ML20024B957
| ML20024B957 | |
| Person / Time | |
|---|---|
| Site: | Palo Verde, Arkansas Nuclear, Waterford, San Onofre  |
| Issue date: | 07/31/1983 |
| From: | Abramson P, Komoriya H ARGONNE NATIONAL LABORATORY |
| To: | Office of Nuclear Reactor Regulation |
| References | |
| ANL-LWR-NRC-83, ANL-LWR-NRC-83-6, NUDOCS 8307120047 | |
| Download: ML20024B957 (111) | |
Text
% t _
PDR.-
h* b g- , - i E P
%. 3 ANL/ LWR /NRC 83-6
[~. [g3 j July 1983 w,
g Decay Heat Removal During a Total Loss of Feedwater Event for a L-E System 80 Plant H. Komoriya P. B. Abramson Light Water Reactor Systems Ar;alysis Section Reactor Analysis & Safety Division ARGONNE NATIONAL LABORATORY
/
9700 South Cass Avenue v Argonne, Illinois 60439 Prepared for:
Division of Systems Integration Office of Nuclear Reactor Regulation U.S. Nuclear Regulatory Commission Washington, D.C. 20555 NOTICE: Tnia infornal document contains preliminary infornation prepared primarity ,
for interim use by the Office of Nuclear '
Reactor Regulation, Nuclear Regulatorjs q ggg Conmission (NRC). Since it does not constitute a fir.at report, it should be Ah s A\
cited as a reference only in special RATOR circumstancec, such as requirements for '""
regulatory needs.
%J l
i 8307120047 830731 PDR ADOCK 05000361 P PDR -
4 - C
1.0 INTRODUCTION
This report presents the results of a series of RELAPS/M001.5 calcula-tions to examine the ability of a C-E System 80 plant to avoid core uncovery during a total loss of feedwater (TLOFW) event. The study included parame-trics with and without concurrent loss of Offsite Power (LOOP), considered the use of the C-E Auxiliary Pressurizer Spray ( APS) for rapid depressurization, and examined the potential value of installation of a Pilot Operated Relief Valve (PORV) as a direct manual method of depressurization and feed and bleed decay heat removal . There are two C-E plants in operation without PORVs: San Onofre-2 and ANO-2, and C-E System 80 plants under construction are Palo Verde-1, -2, and -3 and Waterford-3.
The three major areas of transient analysis addressed in this study are:
- 1. Study the impact of TLOFW (with and without LOOP) with and without operator recovery actions as stated in the C-E Recovery Guidelines O) 2. Perfonn feed and bleed sensitivity studies with respect to PORV sizes and PORY opening time.
- 3. Identify the latest time when AFW can be actuated to avoid core uncovery.
Detailed RELAPS/M001.5 modeling descriptions and initial plant conditions are presented in Chapter 2 of this report. Chapter 3 contains the transient cnalysis assumptions and scenarios. The results and discussions are contained in Chapter 4, and the summary and conclusions are presented in Chapter 5.
Also included in Chapter 5 is a brief discussion of modeling uncertainties and their impact upon the conclusions.
. _ _ . . _ . . _ . . . _ . . _______.__J
4 -
o I
l l
2.0 PLANT DESCRIPTIONS AND INITIAL CONDITIONS O
ANL initial plant conditions for the TLOFW transient analysis are com-pared with C-E data in Table 2.1. These represent 100% full power steady state conditions for a C-E system 80 plant obtained with RELAP5/M001.5 cycle 29 with updates. ( All calculations were performed on the INEL computers.)
During preliminary calculations, the results with RELAP5/M001.5 Cy=29 exhibited significant mass error accumulations in the regions where saturation conditions existed. Code updates provided by the RELAPS code developers were incorporated into Cycle 29 which resulted in a dramatic reduction in mass errors. Calculations reported in this study were performed with Cycle 29 with these updates.
The plant nodal diagram is shown in Fig. 2.1. The system-80 plant is designed with two cold legs per loop and thus also contains four reactor coolant pumps. In the analysis the cold legs were combined in pairs and each cold leg and RCP delivers a combined flow of two plant size cold legs.
O Pressurizer safety valves were similary combined such that one valve had an equivalent flow area to yield a combined steam flow of two valves at the rated pressure. Similarly, when two PORVs were assumed in the analysis, they were simulated with one valve sized to yield a combined rated steam flow at the rated pressure.
The safety engineering features modeled in the analysis are:
- Steam Generator Safety Valves Actuation: Steam line pressure > 1282 psia l Capacity: 20 valves 4 valves 1270 psia 4 valves 1305 psia 12 valves 1333 psia Total capacity 19.0 x 106 lbm/hr g% ,
4 . e e Pressurizer Safety Valves Actuation: Pressurizer pressure > 2525 psia Capaci ty: 2 valves 255.56 lbs/sec e Charging Pump Actuation: started by the operated after TLOFW Capacity: 6.90 lbs/sec (1 pump), 120*F e Auxiliary Spray System Actuation: Hot leg subcooling > 25'F; open.
Hot leg subcooling < 20*F; close.
Capacity: 6.90 lbs/sec (1 charging pump capacity), 120*F e PORVs Actuation: Opan by the operator Capacity: 2 valves 119.71bs/sec (Calvert Cliff type) 254.0 lbs/sec e Accum"htor Injection Actuation: Cold leg pressure < 624.7 psia 1
Capacity: 1536.7 ft3 of liquid volume (per tank) e Auxiliary Feedwater Pump Actuation: Started by the operator Capacity: 121.27 1bs/sec, 100*F l
l 1
h
lI !
]
l r
c
\ m
@ i i igdz " i Y 0
8 8
T 78 ~
1 8
'c 2 2 2 2 h C C 1
0 2
8 p
%iT-1 i 6
0 0
2 c
@ o o
2 L C
r [
o f
r r '
~ i .
2 t 9
5 2 c 2 0 n 8
2 7 2
7 8 -
6 7
2 T# ' 2 c
i l
P a
2 S
- 2s 2 c
0 8
Ig 0
iJ 21
" m 0 2 c
e 2
c k J d, t s
6 8 d c y 2 ,' S 4:
2 z 3 s E c c 2 -
0 c C l 3 4" ^
I '
f g
C j o
- :j I
l k {2TT a m
oe H. v n r 6 g 0
0 6
j=
s B
s 1
0 f ghmc)i
( '
i a
6 s 4T 4 s r ; ,3 D
[5 s 3 3 h sa a C k C c .,
n l a
I'p6 l
0T
,dWy' n L
d N
o 6
8 0
4 3
D .I P
>=1'
. F I- 3, g"
' lT; S.
I D
1 C 8 O D
i a
c M
C 4 /
m hs s S i
C i
s P l c i l
dI c A 2 L 6
1
( s J2 s
" E
' R 9
6 C c it c r
e p Si O t ll' J " .
5 t C S SI o
8 I s 1 C 2 4
C S l
- i c
+ C I
2 g@@ b,~
- 2 O
c i
5 g 7
i o h C" O c i
F l
T -
I C = 3 b @
4 G'
g ,
1 p
0 8i 8 I T
? g o, g o 2 i @ o o
p a
C il C C I
c L
G' hj'.t V r
r i
c s 2 4 6 o 71 7 1 ii t T l c
i 3
} a C"Ns=.,.
[)-
i 4 C .
18 .
C -
l jIf \ ! i!l x
l1 ,!l;;jl
- . l,j1 '
, j ll ll lll I L*' !l
e e l
l T Table 1. CESSAR NSSS Component Thermal and Hydraulic Parameters Plant Steady State Nominal Initial Component Cor.ditions Conditions Reactor Vessel Rated core thernal power, MWt 3,800 3,800 Operating pressure, lb/in.2a 2,250 2,250 Coolant outlet temperature, 'F 621.2 621.6 Coolant inlet temperature, 'F 564.5 566.0 Coolant outlet state Subcooled Subcooled Total coolant flow,106 lb/hr 164 164 Core average coolant enthalpy Inlet, Btu /lb 565 565 Outlet, Btu /lb 645- 645 Average coolant density Inlet, lb/ft3 45.9 45.8 Outlet, lb/ft3 41.2 41.1 Upper head recirc. path flowrate, lb/s 319.4 320 Steam Generators Number of units 2 2 Primary Side (tube side)
Operating pressure, psia 2,250 2,215 Inlet temperature, *F 621.2 621.9 O Outlet temperature, *F Secondary (shell side)
Steam pressure / temperature, psia /*F 564.5 1070/552.86 565.7 1070/552.86 Steam flow per gen., lb/hr 8.59 x 10 6 8.59 x 10 6 Exit steam quality, % 99.75 99.75 Feedwater temperature, 'F 450 450 (at 1070 psia)
Recirc. Ratio 3.25 3.27 i
l e , 1 3.0 TRANSIENT ASSUMPTIONS The transient was initiated following 100 seconds of steady state 1 calculation in the analysis. The figures presented in the next section also include these 100 seconds, however, the time discussed in the text of this report refers to the transient time rather than the computer run time.
Two sets of transient conditions are introduced in the calculations:
- 2. TLOFW without LOOP at t = 0.0 s.
Under each set of transients, the following scenarios are studied:
- 1. a) TLOFW with operator recovery action based on the C-E Recovery Guidelines, b) without
- 2. TLOFW with initiation of feed and bleed.
- 3. TLOFW with actuation of auxiliary feedwater flow.
The general transient assumptions are:
e One charging pump (6.09 lbs/s) was started at 10 minutes after TLOFW.
- APS (6.09 lbs/s) initiation by the operator based on the Recovery l
l Guidelines.
- Only one HPSI pump was available, e HPSI actuation on SIS (1600 psia) when no operator recovery action is assumed, otherwise by the operator at 10 minutes after TLOFW.
e o
e
-=4 *-- - ---e,.e.a__ , . , , . . _
-- a _ - - - - .
l a ,
I e Two PORVs were modeled in the feed and bleed sensitivity study with PORV size and the PORV opening times as variables. Two PORY sizes were simulated:
Nominal size: 119.7 lbs/s (113.4 lbm/hr/MWt)
Giant size: 268.1 lbs/s (254.6 lbm/hr/MWt) of steam at 2400 psia.
- AFW initiation by the operator Trip setpoints and the associated delay times are summarized in Table ,
3.1. There are four operator actions simulated in the RELAPS calculations Table 3.1 Reactor Protection System Trips Used in the Analyses Trip Setpoint Delay time, s High pressurizer pressure, psia 2475 0.55 Low pressurizer pressure, psia 1600 ---
Pressurizer safety valves, psia 2525 ---
Steam generator safety relief valves, psia 1282 to open ---
1218 to close ---
RCP trip on signal 0.55 Turbine trip on signal 0.55 based on the C-E Loss of Feedwater Recovery Guidelines; CEN-152 Rev. 01 and a're the following:
- 1. Auxiliary Pressurizer Spray
^
e on if the hot leg subcooling > 25*F e off if the hot leg subcooling < 20*F, or the pressurizer level > 90% ,
liWFT v
lL . .__ __ _ _ _.. _ _ _ .. _ _ _ _
e - , l O 2. HPSI e on if the pressurizer level < 100 inches, or the hot leg subcooling < 20*F ;
e off if the pressurizer level > 100 inches and the subcool- l ing > 20*F During the time the operator takes control of the HPSI system, the injection is actuated only based on the above criteria, and not by the automatic actuation on a low primary pressure setpoint at 1600 psia.
- 3. PORVs e opened by the operator and lef t open throughout the tran-sient.
- 4. Auxiliary Feedwater e actuated by the operator and left on throughout the tran-sient.
Cases analyzed are summarized in Table 3.2. For the TLOFW with LOOP transient, the reactor trip, turbine trip, and RCP trip occur at 0.0 second due to LOOP, and the feedwater is lost at the same time. Whereas, for the TLOFW without LOOP transient, the loss of feedwater occurs at t= 0.0 s and a short time later, the reactor trips on either high RC system pressure or low steam generator level. At that time the turbine is isolated but the coolant pumps are still operated until they are tripped by the operator as soon as the TLOFW is detected to minimize heat input into the RCS. In the analysis RCPs were tripped at 10 minutes into the transient.
In the feed and bleed sensitivity study, the only operator action simu-lated was opening of the PORVs and leaving them open for the remainder of the transient. Sensitivity to PORV sizes and the initiation time was examined.
Attempts were made in both transients to estimate the latest time at which the operator could open PORVs without resulting in core uncovery. Similary the sensitivity to the AFW flow was studied by detennining the latest time for the p actuation of flow.
A 1
DRMI
i l
Table 3.2. Sumary of TLOFW Transient Cases TLOFW WITH LOOP TLOFW WITHOUT LOOP Transient Description Transient Description Case 1 -- Base Case Case 2 -- Base case ho operator actions No operator actions HPSi on at 1600 psia HPSI on at 1600 psia No APS No APS RCPs tripped at t=10 min Case 3 Case 4 Operator Action at t=10 min Operator actions at t=10 min APS, HPSI APS, HPSI RCP tripped at t=10 min Case 3ai -- Small PORVs Case 4ai -- Small PORVs l No APS No operator actions HPSI on low RCS pressure on at RCP tripped at t=10 min t=10 min PORVs opened at t=10 min PORV opened at t=20 min Case 3bi -- Giant PORVs Case 4bi -- Giant PORVs No APS No operator action HPSI on low RCS pressure on at RCP tripped at t=10 min t=10 min PORVs opened at t=20 min PORY opened at t=20 min .
V Case 3aii -- Small PORVs Case 4ati -- Small PORVs Find the latest PORV opening 1) PORY opened at t=10 min time Find the latest PORV opening time Case 3bii -- Giant PORVs Case 4bfi -- Giant PORVs Case li -- AFW Case 21 -- AFW Find the latest AFW initiation Find the latest AFW initiation time time o DRAFT
4.0 CALCULATIONAL PESULTS AND ANALYSIS In this section we present the detailed results and discussions on four cases related to the ability of the APS to depressurize the system (cases 1, 2, 3, and 4) and eight cases related to the feed and bleed mode of depressur-izing the system (cases 3ai, 3bi, 4ai, 4bi, 3aii, 3bii, 4a11, and 4bii). Also discussed in this section are two cases investigating the latest effective time for restoration of auxilicry feedwater (AFW) flow (cases li and 21).
These cases cover three areas of the transient:
- 1) Total loss of feedwater (TLOFW) flow with and without operator recovery actions.
- 2) TLOFW with initiation of feed and biced.
- 3) Restoration of AFW.
Analysis is focused upon detennination of the effectiveness of CE's operator recovery guidelines and analysis of the relative capabilities of auxiliary pressurizer spray (APS), the Pilot Operated Relief Valve (PORV) and AFW as methods of rapid system depressurization.
Results are presented in the following order; first, cases on TLOFW with concurrent Loss of Offsite Power (LOOP) are discussed, followed by those for TLOFW without LOOP. In the next section, cases where the PORVs are utilized are discussed, and finally we examine the sensitivities to PORV opening time and to AFW initiation time. Results are summarized at the end of this sec-tion.
l 3
"n - "o~ -- = , ...
+ - m. - +-,.
4.1 Total Loss of Feedwater Flow Concurrent with Loss of Offsite Power 4.1.1 Case 1 Normal feedwater flow is assumed to be lost as an initiator of the transient, and at the same time, the offsite power is also assumed lost
("A" in Fig. 4.1) . A signal is generated to actuate the reactor trip, turbine trip and the reactor coolant pump (RCP) trip, which initiates RCP coast-down. Throughout this first transient, we assume that no operator recovery actions are taken to mitigate severe consequences; eventual core uncovery is indicated by the presence of both of the following conditions: 1) complete voiding of the upper node in the core; 2) high cladding (and also coolant) temperatures.
Figures 4.1 and 4.2 show that the RCS pressure increases rapidly due to the secondary side temperature increase and concomitant pres-sure buildup due to the loss of cold feedwater into the steam generators, aggravated by the turbine trip.
- This caused degraded primary-to-secondary heat transfer. The steam generator relief valves opened immediately following l
the onset of the transient and remained open as the steam generators boiled away their inventory. Following the delay time associated with the reactor trip, the RCS pressure dropped instantly to 2130 psia then rose immediately to 2220 psia due to power-to-flow mismatch. The RCS pressure gradually dropped to 2150 psia and remained at this level until the steam generators dried out at 2400 seconds into the transient (Fig. 4.3). At this time, the primary-to- ,
secondary heat transfer was completely lost and the primary pressure rose to the pressurizer safety setpoint (2525 psia) and pennitted the RCS inventory to be discharged through the pressurizer safety valves. Small fluctuations in the primary pressure between points B and C on the curve in Fig. 4.4 corres-ponded directly to the times when the liquid level in the steam generator crossed the computational node boundary, hence to the change of heat transfer regimes (Fig. 4.4). With the loss of energy removal capability, the RCS i temperature rose and the coolant expanded, increasing pressurizer level as I shown in Fig. 4.5. We note that around 3100 seconds after the transient initiation the pressurizer became solid (in Fig. 4.5). The pressurizer safety valves opened around 2630 seconds into the transient and, when the pressurizer O1 DfMI O - -, . - . - . , - - --w - 4.m+
went solid, the discharge from the pressurizer safety valves became single phase liquid (Fig. 4.6). Since the primary pressure was greater than the l
maximum HPSI pump head (~ 1790 psia), without operator actions to depressurize to enable ECCS delivery the primary temperature rose until it reached the saturation temperature corresponding to the pressurizer safety setpoint (Fig.
4.7). From that time on, the primary temperature stayed constant while void was generated in the primary system (Fig. 4.8). Core uncovery began at 5800 seconds into the transient (Fig. 4.9), when the RCS inventory became so low that the void fractions in the top three nodes of the core reached 1.0. At that time, the collapsed water level in che core began to decrease rapidly (Fig. 4.10) and the cladding and core outlet flow temperatures began to rise (Fig. 4.11 and 4.7).
\
l i
'- - ~ . - - ~ + ~ .-_-- ._,._-. _,__, ___, _
-7,, - , .-- m, ---,. - v.g,- p n, -- - - , - ,-e--- m. ,,e ,-, 4
o ,
1 PRESSURE-P2R V ""
-J m" .t.. .Jr -
I . I
] j'
, 7 'p..,g,( # r aese
- sene b asse 3
u a3eo
$8-asse ,
3,
\ n c.
\
==
V w
, 3 v anos seet 3ese 40e0 Se00 asse e 3000 Ebee 35ee esse 55e8 M see Itee Tlee to CASE I Fig. 4.1 J
SG DSME PRESSURES: SGl. SG2 lage tame teos taas k
- t== -
- a. y - - . .
E g stee LtM o.
4146 liao 1806 less 1"
lese e se84 4000 3ete 400e Gese Gaet g See t540 mee m isee see see Time ist S
)
'V CASE I -
i 4 l Fig. 4.2
e .
g WATER LEVEL IN SGI I f \
O
- as 3e S \
~
W N is N
\
N bee tsee as44 asas esse ssee ases flee ts:
CASE I Fig. 4.3 V910 FRACTlWN IN SGI s..
I )
,I ,I ,, , ,
I I I,l1 E .., ,
5 ;
g ...
c e.s -
l 1 l 0.4 i
l' b ] l c y 3 e.e e sees asse sees esse sees sees see asse asse 3see esse snee asse Tlee ts) x CASE I r, a Fig. 4.4 /
MASS INVENTORY IN PZR G) =
1 8
en l \H
~
l I f
2 me
- / l 1
t /
5 " l n.
g j sJ %
=
/
i - .
f ase e
= ,A - _. ,
aos
_ l_
e sees asee sees .ees same sees See 35ee asee 3540 eees See see Tlee ist CASE I s Fig. 4.5 .
m PZR SAFETY VALVE FLOW see f! I
~ '
ii" I,
,,, _r c
$N l
M '
~
n a
e h ;i L ,
l'
= .ee ,
i R il i
, = -e p-- ;
l,
, < l l. i -
- i
.se i
! L lL:a JI
, t i ,
,lt . , . i e sees asse 3ees doce seee esse See 19e0 Neo Neo ese Seet GJe N Tlee ist g ) CASE I v
Fig. 4.6
e e C3RE C80LANT TEMPERATURES: BTH. MID. TOP V -
i too see V+ - a
- e /
s g M l r j
.e (
~
sas s
.ee
( - . _
8
/
L 4 b
- 4 see e sees m aeso eseo sees sees let tsee 3844 3584 . Gee EEse 8888 Time (s)
CASE I Fig. 4.7
- ValD FRACTION IN C3RE:T3P 3 N90ES e.s e.a
- e. , --
E e ...
I G
k .3 1
m
~ , ,1 ,
0.&
i e sees asse seee .eee sees sees l
see ine asas mee .see mee ese.
Time ist CASE I
~j l Fig. 4.8 l
1 1
1 l
e .
RCS MASS INVENTORY
% si.
J ...
s..
I 3 ..a g ...
h 3.s k
$ 3. .
..s N .
..N
.s
. sees amee mees .see s.se sees 54* ssee asse 3see 4see ssee s fl.e (s)
CASE I
% Fig. 4.9
)
CSLLAPSED WATEFI LEVEL IN C3RE
. s..
i...
3 e .a
.... -~
~
$ i...
p ...
i s ...
I t.s
{
1
. i.e. -. n.
U
) ,,.. ..,
CASE I Fig. 4.10
e ,
CLAD TEMPERATURE: TOP N00E 1ee vie
- J ase ese O A- -
sao l
w s - /
~
u j i
u
~
- i
( /
\% 1
/
ese -% -
j see
' I
=
e seee asse sees me sese sees see sees mee asse gee sses ssee Tlas (s)
CASE I Fig. 4.11 N
D i DRAFI
, -- , -,- , . . . - - - ,,-n,- -- ,-4 --n- -n , n,.,-,, -ea
b 4.1.2 Case 3 This transient differs from Case 1 af ter 600 seconds into the transient, af ter which time operator actions based on the CE Recovery Guidelines were simuinted in this case. The purpose of thase actions is to provide alternate me&ns of cooling by activating the APS and HPSI to attempt to maintain decay heat removal capability.
The following operator actions were modeled:
- 1. Actuation of the APS using one charging pump (6.09 lbm/s) with an objective to maintain a range of 20-25'F subcooling in the hot leg. However, the APS was turned off wnen the pressurizer level reached 90% (or more) of the total pressurizer height because beyond that it would merely aggravate pressurizer refilling and would flow out of the safeties directly.
- 2. Actuation of the high pressure safety injection (HPSI) system, using one pump, when the pressurizer level was less than 100 inches or the v
hot leg subcooling became less than 20*F. The HPSI was turned off when the pressurizer level rose above 100 inches and the degree of subcooling in the hot leg was more than 20*F. (Houaver, it is important to note that EPSI mzbmon head is ~ 1800 psi and that for most of this transient the gehn1ry preseure is vell above that value. Thus the HPSIs produced no substantial flou.)
At 10 minutes into the transient, the APS was actuated b;ith the 120*F water) because the subcooling in the hot leg at this point (see Figs. 4.12 through 4.14) was 44.5'F. The primary pressure had decreased from 2157 psia to 1760 psia by roughly 1240 seconds into tne transient, at which j point the subcooling fell below 20*F and the APS flow terminated. The APS was
! actuated on and off four more times in the ensuing 2400 seconds as the hot leg subcooling fluctuated. Each time it actuated, the RCS pressure dipped a little, but the flow had less and less impact as the transient progressed, since by 2400 seconds into the transient, the steam generators were dry and had therefore lost energy removal capability. The steam generators represent l
'~
gy
O the major heat removal mechanism, so once they are lost the primary saturation U pressure no longer decreases. Thus, since APS activation is keyed on the saturation temperature, the APS is not effective. The APS was actuated for the last time around 2830 seconds into the transient and was shut off due to the high level in the pressurizer (Fig. 4.15). As discussed with Case 1, the humps in the primary pressure curve between 1300 and 2400 seconds after the onset of the transient are numerical effects, caused by RELAPS, due to the water level crossing the code node boundaries in the steam generators. As in Case 1, as the steam generators dried out, primary pressure increased to the pressurizer safety valve setpoint. This occurred about 3015 seconds after transient initation. We note that this is approximately 400 seconds later than for Case 1 due to the actuation of the APS flow which depressurized the system roughly 400 psia more than Case 1 (which had no APS flow).
Roughly 1237 seconds into the transient, the HPSI flow was initiated at the rate of a few pounds per seconds when the not leg subcooling fell below 20*F. The flow wa's insignificant however because the RCS pressure was too high for the HPSI pump head to deliver more flow. For the remainder
\ of this transient, the system pressure remained above the HPSI maximum head, and the results became very similar to those from Case 1.
As soon as the steam generators boiled dry, the primary temperature rose (Fig. 4.16) resulting in insurge into the pressurizer with the expanding primary liquid. The pressurizer went solid at about 2950 sec-ands into the transient at which point the flow out of the pressurizer safety valves became single-phase liquid (Fig. 4.17). The PCS inventory was rapidly lost through the safeties (Fig. 4.18), void formation in the core started at about 3880 seconds after transient initation, and by 6030 seconds the core began to uncover. See Figs. 4.19 through 4.21.
)
. ~ . .= .. . = _ _ _ . . . - . . . - _ _ . - .-. -. - .-
4 e , e 4
i RCS PRESSURE PZR I
a
. ~ . , , -m ,, . .
ll l lI l
3 1
i $ "
I us b ^
me. g up .18.
~
4 M g ,
i a
i =
i l e s a s i Y"d 17 i
4 9 19 a 3 a m g 9 Time est CASE 3 l Fig. 4.12 1
i i
i i
DRAFT
- - ***w-w *
-u....m.....wm,,,9_. %. , , . , , , , , , , , ,,,,.
- - - - - - . --. , - - . , - . . , , . - - - . , , , - . . . . , , - - ~ . , . , , , ,_._-s ,,s-
e o I
HOT LEG SUSC031.ING N es "
c) -
ss t !
le es
/3 I
\ l w \
' /\
E"
~
"" \ . / \
" \ r ~W \
i,
\
g
\
3 e
-s l e seso asse sees an mee cow see sees asse sees *ew esse 88e*
Tlee (si CASE 3 Fig. 4.13
'\4 AUX PZR SPRAY FLOW s.s e.o s.s
/ ,
/
s.e j
S 4.s 3
- / [
e.e w
y 3.s
'/ -'
i-DC
/
g 3.e '.
N a.s ,
m y S. .
-/ I x ., .
4.s ,
t.e ,
e.s o.e e tees 7000 300s *ese Sees etee See R$es 3500 3ees esse Sete sete
% Tlee (si
\
CASE 3 l
Fig. 4.14
-1
{
MASS INVENT 8RY IN PZR C ese 4 r T
~ ~
( \fd
~
l l
.5 l
.e,
~ / ,
i l s -
- z 3as
/ NL gig m
h /
- J x , i me
- ,~ ~
/
age e tese asse 3see dose sene esso See 1800 asse 3sse este Gene Wee Time tel CASE 3 Fig. 4.15 CORE CGOLANT TEMPERATURES: BTH. MID. T3P ese l
m fee
~
Tee e ,ee m . ,
in
$ ese Q jfU- :- . .l
~ ///
=
ese
_ y
.e. w. b
~ %_
y .
[
ese e 8000 esse 3 Bee dose Gese esse See toes asse Met dese Wee Wee I Time (si CASE 3 Fig. 4.16
, g
e .
I Veld FRACT!2N IN PZR SAFETY FLOW U -
p p L !
..a ,.
k!
g_q__
g ... . - cf I C l W ... __ .
E e ... j __.
2
..a i
.. i / !
k 1
. tese asse seee ,een gee, go,,
ese im asse asee esse see , em Tle. (s)
CASE 3 s Fig. 4.17 4
, RCS NASS INYCNTORY s..
h +
... s a
y ..s \
g ...
\
E 5
s.s (
(
- ... N s.s <
i s..
N A l N l
n..
- tm em see. .se. n l
) " ** m m Time ts) m m m 3
- I} j
- CASE 3 l
Fig. 4.18 m
e ,
)
,,$ll1LL,APSED WATER LEVEI. IN CSRE, she -
- ia.s a s..
\
u..
sa.e g N A .k[
C g
e p -
N i...
d s.s E
a 6..
s.s e.o vs
. k e seee sees seee eseo seee esse ese tSee see 3544 eless Wese Geee Tlee ts)
CASE 3 N =
Fig. 4.19 V310iFRACTION -tN'C3RETTOP UIN00ES t.e 0.9 0.8 5 ..,
7 X o..
E o .
E q
E
< ._L_
,m 6, m ,
o.
JI
- 9.4 . .m , . =
0 lese 8000 3ose sees tese east See 1988 Sees
- 2 80 asse 3 80 M ASE 3 )
Fig. 4.20
CLAD TEMPERATURE: TSP N30E m
m a
1*
g 3 m 9 /
ta 0 =
g m - f- ,_ ._ .-
W o.o I ,
/
m ( l l
.e. ,
N _
[
see e m ases asse .see see. see.
- m ases asse .ese sees esse flee (si
- CASE 3 s Fig. 4.21 t
I
- - -~ - -- --- . . _ . . ._. ___ _ _ _ _ _ _ , _
l l
e 4.2 TLOFW with Power Remaining Available
(
4.2.1 Case 2 In this case, although normal feedwater flow was assumed to be lost, power was assumed to remain available to the RCPs and the reactor was not tripped concurrent with loss of feedwater. Ten minutes after the loss of feedwater, we assumed that the operator manually tripped the reactor coolant pumps and started pump coast-down (Fig. 4.22).
Without the cold feedwater flow into the steam generators, the secondary side temperature rose (Fig. 4.23) and steam pressure increased while the turbine was maintaining constant 100% load (constant steam flow)
(Fig. 4.24). The primary-to-secondary heat transfer degraded and the primary pressure rose (Fig. 4.25) to the reactor trip setpoint on high primary pres-sure (2475 psia) at 29.8 seconds after loss of feedwater. The reactor trip was assumed to actuate a turbine trip signal. The RCS pressure continued to
- s increase rapidly and ree.ched the pressurizer safety valve s'etpoint, but by that time the control rods were fully inserted into the core and the pressure decreased immediately (spike roughly 30 seconds after loss of feedwater). The secondary side inventory boiled dry much more rapidly in this case (Fig. 4.26) primarily because there was an additional 30 seconds of full power operation after the loss of feedwater and additionally in part because the reactor coolant pumps were still forcing flow through the RCS so that more heat was transferred from the core to the secondary side. Thirty seconds of full power operation generates enough heat to completely dry out the mass inventory of one steam generator; the balance of energy which caused early dryout came from decay heat and pump power. Thus in this case the steam generators dried out less than ten minutes into the transient, in sharp contrast to the situation with LOOP (Case 1) where the heat removed was due solely to tecay heat.
Compare Fig. 4.26 to Fig. 4.3.
The loss of primary inventory out of the safeties began much earlier in this case (Fig. 4.27) and the primary inventory began to deplete correspondingly (Fig. 4.28). It is useful here to compare the RCS inventory history in this case to that of Case 1 (Fig. 4.8 and Fig. 4.28). We note that u.
WR y
l O the slope of inventory decay is nearly identical but that in this case it occurs more than thirty minutes earlier due in large part to failure to trip l
the RCPs.
As we saw before, once the steam generators dried out, the primary inventory began to thermally expand and the pressurizer filled (Fig.
4.29). Because the loss of mass in the steam generators was so rapid in this case, however, the numerical " humps" in the primary pressure (which were also visible as a secondary effect in the pressurizer inventory level) which were so marked in Cases 1 and 3 were not particularly noticeable here.
As the hot leg vapor fraction reached 1001, in the junction to the surge line (Fig. 4.30), the surge line began to draw vapor (Fig.
4.31). (We note that the junction is on the bottom of the pipe and the RELAPS flow regime map forces the surge line to draw only liquid from the stratified flow until no liquid remains.) Once the surge line begins to draw vapor, the i
net inventory in the pressurizer drops rapidly because low quality mixture was still flowing out of the safeties (Fig. 4.32 and 4.29).
O Since the primary inventory was rapidly depleting (Fig.
4.28), the void generation in the hot leg (Fig. 4.30) causing voids in the surge line (Fig. 4.31) was nearly co-incident with onset of voids in the core (Fig. 4.33) and a decrease of collapsed water level in the core which occurred at about 2100 seconds after the loss of feedwater (Fig. 4.34). We observe, as before, that rate of the loss of inventory from the primary decreased as the pressurizer safety flow returned to high quality flow at around 2500 seconds in this transient (see Figs. 4.27 and 4.28), pennitting the core water level l
to temporarily level off (Fig. 4.34) (although inventory is still being lost
- l. from the system). Since this transient was proceeding much faster than Case
- 1, the inventory in the core leveled off for a somewhat shorter period. (This is a compound result of the inventory loss rate through the safeties, somcwhat higher decay heat, and lower primary inventory at the onset of uncovery.)
(Compare Figs. 4.34 to 4.10). As before, this results eventually in core uncovery (Fig. 4.34) and coolant and cladding heatup (Fig. 4.35 and 4.36).
O b
e
$7L
- - , --y
l In eur:mry, w obscrue the mrked negative impact of faizure to tri? the RCPe at loss of feeda:ter, and that it resulte in onset of core uncovery mee than 40 minutes earlier!
I l
l l
l l
1 o
9 0
u
~
.w T.N1
i e ,
k PC FLCMS: HOT LEG. CmLD LEG o.
I q a.e t.8 w
g t.s a
~
s.e U
< e.a rr g i..
i e..
m y e.s x
e.4 e.a L_ __
e.o
? .
t
-e.a
- tese asse seso esse esse sees see tese asse sees asas sees sees Tlee Isl CASE 2 .
Fig. 4.22 l
l 4
l 1 . . - - .- - - .. .. -. . .__.___-_\
SG C3ME TEMPERATURES: SGl. SG2 s
see
.l.
J y -
E f a~
e j
/
3 see -
J us L .x A M v l
see 1"
see
- 1*eo sese 3ese 4eee seus sees see sees asse 3ese mee spee sees Tise ist CASE 2 .
Fig. 4.23 v SG DOME PRESSURES: SGl. SG2 13se SM 4300 1879 Q 18se 3 '"
L .I S ". ... . .
y 128 180s a im u
A 1854 lies lleo le?W g.
logo 9 lose Sees 3ese 4000 9000 ser3 9ee 19eo Sese 3ese 4500 Wee Wee x Ties is)
(y CASE 2 o
Fig. 4.24
- " " ^ * - -m.- . - - . . . _ _ _ , _
4 g RCS PRESSURE PZR TlT1.I , l- .-
I f
ano awe -
$ M E-
==e N !
a m.
O i
- m.
Stee
- t/
v sees e seee sees sees eseo sese esse see tese sees zee ese sees esse Time ist CASE 2 Fig. 4.25 -
WATER LEVEL IN SGI e
38 38 S
E te d if i
l se
\
\
e e test asse 380s dose seBe M ese tese Wee 3ees .ese esse esse
[]
l Tlee ist CASE 2 'n Fig. 4.26
)
PZR SAFETY YALVE FLOW G
LJ -
rh e .
i
. i]- 'l h
s il ti al = - gi 4 l
~ I l l
= -
i' ,
T i i--
5 4 l d i i j 1 ' -
o, ,,,
$ 'l l _1 I
- !' l il
= -l ,J! .
L i
- -4 1.y ,
m ,
i a g-'
l _
asse goes sees O tese stee 3ees asse 3ese egee $50s Wee See lose Time ist CASE 2
~ Fig. 4.27 RCS HASS INVENTORY ese
- 4.0 N
4..
g ...
E W s..
5
\
\
\
\
l N 1..
O tese sees 3ese esse goes sees See lose See 340 goet 3 00 See SE ,
j Fig. 4.28 m
_9-@eewh--- p,.
-.y W
e o MASS INVENTORY IN PZR s.e 1
.49 I \,
~.
see h
. m
/ i y
f \
U
=
J 3
Its y let I
. i m. 9 900 1989 8900 3500 asse 9588 8388 Tlee ist CASE 2 Fig. 4.29 VSID FRACTION IN HOT LEG s.e p
8.0 0.7 E
'i **
j Q
g ...
j s
0.8 6.4
- SEED SERG Suee 4000 Some Gene y
A WB 939 M Ws0 4EES 3 09 W b SUsE 2 d '
Fig. 4.30 0
~~
- e Veld FRACTION IN SURGE LINE s.4 l
... i 1
[A E.
N l g .
O I
.3 %
e..
e ,
..e A_ l
. sees asse same eso same esse suo sees esso see ese some asse Tle, ts)
CASE 2 Fig. 4.31 veld FRACTION IN PZR SAFETY FLOW s.e 7, q!;
e.e ,
'B
... H -
f IT z
2 y
l I
1 J
'l
< ,.i g ..s ,
a y e.. ---
.3 a
P s
. sono asse suse ese w es ses seen asse see ese asse asse Tla, tel s
CASE 2 Fig. 4.32
I e =
a l l
l VSID FRACTION IN C3RE:T3P 3 NSDES s ..e
..e e.e e.7 E
- ..e 4
g e.e o
Q
- e.4 e.s ^
en .
den
- e. '
e.e . ... . . . .Y e sees asse sees esse sees sese see sees asse sese ases ssee esee flee tsi CASE 2 Fig. 4.33 O
v C3LLAPSED WATER LEVEL IN CORE 88
' IS '
'. s s
'.a. to k\
.y. -
.a
- I N- e l
I
. \,
.. e e meno esse asas asas esse meno 1
ese isee sees sees eso esse esse ==
Time ts! , L CASE 2 a e
'v' Fig. 4.34
_,y -- ~ _...%---, --
1 l
I C9RE C99LANT TEMPERATURES: BTH NID. TOP s an.
m i
m m
i C
- I e
- l 9
w f m
E 3 m )
3 g7 , ., ,,
b
- //'/ .
a -///
= LV i 1 P1
= -
. son. m. me m.
. im m. .e m.
Tlee ts)
CASE 2
(~N Fig. 4.35 CLAD TEMPERATURE: TOP N00E N.
~
1 c =
g =
. e ,
w 5 *
- I .
5~* /-
3
- m
/
e Y
_ i /
.e.
. i.e. .es. 3re ass. .se. .88.
_ m.
gJ '::!;'
- m. m. m.
DRMI Fig. 4.36 t' L
4.2.2 Case 4 This case was identical to Case 2 for the first 10 minutes of the transient, at which point the RCPs were tripped and operator actions were initiated. However in this case, as in Case 3, the operator actions were assumed to include APS and HPSI actuation according to operator guidelines.
We recall from Case 2 that by about 510 seconds into the transient both steam generators were dry and had totally lost primary-to-secondary heat transfer capabili ty. In this case, as with case 2, within 100 seconds after dryout the RCS pressure was on a sharp rise (Fig. 4.37). Thus, when the APS flow was initiated (Fig. 4.38) because of high hot leg subcooling margins (Fig. 4.39),
the APS flow had little impact in lowering the RCS pressure because, in con-trast to Case 3, there was N0 heat sink available here. The APS was even-tually turned off when the pressurizer inventory level reached the high level setpoint (90% level) roughly 1015 seconds into the transient (Fig. 4.40).
Nevertheless, the APS did slow down the primary pressure increase to the pressurizer safety valve setpoint by about three minutes. As in all cases uiscussed thus far, the primary pressure leveled off at the pressurizer safety Q setpoint, and the RCS inventory was rapidly discharged out of these safeties (Fig. 4.41). HPSI produced no flow because the primary pressure was too high, and core uncovery began about 3240 seconds after the transient initiation (Figs. 4.42 and 4.43). As before, the code computed cladding and core coolant temperatures rose sharply (Figs. 4.44 or 4.45) when the void fraction in the upper node finally reached 1.0.
Thus va observe that, as in the imediataly preceding case, the impact of continued opension of the reactor and RCPs dominated the tmn-sient. Although opentor actuation of APS and BPSI t.ns assumed here, the single impact from roughly ten minutes of APS flou (initiated immediately titen the opentor first acted - at the same time of RCP trip) ate a mughly ten '
ninute delay in actual core uncovery.
Only one charging pump t,ns assumed available for APS in this study. Since APS togic vitt turn off APS titen there is less than 200F sub-cooling, the use of att three charging pumps vould alter only the timing not
[
\
the substance of this result.
L
i 1 e s l RCS PRESSURE-PZ3
/%
... dea . . L, __
- - 1% ,,) f 1
I Ene -
two I N E
em ,J ju .
R '"
E esse atee
/
"" t/
y esse e lese sees sees esse Seco some See 1948 See See does Wes See Time ts)
CASE 4 Fig. 4.37
- t AUX PZR SPRAY FL3W s.s e.e ,
s.s
/
se 4.$
- e. .
/
y , .
< 3.8 l 5 g ...
J /
ta. e.s E s'e
,/
t.s
/
e.e ,,
..e
/
e tese esse sees esse sees esse See 1989 Wee 2 00 4300 Sese See Tise ts)
CASE 4 i Fig. 4.38 "I k
lier LEG SUBC22 LING r
"E f
1
. \
. 3 s
~ ,e 2 \
at
\
s.
\ i
-se o, toes sees sees gee sees a goe 19e0 ales ,see 49eo N 88 "
Tlee ist CASE 4 Fig. 4.39 O
PRESSURIZER INVEN!8RY ses
, s
=
I \,
E I
=
g l 0 see h %
= e E
3 ase I \
]
i (l
t
~
see v
\K too e sees osso Jene ** " "
ese isse are 3sso *** * "
Time tal s
CASE 4 Fig. 4.40 w I
RCS MASS (fr.'Ft1TcRy e..
s.s A
s.e N
- y 4.s
- \
E E ..s W
5
\ <
. ~
.. e j t.s
... \,
s.s N
. tese m fees sees ee m t** ses asse noe a m .
Tlae ts)
CASE 4 Fig. 4.41 ValD FRACTION IN C2RE: TOP 3 N3CES
..e
.8
~
W ...
E .
j e ...
..e .
1
.., i
,Y 8 lose 8000 3ese .seg gee, e g,,,
U 1988 Else See .ees gges ggg a flee ts)
CASE 4 s Fig. 4.42
COLLAPSED WATER LEVEL IN C9RE
" \
Hm
)
i '
1
\
y . \ .
=
' N
\
\
(s t
. i a sm = ** =
go. 3930 838. 3588 = M Tlee tel CASE 4 Fig. 4.43
- CLAD TEMPERATURE: TOP Ne0E r
185. /
188.
C b **
(
a i. ,
E -
E gi W
J ese 1 -
m
. a see. =. ** *
( p go. 3940 33 3588 Time ts:
8888 N 8E'8
\
CASE 4
) \
l Fig. 4.44 l
- e i
CORE CO3LANT TEMPERATURES: BTH. MID. TOP J -
i c ,_
/
o 8 ***
,iase Q
seee l l '
/ ,
/
J '
l
+ -
noe e less nees 3ese ages sees gene too tese sees sees asse sees asse Time tel CASE 4 Fig. 4.45
4 4.3 Feed and Bleed Operation l l
In this section we present results from four cases where " core flushing" (feed and bleed) was utilized to attempt to cool the core by opening a PORV and aligning the SIS for cold leg injection. " Core flushing" takes place when the cold fluid enters from the cold legs passes through the core and out the PORV.
. Assumptions in this transient are:
- 1) Two PORVs were simulated in each analysis and
- 2) Two different sizes were examined with the following combined (total) flowrates a) 119.7 lbm/s (113.4 lbm/hr/MWt) (nominal) b) 268.1 lbm/s (254.0 lbm/hr MWt) (giant)
, 3) The PORVs were assumed to be manually openeo after 20 minutes into the transient and to remain open.
Other than to open the PORVs and to shut off the RCPs in those cases based on Case 2, no further operator actions were assumed required.
The PORVs were sized in the code to deliver the rated VAPOR mass flowrate at 2400 psia. However it is important for the reader to recognize l that the two-phase flowrate is substantially greater and is not well known.
Thus these (and any other) analyses should be taken to be only representative of the general behavior to be expected and the fine details of the analysis, while interesting, have substantial uncertainties.
n w -
,.---w-.-,ew. ,m-.
I 4.3.1 Case 3ai i
This case and the next case (3bi) are both identical to Case l
1 up to 20 minutes into the transient when the PORVs were manually opened. In
' ths case we assumed the small PORVs while the next uses large PORVs. A signi-ficant. amount of energy was removed through the PORVs when the flow out of these valves was a single-phase vapor; however as the flow leaving the PORVs became two-phased, the energy removal slowed down, hence the primary pressure decrease also slowed down (Fig. 4.46, 4.47, and 4.48) and briefly began to repressurize when the quality in the PORY flow reached 0.0 (single-phase liquid). At that point the pressurizer was solid. This slight upturn in the primary pressure was aggravated by the steam generator relief valve closure until the secondary side pressure butit up enough to open relief valves again.
The HPSI was actuated at about 1255 seconds into the tran-sient. when the primary pressure fell below 1600 psia shortly after the PORVs were opened (Fig. 4.49). As the RCS depressurized, HPSI flowrate into the
("' cold leg increased and because the injected fluid was at 120*F, the HPSI flow
- ( also contributed to further reduction of the RCS pressure. This depressuriza-tion in turn pennitted higher HPSI flow, so that the symbiotic effects re-I sulted in a relatively monotonic increase of HPSI flowrate. -
4 The combination of opening the PORVs, which resulted in loss of RCS inventory, and the HPSI injection of cold fluid, which lowered the RCS average temperature and therefore led to contraction of RCS fluid, eventually
! caused voiding in the RC system (Figs. 4.50, 4.51, and 4.52). Void formation i was evident in the core as early as 1500 seconds into the transient. From that time onward, the primary inventory decreased monotonically and the void l
in the surge line finally reached 1.0 at slightly after 3000 seconds (Fig.
l 4.53). As we saw in case 2, a drop in the pressurizer inventory accompanied voiding in RCS and finally reached the pressurizer through the surge line
! (Fig. 4.54). As discussed earlier, the delay is due to the fact that the surge line is connected to the bottom of the hot leg pipe, and until 3100 -
seconds into the transient the flow in the horizontal hot leg was stratified so that the insurge into the pressurizer was extracting only the liquid por-tion of the flow from the hot leg until the flow in the hot leg became pure steam. A
- . ,,.. .. , . , ,, -. ,. ,--. -..v--,,,, , - . ~ - - - . - - - - - - , - , , - - , - - , . - - ,
The temporary drop in the void fraction in the core at roughly 4000 seconds after transient initiation (Fig. 4.52) was due to a clearance of one of the loop-seals which forced cold fluid to flow into the core (Fig. 4.55).
During the period for which the PORV flow was two-phase mixture the RCS pressure had leveled off, but when the flow momentarily became pure vapor at - 4200 seconds (Fig. 4.48) the pressure dropped again and the HPSI flow increased. The system pressure continued to decrease, eventually reaching the accumulator injection setpoint (640 psia) about 5325 seconds into the transient. By 5000 seconds, the systen had depressurized enough for the HPSI fToo to turn over the RCS inventory. With the accumulator fTou, the net flou into RCS aza positive !?ig. 4.56). Core uncovery aza not computed to occur in this case (Fige. 4.52, 4.57 and 4.58).
- f e
)
W m% O43f w
- e I
RCS PRESSURE PZR
]
asw y, % .,,,,,
esse lose 2
m sees .
us i.ee
( s
\ \ A _
~
- ..e e 1 . .
sees see m
(
.ao ,
see e teme esse sees esse Sue een ese sees sees sees .see seco stee Tlee tal CASE 3Al
\ Fig. 4.46 PSRV FL9W eve A
/\
b 388
. ~/ L -
l e in --1 e ,se 5 Y $ .
5" \I l
h\'\ -
lJ{n N 4 N
e e Rose sees 3see goes case esse see tSee 5 00 Wee seet Neo See , y Time tot .r '\ 'v . 9 CASE 3At 'p
, f ,h-k
-- Fig. 4.47
e .
l l
V8tD FRACTION IN PERY Fl3w
.j e.e . ,
.., 1- Al -d fu
= j l
u
} ..s Q
- ~3 k ^
... L I
e.e
- sm asse 3 ,,,, ,,
'W ises esse ,,, .
Tise tal CA0E 34I
( Fig. 4.48 HPSI FLOW se e SG -
- . [
e f S a y '~ ,.sr E ,
/ Tr/ -
a - / (
e 3 ~
to . __
e e sees esse w . .m- ,,
8888 8 88 3000 ' esse ggme ggge f l
\ Tlae (s) . i CASE 3At .' ,
F1g. 4.49 l
l
e e
, les P. INvfNTWNY S. 4 6.se s.at
- 5. 4.?S
\
g ...
e..
3
! t..
a ... s 3.54
... \
..ee -\
gg D A ._
... l e gene sees 3see eene sees Gene See goes ses see eget gas see flee Isl CASE 3Al Fig. 4.50 n -
U r WW9' F WWI Af t f fl MI'1 f f A f f MI 9 2 fifM. M l l' . 1 Cf' 6es 4 l c /
,, % __1s A.-
, 5[ E -
k hgM :d E O (
. ,e O k.,,
/ b 1
est en e ice. .88. . .e. .e. ==
See tSee met See est Het See flee tal h CASE 3Al
{ ) ,
J
! Fig. 4.51
I e .
l l
/
U' vain rwArT19N I N ( WitF 3 T WT' 'l NWDI '1 e.as I
- e. -
j h p' '1 u
L r, li rj.,
. U I'o a. i aN i
... N"jS, i v '
y ,
l
- ... v l Ild
, ,qc
... ( ,
I
.. e. ,
f
.e ,= ..
. sees sees see. .ees sees sees ese asse asse aseo esse sees ssee Time ist CASE 3Al
- Fig. 4.52 o VOID FRACT!9N IN SURCE FLON g,,,,y p , .i I Y pr I .
l j ,
hg 1, , , ,
4 g ... - --
o *
.3
..a h
I aL ~
. .e. .e. s.o. -
'JS. 1840 Wee 3550 esse 5 09 WBO Time (si
) CASE 3AI ss h v ^^
Fig. 4.53
e .
l i
1
,O MASS INVENTeRY IN PZR l
V "
P _l l I 1
.n I \/ \l I I 7 e l
.e.
- ee .
_-_ i
> Ii .
' i g ses l l NN y i t,[fk a a_
i ry Ib
s'
< I z ,
l.l s g
vs ,
ese
- b, e sees sees somo **se 58 " 8
tes Igg some Ese 6 M M Tlee ts:
CASE gal Fig. 4.54 N
Veld FRACTION IN LeeP-SEALISG21 s .*
7 g e.e l} ,
y e.e sg I
g e.e
. e..
S g . .. e e.2 e.a l
e..
e sees asse asse .ees sees eene
[ ees tese 3 00 M does Wee esse Time ts) i l CASE GAI Fig. 4.55 i
l
- ACCM - PORY - SAFETY tee ese see S "*
~
see U
$ ils e I e see >
3 y *
/
(
w
. . t la\ a. ,3
-tee (A,
s-% f-,
, \/
-seo e sees sees sees esse sees sese Soo 19ee Mae 3 Bee ages 5 00 Wee Tige tal CASE 3Al Fig. 4.56 v
COLLAPSED WATER LEVEL IN CARE
' 52.*
' sa.s .
.e ts.s Yl ,
)
e io..
g N x 4
d g,,, . et!
'\)i,'ya19
" 1
- e. :
frl f l ,'
- I e.c s.s e.e 7.s e sees asse sees sees sees esse see sees asse seee eseo esse sees Tlee (s) y CASE 3Al Fig. 4.57 1
-Mw. %.en emm=p= g% m- a aw 4..r
'N
e 4 j
., j, CLAD IEMPE'.tATURE: TOP NGDE A
" N
% i is.
o " l
] I I
W
}
" \ hw w v No
$ D G = s Nm s
_ N
. 1-M 100. me. 3e3 agog ggge ggg Tlee ts)
CASE 3Al
( Fig. 4.58 i
fu 9
> 99'
_ _ , ----y v v -------r e, -N -
a 7 4.3.2 Case 3bi l
This case is identical to case 3ai, except that we used the j larger PORV size through which more energy can be removed. The RCS pressure l 1
dropped to approximately 1200 psia compared to 1350 psia with smaller PORVs (Figs. 4.59, 4.60, and 4.61). The HPSI was actuated immediately following the opening of the PORVs (Fig. 4.62), and the flow continued to increase as long as the RCS pressure continued to drop (as expected). In that case the system depressurized and cooled down so rapidly that the steam generators did not dryout (Fig. 4.63). More liquid was discharged from the PORVs permitting l
generation of void in the RCS system to take place earlier (Fig. 4.64). The dip in the void fractions in the core at 1900 seconds into the transient was, as in the prior case, due to loop-seal clearance (Fig. 4.65). The void frac-tions in the core increased when the accumulator injection was on because the cold accumulator flow condensed the vapor and cooled the core inlet tempera-ture causing mixture collapse (compare Figs. 4.64, 4.66, 4.67 and 4.68).
The RCS inventory apneared to bottom out by 3500 seconds after the onset of transient, and although it was fluctuating, tne trend seems
(
to be up (Fig. 4.69). The collapsed water level in the core also stabilized, with occasional dips also due to accumulator injection. No core uncovery was computed to occur in this case.
Although none of the other cases analyzed encountered mass error problems with RELAP5, this case originally had significant problems.
With assistance from the RELAP5 code developers who prepared code modifica-tions to cure these errors, we were able to complete the study without any further code problems. Hence, this particular transient was run with these code updates, while the remainder were not. For completeness, selected para-meters were compared for this transient using the original code and the up-dated version. These results are pesented briefly in Appendix 1. We point out here only that the differences in major plant transient parameters are negligible.
v/
- 9 RCS PRESSURE-PZR
\
aese
- c~
ones snee 5 sees E
t 8400 5 taes
\^ s 0
'~
3
_ \
\ -
.e.
es one e 1988 asse 3ese eene sees sees ese tsas asse stee asse sees sees
. Time tel CASE 38I Fig. 4.59 PORY FLOW see a.e I ese G m ' * -
S
)
e =
m N
- ce g see a -
m )'{Ih iA E l f lf 5 toe see $l
'R l l
se e
e i- .see -e .see sees g)
(
see asse ages asse asse sees sete esee y ,/ Tlee tel CASE 391 Fig. 4.60
9 6 l l
VC!D FRACTIC?! IN P3RV Flaw s.e ,,,
f l L
, I \
4 1 I l g ...
i o
l u, f
, I l
1
\ l
\
$ I p
8ustEI S.4 13.39.e8. eM443 e same emme :smo .emo esso amme see use sees see w some ammo Tlee ts)
CASE 181 Fig. 4.61
\
/
a /
a - ,
a /
- 0
- m
^
/
= /
. = i d
E **
x is s
e e some sees seee esse seee seee 948 ISee 5 88 Ese 4000 0000 Ese p Tlas ist
& CASE 381 O ,
Fig. 4.62 I
. . . . ~_ .
0 0 STEAM GENERATCR WATER LEVEL. IN SGI
\
) ..
"' I I ! l I l ;
- m. , I I l
..e .
. n..
w ..
., \
y...
$ ... \ I
.... \
\
18.8 \
se.e \ ---
2 v.s I e ie.e ee - e e ee
.00 19 Mee ages See e.es ' mes flee tel CASE 388 Fig. 4.63 i
VGID FRACTION IN CORE: TOP 3 NGOES e.,
e.e j 4.5 5 k c e.. I '- %
N '
hl f.
- e., -
AML
\
L- ~
e 1 1
a -
I e.e . ...
p }I O tese 3000 Wee 4000 Sage esse see 1940 asse Bee 4see Sees See Tlee ist O' CASE 381 rig. 4.64 e w w -- ,-y .-.,_
C V8ID FRACTION IN LesP SEAL IL93P li
\
3.e - -
l-
..e ,
e.e
- ..e W
g ..e Q
~
0.4 e.3 e.a e.s e.e e 8000 ages 3ese dese Sees sees see sees asse 3see gese sees esse Tlee tel CASE 301
/O Fig. 4.65 V
ACCLH. (NJ. FLOW s
We e.e l l l l l
,,, e li l I A !
e = '!
l g ,
I l
- ll
= p d ! .lu- l e ; g' , _ _.1 E l l l x ies l ! ll n !f-t e - e =s e i e e 1
a.e same am.s ze .m.e e es.O flee is)
\
G CASE 301 -
\
t Fig. 4.66 1
e e
c c*tt C*0'. A!J
- IEMP'.R A rt*E2 ' 9M. "10. W U -
, J i i i i me '
N i
I i
_ l I
_~ 2 1- . . , \. \_
I E~ ,,, ___,
('[i ,
I y , . - - .
< l 1 s I -
g ,,, . _ _ _
_ .T .
l l
,I ,,, _ _. ._ .._ . {
% 7- ,
l
/'
es f' ses e gene seg 3,3 esse M M m em Emes SIS *EE M Tlas tel CASE 381 Fig. 4.67 D C3LLAPSED WATER LEVEL IN CORE 3 -
na
\
\
18
.. J S
S .
b r
I 4
, li "i -
4 6
4 4 1888 8ese 3000 asse 5000 sees
.00 1900 ages M esse g500 eseg Tlee ist S
CASE 381 s
Fig. 4.68
)
1 1
e a WATER' MA$s' IN 'RCS ese '
e.e 0.0 3
0.e a
m 4,3
\
4..
Z
$ 3.8
\ '
5
\
g 3.. -
=
8.S
\ ,
A M. ,
8.8 4 test sees 2004 4000 eses Gese ces test 5 00 2 00 deet Wee eget
. Time ist CASE 381 Fig. 4.69 i
5 4.3.3 Case 4ai This case and the following case (4bi) are identical to Case 2 (ten minutes of RCP operation) up to 1200 seconds (20 minutes into the transient) when the PORVs were opened. As with the previous two cases, in 4ai we study the effect of the small PORVs while in 4bi we study the effect of the lare PORVs. . We note that by the time the PORVs were opened the RCPs had been tripped (at 10 minutes after transient initiation) and both steam generators (this plant is nearly symmetric) were dried out (Fig. 4.26). Figure 4.70 shows that the RCS pressure was already at the pressurize safety valve set-point and the RCS inventory was being' discharged (Fig. 4.71). The pressurizer was nearly solid with water at this point (Fig. 4.72). When the PORVs were opened, the void fraction in the PORY flow indicated that the flow was single-phase liquid, which degraded the energy removal capability and the RCS pres-sure began to rise to the safety setpoint (Figs. 4.73 and 4.74). RCS pressure was, nevertheless, momentarily reduced by 450 psia due to opening the PORVs.
Q Void formation began as soon as the PORVs were opened and D' quickly propagated through the RCS up to the surge line and into the pressur-izer. The inventory in the pressurizer dropped as the flow into the pressur-izer became two-phased and finally pure steam (Figs. 4.75, 4.76 and 4.70).
When the discharge out of the PORVs finally became two phased. and therefore began to remove energy at a greater rate, the RCS pressure began falling (about 2100 seconds into the transient). However, by 3050 seconds into the transient, the system had lost enough inventory that core uncovery was indi-cated by the core coolant temperatures, the liquid level in the core, and the cladding temperature (Figs. 4.77, 4.78 and 4.79).
This case indicates that the amLL PORVs vere unable to depressurizar the plant enough to enable HPSI deliverj to prevent core uncov-ery. As for Cases 2 and 4, 30 seconds of full pocer operation em carbated the transient.
f
- - . ~ - . -
m _ . . . y -m-y . . - - , - . . -
1 I
\ RCS PRESSURE-PZR nse sw [ \
g am
~ I, ) \
m j s
\
e I \
- ~
e
\ l \
- f \ -
E I 0- esse atee ,
/ I i
- >I I v
4
- l e some esse sees - sees sees see t.se .see ssee me esse asao Tlee t.)
CASE 4AI O 1 Fig. 4.70 i,. RCS HASS INVENT 9RY
- Of .se '
-L ' ...
g
..e \
g ...
\ 1 e
, e. .
\
k.
3 s s .
\ .
g ... (
8.. \
8..
g, V
.es es.e
,see Tlee f.)
_e
-e es.o
- . ,) .
CASE 4Al Fig. 4.71 i nm---
^nn es t, ~3 m m
g MASS INVENTGRY IN PZR J -
e .c _
. *ee l
2 i
a
/
x L~
- 3.e s
ese 190 y 18e e sees sees sees .see sees see, ese 195e Wee 3see sees gese gee flee ist CASE 4AI Fig. 4.72 O .
G P9RV FLOW N'
.g.
Wo -
! k n'~ \
$; - t us tM E
ce tse -
19 %
3 too -
n so W .
e teso asse sese asse ease seen
^
See toes See Woo 4ese Eso Wee g flee ist .
D' CASE 4AI Fig. 4.73
. . - . _ ~ __ _ ,. . _ . .
l I
VOID FRACTION IN P3RV FLQM
)
a i.e j e.. g e.s
[
. .?
E f M
g ...
g ...
..a
. see. see. see. esse .ees eso.
. es. see. see. sees ass. . sos eso.
Tim. Is)
CASE 4AI Fig. 4.74 VOID FRACTION IN CORE: TOP 3 N00E c
. i.. ,
5 M
g :
O * .
g ...
l
..a D'
( 9' l . .co. .se. ,es. e. se esse
! h) v
- == == ss=
ri.. i.,
son .es == e CASE 4Al I
i Fig. 4.75 i
l
YSID FRACT!GN IN PZR SURGE LINE
..e c ,
t e.e ~
e.s q y g
e.,
ifn U l
g s.a 4
g e.e o
g e.4 e.3 i
e.a
' ~
e.,
l L !
e tese asse sene esse sees asse ese tsee asse sees esse esse ese flee (si *
. . CASE 4A!.
Fig. 4.76
>V CsRE COOLANT TEMPERATURES: BTM. MID. TOP ese f
ese b
8 =e i m E
w = ,
I j.
.e /PW
.J
.Se Ef J
ese o tese sees 3ese esse esse sees See sees See 3ese ees sees Wee '
s flee tel CASE 4AI Fig. 4.77
1 CSLLAPSED NATER LEVEL IN C3RE I
,I
$3 is i
)
n.l I
Si p b N 1
y c* e :
- l l 1 d
C a ?
\
e
\ .
4
\
3 e tese esso 3ese 4sse sees sees see sees esse zoo ese sees sees Time ts)
CASE 4AI Fig. 4.78
\
CLAD TEMPERATURE AT HOT SPST use less teos o
}
8 es.
a -
4 " 1 5
$ = l _
~
,ee l
~ - - ~ J .
~
u /
r see o tese esse asse esas name esse l ses less Ese gens l
Mme mes see Tlee est I CASS 4Al
-_/
Fig. 4.79
(
4.3.4 Case 4bi This case differs from 4ai only in that here the larger PORV size was utilized. All other conditions of the plant at the time of PORY opening were identical to Case 4ai. The flowrate through tnese larger PORVs was so great that even though the initial flow was nearly all liquid, it was able to pull the voids generated in the core immediately after the PORY open-ing through the surge line into the pressurizer. By 2000 seconds into the transient the PORY flow was pure steam (occurring about 500 seconds sooner 4
tha.1 Case 4ai with smaller PORVs) and the RCS pressure fell rapidly.
In this case, the RCS pressure fell below 1600 psia and HPSI
, was actuated at 2150 seconds into the transient (Fig. 4.80 and 4.81). repres-surization of the RCS continued and at 3120 seconde, the pressure beca.m tou enough (belou 640 paia) for accumulator injection (Fig. 4.82). The RCS mas
! inventory (Fig. 4.83), hooever, turned around before the accumulator actuation because of mpid primary pressure reduction allouing HPSI flou to reach a near 1 p marinwn flourate 300 seconds sooner than for Case 3bi (compare Pigs. 4.81 to N 4.62).
Figures 4.84 shows that the top node in the core nearly l voided at about 2300 seconds into the transient but core uncovery was not l indicated by the core coolant temperature or the water level in the core l (Figs. 4.85 and 4.86).
l m}
<<w -
o--. .m - o-ee vomon. ,
m
1 *. .
O asse RCS PRESSURE-PZR
..se
- /
v '
f X s
2 some y E
- \
itse R $~
3 late lese
_ \
~ \
.e t~
e lese 80se 3000 eese sees 8000 see lees Sete Bee goes seet . 00 Tlee tel CASE 49!
~
Fig. 4.80 HPSI FLSW e*
w M
/
- /
a B - f
.- . /
= n e
e ;
a -
E 15
- 19 8
O e Rose sees 3ees sees sees sees y
) ,00 e _e Time tal
.mo _e .se s
CASE 481 Fig. 4.81
e .
ACCUH. INJ. FLOW O =
ise see S m a
W m a
$ ile .
d ies 1.
$ l } (
, j a_
e a9e e seet seee seee aos esse esse See 1950 asse 3ese 4ee Esse sees flag Isl CA':E 4BI Fig. 4.82 O
, ur ei wA*: r INVf NIMNY s.e s.s s..
\
, , . I L
g e.s g ...
=
3.s
\
S 3.e s.s
\
me l t.s
\ >
e sees sees sese seen sees esse een sees sees 3sse sees see.
Time tel CASE 481 g Fig. 4.83
.- - - . . - . - - ~ ,.. -
e e e
VOID FRACTION IN CORE: TOP 3 NSDES J ..e
..e 6
e.e e.?
I
- ,s 4
'5
- .a
$ e.. '
d : 'I e.3 w J d a
e.t e.o . ...
e seee anse sees aos esse esse see m sees snee ese sees" . sees Time ts:
CASE 481 p Fig. 4.84
\v CSKE CCELANT TEMPERATURES: BTH. MID. TOP eso
.J i
.se LV_ \
i y ,.
E
$ see
\
= -
2 \
ese eo
(: ;
see e sees asse sees asse sees esse ese tese snee snee esse sene eene Tine tel 1
CASE 481 Fig. 4.85 l
l C3LLAPSED WATER LEVEL IN C3RE
' l L
1 is..
is..
is..
u..
\ \.
n=
y; .
i.. .
a y ... k. i -
w
' e..
i l i -
%.J. 4t
}
v..
l
. i 2=. = ==
.= i.= == == a ==
Tlae (si CASE 481 N
) Fig. 4.86
/
-me ee se -m eaet.,e - - - - - - - h-
_ mee #. m ,en .._.9 m y... , , , .
--._m_.....__..-__._m-
4.4 Sensitivity to PORY Opening Time and AFW Restoration Time In this section, sensitivities to PORV opening time and to AFW res-toration time were investigated. In all cases analyzed here, we attempted to determine the approximate time of core uncovery by presenting, whenever pos-sible, results from two RELAPS calculations. The only assumption changed from those discussed earlier is the time of PORY opening. As before, once opened the PORVs stayed open for the duration of the transient. Since plant re-sponses to use of PORVs were similar to the earlier cases, only those para-meters affected by the timing of PORY opening are presented here.
1 V
1 -
tew- % y% g--w,-w , , - - - -v----* -- a e,- e e-- -m ry-. -$- --- .- a-- 4 -_-- ----ve--- -w-
4.4.1 Cases 3aii and 3bii
(/
In Case 3aii, smaller PORVs were opened at 30 minutes after the transient initiation in one calculation (curve 1 in figure) and at 40 minutes in another (curve 2). We recall that in the original Case 1 the pressurizer water level rose rapidly between 2400 and 3200 seconds into the transient and the pressurizer became solid at 3200 seconds (Fig. 4.5). When the FORVs were opened early, the system had a greater period of time during which the PORY flow was two phased than the later opening (~ 700 seconds for early opening and 250 seconds for late opening (Figs. 4.87 and 4.88) because with later opening the primary fluid continued to swell and fill the pressur-izer (Fig. 4.89). Thus, nere energy ma removed by earlier PORY opening, depressurizing the system and permitting continuous HPSI delivery (Figs. 4.90 and 4.91). ilhen the PORVs vere opened late, the system repressurized after PORY opening (Fig. 4.90) terminating the GPSI flou for approvinutely 2000 sec-onds. During this period, there as not cold liquid coming into the RCS in this case and, the RCS inventory txs rettced to a lover levet than when the PORVs were opened early (Fig. 4.92).
By 4900 seconds into the transient, the collapsed water level in the core had dropped substantially for the late opening case (curve 2 Fig. 4.93), the core uncovered (Fig. 4.94), and the coolant temperature sud-denly rose (curve 2, Fig. 4.95) whereas curve i shows that this was not the case in Figs. 4.93 through 4.95 because of continuous HPSI flow. When the HPSI flow was actuated again in the late opening case by eventual RCS pressure reduction, the transient turned over.
Thus we conclude that with the small PORVs, the operator has at most 40 minutes after TLOFTl and LOOP initiation to open PORVs without risking core uncovery.
In Case 3 bit, we perfonned a similar parametric analysis using the larger PORV size. In this study the PORVS were opened at 3000 sec into the transient (curve 1) and at 3600 seconds into the transient (curve 2) to bracket the time when the pressurizer went solid (~ 3200 + seconds into the transient) (Fig. 4.96). Thus with 50 minute opening the initial PORY flow wa o ..
u _ _.
two phased (at least for a brief interval) while for the 60 minute case the PORY initial flow was single phase liquid (Figs. 4.97, 4.98). Because chase PORVs have auch large capaMty, the pressure dropped far enough in both cassa to per-nit RPSI flou (Figa. 4.99, 4.100). the slight repressurization in both cases began coincident with the PORY flow being single phase liquid and turned over as the flow void fraction increased. . Compare Figs. 4.98 with 4.99.)
Although there were indications that the 60 minute case was closer to core uncovery (Fig. 4.101), neither case reached cladding heatup conditions (Figs.
4.102, 4.103).
In both cases the inventory turned around about the time that accumulator injection began (Fig. 4.104).
From this study one observed that the operator vould have on the order of 60 minutes after TLOFW and LOOP to open the large 20RVs -- com-pared with roughly 40 minutes for the emiter ones. ,
s., ,
PORY FLOW 30 HlH VS 40 HIN r h y ,
a.
'\
I y i
a
.M IM 7
\
",rvy
+ I l' W
N 8# f
- k d4 5 '
,". l e %,}
e
. i i- -._
N # k, '
k s z i y
M T L
4e M
9 i .. .
O lest Seat 3ese 4089 gene Mee See 1980 mee 2000 4000 W98 5 08 Time ist . .
CASE 3All Fig. 4.87 i
_ ~ n
--~~--ey,-,we.~ -ww-n em
e e
. V8tD FRACTION IN P3RV FLE4 30 MIN VS 40 MIM a ...
i .. ..
i .f f
I I' u 4 fi
..e g qq-2
- ..e p ,'
7-y .
m ~
a %
o ' l' ;
g e.4
.I
.- 1 ,
\
,, N
~
en l
I
. .j. . ...
e sees snee me. so sees sees nas N.
flee ts) -'
- i. .
s' -
CA.CE 3AII '
'N-
- s .
p Fig. 4.88 ,. N ,
, s ,
, s'% ?% y s ,
MASS INYCiTORY IN PZR (
\ 30 Mt,9 VS 40 M!N k i een
, y a l , --
Xl _- :
[ ( d's I" s
s
. .se 2 ':+ 1, '
~
s,
> *ee ,,
~
e -s --
- i. i
% i 2 L \
ua Me ,' _.
v,-
> s .
, . _ '. 3
, ,,e _ ,
, J t-
-[ 4 see q fi l_ ArW hi '
\l 5 q '
I I %'
t, J g y%
e tems soap asse esse sees esse T
see seen enee seen sees 2 same seen
- ' Tlee tel [. 3,', .
Q t
, CASE SA!! w. s N
(Q ,-s y '~ Fig. 4.89 t, . e '4 .
o u ,
8 rL . .
Av - - , ~ . - - -
~ - _ -, . . . . - :.
1 l
4 RCS PRESSURE - PZR O' a+ee 30 MIN VS 40 MIN i-
^
w; _
__J' asse ,
l Q
- tese rvn
>M A m
?
,te _
/ N a s t
v,- i E (j /
tage Lese e
-N a tees au asse esse sees esse see tese asee' stee sees ssee sees Time ist CASE 3AII p Fig. 4.90 d
,. W S! FL9W 30 MIN VS 40 MIN se l '
e f
.o
.o -
c ) /
! - [M\ [ [ I t
5
. .e t
n swa t e ..
(
"g ts i.
S e . .. . . .
e lese goes zees game sees goes See Lese See Wee dose Wee 3 00 l
(y ri.e is, CASE 3All (
Fig. 4.91
- _m - _ - _ _ _ _ _ _ - - _ _ - _ _ _ - - -
0 .
o CN MATER MASS IN RCS I sie ' 30 MfH VS 40 MIN e.. ,
g NN
..e q
5 e \.\ 1 e.a
, \
\
5 k
\\ v-
= ...
g s
i.e x
\ - -
8.. -
8.e e less name 3sse esse W M gee 6600 .500 See 8088 M M Tlee tal CASE 3A!!
Fig. 4.92 C9LLAPSED WATER I.EVELS IN CORE 30 MIN VS 40 MIN 13 I k lI NYbA ir w % ,
p+
c \
S e d i f s .
\ /
.s t 5
\ !
l s
a e lose sees 3ese goes gese gese ese lese 5 00 Net dess 9500 mee t n .e ..I CASE Gall l
Fig. 4.93 w
l
. l l
I l
1 O
Veld FRACTION IN CORE: T9P N90E
) 30 MIN VS 40 MIN d
s.e 1 e.e e.s e.?
.e. e.e 4
g e.s O
e.4 h AI e.s 1 f k C h0 W
/d tt e.s v ,
j me . .. .
e so.e s.ee 3e *e ee ee s p.e as.e asse ca.e se e Tlas tel CASE 3AII Fig. 4.94 CORE C96LANT TENPERATtJRES: TOP NGOE 30 MIN VS 40 MIN
- i llee Itse j l.es E
. es.
/
_ e.e We a I \
- = .
f 3
=
E n m i l
/
'1% #- - -'
- =
w- v- p-w
= '.-
e.e p
g' e.e
,e ee e
.e e ee es e e_
e Tlee Isl CASE 3Att Fig. 4.95
l l
1 MASS INVENTORY IN PZR m
50 MIN VS 60 MIN s
m at
/; .
C g
i k- 5 e-8 1 V . Q -
V . ,_ 1 " y N
390 e tese asse seee me sees sees les ISM 5ee zee 4See Stes Was Tlee tel CASE 3811 Fig. 4.96 V
PGRV FLOW 50 MIN YS 60 MIN m
ese
~
m %
e, I
%t 1 3 = ,
\
= .
3
- i h-E 5 " j
\
NN r . ..
l e sees asse sese .ese see , esse Oa e - - -
CASE 38I!
l Fig. 4.97
1 V9tD FRACTION IN PSRV FL9W SO HIN VS 60 MIN f
g
... .i .a N g g ... I e
i.. -
g) a
- a l}
. . . . . . .. ... .. .l .f
. m. ==.
t a*. " " "
Tle, ts)
CASE 3911 Fig. 4.98 RCS PRESSURE PZR 50 MIN VS 60 HIN a
q M
1" l
/' ,_ _/' '
/ Y 3 ,
< \.
. \ b
, . \ \
fo'
- 1 s
- ~ \ \
8 \ \
k
\ \
\
e te a 3 t)\
4e 98 ge.
15 0 ae B M .EM M n...
CASE 3911 Fig. 4.99
___.__m .
e e HPSI FLOW 50 MIN VS 60 MIN
- I
- /
l / /
j i i
/ -
/
s_ "'
/ /
sa 3.
) j N I I oc w E -
g .
t 8.
to e
f .
see. .eae sees asse e tese asse Ehet _ 4 88 808. .388
.00 1.e4 T!ae ts:
CASE 09I1
- Fig. 4.100 v
Ve!D FRACTION IN CSRE: TOP N90E SO MIN VS 80 MIN
..e
..a
..?
g 7
g ...
E e ,8
. gr 4
I I n >
..s ,
b v
l l
p . - _ _ _
ab _ _ _
Time ist CASE 3811 Fig. 4.101 i 1
1
~ ~ ~ ~ ~ * - - - - - - . _ _
1
e e l
l l
[ COLLAPSEO WATER LEVELS Ira C3RE
(' n..
50 MIN VS 60 MIN sa.s i - j
- l. . .
. . hb f\, d g ....
-i...
(
O 4 e ...
s . K \
k L
\ %)
l l
?..
a.s
. s aus e a
. e se .s ss es seas s've.
Time ts) ,
CASE 38II O) t
'O Fig. 4.102 CSRE C89LANT TEMPERATURES: TOP N9DE 50 MIN VS 60 MIN
\
l- 1 1%
W
=
1 S
~
\ \
s 2 -
\ \
1 5 \
g i.
i \ \
W e
\ \
\
\
) . -
V ,e e e Time ts)
CASE 38II Fig. 4.103
_ m____________-_m
O WATER MASS IN RCS Bie ' 50 MIN VS 60 MIN s.e
. , s = a
..e 4.5 3 g
~
4.e 1 I . \ \
g .
n s.e l
=
3.8
\.\ /
d 8.8 e less asse 3ese doss Gene Gees See 3 Bee 5 00 Wee dess See See Time 4 1 ,
CASE 3811 Fig. 4.104 1
I 1
I
.. --% - - . . , _ _ m. . . . . . . - . . , _ - .. _
4.4.2 Cases 4ait and 4bii In these two cases we investigated PORY opening time in a manner similar to that discussed in $4.3.1 using case 2 as the base case (i.e., 10 minutes of RCPs).
Case 4ai had resulted in core uncovery (when the PORVs were opened at 20 minutes), so in this parametric the valves were opened at 10 minutes into thrs transient (Case 4ati is the small PORVs). We recall that in Case 2, by 10 minutes into the transient the steam generators dried out since the RCPs were not tripped until 10 minutes after loss of feedwater. In preVf-ous cases, we saw that the smaller sized valves were unable to depressurize enough to mitigate core uncovery under the combination of TLOFW and LOOP conditions. Here again, as in all cases, the ability of the PORVs to depres-surize was governed by the quality of the flgw through them (Figs. 4.105, 4.106).
x Opening at 10 minutes gemitted the BCS pressure to fall enough for HPSI infection, houaver the pressure quickly turned over termi-nating the flou into the cold lege (Fig. 4.107 and 4.108). The sequence of eventa vas very similar to Case dai throughout the transient (although taking place about 250 seconds earlier instead of 600 seconds due to HPSI infection) and similarly ended with core uncovery (Fige. 4.109 through 4.112).
In Case 4 bit, the larger PORVs were opened at 1800 seconds into the transient (curve 1) and at 2400 seconds (curve 2) of Fig. 4.113.
The larger PORVa, as before, vere able to take the primary pressure doun enough for continuous BPSI infection (Figs. 4.113 and 4.114).
Voids forned in the core following PORY opening were pulled through the system rapidly to force the PORY flow to turn to pure steam from the two-phased mixture (Figs. 4.115, 4.116 and 4.117). The RCS inventory leveled off and started refilling on HPSI flow (Fig. 4.118). However in the 40 minute case, complete core uncovery was encountered (Fig. 4.119) and clad heatup was com puted to occur (Fig. 4.120). Thus, un conclude that the operator has roughly -
30 minutes to open the targer PORVs to cool doun the plant and stilL avoid core uncovery.
i l
l e ,
V9tD FRACTION IN P3RV FL94 10 MIN VS 20 MIN
..e , . , . . . .
.. e e.e
.E. e..
I 1
,i M l g .. e g
O g e e.s e.e e.
e . . . . . '. ....
o some same some asse sees esse sese Time ts)
CASE 4All Fig. 4.105 v
PORY FLOW 10 MIN VS 20 MIN see
~ '
/k- /
f ) U" E \ \
~
\n \
e -y' a iN i
15 y o,
en I" w D.%
$ " h
\
M
\
I -
e tese asse 3ese dose spee Goat see seet Cj\,
Rose 3000 dose stes does ,
Time ts)
CASE 4Att Fig. 4.106
e .
^
RCS PRESSURE - PZR 10 MfN VS 20 MIN n u '
i i
- = / ; x
/
1 I N J <
- / ~ \
[f/
5 S ""
" V < /v \
td esse l ,
\
i
- >= d -
t \ -
E "" (j )
'~ \
t ises \
I \
y 13ee 1880
$ lese asse 3ese dose Gene seet est lese 3 00 3B00 esoe Wee Wee Time tel CASE 4AII Fig. 4.107 ,
U HPSI FL3M 10 MIN VS 20 MIN N
1 a . \
A I a
~
l'! "
t A l
3 i.
1 e -
[
1 "
l {
l l s l
l l e . . . . . .. ..
- x. O test asse 3ese sees 98e0 Asse "-
t I, me wee see O mee Time ts) ese een een ,
CASE 4A!!
Fig. 4.108
e e m WATER HASS IN RCS h use ' 10 MIN VS 20 MIN V .. e n +
g ...
g d ( \ .
g k (
g*
g .
e, e ...
f
- \\
x
- .s y
k e 7,000 asse 3ese esse 90se east 900 3958 Wee 3 00 does 9000 Web
. Tlee is3 CASE 4AI!
^
) Ffg. 4.109 d
C9LLAPSED WATER LEVELS IN CORE 10 MIf4 VS 20 MIN i- t{-pg l is y ti g u lit i ,a Wi
\\
\ \ \
\ c \ \
3 '
O .
E
" \\
l
\\
N
\
t
\
\
p .
. e - - - - -
o ,se ,see - -
Tlee ts)
CASE 4AII Fig. 4.110
e (N V9ID FRACTION IN CORE: TOP N90E k 10 MIN VS 20 MIN (O i.. . .
..e e.e 0.7 M
g ...
a i
- 0..
9.3 f
1 I
e sees tese 3 see.
se* Rest ase$ NeG 4 e9 Wee ese$
Time (si CASE 4All Fig. 4.111 CORE C09LANT TEMPERATURES: TOP N90E 10 MIN VS 20 MIN tese
$1e0 y tese
,/
s g
,es.
7
=
o I N 8100
- I 5 *~
/
1 I
y E ese j ,
I //
~ //
ld r 888 wC 99 ov Tlae tel CASE 4AII Fig. 4.112
1
- e I
A RCS PRESSURE - PZR a -
30 MIN VS 40 MIN iN
..e j 3 mas
. J / A \
j .
V ( \
2 i \ g
\
g some g name g . .
0 8"*
1
\.
\
_ \
eso k
e sees asse same m omme emme ese same esso asse ano seso esse Tlae (s)
CASE 48I1 p) z V
Fig. 4.113 HPSI FLOW 30 MIN VS 40 MIN
,es
. ~
- - /
a d 3 f f f .
m i i a M
S e i e s . . . . .
) e Smo WW sus 3sse e em
\ em esse use imme ame imme ame \
Time (si CASE 4811 Fig. 4.114 e wg p.-pise p * *"
_____wwmp_________~__._._m-_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ . _ _ _ _ _ _ . . . _ _ _ _ _
. e g YSID FRACTICN IN CORE: TOP NODE 9
30 MIN VS 40 MIN J n .
3 e.e e.o e.t z l '
g e.e N
E 8e i
o a .
k
" d f e.. (*ml --
e..
e.e . . ..
y e teme same asse <see sees esse We tems See Net dose Mme M Time tel CASE 4811 S
Fig. 4.115 s
PORY FLOW 30 MIN VS 40 NIN ese Me eBe E . W '
M g
4 ese
\ e a
la. y ,,--
?
s ~ (
- " p w ,
O eene uses sees asse sees eneb mee seen amo some k does sees essa Time tel CASE 4811 i
Fig. 4.116
VO!D FRACT!sN IN PORY FLOW 00 MIN VS 40 MfN l
8d . i . ... .
l M
E f Q
g 8.4 I
o.e . .. ,, ,. ,, , , ,
- *
- m m m m m Time ist CASE 49II N
Fig. 4.117 WATER MASS IN RCS
", 30 MIN VS 40 MIN e.e s.s N
m d ...
- . . \\
5 .
\1 s
\(
G
~
a.s
\
\\
s.o q
~
t.s s.e m W
o.s
~ ~ ~ - - _
y I, .
A/ = = = = . . .
Tlee is3 CASE 48II l Fig. 4.118
l l
1 m CELLAPSED WATER LEVELS IN CORE 30 MIN VS 40'HIN l
B se N
.1C\
\ \ .
c' '
\ \/
A g
\
(
i
\ j
\ l
\ /
e
\ -
e seen mese smee mee same e" see some ese same mee - -
Tlas tel .
CASE 481I m Fig. 4.119 CORE C99LANT TEMPERATURES: TOP NSDE 30 NIN VS 40 NIN esse l
j E
ts 8 *
N [
s- /
=
tems I
~'
,% ,. # 'TV '
s' g l
p) me e -
l a
V _ _ _
Time tel CASE 481I l
l Fig. 4.120
r f
4.4.3 Case li and Case 21 (Restoration of AFW)
In these cases, instead of opening PORVs, the AFW was as-sumed to be restored to regain primary-to-secondary heat transfer capabili-ties, providing cold (100*F) feedwater flow into both steam generators. This portion of the study was directed at identifying the latest possible moment for restoration of AFW which would prevent core uncovery.
As soon as the AFW began flowing into the SGs at 75 min into the transient in one case and 90 min in the other case (Fig. 4.121), heat removal began from the primary loop, the coolant temperature dropped, and the system depressurized rapidly (Figs. 4.122 and 4.23). When the RCS pressure decreased below the pressurizer safety valve setpoint, the RCS inventory loss was terminated (Fig. 4.124). As depressurization continued due to feedwater flow, the HPSI came on line (Fig. 4.125) and the RCS inventory began refil-ling.
However, as the reactor coolant temperature dropped, liqui'd contraction took place. Hence, even after the loss of primary inventory was tenninated, some risk of core uncovery still existed since HPSI flowrate could not keep up with the contraction rate, although that uncovery should be brief. In Case 11, onset of significant core voiding occurred when the AFW was restored at t = 90 min into the transient (Figs. 4.126, 4.127). Howver the voiding was not significant enough to result in core uncovery (as indi-cated by the coolant temperature continuing to fall (Fig. 4.122).
We recall that in Case 2, core uncovery began at roughly 3140 seconds into the transient. Commencing AFW flow as late as 3000 seconds would cool down the RCS enough to avoid that core uncovery in Case 21 (Figs.
4.128 through 4.133).
In .;ummary, use of AFW is more effective for rapid system depressurization than the APS or PORVs in that the operator has at least 1/2 1 -
hours before taking action to initiate AFW under the TLOFW and LOOP condi-tions. Even when power is available and RCPs are permitted to operate for 10 minutes after the feedwater is lost, there is still almost one hour after the h
I transient initiation for the Operator to actuate AFW without resulting in core uncovery.
O l
4
=
AFN FLON k 75 MIN VS 90 MIN
.. \
tee Ele les se 5
Lal a:
E d ' ee x
3e se 19 e , . . e s *
- g w gene asse sees - M ese less sees mee sees _ _
Time tel CASC 11 Fig. 4.121 CSRE CeeLANT TEMPERATURES: TSP NGOE 75 MIN VS 90 MIN l
/
i l
< \ \
~
/ \, \
c 7 t
.e
'_ / S \
o s ) 1 3- -
- \
g < (
=-
~
\ \
N\ .
N n
m Time tel CASE II Fig. 4.122
m I
RCS. PRESSURE - PZR
~
75 MIN VS SO MIN J _. ,
~
- / \
c- ,.
y ,
b_
e \.
j
.l
(
s.
E \
t
~
N \
N e some asse snee esse use esse ens seen see snee ese sese sees Time tel CASE 11 Fig. 4.123 WATER MASS IN RCS 75 MIN VS 90 MIN e.e NN
- s. .
\
!. \
,e .
\ .
/
W 5 s
- Nm /
u o,,'
m Tlee isl CASE II Fig. 4.124
HPSI FLOW 75 MIN VS 90 MIN
~ '
/
~ / [
5
- / / '
03 i l d =
su {
U 88 as g as i
d H
e i.
i.
e e . . . . . . . .
e name same same ese ==
en some mee use ame uses emme Tlee ts)
CASE ll O
Fig. 4.125 VEID FRACTISM IN C9RE: TOP Ne0E 7S MIN VS SO MIN s.e o.o e.e e.T
-. e 4
g .
O Q
9.4 e P\be
(
. lj s n
o j
j e temt 5 00 3BBo 48Be SEES 8888 ese ogge age Edo done
! Wee met Time ist CASE i1 Fig. 4.126
e j COLLAPSED WATER LEVELS IN C3RE f(g .8 l 71 MIN VS 90 MIN
= ,
1 u
r)
.j_ t I i c
/en( % ,
e /g f l 3 .
9 e
e e .e eene me. emme .sse 989 Stee met See sees Seae See Time ts)
CASE It Fig. 4.127
'V RCS PRESSURE - PZR 50 MIN VS 80 MIN auss
, 7 ;
I
,)
~
a-E s k E
a-E i 1 4 \ \
488
- (
- \ s ;
i s , ,_ _. - -
\
(
,e
) . - - - -
Time ts)
CASE 21 Fig. 4.128
=,3
HPSI FLOW
(
50 HIN VS 60 MIN (m . ,
n
/
a - /
t d I w
1 N. .
A d =
2
. i.= n
E g.
\
g ...
\ /
u Y
n == ==
/ .
\
Tlas ist CASE 21 i
Fig. 4.130 '
e 4 Vetu FRACTION IN CGRE: TSP NacE
[ 50 MlH VS 80 MIN
( ... o s
?
E
- e..
g ...
O oe4 1
- ~
e.a
( -
9
... . .. ./
e same e.w asse esse esse emme See tone see see esse nos See Tlas ist CMC 21 m . Fig. 4.131 COLLAPSED WATER LEVELS IN CORE 50 NIN VS 80 HIN
- i. %. q
, "=
t l
\
d
- h -
t,
.J
- g
\\ \.
i \ +
8 -
g .
e (y
'N m .
.s.
Tlee tel CASE 21 Fig. 4.132
\
C3RE Ca?LAf4T TEMPERATURES: T3P NODE 50 MIN VS 60 MIN time twee some
~
j
- I e .
eso 2 ~
I a i k /
Y /
s f
Lj %
tuo e tems amme asse m amme m ano asse asse asas me esas ammo flee ts)
CASE 21 Fig. 4.133 l
I l
i l
o #
l *
- 5.0
SUMMARY
AND CONCLUSIONS 5.1 Modeling Uncertainties Like all computer codes, RELAPS is limited by the phenomenological models built into the code by the code developers. In addition, however, RELAP5 permits the user to vary the nodalization (an option not present in CESEC and some other codes) and to utilize the " control functions" to override certain built in RELAPS models. This latter option was not invoked in this analysis, however there are important effects of the built in modeling which definitely affected the absolute values of numbers computed during these transients. Thus, in interpreting this analysis, the following modeling ,
uncertainties and variabilities must be considered.
- PORV flow modeling The PORV flow was initialized at the rated steam flow for those valves. As the pressurizer fills up,' the flow becomes two phase and
\ the computed mass flow rate changes markedly from the rated value.
This is phenomenologically expected to happen, but there are defi-nite uncertainties in the flow rate. No experimenta! data is cur-rently available, and the RELAPS model is probably no worse than any other model, nor can we say with confidence that it is significantly better.
This uncertainty in actual mass discharge rate causes a concom-itant uncertainty in the primary inventory and therefore in the depletion of coolant from the core region. However, we do not believe this uncertainty negates the trends computed as differences between cases. Thus the trend conclusions of the study are not affected by this uncertainty.
e Flow regime modeling RELAPS has a family of flow regime " maps" in which the drag between liquid and vapor flow is computed. The state of the art of a
~
.~
, N(T.s.
_ %s .
modeling flow regimes is weak, and the RELAPS aps refle(t this weakness. When RELAP5 indicates th'at flow changes from bubbly flow (bubbles rising due to density d3fferences) in,which the drag is very high and the slip betwe'en phases small .'to}a high slip flow regime (such as annular flow), . th3 flow pattern changes markedly.
Similarly, the stratified flow in horizontal pipes changes rather sharply as a function of local vOor fraction and vapor and liquid mass fluxes. ,
\
In the primary loop ,
These maps strongly influence the computed distribution of mass around the primary loop (for example because of potential countercurrent flow in the hot legs) and the flow into the surge line (because the stratified flow map couples with the fact that the surge line joins to the bottom of the hot leg) and therefore the PORY ilow.
In the core More importantly, however, the maps dominate the computed behavior in the core. As we observed in the main body of this report, the vapor fractions in the core region tend to hang up in the 25-30% range for the entire axial extend of the core for extended periods of time. When core uncovery is computed to occur, it frequently takes place in such a manner that the local vapor fraction jumps from .3 to 1.0 over a very brief time interval. This is purely a result of the flow regime maps in RELAPS.
That is not to say that other codes do this part of the computation with more accuracy. In fact, there is very little data on vapor distributions in core bundles during the quasi-static Dolloff condi-tions characteristic of a small break LOCA -- which is the situation during the transients computed here. Every computer code has some model for its vapor / liquid drag and that model almost singularly I
6 e governs the mixture level in the core during core uncovery. Most of these models are based upon data gathered at relatively high flux rates characteristic of conditions on the secondary side of steam generators, since that is of major importance and is relevant to most transients of interest. Another large body of data exists for Large Break LOCA conditions of extremely rapid core draining and reficoding because of extensive concern over that accident in the past. Unfortunately prior to the accident at TMI-2, virtually no data existed for the quasi-static boiloff conditions -- and very little new data has been generated since that accident. Further-more, in general, systems code developers have not paid particulcrly strong attention to the low mass flux can'itions. d Thus significant uncetainty exists in the prediction of core mixture level, and therefore in the prediction of the specific time of onset of core uncovery. For computations which border on uncov-ery or uncover only slightly, it is not possible to state with any confidence that those situations would or would not cause core uncovery.
Houever, as uith the other uncertainties, ve believe that the gene nt trends computed are believable. Thus ue believe that over-att conclusions regarding the genent effect of eine of PORY or of 1 the efficacy of APS are valid.
The caution that ue put forth here goes touard the thztidity of any specific time of core uncovem and tornrd interpretation of borderline cases. If the code predicts extensive core uncovery, it is our judgment that the conclusion that core uncovem oculd occur is mennted. Simitarty if the code predicts that the core never uncovere and the vapor fnction never even get near 30%, we feel that a conclusion that core uncovery oculd not occur is a rmnted.
Taken in combination, these can be used to ex.ains the pre-dicted time of onset of core uncovery. When the onsec of core uncovem is predicted under 200 different situations to differ by
.m., _
e, s - ., - - , ,
-_ . ~ - -- --- - - - - . -_ _- -. . -. - - .
'e e l
thouensd esecnds, the trend is very credible. ifnen the onset only diffr*e by ane hundred seconds, ce vould question dethen the models are adequate to accurately mke the distin tion. l 5.2 Analysis Sumary and Conclusions i
In this section we summarize the results of calculations performed to examine the ability of a CE System 80 plant to avoid core uncovery during a total loss of feedwater event accompanying Loss of Offsite Power (LOOP) con-sidering the use of the CE Auxiliary Pressurizer Spray (APS) to lower the system pressure to enable High Pressure Safety Injection (HPSI) flow to cool the primary. We also examine the use of the Auxiliary Feedwater (AFW) system and the potential effect of installation of Pilot Operated Relief Valves (PORVs). In addition, we investigated the impact of LOOP by examination of what would happen without LOOP. A summary of the results are presented in Table 5.1.
Continued full power operation in cases 2 and 4 generated more O~ primary heat to be transferred to the steam generators and therefore more rapidly depleted the secondary side inventories. (These cases were accom-l panied by an additional roughly 30 s of full power operation because the reactor did not trip concurrently with TLOFW, but tripped later on high pri-mary pressure.) The steam generators dried out by roughly 500 seconds into the transient, after which time the pressure in the primary rapidly rose to the pressurizer safety valve setpoint where it remained. By contrast, if the reactor is tripped (which accompanies LOOP), the loss of inventory from the
- steam generators is much more gradual and they do not dry out until ~2400 seconds in Case 1 (and even longer in some cases with PORVs depending upon the PORV sizes and the timing of PORY opening). Once the steam generators dried out and the primary pressure rose to the safety setpoint, the safeties opened and primary inventory began to be lost. HPSI flow could not enter the primary because the pump head was insufficient to overcome the high system pressure.
Thus, once the safeties (and later the PORVs) opened, the inventory depleted at a rate which was dominated by the valve flow, since the primary pressure was constant. (The valves represent the only energy removal mechanism after SG dryout.) Thus inventory decay once tha safeties opened was essentially
- i. ;'
- , _ , , .,-, -- - . , . , - . . - - . - . . . ~ - . . . , , - , . . . - . , . , . . . , , . , , -
,b ,
same between cases 1 and 2, with 30 minutes extra time available in case 1 because of LOOP at the onset of the transient.
The APS system aza not useful either with or without offeite power, since it pas unable to depressurize tha primry pressure to a too enough value to per nit significant HPSI flou. (We note that in this study we assumed that only one charging pump was available for APS; however, since APS logic will turn off APS when there is less than 20*F subcoaling, the use of all three charging pumps would alter only the timing but not the substance of this resul t.) Although use of the APS did pemit depressurization to slightly below 1800 psia, since the HPSI maximum head was only 1775 psig only a trickle of HPSI flow entered the system. The use of APS did delay opening of the pressurizer safeties by roughly 300 to 400 seconds. However, that made vir-tually no difference in primary inventory and therefore had no impact upon core uncovery.
If the steam generators are not available as a heat removat mechan-
/~) ism use of a PORY appears to be necessary to depressurize the system to a 100 enough value to pentit HPSI and eventually accumulator injection to get into the primm and to cool doun in a feed and bleed twde. In this study, we assumed two PORVs were opened by the operator at some point during the tran-si'ent and were left in the full open position thereafter. Both the nominal sized PORVs (yielding a combined flow of 120 lb/sec at pure steam) and the large PORVs (270 lb/sec) pemitted operation in a feed and bleed mode. Nat-urally the larger PORY permits more rapid depressurization and is accompanied by more rapid primary inventory depletion.
After ECEV and LOOP, the operator vould have roughly 60 minutes to open the large PORVe, compared with roughly 40 minutes for the emiter valves. When the offeita power is available, the operator has on the order of 30 minutes to open the large PORVs and still avoid core uncovem compared with Issa than even 10 minutes for the smiler ones. Throughout this study we assumed that only one HPSI pump was available, however, we examined the effect .
of utilizing two HPSI pumps by performing a calculation similar to case 4ati. The results indicate that increasing the HPSI flowrate instead of HPSI pump head does not alter the transient scenario enough to avoid core uncovery, u
_._-s. .r
- e because with smaller PORVs the RCS pressure does not drop enough to permit a singificant EPSI flow.
These REAPS calculations indicate that to be effective the instal-tacion of too smil PORVs should be accompanied by installation of higher head BPCI pumpa, uhereas the larger PORVs oculd suffice withcut the higher head HPSIs. With the larger PORVe, the operator has at least a 30 minute vindou in uhich he can mke a decision to open them and to attempt to restore his AEW.
Housver uith smiler PORVs for some transients the vindou is less than 10 minutes. Thus, it vould appear preferable to utilize too large PORVs instead of two smil ones.
Restoration of AFW is effective in rapidly causing system depressur-ization and preventing core uncovery. After TLOFW and LCCP, the operator can wait at least 90 minutes before actuating the AFW flow, and even when the power is available if the operator can restore the AFW flow within 50 minutes he can avoid core uncovery. It should be stated and emphasized in the Recov-ery Guidelines that the operator should utilize the time available to make all V possible attempts to restore main or auxiliary feedwater systems in order to provide a primary decay heat sink for a controlled reactor cooldown before resorting to PORVs. Use of PORVs is a good backup measure when he is unable to restore feedwater.
v
_ _ ~1r-i__________Tr____._________________
l jljl{, lI1 A -
?
y r
e '
0 - 0 - 0 v 0 0 - 0 - - - - - 0 0 0 o 0 3 - 0 - - - - - 4 4 5 0 - 0 4 c 8 0 9 1 2 0 1 1 1 n 5 6 4 3 3
3 3
3 3 U
e
' r o
C 0 - - -
- - - - 0 00 - - - 0 0 - - -
M - - 5 - - 0 55 - - - - - - 2 7 - - -
U 2 9 62 1 5 C 3 2 45 3 3 C 5 A
3
- 7 . . . . . . .. - - - . . . . ..
I - 3 5 500 0 00 00 - - - 0 0 05 00 S 2 5 550 2 26 40 5 5 57 29 P 1 2 843 2 84 27 6 1 20 37 H 1 125 1 34 55 2 24 33 S - 0000 - - - - - - - - - 0 - - - - - - -
P - 0093 - - - - - - - - - 0 - - - - - - -
A 6468 6
- 222 y
t e 0 0 - - - ' 00 00 2 0 2 - 2 22 22 f 3 0 - - - - 33 33 7 5 7 - 7 77 77 a 6 0 66 66 7 9 7 7 77 77 S 2 3 22 22 R
Z P
t u 0 0 0 0 0 00 00 o
y 0 0 0 00 - 00 00 0 0 0 0 0 00 00 r 0 0 9 54 00 00 0 0 0 0 0 00 00 D 4 4 0 94 44 44 5 5 5 5 5 55 55 2 2 3 22 22 22 - - - - - - - - -
y r
a n
u s r
u .f .
S ee mi 3 3 3 33 3 33 33 0 0 0 0 0 00 00
. F il O Te 5 5 5 55 5 55 55 3 3 3 3 3 33 33 L R 3 3 3 3 3 33 33 T n oG iS
' . t l.
a i
5 t iO e nC . . . .. . . . . . . . , . . .. . .
l IP 0 O 0 D0 O 00 00 0 0 0 0 0 00 00 b C 0 0 0 0 0 00 00 a R 6 6 6 6 6 66 66 T
W - - - - - - - 00 - - - - - - - 00 F - - - - - - - 00 - - - - - - - 00 A 54 06 45 33 V - - 0 00 0 00 - - - 0 0 0 00 -
R - - 0 00 0 00 - - - 0 0 0 00 -
O 2 84 2 06 2 6 2 84
. P 1 12 1 33 1 1 12 p
i r
T r . . . .. . . . ..
8 8 8 8 8, 88 88 t
o O O O OO O OO GO 9 9 9 9 9 99 99 2 2 2 2 2 22 22 c
a e
R F
O . . . .. . . . .. . . . . . . . . .
L O 0 O O0 0 OO 0O O 0 O O O DO OO s n nn n nn s in n n nn n e ii nn e m mi imiimm nn i ii i i
o .
nS iP m mm m mm ii mm .
nS iP ii mm t t lA 0 00 0 00 t lA 0 0 0 00 p ec sa e
dd 2 34 2 56 50 ec sa e 2 1 2 34 00 i 79 dd 56 r a in - - - - - - a in - - - - -
c C .p ua g - - - - - - - - C .p ua g - - - - - -
s - - - -
e eo .I V VV V VV WW eo I Y V V VV D s ao S R RR R RR s .S R R R RR WW pP O OO O OO FF ao pP O O O OO FF BN OH P PP P PP AA BN OH P P P PP AA ll tt l tt l ll i ii l l i ii 0 a aa b bb 11 a a b bb 11 1 1 3 3 33 3 33 11 2 4 4 4 4 44 22 e
s e
s e
s e es s
e s
e s
e s
e s
e s
e s
e se es s s e
a a a a a a a a a a a a a a a C C C C C C C C C C C C C C C I , 'll . ,II ll* ir' .l I j 1 +i j ; !!f ,! ?Iji,
.i.i j l ll1Illllf l! l I
p_ - - - - . - - - . . - . _ .._ -
' gh : "
Q l
Acknowledoments l
This report was prepared by ANL staff in partial fulfillment of a project under the direction of the U.S. NRC Division of Systems Integration, R. J.
Mattson, Director; 3. Sheron, Branch Chief for Reactar Systems; N. Lauben, <
Section Leader; J. Guttmann, Project Manager. L. B. Marsh of the NRC Reactor Systems Branch provided direct technical guidance during the performance of this task.
ANL staff who provided input to this report were H. Komoriya and P. B.
Abramson, authors; K. Rank and J. Bracken, report preparation; and R. D. -
Wright, Jr., drafting services.
I l
O a
h e
t
,e I
l 0
I i
l - - - - . - . _ _ , . _ _ _ , __ _ , _ ,
.. .,_. ___. . ~ , - - . _ _ _ - - - - , . _ - , . _ _ _ _ . . _ - . - . _ - . _ _ _ _ _ - - . . _ - . - , _ ,,---- _. _ ,, ..,,_-,.- .. . ~ . -