ML20024C587
| ML20024C587 | |
| Person / Time | |
|---|---|
| Site: | Crane  |
| Issue date: | 12/29/1982 |
| From: | EDS NUCLEAR, INC. |
| To: | |
| References | |
| TASK-06, TASK-07, TASK-6, TASK-7, TASK-GB GPU-3138, NUDOCS 8307120844 | |
| Download: ML20024C587 (50) | |
Text
f m
.C.
Q c
6 i.
4 I
I i
i 1
n.
s r.,,
9,;.,.; e,
,. ; ' i
"-i -
5 f
4
=4 e5 m.
-.p
'. % 2*
- r. ; -
8307120844 821229 DR ADOCK 050002 9 l
1 t
e O
N ca Td 4
0 y#1b g1, -
f g
8 4
f q
/
w
/
~
s' i
8
/
~
/
D s
/
l
/
u
/
2 i
/
/
f i2 h
O g'
g 5
i a
l
~
~
s 1^
f Ch g-1 e
e-
' -)
r O
t e
C 1
ANALYSIS OF REACTOR COOLANT SYSTEM MAKE-UP DURING THE THREE MILE ISLAND l'
UNIT 2 EVENT c
Prepared for Kaye, Scholer, Pierman, Hays & Handler
(
By
(
EDS Nuclear Inc.
December 29, 1982 Report No. 02-0370-1122
(.
b
~
t i
l
(.
02-0370-1122
('
s c
ANALYSIS OF REACTOR COOLANT SYSTEM MAKE-UP DURING THE THREE MILE ISLAND UNIT 2 EVENT C
TABLE OF CONTENTS Section Page
1.0 INTRODUCTION
1 ir 2. 0'
SUMMARY
OF RESULTS 2
3.0 SYSTEM DESCRIPTION 4
4.0 METHODOLOGY 6
(
5.0 RESULTS AND CONCLUSIONS 9
6.0 REFERENCES
18 l
7.0 TABLES AND FIGURES 19
(;
,(.
c 3
L
.-e--
4 t
e
('
f-
1.0 INTRODUCTION
02-0370-1122 s
r C
(
(
(
i l
(
I C
+
l 1
i l
l k
\\
f
'~
o
?
02-0370-1122 r
Page 1 s
i 0
1.0 INTRODUCTION
on December 14, 1982 the law firm of Kaye, Scholer, Pierman, Hays & Handler requested that EDS Nuclear, Inc. perform an analysis to determine if there was a High Pressure Injection actuation at approximately 5:40 a.m. March 28, 1979.
The primary objective
(
of this analysis was to determine if an observed decrease in the TMI-2 make-up tank level at approximately 5:40 a.m.,
March 29, 1979 was due to a high pressure injection actuation, or due to other sequences of events.
(
This report documents the results of our analysis.
Section 2.0 presents a summary of results; Section 3.0 presents the system description; Section 4.0 describes the methodology used; Section 5.0 presents detailed results and conclusions; Section 6.0 provides reference information; and Section 7.0 contains tables and figures, i
(
(.
l
(
w.
o
6 8
e e
O e
2.0
SUMMARY
OF RESULTS 02-0370-1122 r
s C
C C
C
(.
C.
C.
t
c 02-0370-1122 g
Page 2 s
2.0
SUMMARY
OF RESULTS c.
The Three Mile Island Unit 2 (TMI-2) make-up tank level exhibited continual changes in level during the course of the TMI-2 accident on March 28, 1979.
These changes resulted from combinations of c.
normal reactor coolant system flows to the make-up tank, normal make-up flow from the tank and/or full actuation of the emergency core ecoling system (ECCS) high pressure injection pumps.
From computer alarm printer data, ECCS actuation is known to c
have occurred during the initial few minutes of the accident at 4:01 a.m.,
and again at 7:20 a.m.
Plant computer alarm data were unavailable for a period of time (5:14 a.m. to 6:48 a.m.) so it is not known from the alarm printer if ECCS was actuated 4
e during this period, particularly at 5:41 a.m.
The objectives of this analysis are to determine if the make-up tank level can be used to identify ECCS actuation and, if so, to determine if ECCS was actuated at 5:41 a.m.
C Hydraulic analyses were performed for the make-up/high pressure injection system.
The results of the analyses were calculations of make-up tank level following an ECCS actuation at any point in L
time.
The hydraulic models were benchmarked by comparing these calculated make-up tank levels to measured levels at points in time where ECCS was known to have actuated.
1 L
The conclusions drawn from this analysis are:
l Because ECCS actuation produces characteristic changes in the make-up i
tank level, the make-up tank level can L
be an indicator of ECCS actuation.
t r
L l
r 02-0370-1122 c.
Page 3
\\
e The zake-up tank level increases upon full ECCS actuation under certain make-up tank conditions.
These conditions existed at 4:02 a.m. on March 28, 1979, and caused the make-up tank level to increase when the ECCS
(:
actuated.
Under other conditions, the make-up tank level decreases upon full ECCS actuation.
The decrease is initially very fast compared to the normal C
9 I
make-up outflow, and then decreases 4
asymptotically.
Such conditions L
existed at 7:20 a.m. When the make-up p nk level decreased sharply at first Tand then the drawdown gradually decre_as ed.
py(
The 5:41 response of make-up tank level jp does not -exhibit the characteristics of
'fL e
an ECCS actuation.
In particular, there is no sharp decrease in the tank L" *"
level, and the level to which the i
t make-up tank descended was much lower than could be achieved following an ECCS actuation.
As a result, the level behavior at 5:41 a.m. cannot be attributed to ECCS actuation.
C C
(-
L F
8 6
e e
t 4
f' c.
3.0 SYSTEM DESCRIPTION 02-0370-1122 s
C C
(.'
(
(
(.
C
(
\\
~
N
i i
e 02-0370-1122 c-Page.4 s
r-3.0 SYSTEM DESCRIPTION The make-up/high pressure injection system consists of three pumps capable of injecting coolant to the reactor coolant system during normal operation and under c
abnormal conditions.
There are two sources of injection water: the make-up tank and the borated water storage tank (BWST).
The system is shown schematically in its normal operating mode in Figure 1-1.
C Under normal operating conditions pump 1-B
,is operating and is taking suction from the make-up tank.
At the initiation of the TMI-2 accident, make-up pump 1-B was operating in the normal make-up mode.
The make-up tank is supplied.with water from 0
the RCS letdown line, the RCP seal return line and the recirculation line from the f make-up pump.
Water can also be added to the make-up tank in controlled batches from the domineralized water system, the reactor coolant bleed tank and the boric acid C
tanks.
In order manually to reduce the make-up tank level, the letdown flow can be diverted to the reactor coolant bleed tanks.
An ECCS signal may be generated on low RCS pressure of 1640 psig or high containment i
building pressure of 4 psig.
Upon ECCS-actuation the two high pressure injection pumps 1-A and 1-C are started and the make-up pump 1-B is tripped.
Suction valves to the BWST (DH-VSA and 5B) are i
opened automatically as are the four HPI 1
(
injection line valves MU-V16 A, B, C, and D.
The pump recirculation valves (MU-V36 and 37) are closed.
Additionally, upon ECCS actuation on high containment building
(
i t.
r, 02-0370-1122 c
Page 5 s
C-Pressure, the letdown va.ve (MU-v2A/B and MU-V376), the seal return valves (MU-V25 and MU-V377) and the normal make-up valve (MU-V18 are closed.
The system configuration in the injection mode is shown in Figure 1-2.
Check valves in the C
suction lines from the make-up tank and the BWST. prevent backflow to either tank.
Table 1 lists system flow rates and parameters.
Figure 2 shows the make-up tank pressure operating limit.
C l
l b
c L
L L
\\
e
e g
e 8
9 C'
4.0 METHODOLOGY 02-0370-1122 s
C C
(
C C
L
(.
L l
l r
02-0370-1122 C
Page 6 S
C-4.0 METHODOLOGY Two hydrau'lic models were constructed to evaluate the make-up system behavior following ECCS actuation.
One utilized a lumped flow. resistance approach.
This model was used.to perform sensitivity studies'in order to define general system 0
behavior.
A more detailed network flow resistance model was constructed for use with the FAAST (Ref.1) computer code.
This model.was then._used.to confirm the results of the lumped flow resistance model.
(
Lumped Flow Resistance Model Figure. 3 illustrates the make--up system simulation using the lumped flow resistance.
model.
In modelling.the system the HPI and L_
letdown flow rates'are specified as functions of.. time.' Tna initial pressure and level in, the make-up tank and the BWST level.'are also-specified.
The model then determines the mass flow rates from the make-up t.ank and EifS.T. ~
(
P g +_Agh3 (t) +0gzB BB m mm
- ~
- N s..
<x x
s
-s In"th;e above equation "a.and*m are t
constant flew resi0tances for the BWSTr I
~" suction 1,1,ne and the'make-up tank sucti'on line respective 1;'.'sfrom s The., hydrostatic head.in s
the BWST is updated, e mass balance
^
PT 0b(*3
= -w,N -w N ' ~
.s
~
g y
('
at e
s x
~
.~~,.
s
.h
- Similarly, for the'make-up tank,.tho'
'i
^
hydrostatic head la updated.',N,
~
v
+
i 3
\\
\\
,,*""th OAs M - -w'tt) ++Ia)
's at m
s
~
k
.s
~
y ;b
, [
ay
/#
/
.t
02-0370-1122 c
Page 7 s
e The BWST and makeup tank flow rates are related by w (t) + w (t) = w, N g
L with the limits
. g(t) 3 w,(t) w W
3M gg3 B
L
,g 4
Floy rates outside these limits (i.e.
f' ~
y
~ negative flow in one of the lines) are
'~
pcavented by check valves in each line.
s u y
y s.
- ~;
_i Finally the make-up tank hydrogen cover gas C
', s s
pressure is given by the equation of state for jn ideal gas.
_.y k.
T PV s.=
Conatant
.N This equation assumes there is no flow of L
?.
(
hydrogen to or from the makeup tank during the individual make-up tank level changes s'
y x
.g
\\
being,pvaluated.
Isothermal gas conditions s
. were'assuR d since water sprayed into the r
make-up tank would tend to maintain the
.),
- cover gas at a constant temperature.
The 4-gas expansion coefficient y was therefore w
set equal to.l.'O.-
\\-
y w
.c 1
k e
g L
't
.N
. s g i -"
m.
,s W
i p
e x
f V"
adre
%g
\\
L.
$N g#., y.,
W
- g.
o
.. T
.-e--
-r
I i...
- t s
02-0370-1122 r
Page 8 s
1 t.
Network Flow Resistance Model Figure 4 illustrates the network flow resistance model used to simulate the TMI-2 make-up system with the FAAST computer code.
FAAST is a standard, public-use program for simulation of single-phase C.
fluid systems.
The code contains the following features for detailed simulation of the TMI-2 make-up system:
Capability to determine friction head losses in pipes as a function of local i
C fluid velocity Capability to simulate head losses through fittings, valves, elbows, etc.
as a function of local fluid velocity C
Capability to simulate the head vs.
flow characteristics of centrifugal pumps Capability to simulate the fluid head contribution of elevated tanks L
(unpressurized or pneumatic) as a function of current tank liquid volume Capability to perform time-history fluid system simulations.
C Output of the FAAST program consists of nodal pressures, element flow rates and tank fluid volumes as functions of time.
The FAAST model shown in Figure 4 contains representations cf the BWST (node 7),
make-up tank (node 11), RCS.(node 1),
make-up pump 1-A (element 1),' letdown i
(element 12) and the check valves in the BWST suction line (element 5) and make-up tank suction line (element 8).
The model also includes branch lines to allow flow e
diversion to or addition from other fluid systems, although these lines were not j
active in the analysis of interest.
1 w.
,-,.,,,.--.--..a.,.
..w,-,
-,,,,,-,,--.-,__.,,n_.,
,-,-----,.,..,ng,,-
,,w
--.-n,-
-,,m--
0 e
6 t
C 5.0 RESULTS AND CONCLUSIONS 02-0370-1122 s
C J
9 t
(.
t L
i
p 1
i e
02-0370-1122 l
r Page 9 s
r 5.0 RESULTS AND CONCLUSIONS Upon ECCS actuation, valve DE-V5A opens k
allowing the HPI pump 1-A to draw from the \\
BWST and/or the make-up tank.
Also, upon ECCS actuation, valve DH-VLB opens and pump 1-C draws exclusively from the BWST.
Pump C
1-C cannot draw from the make-up tank due to the check valve in the BWST suction line.
One of three alternative flow behaviors will occur depending on the initial conditions of pressure and level in the make-up tank.
These are:
Regime 1 All of the HPI flow is initially drawn from the BWST, i.e. the BWST hydrostatic prescure is sufficient to close the cake-up tank check valve.
In the absence of any flow into the
(
make-up tank, the make-up tank level will remain constant, as shown in Figure 5-1.
The initial conditions defining this behavior are:
"p h4 f
(o)+pgh(o)+pgs, > P L
P
+pgh (o)+pgz 3
B mH l
Regime 2 The HPI flow is initially drawn from the make-up tank and the BWST.
In the
(.
absence of any flow into the make-up tank, the flow from the make-up tank Ja yantattically approaches zero, as Rhown iii Figure 5-2.
The initial conditions for this flow behavior are:
P,g+Agh ( }+09*B i.
I I*m b B
m H P,g+0gh (c) +99z ~*B H g
B L
t,
..,,._.a
c
/
02-0370-1122 r
Page 10 s
C Regime 3 All of the HPI flow is initially drawn from the make-up tank.
As the make-up tank pressure and level decrease, a point is reached where there is a pressure balance between the hydrostatic heads of water in the BWST t
and make-up tank.
Further decreases in the make-up tank level cause the flow rate from the make-up tank to decrease and the flow from the BWST to increase.
In the absence of any flow into the make-up tank, the flow from 0
the tank asymptotically approaches zero.
This behavior is illustrated in Figure 5-3.
The initial conditions defining this flow behavior are:
P(o)+0gh(o)+pgz
<P
+pgh (o)+pgz d g
g BH Figure 6 illustrates the make-up pressure and level regimes defining each type of
(.
behavior.
Also shown in this figure is the operating limit on make-up tank pressure previously described in Figure.2.
Benchmark Analysis i
Results l
~
The lumped flow resistance model was us'ed to analyze the make-up tank level behavior during three known ECCS actuations at TMI-2.
These were:
approximately 4:02:39am on 3/28/79 l
c l
approximately 7:20:22am on 3/28/79 l
spurious reactor trip on 4/23/76 L.
t
~
e i
02-0370-1122 c'
Page 11' s
(-
4:02:39 AM Actuation The make-up tank level for the first few minutes following the reactor trip on 3/28/79 is shown in Figure 7.
Operator actions and automatic actuations were
(.
reportedly as follows:
On reactor trip the operator stopped letdown, started pump 1-A and opened MU-V16B in order to arrest the decrease L.
in pressurizer level following reactor trip.
1 At approximately 4:02:13 the suction valve from the BWST (DE-V5A) was opened in order to prevent the make-up tank level from being drawn down further.
At 4:02:39am ECCS actuation occurred, isolating the make-up pump recirculation flow.
l
(
The response of the make-up tank level was as follows:
The make-up tank level decreased sharply when letdown was isolated and pump 1-A was started.
0 When the suction valves were aligned (DE-V5A open) so that the pumps could theoretically draw from either the BWST or the make-up tank, the make-up tank level increased.
At this point the t
make-up tank level was low enough that, even if the initial make-up tank pressure was at the limit of the operating curve of Figure 2, the flow would be described by regime 1, i.e.
all flow would come from the BWST.
The o
make-up tank level increased due to recirculation flow from the make-up
' pumps and the seal return flow from the
, RCPs.
t e
,,__ -, $L.-
...._,,..,,_._,,_,_._,.m,
...._,,.m,.,.,,-,_.__,,.,_.,,--___,m.m
,.._.,_._,.._,,,m
o i
02-0370-1122 r
Page 12 s
r Upon ECCS actuation, the make-up tank level continued to rise although at a reduced rate.
The change in rate is due to isolation of the pump recirculation flow.
c In Figure 8 the reactimeter data are plotted along with the make-up tank level computed with the lumped flow resistance model.
The level calculation was performed using the system flow rates listed in Table 1.
The comparison of model results to
(
make-up tank level data is very good.
The major point to observe is that the make-up tank level increased when the BWST suction valve opened.
This behavior was predicted with the model and was observed in the reactimeter data.
This behavior is due to
(
the make-up tank pressure and level being low enough at this point in time that the make-up tank check valve closed.
7:20:22AM Actuation
(
Figure 9 shows reactimeter data for the make-up tank level between 7:15 am and 7:30 sm on 3/28/79.
The following operator actions and automatic actuations were recorded or are inferred from this reactimeter data:
U A full manual ECCS actuation was
'ecorded at 7:20:22 am.
This was followed by an automatic actuation on low RCS pressure at 7:23:53.
Prior to
('
this time pump 1-A was operating in the make-up mode and reactor coolant was being letdown to the make-up tank.
On ECCS actuation DB-V5A was opened theoretically allowing pump 1-A to draw from either the make-up tank or the BWST.
-_.,.__w------,.-
...._..,-,,._%_-.m
_.,.,_--..._.-.._,,_%--_,_..___%,,_rmy.
f r
4 i
02-0370-1122 r
Page 13 s
r The injection valves MU-V16A and B also opened increasing injection flow from the pump.
At 7:25:10 the level in the make-up i
tank suddenly increased.
This is an c.
indication that letdown flow to the tank increased, or that the make-up tank suction valve (MU-V12) closed.
The response of the make-up tank level was as follows:
r The level decreased sharply when the ECCS actuation occurred.
Flow rates in excess of 200 gym are calculated from the reactimeter make-up tank level data.
(
.The level increased at 7:25:10 when either the letdown flow to the tank was g,k/1/1 l7b {
increased to approximately 140 gym, or MU-V12 was closed.
b OT,,"teJ
$8 The level reacned a nearly constant b
C value of 67 inches, either due to a hydrostatic balance being reached g(
between the make-up tank and the BWST~
or, if MU-V12 was closed, due to 4. )f, ' Ig isolation of the letdown flow to the
(,
,s make-up tank.
4 t.
r 02-0370-1122 Page 14' r
In Figure 10 the reactimeter data are i
plotted along with the make-up tank level computed with the IQmped flow resistance model and with the FAAST detailed network model.
At the time of ECCS actuation, the make-up tank level was high (76.5 inches) t and therefore the expected flow regime would be Regime 2.
The initial pressure in the make-up tank cannot be determined at the time of ECCS actuation.
The operating limit on pressure at a level of 76.5 inches
(
is approximately 24 psig.
The observed level behavior is better predicted by the 7
models under the_ assumption of an initial make-up tank prCasure Tf 27 psig.
This is slightly above the operating limit.
Bowever, this event occurred after
(
significant amounts of hydrogen had been generated within the RCS and some of this hydrogen may have been vented to the make-up tank via the letdown flow.
A somewhat greater than prescribed make-up tank pressure could therefore have existed at 7:20 am.
t As seen in Figure 10, the make-up tank level is well described by the lumped flow resistance model.
It is significant to note that the calculated and observed flow
(
rates from the make-up tank during the first 30 seconds following the ECCS actuation were in excess of 200 gym.
A p sudden surge from the make-up tank is 3
characteristic behavior for a Regime 2 or i
Regime 3 ECCS actuation.
(..
f
~
i
.,,.._,_----.u-_.-,.
.-_.,n
--.----,,-,r-.
e 1
c 02-0370-1122 Page 15
(.
t The 7:20 actuation was also analyzed using r.
the detailed FAAST computer model in order to provide a verification check of the lumped resistance model.
The make-up tank level change predicted by the detailed model is compared to the measured level and the level predicted by the lumped model in Figu e 10.
The predictions of the two
(
models are in good agreement both with each other and with the measured data.
Spurious Reactor Trip on 4/23/78 A spurious reactor trip occurred at TMI-2 on 4/23/78.
It also caused RCS Pressure to decrease and ECCS to actuate.
Figure 11 shows the make-up tank level versus time following this reactor trip.
The make-up t
tank level shows the same characteristics as seen during the 4:02:39am sequence on 3/28/79 (Figure 7).
In particular, when ECCS actuated, the make-up tank level increased, even though the flow path from the make-up tank to the pump suction was l
C still available.
This is indicstive of Regime 1 behavior, which is consistent with the make-up tank level at the time of ECCS actuation.
Figure 12 shows the analysis of the make-up tank level transient using flow values from Table 1.
The behavior is well C
predicted by the model.
5:35AM to 5:54AM Mak e-Up Tank Invel change C
Finally, the lumped analysis model was used to investigate the make-up tank level behavior from approximately 5:36am to 5:54ap on 3/28/79.
A relatively constant level decrease occurred over this time interval.
This level change is shown in Figure 13.
l 1
i
-. ~. -.. -
a-
~
e 02-0370-1122 r-Page 16 s
r The benchmarked model was applied to determine if there could have been an ECCS actuation during this time interval at 5:41 am.
In this analysis a constant letdown 3) flow rare of 62 gym was assumed based on g the reactimeter data prior to 5:36 and
(
after 5:54.
Because of the make-up tank level, if there had been an ECCS actuation the flow would have been in Regime 2.
Figure 14 shows the calculated level g
ibehavior had there been an ECCS actuation.
(.
A sudden surge of water would have been drawn from the make-up tank and a hydrostatic balan'ce with the BWST level would have been reached at approximately the 70 inch level.
In fact, no such initial surge occurred.
The level did not c
reach a hydrostatic balance, but continued to decrease to the 59 inch level.
The absence of any surge in make-up tank flow at 5:41 and the continued drawdown to the 59 inch level are both inconsistent with an L
ECCS actuation.
Since DE-V5A would have had to open on an ECCS actuation, a make-up tank pressure of approximately 39.4 psig would have been necessary at the start of the drawdown at.
G 5:36am to continue drawing down the make-up,
tank to the 59 inch level.
This pressure is far outside the range of normal operation for the make-up tank and exceeds the pressure levels known to exist at the start of the accident.
A buildup of C
hydrogen pressure in the make-up tank after the start of the accident and prior to 5:41am would not be likely since this level change occurred prior to sustaining any core damage.
However, a second calculation was performed assuming an initial tank g'
pressure of 39.0 psig.
The results are also e
-,-.,-~,-----,.ne..
.e
-e
---,---.a
.-,--,--------,,-m
--,n,.._-,_,n.,.---,,-vw,,,.,,,-
n,n,
,.,,,-.,--,n-a-,-ng.-..
c-
.r-
r 02-0370-1122 Page 17 r
shown in Figure 14.
The calculated make-up r
tank flow behavior and level change contradict the reactimeter data.
Therefore, even assuming a higher pressure, the actual data are inconsistent with an ECCS actuation at 5:41 am.
c.
The make-up tank level behavior during the 5:36 to 5:54 time period was not different than a series of make-up tank drawdowns which occurred on 3/28/79 that were not coincident with ECCS actuation.
Since there was no observed surge in the flow
(
from the make-up tank (up to 200 gym would have been expected) and since the level decreased well below that which would balance the BWST hydrostatic head, it is concluded that the makeup tank level decrease at 5:41 was not the result of ECCS f.
actuation.
4
.-p
('
C
(
i i
w__.
,....,,..r,
_.---m
n
6.0 REFERENCES
02-0370-1122 p
s
(~
C
(
)
(
i I
(.
n 02-0370-1122 c.
Page 18 s
C'
6.0 REFERENCES
1.
'FAAST Fluid Analysis and Simulation Technique", CDC User Information Manual.
e I.
(.
t
(.
4 l
l 4.
I l
0 8
0
- * =
7.0 TABLES AND FIGURES 02-0370-1122 s
/'
[
(
J t
s-e N
e
02-0370-1122 Page 19 c.
s r
Nomenclature AB Cross sectional area of BWST Am Cross sectional area of make-up tank g
Gravitational constant h
Make-up tank level ha BWST level P
Makeup tank cover gas pressure Patm Atmospheric pressure t
time V
Cover gas volume WE Flow rate from BWST W
HPI flow rate 7g WL Letdown flow rate W
Flow rate from make-up tank m
l
('
2B Bottom elevation of BWST Z
Bottom elevation of make-up tank u
=B Flow resistance in BWST suction line L
=m Flow resistance in make-up tank suction line y
Gas expansion coefficient p
liquid density
(
N*
l l
02-0370-1122 e
Page 20' r
r TABLE 1 System Flows and Parameters
. Letdown Flow Rate 0-140 gym c
RCP Seal Return 4 gym RCP Seal Water 32 gym Normal Make-up Flow 0-160 gym EPI Flow 0-250 gym per leg Make-up Tank Volume 4388 gal Make-up Tank Low Level Alarm 55 inches Make-up Tank High Level Alarm 86 inches BWST Level above 53.1 ft.
(
L L
n n
n
.. _ =
FIGURE 1-1 l
Make-up/Let-down System, Make-up Mode Using Pump 1-B 1
32 c
. ! - ***e s
.e S
- 8
^
-m e,,,,
m as c e a v m..
- g av a.a.
- s u..moe eese w e
-as sa f
,c,,,
,c,,,
- j..
Ia.
A AE
'=0
-- r '
W a_m
=
co u -o q
v :,7 -
- mi. :. ::: c i
Il 1
- =
l **-.
11 11 ise.
- p;
- l5 e.=s,sr m
E E
a cosa.
e g*
e es pa po.pt 8 38 8" ***
e.
==
e.
- M Bos.
ensem aanerne i
7, e
as s.=
j Jin ess me meerie 1
33,,
4,g
's.s 6
'=qi
=r.
'a
- s, sa 'El h;
assi om- ' ines _
1 t a m * ** ****
A um
'M W
gv N4 A
n com-ee-es.mw nasene se g
o, v
s
,T. o c
evm
. l
',O
'aa' M
4 r,,,
. (' **
ea c aavsssso.m ca e as 4.,,.,,,; yc h, m.-
is as s.
e
,.7
-,.e
~.
, s'
-~
_g.
m..-
Ausiliary Building Containsneat Buildeng e.
n.
c.
s FIGURE l-2
\\
I i
e 1
Make-up/Let-down System, Automatic injection Mode i
e
=
i.. _ c
.~s
=
W h,
g
- 8"8 **" e
- UM_ '
i P=es.
_3M_
3^8 0 6Eram O,.
- c
- f. me sae.
-.g ca s en s. -
. se coe i
)
,j p"
ac' '*
ace se e
( a esame )acesa me a.s ace is O r@;;:- :-& EEs' L.,
i,,
l '-:: : : 4 ' i-?-
i I.
t3,
- .. " E evv 4
1 ir-l8a=*
l d'.
g 4
620 eras m.s b
_.y.
f.
436.coce.m _. in' l
- p i
as I.
- ",y, esime.asacres
'%* Q
_S_ _saa i
f.
as som 833 d
at aflA j
a'$
g i 4a, u
,g y,
j }"*
--.^:4 t e,
{
ine I
sa' M 23 ek
<,I s.
= === saa
.9.
1
[
n.
,! 4 g-h W.,,"#
. _ _ _ aTJ l
i.e.aa guo.
"'J
- - " " - ' ~ * " ' * * * * " "
j
,r, in.1 u
w
-l es c an.ase s.ea s j J j!
l E3 are as i
- - - - g$ x q ss em inc
,..r ai.c
- m->-
see Qato eres=2.sene Asamiliary Basilding Contaist.noat Building flGURE 1-2 l
n FIGURE 2 MAKE-UP TANK PRESSURE OPERATING LIMIT r
(Ref. B&W Limits and Precautions for TMI-2, p. 49-0 t
and Fig. 2.0-02) 60 50 C
a 40 3
5
(
S e
30
=
!l m.
(
20
(.
t l
10
(_
0 0
2O 4d 6'O 8d 150 Level (inches) s-p
,____.,_,-__.y-
r FIGURE 3 LUMPED FLOW RESISTANCE MCDEL
^
\\
/
Pata N
C JL L
P Make-Up Tank 9
4
(
2 %
=
z 3
=
Wm 9
Pump 1-A' c
o<,
4x,
= *a
(
=
(-
Pump 1-C w
"y J
t
{
a c
(
s-
,,..-n.-
s Figure 4 FAAST HODEL OF TMI-2 MAKE-UP SYSTEM
_ _ - - =
. = =
TANK
- Node j
~
~
Element 20
- 7 3
34
.g
.g c;);e.g g
- c, _
. 1.e 1.
x
- c,ee.1.e 2
e l
1' A
e e::.'O i
MAKE-UP PUMP -1A 4
k O *3
- 2OYb
^
v y
v,,
A k A k A k b
C FIGURE 5-1
(
FLCW REGLME 1 (All Flow Initially From BWST) f' f
l Make-up Tank Flow
(
e (no flow) e Time e
Make-up Tarik Level e>
3 Time
(
v._m x.-
r FIGURE 5-2 FLOW REGLME 2 C'
(Flow initially from make-up tank and BWST) t r
(~
-y H
Make-up tank Flow C
3 m
l l
C.
Time
(
Make-up Tank Level
,e b
S I
b c
Time L
~
+
4
.,-m--
ww ---
n-----.
e FIGURE 5-3 FLOW REGIME 3 r-(All flow initially from make-up tanic) s 4
f:
f.
W H
Make-up Tank Flow
(
B n.
(
Tirm
(
l Make-up Tank Level a
e>
Time f,
C FIGURE 6 MAKE-UP TANK FLOW REGIttES
\\
C 60 L
50
[
I Regime 3
/
i.
40 Regime 2 c.
3
_2, 30 l
7:20 on 3/28/79 o g
5:41 on 3/28/79 L
20 Reg [me1 4/23/78 l'
4:02:39 on 3/28/79, 10 L'. -
0 e
i i
a i
0 20 40 60 80 100
[
Level (inches) l t
l w
r.
FIGURE 7 c.
MAKE-UP TANK LEVEL 4:02:39 AM ACTUATION 80
(,
e Reactimeter Data e
a 70 -
C-I j
ECCS Actuation
'N 4
c.
g.
9 a
e a
Q
(
60 e.
O
(
DH-V5A Open
(_
50 4:01 4:02 4:03 4:04 Time l
t.
i
\\
6e
=4
- b.
g 1
/
'N a
.+
FIGURE 8
~
~,
o 3
s CALCULATED-MAKE-UP-TANK LEVEL \\
r' 4:02:39 AM ACTUATION "3g s
f 80 -
C
-- e Reactimeter Data,,
s s
Calculation (Luznped flow '
\\
resistance andel) f.
s.
\\.
- g.,
g, 70
\\,
(-
\\.
.,, s x y e>
g.
.3 s
.~
.w a x s ce3 e-eu c
s 44
- ,, e e *
- s' *
- 3~
e s
y 60 -
/,*
. ' ~
s 3
c e
50 r.
4 e
a 4:01 4:02 4:03 4:04 Time L
l l
, ~~
w.,
L-
)
s.g.."
+
, y c-
,x s
s-FIGURE 9
- ~~
~-
s:
MAKE-UP TANK LEt %
r 7:20 AN ACTU TION. N.
t b:
' s.'-
w.
- r. /
- 1
's %
(-
m -
i.
- Reactimeter Data s
ECCS Actuation 80 =
,,. l..
m y
e n
i t
e c
e w
~
f.
s
~3
.wc S
70 -
e e
t Ag E
l sp l
3<-
e t,
e s
s N.,
60 I
t i
e i
1 7:15 7:20' 7:25 7:30 Time
(.
L
FIGURE 10 CAICULATED MAKE-UP TANK LEVEL g
7:20 AM ACTUATION C
- Reactimeter Data c
Calculation (Lumped flow resistance model) 4 Calculation (FAAST network resistance model).
(
80 "
e
'e a
I 3
\\*
x g\\.
g b
70.
g, e
J 2
\\.
g
.....~...~.
y.e..
9 s
i t
60 i
7: 15 7:20 7:25 7:30 t
l Time
(
l t
i t
I FIGURE 11 MAKE-UP TANK LEVEL 7
4/23/78 SPURIOUS TRIP t
F 80
- Reactizneter Data
(
a C-e.c g
70 e
s=4 g
3 C..
t e
e.
.S T.
e e....'
.w c
g 60
(.
C 50 e
a a
O 30 60 90 120 150
(
Time After Reactor Trip (sec.)
e
=
.,...__....,____.,__,_.,,y,
('
FIGURE 12 CALCULATED MAIG-UP TANK LEVEL r
4/23/78 SPURIOUS TRIP g
I e Reactimeter Data 80.,
Calculation (Lumped flow resistance model)
(:
a
.I E
c.
c 70 -
3 ECCS Actuation
...** N
.g N
=
(
b g
I
\\'.
If a=
\\.
e
\\.
\\,.. /_
e *
\\
60 -
/
V
(-
?
(
l i
e a
a i
0 30 60 90 120 150
(
Time After Reactor Trip (sec.)
t l
e
C FIGURE 13 MAKE-UP TANK LEVEL r
5:36-5:54 AM F
90 -
e Reactimeter Data C
85 -
~
.O.
2 1
N c.
C Y 75 -
0
(
i 70 -
2 y
k.
65 -
e L
60 -
L.
55 e
i a
i 6
4 8
5:35 5:40 5:45 5:50 5:55 6:00 6:05 6:10 Time t
C i
FIGURE 14 e
CALCULATED MAKE-UP TANK LEVEL t
5:36-5:54 AM e
go -
e Reactimeter Data Calculation (Lumped flow 85' resistance model) g, a
E c
3 80 _
g
=
ay bo e
3 L\\
(.
75
\\
E
\\
k
[\\
2 5*
\\.
p e
(
E
\\
Initial Pressure 70 -
- ~
= 24.5 psig
\\
e L
65 -
e e
\\
60 s%-
i-
,t.--
Initial Pressure 31.5 psig a
(
55 i
e i
1 4
5:35 5:40 5:45 5:50 5:55 6:00 6:05 6:10 Time v.
e c3
4 9
e B$
se 4,'
s e
' -,+,
e
+
.. ?(
+
4'
's t,-
p
,s, o
t s
f
- .o
, fk,
s.
y p
1
'g.
4
. m
,y, e
- ar_,
I N4" e
&A E
b W
.m
.O..
A ah' e
1 s
r.
e,
.., v,s I
J 4
a g" e
4 4
e e
k g
4 f
R %.
4
%e, a
e i
n O
~
sab i
9 O
c-
A Y
w
~.$
9 h
- " * * * * " * * " * = =
ew
- y#
g p
D 8
4
/*
Q
)s
/
/
l e
/
e g
/
b
/
c
/
v
/
s k
=
t h
1 0'
f E
E e
o 4
&^
Cb 6