIV closure have determined that when direct reactor scrams are ignored, i.e., those originating from position switches on the stop and isolation valves, slightly higher pressure peaks, (less than 10 psi), result from the closure of all main steam isolation valves. The transient analyzed to verify satisfactory margins to the ASME Code limit is thus the three second closure of all main steam isol~tion valves.
It is. assumed that (a) the r8actor is c.-,_t 2527 li'::Jt when the steam line isol.;.tion occurs, (b) the relief valves fail to open, {c) .direct reactor scram ::rom the position sid. tches fails, (d) the backup scram due to high.neutron flux shuts drn*;ri the :n,acotr. Figure IV-3 shows the results of ttis transient.
As the steam flow vs. position characteristic of the isol'.:!tion valves is non-linear, the effects of closinr.; the main steam isola~.:,ion valves are not immedic..te. Hm-.rever, by about 1.5 seconds after the valves start to close, pressure has increased to tlie *i;oint where core void IV-6
2 STH LINE PRES fr~SEtPSll l VESSa PRES R lPSII a TURBI ME PRES I !SE lPSIJ
~ BYPASS VAL VE cw (%)
350. 6 REL !EF VRL VE ow l%l e SAFETY VALVE cw l4J
/,
/
/ ~ ;::::--__
r--::::: .
Cl w 100.
..... 200. - I I I
a:
a:
LL.
c I-z w
u so. so.
a:
w .J v &
- r--_
- . . I
-\
0...
ob. 8. 15. . II. 6. 12. 15.
lft:ID2"11-211 TIME TIME CSECl f
--.J l NEUTR ~ FLUX 2 SURFA HEAT FLUX 100. 120.
(
'.....(
I so.
- CJ w eo.
I-er:
a:
LL.
D I-z w
- 0. u llO.
a: ,..
.W 0...
..p ---211
-sob: II. 8. 12.
TIME CSECl 15.
me!EllM-211
. I.
- 25. so.
CORE FLOW liO
- 75. 100.
. FIGUREJV-3 MSIV CLOSURE WITHOUT RELIEF--FLUX SCRAM ...
collapse results in a significant increase in neutron flux. Flux approximately 390 percent of initial level. The core average surface heat flux peak occurs later atabout 2.8 seconds and reaches a peak of about 117 percent. Pressure increases at a rate of about 10 psi/second until limited by the opening of the eight safety valves. The safety valve setpoints are spread in 10 psi increments between 1210 .and 1240 psig with two valves at each setpoint. St.eamline pressure reaches the setpoint of the first safety valve at about 3.8 seconds and all valves are open before 4.5 seconds. Vessel dome pressure reaches a peak of 1247 psig and the peak at the bottom of the vessel is approximately 1276 psig which is 99 psi below the peak pressure allowed by the ASJ*iiE Section III Code, (110% of vessel design pressure of 1250 psig or 1375 psig )
- These pressures are actually sligbtly lov.1er than the peak pressures of the original design analysis in Section 4.4.3 of the F'SAR.
C.3. Nominal Turbine Trip A turbine trip is the primary turbine protection mechanism and is initiated ~*men ever various turbine or reactor system malfunctions occur which may threaten turbine operation. The turbine trip initiates fast closure (approximately 0.1 second) of the turbine stop valves. The load demand is also set to zero such that reactor pressurff control is quickly transfered to the bypc;ss valves. iteactor scram is initiated immediately from position switches mounted on the turbine stop valves.
Figure IV-4 illustrates the results of this transient when occurring at the maximum reactor pm. :er level of 2527 MWt.
The sudden closure of the stop vo.lves causes a rapid pressurization of the steam line and reactor vessel with subsequei1t void collapse and a small reactor power increase. Because of the fast action of the IV-8
t_ ,,
150. 350.
/
/
/
/
/
c; LU 100. 2CXJ.
I-a:
a::
LL.
0 I-z:
LU u
so. 50.
a::
lJ.J a..
~
ob. e. 12. -1000. ij. e. 12. 18.
\0 TIME CSECJ TIME CSECl
-soo~.....................................~ij~.~~~~-:!e.'--~~~---:1+/-2.~~~~-:-:1s~.--"lftSEll--==""""'""""""
TIME CSECJ
?'igure IV-4 DRESDEN 3 RELCJAD, CYCLE 2 TURBINE TRIP WITH BYPASS
. *,, ... ' ! ~ '*. ** * .l* *i*
.,~ .... ~ ' . . ...
- '* *'.t
~ ~. ~* ,:
- ~' ' ~ ' ., I turbine trip scram, the neutron flux peak is held to 102% of initial
value-.-Cere---ei.-vera-ge-su-rf-ace-heat-f-lux-decrease-s-throughout-the-transrent--
and no decre.1se in MCHFR occurs. As the bypass valves open almost irnmedi:itely, the vessel pressure increase rate is limited to less than 50 psi/second. The action of tbe bypass valves limits the peak pressure rise at the loc~tion of the relief and safety valves to 1098 psig which is 112 psi below the setpoint of the first safety valve. Vessel pressure peaks at about 1100 psig.
c.4. Loss of Genera.tor Load Loss of generat9r load is quickly sensed by the power-load unbalance
- circuitry in the Turbine Electro-hydraulic Control (EHC) system. This circuitry energizes the fast acting solenoid operated valves which open disk dwnp valves on the turbine control V<~lves such that the control V?-lves close rapidly and excessive turbine-e;enerator ovcrspeed is prevented. Byr_Jass valves are orr=;ned rapidly .-.rhen the load demand is 1
stepped to zero and reactor pressure control is transferred from the control valves to the bypass valves. H.eactor scram is initiated immediately from position s-v1i tches mounted on t:he fast acting solenoid valves.
As the closure of the control valves is essentially as fast as the closure otJ the ti.:.rbine stop valves, the transient is almost identical to that of the Turbine Trip transient described in C.3.
c.5. Loss of Main Coiidensor Vacuum If condenser vacuum is lost while the un:i. t is in operation the followinr; trips will occur:
IV-10
Alann at 24" Hg vacuum
_ _ _ _ _ _ _ _ _ _ _S_crarn_at -22.!.!_Hg-vacuum.---
Turbine Trip at 20" Hg vacuum Closure of Bypass Valves at 7." Hg vacuum The worse case for t:;is type of event would be instantaneous loss of vacuum with the unit operating at 2527 M~'rt. The transient for this
. case becomes identical to the turbine trip without bypass transient.
Slower losses of condenser vacuwn will produce less severe transients because the scram will precced the stop valve closure resulting from the Turbine Trip and some bypass flow will be ;;emitted to remove stored heat.
C.6. Main Steam Isolation Valve Closure - Position Scram The closing time for the main steam isoL1t:i,ori valves can be as fast as 3 secuuds. The effects on the reactor are not imroedia te as the flow characteristics of the valves are non-linear; however, a significant vessel pressure rise results when the reactor is suddenly cut off from its prin1ary heat sink. Position switches on* the isolation valves initiate reactor scram before the valves reach the 10 percent closed position.
The transient resuJ.ting from :i.nadver.tent closure of all main steam isol:.1tion valves from 2.n initial power level of 2527 MWt is illustrated in Figure IV-5.
There is no increase in neutron flux or core average surface heat flux as the position scram becomes effective before significant valve area reduction occurs. MCHFR therefore increasesthroughout the transient. Core pressure does not rise significantly until about 1.5 seconds after the isolation valves start to close. By 2.5 seconds IV-11
0 100.
w t-cc cc IL..
0 t-z w 50.
u cc.
w Q..
ob. - 1000~_.......................L...L.~~~-~~~~-::8~.~~~~~1+/-2.~~~~-:-:!16~.--:IHiiEi""""'~-,,,,,,_,~~
12.
f.....
I\)
CSECl TIME CSECl 100.
-50b~..............................~~~.~~~~~e.'--~~~--=1+/-2.~~~~_,_,1a~.~IH5Ei~~*~~==-
TIME CSECl Figure IV'.'"5 DRESDEN 3 RELOAD. CYCLE 2 MSIV. CLOSURE WITH RELIE.F VALVES
the pressure rise rate is approximately 60 psi/sec and pressure reaches the first relief valve set Qoint at abQU+/-,_3._6_seconds.* _ A l l - - - - - - - - - -
relief valves have lifted prior to 5.0 seconds. 'i'he relief valve flow removes the excess stored hea.t and holds peak steamline pressure to 1155 psig which is 55 psi belm*1 the set:;;oint of the first safety valve. The isolation condenser would be initiated during the first pressure increase as vessel pressure remains above the 1060 psig level for more than the 15 second. time delay required for actuation. Vessel pressure peaks at 1155 psit;.
D. fl.'EDt*JATE:R SYSTf'l'il TR.L\J~SIENTS D.l. Feedwater Controller Failure - Maximum Demand
'rhe response of the. plant to an excess feedwater flow tr.:HJsient was reanalyzed and is illustrated in Pigu.I,'e IV-6. The transient was initiated from a typical low po1r~er condition with reactor power at 65 percent of 2527 l!iWt and core recirculation flow a:t 49 percent of rated.
'.I"ne feedwater controller was assumed to fail such as to demand maximum feedwater valve opening resulting in about 110 percent of rated feedwater flow. Because of the low initial power level, a large mismatch between steamflow and feedwater flow results and the level transient is more severe.
'I'he maximum rate of level rise is about 2.8 inches per second.
The insurge of excess.feedwater initially causes a slight reactor pressure and power increase. At abmit 5 seconds, the core inlet flow becomes cooler and a faster increase in reactor power results.. Sensed water level reaches the high levd turbine trip setpoint at 7.2 seconds causing fast closure r..;;: thE;? turbine stop valves and sj.r.iul taneous opening of the bypass valves and a r2 .ctor scram. As the bypass capacity IV-13
i VEssa P!lES Rifi tPSTl 2I NEUTR0-1 FLUX PERK. FUEL '* TEHP CENmt 2 STH LINE PRES !SE IPSIJ a AVE SURFACE FLUX I TURBINE PRES : !SE lPSll Q FEEllWAlCll sso. G BYPASS VAL VE aw w
~ FELtEF Vl'U.V'E F L..Cll t%1 a SAFETY VIII.VE FL..Cll lXJ fi3
..... 100.1--~~~-tr--1-~~~~--;~~~....,.......,...-1-~~~~--1'--~~~~ 200*
a:
a:
IJ...
0 z
w u so.1--~~--1f-+-H-<--"..,.....~~--i~~~~....,...-+-~~~~--1,__~~~~
so. -- -
a:
w
- 11 ~ 'I\ . . . . I .
Q..
~ ...
I s -
I 2..
'".
- I
- 10. 20. l,IO.
!lelD 203 -
30
- TIME lSECl I
2 NE\11;~ ~FLUX SURF HERT FLUX 120.
D w 80.
a:
a; J IL 0
zw
- 0. u a:
w Q..
- ~
-so~.~..................................~10.~~~~--,20~.~~~~~30~.~~~~~l,IO~.--=OESlEii===-= =-=
2111
. 1 .zs. / so* 75. 100.
llll!Sl!ll 2111 .Jn TIME lSECl CORE FLOW (i.)
FIGUREIV-6 FEEDWATER CONTROLLER MALFUNCTION - INCREASING FLOW - HLTT AND FWPT
is smaller than the steaming rate at the time the turbine trip occurs, reactor pressure increases. The_sj;_o_p_v-al-v:e_closure--scrcam-1-imi-ts--'--------
the peak neutron flux to 80.5 percent of the 2527 MWt level and the peak average surface heat flux to 68.6 percent. As core inlet sub-cooling is increasing, there is no significant decrease in thermal margines. The action of the bypass system and the reactor scram limits the vessel pressure rise to about 42 psi. Peak steamline pressure is about 990 psie; which is well below the setpoint of the first relief valve.
The analysis performed assumed that the reactor feedwater pumps were also tripped ori high water level, thus terrnin:.tting the increase in reactor water level.
IV-15
- v. PROPOSED TECHNICAL SPECIFICATION cm. NGES A. Scope of Changes The principal change concerns the improved control rod scrain times *
.This change is consistent with the new assumptions used in the transient reanalyses and is discussed in detail in Section III B above. Other changes are. those.associated with the results of the.transient reanalyses discussed in Section IV above. None of these are of a crucial safety nature and mostly.aff~ct statements about margins for various pressurization trans.ients.
B. Specific Changes Item* Location Ch:rnge Safety ;:::valuation Basis Page 13, ln the fifth sen- 'l'his change is consistent statement para 3. tence c.han:_i;e "10/ with tlle nf:w *scram time for 2.1 to "5~{ and 20/". in1;1ut p;}r;:1meters used in these transient reanalyses.
Basis Para 5 on The f'ourV1 sen- This change is consistent with sL:ternent Pa.ge 20 te11c.e clnnge to the results of the.se reanalyses for i.2 read:" *.* limit ar.d. co:rl})lies \.,ri th *design the reactor criteria *.
pressure to 1185 psig ( 5) and ( 6) which is 25 psi below *** "
Basis Refs. at Add reference The statements are supported statement bottom of to this report by these transient rea11alyses for 1 .2 rage "20 Spec. Tables on Change the h!O These transient reanalyses were 3.3.c.1 page 58 tables, ir. done with these new scra'11 time seo_uence, to the req_uirements. These changes are following: consist.ent with input parameters used in these raanalyses. These new scran time are more restrictive in the safe direction; therefore no 1;.nreviewed consequences are involved with these changes*
V-1
.%Inserted From Fully Average Scram Insertion Withdrawn Time (secs) 5 0.375 20 0.900 50 2.00 90 5.00
,%Inserted From Fully Average Scram Insertion Withdrawn Time (secs) 5 0.398 20 0.954 50 2.120 90 5.300 Item Location Change Safety Evaluation Basis 4th, 5th, Footnote references This change is consistent with statement 6th sen- from Fig.3,5.2.of the scram curve input parameter 3.3.c tences on the SAR to Fig. used in these transient reanalyses.
pages 63 I-1 of this sub-and 61+ mittal.
V-2