ML19347E921
| ML19347E921 | |
| Person / Time | |
|---|---|
| Site: | Zion File:ZionSolutions icon.png |
| Issue date: | 04/30/1981 |
| From: | Fletcher C EG&G IDAHO, INC., EG&G, INC. |
| To: | Disalvo R NRC OFFICE OF NUCLEAR REGULATORY RESEARCH (RES) |
| References | |
| CON-FIN-A-6354 EGG-CAAD-5428, NUDOCS 8105140271 | |
| Download: ML19347E921 (45) | |
Text
h EGsG,..
N FORM EG&G 390 I
(Rev.11 F9) j INTERIM REPORT Accession No.
Report No.
EGG-CAAD-5428 Contract Program or Project
Title:
Code Assessment anc Applicatio'ns Division Subject of this Document: Accident Mitigation Following a Small Break with Coincident Failure of Charging and High Pressure Injection for the Westinghouse Zion 1 Pressurized Water Reactor Type of Document:
Informal Report C. D. Fletcher
- esearci anc,,,echnical Author (s):
Assistance Report * */
Date of Document:
April 1981 Responsible NRC ladividual and NRC Office or Division:
R. DiSalvo, NRC-RSR This document was prepared primari'y for preliminary or internal use. it has not received full review and approval. Since there may be substantive changes, this document should not be considered final.
EC&G Idaho, Inc.
Idaho Falls, Idaho 83415 Prepared for the
.U.S. Nuclear Regulatory Commission 4
Washington, D.C.
Under DOE Contract No. DE AC07-761D01570 NRC FIN No. A6354 INTERIM REPORT J
Sp 5/ 9's> 7 /
O a
ABSTRACT Four RELAP4/ MOD 7 computer code calculations were performed to simulate the response of the Westinghouse Zion I Pressurized Water Reactor to certcin small break loss of coolant ' accident mitigation techniques.
The accident sequence modeled was a small primary break with coincident failure of the charging and high pressure injection systems.
Unless corrective action is taken this sequence 19 ads to a core uncovering and subsequent damage. Mitigation te hniques investigated were the opening of steam generator secondary atmospheric dump valves (ADVs) and pressurizer power operated relief valves as a means for depassurizing the primary system sufficiently to allow continuous low pressure injection.
The calculation results indicate opening all ADVs at 10 minutes following the break effectively mitigated the accident.
Since the results of this preliminary analysis are encouraging, recommendations are made for further investigation.
l l
h O
ii
SUMMARY
I V'
This report presents the results of a preliminary investigation into the feasibility of mitigating a pressurized water reactor (PWR) small break accident with coincident failure of charging and high pressure injection (HP I).
The Westinghouse Zion I PWR was modeled using the RELAP4/ MOD 7 computer code.
The plant was assumed to be operating at full power best estimate conditions.
Four calculations were perfomed using break diameters of 2.54 cm (1 in. ) and 5.08 cm (2 in.).
For the Zion plant the HPI pumps are called the safety injection pumps.
Because of the small break size and failure of injection systems, the primary system pressure remains alove the pressures required for accumulator or low pressure injection (LPI) flow.
If no operator corrective action is taken a core uncovering and subsequent damage occur.
The corrective actions represented by the calculations include the opening O
of one Atmospheric Dump Valve (ADV), the opening of one ADV and two power operated relief valves (PORVs), and the opening of all four ADVs.
The purpose of '.he corrective action is to hpressurize the plant primary sufficiently so that the LPI system may be continuously used to replace primary fluid lost through the break.
The results of the calculations indicate that all four ADYs should be opened.
Failing to do so leaves some of the steam generator secondaries 3
hot thus creating a mechanism for repressurizing the primary at a later f
time.
Opening all ADYs at 10 minutes was found to be an effective l
technique for preventing 6 core uncovering for both break sizes.
Recommendations are made for additional studies to detemine the latest time corrective action will be effective and the effect of only 1
partially opening the ADVs to allow a slower depressurization.
The results
{
l of these studies are expected to be highly dependent or. break size.
r l
a I
i iii L
I
ACKNOILEDGEMENTS The author wishes to thank A. C. Peterson, J. E. Koske and S. J. Bruske for their independent review of this report, C. Polk for her timely contributions in preparing the graphics, and the typists and proofreaders in TSF/TSB Word Processing for their support.
m O
W I
i f
I O
iv i
CONTENTS i
AB ST R A C T............................................................
1 1
SUMMARY
..............................................................i1i i
1.
I NTR O D UC T I O N....................................................
I 3
2.
TECHNICAL APPROACH..............................................
2.1 Ca l cul a ti on s Pe rf o rme d....................................
3 5
2.2 Computer Code.............................................
2.3 Mo del a nd Co nd i ti o n s......................................
5 3.
RESULTS........................................................
13 3.1 Ba sic Sequence Descri ption...............................
13 3.2 Run 1, 5.08 cm (2 in. ) Diameter Break, One ADV Opened at l
1 0 M i n u t e s...............................................
14 3.3 Run 2, 5.08 cm (2 in.) Diameter Break, One ADY Opened at 10 Minutes and 2 PORVs Opened at 25 Minutes..............
19 3.4 Rt.n 3, 5.08 cm (2 in. ) Diameter Break, Four ADVs Opened at 1 0 M i n u te s............................................... 2 5 3.5 Run 4, 2.54 cm (1 in. ) Diameter Break, Four ADVs Opened at 1 0 M i n u te s............................................... 3 0 4.
CONCLUSIONS AN D RECOMENDATIONS................................ 36 FIGURES 1.
Comp u te r c o d e u p d a te s........................................... 6 2.
RELAP4/ MOD 7 nodalization for the Westinghouse Zion I PWR........ 8 3.
Run 1 prima ry an d seconda ry pressures................... c....... 15 4.
Run 1 hot leg quality in l oops with closed ADVs................ 17 5.
Ru n 1 accumul a to r wa te r l evel.................................. 17 6.
Run 1 expanded scale primary and secondary pressures........... 18 Oi V
V
7.
Ru n 1 ves sel mi x tu re l evel.....................................
18 8.
Run 1 upper plenum and secondary fluid temperatures............ 20 9.
Ru n 1 pres suri zer mi xture l ev'el................................
20 10.
Run 1 s econdary mi xtu re l evel s................................. 21 11.
Ru n 2 prima ry an d seconda ry pressure s..........................
23 12.
Run 2 upper plenum and secondary fluid temperature............. 23 13.
Ru n 2 accumul a to r wa te r l evel..................................
24 14.
Run 3 pri mary a nd seconda ry pre ssure s.......................... 27 15.
Ru n 3 accumul a to r wa te r l evel.................................. 2 7 16.
Run 3 upper plenum and secondary fl uid temperatures............ 28 17.
Run 3 seconda ry mixture l evel.................................. 28 18.
Run 3 pressuri zer mi xture l evel................................ 29 19.
Ru n 3 ves sel mi xtu re l evel..................................... 29 20.
Run 4 prima ry and secondary pressures.......................... 31 21.
Run 4 accumul ator water l evel.................................. 31 22.
Ru n 4 upper pl enta mixture l evel............................... 33 23.
Ru n 4 pressuri ze r mi xture l evel................................ 33 24.
Run 4 upper pl enum and secondary fluid temperatures............. 34 25.
Ru n 4 seconda ry mi xture l evel.................................. 34 TABLES 1.
Summa ry o f Cal cul a ti on s Pe rfo rtne d............................... 4 2.
In i ti al Pl an t Co ndi ti on s....................................... 10 3.
Bo u n d a ry Co n d i ti o n s............................................ 1 1 l
O vi
1.
INTRODUCTION I
'v' Four basic emergency core coolant (ECC) sy;'. ems are used on pressurized water reactors (PWRs) to replenish primary coolant lost during a loss of coolant accident.
The charging pumps have the capability to inject water at pressures up to or above the highest pressures allowed (as determined by the code safety valves).
Following an ECC injection signal the charging pumps become part of the ECC system.
1he high pressure injec? ton (HPI) system has a capability to inject fluid up to pressures near the low pressure scram setpoint.
In some plants the charging and hiyn pressure injection functions are performed by the same pumps.
The low pressure injection (LPI) system is capable of injection only if the primary system pressure is below approximately 1.3 MPa (189 psia).
The charging, HPI, and LPI systems are capable of continuous injection.
Initially, suction is taken from the refueling water storage tank (RWST) and, upon depletion of the tank level, suction may be switched to the containment sump.
The accumulators, containing a fixed amount of water, inject as a function of primary system pressure when the primary pressure falls below
^'
/
approximately 4.14 MPa (600 psia).
t V
Depending on the size of break in the primary system, one or more of these systems is used to prevent an uncovering of the core and subsequent core damage.
For a circular break with a diameter greater than about 6 cm (2.4 in.1 the primary pressure declines below the accumulator setpoint and accumulator injection occurs.
For smaller breaks, however, the removal of energy 3t the break is insufficient to allow a depressurization to the accumulator injection setpoint.
For breaks smaller than about 6 cm (2.4 in.) diameter the performance of the charging and HPI systems is crucial to the safety of the plant.
If both of these systems fail, there would be no way to replenish primary fluid lost through the break rince the primary system pressure would remain above the effective range of the accumulators and LPI. While sucn a n
LJ 1
/
sequence of failures is unlikely, it is not impossibic.
In.1.t.i t t f on t o the possibility of independent mechanical or electrical failures within the systems, there may be possible common failure modes within and between the systems.
Because the potential for core damage and radiation release exists for this multi-failure sequence, potential opportunities for accident mitigation warrant investigation.
The mechanism for accident mitigation which offers the most promise is a depressurization of the primary system to pressures where the accumulators and LPI systems may perform the functions of replacing primary fluid lost through the break and removal of the core decay heat.
Existing PWR equipment likely to provide a depressurizing mechanism include the atmospheric dump valve (ADV) located on the steam generator secondary and the power operated relief valve (PORV) located on the pressuri zer.
The opening of an ADV will depressurize and cool the secondary system which will in turn depressurize and cool the primary system if the two systems remain well coupled. The systems will not be well coupled if the inside or outside of the steam generator tubes are dry. The opening of a PORV will directly depressurize the primary by effectively enlarging the primary system break size.
This report investigates the effectiveness of using ADVs or PORVs to mitigate a small break accident with failures of charging and HPI systems for the four-loop Westinghouse Zion I PWR, The investigation was perfomed by analyzing the results of RELAP4/ MOD 7 couiputer code calculations simulating the plant response to accident conditions.
Section 2 describes the technical approach used including the selection of calculations to be performed and detailed descriptions of the model and conditions.
Section 3 presents the results of the calculations and Section 4 presents the conclusions of the study.
O C
2.
TECHNICAL APPROACH The following sections detail the calculations perfomed, and the computer code, model, and conditions used.
2.1 Calculations Perfomed The Westinghouse Zion I PWR was modeled using the RELAP4/ MOD 7 computer code.
Details of the model and code appear in subsequent sections.
In determining the specific calculations to be performed the following considerations were made:
(a) a range of break sizes should be investigated, (b) definition of the minimum mitigation techniques effective in preventing core damage (for exaniple, define the fewest number of open I
ADVs required to prevent core damage), (c) definition of the times at which operator intervention would be effective, especially the latest time (it was assumed the minimum time at which the operator would intervene was at 10 minutes after the time of the break).
IOU To fully address these considerations would require a large number of calculations.
Therefore, a limited number of calculations were selected which would:
(a) provide an indication of the effectiveness or 4
non-effectiveness of using the ADYs and/or PORVs to mitigate the accident and (b) if effectiveness is indicated, provide a basis for the efficient selection of additional calculations.
I A summary of the four calculations performed is shown in Table 1.
The break size and operator action used for Run 1 were selected to represent the largest break which will not allow a prinary depressurization to the accumulator setpoint and the minimum possible operator intervention.
Run 1 was perfomed and the results analvzed before the specifics for Run 2 were set.
Thus, the specifics of each run were detemined only after an' analysis of the preceding runs.
The logic of the selection process for Runs 2, 3, and 4 appears in detail in the results section.
o I
TABLE 1.
SUMfMRY OF CALCULATIONS PERFORED Time of Break Number ADV Number Time of Diameter ADVs Opening PORVs PORY Opening Run cm (in.)
Opened (Min)
Opened (Min) 1 5.08 (2.0) 1 10 0
2 5.08 (2.0) 1 10 2
25 3
5.08 (2.0) 4 10 0
4 2.54 (1.0) 4 10 0
i O
l l
l k
e O
4
2.2 Computer Co_de
~
The calculations identified in Section 2.1 were performed using the RELAP4/ MOD 7 computer code (Version 100, stored at the INEL under Configuration Control Number H01344IB) with the correction updates shown in Figure 1.
2.3 Model and Conditions The RELAP4/ MOD 7 nodalization for the Westinghouse Zion I PWR is shown in Figure 2.
The plant was modeled using 44 volumes, 65 junctions, and 22 heat slabs.
Two primary coolant loops were modeled thus providing the capability of modeling the opening of one, three, or four atmospheric dump valves ( ADVs). The loops re.present a single and a triple primary coolant loop with themal-hydraulic characteristics of the latter loop adjusted to simulate the flow through three loops of the PWR.
Input decks for these calculations are stored under Configuration Control Number F00138 at INEL.
The primary coolant break, accumulator injection, and low pressure V
injection (LPI) were modeled at the lower plenum.
The RELAP4/ MOD 7 code is not capable of adequately calculating the phenomena associated with the injection of cold accumulator or LPI fluid into a partially steam-filled volume.
For these conditions the code calculates an atypical depressurization and, since depressurization phenomena are of primary concern to the study baing performed, this code limitation was of great importance. By moving the injection, location to the water filled lower plenum, the effects of this code limitation are minimized.
The break location was also moved to the lower plenum since, for a cold leg break, the injection and break locations coincide. Thus, the response of the
~
model with a lower plenum break and injection should closely correspond to a cold leg break.
I fL' r
l 5
i ID INFIXXXJT THIS UPDATE CORRECTS AH ERROR IF 3H + 1 ENTRIES ARE ENTERED
/
OH A FILL TABLE. THE ERROR CAUSED ONE GARBAGE HUMBER IH
/
THE. FILL TABLE I
l I INFIG99HH.65 IF (IUNITS.LT. 3) GO TO 617
'ID HTS 2---SB THIS UPDATE CORRECTS AN ERROR IHADUERTANTLY IHCLUDED IN H007.
~/
THE UPDATE IS HECESSARY IF THE YOID FRACTION IN A UOLUHE
/
/
HITH A HEAT SLAB EXCEDES B.92.
_D HTS 2G99CH.1 DATA ACRIT /8.96E8/
ID INTB---SB THIS UPDATE CORRECTS AN ERROR IN CALCULATING DEAD END UOLUHE
/
/
PRESSURES DURING SELF-INITIALIZAT10H.
D INTBG99KS.4
- HJSC,IJUND,HDEAD,HJD,FIRST1,LAST1) m
/
ERRORS HAVE BEEN FOUND IN THE FILL TABLE INPUT ROUTIHES HHEH
/
/
SI UNITS ARE USED.
CORRECTIONS HAVE HOT YET BEEN FORHULATED.
i
/
THIS UPDATE CORRECTS AH ERROR IN THE REACTIVITY CALCULATICH.
l
/
H0 ERROR EXISTED ONLY IF THE CORE SECTIONS BEGAH t
/
/
HITH THE FIRST SLAB AND HERE HUMBERED CONSECUTIUELY.
THIS UPDATE ss.NECESSARY ONLY IF REACTIVITY CALCULATIDH
/
l
/
IS HAHTED l
ID RINI-2-SB l
I RINI.17 II = 0 D RINI.26,29 l
II = II + 1 i
RDCAL = RDCAL + DOPWT(II)*POLATE(DOPRO,TH(K),HDOP,IDOP(II))
RHCAL = RWCAL + ALPHTH(II)*TEMEFF(J) l RFCAL = RFCAL + ALPHTM(II)*TH(K)
Figure 1.
Computer code updates.
G O
RUCAL = RUCAL + 001DWT(II)*POLATE(UDIDRO 1.,HUDID,IVOID(II)
- ID REAC-2-SB
- I REAC.251 C
II IS THE TABLE LOCATION FOR DOPWT,ALPHTW,ALPHTM,UOIDWT II = 0
- I REAC.254 II = II + 1
- D REAC.255,257 RD = RD + DOPWT(II)*POLATE(DOPRO,TM(I),HDOP,IDOPCII))
RF = RF + ALPHTM(II)*TM(I)
- D REACG68JT.4 RU = RU + UOIDWT(II)*POLATE(VOIDRO,u010,HUDID,IUOID(II))
- /
THIS UPDATE CORRECTS A PROBLEM HITH TOO MUCH PRINT WHEN LIQUI
- /
HASS DEPLETIONS DCCUR.
AFFECTS PRIHT OHLY WHEH ' HEW" WATER
- /
PACKING IS USED (W2-R CARD 839004.LT. INFINITY).
- ID HIFT---DS
~
- I HIFTG94DS.383 PHASE (1,I) = 1
- /
THIS UPDATE CORRECTS AN ERROR IN TYPE 21 CHECK UALUES IF USED
- /
A LEAK JUNCTION AND THE LEAK TABLE HUMBER WAS GREATER THAH 1.
- ID CHKU---SB
- D CHKUG9eSB.7 IF(H.GT. 9) XSIHK = SIHK(H)
- ID PULL---SB
- /
THIS UPDATE CORRECTS A RESTART ERROR IF REACTOR KIHETICS IS
- /
USED.
THE ERROR INCORRECTLY EXPECTED A POWER US. TIME TO BE
- /
READ IN OH TAPE 2 IF HODEL (POWER TYPE) WAS EQUAL TO 2 OR 3
- /
AHD THE CALCULATED HORMALIZED HEUTROH FLUX WAS LESS THAN 1E-8
- D PULL.291 HDDELX = 8 IF(HODEL.EQ. -1) NODELX = 1 IH2 = MAX 9(IH2,HODELX)
Figure 1.
Cont':ued.
l k,Cc:le Safety Valve Double PORV 63 STGEN (to cont 41rment)H
- Outlet, 42 ATH 64 43 44 STGEN SiGEN ATM a,
DumpE Safety i50 STGEN Outlet "
Dump i
Relief 41 Relief 47 46 E 65 25 39 26 It 12 Motor 27 8
15_.. 10
_Ji 17 Motor 16 Drive i 26 Dri Aux
\\
- 51. [3 11 /
1
{
}16 9//
61 Feed
.L Aux feed 60' \\ l3 7
14)_
6 3
Main 39 15 18 Main feed 42 l_
39 tinner Hemi e
7 1
i2 Y-*__
15 J38 41 13"_,
45 feed 6
2 27 34 19 i
17 7
28 38 12 2 'N 18 ll S Q
-*2 Upper Plenum [
~~~
35
/ 11 29 i 33 43 16 r_
SPFdY 4ater 8
6 33 10 5'
Concrete H62 6)
__ T 32 4
11 9 (*
35 40 4
?
fore 28 Fan L
2mtainment Cooler 20 3 >-
i3I 9
- System k
b 59 Leakage 2
31 22 24 1 h 5;NGLE LOOP J
30
.3 23 Break
>l 37 21 i37 1
22 T 20 Volumes 34 36
(
5 Junctions LT*F Plenun l
19 24 25 3
Heat Slabs 23 21 K LPIS ATH Atmosphere (0utside containment 22 Volume 44)
N Figure 2.
RELAP4/ MOD 7 nodalization for the Westinghouse Zion 1 PWR.
O O
O
.The initial plant conditions, shown in Table 2, represent full power plant operation.
Boundary conditions for the calculations are shown in Table 3 and represent a best estimate of conditions expected during the I'
accident in the _ prototype plant.
j t
r.
2.
l 4
i i
i j
e h
i e
i i
I i
l
]
l I
f i
'I i
i
. i I
I t
f i
i' i
s f
l' i
i i
l l
i l
l l
I t
t l
9 i
, _ _. _ ~. _ _ _ _. _.,....
J
TABLE 2.
INITIAL PLANT CONDITIONS O
Core Power:
3238 MWt Hot Leg Temperature:
582 K (589'F)
Cnid Leg Temperature:
550 K (530*F)
S.G. Secondary Pressure:
5.19 MPa (753 psia)
~
Upper Plenum Pressure:
15.6 MPa (2262 psia)
Mass Flow Rate Per Loop:
4606 kg/s (10155 lb/s) i Feed / Steam Flow Rate Per Loop:
444.7 kg/s (980 lb/s)
Feedwater Enthalpy:
1016.5 (406 Btu /1b)
Pressurizer Liquid Mass:
19688 kg (43400 lb)
Pump Speed:
123.6 rad /s (1180 rpm)
S.G. Secondary Liquid Mass Per Loop:
39975 kg (88130 lb)
O O
9 l
0; l
10
i TABLE 3.
BOUNDARY CONDITIONS V
A.
Nonnalized core axial power profile Top 0.0397 0.1791 0.2122 0.2139 0.2101 0.1450.
i Bottom B.
Decay heat calculated by the new ANS standard plus a contribution from actinides C.
Low pressure injection volume flow was dependent on pressure (2 pump outpu t). LPI fluid temperature was:
310 K (100*F)
Primary Pressure Injection Rate Per Loop l
MPa (psia)
Liter /s (gal /mi n)
Less than 1.02 (148) 141.9 (2250) 1.10 (160) 126.2 (2000) 1.16 (168) 110.4 (1750) 1.21 (176) 94.6 (1500)
/
1.25 (181) 78.9 (1250) 1.29 (187) 47.3 (750)
More than 1.30 (189) 0 D.
Secondary relief valve mass flow was dependent on pressure Secondary Mass Flow Rate Per S.G.
Pressure MPa (psia) kg/s (lb/s)
Less than 7.135 (1035) 0 t
8.204 (1190) 112.1 (247.1) 8.238 (1195) 112.1 (247.1) 8.245 (1196) 154,2 (340.0)
More than 8.617 (1250) 635.0 (1400.0)
E.
A full flow area of 0.01015 m2 (0.1092 2
ft ) was used for each atmospheric dump valve.
This was detennined to be the area required to pass 112.1 kg/s (247 lb/s) of saturated steam from 7.72 MPa (1120 psia) to 0.1 MPa (14.7 psia) based on the valve rated conditions.
l l
I O
11 i
F.
A full flow area of 0.001057 m2 (0.0114 2
ft ) was used for each power operated relief valve.
This was determined to be the area required to pass 26.46 kg/s (58.3 lb/s) of saturated steam from 16.2 MPa (2350 psia) to 0.1 Wa (14.7 psia) based on the valve rated i
conditions.
G.
Motor-driven auxiliary feedwater flow was controlled to be fully off when the secondary level exceeded 13.1 m (43 f t) and fully on when the level was lower providing a flow rate of 14.08 kg/s (31.04 lb/s) to each steam generator.
Turbine-driven auxiliary feedwater was not used in the calculations as it would not be available during periods when the secondaries are being depressurized.
H.
Accumulator data
~
Pressure:
4.136 MPa (600 psia) 325 K (125"F) 3) per accumulator.
Temperature:
23.2 m3 (819 ft Water Volume:
I.
A primary leakage rate of 0.63081/s (10 gpm) was used throughout the calculation.
This rate represents the technical specification limit on identified primary leakage.
J.
The safety injection (SI) signal was generated based on the pressurizer pressure falling helow 12.62 MPa (1830 psia).
The charging and high pressure injection flows would nonnally be expected to start 5 seconds after the signal but are assumed to fail in the calculation s.
K.
Reactor scram was initiated upon pressurizer pressure falling below 12.82 MPa (1860 psia).
Ikactor coolant pump power was tripped off at the time of scram.
The steam generator outlet valve was closed linearly between 1.0 and 4.0 seconds after scram.
The main feedwater flow was ramped off linearly between 5.0 and 10.0 seconds after scram. Auxiliary feedwater flow was initiated 60 seconds after scram.
L.
Containment fan coolers were initiated if and when the containment pressure reached 0.1323 MPa (19.2 psia).
The fan cooler capacity was 71 MW.
Containment spray cooling was initiated when the congainment pressure M.
reached 0.2599 MPa (37.7 psia) and delivered 0.491 m /s (7782 gpm) of water at 293 K (68*F) to the containment.
9 O
12
3.
RESULTS The results and analyses of the calculations performed are presented in this section.
4 3.1 Basic Sequence Description This section describes the basic accident sequence following a small break and before operator intervention.
Only automatic safety system operation is considered and, since the charging and high pressure injection systems fail, the sequence leads to core damage unless the operator intervenes. The results of the calculations following the time of operator intervention are emphasized in the remaining part of Section 3.
At zero time a 5.08 (2 in. ) or 2.54 cm (1 in.) diameter' break occurred in the primary system.
The primary pressure declined causing a scram'at 56.8 seconds' for the larger break and 181 seconds for the smaller break.
In both cases a safety injection signal was generated but the high pressure
. injection and charging systems fail to deliver flow to the primary.
A turbine trip signal was generated and the secondary systems were isolated by a terminatien of main steam and feedwater flows and an initiation of i
l l
Because the break size was not large enough to remove all the core I
decay heat, some heat removal was required from the primary to the secondary system. This caused a pressurization of the secondaries to the j
relief valve setpoint and arrested the primary depressurization such that j
the heat ' removal rate to the secondaries equaled the difference between the.
e core decay heat rate and the break energy removal rate.
The primary pressure was stabilized well above the pressures at which either accumulator or low pressure injection would occur.
1 ce l'
l i
i O
u 13
1 For economic reasons the calculations with no operator intervention were not carried out to the time of core uncovering.
It was estimated, however, that core uncovery will occur at approximately 30 minutes for the 5.08 cm (2 in.) diameter break and 2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br /> for the 2.54 cm (1 in.) diameter break.
The time for effective operator action is rather short.
It would be advantageous for the operator to know the size of the break since the urgency of his actions will be greatly influenced by it.
One pos:ible means of estimating the break size would be to observe on a strip chart the time required for the primary pressure to decline to the scram setpoint and correlate an effective break size with that time interval.
Also, because of the short time available for operator action, no attempt should be made to use the LPI pumps in the residual heat removal mode.
The pumps should be aligned for the injection mode only.
3.2 Run 1, 5.08 cm (2 in. ) Diameter Break One ADV Opened st 10 Minutes The themal hydraulic behavior of the plant prior to 10 minutes after the break is described in detail in Section 3.1.
At 600 seconds the atmospheric dump valve (ADV) on one steam generator was opened.
Figure 3 shows the primary and secondary pressure responses which are available for observation by the operator.
The secondary with the open ADV depressurized rapidly until about 750 seconds.
This pressure plateau, lasting until about l'.00 seconds, represented an equilibrium condition where the single steam generator was being used to cool both the primary system and the secondaries with closed ADVs.
Note at 800 seconds the primary pressure decreased below the pressure of the secondaries with closed ADVs.
/.fter 1100 seconds the primary sjstem was effectively decoupled from the se::ondaries with closed ADVs when the loops with those steam generators dried out. This dryout was indicated by a rapidly 9
l 14
i O
i
' ~
?
20000 g
g l
l l
4 Primary Secondaries i
with closed i -
10000 ADVs LPI j
initiation i
' pressure a.
- 5000, Secondary with j
open ADV N
i l
Prima ry l
i l
I l
0 i
t 0
500 1000 1500 2000 i
1 l
Ttse tal j
i I
Figure 3.
Run 1 primary and secondary pressures.
l l
l l
i f
I t
i i
?
I I
15 I
)
increasing hot leg quality, which is not available fo' operator observation, as shown in Figure 4.
The decoupling of the three isolated secondaries allowed an increased depressurization rate for the primary and the secondary with the open ADY.
The accumulator injection pressure was reached at 1485 seconds and the accumulators injected at an increasing rate as the primary depressurization accelerated.
The accumulator level, observable by the operator, is shown in Figure 5.
The depressurization of the ' primary appeared to be proceeding toward an initiation of low pressure injection.
An expanded scale projection of primary and secondary pressures is shown in Figure 6.
The primary depressurization had been stopped and the primary was actually repressurizing following 1755 seconds.
The repressurization was caused by a recoupling of the isolated secondaries to the primary.
The recoupling v:as caused by the injection of accumulator fluid refilling the primary loops to those steam generators.
This refiliing was indicated by the hot leg quality decrease shown in Figure 4.
The vessel mixture level is shown in Figure 7.
Momentary core uncoverings were calculated during the period of accumulator injection after 1458 seconds.
These uncoverings, caused by hydraulic oscillations during the injection of subcooled accumulator water, were not of sufficient duration to cause any significant core heatup.
At the end of the calculation (1762 seconds) the following conditions existed:
(a) ~,1e primary system was repressurizing away from the LPI initiation pressure, and (b) the accumulators were nearly empty, and (c) the vessel level was only slightly above the core.
By observing the trend in the primary and secondary pressures it was concluded that a depressuriza. ion of the primary to the LPI initiation pressure will not occur before a significant core uncovery occurs.
O 16
l l
l I
j i
O v
.l 0.8 f
0.e N
l 0.4 I
0.2 I
I 0.0 O
500 1000 1500 2000 Time (s)
Figure 4.
Run i hot leg quality in loops with closed ADVs.
Norma level p
3 e*
2 2"
i 1
Bottom of accumulator 0
500 1000 1500 2000 T!as (s 3' t
Figure 5.
Run 1 acct.nulator w. iter level.
17
8000 g
g g
Secondaries 6000 with closed ADVs 4000 Upper plenum
~
h
(
2000 with open ADV LPI initiation pressure i
u
~
i l
l 1650 1675 1700 1725 1750 1775 Tlas (s)
Figure 6.
Run 1 expanded scale primary and see.ondary pressures.
O I0 l
i I
I I
Top of upper plenum 1
8 o
I E
- - g.
_ yyhk 4
E Top of core 2
j Bottom of core I
/
I I
g O
500 1680 1500 2000 Time (s)
Figure 7.
Run 1 vessel mixture level.
18
The primary and secondary fluid temperatures, shown in Figure 8, basically followed the saturation temperatures associated with the V
pressures shown in Figure 3.
The pressurizer level shown in Figure 9 declined until the pressurizer emptied at about 100 seconds, was reestablished as the primary fluid temperature increased from 120 to 400 seconds, then was lost for the remainder of the transient. The lack of a pressurizer level will indicate to the operator that a core uncovery is possible.
It is important to note, however, that core uncovery did not occur until about 20 minutes following the loss of the pressurizer level indication. The secondary level of the steam generators with closed ADVs returned to normal following auxiliary feedwater initiation as shown in Figure 10. The level of the secondary with the open ADV declined steadily following the opening of the ADY.
The auxiliary feedwater flow rate to this secondary was arbitrarily tripled at 1722 seconds to prevent drying out the secondary.
The operator would be expected to perform similar action to protect the steam generator.
In conclusion, the technique of opening a single atmospheric dump valve at 10 minutes has been shown not effective in preventing core v
uncovery following a 5.08 cm (2 in.) diameter break in the primary system.
The reason for ineffectiveness is that the three steam generator secondaries with closed ADVs contain a large volume of water at a temperature far above the saturation temperature required in the primary system for LPI operation, lhese secondaries act as heat sources to the primary which inhibit primary depressurization.
3.3 Run 2, 5.08 cm (2 in.) Diameter Break, One ADV Opened at 10 Minutes and 2 PORVs Opened at 25 Minutes The thermal hydraulic behavior of the plant up to 25 minutes after the break was the same as that described in detail for Run 1 in Section 3.2.
r P
. V 19
600 g
F Upper plenum
[
Secondaries
/with closed
~
f g ADVs t
550 Secondary /
Upper I
g plenum with open ADV 3
500 E
I I
I 450 0
500 1000 1500 2000 j
flae ts)
Figure 8.
Run 1 upper plenum and secondary fluid temperatures.
l l
15
- Normal level F
10 3
2 2*
5 B
l
/ ottom of pressurizer l
I' I
0 0
500 1000 1500 2000 Tlae (s)
Figure 9.
Run 1 pressurizer mixture lavel.
20 i
15 g
g g
g econdaries isolated S
Secondaries
~
with closed
\\uxiliary ADVs 10 A
feedwater initiated Secondary I
/withopen g
ADV 5
Auxiliary feedwater rate tripled h
i I
0 O
500 1000 1500 2000 Time to)
Figure 10. Run 1 secondary mixture level.
A a
f 21
This calculation was perfomed to determine the effect on the primary depressurization of opening both PORVs.
Run 1 was restarted at 1500 seconds (time of accumulator flow initiation) with both POWS open.
The calculation was terminated at 1580 seconds for economic reasons.
Figure 11 shows the primary and secondary pressure responses.
Upon opening the PORVs at 1500 s the primary depressurization rate increased significantly from about 5 KPa/s (0.7'3 psi /s) to 17 KPa/s (2.47 psi /s). A similar increase occurred in the primary and secondary temperature cooldown rates as shown in Figure 12.
The accumulator level, shown in Figure 13, was rapidly falling and l
depletion of the accumulator was imminent.
l The infomation presented in Figure 11,12, and 13 would be available l
for operator observation.
It is estimated the increased cooling of the primary provided by the open PORVs will allow the primary pressure to be reduced sufficiently to allow LPI operation at about 1700 seconds.
Since the accumulators will be depleted at that time, a primary repressurization above the LPI shutoff head would result in a termination of all emergency core coolant delivery.
The mechanism for such a repressurization is the same as was calculated in Run 1, which was the recoupling of the hot steam generator secondaries with the closed ADVs to the primary system when the loops were refilled with water.
It is likely the system will be repressurized for a sufficient l
period of time to cause a core uncovering and subsequent damage.
These estimates were made after studying the repressurization phenomena calculated at the end of Run 1.
The net heat addition rate from the four secondaries was about 425 MW at that time the core power was approximately 60 MW, thus a minimum of 485 MW must be removed from the primary to avoid repressurization.
At the I.PI shutoff head pressure, the break in addition O
22
20000 g
15000.
S.i 10000 Secondaries j
h closed POPVs
\\
opened g
5000 Secc4dsry/
with open Primary ADV 0
O 500 1000 1500 2000 Time ts)
Figure 11. Run 2 primary and secondary pressuru.
'O 600 F
Secondaries s
nd
~
550 4
l Secondary /
with open ADV Primary l
3 500 l
2 l
i l
I I
I 450 l
0 500 1000 1500 2000 Time ts)
I Figure 12..
Run 2 upper plenum and secondary fluid temperatures.
l 1
23
O, i
S 3.5 I
e i
i 3.0 l
2 a
2 3>.
2.5 f
4 I
I I
2.0 O
500 1000 1500 2000 P
Tlae (s)
Figure 13. Run 2 accumulator water level.
I f.
c l
9 I
l 24 i
i
to two P0DVs are only capable of removing steam at a rate that removes
)
about 25 MW from the primary system. A net primary heat addition rate of about 460 MW is expected to repressurize the primary above the LPI shutoff head.
In conclusion, the opening of both PORVs at 25 minutes was detennined to greatly improve the primary depressurization rate calculated following the opening of a single ADY at 10 minutes.
The open PORVs did not, however, create a large enough steam relief capability to prevent a repressurization of the primary system above the LPI shutoff pressure. The opening of a single ADV followed by an opening of both PORVs was therefore not effective in preventing core uncovering and subsequent damage following a 5.08 cm (2 in.) diameter break in the primary system.
3.4 Run 3, 5.08 cm (2 in.) Diameter Break, Four ADYs Opened at 10 Minutes Based or the experience gained from performing Runs 1 and 2 for O
opening a single ADV, it was decided that all further runs should consider the opening of all four ADVs at the same time.
The thennal hydraulic behavior of the plant prior to 10 minutes is described in detail in Section 3.1.
This section concentrates on the behavior following the opening of all atmospheric dump valves (ADYs).
Because of the rapid depressurization rates of the primary and l
secondary systems calculational difficulties were encountered.
From-the time the ADVs were opened (600 seconds) until initiation of accumulator injection (796 seconds) the thermal hydraulic behavior of the two reactor coolant loops exhibited scme asymmetric irregularities. The irregularities, believed to be caused by pressurizer effects, were indicated by minor differences in secondary level, temperature, and l
pressure during that period.
The effects of these differences are not i
i (v) 25
significant to the overall results of the calculation and their detailed presentation would complicate the analysis of significant effects.
Therefore the secondary parameters presented in the following figures represent the average conditions calculated for all secondaries.
A more serious calculational difficulty was encountered at 870 seconds when, for economic reasons, the calculation was terminated.
Because of the rapid primary depressurization rate, injection of cold accumulator fluid occurred at such a rate that the calculation progress was slowed beyond the limit of economic feasibility.
Rather than attempt to atypically restrict or teminate accumulator injection, the calculation was stopped since the overall conclusions from the calculation have been ascertained without completing it.
The primary and secondary pressure responses are shown in Figure 14.
The primary pressure fell as a result of the secondary depressurization and cooldown following the opening of the ADVs at 600 seconds.
Accumul ator injection began at 796 seconds as indicated by the decreasing accumulator water level shown in Figure 15. The upper plenum and secondary fluid temperatures, shown in Figure 16, indicate both temperatures were near saturation and decreased as the pressures decreased.
The secondary level response is shown in Figure 17.
The level decreased rapidly following the opening of the four ADYs. A pressurizer level was not reestablished following the initiation of accumulator injection, as shown in Figure 18.
The infomation presented in Figures 14 through 18 would be available for observation by the operator.
The vessel level, shown in Figure 19, would not be observable.
Following the opening of the ADYs the vessel level remained above the core until the end cf the calculation.
This dip in vessel level was not the beginning or an extended core uncovery but rather the result of a momentary shift in primary fluid in response to the l
injection of cold accumulator fluid. The phenomena was the same as that I
shown in Figure 7 for Run 1.
i 1
26
1 20000 I
I I
I V
15000 e
i N
{
Primary All ADVs j10000
/
opened I
N 1
E Accumulator i
{ injection 5000 Secondary LPI I
I I
I 0
O 200 400 600 000 1000 l
?
T!se ts) l t
Figure 14. Run 3 primary and secondary pressures.
i N
I i
4 I
l l
1 y onnal level i
h i
3 l
I 2
i
~
g 2
=
t i
1 l'"
f Accumulator empty l
i I
I I
O O
200 400 630 800 1000
' h i
Tlas to) i t
Figure 15. Run 3 accumulator water level.
~
k a
I I
l l
Primary O
550 N
Secondary 1*
e 3
500 C
l I
I I
450 O
200 400 609 800 1000 T!as (e)
Figure 16. Run 3 upper plenum and secondary fluid temperatures.
9 I4 I
I I
I i
(
Normal level Four r-ADVs opened
/
12 2
i t
i a
i i;i Auxiliary 10
~
feedwater initiated S
i I
I 1
g
[
0 200 400 600 000 1000 g
ri..
l Figure 17. Run 3 secondary mixture level.
t 28
\\
q
,/
+4.//
a*
...e... <e 1,.
TEST TARGET (MT-3) l.0 5 E8 BM y l8 EE a
I.1 \\] '8 IE
.)
~
~
/
j _g I.25 1.4 1.6 E-7 4+//%
++4
'4 4
4fA)f p,,,fff 4$,ga
.i --
-.m
..ww.-
M%
$+4) o W
IMAGE EVALUATION TEST TARGET (MT-3) l.0 gasugg y ll IlE i I,I %,i '8 Ill&&
/
.8 1['l.25--
i_
l IA i1.6
,/
jc 6"
=
i b
5%
4 ///h
- }jp4,h Af,,,&/
3 t
w
- w-h we...A r*.
.4<.
,..e....<e.-
TEST TARGET (MT-3) l I.0 d'EME34 Ena' ilE m
l,l5,4 12 l g
.=
l11 i.25 i.4 g
/
c gn y
Ar,,,#
- 'hy@
O j,
3 i
l
15 j
g g
g g
Normal level l
b 10
~
. s
~
r i
E 5
Accumulator injection i
Pressurizer initiated empty 4
l
[
l I
(
0 0
20e 400 600 808 1,000 Time to)
Figure 18. Run 3 pressurizer mixture level.
.I 10 g
g g
y i
g Top of upper plenum 6
g A%
4 i
E Topof/
core I
2
, s Bottom O
f i
l
~
0 200 400 600 800 1000 l
Tlas (e3
[
i s
Figure 19. Run 3 vessel mixture level.
i l
h Y
- --- - - --- - - "-- " ~ ~ I
Analysis of the calculation results indicates that opening four ADVs was an effective technique for depressurizing the primary sufficiently to allow continued LPI operation.
This conclusion is supported by comparing the results of Run 3 with Run 1 in which a single ADY was opened.
If the repressurization caused by the recoupling of the secondaries with closed ADVs had not occurred, Run 1 would have indicated the technique was effective.
The results of Run 3 were similar except the mechanism for a later repressurization was not present and, because the depressurization was more rapid, less primary mass was lost through the break before the LPI initiation pressure was reached.
It is estimated that the LPI initiation pressure would be reached at about 1200 seconds.
In conclusion, the opening of four ADVs at 10 minutes following a 5.08 cm (2 in.) diameter primary break has been shown to be an effective technique for depressurizing the primary system so that continued LPI operation may prevent core uncovery and damage.
3.5 Run 4, 2.54 cm (1 in. ) Diameter Break, Four ADVs Opened at 10 Minutes The thermal hydraulic behavior of the plant prior to 10 minutes after the break was described in detail in Section 3.1.
This section concentrates on the behavior following the opening of four atmospheric dump valves (ADVs) at 600 seconds.
The primary and secondary pressure responses are shown in Figure 20.
The primary pressure fell rapidly as a result of the secondary depressurization and cooldown.
The accumulator injection began at 945 seconds causing the primary pressure to fall more rapidly until 1005 seconds when the primary system (except for the pressurizer and upper head) had been refilled with water.
Between 945 and 1005 seconds the accumulator level fell sharply as shown in Figure 21 and the upper plenum 9
30
20000 N
y A
Primary i
2 ADVs
~
' ~
d opened k
j 10000 3
2 Accumulator flow initiated
[
Accumulators 5000 Secondary empty LPIS initiatec r
i i
i t
0 0
500 1000 1500 2000 2500 Time to)
Figure 20. Run 4 primary and secondary pressures.
I I
I I
3
- Nonnal level
_3 a
f 2
5 1
1 Accumulator l
I I
I O
O
' 500 1000 1500 2000 2500 Tlas to)
Figure 21. Run 4 accumulator water level.
31 l
level, which had been falling, increased quickly to the top of the upper plenum as shown in Figure 22.
Note the upper plenum level would not be available for operator observation.
All other parameters presented would be observable.
Following 1005 seconds the primary depressurization rate slowed because there was a limited amount of steam available for condensation in the flowing portion of the primary system. The accumulator discharge rate
~
also slowed markedly, as shown in Figure 21.
The accumulator injection rate exceeded the break flow rate and a pressurizer level was reestablished at 1530 seconds as indicated in Figure 23.
At 1755 seconds the accumulators were emptied.
The rise in pressurizer level at that time was caused by a reheating and expansion of the primary fluid, as shown in Figure 24, upon tennination of cold accumulator fluid injection. The reheating continued until about 1975 seconds when a larger primary to s?condary fluid temperature differential had substantially increased the heat removal rate to the secondari es.
This effect was also indicated by the secondary level as shown in Figure 25.
Upon initiation of accumulator injection the secondary level stopped falling and after 1975 seconds the rate of level increase was slowed.
After 1974 seconds the primary was cooled by the secondaries and at 2160 seconds the primary pressure had decreased to the LPI initiation pressure.
The calculation was terminated at 2266 seconds with the LPI flowing, the entire primary system (except for the pressurizer) filled with subcooled water, and a rapidly increasing pressurizer level.
Upon refill of the pressurizer the primary pressure will rapidly increase and stabilize at a value where the injected LPI mass flow rate equals the break mass flow l
rate.
The primary heat removal rate from LPI injection and break energy removal will be much less than the decay heat.
The opening of the PORVs l
will not greatly aid this condition.
It will be important, therefore, to l
maintain the secondary heat sink by keeping the ADYs open and continuing auxiliary feedwater delivery.
O' 32
I t
i I
. O NTop of upper plenum 3
?
~
a 3
2 A
Accumualtor
^
flow t
initiated Top of core i
I I
I g
O 500 1,000 1500 2000 2500 ftse to)
Figure 22. Run 4 upper plenum mi- +ure level.
15 g
g g Nomal level 10
_a
~
LPI initiated 1
2 i
5 v
i Pressurizer f
empty
\\,
g-Accumula, tors empty N
g 0
500 1000 1500 2000 2500 Time to)
Figure 23. Run 4 pressurizer mixture level.
33
600 f pper' plenum U
Accumulator flow initiated 550 3
{
Secondary Accumualtors empty
{
500 Four ADVs y
initiated opened 3
o E
450 I
I I
I 400 O
500 1000 1500 2000 2500 TIse ts)
Figure 24. Run 4 upper plenum and secondary fluid temperatures.
O:
1
.0
,ADVs I
I I
1 opened f
12 5 E
i i
=I 10.0 j
Accumulator initiation d
Accumulators empty 7.5 5.0 I
I I
I O
500 1000 1500 2000 2500 Time ts) l Figure 25. Run 4 secondary mixture level.
j i
34 i
0 At the time of pressurizer refill there will be a potential for
[
water-hammer effects within the primary and LPI systems.
Since continued operation of the LPI system will be critical for continued core cooling further investigation of this situation is warranted.
i In conclusion,-the opening of all ADVs at 10 minuter following a 2.54 cm-(1 in.) diameter primary break has been shown to be an effective
~
technique for depressurizing the primary system so that continued LPI~
operation may prevent core uncovery and damage.
4 i
i 4
_w 4
f 8
{
[
l i
~
l l
t l
l 35
4.
CONCLUSIONS AND REC 0fHENDATIONS The largest break diameter which will prevent a depressurization to the accumulator injection pressure is about 6 cm (2.4 in.).
For breaks of smaller diameter the failure of charging and high pressure injection sydems will result in core uncovering and damage if the operator does not take corrective action.
For breaks of larger diameter, previous calculations have indicated the primary is depressurized to the accumulator setpoint.
The small break sequence with coincident failure of charging and HPI rapidly leads to a core uncovering but sequence timing is directly affected by break diameter.
For a 5.08 cm (2 in. ) diameter break, core uncovering begins at about 30 minutes.
For a 2.54 cm (1 in.) diameter break it begins at about 2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br />.
These figures indicate an estimate of break size by the operator will be important in detennining the urgency of corrective action.
It may be possible to estimate the break size by observing on a strip chart recorder the interval from the break time to scram time or time of loss of pressurizer level indication.
Regardless of the break size the sequenco develops so rapidly that no attempt should be made to use the residual heat removal mode for the LPI pumps.
The pumps should be aligned for the injection mode.
The opening of a single atmospheric dump valve ( ADV) is not effective in preventing core uncovering.
The steam generators secondaries with closed ADVs remain hot and are decoupled from the primary by significant voiding of the corresponding reactor coolant loops.
The hot secondaries will eventually be recoupled to the primary system when the loops refill from accumu.ator and low pressure 9
36
4 injection.
This recoupli ~7 repressurizes the primary and terminates the l [
')
low pressure injection.
This same repressurization effect would be present if two or three ADYs are opened.
The opening of all four ADVs at 10 Minutes has been shown to be an effective means of preventing core uncovering following 2.54 cm (1 in.) and 5.08 cm (2 in.) diameter breaks.
N1ations perfomed indicate the feasibility of using this correct.,
e technique.
Additional analysis will be required to investigate the latest times at which the technique will be effective and to detemine the effect of only partially opening all ADVs.
These two concerns are important because the operator will delay a decision to take this emergency action as long as possible and, if he uses the ADVs, will want to depressurize the plant as slowly as possible becrJse of thermal stress considerations associated with high plant cooldown rates.
In simplest terms, the break size will detemine the time available for corrective action and, the sooner the ADYs are opened, the more slowly the plant may be depressurized.
Following a successful depressurization of the primay and initiation of Low Pressure Injection (LPI), the primary system will be rapidly refilled with subcooled water.
The possibility and effects of water hammer during refill should be considered.
The LPI system will be the only injection system operating to cool the plant.
Therefore, challenges to the continued operation of the LPI system will be important. The possibility of damaging the LPI system through two j
water hammer effects should be considered.
First, when LPI is initiated there is potential for water hamer in the pipes downstream of the pump.
Second, the primary will quickly refill with subcooled liquid following LPI initiation and there is a pntential for water hamer when the primary completely fills.
O
, V 37
It is recommended that additional work be performed to determine the latest time at which the opening of four ADVs will prevent core uncovering and to determina the slowest secondary depressurization rate which will prevent core uncovering.
9 O
l 1
91 i
38
--