ML20204F041
| ML20204F041 | |
| Person / Time | |
|---|---|
| Site: | Crystal River  |
| Issue date: | 04/05/1983 |
| From: | Willcutt G LOS ALAMOS NATIONAL LABORATORY |
| To: | NRC |
| Shared Package | |
| ML19284E734 | List: |
| References | |
| CON-FIN-A-7272 LA-UR-83-1078, NUDOCS 8304150254 | |
| Download: ML20204F041 (23) | |
Text
r L'A-U R 1o78 4
b 5cf EI6rgrunoer contract w-7 05.ENG.36 tos 4:amos hat.onar taeoratory is operates or me universny of caidornia for me unnea States Deoa TITLE:
TRAC-PD2 CALCULATIONS OF THE CRYSTAL-RIVER-3 TRANSIEh7 OF FEBRUARY 26, 1980 USING REVISED ASSUMPTIONS AUTHOR {S):
G. J. E. Willeutt, Jr.
SUBWTTED TO:
US Nuclear Regulatory Co::: mission
-~
Washington, DC 20555 ty accepiance o'in i a<i.cie tne puoi sner recogne.s met me u S Government reta.ns a nonencius~e reys'iy. tree ense io puoi.sn or repro co ene puesisned form of tnis conteewt on. or to sitow others to ao so. for u 5 Government purposes Tne Los Alamos Natenar Laboratory recuests tnat tne publisne' ecentity tnis article as wo a performee under the auspices of the U.S Dece tment of Energy O@ d w
n @ LosAlamos NationalLaboratory G
_ d
_ v G Los Alamos,New Mexico 87545
,po o n 8304150254 830405 st.e na DOCK 05000 2S
1 I
TRAC-PD2 CALCULATIONS OF THE CRYSTAL-RIVER-3 TRANSIENT OF FEBRUARY 26, 1980 USING REVISED ASSUMPTIONS
- By G. J. E. Willeutt, Jr.
ABSTRACT In this report we present TRAC-PD2 calculations of the shutdown incident that occurred at the Crystal-River-3 Nuclear Station on February 26, 1980.
These calculations use revised assu=ptions based on Ref. 1.
The predicted system depressurization and repressurization agrees much closer with the data than our earlier analysis.2 A brief synopsis of the event is presented, plant, model assumptions are discussed, the event occurrences are compared with the TRAC results, and conclusions and reco=mendations are presented.
I.
EXECUTIVE
SUMMARY
An automatic reactor shutdown occurred at Crystal-River-3 on February 26, 7 980.
Interruption of a power supply to the non-nuclear instrumentation caused erroneous signals to be supplied to the integrated control system (ICS).
The ICS then reduced the feedwater flow, increased the reactor power, and opened the power-operated relief valve ( PORV).
These actions produced a transient that partially depressurized the plant and then repressurized it to the safety-relief-valve (SRV) set point.
3 We simulated this transient using the TRAC-PD2 code and a generic model of a Babcock-&-Wilcox (B&W) lowered-loop plant. There are some differences between cur generic model and the Crystal-River-3 plant.
Our major intent was to deter-mine if we could calculate major trends with the generic model.
cWork performed under the auspices of the United States Nuclear Regulatory Co==ission.
I
+$
7-,P 2-y 2
An earlier TRAC calculation was based on information contained in Refs. 4 I
cnd 5. Additional information in Ref.1 indicates that feedwater was lost much g
earlier in the transient than was indicated in Refs. 4"and 5. Reference 1 also carh accurately fixes the tilne' of PORV reclosing, which was originally only known Eithin a 4-min period.
11 sing the revised assumptions, the transient was
~
rerun for two cases.
Case 1 nse[a 70RV model that gave 110% of the design flow cud calculated the Loop-A secondary pressure based on relief capacities and primary-to-secondary heat transfer.
Case 2 used a PORV model that gave 100% of i
the' design flow and calculated-the Loop-A secondary pressure in the same manner os Case 1 until 225 s.
3etween 225 s and 510 s, Case-2 ramped the Loop-A secondary pressure down to the isolation pressure of 4.24 MPa (600 psig) and then maintained that value for the rest of the calculation.
Both cases agreed asch closer with the system depressurization than the earlier analysis and also predicted the repressurization well.
Both calculated' the Loop-B natural circulation well.
Comparison with upper plenum temperature data indicated the decay power was too large when the initial TMI-2 power level and an infinite irradiation decay-heat assumption were used.
3~
i Af ter the system repressurized, Case I was continued to 1800 s; there was no established flow in Loop A because there was no steam generator cooling to drive it.
High pressure-injection-system (HP1S) flow collected in the cold leg, cnd because the HPIS injection location is close to the pump, the cold water flowed back through the pump to the loop seal producing intermittent gravity-i driven reverse flows.
HPIS water flow to the downeomer was thus reduced, and l
downcomer temperatures never decreased below 520 K (476 T),
0 i
f.
We recommend that tests be conducted in large pipes to determine if this reverse flow is a potential phenomenon of concern or is. just an artif act of cne-dimensional models.
1
]
II.
INTRODUCTION i
We simulated the Crystal-River-3 transient using revised assumptions based on 'information obtained since an earlier analysis.2 We used the TRAC-PD2 code and a generic model of the B&W lowered-loop plant.
Section III of this report l
.,/
.r' f '.
~
~
_3 i
summarizes the observed transient event _s and discusses the most important changes in assumptions since our earlier analysit.
See, tion IV -summarizes the
~
TRAC generic B&W lowered-loop model and the initial and bouhdary conditions.
Sec. tion V presents the TRAC results and compares them with actual transietit~
plant parameters.
Section VI contains our conclusions and recommendations.
III. SLHMARY DESCRIPIION OF TRANSIENT A large uncertainty exists about when events actually occurred in the transient because so inich instrumentation data was temporarily lost.-
Our original calculation was based 'on a B&W analysis of the transient," which was
-1 ccmpleted only 11 days after the transient, _ plus' a Nuclear-Safety-Analysis-
' Canter report completed the month af ter the transient.5 Since then, there has baen an improved understanding of what probably occurred.1 N st'of our analysis is based on Ref. I with backup information and data' comparisons f rom Ref. 4.
The transient started when interruption of a power supply to the non-nuclear instrumentation caused erroneous signals to be supplied to the ICS (time =0).
The ICS then increased the steam flow to the turbines, reduced the 7eedwater flow, and increased the reactor power.
These combine-effects in-creased the reactor-coolant-system (RCS) pressure.
The power-supply failure caused the opening of the PORV af ter 1 s, and it was held open because of the powe r-supply failure.
Even with the PORV open, the RCS pressure reached the overpressure-trip set point of 15.96 MPa (2300 psig) sometime between 10 and 25 s, and both the reactor and the turbines tripped. The RCS pressure reached a maximum of 16.10 MPa (2320 psig) and then decreased because of the combined offects of decreasing reactor power, post-reactor-trip cooling, and~the open PORV.
At 201 s, the RCS pressure had decreased to 10.44 MPa (1500 psig), and the amergency-safeguards system activated the HPIS with the full flow of three HPIS pumps injected into the RCS.
At approximately 225 s, all four RCS pumps were tripped.
Sometime between 280 and 520 s :he operator shut a block valve (Ref.1 estimated this occurred at 450 s); this stopped the PORV flow, and the RCS pressure increased until the SRV opened.
I i
I
...-,--,_,,,.-..-.s,,-__,__.-
.l s,
(
,4, Reference 1 c'oncluded from pressure data that the main feedwater to both oteam generators was ramped down'f rom full flow.at 1. s to zero at 10 s, whereas carlier Ref. 4 had indicated the main feedwater was ramped'down between 13 s and 85 s, providing more cooling.
The effect of the later rampdown was observed in 2
cur earlier TRAC results where _the code-calculated depressurization was much fcster than actually occurred.
In our earlier analysis, we closed the PORV at a linear rate between 280 and 520's.
la the analyses presented in this report, we ecsumed, based on Ref.1, that the PORV was closed at 450 s.
In the earlier cualysis, we assumed the reactor tripped at 25 s, but with the feedwater ramped dsvn earlier the reactor tripped on high pressure at 17.5 s.
IV.
MODEL AND CODE DESCRIPTION A.
TRAC Noding Tigure 1 shows a TRAC noding diagram for the B&W lowered-loop model representing Crystal River-3.
This model was based on one developed f or Tu.1-2 using boundary conditions from the Crystal-River transient.
The model includes two' loops, identical except that the pressurizer is connected to Loop A.
Each
__ loop includes a hot leg with candy cane, a steam generator, and two cold legs combined to increase calculational efficiency.
Each combined cold leg includes a loop seal, a pu=p, and a HPIS connection.
The RCS pumps are modeled with the LOFT pu=p characteristics built into TRAC, but scaled with TMI-2 pump data.
Each steam generator secondary is attached to a main-feedwater inlet, an auxiliary-feedwater inlet, and a long pipe to the steam outlet with a side con-nection to a saf ety valve that vents to the atmosphere.
The TRAC-PD2 steam-generator model did not include an aspirator model, so the mixed feedvater plus-espirator flow was supplied as a boundary condition.
We noded the vessel with two azimuthal segments, two radial segments, and oev'en levels.
The seven levels include a lower plenum, three active core levels, two levels in the upper. plenum to permit the vent valves (Level 6) to be obove the hot-and cold-leg connections (Level 5) in case of water level changes, and an upper head.
The model includes connections from the upper head to each hot leg to simulate the upper-head circulation observed by B&W in their flow tests.
These connections are needed because the flow up out of the core,
/ f'.
through the center of the upper plenum, into the upper head with turning cutward, and back down into the upper plenum is not modeled yell with a single radial ring.
We adjusted the resistances to give the full-load flow split with 36% going through the upper head and 64% going directly from the upper plenum to the hot leg.'
Geometry and other plant data were obtained from the THI-2 Final Safety Analysis Report, other TM1-2 data sources, and Crystal-River-3 data sources.
Noding changes to improve our modeling include a 20-cell pressurizer (vs 4 cells in the earlier model) for the refilling process, a 9-cell PORV and 9-cell SRV (vs 2 cells) for the liquid-discharge process, and 6-cell cold-leg
' tees (vs 3 cells) for the HPIS injection location.-
B.
TRAC Code Description We used the TRAC-PD2 code with the following additional features.
We added a vent-valve model to the vessel and an auxiliary-feedwater system to the steam generator with control based on a steam generator level calculation and operator
-at t ion.
L e auxiliary feedwater enters near the top of the steam generrtor.
We also improved the modeling of the mixing of liquid and vapor between one-dimensional cells in horizontal and vertical low-flow regimes.
His TRAC varsion also included several other improvements and corrections to the TRAC-PD2 code. This same modified code version was used in our previous calculation.2 C.
Initial Conditions and Boundarv Assumptions Two cases were modeled.
Case 1 used a PORV based on 110% of the design flow of 12.6 kg/s (100000 lb/hr) of steam at 15.65 MPa (2255 psig) whereas Case 2 modeled the design flow.
Case 1 modeled the steam generator secondary pressure in Loop A using a relief rate based on the secondary relief capacity including the turbine-bypass capacity and the prima ry-to-secondary heat transfer.
Case 2 used a PORV based on 100% of the design flow and calculated the' Loop-A secondary pressure in the same manner as Case 1 until 225 s.
Between 225 s and 510 s, Case 2 ramped the Loop-A secondary pressure down to the isolation pressure of 4.24 MPa (600 psig) and then maintained that value.
The following initial conditions and boundary assu=ptions apply to both cases.
i
~
^*,
+
The model used the TMI-2 2772-MWth initial power and the 1979 ANS decay curve based on infinite irradiation and including the. actinide contribution.
~
' Table I shows the total EPIS flow delivered vs pressu assuming three, pumps running.
The numbers in.the table are from Ref. I for pressures. above 10.44 MPa (1500 psig) and Ref. 6 for lower pressures.
Each loop received half a
of the HPIS flow.
The HPIS was turned on after a 35-s delay when the RCS pressure decreased to 10.44 MPa (1500'psig)..
A 9-cell SRV model was used to get - the design steam flow of.' 39.3 kg/s (311700 lb/hr) at 17.34 MPa (2500 p31g). 11 sing this SRV model, a separate TRAC a
calculation determined that for saturated water at 17.34 MPa (2500 psig), the
~
, flow rate would be 71.8 kg/s (569000 lb/hr).
We modeled the period after re-pressurization with a negative fill that began opening at 16.65 MPa (2400 psig) and was fully open at 17.34 MPa (2500 psig) with a flow of 71.8 kg/s.
(569000 lb/hr).
This was used only for Case 1 because we stopped Case 2 when the system repressurized.
' Based on Ref.1, feedwater to both loops was linearly decreased to zero
__hetween 1 and 10 s in the transient.
Af ter 10 s, no feedwater was supplied to Loop A for the rest of the transient.
Af ter the RCS pumps tripped, the Loop-B feedwater was restarted at 226 s with 76.2 kg/s (1400 gpm) and remained on until the auxiliary feedwater was started at 510 s and the main feedwater was turned
{
off.1 Af ter the Loop-B auxiliary feedwater was started at 510 s, it remained on at 46.5 kg/s (740 gpm)1 until the level reached 92% on the operating range, and j
it was then throttled back to maintain that level.
The turbine stop valves were closed when the reactor tripped.
Subsequently, the secondary pressure-relief system was modeled by a flow vs pressure table (Table II) that combined the turbine-bypass capacity with the relief-valve capacity.
e
- J
~
['.
r o
V.
CALCULATION RESULTS e
Table III summarizes the significant transient events 'for the two cases (cee Sec. IV.C for description of the two cases).
The non-nuclear ~
instrumentation f ailure caused the PORV to open af ter 1 s and the feedwater to both stea= generators to be reduced from full flow at I s to zero at 10 s.
The raduced feedwater flow caused an increase in RCS pressure, a high pressure reactor trip at 17.5 s, and a coincident turbine trip.
Steam lost through th"e PORV depressurized the system as shown in Fig. 2.
Case 1 with the larger PORV matched the depressurization process better. Note that Ref. I used an even larger PORV with 155% of design flow and a si ilar
- TRAC-PF1 calculation will be presented in Ref. 7.
In the plant transient, the RCS pu=ps were tripped upon HPIS initiation (after a 24-s delay) as required by the USNRC s=all-break guidelines.
We x:odeled the pu=p trip at the actual transient time of 225 s.
Ihe HPIS flow was initiated in the calculations when the pressure decreased to 10.44 MPa (1500 psig) as it did in the plant (after a 35 s delay).
In bothi calculations
-the pu=p trip was forced at 225 s even though HPIS initiation occurred at 282 s for case 1 and 305 s for Case 2.
After the RCS pump trip, the Loop-B t'ain-feedwater flow was re-established for both cases.
Following RCS pump trip, the system continued to depressurize until the liquid became saturated and voids began to form (Fig. 3).
At this point, the depressurization slowed as the liquid flashed.
During this period, natural-circulation flow occurred in Loop B (Fig. 4) because of the secondary cooling, whereas the Loop-A flow almost halted (Fig. 5).
Note the data used for these two figures is taken f rom Fig. III-13 of Ref. 4.
We assumed the data labeling was switched in that figure because it shows the Loop-A flow continuing when the Loop-B flow stopped even though Loop B was cooled and Loop A was not.
Void formation continued and voids collected in the Loop-A candy cane and upper head, cspecially for Case 1.
(
)
,, / -.^ *
~ l p
For Case 2, decreasing the Loop-A secondary pressure to 4.24 MPa (600 psig) j batween 225 s and 510 s (Fig. 6) provided a small amount..of additi,onal cooling i
that kept the Loop-A flow from stopping co=pletely and limited the void formation in the Loop-A candy cane.
The later Loop-A secondary pressure d crease for Case 1 was caused bf secondary-to primary cooling.
As a result of the flashing and of the injected HPIS flow exceeding the PORV flow (Figs. 7 and 8), the pressurizer water level increased until the pressurizer filled at 360 s for case 1 and 405 s for Case 2.
Following the pressurizer filling, the primary system repressurized as the HPIS flow continued to exceed the PORV liquid critical-discharge until the SRV opened at 704 s for Case 1 and 660 s for Case 2.
Case 2 was halted once the SRV opened, and the
' C:se-1 transient was continued to 1800 s.
Figure 9 shows the calculated core outlet temperature (lower part of upper plenu=) for both cases compared with plant data.
The plant data was obtained from subcooling alarms until RCS pump trip and from thermocouple data thereafter.
Until.the system 'repressurized, the curves are fairly close together.
- However, the plant data does not show the small temperature
~~escillations that arise both from the collapse of voids prior to 700 s and from j
Loop-A density flow oscillations between 700 and 1800 s.
For later times (Case 1 only) the TRAC result does not decrease as rapidly as the data.
The main reason for this is probably that the TRAC-calculated decay power is too large because of an initially larger power (based on TMI-2 rather than Crystal River) and an infinite irradiation assumption.
A smaller effect is that warm water from the Loop-A candy cane is brought into the upper plenum by intermittent reverse flows.
After the system repressurized to the SRV set point, we continued the calculation to 1800 s to determine how cold the cold-leg and downconer water would get.
From 700 to 1800 s, the Loop-B flow continued because of steam-generator cooling, and there was no significant HPIS cooling of the Loop-B cold-leg fluid.
However, the Loop-A flow was nearly zero for long periods be-cause there was no steam-generator cooling to drive it.
During these quiescent periods, the HPIS water entered and started spreading both ways from the injection location.
When this cold high-density water reached the pipe leading t
a
.! f'.
down into the loop seal, the density difference produced a flow back through the loop and out the hot leg to the vessel.
This reverse" f-low. rapidly damped as warm water from the vessel entered the cold leg and mixed with the HPIS flow to remove the driving force (Fig. 10).
Figure 11 illustrates this density-variation driving force by showing the products of density, gravitational I
6 constant, and height for the fluid from (1) the loop-seal low poin't to the pump nidpoint, (2) the loop-seal low point to the candy-cane high point. (3) the hot-leg horizontal section to the candy-cane high point, and (4) the cold-leg horizontal section to the pump midpoint. This intermittent flow prevented cold I
un=ixed HPIS water f rom reaching the vessel and thus the downcomer temperatures remained above 520 K (4760F) even though temperatures in the Loop-A cold leg near the HPIS location were as low as 310 K (98 F) during the quiescent periods (Figs. 12-14).
The intermittent flow would probably eventually stop because cach surge produced a colder loop-seal temperature (Fig.15), and as the loop seal filled with cold water, the driving potential would be removed.
4 Another f actor besides the intermittent reverse flow in Loop A that keeps the downcomer f ro= getting very cold is the flow through the vent va.lves between.
the upper part of the upper plenum and the upper downcomer.
The ! t=perature of
$he fluid entering the vent valves and the vent-valve liquid veloctries for the two azimuthal segments are shown in Figs.16-18.
We have two concerns about the inter =ittent reverse-flow process.
- First, it may be a characteristic of a one-dimensional model.
It perhaps would be i
eliminated by a multidimensional loop-seal model that permitted cold water to flow down one side of the pipe while warmer water moved up countercurrent to it.
S3cond, if it does occur, a model that included both cold legs in each loop night show a flow from the vessel into one cold leg and back out the other cold log to the vessel.
More downcomer cooling could result as the cold HFIS water from one of the cold legs flows into the downcomer.
VI.
CONCLUSIONS AND PICOMMENDATIONS Two TRAC-PD2 calculations of the Crystal-River transient were performed.
C:se 1 used 110% of the PORV design flow, whereas Case 2 used 100% of the FORY design flow.
Case 1 assu=ed the Loop-A secondary pressure was as calculated l
l
=_-
7 y
{'
.A f'.
s t
based on relief capacity and energy transfer to and f rom the primary, whereas Case 2 linearly decreased the Loop-A secondary ressure. from 'the calculated
[
value at 225 s to 4.24 NPa (600 psig) at 510 s based on an isolation signal.
f
~
The system depressurization was calculated accurately for both cases, and the differences reflect the different PORV areas used.
Following the RCS pump l
trip, feedwater was re-established to Loop B, resulting in natural circulation i
in ' Loop B, which was calculated accurately for both cases.
The system rcpressurized following the PORV
- closure, and the repressurization was calculated accurately for both cases.
However, because of the nature of the.
l t
1 i
j candensation
- model, the code calculated condensation-induced pressure I'
cscillations during.the repressurization.
1 Comparison of calculated upper plenum temperatures with data indicated that 1
the decay power was too large because the calculation assed the initial TMI-2 i
power level and an infinite irradiation decay heat.
i f.
, Af ter the system repressurized, Case 1 was continued to 1800 s and there t
was no established flow in Loop A because there was no steam generator cooling lodriveit.
This resulted in the HPIS flow collecting in the cold leg, and because the HPIS injection point is close to the pump, the cold water flowed I
back through the pump to the loop seal producing gravity-driven reverse flows i
i through Loop A.
Therefore, the HPIS water flow to the downcomer was reduced, i
1 cnd downconer temperatures never decreased below 520 K (476*F).
i
)
We recommend that tests be conducted in large pipes to determine if this reverse flow is a potential phenomenon of concern or is just an artifact of l
ene-dimensional models.
i 4
REFERENCES s
i 1.
W. Brown, C. Caldwell, B. Chexal, and W. Laysan, "Dermohydraulic Analysis of Crystal River Unit-3 Incident," Nuclear Safety Analysis Center report NSAC-15 (June '1981).
2.
G. J. E. Willeutt, Jr., J. W. Bolstad, and M. W. Burkett, " TRAC-PD2 1
}
Calculation of the Crystal River-3 Transient of February 26, 1980," Ims j
Alamos National Laboratory letter report LA-SBTA-TN-81-7 (February 1982).
l.
. ~ -
,.___,.,,___--_,_--._,--.__.--,-.-..._.._-.,,,.,_--_-.,.-,_.,_.----_-,---~,.__.,...__.~r-,-.~,.~
1
~
3.
" TRAC-PD2 An Advanced Best-Estimate Computer Program for Pressurized Water Reactor Loss-of-Coolant Accident Analysis," Los Alamos National L'aboratory report LA-8709-MS, NUREG/CR-2054' ( April 1981),-
~
4
" Transient Assessment Report--Reactor Trip at Crystal, River"3 bbelear s Station on February 26,1980 (Preliminary)," Babcock & Wilcox report 07-08-02, Rev. 02 (March 9,1980).
5.
" Analysis and Evaluation of Crystal River--Unit 3 Incident," Nuclear Safety Analysis Center report NSAC-3 (March 1980).
6.
W. L. Jensen, United States Nuclear Regulatory Commission, Personal communication.
7.
P. Coddington, " TRAC-PT1 One-Dimensional Analysis of the Crystal River Unit-3 Plant Transient of February 26, 1980," Los Alamos National Laboratory report (to be published).
TABLE 1 TOTAL EPIS TLO'.J VS PRESSURE Pressure Flow Rate (MPa)
(esig)
(kg/s)
(gal / min) 0.10 0
96.6 1530 3.55 500 87.2 1380 7.00 1000 76.8 1215 10.44 1500 71.6 1132 11.13 1600 69.2 1095 12.51 1800 64.0 1012 13.99 2000 57.8 915 15.27 2200 51.2 810 16.65 2400 44.2 700 17.34 2500 39.2 620
.r' ;f'.
~
12 TABLE II SECONDARYPRESSURERELIEhCAPACITY
~
Pressure Flow Rate (MTa)
(psig)
(kg /s)
(Ib/hr) 0.00 0
0.0 0
7.00 1000 115.8 919000 7.35 1050 238.5 1893000 7.49 1070 378.7 3006000
~
7.62 1090 554.0 4397000 7.69 1100 799.4 6345000 TABLE III EVENT SEQUENCE Time Event (s)
Case 1 Case 2 0
0 Non-nuclear instrumentation f ailure 1
1 PORV opens Feedwater begins to linearly decrease.
10 10 Feedwater completely off 17.5 17.5 Reactor trip 225 225 RCS pumps tripped.
226 226 Feedwater re-established to Loop-B steam generator.
282 305 HPIS turned on 333 375 Loop-A flow stops on high void fraction in candy cane.
360 405 Pr'essuriser filled (liquid flow out of PORV) 450 450 PORV closed 510 510 Main feedwater turned of f and auxiliary feedwater turned on to Loop-B steam generator 704 660 SRV opens first time
,-'e-
-n-4
'f De
.. n. 9),...,
y t
1 11
... a
()
$e O
'"re e
',.,. '~ a e
Q=
,.. 9......n, p
9 l
-gw, O '.' B
'd 1,
...<..x a
g, y (,
l,$@-,,g
- < w.D@ W c6 Da=.
, e._.
e......_.,
,,,,,.....e.....{o.,
cc
_ O=
u O,
. ~ D,. e...
@D.
i.eis z.ss e 8"'87
.iess e
g ILOOP B) l LOOF A l ygg gg Tig. 1.
TRAC model of a B&W lowered-loop plant.
i I
+
s f
e
~
14 4
2.co s'
^*
(Dasw) TaAC CA$t 2 SOUD) TRAC CASE 1 g
2?no too.c'-
(CMAIN DASH) TRAN$l[NT DATA s
2500 s
tao,e'-
e,s
=
m E
\\
3tto w
I
.b c.
140.t'-
I f
m 3000 w
g f,,i.-
[
w use'-
\\.,;
,,,,o sm'.'
/
. soo
..)
, m.
200
- oo soo son woo woo woc moo see Tiut (s) l Tig. 2.
I Comparison of calculated and measured primary pressures.
s2 m
($0ilD) TeaC Cast 1 v.
(D45*4) TRAC CA$[ 2 i
T 4
sa-1 5
- u es-4=
g c.4 -
e.
I l
on
=
j... :ji, o
-os e
zoo soo soo soo eco voo wee
.oo woo flWE(s)
~.
i Tig. 3.
TRAC-calculated Loop-A candy-cane vapor f ractions.
l
..,f 100cc 5
- SO' lD) TRAC CASC 1-i g-N._
5
,DASM) TRAC C ASE 2 i
seco--
l CNMN DASH) TRAN$t[NT DATA.
- ~
i o
9 6
.e w
4 W
O 4000-S d
1 w
\\.
E w
w j
seco-
=
y e.
-noo C
100 300 300 400 600 m
Tiut (s)
Fig. 4 Comparison of calculated and measured Loop-B flows.
- 00
" I'*8 S0' 10) TRAC CA$t 1
]%
DASN) TRAC CASC 2
.)
CMAIN DA$H) TRAN$ TENT DATA 17900
- 000 7
4000-s i
m O
w h
.000-o3 e
.s to0:
g
~
m m
3000-2 3400 8
' *bM 3000 o
se too 30o
.so m
Tiut (s)
Tig. 5.
Comparison of calculated and sessured Loop-A flows.
s'.s
'I I
g,,,
(SOLID) TRAC CASE t
~-
(DAsw) inAt cast 2 i
o _..
~
.h
.co 9
~-
i 2 owe'-
..oo a
l
\\, ' '
g T
.co
=
1 s
g aco 2.co s*-
.,oo c o*
3 C
2Co 400 60o 000 WCo 8 00 WDC
- Co suD Tiest (s)
Fig. 6.
TRAC-calculated 1,oop-A secondary pressures.
l no (50L10) TaAC CASC 1 (DASM) TRAC CA$t 2 po.
- Wo i
l so 7
N N
E n-E v
i w
3 3
O o$
=-
oc
.o.
m m
4 4
2 3
le M-
.g o
o o
too 400 000 000 soo uso woo soo moo flut (s)
Fig. 7.
TRAC-calculated total HPIS flows.
m
~
./ ln'
,o e'
c s
as
,a (souc) TaAC cast t (DASN) 1RAC cASC 2
- 0-
/ -ae 7
i n
./
s2 w-j
=
w
-30 B
t t
S
./ '.
I o
s m
.. ~....., '
y t
w e-g
~ -.~. -......
... - t
<x 3-,
i e
0 60
- C t&C 200 250 400 360 400 460 TlWE (s)
Fig. B.
TRAC-calculated PORY flows.
~
s00 scuo) inAC cast i T
DAlw) T8AC CASC 2 e00 C>AIN CASH) TR AN$1[NT DAT A j
g N. %..
373 m
n5
- i....
F-g N-E
.0 N %~y MO -
.a
=y g
s3s 3
%,~
640-O o
~
=
8 g
e and
". 47$
490 0
300 400 400 000 E0 U#
W W M
TlWE(s)
{
te outlet (upper P enum) temperatures.
l Co:::parison of Calculated and measured
- 3 r
- ~.
-l8-i e
et' t
n' 400--
- ... 900 200-sD0 5
i
=
r 3
0
=
=460 o
Q
- M-W Y
e w
w
-. 900 m
e
=400-j 4
43 2
=400-
-.. s 60
-400-
~
-.. g
-*00 400 SCO IDCC
=00 WOO 900 3 00 Tiut (s)
Fig. 10.
TRAC-celculated Case-1 Loop-A flow details.
280000
)
{t AL TO Pup CNb l
N T
N T oL LEG fo Puw popogo.'
,........... '~ ~*" - ~~~~~~*-~~~***** ' '~
n 7.,
see00 5.
30 m
2 o
[ wee 00-,.... -
... _.. - -.. -.. - _...... a f
_2_
i f
9e000-
~.
o
- 00 eCo se0 use use
.0 Tiwt(s)
Tig. 11.
TRAC-calculated Case-1 Imop-A Tho 3-h products.
4
U
/
.s.
e (SOLID) TEAC CASE 1 (DASH) TEAC CASE 2 eec t
bec-4'
$73 m
m5 l
'.r b
I it.
960--
1!.
- *SSO f !.
/
N f
Rn I
=
\\
)-
- 52S e
MO-e t\\
o 3
}t c
g I
too g
5
~.;
320-.
a73
.e40 See O
200 400 000 800 SCG uso MD0 S 00 ese TiWE (s) i Tig. 12.
TRAC-calculated downcomer liquid temperatures in cell attached to Loop A.
600 (50 LID) Teac CA1[ 1 (DASH) TRAC cA$t 2 uo.
4 S79
^
E' I
D W
E wo--
..m q
jf
'S25 w
~
- u. -
e a
9 t
.we 8
I D
33,.
-. tS ele 300 m
m soO
=0 u00 woc woo wa TlWE (s)
Fi g. 13.
'rRAC-calculated downcomer liquf d temperatures in cell attached to Loop B.
t s
o
~
,-f*
850
,y,,
(SOLID) 70AC CASE 1
, * ~
(oASN) T0AC CASE 2 000 s
^
- ft.[*.
^
t.
g s
m.
- 9 s
-800
- .I !.;
I-I SCO-
=
p (l j:k\\
l r.
=
=
=
v..
t e
300
-8 8
.00-
.e
.e 300 350-100 300 0
2C0 00 000 000 1000 uG3 1400 1000 1000 Tlut(s)
Fig. 14.
TRAC-calculated loop-A cold-leg temperatures at HPIS connection.
~
660 700 (SOLID) TaAC CASC 1 (DASH) TRAC CASE 2 600-800 a
m E
se0-g s00 C
5
?
~-
g 400 W
.30-g 9
300 9
m-8 s
,m 390-30 0
300
.00 M0 000 1000 1893 M00 SW t000 flut(s)
Fig. 15.
TRAC-calculated Loop-A cold-leg temperatures at bottom of loop seal.,
e
t t
- ,o
,r f'.
n.
\\
soo
($0 LID) TRAC CASE 1 (DASH) TRAC CA1! 2 M~
\\
see.
p v
w see.
. > tse 4
4 m
ac k
12S e
w 640-O O
'I88 s
s S20-.
4ys
- bO Me j
0 200 400 400 000 sce uos teoc soo use i
Tiwt(s)
Fig. 16.
TRAC-calculated vent-valve supply temperatures.
os (SOLID) TRAC Cast 1 0.4 -
(DASW) TRAC CA$t 2 s
7 7s s
3.,
u-q.
t
/*.
[V t
8 3
8 d
"W
.u.
-oa.
S
=
=
9
\\
9 g
g a
a
.3
.i 0
300 800 000 800 10 :3 1200 6400
- 00 1000 Towt(s)
TLg. 17.
TRAC-calculated vent-valve velocities on Loop-B side of vessel.
~
r 84 r'
(SOLID) TRAC Cast i e.4 -
.(DASN) TRAC Cast 2
- a+f.
7 e.
8 8
.a a
Y Y
.O.2 -
e a
2 2
S
.e.4 -
a a
9 i
f
,4 g
g 5
ad e
04-i a -4
.t c
200 400 6co aos woo 2 00 ' wec 3 00 eso Tsut (s)
Fig. 18.
TRAC-calculated vent-valve velocities on Loop-A side cf vessel.
e 6
ed 4
6 e
a