ML20236Q309
| ML20236Q309 | |
| Person / Time | |
|---|---|
| Site: | Crystal River  |
| Issue date: | 10/30/1987 |
| From: | Ellison R, Parece M BABCOCK & WILCOX CO. |
| To: | |
| Shared Package | |
| ML20236Q289 | List: |
| References | |
| 51-1170352, 51-1170352-00, NUDOCS 8711190306 | |
| Download: ML20236Q309 (18) | |
Text
4 BWNP 20440 5 (12/86' C"
[ ENGINEERING INFORMATION RECORD Document !dentifier 51 1170352-00 Title FPC Diesel Loading Evaluations
(
PREPARED BY: REVIEWED BY: l Name Randy H. Ellison Name Martin V. Parece i Signature , MN Date /o//o/P7 Signature MI b Wate /odeI7 f-1 Technical Manager Statement: Initials i Reviewer is Independent. l l
l Remarks:
l' This document provides a review of diesel loadings through a scenario based review. The intent of this document is to provide additional review beyond that provided by Gilbert Commonwealth Inc.
to support FPC in discussions with the NRC. Additioanlly, inputs for Gilbert Commonwealth are also provided.
[
4 8711190306 871116 PDR P ADDCK 05000302 PDR Page 1 of '7
y j J
51-117h352 00' I i
Ba$kground:
During an Operational Safety Team Inspection by'the NRC at Crystal River 3, the NRC identified an inconsistency between the emergency diesel generator (EDG) loadings and the surveillance testing performed at the plant. The team also ,
identified the overload condition for the EDGs was less than the ==v4==- {
expected loniing. The NRC requested a review of the inconsistency and a presentation of that review to the NRC in October. Gilbert Ctemonwealth, Inc.
(GCI).-began addressing the diesel loading insam for Florida Power Corporation
. (FPC) . The calculations evaluated the actual electrical loadings due to instrumentation and other arri,==nt loadings = par +M to be automatically loaded following.a' loss of offsite pcuer in conjunction with an Engineered Safeguards Actuatim Systen (ESAS) actuation. The loadings are higher on EDG "A", so a failure of EDG "B" was postulated for this evaluation. By postulating *B"Jdiesel failure, the "A" diesel will be-required to handle all of the load frtan safety systems for a langer period of. time. Thus, if the "A" diesel can be shown to have acomptable loadings, the "B" diesel, by otsparison can also be shown to have acceptable loadings. Among the largest individual-loads are those associated with the high pressure injection ,(HPI), low pressure injection (IPI),. biildig spray (BS),
Since for various scenarios, eam puup,and emergency will have feedwater a different (EEW) loading,,it pungs.
was decided to perform an evaluation of the diesel loadings hamad on event s
scenarios.z GCI developed a. series of scenarios to bound the FSAR accidents.
The scenarios had as a ocamon assunption a loss of offsite power, ani thus a diesel start, at the initiation of the event. The flow ratas expected and the amarviated horse power requi,_._tus for the individual pungs were examined to
' arrive at conservative Irmaings for each event. Due to the nature of the task,
==vimization of diesel 1r'ading, same of the scenarios develop differently than described in the Crystal River 3 PEAR. Maamd on this, the scenarios whi d diverge form the FSAR scenarios will not be evaluated for specific thermodynamic responses since the design basis of the plant runnins the same as that presented in the FSAR.
This rimunant sunmarizes a review of the GCI scenario thunant ard provides additional detail to Pupport the system responses used in GCI diesel loading calculations. Therefore, this rimnnant ws. _As an irdei;am4=4 review of the
. GCI dm==ntation with backup information where required. Additionally, this dmunant highlights the areas of expar+M NRC scrutiny and rarvvanandations for f changes to the GCI rimnnantation to address these issues.
l
{ Diamanicriof Scenarios:
l The scenarios chosen by GCI represent a wide range of thermal hydraulic canditions. Because of this, the events provide a good basis for determining the range of diesel loadings a =
- M for design basis accidents at the Crystal River 3 plant. For each event type, the scenario is provided with a description of the expected flow ratas frcan the major pumps. Again, the reader is reminded that the intant is to determine conservative diesel loadings and not define additional design basis events for the Crystal River 3 plant.
In all cases where EPW is actuated, it is maannned that flow will be required to ,
reach the natural cirt:ulation steam generator (SG) level. This value has been ]
irwr- i frtan the BW generic 50% SG level to allow for ar=51% calibration I z
s ,
53-1270352 00
, and instrument uncertainty. The SG level for all power levels will be required 0
' , to be' increased to the natural circulation point since follwing reactor trip, the steam generator levels collapse to a liquid pool. Since endt event is amannad to exist coincident with a loss of offsite pcWer (IDOP), DW will be
~ initiated m the_ loss of reactor coolant (RC) pumps. The prTnaad modification
. for tripping the motor driven EFW pump during large and irta"nadiate break IDCAs will' eliminate the EPW pung load from the EDG for those events. The ar-+=hility of this modification will be diamamed later in this dennant.
IARGE IREAK IOCh The large break IOCA is typified by a rapid primary system depressurization as the break ' flow radn'an the primary system inventory. The rapid d6 pressurization initiates HPI ard IPI on the icw reactor coolant pressure l function of the ESAS. Additicmally, the BS will be actuated by the ESAS high-high reactor biildtrg pressure function when the reactor biilding pressure reaches the 30 psig .setpoint. The primary system pressure response has been predicted by approved evaluation undals in Reference 1. 1he cantairanent response is described for different break sizes in the Crystal River 3'FSAR.
For large break IOCAs, the primary system pressure' will be wwcimately 100
. psi at approximately 15-25 seconds after the rupture. . The actual timing is i dependent upon the actual break size ard location. In all cases, the pressure i is sufficiently lw that IPI flow is aw=r+ad anoe the puup is inadad onto the EDG. Since the primary system pressure will remain low during the entire event, it is amannad that the IPI flow will be at capacity. 1his will ==vi=4ze the. diesel loadings for this event.
During large break IDCAs, the primary system voids rapidly, th 4fra=4e ally decoupling the primary and seemdary systen. Virtually no heat transfer, frun the primary to the secondary, occurs after the first few seconds of the event.
In fact, the IOCA analysis amannan EFW flow to provide a secondary system heat source for energy addition to the primary. Therefore, although EFW will be actuated and will begin to fill the SGs, a safety function is not strictly being acccaplished since the EFW flow does not assist in event mitigation. . A tarporary modification for tripping the motor driven EFW pung when a large break IOCA occurs coincident with a IDOP will be aamunad in place for this ,
evaluation. Thus, the EFW pung will not be added to the EDG loading. A more ocuplete diamanian of this modification is described in the MAR for this modification, included as Appendix A to this dennant. Additional informaticm ;
relative to the inpact of this modification on plant safety analysis is provided in Appendix B of this report.
I Within the first 30 seconds of the event, the HPI, IPI, and BS puups will be !
, expected to be loaded onto the energency diesel generator. Idoensing analysis aaannaa that the HPI flow reaches the reactor coolant systen by 35 seccmds.
Due to the low pressures, the HPI system will be in nearly a runaut candition.
This flow has been determined to be 600 gpn, based on the pump performance '
curves. - Also, pre-throttlirg of the valves in the HPI lines will prevent
,_ runaut of the HPI pungs. The IPI system will be up to full flow plus recirculation, since the primary system pressure will be below the 100 psia IPI design point (3000 gpn per punp at 100 psig). Full flow of 3150 gpn is assumed. The building spray punp will also be in operation with its flow ;
beginning at 56 sectmds and full flow reached by 71 seccmds. niilding coolers '
3
, ,a
. . J 1170352'00' f are a===d operational within 25 secords of initiation whicti.is consistent I with PSAR analysis assumptions.
Haus, tho' equipment required to be operating and the appropriate flow rates are mamarized below for the IR.DCA event.
HPI - 600 gpa IPI - 3250 ga, including 100 gpn recirculation BS -
1600 gpa EFW -' EPP-1 not actuated with the proposed modification-INTERMEDIATE IEtEAK IDCAs II*a-diate break IDCAs are defined as those break sizes whidt result in a final event cordition of severely depressed primary system pressures to the point of IPI flow, but do not exhibit the phenomena a==neiated with a large break IOCA (i.e. blowdown,. refill, and reflood) . Thus, during the early portions of the event,.the primary and secondary systems are still coupled.
211s nears that EFW flow will result in heat transfer early in the event. Once the primary system levels decrease below the secondary sufficiently, the only heat' transfer which can occur is between the secondary systen liquid and the primary systen steam. itan the primary systen depressurizes below the secondary _ system pressure, the heat transfer frun primary to secondary is -
reversed and the secondary systen acts as a heat source and not a heat sink. !
At'this point, the fill rate:in-the steam generators will easily be met as very little'staaming.is occurring. As the event evolves frun a secondary cooling event into a ence through oore cooling with HPI and IPI, the EFW punp
< trip modification will ocne into play e.nd retuove the EFW pungs frun the EDG.
The EFW pump will be removed frun the EDG before the IPI pung loading incrmaaaa
.significantly. Thus, the EDG loadings for this event will range frun those for the small break IDCA, to be diarmaad later, to the loadings for the large break IOCA, described earlier. The caly differences are the timing of the equipment actuations and not the actual loadings tbamam1ves.
Due to the extended nature of the events, the ESAS high-high reactor building pressure setpoint will be reached later in the event, as empared to large break IOCAs. A review of the various IDCA analyses for this type of break size -
indicates that the BS will normally be required prior to IPI flow initiation.
Therefore, the aammption that BS is a load durity the entire scenario is a conservative ====rtion and will lead to higher diesel loadings. The resultirg equipment flows following the depraammizaticr1 to the IPI flow range are sunnarized below.
2 HPI - 600 gpn IPI - 3250 gpn BS -
1600 gpa i EFW - No flow. Tripped due to IPI flow and EDG loading.
i-k L: .- - - - _ - ----m----- - -- - - _ _
-l 51-1370352 00 smtL amx IOcas This clama of events include those break sizes which result in primary pressures above the IPI cutoff head'and primary levels which runnin within the . i tube region of the steam generator. With a loss'of offsite power (ICOP) at the
- initiatics of the event the rods will fall, the RC punps will trip, the MFW pumps will trip, the diesel' will start and sequencing of loads will p.veed.
The event will result in varying rates of primary system depressurizations, depending on the break size. The secondary and primary systens will still be' ocupled since the primary levels will rammin within the steam generator tube region. EFW flow is desired and required for these break sizes. Based on
. Reference 2, the limiting break is the 0.04 ft2 break. For diesel loading considerations, this will.also be.the limiting break since EFW is actuated and the. primary pressure is rudimd, thereby allowing higher HPI flows. The
-ptimary system response results in a pressure of 1000 to 1100 psia within the-
. first 15 minutes. The IPI system although actuated will run in recirc mode.
The HPI pung will be puuping at a decreasing pressure during this time, although the majority of the time will be at a pressure of about 1100 psia. It can be conservatively assnraed that the HPI flow is 600 gpn throughout the event. The EFW system will be operating,' and amanning the motor driven punp shares the total load for SG feed,. the flow will be required based on the boiloff plus the ' additional flow r='a==ary to raise levels to the 65% point on the operating rarge. The figures given on the following pages provide scene insight into the EFW flow required. These figures are taken frca analysis of the small break IOCAs in support of the B&W Owner's Grmp Abnormal Transient Operator Guidelines (ATOG) ww.am, Reference 5. Based on the figures, the EFW flow, per generator, which will be required for the 0.010 and 0.015 breaks is apprescinately 550 gpn (450 gpm plus 100 gpn recirc). This amnits for both the boiloff and the level increase. Since these break sizes result in more steaming due to the delay in the primary pressure drop below the secondary pressure ocupared to the 0.04 ft2 break, the EFW flows for these will bound that expected for the 0.04 ft2 break. Thus ====ing flow rates' consistent with these will ensure a conservative EFW flow for use in loading analysis. These figures are taken frun Reference 5. Billding pressure is not expected to exceed the high-high reactor biilding pressure setpoint during the first 30 minutes of this event, so the building spray punp should not be actuated and its load is not included in the diesel loading for this reason. This jr-(--it is haama on buildirg analysis performed for small break LOCAs for the another 177 FA plant, Reference 3. The analysis showed for a 0.04 ft2 break, the 30 l psi setpoint would be reached at about 2000 seconds. Since the analysis was !
performed ===4ng a smaller cx:ritainnent volume and lower capacity building fans than the Crystal River 3 cxmtainnant (Reference 4), the analysis is conservative. A later spray actuation would be expected at the Crystal River 3 plant.
4 The agirmant required for this analysis arrl the flow rates expected are given i below.
l-HPI - 600 gpm IPI - 100 gpn recirculation flow l
' BS -
No flow without the initiation ;
EFW - 550 gpn
[
, _ - = - - - - - - - - _ - - _ . - _ - - _ _ - _ = - - - - _ - - - - - - - - - - - - - -
., e . . 0:
" ' 0. ,o-
'8-0-
8 0; sl-1170352'00 M ,m CD CD t.O - N 7 9 2~ 2 O O
- -.J I u- L.,
E' E O
-O 1 ; ,
l i
%~
O w
_ 1 t
4 C t.l C. ~
LJ'
=l C!
U- O ,
f L '
o -
i 4 c?
.u._
,- !~
h I
n!. '
.g O
c',a' l'.
~d
, -.J Oi CCI ' ~
VJ v ol -
~E OZ I L._
OCi C
-S i- CD C ~ ;' n a
-t LO
-Z t ~
Ny ~
f4 OZ-J Cl @
i JI; '
~t C -i L. 2 U 2 N b:" -
NC - y
- LJ CC ,
N CD iO O ~
- OW' N C
Wu" L . _ - ~
rw w Cb -
.-i G.J k . e Cg t.
f, C! (
l O- 4 W
- CO I I I I t l O e, ~ .m-J. -
ww* .C .en .m .n, .Q
.< wb VI .. L S / W E'l ' C ',3 M jb
gw----_-..---,,_..__, , ._ _ _ _ --- .
'}
'a O
O O-O O
!: R cn R
cn Il - 13 7 0'3 5 2 00 i e.- n- i
.q .y 1 1 ' 1 O O J J
,g
'r :
- Lt.
E u'
E j o )
O w "'
I 1 l t 4 jQ O.
se.e
= . j w i i
,,,, t4
-p C
LJ I
I
> l C
La d
c - -
O, m -
C O C
~: >W
-_s . 1 O
.J 3 J C L, t 5g-i D -
g _.
A e - L.
3 6 E; C
'I CCt O
Cj .
. C)
(,0
, ~ C'
- I !
% F -
w Mui y O .' l JZ ^ {*
~
.n s "i ed.m
&rp -
- NC *
- L.l ,
',. CC &
NO N O N OO N rc o N N Ot) #
g -
- m. ~ -
N C g 3
CN o
~~
.Lw ()
[f f,n O vI l O<
c tn Gs go i ! I I i g a.
.. .. .i . ..
._ . r. c. .. .
.n, 9
"I e ytn-cf ao nb s v s ;. >
I< a vJu t'
, u .
51-1170352 00-
,y< IDSS OF MAIN FEEDIOLTER (IDFW)
U '
This event dan ocapled with'a loss of offsite power is similar to the ocuplete
. loss'of unit AC power as dancribed in the Crystal River 3. FSAR. The event is initiated with the loss of station AC power. This results-in a' turbine trip, a
' trip of the control. rods due to loss of holding power, a trip of: the reactor
' coolant pungs, and a trip of the main.feedwater pungs (usually based cn tho'
. loss. on condensate booster pumps) .- Ramad on the loss of the AC power, the EDGs will start. The makeup puup will be inadad to ensure long term' cooling of the -
-RC pumps' seals and the core is provided. The EFIC will initiate EFW with both the turbine driven and motnr. driven pungs starting.- Since the event is an -
overheating event.without releases to the contaiment, the HPI, IPI, and BS will not be actuated. The EFW will begin flow and raise levels to the natural
. circulation point of 65%;on the operate range. The analysis of the average EFW 'l pump loading by GCI used decay heat values which are ocnservative~and bound '
. that which would be calculated haamd on 1979 ANS 5.1. Since the ==4--- 1 it loading for this event is less severe than that for the small break IDch, this event will not +=i-it a lianiting condition.
The a? != ant required to be 4
functioning for this event are listed below.
HPI -
Makeup flow continues since HPI not initiated IPI -
No flow since it is not initiated BS -
No flow since it is not initiated EFW -
Flow rate depa.6L upon decay heat level as shown in GCI
<b ==nt STEAM LINE BREAK EVDrr The steam line break event is a severe overcooling event caused by the failure of a main steam line. The increase in steam flow and the depressurization of the secondary systen incraaaaa the primary to secondary heat transfer. This will lead to a primary system ocntraction. Amannning a IOOP at the initiation of the event, results in a trip of the reactor, trip of the BC pungs, trip of the main feedwater pungs, and trip of the turbine. The EFIC system will initiate EFW based on the loss of the RC pungs. The break will also result in EFIC isolating the steam generators based on low SG pressure. This signal will occur nearly instantaneously as the pressure wave frcan the opening of the break moves through the secondary systen. The EFW flow will be controlled to raise the SG level to the natural circulation setpoint of 65% on the operating range.
If the break is inside cantalment, EFW is fed to the unaffected SG while the affected SG is allowed to boil dry. If the break is outside contalment, the break will be isolated fran both generators, and EFW will be delivered to both SGs. The EFW flow required can be calculated based on the fill rate desired
.and the staaming expected. Using an area of approscimately 66 ft2 for the secondary tube region and dowrmnar, it will take approximately 330 gallons of water to raise the level 8 inches. Since the maximum fill rate for EFW is 8 inches per minute, the fill rate alone will require a flow of 330 gpn. The steaming rate can be estimated as 470 gpn total based on a 3% FP decay heat a--tion. This assunption is justified since the decay heat will have i w i2 to approximately 3% within the first 2 minutes, Reference 6. The 470 gpn can be calculated by ===4ng an approximate 1100 BIU/lba enthalpy increase fran EFW injection until it is released as saturated steam thraxjh the MSSVs. Thus the total flow required is a sua of these two values plus a
[ .
51-1170352 00 recirculatical value. Be recirullatical flow rate is taken as 200 gpn for i either one or two punps operating. %erefore, the total required flow rate is 1000 gpn (330+470+200) . With the two EFW punps sharing the flow dmand, the motor driven EFW puup flow will be half of this total, or 500 gpo. Although this flow rate will decrease during the event, aqwially after the natural circulaticm level has been reached, it is ccriservative to aamima a ocostant I flow of 500 gpn for the event. If the turbine driven EFW punp is assumed to fail, instead of the "B" EDG, the motor driven pung will be required to provide the total desired flow. However, the flow frcan one EFW pung to one SG, with the other isolated, is 940 gpn including a 200 gpn recirculation flow. mis l limitation is due to a change in the EFIC logic which provides flow limitation for overcooling and tube inpie-nt concerns. mus, for the failure of the turbine driven punp, the flow demand frcan the notor driven punp will be 940 gpn. .
For a steam line break outside ocmtainnent, both SGs will be available after isolation, so the required flow will increase by the fill rate requirement on the extra steam generator. %e total required flow 'vte will then be 1330 gpn (330+330+470&200) . Splitting this flow between two EFW punps will place a flow requirement of 665 gpn on the motor driven punp. If the turbine driven punp is assumed to fail for the same break location, the motor driven punp will try to l fulfill the needs of both steam generators for both fill rate and staaming. i However, the' motor driven punp will be limited by the total flow available frun the EFW system hamed on previous GCI analysis of the EFW systan. mis flow rate, at full secondary systen pressure is approximately 1070 gpn. mus, the motor driven EFW punp will have a flow rate of 1070 gpn for the case of a steam line break outside containment with a failure of the turbine driven EFW punp.
Since this flow rate does not meet the EFIC demand, the refill period will be slightly extended. B e fill rate is within the allowable EFIC band at all times, even with the EFW flow rate less than the demand.
A review of the mass and energy relaaaaa for this event was made to determine if the BS is actuated. Raaad on the conservative FSAR ammmpticris, BS would not be expected for this event. Additionally, an imediate MEW pung trip coincident with reactor trip and RC punp trip at the beginning of the event ensures mass and energy relaaaaa to the contairinent will be less severe than that predicted in the FSAR. Consequently, the reactor building praamtre will remain below the high-high reactor building pressure setpoint.
%e HPI flow can be conservatively assumed to be at the mavimm flow of 600 gpn while the systan is depressurized. In actuality, the HPI flow will be depe= Tant upon the systan pressure and will ocntinue to decrease during the i event. Once the primary system repressurizes, the HPI flow will be haned on j the a systan pressure of approximately 2500 psig. %e a=mertion by GCI of a q change in HPI flow at 10 minutes is conservative for diesel loadings arrl will )
bound that expected for this type scenario. %e equipnent required to operate )
during this event are listed below.
HPI - < 10 minutes 600 gpm
> 10 minutes 295 gpn 100 gpn recirc IPI -
BS -
No flow since not actuated
'I
51-1170352 00 EFW - Break inside containment 500 gpn Break outside containment 665 gpn I Break inside w/ EFP-2 failure 940 gpn break cutside w/ EFP-2 failure 1070 gpm \
F m EATER LINE BREAK h feedwater line break event is very similar to the IOFW event. Again since this is an overheatify event, the HPI and IPI will not be initiated unless the break is inside ocntainment. If the break is outside containment, tha HFI and IPI pumps will not be actuated. h mass and energy relaawa for this event (
are insufficient to cause the building sprays to actuate. hast.cicted HPI flow for this event will result in the primary system remainirs pressurized at the pressurizer safetv valve setpoint aM the system will eventually go solid.
The HPI will then pung against this pressure until operator action to cool the plant down ard centrol HPI flow. EFW will be initiated on low steam generator pressure.
signal. The Also, the feedwater and steam syste s will be isolated by the same -
flow frcan EFP-1 will be controlled by EFIC to raise the steam generator level of the unaffected SG. Iong term cooldown will be based on the (
cooldown decay heatrate desired by the operator, the atmospheric dunp capacity and the levels.
below.
h a?H mant required to operate for this event are listed HPI - Flow hw=1 on RCS pressure if break inside containment IPI - If break inside containment, recire flow. No flow if break cutside containment BS -
No flow since not initiated EFW - Flcw based on demand for steaming ard leve.1 increase -
STEAM GENERATOR IUBE RUPIURE (SGIM The steam generator tube rupture event is a small break IOCA frun the primary system to the seccndary system. The event is different frun other SBIOCAs in that the containment building is bypassed. This results in higher doses for this event than others. The event is analyzed to determine the length of time r=caly to terminate the leak and prevent further radioactive releases to the atmosphere through the lifting of the E SVs. A rapid, controlled primary depressurization is desired to minimize the primary to secordary leak f1cu and to permit the isolation of the affected steam generator without further MSSV lifts. The FSAR analysis was a sophisticated hand calculation which was used to determine the lergth of time for the radioactive releases and therefore the total amount of radioactivity released to the atmosphere. The operators were a===4 to take ccntrol of the oaoldown early in the event and cooldown at the emergerry limits to a primary pressure below the lowest ESV setpoint. Once the primary pressure was r e vwi, the affected steam generator was allowed to fill, although the releases were then terminated. During the cooldown, the makeup /HPI flow was naarini to offset the leak flow aM the contraction of the primary system. The EFW systen was required to supply flow to the unaffected steam generator to allow the cooldown to continue. The operators were to manually reduce the secondary pressure by either the condenser dunp valves, if the condenser was available, or the atmospheric dtmp valves if the aardenser i
lo
h' f
51-1170352 00 l
i was unavailable. Assuming a I.00P, the operators would be required to use the l a W ic dunp valves to cool the plant down. l During this scenario, the plant trips, the RC pungs trip, the main feedwater puups trip, and the turbine trips at time zeru as a result of the ICOP. The diesel will be started and the makaup punp will provide flow as it is Irwhi on the diesel. Since initially, the makeup punp cannot matd1 the volume lost out I the tube rupture, the prirary system will deraisiiurize to the 1500 psig ESAS j setpoint. The makeup pung will shift to HPI mode and the LPI punp will start in recirculation. At scane pressure, the HPI wdl matd1 the leak flow and maintain the primary pressure at a nearly constant value until the operator can take action to cool the plant down. It can be acreervatively assumed that the HPI punp runs at mav4== flow, although in reality the punp will be at a scanewhat lower flow. Based on the loss of RC punps, EFIC will start EFW. For this event, the EFW system is required to feed the steam generators to ensure the decay heat removal and the fill rate requirements frran EFIC are met. I Earlier the fill rate was calculated as 330 gpn per generator. With the break !
faariing one generator, it is assumed that the break will be equivalent to 1 330 gpn EFW flow. Therefore, the EFW flow requirements for this event are l 1000 gpn: 330 gpm fill rate, 470 gpn heat renoval, and 200 gpn recirculation. I If the turbine driven EFW punp is failed, the motor driven EFW punp will be required to provide all the flow, otherwise it will provide only half of the flow, 500 gpn.
Since the break does not release mass and energy to the reactor building, the building spray punp will not be started. Also, since the primary pressure for {
this event will runain above the cut-off head of the IPI pung (the HPI can i match the leak flow at lower pressures) the IPI p.mp will remain in I recirculation.
The required aqdynant for this event are listed below.
HPI - 600 gpn 1 IEI - 100 gpn recirculation flow BS -
Never initiated .
I EFW - 500 gpm for both EFW punps 1000 gpn if the turbine driven pump fails
)
l l
4 4
II
51-1170352 00 References -
Revisia) 3", Dated July 1977.
- 2. BAW-1976, "Amall Break Ioss of Coolant Accident Analysis for B&W 177FA Iowered Icop Plants in Response to NURB3-0737, Itan II.K.3.31", Dated SeE* amber 1986.
- 3. 86-1103119-00, " Containment Risipcisse to a Rmall Break IDCA", Contract 620-0012,0013, Thak 04-001-001, July 10, 1979.
- 4. 86-1103119-01, "Contalment Rimipcimie to a Ama11 Break IDCA", Contract 620-0012,0013, Task 04-001-001, %2 4,1979.
- 5. 32-1159004-00, " Task AS-4 Operator Acticals/N Cire", Contract 582-7454, Task 999, February 1987.
- 6. 32-1150622-00, "ANS 5.1 (1979) Decay Heat Envelope", Contract 582-7151, Task 502, September 26,.1985.
l l
l 1
l2.
51-1170352 00 A;perriix A EFP-1 Trip MAR For Infonnation Only i
9 l
l ii 1
l l'
t
c - - - - ~
s . . . ~
. .. w . ar L cu.w-IRIP GIRCCII MODIFICATION SAFETY EVALUATION MAR NO. T87.10 . 04,, 01
-- Shast 1 Of 3 SAFETY EVALUATION:
(use attachment if necessary). Answer the following questions 00 and 1.
Is tne probability of an occurrence or the consequence of an accident or malfunc of equipment important Report, INCREASED? - YE5 to safetyNO X as previously evaluated in the Final Safety An Because: i See Attached Sheet
\
2.
Is the possibility for an accident or malfunction of a dif ferent type than any previously Because: evaluated in the Final Safety Analysis Report. CREATED? NO X YES See Attached Sheet
' 3.
Is the margin REDUCED? YES of safety, NO X as defined in the basis for any Technical Specification ,
Because:
.i See Attached Sheet f.! CENSE REVf5!ON REQUIRED: Final Safety Analysis Report:
Tecnnical Specification: YESX NO NRC Authorization for Change Required:
YES _ NO _A Semi Annual Reporting to NRC Required:
YES - NO 7 YES[ NO 3 ICCFR30.39 CHECK!.!ST Does the proposeo action cnange the Final Safety Analysis Report or require accational description to be added to the Final Safety Analvsis Reoort?
YE5 t )
NO (X )
Notif y Manager, Nuclear Licensing- y
_and Fuels Manaeement s *
, Y_
is a Change to the Tecnnscal Specifications Recurred?
YES( ) y NO (X )
ls any unreviewed saf ety question involved, i.e.,
(1) is tne probability of an occurrence or the consequences of an accicent, _ j or malfunction of equipment important to safety previously evaluated in l tne Safety Analysis Report increased? YE5_ NO X l (2) . Is the possibility for an accident or malfunction of a c_tiferent type l tnan YES any previously evaluated m the Safety Analysis Report created?
NO X (3) is the margin os safety, as defined in the basis for any Technical Specification reduced? YES ,,, NO L
,, Any answer YES ( ) y 6
. 4 All answers NO (X)
Request and receive NRC .
Authorization for chance Y
' Document Change including _
(1) Description of change
, (2) Written Safety Evaluation which provided basis f or items (1), (2), and (3) above.
Authorization Received ( )
Description Safety Evaluation Complete h E .
l Initiate installation of Modification l t
- Recuired enanges to Technscal Specifications snouac ce processed an parallel to tna checklist. Prepared by Name Date
__ ______ _ ______-_---- - - ~
si-1170352 00 _
Florldn Power ANALYSIS / CALCULATION 9 Corcor.t.on i
Crystal River Unit 3 SHEET 2 OF 3 REl/ MAR No.
MAR T87-10-04-01 oate OCTOBER 19,1987
- i""
MODIFY EFP-1 AUTO-TRIP CIRCUIT ATTACHMENT TO MODIP! CATION SAFETY EVALUATION 1.
Tne proposed modification replaces a time delay whose actuation energizes an auxillary relay (BD). The contacts of the BD relay are wired into the START and STOP circuit of EFP-1. The initiation of the timer circuit is started when the diesel generator enters intont.s 30 minute 3000KW rating.
This action occurs for every accident scenario analyzed in the FSAR if the Diesel Generator 'A' is required to operate above its 3000KW load rating. A switch and indicating light is provided in the Control Room PSA Panel for warning and reset functions.
The proposed modification would utilize an AND logic of two relay contacts connected in series through a selection switch to the same relay (BD). One of the contacts exists in a diesel generator loaded auxiliary relay and the other contact would be derived from a bistable which would trip on LPI flow at a predetermined setpoint. The AND logic of this circuit would state
" diesel loaded AND LPI flow AND the select switch in the NORMAL position then energize Relay BD to trip EFP-1".
The consequence of a malfunction of Relay BD will not increase the occurrence of accidents as previously defined in the FSAR. The consequence of a malfunction of a single time delay relay has been decreased by the replacement of that relay with the AND logic contact arrangement. The malfunction of a single component of the logic circuit will not increase the i; occurrence of accidents as previously defined in the FSAR.
- 2. The EFW System does not create accidents of a different type than any i evaluated in the FSAR. A malfunction or single failure of either the 'A' (
Battery, 'A' Diesel, or EFP-1 has already been considered. The 'B' Battery, )
'B' Diesel, or EFP-2 is the redundant system to provide EFW. However, for l large LOCA scenarios, the RC pressure decreases to values allowing LPI J flow. At these pressure values, heat transfer from the Primary to Secondary System cooling is not occurring, and therefore, steam is not available to i drive the turbine driven pumps, i.e. no EFW is occurring. Hence, new accident scenarios are not created. For other scenarios (with an EFP-1 s failure) where steam is available, the turbine driven pump provides EFW flow. {
\
l
)
{
- o. gn ensin.., o.i. v.,,viationingin.., o. . sop.,,,sor. u uci.., ing n..,ing o.i.
j i
Rev 7,81 912244
]
IS~ l
a-1170352 00 !
)
f ph ANALYSIS / CALCULATION 9cu(orek".
Rfl/ MAR No.
tton Crystal River Unit 3 SHEET 3 OF 3 i 1
M AR T87-10-04-01 Date OCTOBER 19,1987 l
P'oi"t :
MODIFY EFP-1 AUTO-TRIP CIRCUIT 1
- 3. i The margin to safety, as defined in the basis for Technical Specifications for diesel generator loading above 3000KW, has not been reduced by the i modification of AND logic circuit to trip EFP-1 on LPI flow coincident with i i
diesel generator loading. Tripping of EFP-1 upon conditions of LPI flow does i not reduce any margin to safety as defined in the Technical Specification.
)
i i
- l A DDITIONA L INFORM ATION The proposed modification for tripping the EFP-1 pump upon initiation of LPI flow has been evaluated for its impact on plant safety. The modification causes an automatic trip of the EFP-1 pump upon coincident indication of LPI flow > 400 gpm and diesel "A" loading. These indications are chosen to define a large break LOCA with a coincident loss of offsite power. Under these conditions, EFW flow i is not required for accident mitigation since the Primary and Secondary Systems are thermodynamically uncoupled.
LPI flow to the RCS cannot occur for RCS pressures above ~185 psi, the shutoff head of the LPI pump. Thus, events which result in primary pressures above this value will not be impacted by this modification. The only event, other than LOCA, which can result in a large Primary System depressurization is the steam line break accident. The steam line break accident, however, cannot result in pressures below ~600 psi without multiple safety grade equipment failures. Thus, consideration of events other than LOCA is not required.
For large and intermediate break size LOCAs, the Primary System pressure will eventually decrease and allow LPI flow. By this time in the scenario. the ;
secondary cooling provided by EFW flow is not required and has stopped due to l extensive Primary System voiding. l l
A small break LOCA will not result in a sufficient RCS depressurization to allow !
LPI flow. A review of the LPI injection line concluded that a break in this i flowpath would not result in a small break LOCA (when EFW flow is desired) with i the LPI pump feeding the break at high flows. Therefore, the only events which l will result in LPI flows are large and intermediate break LOCAs, when EFW flow '
is not required for accident mitigation. This potential scenario was assessed to determine if the trip logic could be initiated for circumstances requiring EFW flo W.
l oesign Engineer oate Verification E ngineer oate Supervisor, Nudear Engineering oate l Rev 7 81 OM Il
W 51-1170352 00 Appendix B EFP-1 Trip Diamanion 17he prgomed meriification for tripping the motor driven EFW pung, EFF1, upon initiation of IPI coincident with a loss of offsite power has been evaluated for its-impact m plant safety. The modification, as described in Appendix A of this report, initiates an automatic trip of EFP-1 upon coincident logic of IP flow >400 gym and "A" diesel generator loading. These indication are chosen to define a large break IDCA scenario with a loss of offsite power. Under these canditions, EFW flow is not required for accident mitigation since heat transfer fran the primary to secondary systems has stivrad or reversed.
A review of other events whid form the licensing basis for Crystal River 3 has been made to determine if other events, besides a IBr.DCA, can result in IPI flows, thereby satisfying half of the EFP-1 trip condition, when EPW flow is still required for accident mitigation. The review determined that the only events which can lead to severely depressed primary system pressures are IDCAs and design basis overcooling su& as a steam line break event. All other
' design basis events will result in only small reductions in primary system prammim, or will result in primary system pressure increases. The steam line i break event, however, will not result in primary pressures below the cutoff i head of the IPI pung, approximately 185 psi, without multiple safety system failures. This is due.,in laije to the limited secondary steam generator inventory av pae'for heat removal fran the primary systen. Worst case steam line break 1inalysis presented in the Crystal River 3 FSAR predict mininum
,predsures of approximately 600 psi. Although this pressure would result in the 4 modification, will not <mm. Therefore, the only event for cmsideration in y [,' / start
. reviewing ofthisthe IPI punp, modification flow into the RCS, which i is the IDCA.
For certain small break IDCAs, the primary system remains pressurized and secondary heat removal is required to maintain core cooling. The break size is too small to adequately remove heat;through its break flow. A detailed review of the IPI piping was made to determine is a break in the flow path could result in a small break IOCA, where EEW is required, but would result in the IPI punp feeding the break resulting in the IPI flow amaading the flow setpoint arid resulting in an EFP-1 trip. The review indicated that due to the redundant nature of the valving arraced., two check valves and one motor operated valve, a break location does not exist whi d could result in a small break IDCA with the IPI punp feeding the break. Therefore, the large break IOCA is the only design basis event whi& can lead to IPI flows arx1 the mihaar=iant trip of EFP-1 aa==ing a coincident loss of offsite power.
As daarvibed in the MAR for this change, this logic will replace the carrant logic whi& provided for an autanatic trip of EFP-1 after a thirty minute time p period of "A" diesel operation. Thus, this modification does not result in the L potential for a new type of failure, an inadvertent trip of EFP-1, beyond that currently configured at Crystal River 3. Additionally, this logic provides for '
EFP-1 to remain loaded during events where its flow is desired for long term ;
cooling. This means the operator is not required to reload EFP-1 after a 30 l minute time period, which would have been required prior to this modification.
I ~l
\ . - . _ _ _ _ _ . ____-___ -
~
.. x '
o n 11 1100 50 006 ,
j
'1 Appendix B l EFP-1 Trip Discussim 1
. 'Iha pn:cposed modification for trippig .the actor driven EFW pump, -EFP-1, upon i initation of IPI coincident with a loss of offsite power has been evaluated .
I for' its i==+ cn plant safety. . 'Iha modification, as described in Appendix A of this report, ' initiates an autcunatic trip of EFP-1 upen coincident logic of
. IP flw >400 gym and "A" diesel generator. loading. 'Ihese indication.are chosen
' to define a large break IDCA scenario ' with a loss of offsita power. .Under I these cmditions, EFW f1w is not required for accident mitigation since heat j
transfer. frun the primary to secondary systens has stc5 ped or reverried. !
A review of other events whim form the licensing basis for Crystal River 3,has been nede to determine if other events, besides a IBIDCA, can result in IFI 1 flows, . thereby satisfying half of the EFP-1. trip condition, when EPW flw is l still required for. accident mitigation. 'Ihm review determined that the enly I events whi& can lead to severely depressed primary system pressures are IOCAs and design basis overcooling such as a steam line break event. .All other design basis events will result in only small reductions in primary system i
pressure, or will result in primary system pressure increases. 'Ihe steam line break event, however, will not result in primary pressures below the cutoff head of the IPI punp, approximately 185 psi, without m21tiple safety system failures. 'Ihis is due largely to the limited secondary steam generator inventory available for heat renoval frun the primary system. Worst case steam l line break analysis presented in the Crystal River 3 PEAR predict . minina pressures of approximately 600 psi. Although .this pressure would result in the l start of the IPI pump, flow into the RCS, which is maamned for this modification, will not occur. 'Iherefore, the only event for oansideration in reviewig this modification is the IDCA. ,
Ebr certain small break IOCAs, the primary system remains pressurized and soudary heat renoval is required to maintain core cooling. 'Ihe break size is too small to ariarymtaly remove heat through its break flow. A detailed review of the IPI piping was made to determine is a break in the flow path could result in a small break IOCA, where EFW is required, but would result .in the j IPI puny ' feeding the break resulting in the IPI flow avr== ding the flow j setpoint and resulting in an EFP l trip.. 'Ihm review indicated that due to the '
redundant nature of the valving arrangement, two deck valves and cre motor operated valve, a break location does not exist whis could result in a small break IOCA with the IPI pung feeding the break. 'Iherefore, the large break IOCA is the only design basis event whi& can lead to IPI flows and the j anhaarymnt trip of EEP-1 ==ing a coincident loss of offsita power.
As <iaar ribed in the MAR for this change, this logic will replace the alrrent ,
logic whi & provided for an automatic trip of EFP-1 after a thirty minute time !
period of "A" diesel operation. 'Ihus, this modificatim does not result in the potential for a new type of failure, an inadvertent trip of EFP-1, beyond that l currently configured at Crystal River 3. Additionally, this logic provides for !
EFP-l' to remain lemtimel during events where its flow is desired for long term cooling. 'Ihis means the operator is not required to reload EFP-1 after a 30 J 4
minute time period, which would have been required prior to this modificatim. j l
l l
4
)
_ _ _ _ _ _ _ . _ _ _ _ _ _ _ _ . _ . _ . _ _ . _ _ . _ _ _ . . - .