ML20093A344
| ML20093A344 | |
| Person / Time | |
|---|---|
| Site: | Calvert Cliffs  |
| Issue date: | 05/31/1984 |
| From: | LOS ALAMOS NATIONAL LABORATORY, OAK RIDGE NATIONAL LABORATORY, PURDUE UNIV., WEST LAFAYETTE, IN, SCIENCE APPLICATIONS INTERNATIONAL CORP. (FORMERLY |
| To: | NRC |
| Shared Package | |
| ML20093A335 | List: |
| References | |
| REF-GTECI-A-49, REF-GTECI-RV, TASK-A-49, TASK-OR NUDOCS 8407100360 | |
| Download: ML20093A344 (272) | |
Text
r- -
9 h
l Chapter 4. Thermal-Hydraulic Analysis of Potential Overcooling f
l Transients occurring at Calvert Cliffs Unit 1 Nuclear Power Plant of A PRESSURT7FD THERMAL SHOCK EVALUATION OF THE CALVERT CLIFFS UNIT 1 NUCLEAR POWER PLANT written by The PTS Study Group -
of Engineering Physics and Mathematics Oak Ridge National Laboratory, Los Alamos National Laboratory, Purdue University and Science Applications, Inc.
Date of Draft: May 31, 1984 NOTICE: This document contains information of a preliminary nature. It is subject to revision or correction and therefore does not represent a final report.
l l *Research sponsored by U.S. h'uclear Regulatory Commission under Contract No. DE-AC05-840R21400 with the Martin Marietta Energy Systems, Inc.
i 8407100360 840618 PDR ADOCK 05000317 P pon
_J
r A PRESSURIZED THERMAL SHOCK EVALUATION OF THE CALVERT CLIFFS UNIT 1 NUCLEAR POWER PLANT List of Chapters Chapter 1 Introduction Chapter 2 Calvert Cliffs Unit 1 Nuclear Powec Plant System Descriptic-Chapter 3 Development of Overcooling Sequences for Calvert Cliffs Unit 1 Nuclear Power Plant Chapter 4 Thermal-Hydraulic Analysis of Potential Overcooling Transients Occurring at Calvert Cliffs Unit 1 Nuclear Power Plant i
Chapter 5 Probabilistic Fracture-Mechanics Analysis of Calvert Cliffs Unit 1 Sequences Chapter 6 Risk Integration of Potential Calvert Cliffs Unit 1 Overcooling Sequences Chapter 7 Sensitivity and Uncertainty Analysis Chapter 8 Summary and Conclusions i
l 1
4 i
i 4.0. THERMAL-HYDRAULIC ANALYSIS OF POTENTIAL OVERC00 LING TRANSIENTS
' OCCURRING AT CALVERT CLIFFS UNIT 1 NUCLEAR POWER PLANT 4.1. Introduction 4.2. Selection of Twelve Sequences i
4.2.1. Sequences Initiated by Large Steam-line Break at HZP 4.2.2. Sequences Initiated by Small Steam-line Break at HZP 4.2.3. Sequences Initiated by Large Steam-line Break at Full Power 4.2.4. Sequences Initiated by Small Steam-line Break at Fall Power 4.2.5. Sequences Initiated by a Reactor Trip 4.2.6.
4 SequenceIg)itiatedbyaSmall-BreakLOCA (10.016 f t 4.2.7. Sequence {nitiatedbyaSmall-BreakLOCA
( 0.02 ft )
4.2.3. Sequences Initiated by LOCAs with Potential
' Loop Flow Stagnation j 4.2.9. Sequence Initiated by Loss of Main Feedwater with Subsequent AFW Overfeed 4.2.10. Summary l
4.3. LANL TRAC Analysis 4.3.1. TRAC-PF1 Model of Calvert Cliffs Unit 1 4.3.2. Steam-line Break Calculations 4.3.2.1. Transient 1: 0.1-m2 Main Steam-line Break from HZP 4.3.2.2. Transient 2: 0.1-m2 Main Steam-line BreakfromFullPowgr 4.3.2.3. Transient 3: 0.1-m Main Steam-line Break from HZP with Two Operating Reactor Coolant Pumps 4.3.2.4. Trossient 4: Double-Ended Main Steam-line Break from 52P with Failure to Isolate AFW Flow to Broken Steam Line 4
4.3.2.5. Transient 5: Double-Ended Main Steam-line Break Upstream of MSIVs from HZP with Two Stuck-Open MSIVs 4.3.2.6. Transient 6: Small Steam-line Break Downstream of MSIVs from Full Power l
4.3.2.7. Transient 7: Small Steam-line Break Down-l stream of MSIVs with Failure of One MSIV to Close from Full Power
4.3.3. R:n: wry Focdwater Ev:nts 4.3.3.1. Transient 8: Runaway Main Feedwater to 1 Both Steam Generators from Full Power 4.3.3.2. Transient 9: Runaway Main Feedwater to One Steam Generator from Full Power 4.3.3.3. Transient 10: Runaway Auxiliary Feedwater to Two Steam Generators from Full Power 4.3.4. Small-Break LOCA Events 4.3.4.1. Transient 11: 0.002-m2 Hot-Leg Break from Full Power 4.3.4.2. Transient 12: Stuck-Open Pressurizer PORY with Stuck-Open Secondary ADV from Full i Power
]
4.4. Downcouer Fluid Mixing Behavior 4.4.1. Stratification Analysis of Twelve LANL
' Transients 4.4.2. SOLA-PTS Mixing Analysis of Selected Transients 4.4.3. Total Loop Flow Stagnation 4.5. Heat-Transfer Coefficient Evaluation 4.6. Estinations of Pressure, Temperature and Heat-Transfer Coefficient Profiles 4.6.1. Methodology
, 4 . 6.1.1. General Approach I 4.6.1.2. Sequence Grouping
- l 4.6.1.3. Temperature Evaluation by Cooldown Model
/ 4.6.1.4. Pressure Evaluation by Coolant Swell Model l
4.6.2. Results of Simple Model Evaluations
! 4.6.2.1. Large Main Steam-line Breaks at HZP 4.6.2.2. Small Main Steam-line Break at HZP
- 4.6.2.3. Large Main Steam-line Break at Full Power 4.6.2.4. Small Main Steam-line Break at Full Power 4.6.2.5. Reactor Trip Sequences i; ! 4.6.2.6. Small-Break LOCA (10.016 f t2 )
Small-Break LOCA (~0.02 f t2 )
i 4.6.2.7.
i l
I I e
=
_,- ,~_. - . - , - - -
--._v . - _ . _ , , . - , . - . .,%. ,_-, - . _ . - . - ,--
-..-,m. -
_ , . . , , . - , - , - . , . . - ,n- . ,- ,-w..
l l
l Chapter 4 Appendices F.* 1RAC-PF1 Analysis of Potential Pressurized-Thermal-Shock :
Transients at a Combustion-Engineering PWR l G.* Brookhaven National Laboratory Review of the TRAC Analysis I H.* Buoyancy Effects in Overcooling Transients Calculated for the Calvert Cliffs Unit 1 Nuclear Power Station I.* Three-Dimensional Calculations of Transient Fluid-Thermal Mixing in the Downconer of the Calvert Cliffs-1 Plant Using SOLA-PTS J. Estimation of Pressure, Temperature, and Heat Transfer Coefficient
- To be published as a separate report.
I l
1 1
4
_ , - . . . - . . , _ . - . - , - . . - . - - , ...-,,,.c.-- - , . . , . -. .v,.- . ,-,,_-,m. , _ - _ . m.-._--,.. --, . . . - - , , . . - - , - , . , , ,..v.,.,
- 1.IST OF FIGURES l
4 .1. TRAC noding diagram for the primary side at Calvert Cliffs-1 (from Ref. 37) 4.2. TRAC noding diagram for the feedwater train at Calvert Cliffs-1 (from Ref. 37) 4.3. TRAC noding diagram for the steam lines at Calvert Cliffs-1 (from Ref. 37) 4.4. Transient 1: Downconer liquid temperature during 0.12 main steam-line break from HZP 4.5. Transient 1: Primary system pressure during 0.1-m2 main steam-line break from BZP 4.6. Transient 2: Downconer liquid temperature during 0.1-m 2 main steam-line break from full power i
! 2
) 4.7. Transient 2: Primary system pressure during 0.1-m ,
main steam-line break from full power Transient 3: Downcomer liquid temperature during 0.1-m 2
~
4.8.
main steam-line break from EZP with two operating RCPs
~
4.9. Transient 3: Primary system pressure during 0.1-m 2 main steam-line break from HZP with two operating RCPs t
4.10. Transient 4: Downconer liquid 6emperature during double-ended main steam-line break from HZP with failure to isolate AFW flow to broken steam line 4.11. Transient 4: Primary system pressure during double-
! ended main steam-line break from HZP with failure to isolate AFW flow to broken steam line i
4.12. Transient 5: Downconer liquid temperature during double-
! ended main steam-line break from HZP with two stuck-open NSIVs 4.13. Transient 5: Primary system pressure during double-ended main steam-line break from HZP with two stuck-l open NSIVs l
4.14. Tra.tsient 6: Downconer liquid temperature during small steam-line break from full power with stuck-open TBV 4.15. Transient 6: Primary system pressure during small j steam-line break from full power with stuck-open TBV 4.16. Transient 7: Downconer liquid temperature during small steam-line break from full power with one stuck-open
i
, . - - , - - - - , - - , , , , , - - , ,,--._,.w-----r-,,--..., -,--wemw- , , , . . .- - - . - -.-.w.-, - ,r,-,y,er~e-,----,-m,---------,,.,-,,-e-mw,-%. - , - -
4.17. Transient 7: Primary system pressure during small steam-line break from in11 power with one stack-open MSIV and a stack-open TBY I 4.18. Transient 8: Devacomer temperature during ranaway main feedwater to two SGs fram in11 power 4.19. Transient 8: Primary system pressure during runaway main feedwater to two SGa from fall power
- 4.21. Transient 9
- Primary system pressure during runaway i
main feedwater to one SG from fall power 4.22. Transient 10: Downconer temperature during ranaway anziliary feedwater to two SGs from fall power 4.23. Transient 10: Primary system pressure during ranaway anziliary feedwater to two SGs from fall power 4.24. Transient 11: Downconer temperature during 0.002-m2 hot-leg break from fall power 4.25. Transient 11: Primary system pressure during 0.002-m 2 hot-leg break from fall power.
4.26. Transient 12: Downconer temperature during break from a j stack-open POEV pins a stuck-open ADV from full power 4.27. Transient 12: Primary system pressure during break from a stack-open POEV pins a stack-open ADV'fron full power l 4.28. Transient 1: Downconer finid mixing behavior l 4.29. Transient 4: Downconer fluid mixin3 behavior l 4.30. Transient 12: Downconer finid mixing behavior 4.31. Temperatures associated with loop flow stagnation event 4.32. P, T, and h estimation approach 4.33. Mass and energy flows for two-node cooldown model 4.34. Comparison of TRAC and cooldown model temperature profiles for PORY LOCA with stack-open ADV (LANL transient 12) 4.35. Extrapolated downconer temperatures for large main stensr-line break at HZP I
1
..------.,,m.v.a, e s. w w.=.--g-- ,w.- . -. . - . - _ . = , -g-.,.e+rww.y-,-e-y -y,, -y.-m eg, m, y v we %-,--p rs,rm-- y--r- .-wr =,-
g.ry7- --- = - ""*
4 4.36. Extrapolated dowmoomer pressures for large main steam-I line break at EZP 1
4.37. Extrapolated dowmoomer heat-transfer coefficients for large main steam-line break at HZP
! 4.38. Extrapolated dowmooner temperatures for small malm
! steam-line break at BZP j 4.39. Extrapolated dowmooner pressares for small main steam-line break at EZP l 4.40. Extrapolated down' c omer heat-transfer coefficients for l small main steam-line break at BZP i
j 4.41. Extrapolated dowmoomer temperatures for large main 1 steam-line break at full power j 4.42. Extrapolated dowmoomer pressures for large main steam-j line break at full power i
i 4.43. Extrapolated dowmoomer heat-transfer coefficients for l large main steam-line break at full power l 4.44. Ext:apolated dowmooner temperatures for small main 1 steam-line break at full power 4.45. Estrapolated downoomer pressures for small main steam-i line break at fall power i 4.46. Extrapolated dowsooner heat-transfer coefficients for i
maall main steam-line break at fall power i
4.47. Bztrapolated downoomer temperstares for reactor trip
! (Segmences 5.18, 5.19, 5.22, 5.35, 5.36, 5.21 A and B)
I 4.48. Extrapolated dowmoomer temperstares for reactor trip (Sequences 5.25 A and B, 5.26 A and B, 5.27 A and B)
! 4.49. Extrapolated dowmoomer pressures for reactor trip (Segmences 5.18, 5.19, 5.22, 5.35, 5.36, 5.21 A and B) l I
j 4.50. Estrapolated dowsooner pressures for reactor trip i
(Segasmoes 5.25 A and B, 5.26 A and B, 5.27 A and B) 4.51. Estrapolated dowmoomer heat-transfer coefficients for reactor trip (Segmences 5.18, 5.19, 5.22, 5.35, 5.36, 5.21 A and B) 4.52. Estrapolated dowmooner heat-transfer coefficients for reactor trip (Sequenees 5.25 A and B, C.26 A and B, 5.27 A and B) 4.53. Estrapolateddogmoonertemperaturesforsmall-break LOCA (<0.016 ft )
I l
4.54. Extrapolatgd downconer pressures for small-break LOCA
(<0.016 ft )
4.55. Extrapolated downconer heat small-break LOCA (<0.016 ft 3) transfer coefficients for !
i 4.56. Extrapolateddgwnconertemperaturesforsmall-break l LOCA (50.02 ft )
4."7. Extrapolageddowncomerpressuresforsmall-breakLOCA (9 02 f t )
4.58. Extrapolated downconer heaj-) transfer coefficients for small-break LOCA (50.02 ft I
9 i .
i 1
I LIST OF TABIE.S 4
4.1. Samssary of twelve postulated overcooling transients 4.2. Comparison between TRAC and nessured plant data at hot zero power conditions i 4.3. Comparison between '11 TAC and nessured plant data at full power conditions I
4.4. Fluid film heat-transfer coefficients for twelve LANL transients 4.5. Cooldown nodel assumptions 4.6. System state changes for extrapolation of overcooling sequences by the cooldown nodel 4.7. Estinates of initial steam volumes for Calvert C11ffs-1 transients for repressurization from HPI shutoff up to PORY opening i
i 1
I
\
l 9
._. __ . . _ _ _ _ _ _ _ _ _ - _ _ . _ _ _ . , - . _ _ _ _ _ _ . - . , . _ , _ . _ _ . _ . ~ . . - . _.
l l CC-4.1
. 4.0. 'runnuAr-HYDRAULIC ANALYSIS OF POTENTIAL OVERC00 LING IRANSIENTS
- OCCURRING Ar CALVERT CLIFFS UNIT 1 NUCLEAR POWER PLANT
- 4.1. Introdnation In Chapter 3,115 sequences were identified for potential thermal-hydraulic analysis. Even though many of these sequences have relatively slow cool-down rates (less than 100*F per hour), some thermal-hydraulio data must be
[
l generated for each sequence or at least for each class of event.* Clearly, i
q an extensive thermal-hydraulic analysis of each sequence would be unneces-i sary. Therefore, the approach used was to analyze 12 selected sequences to e
i provide data that could be ssed either directly or to estimate the thermal-j hydraulio characteristics of each of the 115 seguences.
I
! The selection of the 12 sequences is described in Section 4.2. For each one, an analysis of the system response over a two-hour period was per-formed by Los Alamos National Laboratory (LANL) as described in Sec-
- tion 4.3. Two topics were identified as requiring special attention:
1
- (1) mixing in the downconer region, and (2) the heat-transfer coefficient at the surface of the reactor vessel wall in the downconer region. These j two characteristics were examined by T. G. Theofamous of Purdue University, i
j and the results are presented in Sections 4.4 and 4.5, respectively.
3 Finally, the results of the analyses discussed in Sections 4.3, 4.4, and 4.5 were used to estimate the thermaA-hydraulio characteristics of those I sequences for which a specific calculation was not performed. The process 7
i j applied and the results obtained are presented in Section 4.6.
I
!.!..!:.!:.i;'::::'i..,,:':::
. u..-r,.... :::'.:'.3:J::::'
... . . . u . i ...u*!!!di . . . .: a1.;*..!.!*::.lll;:
. ...... ....u" lt*::.!*::!!'::*::'. : *!! ;!l!J:10."*"*"* "' """" '"' "
- 1 i
_ ,__ . - _ - . _____.._.__.-._-__..._.___-._.,..__r, .m_. _ _ _ _ _ _ . _ , ._._ -_. , _ , , . . ~ . _ _ _ . . _ _ , , , - . . , . ~
CC-4.2 2
4.2. Selection .gf Twelve Seamences i
The primary objective of the selection process was to identify sequences that would provide information on the impact of the initiating events, potential equipment failures. and operator actions on the primary system cooldown rate and pressure. As a result, many of the sequences chosi.a are low-frequency probability sequences.
. 4.2.1. Segnances Initiated by Large Steam-line Break at HZP i
)
i Three sequences [ sequences 1.4 and 1.7 plus a sequence to represent the t
y residual * (1.3)] were chosen to provide information for sequences initiated i
[ by a large steam-line break at BZP. Two large break sizes are covered by
)
0.1 m2 and a full double-ended main steam-11ae break.
the three sequences:
1 The two different break sizes will be used to examine the effects of the 2
range of sizes in the large-break category. The 0.1-m break size was used for sequence 1.4. and the in11-break size was used for sequence 1.7 and the
> residual sequence. These three sequences can be used to provide the fol-
! lowing information for analysis of the large steam-line break sequences:
j (1) The silect of a variance of break size, t
- (2) The effect of continued feeding to the steam generator on the broken line, i
(3) The effect of the blowdown of both steam generators.
I l In Chapter 2 it was stated that Baltimore Gas and Electric is considering a procedures change at Calvert C11ffs-1 which wonid leave two reactor coolant
- n. . .4 t.r..,in........ is,.t e s.:s.i... _ the _ _ at =e nsi. .f;--. . 3.ts . . _ ..
g . . c. . - - g asn.. . .u.. 4 .:t ...
in e.,r er. . ..e.,,. . _ . . . ,,
. .. . . .._s,-
- n-l . - - s=. ves. =;
m v. ; :_.--
.. ..e v, % w,. x m.-r:::: [ yv .x.r
.. . n -
- n. e----- .. -
-; -- ru - s e .~.-:. . .. ,
.w. 4
CC-4.3 pumps ranning during a cooldown caused by a secondary system initiating event. Although this segnance has not been identified as part of the sequence tables, it was felt that a full calculation of this effect was l l
! necessary to evainate the potential effect of this procedures change. Thus
! sequence 1.4 was analyzed with two reactor coolant pamps lef t ranning.
l 4.2.2. Sequences Initiated by Small Steam-line Break at HZP At the time the 12 sognences were chosen, it was felt that data from the j large-break cases at both RZP and fall power, along with data from small-l break cases at fall power, could be used to estimate the small-break I
seguences at EZP. Thus no elaborate calculations were performed for small steam-line break sequences at the RZP condition. In retrospect, even though we were able to estimate the temperature and pressures associated with these transients, the evaluation would have been greatly simplified l with at least one elaborate evaluation of a small steam-line break at EZP.
4.2.3. Seguences Initiated by Large Steam-line Break at Fall Power l i i
i i One elaborate evaluation was performed for this initiator. This sequence
- involved the 0.1-m 2 break with failure to control repressurization and fallare to throttle anziliary feedwater.
l f 4.2.4. Seguences Initiated by Small Steam-11ae Break at Fall Power l
Small steam-line breaks at full power are dominated from a frequency stand-point by failures of ADVs and/or TBVs. As stated in Section 3.3.3.3,
{CC-4.4 i
4 failures of these valves at full power are treated as f ailures following a f
reactor trip initiator. Thus, the data used to estimate sequences in this :
i
- category will ooes from caleslations performed for the reactor trip initia-tor as described in Section 4.2.5.
- 4.2.5. Sequences Initiated by a Reactor Trip 4
Elaborate calculations were performed for four sequences associated with the reactor trip initiator. TWo of these sequences deal with' steam-line j valve failure, while the remaining two are steam generator main feedwater overfeed seguences.
\
l Both of the steam-line valve failure sequences involve the failure of a 4
TBV. In the first sequence, the MSIVs close as required. This provides i
- information on small steam-line breaks which are downstream of the MSIVs and involve isolation of the broken valve by closure of the MSIV. The second TBY failure sequence includes the failure of a MSIV to close. This not only provides information on small breaks downstream of the MSIVs when
} a MSIV fails to close, but also represents small breaks upstream of the I
1 i
i The two overfeed seguences involves (1) the overfeed of one line, and i (2) the overfeed of both lines. The overfeed on both lines represents the i
maximum main feedwater overfeed. The single line overfeed was examined to
- evaluate the potential for loop stagnation due to the asymmetrio cooldown.
l In all four sequences, operator actions to control repressurization and APW
! . . i i
CC-4.5 were not considered. This was done since it was deternised to be much easier to extrapolate from the osse where these operator actions are not I performed to the case where they are performed than it would be to extrapo-late from the case where these operator actions are performed to the ease i
where they are not performed.
4.2.6. Sequence Initiated by a Small-Break LOCA (10.016 ft2 ) ,
l .
1 The sequence choses to provide information for this ostegory was a PORV-sized break with a failure of one ADY to close. The PORY size was used to ensure that the pressure remains reasonably high during the transient. The additional fallare of an ADV to close provides information on the compting of a small-break LOCA and a small steau-line break. As in previous cases, the operator action to control APT flow to the intact steam line was not i
l considered for the initial calestation.
l 4.2.7. Segasmoe Initiated by a Small-Break LOCA (20.02 ft2) f The most probable break size in this category is a 2-in. break besasse of 1
i the assy 2-in. 11aes that come off of the mais primary piping. A 2-in.
1 break represents a flow area of ~0.02 ft2 (0.002 m 2 ) . Thus the calculation performed to provide information on this class of event was a 0.02-f t2 break.
i 4
4 l
t 4
i . .
CC-406 f i
- 4.2.8. Sequenses Initiated by 1ACAs with Potential Loop Flow Stassation l
l
! - In Seetion 3.4.5. several sequences were defined which cos1d potentially lead to loop flow stessation. As stated la that section. It was determined that loop flow stessation weald be assumed for these esses as a screenlas f
mechanism. Since loop flow stessation is assumed, the downeomer tempera-i 1
tare becomes a mislag analysis, nas these seguemees were smalysed as part l l.
of the minias analysis disenssed in Section 4.4.
l I I
1 4.2.9. Sequesee Initiated by 14ss of Mais Feedwater with 4 i j Subsegment AFW Overfeed
?
i
}
i In Section 3.4.6, a bosading ease was identified to represent this estesory 3
i of event. D as a transisat smalysis was performed for this seguesee. In addition to the segsease description as gives la Section 3.4.6. it was l assumed that the repressurisation was not oomtrolled by the operator.
l I
4.2.10. Summary no 12 seguesses identified saa be grouped mader three sategories: (1) a i
! steam-line break, (2) ressway feedwater, or (3) small-break LOCAs. A sun-
! il l , /l* mary of these transients is presented in Table 4.1. It should be noted I
that eseept as speelfied in Section 4.3. several operator actions /imastions l
i l
, were assumed to be sommes to all IANL treasiest estentations.* nose
- p .'
I **
I assumptions were 1
l
'g.g.. y a m .. ..... ... .ui . .. .. t.u. .. .... . ...... i.
I .
l L
CC-4.7 4
Table 4.1. Sammary of twelve poststated overcoollas transients j 11 Steam 1-m 3main ae Brdaks atens-line break upstream of NSIVs (1) From EZF (2) From in11 power (3) From EZP with two operating reactor coolant pumps Double-ended main steam-line break upstream of MSIVs (4) From EZF with continued APW flow to broken stems line (5) From HZP with two stack-open NSIVs l Small steam-line break downstrema of NSIVs (6) From fall, power i (7) From full power with one stask-opes NSIV Ramaway Feedwater (8) Runaway mala feedwater to two steam generators from full power (9) Runaway mais feedwater to one steam generator from is11 power (10) Ramaway anziliary feedwater to two steam semerators from fall power i' Sea 11-Break Long-of-Coolant Aeeldents (11) 0.002-m hot-les break from is11 power (12) Stack-opea pressuriser PORY with staok-open secondary system ADV i from fall power l
)
l l
J l
l l
i n
CC-4.8
\
(1) Operator turns off all RCPs 30 seconds after SIAS based on low pressariser pressure.
I (2) Operator fails to turn off charging pumps prior to in11 repres-satisation.
(3) Operator fails to control repressarization.
(4) Operator fails to maintain level in intact SG.
(5) Operator fails to respond to high SG alarm at 30 inchss.
l (6) Operator fails to respond to high SG alarm at 50 inches.
J l
e
-,--,,,e , , , ,-- - , - - - - ~ . - . --- e--- - - - - - - - -
.-. __ ~ _ - _ - - .. . . . . .. - - _ - - _
1 . .
' CC-4.9 1 l l
i l 4.3. Mlg,23&G Analysis Los Alamos National Laboratory (LANL) participated in the PT3 program by l
usin; the TRAC-PF1 computer code to provide best-estimate thermal-hydraalia analyses of the 12 postalsted overocoling transients identified la Table 4.1. Each of the 12 transients was to be analyzed by LANL for a 2-hour transient period.* A summary of the TRAC model ased and the results d
/ obtained for each transient smalysis are presented in Sections 4.3.1-
/
4.3.13. A separate report, " TRAC-PF1 Analysis of Potential Pressurised-Thermal-Shook Transients at a Combastion-Engineering PWR. " has been pub-a lished by 1ANL which describes in great detail both the model and the tran-sient analysis performed. This reycrt has been isoladed here as 4
Appendix F.
i Since the thermal-hydraalle characteristics of the transients are la some j Instances a result of complex intra-system cooling mechanisms and since in many instances small differescos in temperature can have signifloant effects on the fracture-sechanics analysis, a separate review of the TRAC
! analysis was performed by Brookhaven National Laboratory. The report sun-
! marizing the BNL work is isoladed la this report as Appendiz G.
I i
i i
l
- o.... .....u rie
- n. :-n.n .in...ini e etw .... . uu..w n.tu.t euiro
.un..:.t e i+-n = te uins
- i. in.
..i.o.u i .
...u.. a.w,.ai t
' . 4 inu
...i n. .i4u. ner
.u.u.nio 4..au .s.ono re unw.us..in..iu n. uw. .nu..e..i.nntens
- n. :-n=e .uires.
no .ar mon.e.ne.,: .
l s.u.un .uiro..
t:-n..,no n.ia n. w in.iui.4 ein. .
. . n. .a ie .i 4 .. . .u. .no. asin. n in. ..e.u..
au ru ou i.e.u ,r :....w. n..i n.-
.i.n eu u
.i..4..u.
.n. o..u.ie un. nen. :.e en.uu..u rr u u....
i
.l.e...
- i. :.n.. n.e.u
= enluu an.
n..... ,s -
j .un . i.e.s.umer..n.u.
. .. .i n.uoa. u.n u.u.u.e.n.ie .uir.i i. : n.
n.r.e. o . 3-n.n. ew u..2-en.n..u u. re e.n u . a a u o e au. .i. c linur.,u.ie.w u . n u m.eu.u.,,a, u .
4 or uu- nr-r,,...wi..n n u.. ,....r-i.w u. o a .n.
l 1
i . .
CC-4.10 I
j
\
j 4.3.1. M AC-PF1 Model of Calvert Cliffs Unit 1 l l
1 he RAC model used for the Calvert Cliffs-1 analysis (nAC-PF1) resulted from a evolutionary process involving several interactions with the plant owner, Baltimore Gas and Electrie, and the plant vendor, Combustion i
Engineering. n e H AC noding diagrams for the primary system, feedwater train, and steam lines are presented in Figures 4.1, 4.2, and 4.3, respoo-l l'
- I tively, n e development of the moding and control signals for each system s is described in detail in Appendix F.
Two initial condition models, hot-sero power and fall power, were required i
to analyze the 12 transients. For each initial condition model, a steady-state calculation was performed and compared with plant data.
I j The RAC-PF1 HZP steady-state calculation for Calvert Cliffs-1 yielded very stable primary-side conditions but oscillatory secondary-side conditions.
The fundamental difflonlty in determining the secondary-side conditions
) during BZP occurred because the vapor generation rate was very small and i
j appeared to destabilise the steady-state solution for the SG wodel.
1 :3 y Table 4.2 compares the actual plant conditions with the conditions gen-1 I
l* ersted by TRAC af ter 15 minates (reactor time) of the steady-state HZP cal-culation. The comparison is reasonable with the exception of secondary l steam flow. A simple energy balance dictates that, in the steady state, .
l l
the tabulated plant vaine for the steam flow is correct. ne over-l prediction by DAC suggests that the lignid temperature entering the SG
{
l riser was too high because the temperature profile in the downconer was not
. - _ . -- . - . _ - - - - _ _ . .- . ._ - . - . - - - -- = - - __
CC-4.11 1
Table 4.2. Comparison between TRAC and measured plant data at hot zero power conditions Parameter Measured Plant Data TRAC Predictions i '
Primary Side
- 1. Pressure 15.5 MPa (2250 psia) 15.5 MPa (2250 pata)
- 2. Fluid temperature 550.9 E (532*F) 551.8 K (534'F)
J
- 3. Power 100 hr after shutdown 9.38 NW decay heat
+ pump power + 17.38 NW from the pump
- 4. Mass flow 19,300 kg/s (153 x 10' lb/h) 19,700 kg/s (156 x 10 6 lb/h)
- 5. Precanrizer 3.66 m (144 in.) 3.66 m (144 in.)
Secondary Side
- 1. Pressure 6.20 MPa (900 psia) 6.17 MPa (896 psia)
- 2. MFW temperature 300 K (80*F) 300 K (80*F)
- 3. Steam flow 10.1 kg/s (22.2 lb/s) 11.8 kg/s (26.0 lb/s) l
- 4. SG inventory 95,000 kg (210,000 lb) 102,000 kg (225,000 lb) 4 5. 7BV flow aren 7% open 5% open I
l i
I i
, , , w - - ~ . --- , , , - , ,y-- ~ .. . - - - -, ~ . , .,. ,,.---w.sm, . , -
1 CC-4.12 l
(
j th a@:,..i.........l*.g si@i .
nci- -
- s. L
- , p i
/ 3, w .cs N *
~ "f g'.: -w c- g ; 4 n
lj-a ap beui Illi Il y P N* u [j3D i n &
g g, ; ia
.f.-
! @hE!@N f ai" @ r }r = o @ % m ,
i ds 5@ 5 !@ E 8 e a g s E g e !!;b 96 wayps ys 8g.n9l% ajN! g s ga i ) s. g ii n, n na "11 8 '
@ u ~ , 4. - f g!y! @ -@m **
%*. =gE ,,
a a i 3
[4 p e@# p .f $kI~J
+-:w;Tg,1@elIg@gy@g Je geesged nemi 60s th,
{
/_2., s 45_7its. 2 t ul -
,., 2 em= - -. --~ i 4~e I .D . T! kg8]-@@i -- l }
..f-----3 .
1 f P b 6 4
\CC-4.13 I
E E
. N -- -
4
*!!5hl eiI
*lEl eiI mel i s
! ,i eisse i pi i
eg! E 4 l c ~ egl i;
- k g! I 51 ' @"i 21((ti n
{sts $ 3 I li.cm j b .....g e ..... d #'l , f
@]B ,
e z e,I;si @it is i4
@55 't J
1 aw l f 3 5 g
%# l 1 5
p
@h - *[! $ @e
. . . ....- u tu n . % ,2l
- :- ik= 3 a
ggey my -'a- ] @ ,
@]aMi =k .
l' fu 8l y @li l e!;b - , . , . . . 5lb i s te
=
ell;;
.i
@![!
s i@'i I eb"! Gb s l' Is 'i, g
*I' ~; '
8 I*gela*!!
= . z ei> ; y 1
, @gL-] _.
4 S@ s@ aE)i aii -i-s : :-is
O e CC-4014 i
~
t
~
li .
+ i,i i
; }
l Ii! e 2 3 I
*5"5 d ,,....... ,
5 sip p a I-
~ 3
@f , [ '
h5. 1
- @E j i sfi . ei -
saa an, , I
' [
! i $ @l igxa< {
ibx@l 22 .
,d
-,d .g .
l
\
l 4 t
CC-d.13 is11y established. H e EEp toeperatore prof!!ee appear reasonable and reveal a sitsatten la which the sold feedwater heated to eatsraties by the time it entered the riser seetles. In the riser, the sea 11 vaper generation rate yielded a very sea 11 veld freetles natil the liquid surface was reaeked. Even with the esaggerated steam flow, the LAN1. analysts.believe that the MAC EEP steady-state seistles was eless to the setual plant senditions. h is belief la supported by the corroependesee of the TRAC seisties to the senditions is.uitively espeeted when the power was nearly sere. that is, a primary temperature that was eJeestially selfore and a seceadary tempera-tore that reseebied a stewly staaering pet of botting water. Of the 12 trasetents. eight were initiated free is11-power steady-state senditions. Darlag is11 power the reseter operated at 2700 W with as addittomat energy impet of 17.38 W free the RCPe. he salestated tempera-tsee ineresse screes the veneel was 24.4 I (47.4*F) with an intet toepera-ture of 589.3 I (847.0*F). he pressere drop through the 1eep was 0.54 Nya (78.7 paid). Nakesp/ letdown flew reestated the prosestiser levet to S.44 m (215 in.). Neat was traseferred throsch twe See to the secondary leep. The feedwater flew was regulated to estatain a specified 18tsid levet by the MPRY setas a three-mode oestretter. H e valve area wee deternised free the SG 1evel and feedwater flew /eteam flew mionateh na deseribed in Appendia F. De NFW-peep speed was adjusted to asistain a seastsat preessee drop of 0.71 NPs
CC-4.16 l i i u 0s ,sid> serees the n n e. He feedwater was heated to dys : <431. ,) by
- l i
two high preessee feedwater heaters and foer low-prosesse feedwater heaters. He 11gsid mese la each M was M2300 kg (136400 lb). l h e is11-power transients were initiated fees differest is11-power steady- , ! state eatestations. As the Calvert C11tfe-1 model evetved daring the sol- : \ estation of the treasiente, it was messesary to rersa a steady-state sales-l M 1sti.a wk.a.,., a e.det was e.difi.d. hbt. 4.3 .i .. a .es,ari..a b.tv.es
'l
- the TEAC eelestation and the seassrod plaat sendittens for the last steady-l state estestation. H e reestte are la good agressent, as were these fees ;
l t l l the previese salestations. l l l l 4.3.2. Steaartime Break Catestations i f I h e steam-line breaks seasidered la this analysis reased free a desble- l 1 l l ended set 11stine break to a elaste etsek-open D V. He general evente fel- [ t & j ; lowing a steam-line break were as fetteve. After a break er esseh-epes ! 1 i l valve oessered la the stone line, seeesdary depreessaienties reestled. If ; I the plant was at is11 power, the reaeter and tastine tripped (probably es j i t i 11 geld levet la the M) sad the NFW flew ran beek. Seensee the seesadery f l i lignid looperatore deeressed with the satsraties toeperature, the primary l ll temperature was seversed by the AT screes the tubes in the Ms. { l l D e dessesse la seeendary prosesse esseed an MIS, ialtiatlas steente of t the MIVe and NF1Ye. If these valves operated eerreetly and (selsted one i M free the break, aeyesotrie senditions were indseed en the peleary side, j r i As described la the SAC-Analyste-Hethodelegy sostion of the LANL report, i ! I
. o 1
CC-4.17 Table 4.3. Comparison between 7MAC and measured plant data' at is11-power senditions Parameter Measured Plant Data TRAC Prediations Primary Side
- 1. Core power 2694 MW 2700 W
- 2. Vessel flow 25.27 m3 /s (401.100 spe) 24.9 m3 /s (395,230 spa)
- 3. AP,,,,,g - 0.23 MPa (33.3 psid)
- 4. AP 0.19 MPs (28.13 paid) 0.19 MPs (28.15 paid)
SG S. 0.54 MPs (78.73 paid) 0.55 MPa (80.5 paid) APg ,,,
- 6. 7,,gg 339.3 K ($47.0*F) 559.5 K ($47.7'F)
- 7. AT,,,,,g 24.4 K (47.d*F) 24.0 K (47.0'F)
Secondary Side
- 1. Feedwater flow per 80 749 kg/s (3.95 s 108 lb/h) 737 kg/s (5.83 a 10 8 lb/h)
- 2. 50 done pressere Loop-A SG S.90 MPs (Old peig) 3.9 MPs (851 pois)
Loop-B SG 5.86 MPa (850 peig) 3.9 MPs (852 pois)
- 3. MFW pump disekarge prosesre Loop-A SG 7.8 MPa (1130.7 psia) 7.66 NPn (1123.7 psia)
Loop-8 54 7.63 NPa (1104.7 paia) 7.56 M (1111.0 psia)
- 4. MFW temperature 494.8 K (431.0'F) 4H.2 K (433.S'F)
- 5. MFRV flow ares (% open) ~90 93
- 6. So 1iguid mass 62.350_ha (137.458 lb) 63.000 ha (138.600 lb)
CC-4.18
\
Appendix F, this asymmetry could result in temporary flow stagnation in the i l
" intact" SG so that it eventually boiled dry. AFT filled the intact SG and f
i because of assumed operator inaction, the intact SG overfilled. If neither l or both SGs were isolated, symmetric conditions would exist on the primary side and AFT would be delivered to noth SGs if a low liquid level in the SGs was reached. 4.3.2.1. Transient 1: 0.1-m2 Main Steam-line Break from HZP l The downconer temperature and pressure profiles for Transient 1 are il rb ( presented in Figures 4.4 and 4.5, respectively. f*I ~
- y-l The temperature profile was divided into three phases. Phase 1 (0 - 1300 seconds) was dominated by the blowdown of SG A. The blowdown was limited by choked flow through the 0.1-m2 (1.0-ft 2) break. Data from Combustion Engineering indicated that the moisture separators would continue to be effective at the predicted steam flows and thus the fluid exiting the break was 100% steam. As SG A depressurized, the saturated liquid flashed and the secondary temperature decreased according to the saturation curve.
Power extraction slowed as the liquid inventory depleted because the tubes began uncovering, which steadily decreased the heat-transfer area. Because AFW was valved out to SG A based on an asymmetric-SG pressure signal, its secondary eventually voided completely. This event marked the end of Phase 1 (at 1300 seconds).
- Phase 2 (1300 - 4200 seconds) of the downconer liquid temperature was the
' period after SG A dryout and before natural circulation was established in i______ _ _ . _ _ - _ w m,,m.--,w-,,,--c.y - m-, ep ww,p , , , p.,,.gy---c. __,-y-,, , . , , , ,,_,,,_,v -,,,,,y,-,-9
,-ee'--'--,- ,y- --pgmeey e w- -v e -w- g--w-w-%,- --
CC-4.19 590 . . - 600 ; f 560-530-j .-500 SCO- g k -
-400 s s- 3 3 o E
h 440 5 ,.
-300
$ 44- 1 2' ]
s , . 380- G A dryout
-200
~
350- 1. blowdown of SC A
- 2. heating from core and slabs; I n energy removal fr a SG B (or A) .
320 -
. 3. natural circulation and energy . t00 removal in Loop B 290 . .
0 1000 2000 3000 4000 5000 6000 7000 8000 he (s) Figure 4.4. Transient 1: Downcomer liquid temperature during 0.1-m2 main steam-line break from HZP. (Note: This transient assumes MULTIPLE operator / equipment failures; see Section 4.1.10 for failure assumptions.) 18 ,
- - 2450 7 4 *1 i ,-
lfl
. 2100 g.
n- ~ -- 030 10 - ~. uco 3 8 a 3
$ 6-0 E -1050 [
' I 6- f
-v' - 700 f
4- l l,
- - 350 2- ,
O . . . 0 0 1000 2000 3000 4000 500's 6000 7000 5000 me (s) Figure 4.5. Transient 1: Primary system pressure during 0.1-m2 main steam-line break from HZP. (Note: This transient assumes MULTIPLE operator / equipment failures; see Section 4.1.10 for failure assumptions.)
CC-4.20 Loop B. The downconer temperature went through a maximum of 435 K at 4200 e seconds. Energy was added via the core and heat slabs. H e RCPs and SG A had no input during this time period. SG B added energy only to the tu'oes because Loop B was-stagnant throughout Phase 2. The deadhead of the HPI pumps was reached at 100 seconds so HPI flow was zero in Phase 2. Charging flow continued throughout the transient. The PORVs opened at 3120 seconds,* relieving the fluid input from the charging system but at a much (/ / higher temperature. Phase 3 (4200 - 7200 seconds) began with the onset of natural circulation in Loop B. Because it was assumed the operator failed to throttle AFW, the liquid level in SG B rose above the moisture separator deck and natural circulation was established on the secondary. AFT mixed with the warmer liquid in the riser, making SG B an effective heat sink. Energy removal by SG B induced natural circulation on the primary side. It is interesting that as the primary fluid mixed in the system, the downconer temperature approached the bulk temperature circulated at 4200 seconds. He circula-tion was ended at 7200 seconds with the primary temperature decreasing l slightly. SG B was slowly becoming a colder heat sink with continued injection of AFT. The primary temperature was also decreasing as charging
/, flow replaced the hotter fluid leaving through the PORVs.i I. .
4.3.2.2. Transient 2: 0.1-m2 Main Steam-line Break from Full Power With the exception of the initial power condition (full power vs HZP), Transient 2 is identical to Transient 1. The temperature and pressure It is assumed for this esisslaties that se attempt is made by the operster i I to sentret the repressortsation. This allows the pressere to reesh the pelat at which the 70sY w!!! eyes. This assuses that the 70sY is allowes to oestineally eyste opes and shut for the duraties of Phase 3
~ ..Yb'- *k- ._. T ! { -y/.,,
-yhhL..-.,j- .i1, _ h [. . i ( ]. - -~- J.feC bN.-7,'.11 ;
i
, , - . _ . . _ . . . _ , . - . _ _ . - . . . _ . ~ . . . - , . . , . . _ , _ _ . . . . _ _ , . - , , . . _ , . . . _ , _ , . , _ _ , , , _ . .
CC-4.21 N l j/ (
- s l-} profiles are presented as Figures 4.6 and 4.7, respectively.
(, ' Again Phase 1 of the downconer liquid temperature profile was the period during SG-A blowdown. Because the system energy was higher and the SG mass lower, SG A dryout occurred much earlier (at 300 seconds) than for the same j i transient from HZP. MFW [~5000 kg (11000 lb)] was added to each SG for 15 l seconds af ter the reactor / turbine trip, but this was balanced by steam flow through the 7BVs. Because loop flows were very low in Loop B from ~250 to 750 seconds, the downconer liquid temperature varied as much as 30 K (54*F) I in the azimuthal direction. The total energy-removal capability of SG A was 98.1 GW-s. SG B removed 30.9 GW-s before SGIS at 44 seconds. After I this, SG A cooled the primary below SG B, and the resulting energy addi-tion, though small, severely slowed the flow in Loop B. Phase 2 (~300 - 800 seconds) was a period of relatively rapid heating fol-lowing SG-A dryout. Because Loop B was close to stagnation, less primary fluid was available to receive the energy deposition from the core, and so thi specific energy of the flowing fluid increased rapidly. As the primary temperature increased, SG B became an effective heat sink. In Phase 3 (800 - 2500 seconds), the average core power was ~46 MW. SG B removed ~24
, /f MW and the PORVs removed some energy after they opened at 1975 seconds,*
l'l but the primary fluid continued to heat. l l Phase 4 (2500 - 7200 seconds) was extrapolated from a previous calculation of the same transient. As the core power decreased, a balance was achieved i with the energy removal by SG B and flow through the PORVs. A quasi-I equilibrium state existed in Phase 4 with tle downcomer temperature at I *It is assmed for this eatsalattes that se attempt is made by the operater l to oestrel the repressurisaties. This allows the pressere to reach the pelat at which the F0EV will spea. This assumes that the POEY is allowed to oestisaally eyete open and shot I for the duraties of Phase 4. I
" I, ?T.y - f-T{,{ '
'N,,57- S ,{ di. _ gcp.t.Q[ #^ hY" *~ L **
l
._ _ _.- . _ _ _ _ . . ~ _ _ . . . . _ . _ . . . _ _ _ _ . _ _ _ _ _ , _ . . _ . . - - _ . . _ . ._.
CC-4.22 590 -
. . i . . - 600 560- -
530- g, g i i .-500 4 ^ I SCO- 3 - 5 470 - g\ 5
-400 Y -
3 - 3 ;2 SG A dryout 2 6 E g 440- - p y 1. SG A blowdown c
-s
~
- 2. Low flow in Loop B - 00 y 410 -
- 3. Natural circulation in Loop B ~
- 4. Extrapolatica (based on previous ]
,g-380- **I*"I*EI'") -
-200 350- , -
i 320- -
- 10 0
- 290 , ,
0 1000 2000 3000 4000 5000 6000 7000 8000
- m. (s)
Figure 4.6. Transient 2: Downcomer liquid temperature during 0.1-m2main steam-line break from full power. (Note: This transient assumes MULTIPLE operator / equipment failures; see Section 4.1.10 for failure assumptions.) 18 , 1 A "# q a 18 - as = = - j g, - 2t00 I 12 -,- --1730 2* d 10 - O e .- n+00 N E E s- - a- e d
- extrapolated -1050 g.
6- -
- 700 4- .
. 2-~ -
- 350 ;
O .
. , , , . 0 0 1000 2000 3000 4000 5000 6000 7000 8000 nm. (s)
Figure 4.7. Transient 2: Primary system pressure during 0.1-m2 main steam-line break ( from full power. (Note: 'Ihis transient assumes MULTIPLE operator / equipment failures;
, see Section 4.1.10 for failure assumptions.)
l l
CC-4.23 530 K (495'F). The primary temperature would decrease slightly with time 1 because: , (1) The decay heat was decreasing; _ (2) SG B was becoming slightly colder with continued AFW; and (3) charging flow at 300 K (80'F) was replacing hotter fluid that left through the PORVs. 4.3.2.3. Transient 3: 0.1-m2 Main Steam-line Break from HZP with Two Operating Reactor Coolant Pumps This transient was identical to Transient 1 except that two diametrically opposite RCPs remained in operation throughout the transient. The princi-pal effect of leaving two RCPs in operation was that loop-flow stagnation did not occur in Loop B and SG B became a considerable heat source during the initial part of the transient (0 - 500 seconds). 2
,/ b Figures 4.8 and 4.9 show the downconer liquid temperature and the downconer
/ pressure for this transient. Again the time for the downconer temperature I
(: . was divided into three phases. Phase 1 (0 - 500 seconds) corresponded to l ' SG A blowdown and ended at the time of minimum downconer temperature. Because two RCPs were still operating, energy-transfer rates were much higher' than when all four RCPs were tripped and SG A dried out at 500 seconds. The forced circulation in Loop B allowed SG B to deposit consid- I erable energy into the primary while it was being cooled by SG A. l t
,.--,.-.,,%,..p ._ y._ ,--..__..4 m._, ,%- . . , _,._,_,..,.,._,c.. , _ _ _ , _. __ . ,,4.,- _ , , - _ , . - . . . . . _ . , - - _ _ _ . _ . . - - , _ ., -
CC-4.24 590 , . . . . -600 l 550- - S30- --500 500- h . y G A dryout -400 Yg B M- i 3, 1 3 4 E
'4' l
- 1. 81owdown of SC A; SC B was a heat source -300 j I 410 - 2. Heating from core and RCPs as SC became heatsink -
m I
- 3. Quast-equilbrium between decay heat and '3 t^
380- 4 SG B/PORVs Extrapolated- - [
-200 350- -
M0- - .
-100 290 . . . . .
0 1000 2000 J000 4000 5000 6000 7000 8000 rm (s) Figure 4.8. Transient 3: Downcomer liquid temperature during 0.1-m 2main steam-line break from HZP with two operating RCPs. (Note: This transient assumes MULTIPLE operator / equipment failures; see Section 4.1.10 for failure assumptions.) , 18 ,
- 2450 96 - t -
- 210 0 to - .
12 -- -1750 ;, 10 -
--1400 5 * ;
E , & e 8- .
- s e E -
-105 0
[ 6- -
-700 4- -
-350 0 . . . .
. . O j 0 1000 2000 3000 4000 S000 6000 7000 8000 Tme (s) .
2 Figure 4.9. Transient 3: Primary system pressure during 0.1-m main steam-line break i from HZP with two operating RCPs. (Note: This transient assumes MULTIPLE i operator / equipment failures; see Section 4.1.10 for failure assumptions.) l l 1 I l l
CC-4.23 Phase 2 (500 - 1900 seconds) was a period of primary fluid heating (from the core, the two operating RCPs, and the primary side heat slabs) before SG B became a significant heat sink. Not much cooling was provided by HPI and charging flow. Phase 3 (1900 - 5300 seconds) began with a significant increase in the heat-transfer rate across the tubes in SG B. This abrupt increase was a result of an inadequacy in the TRAC code but perhaps was physical to some extent. As the secondary side of SG B filled with AFW l above the moisture separator deck, the liquid began to spill over into the steam space in the SG downconer above the feedwater ring. As seen from the rapid depressurization, TRAC overpredicted the condensation rate that would result from this spillover. The resulting depressurization caused the cold
' AFW that had accumulated in the SG downconer to flow into the riser region.
The colder liquid came into contact with the tubes and the energy-removal rate from the primary increased. Af ter 2500 seconds, a quasi-equilibrium state was reached. The PORY _J opened,* removing approximately 8 MW. SG B removed the remainder of the
-' energy input from the core, the heat slabs, and the RCPs, which amounted to
~15 MW. The calculation was terminated at 5300 seconds with the system in this quasi-equilibrium state. The system was cooling slightly with time because SG B was becoming a cooler heat sink with continued AFT and chars-l ing flow was replacing the hotter fluid leaving the PORVs.i 4.3.2.4. Transient 4: Double-Ended Main Steam-line Break from BZP l
with Failure to Isolate} APT Flow to Broken Steam Line f .,
/-[
The downconer temperature and pressure profiles for Transient 4 are
*It is assumed for this salentaties that as attempt is made by the operator to oestret the represserisaties. H is allows the pressere to reach the pelat at skish the Pop will eyes.
'nis assumes that the FON is allowed to esatissally eyele eyes and shot for the daraties of Phase 3.
A lastadse both fallare of automatie system and fa11ere of operater te
....~ . n w . .E**F**** .w--~~- e.=.- w==, $ g* ~ -- ~~: ~ ~ ~ v - -- '-' *w -
-E*'- +- **
- b , MQ& ., ,
[*.9 \% M #
=. w W
-e. . m ~.~~-~e* m --
.. . q .p . - ~ .....w..
- m.:~. . ,v. w- : .a.m ... .
m.__--- A-_ -,.e- m -w--- -.- .m-+irg-v,osyo-..-----m---,-.-'e-- -e-w,g-- , - - - . ---,-----e5 w,,--y..w*e-% ..--------ynys.-+--.--.g---,%,siyeg
-mg-%9q--ie.w.
CC-4.26
/ ,ill.,il 1
presented as Figures 4.10 and 4.11. As shown in Figure 4.10, the downconer g1 { , temperature was again divided into three phases. The first phase (0 - ~800 seconds) is characterized by severe overcooling of the primary caused by the rapid blowdown of SG A to atmospheric pressure. Although the blowdown rate was limited by the flow restrictors downstream of the SGs, the mass flow out of the SGs increased by more than three orders of magnitude over l the HZP steady-state value. Furthermore, the assumed failure of the asymmetric-SG pressure signal to effect isolation of AFW to SG A ensured that the atmospheric heat sink would not be lost because of SG dryout. During this period, the flow in Loop B stagnated following the RCPs being ) tripped because of reverse heat transfer in SG B following SGIS. Also, I during this period the upper heat of the vessel voided briefly (90 - 350 seconds) because the primary fluid contraction initially exceeded the HPI/ charging refilling capacity. A model input error caused closure of the MFIVs on AFAS at 2 seconds instead of on SGIS at 9 seconds, but this error has no significant effect on the results. The second phase (~800 - 3275 seconds) is characterized by repressurization
/
of the primary caused by unrestricted operation of the charging pumps. l j During this phase of the transient there is an approximate balance between i decay heat, heat transfer from the structure to the fluid and heat rejec-
- tion to SG A. However, because the HPI and charging flow added substantial mass to the primary [~46000 kg (101000 lb) during 0 - 800 seconds and
~30000 kg (66000 lb) during 800 - 3275 seconds to an initial mass of 224000 kg (493000 lb)] but very little enthalpy, the average specific internal energy decreased slightly. By 3200 seconds the downcomer temperature had leveled off at 380 K.
-<-,----,,e,---en-,--,---,n.,,-m.---,e,------- - , . - - . - - ---,rw----e,-----,nw o e , - w ,,e v e w ww- w,w --
CC-4.27 590 . . . - GOO s60-l -~%0 8
$30 J
'\ -
500-l' 7
-400 l -
} CD- s y
? O 440-
; - -300 y 410 - g i
A i 1 ,
. 2 3 - 200 350
- 1. SG A blowdown
- 2. Continued ArW to SG A
- 3. Extrapolated 320-
- - 10 0 290 . . . . . . . .
0 1000 2000 3000 4000 5000 6000 7000 8000 me (s) Figure 4.10. Transient 4: Downcomer liquid temperature during double-ended main steam-line break from HZP with failure to isolate AFW flow to broken steam line. (Note: This transient assumes MULTIPLE operator / equipment failures; see Section 4.1.10 for - failure asst.mptions.) 1s . . i g g OSO
~
15 --
- 210 0 i .
12 - -
-1750 7 1
- c x ,
& to -! -* bl.
I l i j i T l 2 3 . e a- . 8- -1050 [ 6- - d 1 extrapolated
- 700 4 .
- 350
- 0. . . .
0 0 1000 2000 3000 4000 5000 6000 7000 8000
. me (s) _ _ ,
Figure 4.11. Transient 4: Primary system pressure during double-ended main steam- l line break from HZP with failure to isolate AFW "ow to broken steam line. (Note: This transient assumes MULTIPLE operator / equipment iailurcs; see Section 4.1.10 for failure l assumptions.)
CC-4.28 The problem was terminated at 3275 seconds because the transient had sta-bilized with respect to downcomer temperature and pressure. POEV cycling , between 15.7 MPs and 16.5 NPa would lir.it the pressure because POKY capa-city was more than adequate to relieve the charging flow. Furthermore, the decay power was sufficient to heat the AFW to SG A to the atmospheric boil-ing temperature; therefore, the liquid temperature in the downcomer of the vessel would not fall below 373 I (121*F) within 7200 seconds (Phase 3). 4.3.2.5. Transient 5: Double-Ended Main Steam-line Break Upstream of MSIVs from HZP with Two Stuck-Open MSIVs Transient 5 is the same as Transient 4 except that the MSIVs failed to close upon receipt of SGIS and blowdown of both SGs continued. Also, the l operator terminated AFW flow at 480 seconds (8 minutes) . The transient may be divided into three phases as shown on a plot of the yF 5 downconer liquid temperature in Figure 4.12. In Phase 1 (0 - 1000 I seconds), a minimum temperature of 376 I was reached, which was a few degrees above the temperature of the liquid remaining in each SG-secondary af ter the blowdown to 0.1 MPa (14.7 psia) . Each SG removed ~97 GW-s of energy from the primary, which included AFT flow for 390 seconds. The heat slabs added 33.1 GW-s to the primary fluid. Af ter the AFW ended at 480 seconds, the primary temperature leveled off a few degrees above the secondar'y temperature (Phase 2). The downcomer tem-perature increased slightly after the termination of HPI flow at 1000 seconds. In extrapolated Phase 3 (3300 - 7200 seconds) the power from the l l
CC-4.29 590 -
. . . 4 i -600 560- -
- 1. Blowdown of both SGs
- 2. Quasi-equilibrium with both SGs 500 330-,
- 3. Extrapolated; equilbrium maintained .
5c0- - l T e- -400 7g y -
~
E O 1 E m. - E
$ -3co 3
AFW terminated y 4to. - HPI flow ended saQ- 1 m. y - a i 1 - t 2 3
*0 li 350- -
s20- -
- 10 0 290 . . . . . , , 0 1000 2000 3000 4000 5000 6000 7000 8000 me (s)
Figure 4.12. Transient 5: Downcomer liquid temperature during double-ended main steam-line break from HZP with two stuck-open MSIVs. (Nate: This transient assumes MULTIPLE operator / equipment failures; see Section 4.1.10 for failure assumptions.) i l l l l l a.-- ,_*--y ,,,--r--,- 7w.% --w... ---,y- y..- . <, -i- , *y - --,
,--g .,*y9 y y,
CC-4.30 Primary is expected to slowly boil the remaining liquid in each SG [-18000 kg (~40000 lb)] . At 3300 seconds, the power from the heat slabs was ~7.5 MW. Together with 9 MW from the core, a steaming rate of ~4 kg/s would be produced in each SG. With this rate as a maximum (heat input from the 1 slabs would decrease in time), the SGs would dry out in another 4650 ) i seconds (t = 7950 seconds), which is past the end of this transient. Thus, the temperature is expected to remain at ~378 K (221*F) for the remainder of the transient. G, j l-l s Figure 4.13 gives the system pressure. The blowdown of both SGs caused the system to depressurize to 4.1 MPa. BPI flow reached a maximum of 60 kg/s l ' to make up for the prim'ary liquid contraction. The upper head voided dar-ing the 50- to 900-second time frame. Charging flow eventually repressur-
. l- ized the primary system to the PORY setpoint* where it was assumed to l/ remain for the rest of the transient.t r ,--
I - 4.3.2.6. Transient 6: Small Steam-line Break Downstream of MSIVs from Full Power . The f ailure of one TBV to resent af ter opening on a turbine trip is postu-Isted in this transient. One full-open TBV is about half the size of the 0.1-32 (1.0-f t 2) break described previously (0.05 m2 /0.51 ft 2) . Because the TBVs are downstream of the MSIVs, a stuck-open TBV is isolatable, whereas the 0.1-3 2 MSLB described previously was not. The " break" communi-cated with each SG identically and so the thermal-hydraulic events on both the secondary and the primary side are symmetric.
- i :t./::,::::.:::::i.:!'."dt:'.it.::';t:',i:.::':
. . .u. . , ..u . .
1: '!:.:' ::'"
^
!!!'J:'::::.it:'.;';a."d' '""" " "'""'"""" *" "' "
1 l
l l CC-4.31 18 . . .
-240 nui 4dul a A A 16 - - -
.,wnspq -
- 210 0 u- -
12 -
-* U50
'l N
} 10 - .
. u00 $
5 5 s- -
*
- extrapolated T ~* 5 l
6- -
- 700 4- -
l ,_
,-350 0 . . . . . 0 0 1000 2000 J000 4000 5000 6000 .7000 8000 Time (s)
Figure 4.13. ' Transient 5: Primary system pressure during double-ended main steam-line break from HZP with two stuck-open MSIVs. (Note: This transient assumes MULTIPLE operator / equipment failures; see Section 4.1.10 for failure assumptions.)
-.- - - - - - - - , . , ,--,,p. v - , -- ,, - - - , .-e- e-m- , -a, -
-,~-p -
p e rg,
e . I l' I*
, 'Ihe temperature history in the downconer (Figure 4.14) was divided into i
' l five phases. Phase 1 (0 - 510 seconds) was the time before the stuck-open TBV was isolated frce the SGs as a result of the closure of the MSIVs fol-loving SGIS. The initial ~50 seconds of the transient should have been h identical to a loss-of-load.* The TBVs resented as the primary temperature f/ decreased. Then one failed, a relatively slow depressurization began in both SGs. The secondary pressure decreased until the setpoint for SGIS was reached. This marked the end of the cooldown caused by the stuck-open TBY.
Phase 2 (510 - 1050 seconds) was a time of primary fluid heating ending with opening of the ADVs on high primary temperature. Boiling on the SG secondary continued to remove energy but at a slower rate as the secondary repressurized. The ADVs were open in Phase 3 (1050 - 4200 seconds), modu-i lating to maintain the average primary temperature at 552 K. The TBVs also opened, but they had no effect because the MSIVs were closed. Boiling in the SGs continued and mass was depleted throngh the ADVs. AFAS i was received at 4200 seconds based on low level in both SGs.t Phase 4 i , ./:2 (4200 - 5800 seconds) began with AFW flow to both SGs. A cooldown ensued l as the AFW mixed with the boiling liquid in the riser phase. AFT flow affected the primary temperature in this transient more than in others
! because it was initiated to both SGs (no asymmetric-SG pressure signal) and
! both SGs were low in inventory. Also, both loops were in natural circula-tion on the primary; this allowed rapid feedback to the primary side. The cooldown is expected to continue at the same rate until 7200 seconds, reaching a minimum of ~510 K. Phase 5 (5800 - 7200 seconds) is the extra- j i l I polated temperature history. l
'Seesese of an esser is the hitial liquid temperatures la the presser!ser.
the primary side depresserised eset too rapidly. This ealsalaties was to be redese, bet bessese it was already p.ediated set to be of FT5 semeers. as adottlesel fallare of ese NSIV was speetfied. The reestentattes is reported la the sent seetten. The period (0 - 570 seeesdal before 3GIS was identiss1 to the specificattees of this transient. This transient is imeladed to give details of a 7200-sessed transions with the f a11sre of one Tsf esty. in.). esThis ..=~e.. .y .~ .w m, -- . . I 'AFAS wee based on a At messarement of ~d.3 m (-170Desed a se11 speed T . "
- T' i -
--" "W # eerresponded to a lignid tavestery of -17000 ks. .s-
"' .~
Itgeld messarement. AFAs woeld onest with 45.060 kg remalaims in the SGs. . .
. ~ ~ ..C It is askmews which mothed is sete sorteet, het AFA5 prehebly was s.st ,.
* * " ' ~
later them it should hav. Does. I _< , .- - - .-, _ _ _. . ..---..-,,__ ., ,m _ - _ _ . ~ . , - . _ , - . __ _ .,, ..._ .-,. 4 .._.._..._...-,.,.__._....-m-
CC-4.33 590 - - 600 560-
.-500 530-I 5 SCO-
~
40 $
$ 470 < 3 o "Ei o -
U h 440- 1. Slow blowdown of both SGs h J! through TBV -300
~
"5 v 2. Heating from core .
410 -
}
" 3. ADVs regulated primary temperature
~
- 4. AFu to both SGs
, - 200
- 5. Extrapolated 3 ,
3:0
- - 10 0 290 , , , . . .
0 1000 2000 3000 4000 5000 6000 7000 8000 r<ne (s) Figure 4.14. Transient 6: Down:omer liquid temperature during small steam-line break from full power with stuck-open TBV. (Note: This transient assumes MULTIPLE operator / equipment failures; see Section 4.1.10 for failure assumptions.) 18 a g - 2450
/y hjIei iff,El(IIs$(hhh,/ $ fg* 2 m .
16 -
- - 310 0 Q- - -- 050 to-,
. t40a [
5 8 l f e- .
-1050
[ , 8- ! A A m a extrapolated 700 4-
-350 2- -
l
- 0. .
. . .0 l 0 1000 2000 3000 4000 5000 6000 7000 8000 Time (s)
Figure 4.15. Transient 6: Primsry system pressure during small steam-line break from full power with stuck-open TBV. (Note: This transient assumes MULTIPLE operator / equipment failures; see Section 4.1.10 for failure assumptions.)
l CC-4.34 i t-
/. The pressure history for this transient is given in Figure 4.15. Energy removal, and consequently depressurization ended at 510 seconds when the
- SGs were isolated. As mentioned earlier, the initial depressurization was too rapid because all the initial liquid in the pressuriser was not saturated. SIAS should not have been zeached at 28 seconds. After the initial depressurization that was caused by the reactor / turbine trip (which would have brought the system to about 13.2 MPa), a slow depressurization continued because of the slow blowdown of both SGs. Charging flow repres-s
, sarized the system to the PORY setpoint* af ter energy removal ceased at 510 f/_ /,'
- seconds. The system pressure was never low enough for HPI flow. The pres-sure is assumed to remain at the PORY setpoint for the remainder of 7200-
/y, 3 second time period;i ',
[/ 4.3.2.7. Transient 7: Small Steam-line Break Downstream of MSIVs with s Failure of One MSIV to Close from Full Power
. /
This transient is the same as the previous transient with the additional failure of the MSIV on one loop after SGIS. Thus, one SG blev down com-pistely in this transient. t t . As shown in Figure 4.16. the downconer liquid temperature was divided into i' l lf
- 1 I four phases. Both SGs blev down through the stuck-open TBV during Phase 1 (0 - 570 seconds) . The end of this phase was marked by the closure of one I MSIV and the failure of the other MSIY after SGIS. The energy-transfer nochanisms were similar to those described for the previous transient. The minimum temperature for Phase 1 would be the minimum reached for the satire transient (as a result of a stuck-open TBV)* if only the IBV had f ailed as
( l a..i....,.,,......i.u......u.........,.....
..i n. r. ....ei...t... n6..:i . a........i....a a.
i
.t.a . .nt.h a . rose . u a . ..
...a tt. i ty .y.1. . .ad .b.:
T,hi.
.t th. d.r.tt.
. th t .t,h.Ph.POEF
- 3. i. 11 . t.
%. i . a. . . . n . . ..u.. .. .. ... .... ., a. =m..
l M g T =i.qi y + = g : , m s c. mezm .3 :s;..:yn.~.=-wm.:ay
. _ - . . _ _ . _ . = _ - - .- _ _ _ . - . . ..
.CC-4.35 590 - -600 580- S gN- A^#7""
.-500 530 _i .a. .i.
2 # 3 4 - 500- C
?
~
~ '00 co.
- 1. Slowdown of both SCs through TSV }*
440- 2. Continued blowdown of SC A -
- 3. Heatina by core after SC A was lost as
-300 a host sink
- 4. r.xtrapolated; quasi-equilibrium reached 410 -
- 3 between core and SC 8/PORVs 3sa.
- -200 350 -
320 -
. -iOO 230 .
0 1000 2000 3000 4000 $000 6000 7000 8000 The (s) Figure 4.16. Transient 7: Downcomer liquid temperature during small steam-line break from full power with one stuck-open MSIV and a stuck-open TBY. (Note: This transient assumes MULTIPLE operator / equipment failures; see Section 4.1.10 for failure assump-
! tions.)
3a
- - 2460
+ 1 1
- -110 0 g.
12
- --1730 1 9' '
7 O y . -i400 h g s- - E - -1050 [
^
i extrapolated ,
- - 700 l
j 4 t
-150 2--
0 . O i 0 1000 2000 3000 4000 b000 6000 7000 n000 Tw. (s) Figure 4.17. Transient 7: Primary system pressure during small steam-line break from~ full power with one stuck-open MSIV and a stuck-open TBY. (Note: This transient assumes MULTIPLE operator / equipment failures; see Section 4.1.10.for failure assump- , l tions.) I l
CC-4.36 specified for the previous transient. Phase 2 (570 - 1750 seconds) was a period of asymmetric SG conditions. One j l MSIV closed, isolating SG B from the stuck-open TBV, while SG A continued to blow down. AFW was delivered to both SGs until asymmetric SG pressures i were detected at 640 seconds. APW was then delivered to SG B only. Some azimuthal differences in the dowacomer temperature existed because higher I heat-transfer rates caused the primary fluid to flow preferentially to Loop A. The dryout of SG A marked the end of Phase 2. t Phase 3 (1750 - 2500 seconds) was a period of primary heating after SG A l - l dryout. The PORVs had not yet opened so SG B was the only heat sink for the energy deposition from the core. Phase 4 (2500 - 7200 viconds) was j extrapolated based on the 0.1-3 2 main stensr-line break from full power (the f originel run for 0 - 7200 seconds) . The heatup to a quasi-equilibrium l state should be similar for both transients, because the energy transfers 3 were similar. In both transients, SG B and the PORVs were removing the I decay heat, and the primary side heat slabs, RCPs, and SG A no longer influenced the transient. A quasi-equilibrina state is expected to be reached at ~525 K (486*F). 1 rf II Figure 4.17 shows the pressure history. The first 50 seconds corresponded to a normal loss of load. Then one TBY f ailed to resent at 50 seconds, the f i pressure continued to drop with a sharp decrease after the RCPs were tripped at 500 seconds. The pressure leveled at 11.2 ; ipa as the cooldown i f slowed end the primary liquid contraction ended. The PORY setpoint was ! I reached
- Just as the calculation was terminated. 7he system pressure is i
L i wh h he i
* * . - - l n
l CC-4.37 assumedtoremainatthePORYsetpointduringtheremainderofthe7205 j second time period.* SIAS was . received at 470 seconds, but the system pressure was never low enough for HPI. ! l 4.3.3. Runaway Feedwater Events Three transients were analyzed in the runaway feedwater category. The first two transients involve runaway main feedwater and the third transient involves runaway auxiliary feedwater. i 4.3.3.1. Transient 8: Runaway Main Feedwater to Both Steam Generators
, from Full Power 4
This transient was initiated by a reactor / turbine trip from full power at t = 0 seconds with an assumed failure of both MFRVs to close. The downco-Q I j, I", mer temperature and pressure profiles are presented in Figures 4.18 and e ' t'
/' g./ 4.19, respectively.
As shown in Figure 4.18, for this case the downconer temperature history was divided into three phases. The first phase (0 - 283 sect,nds) shaws a rapid decrease in downconer temperature. The initial 10 K (IS*F) tempera-ture drop that occurred between 0 and 60 seconds is the normal temperature f decrease that occurs when the reactor scrams. The significant decrease in core thermal power caused the AT between the primary and secondary sides of the SGs to reduce to a much smaller value that still permitted dissipation l of the decay heat. The energy removed from the primary fluid during this interim was ~22 GW-s per SG. At 60 seconds after the scram, the relatively h' ::':::: 0:'.?;a"J' '""" *"""-"""" *""" '"'
l
-~
cc-4.3s l 590 . . -600 560-. -
)
530- 3 --500 2 ; 500- ' - l
,.1 ' '
4o0 I, j 470 - - g , o 3 2 440-f y 1. Runaway main feedvater to both SGs h c o
- 2. Heating from core until primary temperature -300 y
} 410 - above secondary -
3 3. Quasi-equilibrium reached - slow boiling 3,g, in SGs _ d
-200 350- '-
320 - -
- 10 0 290 . . . . .
0 1000 2000 3000 4000 5000 60C0 7000 8000
. he (s)
Figure 4.18. Transient 8: Downcomer temperature during runaway main feedwater to j two SGs from full power. (Note: This transient assumes MULTIPLE operator / equipment failures; see Section 4.1.10 for failure assumptions.) is , ,
- 2450 16 - fg h h[J$h,ff[;$hYf;khYp"tf}ffh -
- 210 0 12 -- --175 0 2 ~
? ;
'-1400 5 $
; a- '{,'c E
n - ,
-1050
[ 3- - 700 4- }-
-350 0 . , , . , ,0 0 1000 2000 3000 4000 5000 6000 7000 80C0 Trne (s) ..-
Figure 4.19. Transient 8: Primary system t iressure during runaway main feedwater to ' two SGs from full power. (Note: This transient assumes MULTIPLE operator / equipment failures; see Section 4.1.10 for failure assumptions.) . J l ~ l r , 1 1 I
- - - .. , - , . ., - - . ~ ,
( /9. CC-4.39 cooler liquid that was in the feedvater pipes downstream of the high-pressure heaters was swept into the riser region of the SGs, pushing the hotter liqu.*d in the riser region into the steam-volnae region above the tubes. The effective lower secondary side temperature began to extract energy from the primary side at a rate of ~200 MW per SG. At 218 seconds, the MW pumps tripped on low-suction pressure because of depletion of liquid I inventory in the condenser hot wells. At this point, the liquid in the riser could no longer be replenished with cooler liquid. The riser region stagnated and quickly approached thermal equilibrium with the primary liquid temperature. The energy transferred to each SG decreased to ~15 MW. However, the thermal power produced by the decay heat is adding energy to the primary liquid at a rate of ~75 MW. As a result, the primary liquid begins to heat again. The downconer liquid temperature reaches a minimum temperature of 477.5 K (399.8'F) at 283 seconds. l Phase 2 (283 - 4800 seconds) shows a relatively slow heatup of the primary fluid following the trip of the main feedwater pumps. As the primary ten-perature increases, energy is continually being transferred from the pri-mary into the secondary. The stagnant liquid in the SGs begins to heat up until it reaches the saturation temperature corresponding to 6.2 MPa (900 l psia), the pressure setpoint of the TBY. The primary temperature levels l off at a small AT above the saturation temperature of the liquid remaining in the SGs. A slow boiling process now begins (Phase 3). The small amount of steam being produced in the secondary side of the SGs is being vented by both the ADVs and the TBVs. [The ADVs and TBVs opened because the primary 1 1 1 j side temperature exceeded 552.6 K (535'F) .] l l i
l 3 c CC-4.40 4.3.3.2. Transient 9: Runaway Jain Feedwater to One Steam Generator from Full Power This transient was initiated by a reactor / turbine trip from full power at
.m
- t = 0 seconde with an assumed failure of one MFRV to close. The tempera-
./
f;.c.,j ture and pressure profiles are presented in Figures 4.20 and 4.21. F *a . p.s. Figure 4.20 shows that the downconer temperature history was divided into five phases. The first phase (0 - 363 seconds) shows a rapid decrease in the downconer temperature. As with the transient discussed in the previous section, the initial 10 K temperature drop that occurs between 0 and 60 seconds is the normal temperature decrease that occurs when the reactor scrams. The energy removed by each SG during this interim is ~22 GW-s. At 60 seconds af ter the scrsa, the relatively cooler liquid that was in the feedwater pipes downstream of the high pressure heaters feeding SG A has been swept into the riser region of SG A. The effective lower secondary side temperature in SG A begins to extract energy at an average rate of
~260 MW. At 303 seconds, the main feedwater pumps trip on low suction pressure because of depletion of the condenser hot-well liquid inventory.
(Unlike the runaway main feedwater to two SGs, failure of one MFRV to close produces a feedwater flow to that SG of ~1000 kg/s. This depletes the con-denser hot-well liquid inventory in ~300 seconds.) At this point, the liquid in the riser region of SG A can no longer be replenished with cooler i l liquid. The riser region stagnated and quickly approached thermal equili-brium with the primary liquid temperature. The energy transfer in SG A decreases to ~28 MW. However, the thermal power produced by the decay heat is adding energy to the primary liquid at a rate of ~75 MW. As a result, i i i
.-. - .---w - . . 2.,- _-.., ., -.,,---.-,-,-wc-,.mm,v -----y%,e-,,,,,nw,-- c,w, ,-v,,--,,,,-,,,-3,,-,,-.,-,--, ,,,,,,-,p,w---y.,w-
CC-4,41 O 590 -
, . -600 MO- -
2 3 .-500 530-~ 2 [4 i 5
~
500-C f 470 -
- 1. Runaway main feedwater to one SG ~
- 400 7 3 1
- 2. Heating from core with Loop 8 stagnation -h T:s o e 2. Quasi-equilibrium e f 440- 4. AFW to both SCa
~
.s 5. Extrapolated f ,
-300 y '
a
~
410 -
.}
380- ~ E
- - 200 350-320- -
- -100 290 . . . . . .
0 1000 2000 3000 4000 5000 6000 7000 80C0 Trne (s) Figure 4.20. Transient 9: Downcomer temperature during runaway main feedwater to one SG from full power. (Note: This transient assumes AfULTIPLE operator / equipment l
^
failures; see Section 4.1.10 for failure assumptions.) i 18 , 16 - g.h ,, . ,ij jgg 'lh g gfif) g$ji,I,fi lI 1 - '
'l t
- 2100 12 - --f750 i5 '
'0- -
- a00 5 2
- s 8- A A extrapolated -
g j
-1050 g 6- -
- 700 4- .
-350 0 . . .
. . 0 0 1000 2000 3000 4000 5000 6000 7000 8000 Trne (s)
Figure 4.21. Transient 9: Primary system pressure during runaway main feedwater to one SG from full power. (Note: This transient assumes AfULTIPLE operator / equipment failures; see Section 4.1.10 for failure assumptions.) -
. . e r CC-4.42
\ -
the primary liquid begins to heat again. The average downconer liquid ten-parature goes through a minimum temperature of 491.0 K at 363 seconds. t Phase 2 (363 - 3200 seconds) shows a relatively slow heatup of the primary fluid following the trip of the main feedwater pumps. This is similar to the heatap observed in the runaway main feedwater to two SGs discussed in the previous section except that the heatup that occurs in this transient has only one heat sink - SG A. The other SG cooled only slightly during ! the runaway feedwater portion of the transient. As a result, the decay hcat added to the primary fluid could be dissipated through only one SG rather than two. Hence, the primary fluid heated up more rapidly for this } case. After SG A was heated again to the saturation temperature ] corresponding to 6.2 MPs (900 psia), both SGs shared the heat load equally. l The primary temperature leveled off at a small AT above the saturation ten-perature of the liquid remaining in the two SGs. A slow boiling process
. began (Phase 3). As in the transient discussed in the previous section, the primary fluid temperatsre during this period exceeded 552.6 K (535'F). I Both the ADVs and the TBVs reopened. This vented the steam being generated l
by the boiling process. Subsequently, about one-third of the decay heat was removed by each SG. The remaining one-third of the decay heat was removed by convective mass transfer associated with injecting cold charging . flow into the primary system at a rate of 8.3 kg/s (6.59 x 10 4lb/h) and rejecting, on an average, the same mass flow rate through the PORVs with a much higher temperature.
- i i
l Because the mass inventory in SG B was initially depleted somewhat at the beginning of the transient and was not replenishing during the runaway l l l
g ' . ** 4 CC-4.43 L. . _ . .i ) feedwater portion of the transient, the slow bolllag process that oossered in Phase 3 continued to boil away the remaining liquid la SG B. At 4800 t seconds, the level in SG B was finally low enough to activate APW to jutth i SGs. He contimsons addition of cold 277.6 E (40*F) liquid to each of the
' SGs resulted la a contimacas reduction of the secondary side heat sink ten-i perature. h is, in tara, prodsood a doorease in the primary fluid tempera-tare (Phase 4). Omos the primary side temperature decreased below 552.6 K (535'F), both the ADVs and the DYs reolosed.
n e caloslation was terminated at $g00 seconds. Nowever, it was antioi-pated that the primary finid temperature would contiano to doorsase at i approximately the same rate observed in Phase 4 for the interim from 5800 i to 7200 seconds (Phase 5). , i i i 4.3.3.3. Transient 10: Ramaway Anziliary Feedwater to Two Steam Generators l from Fall Power J
! n is transient was initiated by an amanticipated trip of both main feedwa-ter pumps from is11 power at t = 0 seconds. It was assmed that the AFW l system would fail to start following AFAS. At 1200 seconds (20 minutes)
} into the transient, APW was recovered to both SGs at its prescribed naziosa l i flow rate of 25 kg/s (400 sym). Furthermore, it was assumed that the
! operator would secure AFW to both SGs 3 minutes after the marrow-range i
level indication la either SG reached the +50-in. high-level alare. De , j). ' downoomer temperature and pressure profiles are shows la Figures 4.22 and
- j j 4.23.
I
l CAK RIDGE NATIONAL LABORATORY [3',d','C' ' ,",' ,37,3, cesaarto sv ua=rn unama t%enov systrus ac. . Review Document Number 5 TO: Distribution
SUBJECT:
Review of Calvert Cliffs PTS Report Chapter 4 - Thermal Hydraulic Analysis of Potential Overcooling Sequences Please find enclosed a copy of Chapter 4 of the Calvert Cliffs Unit 1 i PTS report. This material has now been reviewed internally and cleared as draft information for comment. Please note the following icens about the appendices:
- 1. Chapter 4 makes reference to five appendices (F, G, H, I, and J).
i
- 2. Appendices F and G are separate reports written by LANL and BNL respectively. These reports have previously been issued (or very soon will be) as draft reports for review. Therefore they are not included in this mailing.
! e j 3. Appendices H and I are in preparation and will be sent as soon as possible. 1 4 Appendix J is enclosed in an c3rly draf t format. Some editing requirements have already been identified, but the appendix has been included as-is to allow an initial technical review. Please review and return comments on Chapter 4 Sy Friday, June 29. If i additional time is needed for review, please let me know. D. L. Selby , Engr. Phys. & Math. Div. ! DLS:nc Enclosure Distribution G. F. Flanagan (w/o encl) C. E. Johnson, NRC
- S. M. Mirsky, BC&E D. A. Peck, CE -
T. G. Theofanous, Purdue U.
- - . - - _ . -. . _ - _ - - _ - _ - _ _ _ _ -. . ..~. . . .
, CC-4.44 . -- [
\
$90 , , , -s00
~
Ss0-1 _ f - S30- a - '500 t SCs dried (3 .. 4 out
- 500- gg - I 4 6
- 3
- 40 ,
Jo 470 - 1. Loss of main feedwater before reactor scram
- 2. Loss of main feedwater af ter reactor scram
{ 3 3. SCs dried out followed by AFW initiation 440' 4. Manteum AFW delivered
. 3. AFW off -300 y f
340 -
-:00 ,
! + 3S0- - 3n. .. .
- iOC 2o0 0 1000 2000 3000 4000 6000 6000 7000 s000 1
r,ne (e) Figure 4.22. Transient 10: Downcomer temperature dwing runaway auxiliary feedwa-ter to two SGs from fell power. (Note: This transient assumes MULTIPLE . i operator / equipment failures; see Section 4.1.10 for failure assumptions.) 1s . . . i 24s 0 is-
'IMM [hh - .
O2e f - 210 0 , / i : j 12 -- --050
- T 4 I,
10 - ,
--1400 $S l 1 C s- -
IJ c . . ,oso j i s- -
- 700 4- .
g-1 - 350 l i I O 3 ! 0 1000 2000 3000 4000 6000 s0C0 7000 5000 Tirne (s) Figure 4.23. Transient 10: Primary system pressure during runaway auxiliary feedwn. ter to two SGs from full power. (Note: This transient assumes MULTIPLE operator / equipment failures; see Section 4.1.10 for failure assumptions.) l
/
i i CC-4.45 L The first phase (0 - 34.7 seconds) shows a slight temperature increase i prior to the reactor / turbine trip at 34.7 seconds. This temperature incrasse was produced by'$he degradation of the heat-load capacity of each SG following the loss of main f'eedwater flow. Because the reactor power was programmed not to change during this interim, a not energy transfer of 0.9 GW-s into the primary fluid resulted, causing the primary temperature to increase a few degrees. The isitie.1 SG inventory of ~63,000 kg per SG
. was reduced by 30% during this period.
i N As previously mentioned, the reactor / turbine tripped at 34.7 seconds because of low SG narrow-range level indication. He primary liquid ten-perature quickly dropped to a quasi-static equilibrium temperature a few degrees above the secondary side liquid temperature (Phase 2). The decay heat produced by the reactor during this period was dissipated equally by l both SGs at a rate of ~40 MW per SG; this heat continued the baillas pro-i cess in each SG. This continued to deplete the liquid inventory in each SG and subsequently led to AFAS at 35.5 seconds. The average primary temperature during Phase 2 was higher than 552.6 I l (535'F), which caused both the ADVs and TBVs to be op n. Together, they
> vented all the steam that was being produced. After the SG liquid inven-tory was depleted, the heat-load .s, capacity of each SG decreased to less than s
1 MW. The decay heat produced by the reactor could no longer be dissipated i from the primary fluid. 1$te temperature began to rise sharply (Phase 3). This caused the ADVs an the TBVs to open fully, which caused each of the i SGs to depressurize. As a rescit, SGIS nccurred at 864 seconds. The MFRVs and MSIVs closed and isolated the SGs s from the TBVs.
,. ,-n .--,-,----w------ .-,-,e,.--..nw,,n,v,-, ._.n--,-n ,,.-,-.,-,.,.-e,,-,,-,,,.--.e,,---,.-n--,,,,--n-,,-.-e--,.~,.-,--
\CC-4.46
\
i At 1200 seconds the AFW flow was recovered. The initial surge of cold aux-iliary feedwater to enter the SGs vaporized rapidly. This removed 15.4 GW-l s of energy fram the primary fluid over the next 300 seconds. The injec-tion of cold charging flow over the same period of time resulted in a further decrease in the temperature of the primary fluid. The not result , I was a rapid temperature decrease of 22.5 I (40.5*F). The average primary ) temperature dropped below 552.6 K (535*F), causing the control system to close the ADVs. This bottled up both SGs for the remainder of the tran-sient. The continued addition of cold AFW.to both SGs resulted in each SG I removing energy from the primary fluid at an average rate of ~19 MW. This energy did not boil the auxiliary feedwater. Rather, the energy was added assensibleheattotheliquid,causingitstenhoraturetoincrease. The increase in the secondary side liquid temperature, however, occurred for J .- only a short period of time. The secondary side liquid temperature peaked I at ~540 K (512*F) at ~1600 seconds. The rate at which energy was being i
) added to the secondary side liquid as sensible heat was offset at this time I
by the continued addition of cold AFW. The not result was an increasing I liquid inventory in each SG with a modestly decreasins liquid temperature. On an average, the primary fluid temperature decreased at a rate of ~32 K/h (58'F/h) over Phase 4 because of convective cooling. Had the ope'rator throttled the charging flow at the time of level recovery in the pressur-izer, the primary liquid temperature would have remained constant during Phase 4. l At 6590 seconds, the AFf system had refilled the SGs to the +50-in. level. ,i Per the transient specifications, the operator turned off the AFW system 3
- - - + - . ~ . - . . , , _ , __,r.._. --,.,,.-..,,-,,r -
,,,.%w. ,,,_-,,.,.-,,m_._wy,,y,-- -. -., 3,..,-w-w---,_y.y--,--,---m.- ,g--- rw, 3-,, _-.-- r
_ - . . _ . - - .=. - -- (CC-4.47 minutes later. The energy that was being dissipated through the SGs began to heat up the liquid in the SGs. As the secondary side temperature increased, the heat-transfer rate to the SGs decreased. The decay heat from the core finally exceeded both the rate at which energy was being removed via the steam generators and the convective energy transfers asso-ciated with the charging flow. The primary fluid began to heat again (Phase 5) 100 seconds af ter the operator turned off the AFW. 1 4.3.4. Small-Break LOCA Events i In the absence of SIS flow, the depressurization caused by a LOCA will cause the primary system to follow the saturation curve - a condition that is not likely to induce PTS. The break must be large enough to depressur-ize the system to the SIAS setpoint if it is to generate PTS. However, if the break is too large, the rate of_depressurization will be sufficient to maintain a pressure-temperature relationship close to the saturation curve despite the effect of the cold SIS water. Because the HPI flow rate is strictly a function of system pressure, reasoning suggests that the threat of PTS will be increased by any mechanism that localizes and concentrates the effect of the HPI water in the vicinity of~the critical vessel welds. One such mechanism is loop stagnation. Loop stagnation not only localizes the HPI effect along the downconer wall by promoting stratification in the cold legs, it also inhibits reverse heat transfer from the hot SGo that would mitigate the effect of the HPI. Consequently, there is some concern that certain break sizes may generate conditions conducive to loop stagna-tion yet limit depressurization sufficiently to cause PTS.
CC-4.48 l To address this concern, two small-break LOCA transients were selected for investigation. The first was a small hot-leg break LOCA with a break size of ~0.002 m (0.02 ft 2) in the range suspected of causing loop stagnation. For that calculation the full-power model was modified to include a break in the hot leg of Loop A with a prescribed pressure boundary condition of 0.1 MPa (1.0 sta). He second transient was a small small-break LOCA hav-ing a size of 0.001 m2 (0.01 f t2 ) and caused by the failure of one of the e two PORVs to close fully. In addition, it was assumed that the SG A ADV l failed to close when it should have. These two transients are described in the following sections. 4.3.4.1. Transient 11: 0.002-m2 Hot-Leg Break from Fall Power + j ,- j,- g The downconer temperature and pressure curves for Transient 11 are shown in x . . . ,
/-/
- Figures 4.24 and 4.25. The analysis of these curves can be divided into y
two phases. The first phase was characterized by a rapid depressurization of the primary that was halted by flashing in the upper head of the vessel i I at 110 seconds. During this phase of the accident the energetics were dom-insted by overcooling by the SGs following the reactor trip. Heat rejec-tion to the SGs decreased rapidly with the loss of forced convection fol-I lowing the RCP trip, however, and by the end of this phase of the accident, energy removal by the SGs was almost 90% completed. The second phase (-110 - 6636 seconds) was characterized by the emergence of an approxinate balance between the mass discharge rate from the hot-leg break and the SIS injection rate, a gradual decrease in primary pressure and temperature, and extensive voiding in the upper plenum. At 502 l . _ .- _ . - - _ . - _ . - - , . . _ , _ - _ , _ - _ _ - ~ , _ . ~ . . - _ , _ , , . . . . . . - _ _ . _ . _ _ . _ _ . _ . . .
- e CC-4.49 .
I
$90 -
-600 560- -
333_- _-500 500- - g
- - 400 (
) 40- -
g , (
! . )
f 440- - 4 - -300 3 4t0 - {
=
l ,g
, x 380- ,
- 200 350 - ,
320 -
- -10 0 290 . . . .
0 1000 2000 3000 4000 5000 5000 7000 8000 re (s) Figure 4.24. Transient 11: Downcomer temperature during 0.002-m2 hot-leg break from full power. (Note: This transient assumes MULTIPLE operator / equipment failures; see Section 4.1.10 for failure assumptions.) : 1s .
- - 2450 16 -
- 210 0
~
. 4. L f 12 -- -- riso l
2 ? I-
'0 - -
'400 5.
i! B s- - 5 -1050 [ s- -
'00 '
4-
,-350 0 . . . .
0 0 1000 2000 3000 4000 5000 6000 7000 8000 Tee (s) Figure 4.25. Transient 11: Primary system pressure during 0.002-m2 hot-leg break from full power. (Note: This transient assumes MULTIPLE operator / equipment failures;
- see Section 4.1.10 for failure assumptions.) ,
1 l l .
/: g i CC-4.50
\. I seconds, SGIS was predicted to occur based on an auxiliary calculation presented in Appendix F. This analytical calculation was necessary because the TRAC model did not include the containment. The decrease in pressure and temperature during this phase was attributed to the gradual but per-sistent decline in primary energy resulting from replacement of the hot fluid issuing from the break with cold SIS water.
An interesting feature of this phase of the calculation was the non-equilibrium between the steam in the upper plenum and the water beneath it. The TRAC non-equilibrium condensation model predicted that conditions at the liquid-vapor interface were not conducive to rapid phase change; hence, condensation could not cool the vapor as quickly as HPI flow cooled the liquid. Another interesting feature of this phase of the calculation was the reduc-tion in the loop flows that culminated in flow stagnation in Loop A at
~6500 seconds. After the ADVs closed at 968 seconds, the SG could no longer reject heat to the atmospheres hence, the primary temperature fell below the secondary temperature. The resulting reverse heat transfer j cooled the secondary, but it also retarded natural circulation in both loops. The reverse heat transfer and reduced flew downstream of the hot-les break caused voiding in the top of the U-tubes in the Loop A SG at l
~6300 seconds. This voiding caused the stagnation that occurred about 200 seconds later.
(
)
,y CC-4.51 1
4.3.4.2. Transient 12: Stuck-Open Pressurizer PORY with Stuck-Open Secondary ADV from Full Power The downconer temperature and pressare profiles for this transient are i
/ b ,<! shown in Figures 4.26 and 4.27. As in the previous case, this transient uf. # can be characterized by two phases. The first is distinguished by a rapid
/4 depressurization of the primary that was halted by flashing in the upper j head of the vessel at ~210 seconds. During this phase of the accident, the
, energetice were dominated by overcooling by the SGs following the reactor i trip. Heat rejection to the SGs decreased rapidly with the loss of forced convection following the RCP trip, however. The sdcond region (~210 - 7200 seconds) is characterized by a gradual decrease in primary pressure and temperature, stagnation in Loop B result-ing from overcooling by the Loop-A SG, and complete refilling of the pri-mary by the SIS. Most of the decrease in primary temperattre can be attri-bated to fluid exchange between the SIS and PORY discharge with the balance I of the decrease being caused by continued heat rejection through the stuck-open ADV (Loop A). l } Furthermore, the stuck-open ADV was responsible for the stagnation that i occurred in Loop B. Following SGIS, SG B could no longer reject heat to j the atmosphere, and Loop B lost the density head through the U-tubes that helped to drive natural-circulation flow. Because the ADV near SG A was stuck open, however, SGIS did not isolate SG A and it continued to depres- ! surize. In fact, the steam flow out of SG A essentially doubled following l SGIS because the total flow out the ADY did not change but the flow from
...,.-,.r. - , , - , , _ _ , _ _ , _ , , , . , . ,,, __,,, - _ , , , .y,_w,., ._.,_,.,,-m, - - _ - _ _ . _ _ -- . - - . . ,--c._-,, . -.,w. , , _ . , _ . . _
. . tCC-4.52 -
', , j
\
i 590 -
. . -600 560- -
$30_- --500 s
sco- - 7 470-
-400 7g 3
i ~ e g 440- { 3 -s00 t y 4t0- - - g g 330 -
- 200
- sso- }
s20- - -
- 10 0 290 . . . - . .
0 1000 2000 J000 4000 6000 6000 7000 8000 5 Time (s) Figure 4.26. Transient 12: Downcomer temperature during break from a stuck-open PORY plus a stuck-open ADV from full power. (Note: This transient assumes MULTIPLE operator / equipment failures; see Section 4.1.10 for failure assumptions.) 18 . .
- 2450 16 -
210 0 l 12 -- -- 175 0 , 7 ? y 10 -
--1400 O C 8- -
E -
-1050 ,
[ _8- r~' -
- 700 4- -
l
- 1so
- 0. . . .0 0 1000 2000 3000 4000 $000 60C0 7000 80C0 Tu11e (s)
I Figure 4.27. Transient 12: Primary system pressure during break from a stuck-open i PORV plus a stuck-open ADV from full power. (Note: This transient assumes MULTIPLE operator / equipment failures; see Section 4.1.10 for failure. assumptions.) d . e-1
I t
./ r.g -
CC-4.53 SG B was terminated. Consequently, the heat transfer to SG A p actically doubled following SGIS and the increased heat transfer enhanced the density head in Loop-A. The primary temperature decreased throughout the transient and the downconer temperature had f allen to 425 K (8306*F) by 7200 seconds. e
, , _ . . - . _ . - . , , ,_.,,.#., ,,_ , , ,.__,,, ..... ,_~...-_, - , - > ~ ._ _ . ,
CC-4.54 i i 4.4. Downconer Fluid Mixinn Behawlor A review of many of the transients perceived for Calvert Cliffs Unit i revealed several instances in which the flow in one or more cold-leg pipes was very small. This could lead to a stratification phenomenon which would produce localized vessel wall temperatures in the downconer region that are 4 significantly lower than the bulk fluid temperature as calculated by 'IltAC. a As a result, it was necessary to evaluate this phenomenon and evaluate its l potential effect. Three sets of analysis were performed to quantify the effects of partisi or total loop flow stagnation. The first analysis, discussed in Section 4.4.1, was performed at Purdue University by T. G. Theofanons. This analysis involved an evaluation of the 12 LANL calculations to identify the potential for and the effects of stratification phenomena associated with those transients. In addition to the above analysis, an evaluation of the mixing phenomena associated with the LANL transients was performed at LANL by B. J. Daly using the SOLA-PTS mixing code as discussed in Section 4.4.2. , Finally, Theofamous was asked to calculate the downconer temperature pro- ! files associated with total loop flow stagnation. This information was necessary for evaluation of those sequences for which stagnation was assumed. The results of this analysis are presented in Section 4.4.3. 4.4.1. Stratification Analysis of Twelve LANL Transients l This evaluation was performed utilizing a stratification criteria screening process and a regional mixing model (RMM) which had been benchmarked l {
CC-4.55 i against experiments carried out in a 1/2-scale facility rigged up to the Calvert Cliffs-1 injection geometry. A sammary of the results of this evaluation is presented in this section and the detailed evaluation process and results are presented in Appendix H. The initial stratification criteria screening identified three transients (Transients 1, 4, and 12) as requiring further analysis. The RMM was then used to evaluate these three transients. The results showed that for Tran-sient 1, high pressure injection occurs for the first -1,000 seconds. i Loops Al and A2 run well-mixed at strong natural-circulation rates and cool i rapidly in the 400 - 425 K range. Loop B2 goes into momentary stagnation (and stratification) at ~500 seconds and reverses flow for the next 2,500
- seconds. loop B1 exhibits two stratification periods of ~250 seconds each around ~500 and -1,000 seconds respectively. The possible effect of such i short-duration stratification was determined by running the RMM calculation
- for the cold-leg / pump / loop seal system. The results are shown in Fig.
1 4.28, along with the TRAC temperature traces for loops Al and B1. It is apparent that loop A1 (and hence the downconer and lower plenum) cool much faster than the stagnated loop B1. Note that the " cold stream" in B1 (Ble ,) is warmer than the Al outflow for the duration of the stratified condition. In fact, this is the reason for the choice of the mixing con-trol volume as indicated above. It can be concluded that downconer ten- ! i peratures will be dominated by loop Al and A2 flows and their temperatures even for the period of stratification in loops B1 and B2. l The characteristics of Transient 4 are very similar to those of Transient 1 1, with one addition. Here loops B1 and B2 both exhibit back-flow at ~750
CC-4.56
\
i i l l 600 - - 620.6 550 - 530.6 E i . E -\ _ C g i o .v \ m u 500 -\ - Ex 's 440.6 e $ o s < Q - Bj (TRAC) - y Ei \ 9: 4 450 - s
's 350.6 b$
'H ',, .
,,,,,, Bics (RMM) 400 -
Aj (TRAC) 260.6 ,
= r Stagnation period 350 17 0.6 0 500 1000 1500 TIME (SEC)
--- ==
Figure 4.28. Transient 1: Downconur fluid mixlag behavior. B1, stagnated; Al, circulating; therefore, downcomer well mixed, forced flow. s
o . CC-4.57 l seconds, which is slow enough to establish a relatively low temperstare condition before a stagnatica condition for 750 to 1000 seconds is 4 obtained. The possible effects of this stratification, i.e., any addi-
} E 6,k tional cooling, was also determined with an EMM caloslation with an initial V
" ambient " temperatare of 375 E. The results are shown in Fig. 4.29. Here 1 /
I- the cold stream is only ~30 K cooler than the downcomer tamperature (Al 1 outflow). This, plus the strong flows in the downconer from loops Al and A2, indicate that any additional cooling effect due to stratifloation in i loop B2 would be negligible.
- In Transient 12, loops Al and A2 again remain at well-sized conditions, J
! with strong natural circulation. I4 ops B1 and B2 stagnate for times beyond 2,000 seconds under HPI of ~10 kg/s. The effects of the resulting stratif-l ication were scoped by assuming that the strong A1 and A2 loop flows estab-lish the downoomer temperstare history. With this taken as the " ambient"
! - in the RMM calculation, a cold stream temperature in the B1 (and B2) cold
; }
;. legs was obtained as shown in Fig. 4.30. The modest degree of stratifica-
- l. tion seen (~30 K) is the resnit of the strong mixing within the injection line un fee the prevailing low injection Froude Numbers (Fr ~0.2) . This mixing was determined experimentally in our 1/2-scale facility and found to be considerably higher than that observed at Fr ~0.6, which was examined earlier in connection with Westinghouse reactors. The resulting "pinnes" I
into the dowmooner would be extremelv 21Ak nader these conditions and wonid six quickly with the Al and A2 loop flows, which hence will dominate the 1 dowmooner response. l l l l In ocacission, it was determined that, at least for the types of transients l l
CC-4.58
/
450 , , , , , , 350.6 g g g g 305.6 E.-- 4 0 0 - - 260.6 Ei b y AI (TRAC) v -
\ ',,,.
g - 215.6 D Q Q rg B2 (TRAC) d a,, W A :s CL.
' N
,@ 350 - o -
17 0.6 f B2cs (RM) 125.6
' ! ' I ' I ' I ' I '
300 80.6 750 800 850 900 950 1000 1050 TIME (SEC) 1 Figure 4.29. Transient 4: Downcomer fluid mixing behavior. Al, A2, circulating; B1, B2, stagnated; therefore, downcomer well mixed, forced flow. e _ . _ - --r
1 CC-4.59 600 i i i i 620.6 g i l l Frinj = 0.2
' "' " " "I * '" " I "
550 - 530.6 . p - b C y 500 - 44o 8 6
,yo Downcomer(TRAC) $
_ Taken as T,g - g Q m ,,%~'., ' m w M 450 - 350.6 $ g ,, g w
,,, e b -
B '~' ics and B2cs ( ) - - -
~
'~ --- - ,___
400 - aso.s : I 350 ' l ' I i l i I ' i r70.6 2000 3000 4000 5000 8000 7000 TIME (SEC) 9 Figure 4.30. Transient 12: Downcomer fluid mixing behavior. Al, A2, circulating; B1, B2, stagnated; therefore, downcomer well-mixed, forced flow. - 1
CC-4.60 covered by the 12 LANL transients, stratification phenomena are of no PTS significance for the Calvert Cliffs-1 reactor. 4.4.2. SOLA-PTS Mixing Analysis of Selected Transients A mixing analysis was performed at LANL for those transients for which mix-ing was considered to be important. A separate reroet that documents the results of this analysis is included here as Appendix I. The conclusions of this analysis were very similar to those obtained by Theofanons. One exception was that for some transients, a very narrow but strong thermal plume was established below the broken loop cold leg. How-ever, as will be discussed in Chapter 5, this narrow plume was not con-sidered to have an impact on the vessel welds of interest. Thus the con-clasion of this SOLA-PTS analysis was that the TRAC bulk temperature vaines I were appropriate for use in the fracturo mechanics analysis. 4.4.3. Total Loop Flow Stagnation In performing the thermal-hydraulic analysis for Calvert Cliffs-1, it should be noted that no calculations performed predicted total loop flow stagnation. However, it is also clear that situations exist for which the potential for total loop flow stagnation is greatly increased. As stated earlier in this report, total loop flow stagnation, or at least very low flow, has been assumed for these cases. Theofanous was asked to determine the temperature profiles in the downconer region under the assnaption of total loop flow stagnation.
f CC-4.61
.I l Two temperature regions were identified. The first region includes the initial planar pinne exiting the sold leg. This plume area covered a vert-leal strip in the dowmoomer that was two cold-leg diameters wide and about five cold-leg diameters loag. The second region laolades everything out-4 side the pinne region and is called the well-mixed region. De tempera-g .-
tures associated with each region are shown in-Fig. 4.31. These are the
}t \ /
- temperature profiles used to analyse those sequenoes for which stagnation is assumed.
4 e 4 J d
- , - -- - , - - - , - - -- ..--- , . - - - , _ . - , , , -- , - . - - - ~ , . . - . , . . - . . . -
i l f [ l l em , g
, , , , , , , , 820.8 g , , , g ;
~
Well Mixed Temperature - l Initial Planar Plume Temperature 500 \, 440.6 l ,. g-
..,,,' g-400 . -
260.6 v - ,, r4 _____ ei a: _.
- ___ ~_
g ... , _ l 5 300 m 80.8 .2 m $. s-
*M -
200 -
-99.4 E i I i l i i i I i i i l i I ' - m.4 0 1000 2000 3000 4000 5000 6000 7000 0000 nus (sec) .
n ome us. Tm p e.eu s th he, a - sensmessen m
* ' ' " " - - - " - + = -.
O
t
. CC-4.63 i
.l \_. 4.5. Heat-Tramafar Caaffinlant Evaluation A time-dependent heat-transfer coeffleient was calostated by 1EAC for the fluid film oosdition associated with daek of the transients salesisted by LANL. n e fluid film and the vessel wall eoastitute two therms 1 resistaneos la 1 series. Thus the " total" sondsetivity is 1 h~ p . 6. 6 , 4 I I f s b I 4 i l where kg = thereal condustanee of fluid film, k, = thermal condsotivity of claddias, I k b
= thermal conductivity of base material, j Ar, = thickness of eladdias.
Ar b = effective thiekness of base material (time dependent). 1 Then the resistanee of the fluid film (1/k ) gis small compared to the I resistanee of the vessel wall (Ar,/k, + Arb /Ib), the fluid-film eondse-tivity has 11ttle effect on heat removal from the well. For instance, kg= 1000 Bts/href 4 2 ,,F (pamps on) is a "1erge" value, and but even larger values (somentary boillag) have little effect on the severity of the tram-sieet.
i l CC-4.64 When the resistance of the fluid film is large (small value of h g, such as 100 Bta/hr.f t * *F), the film resistance is dominant.' As hg approaches sero, the potential for vessel failure disappears. I i Plots of the heat-transfer coefficient caloslated by TRAC for each tran-i sient are presented in the LANL report (Appendiz F); however, it was l discovered after all the transients had been run that TRAC was not cales-l lating the dowmooner heat-transfer coefficient correctly. For the two-1 l dimensional flow field that occurs in the vessel dowacomer (asianthal and vertical flow), the magnitude of the velocity ze.glaK should have been used i i to evaluate the Nasselt number in each of the fluid cells in the dowmooner annulas. However, boonase of an error, only the vertical component of the ) velocity was considered. In transients in which one loop stagnates and the i -i other loop is flowing, significant animathal flows occurred in the dowmoo-l mer annulas. In cells in which the velocity component in the asinathal a f direction is large and the ve10eity component in the vertical direction is ' small, the Nasselt number was underestimated and a natural circulation flow regime was predicted. Consequently, the heat-transfer coeffiolents for those cells were underestimated. i i i Because of this error, the 11AC-oalculated heat-transfer coefficients were modified for use la the frasture neokantes analysis. In the modifloation i
!, the lattisi drop in the fluid film heat-transfer coefficient was not changed since it was felt that the TRAC ealoslation for this time frame was quite adequate. For the remainder of the analysis time, it was assumed that the minimum heat-transfer coefficient was 400 Bta/href2 t ,,,, gg, i
i value was chosen for two reasons: (1) After a review of the TRAC l
- - , _ . - - _ _ _ . . _ - - ~ - - - - - - - - - _ , - . . _ , . _ _ . _ _ . . _ . - _ - . _ , - . _ _ . - _ ,----,_ -_.-- - ,
i CC-4.65 i . calculations, it appeared th at the heat-transfer coefficient would stabil-ize in the range of i 100 Btu /hr.ft **F of this"value, and (2) the minimum value is large enough so that the total heat transport is not significantly sensitive to the value of the fluid film heat-transfer coefficient (i.e., i it is much larger than 100 Btu /hr.ft *F). I As the fracture-mechanics calculations progressed, Theofanous was asked to review this assumption. He used the TRAC velocity histories to calculate fluid film heat-transfer coefficients. A discussion of the complete analysis is included Appendiz H. In general, it was determined that typi- ! cally the forced convection augmentation was overshadowed by the corresponding reduction in the forced convection component (as the velocity
- decreased) such that the resulting spread in heat-transfer coefficients was much smaller than the variation in the individual " free" or " forced" con-vection components. The variation in calculated wall temperatures was even smaller.
1 i II l ;- The calculated fluid film heat-transfer coefficients are shown in Table 4.4 i for all 12 LANL transients. As shown, the coefficients are almost all covered by the 400 1 100 Btu /hr ft 2 .*F range. Thus it was concluded that l the original assumption was valid. e l p---. _. ,.,., . . . . , , - -.-.-.-wy- -.--, -- .,-e-.-- .,-----.y--,,,_. -,.-,.,,--,.,---,r,,.-,- -
-y-., +,. ,- ----,
\CC-4.66 i g
Table 4.4. Finid film heat-transfer coefficients for twelve LANL transients Fraction Fractics LANL of h of h Transient h*mixeg from Forced From Free Number NU/NU,e (BTTI/hreft **F) Convection Convection 1 1.12 330 0.53 0.47 2 1.00 454 0.98 0.02 3** 4 1.00 365 0.55 0.45 5 1.20 345 0.40 0.60 6 1.00 510 0.98 0.02 7 1.00 480 0.91 0.09 8 1.00 46 0 0.98 0.02 9 1.00 590 1.00 0.00 10 1.00 500 0.59 0.41 11 1.00 515 0.90 0.10 . 12 1.03 477 0.85 0.15
- Based on maximum velocity in downconer region at 2000 seconds for each transient.
**With two reactor coolant pumps in operation throughout the transient period, the heat transfer coefficient is assumed to be very large.
i t l l I i l I
. , - - - - -,,,.v..,-,,--...,-._,.-.--.,.,-----,.r._,,,,...--
_,._ - - , , , , , , - - , , . , , . - , , , , , . ..sn, . - .- . ... , , . ..,.g. .-
CC-4.67 4.6. Esti==tions of Pressure. Tennerature and Heat-Transfer Coeffielent Profiles , l 1 l l The evaluation of the risks of pressurized thermal shock (PTS) entails the coupling of overcooling incident event trees to fracture-mechanics calcula-tions of the probability of vessel crack propagation. The link between an event tree end state and the fracture-mechanics calculation is the tran-sient behavior of the pressure (P), temperature (T), and heat-transfer coefficient (h) in the reactor vessel downcomer region. That is, the P, T, and h transient profiles from the sequence defined by an event tree end i
- state become inputs for the fracture-mechanics calculation.
i I There are potentially several million end states produced from overcooling 4 transient event trees and the cost and complexity of thermal-hydraulics and fracture-mechanics calculations preclude the evaluation of every end state f
- separately. Therefore, it becomes necessary to (a) reduce by similarity grouping the number of end states to be evaluated and (b) reduce the number of detailed thermal-hydraulic calculations to be performed through the use of less rigorous estimation techniques. This section summarizes the 1
approach used to group the sequences and estimate P, T, and h profiles for i the Calvert C11ffs-1 PTS study. Section 4.6.1 describes the estimation j methodology developed for the study and the approach and rationale for sequence grouping, and Section 4.6.2 summarises the results of evaluations for each of the major initiating eventet
a - - ..i.__- - - 4 _ _r . .-- a. _ _m A CC-4.68
- 1. large main steam-line break at BZP,
- 2. Small main steam-line break at BZP,
- 3. Large main steam-line break at full power, l
- 4. Small main steam-line break at full power, I
- 5. Reactor trip,
] i 6. Small-break LOCA (10.016 ft2 ),
- 7. Small-break LOCA (4 .02 ft2 ),
- 8. IDCAs with potential loop flow stagnation, and i
- 9. loss of main feedwater with subsequent AFW overfeed.
l l The last two categories involved sequences for which P, T, and h values i were determined in earlier sections of this chapter and thus they are not discussed in this section. The estimated P, T, and h transient profiles presented here are based on the IRAC-PF1 calculations reported by Los Alamos National Laboratory and j described in Section 4.2. Computer tapes of 11 TAC plot output files for l these calculated transients were also employed in the development of param-eters applied to the temperature and pressure estimation procedures.
o . CC-4.69 The sole and extensive use of these TRAC calculations in estimating the P, l T, and h profiles for the variosa sequences implies that the estimations are subject to the same modeling assumptions and code characteristics driv-ing the uncertainties in the TRAC-onleslated results. Additional uncer-tainties introduced by the estimation procedure have not been fully evaluated. Such uncertainties were minimized by using the estimation pro-cedure to dupliosto portions of the transients calculated by TRAC and thereby check the validity of the assumed parameters and extrapolation models. i The estimated P, T, and h profiles presented in this report represent a
" single point" estimate of dowmooner conditions. That is, the estimated i
conditions are assumed to hold for the entire dowsooner region without any i azimuthat or szlal variations. Sinoe the detailed TRAC calculations demon-strated both azimuthal and axial variation in fluid temperatures and heat-l 1 l transfer coefficients, the cooldown model used in the estimation procedure 1 l was conservatively set up to yield the espected temperature of the coldest subregion of the dowmooner rather than the overall average temperature for the whole downoomer region. l l ! 4.6.1. Methodology i i 4.6.1.1. General Approach l I i Af ter sa initial survey of the data resources and the sequences identified for estimation, the five-step process depicted in Figure 4.32 was employed in estimating the Calvert Cliffs-1 P, T, sad h profiles. This approach
. .. , _ . ._ _ , ,-, .-,--_._3..~-.- .
, y-, . , - . - - , , _ . , , . _ , _ ,
..__,.-,--_.u_.___-,__.__,.,_,,_,-----m- , . . , ,
_-_ . .~ . - . . _ . - _ . . .. ._ _-_- -_ - _ - _ _ . .-- , _ _ . l i
*l
~ ~
Resource Data t 12 e ac.amataa. by LA84L (TRAC) **P
- s.ea 4 Detesadne applicalde Check consistency of -
stease. fine besaks - TRAC casestentsact : ; pasaenesess by TBW Iailuse . selevant pasaanetess. duplicating TRAC pOftVLOCA sesults in needlesse besak LOCAL 4 h4 c008d**e8m*8884 BAFW everteed EFW ever feed < ' Less of lea:s ORNL SpecNied E:;r:nres Teteloverles cases ,
- g Steaan Ilme beeaks:
l L::;: i::" at inet rose power I~ l Senameseek at het asse power 588P 1 S88P 4 Step 5 Lasse bseak at lislI p - _
- -+ n ,-
"{
+
Sane 864 seek at luu power Geetspepecialed Evainsate ,,, seepseseces by sannitasily
- A. Tessperatuses by .
Tambine esips: "
;"::: :r selection of
- Ovesteeds TRAC cesswes and esse of TSW ADV talluses go gsg ,,, ,,,ogog, -
i ~ SenaHM LOCA:s R. Pressenes by ;"::: *:- s selection el TRAC caerves - er by coolant sweN . l calcesiations . C. Heat Trans8er Coedlicient ! by ;"::: *:r selection of l TRAC cusses and cesrected Ilsniting vatise. Fiysse 4.32. P.T. and b h appseech. l I l
._ _ - _ ~ .__,. _ - _ , _ - - _ . - _ _ _ - _ _ _ - . _ _ _ - . - . . --.-- -----_ _ _ _ _ _ _ - _ _ _ _ - _ _
t o - } i CC-4.71 I l i allowed logical redsettom of the number of esses to be evaluated and i 4 derived the greatest benefit free the information in the M AC ealestations. l l l L
! The first step involved the grouping of siellar sequesses withis eseh tras- i sient initiator table. An evaluation of the MAC eelestations for the t effects of different operatias states provided the eriteria for ass 16 ament l 7
of sequesses lato groups. In step 2 the parameters were developed for the ! l i eooldown (temperature) and coolant swell (pressure) models used os oeessies for this study. To sessee correct laterpretaties of conditione daring lj e j esquenses, the appropriate parameters were applied to the eeoideva model to l dupliente portions of seguences e.lestated by TRAC. 7tese ve11dation I l
} efforts took place la step 3 (see Seetten 4.6.1.3).
i I f j In step 4 the prosesse, temperature, and heat-transfer coeffielents were j estimated. Temperstere ess1d be estimated by piecewise application of MAC reesite and/or by salestaties seing the eeeldown model. He method selee- [ j tion depended on the semplealty of the segsease and the availability of { l applicable data from the MAC ealestations. Early pertiene of many l I evalsated seguences had stated configurations identical to these of a par- ( j [ i tiestar MAC salestaties, so piesowise see of the MAC results was applied. ! I no eeoideem model was then seed to evaluate the remainder of the transiest i f out to 2 hours. Certain said (i.e., high-temperature) transients were not ! esplicitly eve 1 sated, nees mild sogneases were as,igned the P, 7, and k i i i profiles of a MAC saleslettaa er the estimat' de segsense whlek meet closely represented the setteipened reopesse of the sequesee. Presense estimates were lerived free observation of pressure trende in the i l
0
- I I
. i 6
, CC-4.72 l i
i t r j MAC ealesistions and by a preessee prediation model (the eeelaat swell ' l 4
! model). The Calvert Clif fe-1 plant features a low-head EPSI system which re..... .. tk. pe - r , a..v. tk. . . p ah.t-off h..d .t ,,,, ,ais. ;
] .. .t j he chargias pumpe saa represessise the primary up to the p0RV setpoint i ( (2400 peia), but does se at a very low rato due to low flew espeetty. He j shargias pumps were set throttled la say of the MAC esistiations. Here-fere, there are a amber of esses available for evalsation of the contribs-j ties of the ehargias pumps to system represessination. De ecelant swell i ; model accesats for pressere effects due to ecolant espassion es reheating. ! l his model is seed for evaluation la these esses where charsias peep flev ( l is throttled. t r West-tressfer eeef fieleats were based en the piesevise selection of DAC l data and the reesits of modeling performed by heefasess at Purdse Univer- I sity (see Section 4.5). In general, the M AC estestations prediated rela-I tively sesotaat values while the reseter see11ag pumps (RCP) are reasies fi and a step ahange to a lower but searly seastsat valse after RCP trip and establishment of satsral strestation. Dee to problems la the M AC heat- ] l transfer regime seleetles teste, the TRAC values were systematica11y seder-1 l prediated. Deefasess feesd that the sentributica of free soavoeties to [ J the dewseener heat-transfer seef fielest offset ineresses er deeresses la l s 2 ,,F l I forced eeavesties snea that a total valse of 400 5ts/href t (1270 W/m2 ,g) ,,, ,,g,g,g,,4 ,,,, , ,gg, ,,,,, ,g ,,,,,,g.,g,,,g,gg,,gg,, ,,,gg. i tiens. He sequesee evaluatione presented la this seetles see a sempesite ' i of MAC-ealestated heat-transfer esof fleients for pre-RCP trip regimes and eerrested estimates for natural-eirestaties regimes. I i 4 l
l i
, CC-4.73 i
l The completed estimations were documented in step 5. This documentation 1 1 comprises Sections J.3 - J.8 of Appendix J. 4.6.1.2. Sequence Grouping When all PTS initiators and failure branches are set up in event trees, several thousand end states result. To obtain a tractable yet representa-tive set of PTS transients requires some method of sequence grouping. Chapter 3 describes the construction of the event trees and the process used to eliminate "non-contribution" states (i.e., component failures made irrelevant by the action of othsr systems or components). The collapsed event trees from this process still contain a large number of end states. Section 3.5 describes the screening process used to separate end states into a set of discrete sequences for evaluation and a set of residual sequences for which no farther evaluation was performed. Sequences representing identical combinations of failures were collapsed to a single group and the corresponding frequencies were summed. Sequences with fre-quencies between 10-7 and 10~8 per year, which would normally fall into a residual group, were examined for similarity with the discrete sequences and were collapsed together with specific discrete sequences where appropriate. This approach minimized the cumulative frequency of the resi-dual. Tha resulting set of discrete sequences are found in tables presented in Section 3.5. Altogether,115 sequences emerged fram this grouping process, including 11 i residual groups. The grouping processes of Chapter 3.0 were based on sys- l tem configuration and event frequency. Further grouping may occur based on i
l CC-4.74 the thermal-hydraulic impact of the configuration. The impact of a partic-ular component or system can be evaluated from observation and evaluation of the effects of its operation or failure in the TRAC calculations. In this way the importance of failures or actions could be classified as dom-inant, minor, or inconsequential. Sequences with the same dominant features were grouped together for analysis. In later stages, the influ-ence of minor events was evaluated to check the consistency of the group-ings. This checking accounted for the thermal-hydraulic interaction or feedback due to the combination of failures. Some sequences were reas-signed to other groups as a result of such checks. The groupings for each of the initiators are discussed in Appendix J. Sec-
'~
tions J.3 - J.9. 4.6.1.3. Temperatro Evaluation by Cooldown Model The temperature response of a transient is a function of the system's con-figuration during the sequence, including the timing of configuration changes (e.g., RCP trip; MSIV, MFIY closures; AFAS, etc.). The sequences 1 from the LANL TRAC calculations represent only 12 of the thousands of sequences on the overcooling event trees. The cooldown model is a means for applying the information generated by the IRAC calculations to other sequences requiring temperature response estimation. The approach used in the cooldown model was to obtain separate mass-energy balances around the steam generators and the reactor vessel (i.e., balance of the primary cool-I ing system) to predict the rate of temperature change. All pertinent cool-ing and heating mechanisms were included. In obtaining these mass-energy
._ __ _ ..~.r .. .- ,._ __,-. . _ . _ - - _ ._._.__,_m-.y--m.,% _ , _
- m. , _ , -y ..,.w, -
,w,,, , , , , -,-y, ,m -- -__,y
CC-4.75 balances, it was necessary to make the assumptions listed in Table 4.5 to simplify the system to a two-node model. The assumption of no steam generator heat-transfer resistance will result in the prediction of slightly lower primary temperatures than are reported by TRAC, the error being proportional to the rate of heat transfer. The error will be less than 10*F for large steam-line breaks (LANL transients 1-5) and less than 5'F for small steam-line breaks (LANL transients 6 and
- 7) under conditions where natural loop circulation prevails.
The assumption of thermal equilibrium in the steam generator secondary allows the use of simple choke flow models to predict steam flow rate. Conditions close to thermal equilibrium are obtained by TRAC for steam gen-erators during blowdown. Division of the reactor coolant system into only two nodes coupled with the assumption of perfect mixing within a node
" smears out" the temperature differences around a loop, thus losing ten-perature lag information available from a finely noded model such as that used in TRAC. Therefore, the cooldown model will respond faster to input parameter changes than will the TRAC model. Direct comparison of the cool-down model's extrapolated temperature response with TRAC resulte suggest that this effect is small for cases where natural loop circulation remains large (>500 lb/sec).
A final assumption that allows the use of the cooldown model is the assump-tion that TRAC-calculated mass flow data from the 12 LANL transients may be applied to the evaluation of other sequences. This assumption is necessary 1 because the mass flow information is required to implement the cooldown l
O Table 4.5. Cooldown model assumptions Assumotion Justifications Limitations NUkkkEEi!EiIk
- 1. No heat-transfer (HT) Large HT area; large HT loss of heat flow lags Simplifies calculation resistance between coefficient for boiling, and disequilibrium at expense of accounting primary and secondary condensation. Information. for SG primary temperature lag of 5-15'F.
- 2. SG secondaries in Same as for assumption 1 Not a good approximation Allows use of enthalpy thermal equilibrium plus good approximation where overfeed is transport model based for SG blowdown conditions. compressing steam in on choked flow pressure, isolated steam generator. enthalpy conditions.
l
- 3. Water inventory is Same as for assumption 1 Eliminates space-time Allows use of two-node well mixed within a plus natural circulation flow effects; difficult to mass-energy balance. n mode (onergy is is generally much larger quantify flow stagnation ?
, uniformly distributed) than HPI and secondary effects.
- flows, allowing equilibration y
, or approach thereto. l l l i i 4
CC-4.77 model calculated from a simple two-node s;odel. Engineering judgement is used to identify segments of the IRAC calculations relevant to the sequence being evaluated. Pertinent mass flow data are then extracted from the identified TRAC calculations for application to the cooldown model. The required parameters for the model are listed in the derivation of the model as described below. 4 Model Derivation and Characteristics. The cooldown model consists of two simultaneous nonlinear differential equations describing the mass-energy balance of a primary node (i.e., vessel, loop piping, and RC pumps) and a steam generator node as follows (see Figure 4.33): d(MU),,1 . (4,1) Hg + QD (t) + ORCP
- k ~ O sec
- dt " "HPI H HPI d(MU)gg .
(4.2) dt " "FW Hg - mST HST + G sec ' where H " Product of HPI mass flow and specific enthalpy at HPI HPI nominal temperature (THPI) vs. thermodynamic reference temperature (T,,g) C, (THPI-T,,g), I mgHg = product of primary leak flow (pressurizer surge line or break) and specific enthalpy at hot leg temperature (TH) l
= a gC p (TH-T,,g) (valid for liquid flow only),
Hg = product of feedwater mass flow and specific enthalpy at feedwater temperature (Tg)
"pg C, (Tg-T,,g),
..-. ---w -- w--y.--,wy , , , - .,,,,.,-w .-.m y 9 g-,,-, -p3 ,m---v-m4,y.,, ,,_-- .,y, .,nyyew.,
O . n7* .$ o l l l l iI l ll i l J
=
e u o a c c e o.
.M T e
. l ol8a4 o l n, e x d o
i, m m e l lI I I l I n w
, o d
O l o
. o
. ce c . d o t c -
o o T . m-o L S P , .. tw H ,, (
)
MT r
.f o
s u w T l o
. . f t
y
.M L J
g r e e n j e
,( n s
a d O - na s u _ s
.M T
, a s a o T M d
o
.M. T o % 3 3
n, 4 o e
- i. r
. u g
n - i
. F c
m i s , o
.msri
.e in se G
l l I liI l I l l l g s a ,. w s , MT .M T
. 1 i < , - 1
t CC-4.79 b ST Hg = product of secondary steam flow and specific enthalpy for saturated steam at steam generator conditions (TSG)
=
ST (AHv ,Tg )+Up (T3g-T,,g) , 03(t) = decay heat input as function of time
= ANS Decay Heat Function for transients from full power
= constant value for transients from hot standby, Q
RCP
= Pump power deposited in coolant Q
T
= heat transferred from vessel wall to coolant Q ,,, = heat transferred from primary to secondary.
In the absence of heat-transfer resistance, Q,,, is limited only by the transport of energy to the steam generator by the hot les flow (mH' ! Q ,,, = $5 C, (TH - TSGI
- i The lefthand sides of Equations 4.1 and 4.2 may be expanded by use of the chain rule Mdt=M $R+gA '
dt dt 1
- where
< M = total mass, U = specific energy = Cy (T - Ty g) , dU/dt = Cy (dT/dt), dM/dt = [E=massflowacrosssystemboundaries. Substituting into the lefthand sides of Equations 1 and 2, d(MU),,1 dTg (4,3) dt pri Uv dt v(H - T,,g) $ppy - my , y,y -- -r , , ,wri--------%---,----w. - ----g- ..,,,,we<w-w-w - - + - , .----,.ry,w- .w, e---w-%, -.,--...--a-,_ ,--aw ws+---.w -er==-,-e=
CC-4.80 dT (4,4) d(MU)SG gg + -m dt SG v dt v (TSG - T,,g) (app ST and then placing these expressions with their respective righthand sides yields dT pri Cy + C, (TH - T,,g) hg - Yy p(H
" U ~
I p HPI - ref - ref
+ QD It) + ORCP + OW - HU p (TH - SG) d for the primary node and M
SG Cy +C y (T ~ ( SG ref ~ "ST'
= Cp W pg - T,, g ) - mST y,T3g
*N
+ C, (TSG - T,,g)] + Eg C, (Tg-TSG) for the steam generator node. For liquids, C may y be assumed to be equal to Cp. Using this assumption and collecting common terms yields dTg _ kg,7C (T,,7-T,) QD (t) Q RCP OW a C,(T g-Tgg) 4.0 dt Mp ,gCp Mp ,gC , Mp ,gC , Mp ,gCp Mp ,gCp for the primary node and dT gg _ dp ,C,(Tg -Tgg) "STIUv,T gg ) a p(T H -T gg) pri p SGU p SGC p for the steam generator node. In this form, the thermodynamic reference state (T,,g) has been eliminated, leaving only the expressions for heating and cooling mechanisms.
=
- -n.- , .- . . . . , - . . .. _,
CC-4.81 i Flow rates for HPI, leak, hot leg and feedwater are independent parameters extracted or estimated from TRAC calculations. Steam flow rate is a func-tion of steam enthalpy and pressure, break (or valve) area, and flow resis-tance. The estimation of steam flow is based on an isentropic choked model altered to account for these elements. The model is of the form (4 9) EST = f(P,H) AkP , where f(P,H) = choked isentropic mass flow [1b/hr/in.2-psia (upstream pressure)] as a function of pressure and mixture enthalpy (see ASME steam tables, 4th ed., Figure 14), 4 A = break (valve) size (in.2), k = factor by which ef fective area of break is reduced to compensate for flow resistances in lines and valves, P = pressure (psia). By evaluating this expression for saturated steam enthalpy at various tem-peratures and taking a power curve fit against corresponding saturation temperatures, the expression was converted to b = Ak x 1.87045 x 10-4 Tf4 (4.10) 3 32991 (1b/sec) ST which has an accuracy better than 13% between 200*F and 500*F upstream steam temperature. The choked flow condition holds over this range for TBV flows to the condenser, but becomes invalid at low temperatures for breaks to the atmosphere. With the expression for steam flow substituted into the cooldown equation i
- - , , - - , , , n., , . .- , , . , - . , - - , , , , , , , - , .
CC-4.82 1 for the steam generator, the total model becomes C nIIE -I (4'11) bdt E nPr C (TNPT-I N ) + O D (A) + ORCP + Of~ H SG) M pri Cp with Mp ,g - Mprio + [ I ~ "L) dt I I E ,C (T ,-Tge,) - Ak x 1.87045 x 10-11 dj32991 + ;n c,(T -TSG) .12) l g g g M dt SG Up with MSG " ISGO + ! PT ~ "ST) dt , which is a set of simultaneous, nonlinear differential equations which can be solved numerically to obtain the primary hot leg temperature (T ) and H steam generator exit temperature (TSG). The dcwnconer liquid temperature is obtained from the following equation: Yg Han + EnPT E MPT + ORep + OW (4'13} T DC " * - (mH + "BPI) C, with all quantities as defined above. This equation defines the downconer ) i temperature in terms of the mixing of loop flow and HPI and the heating of the fluid by RCP power input and heat transfer from the vessel wall. This equation does not affect the mass-energy balances (Equations 4.11 and 4.12) described above but is used to define the local fluid temperature in the downconer. l \ 1 1 _.~m . - - , ,_
. __ - _ .~. - . _ -- _ _
SCC-4.83
\
Ano11eation of Cooldown Model. The cooldown model calculates temperatures for the hot leg, steam generator, and vessel downconer using only a two-node energy balance. The Calvert Cliffs-1 plant is equipped with two separate cooling loops which may be subjected to an asymmetric operating condition (e.g., one steam generator blowing down while the other is iso-lated). Such situations require application engineering judgetent to fit the existing conditions to the model. Judgement is also required to . develop the required mass flow data for input to the model. As described at the beginning of this section, the general approach for evaluating a particular scenario is to first identify which of the TRAC calculations most closely matches the description of the scenario. Often the TRAC calculation and the evaluated scenario are identical out to some specified point in time or particular event (SGIS, RCP trip, etc.), after which the evaluated sequence becomes different from the 11 TAC calculation. Temperatures and mass inventories of the primary system and the steam gen-erators are extracted from the 11 TAC calculation at this point to set up the initial conditions for the extrapolation of temperature by the cooldown model. Also, the effective valve area for the model choked flow calcula-tion is selected so that the model will closely follow the steam flow trends observed in the 11 TAC calculation. The initial mass inventories in the primary loops and steen generators may be distributed in different ways to account for asymmetric loop operation. For example, when a steam generator is totally isolated from the rest of the primary system (no heat transport possible) due to flow stagnation in that loop, the water mass and its energy content (temperature) are left out
-, - - - - , - - -._,-n------,.,,.----,------...-wr -,,m-,,--,,e --.m.,-,-w,,--, e, ,...,o,-,,--m..., --.c..c-,-,---
I o . l CC-4.84 of the model, since they cannot influence temperature trends elsewhere. Should the loop flow be restored later, the water mass and the energy would , be put back into the model where they can influence total system heating or cooldown. Another example is when one steam generator is undergoing cool-ing by blowdown while the other steam generator is losing heat to the pri-mary loop due to continued loop flow. In this case, the inventory of the steam generator would be added to the primary mass since both are working together to retard the cooldown of the system. Should any of these condi- ! tions change to a symmetric condition or to another form of asymmetric con-dition, the extrapolation should be stopped for adjustment of primary and steam generator node masses. Other system state changes will require interruption of temperature extra-polation to alter input parameters. Some of these system state changes are listed in Table 4.6. Thenever one of these state changes is encountered, i the current values of the hot leg and steam generator temperatures as cal-i culated by the cooldown model are applied as input to the next extrapola- ! tion segment, together with altered values (as necessary to match the new , system state conditions) of the primary and steam generator mass inven-
- tories, total loop flow, HPI flow, primary leak (pressurizer surge line) flow, feedwater flow, feedwater temperature, heat input rate from wall heat transfer, decay heat factor, RCP heat, and secondary side break (valve) area. This process continues until the entire 0- to 7200-second period is i evaluated.
By estanting the temperature profile of a TRAC-calculated transient, the validity of data interpretation relating to the transient response can be _ - . . _ _ _ _ _ _ , , _ _., . , _ . _ - _ , - _ _ _ . _ _ , . . _ _ __ __ __- _-__._ _,__ __--..____ .-._~ _ __ -__._
CC-4.85
'lable 4.6. System state changes for extrapolation of overcooling sequences by the cooldown model Trigger Condition Significance Action l
]
RCS cools below 535'F. 7BVs and ADVs close. Adjust valve area. RCS sools below 537'F. RCS pressure falls below Initiate charging flow. 1740 psia. HPSI time + 30 sec. - Trip RCPs and begin 100-sec coastdown. Extt:polated pressure below - Initiate HPI flow as per heat 1285 psia. capacity. . SG ocols below 498'F (685 psia). SGIS Close MFIVs, MSIVs. SG inventory below 99,000 lb. AFAS Initiate AFT to one or both SGs. Ceczistence of " broken" and - Isolate AFW to " broken" steam isolated steam generators, generator. SG dries out. - Set secondary break (valve) area to zero. l Hat-les temperature drops Loop stagnates. Adjust mass inventories. < balcw SG temperature.
. SG icvel reaches +22 in. -
Throttle APW flow to SG.- (250,000 - 300,000 lb). Hot-leg temperature becomes Natural circulation Adjust mass inventeries. greater than stagnant SG restored. terp:rature. Ctan:ncement of primary Repressurization to HPI Eliminate HPI flow. system reheat. shut-off head. Sequ:nce specified closure - Adjust parameters accordingly. of valve. 1 i i l m
, , , _ m-- --. , , - - _ _ - , w. gw
,.---.7- . . - - . _ _ . _ . . _ ---,-9 --y *--r-. ._.,. ----- - ,,.w_- , - -g- w--r-.y..pg. , y-
- s
\CC-4.86 checked. When the extracted parameters are correct, the extrapolation will closely follow the TRAC calculation. For example, the times to SIAS and 2 main steam-line breaks at hot zero power (LANL SGIS signals for the 0.1-m transient 1) and at full power (LANL transient 2) as estimated by the cool-down model are not significantly different.
l .; An example of a full 7200-second extrapolation is given in Figure 4.34, which compares cooldown model and TRAC results for the case of a PORY IDCA with a stuck-open ADY (LANL transient 12). This case was selected because it features a secondary side break that causes a general system cooldown coupled to a localized cooling due to significant HPI flow. The two TRAC curves represent the downconer condition under the nozzles of the stagnated (B1) and flowing (A1) loops which represent the expected range of condi-tions. The cooldown model always assnes that all HPI flow is mixed with the flowing loop, thus yielding a temperature lower than the average for the two loops. In this case the extrapolated temperaturo stays within 10 to 50*F of the calculated minimum temperature loop values, i 4.6.1.4. Pressure Evaluation by Coolant Swell Model ; 1 ! An overcooling event will cause the primary coolant to cool down and con-tract, drawing water out of the pressurizer via the pressurizer surge line. As the water level drops in the pressurizer, the steam layer expands and I the system pressure decreases. As the pressure decreases, SIAS initiates char;;ing pump flow and the safety injection pumps are started. If the 1 pressure then decreases to below 1285 psia, high pressure injection flow l commences. These injection flows help to stabilize system pressure during
, .-_y__ --, . _ - ,
vme-
.p . . . - - , . _ _ . - . , . , ,-,w., ,es.. ,s..+ e w w- e-
4 E o COOLT , A.!ePlele! ~
+ LA N L51 D. .I b
] '. ,
H
~
\ .
- ^
h@ R. .. . K* N
,'t E,9 . ,
sl b E 9N
- \'N ,,'-
gg n 7 N.N*N, .' . ao 4
,N %, -
..a
%. 's I
R E 3 A A A A e A A TIME ( seconds ) Figure 4.34. Comparison of TRAC and cooldown model temperature profiles for PORY LOCA with stuck-open ADV (LANL transient 12). I
4 CC-4.88 x the rapid cooldown portion of the event sequence. If the injection flow volume is greater than the shrinkage rate, or if the system enters a reheating mode, the pressurizer water level will increase, compressing the steam layer and increasing the pressure. The rate at which the pressure recovers is of importance because of the contribution of pres-sure in the fracture-mechanics calculations. To determine the best algorithm for estimating pressure recovery rate, the TRAC calculations for Calvert Cliffs-1 were examined in detail. PTS cases calculated by TRAC and also by REALP5 for the Oconee-1 (Ref. and ) and H. B. Robinson-2 (Ref. ) plants were also examined. It was observed that the codes predict that the system pressure variation with pressurizer water level is essentially linear. Furthermore, the PORY set point pres-sure is reached when the pressurizer is on the verge of becoming water solid. A theoretical model of the ideal adiabatic compression of the pres ~ surizer steam layer yields nonlinear pressure vs. pressurizer uster level response and predicts an exceedingly fast repressurization to the PORY set-point pressure. Clearly the ideal adiabatic compression model is not representative of repressurization rates predicted by TRAC and REALPS. Therefore, the observed linear relaticuship between pressurizer level and system pressure was employed for this study. In most of the Calvert Cliffs-1 sequences that were evaluated, the system pressure dropped below and then recovered to the HPI pump shut-off head of 1285 psia. At this point, system cooldown mechanisms have been isolated or corrected and the system has commenced reheating. Injection flow from the i l l l
. a i \
' CC-4.89 i high pressure injection system has ceased and injection flow from the charging pumps may or may not be throttled, depending on the specification of the sequence. The reheating of the coolant will cause the coolant volnae to swell and (with the charging pump flow) refill the pressurizer. The required increase in temperature to cause total refill of the pressar-izer, and therefore repressurization to the PORY set point, may be deter-mined by the following equation: y(Tg, 2400 psia) = (1 + ) I(T ,g 1285 psia) , pri where Y(T,P) = spec:lio volume of water at specified temperature and pressure, Tg = limiting average primary temperature at which coolant swell (and accanulated charging pump flow) volnae equals available pre:surizer steam volume, Tg = initir.1 aversge primary temperature at start of system reheat, V " P'allable steam volume in pressurizer at start of reheat, ST V = volume of primary system susceptible to rehesting Pr
=
prir.sry volnae without, pressurizer or HPI line volume
= 9601 ft3 - 346 ft3 = 9255 ft3 .
} This empirical relationship ignores the action of the pressurizer heaters. j This equation also assumes that there are no primary steam voids outside the pressurizer and that the pressurizer steam volume (VST) is known at the beginning of repressurization. Table 4.7 contains estimates of effective steam voinnes for the repressurization phases of the LANL transients. i These volumes represent the amount of volume change which results in attainment of the PORY set point pressure and do not necessarily represent
-_ . . _ _ - - . . . - . - - -- . . . . - _ _ _ - - - -- - - . - ~
1
= ,
CC-4.90 Table 4.7. Estimates of initial stesa voinnes for Calvert Cliffs-1 transients for repressurization from HPI shutoff up to PORY opening" Time to Initial / final AYolume AVolume Total ) Trcsient repressurize temperature due to coolagt duetochargigg Effectivg) C71eslation (sec) ('F) reheating (ft ) pump flow (ft ) Volume (ft 2120 258/310 240 670 940 LANL1 LANL2 800 405/467 486 284 770 b LANL4 1980 224/221 0 600 600 LANL5 1200 216/218 0 363 363 i LANL6 800 510/540 250 470 720 LANL7" - - LANL8 1810 438/497 455 665 1120 LANL9 1250 432/4898 540 460 1000 8 - - - - LANL10 - d _ _ _ _ LANL11 _ d _ _ _ _ j LANL12 _ n j Repressurization times are calculated assuming no operator actions to control pressure. Caso not smalyzed.
'Rapressurization commences before system reheat; VST not defined.
LOCA case; system does not repressurize.
J i ,
- CC-4.91 4
the actual steam volume in the pressurizer. - l j Engineering judgement dictated the selection of VST for the estimation of j repressurization rate. In evaluation of sequences similar to a LANL tram-sient, the corresponding value of V ST wonid be applied to Equation 4.14. In other cases, generalized values reflooting the trends in Table 4.7 were I solested. Bot zero power sequences were evaluated using a VST of 600 ft . A vaine of 700 ft 3was applied to severe transients at in11 power and j values between 1000 ft3, and 1500 ft3 was applied to milder transients at i fall power. l l For each sequence estimation, the steam volume (VST) and initial average I system temperature (T g ) were applied to obtain the average temperature at which fall repressurization is obtained. The sequence temperature extrapo-4 lation was then examined to obtain the time at which this temperature is achieved. If charging pump flow continued over this period, the accana-j lated voinne over the interval was subtracted from YST and the final aver-age temperature was recalculated. This was repeated until convergence was i obtained. The rossiting seguence time represents the point at which the
)
POKV set point pressure is reached. Pressure between the beginning of I reheat and attainment of fall pressure is obtained by linear interpreta-l tion. 1 l Due to the assumptions involved in the coolant swell model, the prediction 1 { of repressarization rate is imprecise. In most cases the nacertainty in l the esiculation would be conservatively bounded by the use of the repres-l sarization ourves caloslated by TRAC. Some mild transients may t 1 I w
- . - , , , , - - - , . . . , , . ,,,,----.-.,.r, - ,.-r - - - - - - - , - - . , _ , - - . , , - . . , , , - - , . , , . , , , . ,, ,nn-
o . CC-4.92 i a repressurize faster than the rates predicted by TRAC, but this is not expected to affect the fracture-mechanics analysis. 4.6.2. Results of Simple Model Evaluations i 4.6.2.1. Large Main Steam-line Breaks at HZP The sequences related to a large break (},0.1-m 2 ) in a main steam line with ] j the unit at hot zero power (HZP) are described in Table 3.7 in Chapter 3.
- e The seven sequences in the table refloat a variety of combinations of equipment and operator failures. Section J.3 in Appendiz J relates the details of extrapolation development and Figures 4.35 - 4.37 sumasrize the results of the temperature, pressure, and heat-transfer coefficient extra-polations. Sequences 1.1 - 1.6 are represented in the figures. Sequence 1.7 is very similar to LANL transient 4 (see Figures 4.10 and 4.11. for ten-i perature and pressure profiles respectively), and sequence 1.4 is equivalent to LANL transient 1.
l l l e s --,n-, -n---v ,- - , ~ g-,n.-,-n, ,,n, --.,w----,r., - - - - - ,--,,n . , ,,-- a, - ,,
CC-4.93 1 l The temperature curves in Figure 4.35 show the influence of the various l
. i failure combinations in Table 3.7. He six curves fall into three ranges or families on the figure. Sequences 1.1 - 1.4 are all identical to LANL transient 1 out to 1400 seconds, at which time the affected steam generator dries out. The termination of charging pump flows yields local temperature increases and reduced cooling loads for sequences 1.1 and 1.2, the two war-mest sequences for this initiator. These two curves split at about 3500 seconds owing to the failure to throttle AFW to the intact SG in sequence 1.2.
Sequences 1.3 and 1.4 remain cooler than sequences 1.1 and 1.2 because the charging pumps are left running. The separation of these sequences after 4200 seconds is again due to the failure to throttle AFW in sequence 1.4 (LANL transient 1). Sequences 1.5 and 1.6 (and 1.7) drop lower than the others and do not reheat. In the case of sequence 1.5, the drop is due to the failure to stop flow to the affected steam generator. In the case of sequence 1.6 (and 1.7), it is due to greater blowdown from MSIV failure. These failures provide a cooldown mechanism over the entire period and thus prevent reheating. The minimum temperature for sequences 1.1 - 1.4, 253*F (396 K), lies in the portion of the profile extracted from LANL transient 4. The minimum ten-l l peratures for sequences 1.5 - 1.7 are 212'F (373 K), 211*F (373 K), and 212'F (373 K), respectively.
e a
- - . + - - .
0 SEQ 11 g
..A...SEg ,1,,2 E .t.H9.'A l ll x SEQ l 4 3 o SEQ l 5 g V SEQ 16 9 -
3 - g n ~
~$
d Q:
- -.3.,,,,,,
---aannr -nt~. 4 .-
" ---== =.+
n
, e - . _ _ - - - - - - - - . _ . _ .~~~-K
__. 9% 9
\ s GR d'po#
G
- w g
R s; -- g I R
.E R
3 A A A A A A e TIME ( seconds ) Figure 4.35. Extrapolated downcomer temperatures for large nisia steam-line break at ilZP. 0
= .
CC-4.95 The pressure curves in Figure 4.36 show the influence of charging pump operation and system reheating on repressurization. Sequences 1.3 and 1.4 include charging pump flow and system reheating, which cause total repres-surization by 3000 seconds. Sequence 1.7 does not reheat, but also repres-surizes by 3000 seconds as predicted in LANL transient 2. The charging pumps are turned off in sequences 1.1, 1.2, 1.5, and 1.6, and sequences 1.5 and 1.6 do not reheat or repressurize. Sequence 1.2 reheats slowly and repressurizes to 2000 psia at 72000 seconds. Greater reheating in sequence 1.1 promotes repressurization to the PORY set point, 2400 psia, by 6000 seconds. Figure 4.37 shows the heat-transfer coefficient profiles for sequences 1.1 The minimum assumed value, 400 Btu /hr ft 2 **F,
- 1.6. persists throughout the period fc11owing RCP trip. The profile for LANL transient 1 (sequence 1.4) is shown for comparison purposes.
4.6.2.2. Small Main Steam-line Break at BZP The sequences related to a small main steam-line break at HZP are described in Table 3.8 in Chapter 3. The eight sequences in the table reflect combi-nations of MSIV failure, AFW isolation failure, and failure of the opera-tors to turn off charging pump flow and to throttle AFW, Figures 4.38 - 4.40 present the temperature, pressure, and heat-transfer coefficient pro-1 l files for representative sequences 2.1, 2.4, 2.5, 2.7, and 2.8. Due to similarity of conditions, sequence 2.2 was grouped with 2.1, sequence 2.3 was grouped with sequence 2.4, and sequence 2.6 was grouped with sequence 2.7 for the purposes of this summary. Detailed discussion and individual
9 I ^ o SEQ t t f . 4...S.EQ ,1,2 a
" /
f .t.39.12 X SEQ !4 o SEQ 15 g ' / -
.. V SEQ l6
/
n j [ **" a n
<g / '
'... .a9 IC
- ,j. . . . . . . .
$ .f .
pl i n, o.
- n. .s I
80 O BM 2doo 3d00 4doo edD0 edhe 7due a000 Tike ( seconds ) Figure 4.36. Extrapolated downcomer pressures for large main steam-line break' at IIZP. a
O SEQ 11
. 4...89..I .2
.t.H9.G H
' ' x SEQ l 4 N o SEQ 15 l 7 SEQ 15 5 L M n
8 n gR s ? 8 g= e, R t. 4 M p 1
%n -
- . _ . ,_ r - Q - ~ ..
e glos m ados 4d00 m m N N TIME ( seconds ) Figure 4 37 Extrapolated downwmer heat-transfer coefficients for large snais steam-line break at IIZP. e
CC-4.98 (x) suzuyuam a m m a w e e e a a a a :! = z'
$ $N,$'$ d e ,;..x . . =
I %
, -l 1
b i j- .g g l,
.5 I
'l -{
l "l - 4 ( I, $ 8
" n
?
I i
-\ %-
. f M i 3- l t i -
I -l 1 = a s I Y, 2
-I l'
k* e 5 W l
. .I ,
I u 5
- / .
= m- = = = = = = = = m (J) 3HMYH36
t
.- ~
.k 7 o SEQ 21 -
..A ..E.EQ,2,3, 2 ii /' .!. peg 2)
/ X SEQ 27 0 S5Q 28
/. s
/ 2
- sg ab N ~
/ ..
- - - - - : u u -
5 / . n y 2
~8 ./ -
b
/
I N
- O aies sine .Ja. ,g,. 4 ,
TIME ( seconds ) , Figure 4.39. Extrapeissed dowacesser pressures for samau unais stesse-line break at HZP. l
1 O SEQ 2I -
.e. EE9.23.
_t.EE9.E
'
- x SEQ 27
. o SEQ 28 n l g . R-M gi $
g l~B 8 a 7
% 6 bl 8 8 :a w
o e
- R lc e dos ados some M adoe *Ese O sooe .
TIME ( seconds ) .
~~~~
Figwe 4.40. Extrapolated downcosner best-transfer coefficients for seuil mania steam-line break at HZP.
l l CC-4.101 plots of pressure and temperature profiles are provided in Appendix J, Sec-tion J.4. The temperature profiles show two principal regimes: (1) single SG blow-down and dryout with subsequent reheating and (2) extended blowdown from both steam generators without reheating. Sequences 2.1 and 2.4 feature single SG blowdown to dryout with resulting minimum temperatures of 250*F (394 K) and 242'F (390 K), respectively. The failure of the operators to turn off the charging pumps and throttle AFW in sequence 2.4 causes the temperature to remain cooler than in sequence 2.1, where these operator a ctions are carried out. The effect of these operator action failures is 80'F (44.4'C) at the end of the sequence (7200 seconds), as illustrated by the two apper curves in Figure 4.38. Sequences 2.5, 2.7, and 2.8 do not exhibit reheating because KSIV failures or feed isolation failures aug-i mented the amount of water available for blowdown such that SG dryout does not occur. Sequence 2.8 is 10*F (5.5*C) warmer than sequences 2.5 and 2.7 due to operator actions that tensinate feedwater flow to the affected steam generator and terminate cha'rging pump flow. t The pressure profiles for these sequences are shown in Figure 4.39. Sequence 2.1 is assumed to display a mild depressurization which persists l l until CG dryout, where the ensuing reheat of the system causes repressuri-zation to 2210 psia (15.2 MPa) by 7200 seconds. In-sequences 2.4 and 2.7 i the charging pumps are not turned off, so early repressurization.such as that in LANL transient 1 was projected to occur. Sequences 2.5 and 2.8 have neither charging pump flow nor reheating and thus the pressure is assumed to stay at thu HPI flow-limiting pressure. N
--. _ - - - . x. = - . . _ . - - . - - - . _ . . _ . _ - . .
l CC-4.102 Figure 4.40 shows that all of the sequences were assigned the same heat-transfer coefficient profile. The initial value of 4230 Btu /hraft2 ,,, I (2400 W/m
- K) holds until the RCPs are tripped at 120 seconds. By 250 2 2 seconds the assumed minimum value of 400 Btu /hr*ft e*F (2270 W/m K) is obtained and held for the rest of the sequence.
4.6.2.3. Large Main Steam-line Break at Fn11 Power The sequences related to a large break at full power are described in Table 3.10 in Chapter 3. The nine sequences include combinations of failures of i MSIVs and/or ADYs to close, failure of feedwater isolation, and failures of the operators to control repressurization or throttle AFW. Figures 4.41,
'~
4.42, and 4.43 present the temperature, pressure, and heat-transfer coeffi-cient profiles for sequences 3.4 - 3.8. Sequences 3.1 - 3.3 are grouped with sequence 3.4, which is itself identical to LANL transient 2. Sequence 3.9 is grouped with 3.8 for similarity reasons. Detailed discussion of the individual sequences is provided in Appendiz J, Section J.5. l i i The temperature profiles in Figure 4.41 show a wide range of sequence out-come based on whether or not blowdown is stopped. The higher decay heat levels associated with full power operation render the operator actions to throttle AFW to the intact SG or to turn off the charging pumps of minor importance to the temperature trends in the sequences. This is signifi-cantly different from the HZP cases when the same operator actions greatly impact the trends. In sequence 3.4 (LANL transient 2), SG dryout occurs at ! about 400 seconds (minimum temperature of 358'F) and then the primary sys-l l tem reheats under the' influence of core decay heat. In sequence 3.5, AFW
o
.I o SEQ 3.4 g
..^...EE9. 3.P.
i' .t .EE9.3 E X SEQ 37 [ o SEQ 38~ { \ _
~
/
0 - 3
\. -
g 8
, ,. y
..s.....
* --.-. x n g
_ g ., _.....,... ..
/ '-,.
g .. g
- M
. y . . kA T. .. A .
t s.4....--.............,...,,,,,,,,.....................+......., . b H R R
. M Y. a. 6 6 6 6 6 6 m TIME ( seconds )
Figure 4.41. Extrapolated downcomer temperatures for large main steam-line break'at full power.
n9**woA m " A a 2 $ t 2 - . . = m 1 p, g 7 8-3 3,3 3
, 3 Q QgQQ r E E e S S.gE .S E
S w o p O 4.t.x o
- l 6 l f
u t a k a e r b o e o d o i n l m a t e _ s l a n a i o m m e g
)
r s a d l n r o c f o s e s d o s e a( r s u E s M e I r T p r e o m d o o a c n w o d d t e o l a l o a o p _ a
/ I t r
x E _ I 2 I 0 4 0
/ d3 4 m e
r _ u g i F l n e l l E n M-
,' a
I O SEQ 3 4
. 4...M9. 3.9.
.t.gg 38 g x SEQ 37 I o SEQ 38 m lNm
$=.
E gR H l~b' 8 e
?
e
- bg m g
8 . (Je H i se - '^ O c = c o 0 Ilke 2E00 30kle 4doo 6000 Gehl0 1800 Moo TILIE ( seconds ) Figure 4.43. Extrapolated downcomer heat-transfer coefficients for large mala steam-line break at full power.
e l I CC-4.106 isolation failure to the affected steam generaton provides 320 gal / min of ! flow with which to continus blowdown and cooling. However, the cooling provided by this flow did not exceed the decay heat input antil 2000 seconds into the sequence. The temperature rises slightly before declining to the minimum of 240*F (388 K) at 7200 seconds. In sequence 3.8, a main feedwater overfeed to the broken steam generator loop prolongs steam gen-i erstor dryout to about 800 seconds with a minimum downcomer temperature of 276*F (408 K). Decay heat and natural-circulation flow effects cause a rapid recovery in downconer temperature. The pressure response as shown for sequence 3.4 (LAML-transient 2) in Fig-ure 4.42 predicts in11 repressurization by 2000 seconds. Sequences 3.5 - 3.7 experience no repressurization beyond recovery to the HPI shut-off head pressure. Sequence 3.8 experiences rapid repressurization on the basis of j system reheating. I I Figure 4.43 shows the assumed heat-transfer coefficient profile for the se quence s. The profile for LANL transient 2 is presented for comparison purposes as sequence 3.4. l 4.6.2.4. Small Main Steam-line Break at Full Power The sequences related to a small main steam-line break at full power are described in Table 3.11 in Chapter 3. 7he 12 seguences include all of the f ailure combinations examined in the large-break case: MSIV failure, MFW l rumback failure, ADV fallare, AFW isolation failure and operator failures to control repressurization and to throttle AFW. Figures 4.44 - 4.46
O t I O st:Q 4 2 g Ho -
. . . . ._t _ H9.16. . - . .
x sto4a 3 o SEQ 4ii I > [, " _v seq .4 i2'
~'~ '
-~~~*~
./ __...
e ...... n
... ~
~ < ....-
'. ...-- 6:
4 s6 g
/"p-- ,, , sa -
II y A
,% N.f X.
I . - s-
-F-----
I d H o
-4
-...._,,,,,----....--------------. __________ ______ , g H .
R R a R 8 . .
~ 4 m m
e .se m we ~ TIME ( acconds ) Figure 4.44. Extrapolated downcomer temperatures for small main steam-line break at full power.
+
~
I n A* wooa _
~h- -2 Ma.
S .~ , a = * . M o o o s 2 190 1.2 4 114 1.y 4 r e Q g E S 8WQQ E5 S S E w o
- 3. !. S p
v 4 3 t.x 0 O l
.o00 l u
7 f
. t a
' ._ 'k
. a
. e
. r
. b
,ocu e
. i n
. s l m
a _ t e
. s i
n
,o o a o
s m .
. l l
)
a s . m s
. d
[ ' _ ,0 0 0 n o c e s f r o s e _ r 4( u s
- n. s
- E e
._ M r
_ I p
, _
- T
. - r
. - e
/ .
. .oou m
o c
/ /
s n w o
~
. - d
. d
- . - t e
- a
- ._ ,0 l
o A~# - _. 0 0 2 p _ a r _ t x E 5 4
- o o
,os 4 e
r u g i F
' o 1 22 $~ og I o ndE :)
p: mp.
i
,0 SEQ 4 2
.0...S!ig,,ti
.t . H9.18 ao X SEQ 40 l o SEQ 411 i v SEQ 412
- g s Rm M
{. N ll lBO m 9 ok u m y s< o x k
- e e sooo muo 2000 4000 sooo sooo m em TIME ( seconds )
Figure 4.46. Extrapolated downconner heat-transfer coefficients for small main steam line break at full power.
l l i l i l CC-4.110 1 contain the temperature, pressure, and heat-transfer coefficient profiles for sequences 4.2, 4.4, 4.6, 4.8, 4.11, and 4.12. For the purposes of this section, sequence 4.1 is grouped with 4.2; 4.3 with 4.4; 4.5 with 4.6; and 4.7, 4.9, and 4.10 with 4.8. Detailed discussion of these sequences is available in Appendiz J Section J.6. The temperature profiles in Figure 4.44 show a wide range of sequence out-come based mainly on whether or not extended blowdown occurs. The smaller break tends to draw out the period required for SG dryout. This translates into higher minimum temperatures than were obtained for the large-break cases. Also, the delay of reheating to after 2000 seconds reduces the dom-inance of decay heat and makes the effects of operator actions more notice-able. For example, in sequence 4.2 the operator is to turn off the chars-ing pumps, whereas in sequence 4.4 the operator takes no action. Both sequences behave the same through the affected SG dryout (minimum tempera-ture of 337'F (442 K) at 860 seconds] and begin to diverse thereafter. Sequence 4.8 suffers a MFW overfeed to the affected SG, which extends dryout to 1700 seconds. This case also reheats quickly. Sequences 4.11 and 4.12 feature a stuck-open ADV on line B opposite the break. The addi-tional blowdown extends the time of SG A dryout to 1650 seconds at a minimum temperature of 296*F (397 K). l Figure 4.45 shows that all sequence pressure profiles except that for sequence 4.6 return to the PORY set point pressure, 2400 psia (16.6 HPa) . Sequence 4.4 reaches this pressure first based on mildest cooldown and con-tinued charging pump flow. Next comes sequence 4.8 based on rapid reheat-ing. Finally, sequences 4.11 and 4.2 follow based on their slower reheating rates. 1
l . . CC-4.111 Figure 4.46 shows the heat-transfer coefficient for all sequences. The 2 initial value of 4230 Bta/href t 2 **F (24000 W/m *K) holds until the RCP trip. The final value of 400 Btu /hr*f t * *F is obtained 55 seconds follow-ing the trip. 4.6.2.5. Reactor Trip Sequences The sequences related to reactor trip from full power are described in Table 3.13 in Chapter 3. These 43 sequences involve various combinations of failures, including failure of the turbine to trips failures of the ADVs, TBVs, and MSIVs to close; failure of the NFW to run back; failure of the AFW isolation; and failure of the operators to turn off charging pump ) flow and throttle AFW. The P, T, and h profiles for some selected ) 1 sequences are presented in Figures 4.47 - 4.52. Table 3.13 summarises the
~
groupings of sequences for this initiator. Detailed discussions of indivi-dual sequences may be found in Appendix J, Section J.7. Figures 4.47 and 4.48 give the temperature profiles for sequences with failures of one TBV (sequences 5.18, 5.21A
- 5.21B,* and 5.25B*), two TBVs (sequences 5.22, 5.26A,* sad 5.26B*), three TBYs (sequences 27A* and 27B*),* one ADV (sequence 5.35), and two ADVs (sequence 5.36) to close.
- For turbine bypass valve failures, there is a potential for manually closing the valve at the valve location. The " A" membe r of each se t represents failures to isolate the valves such that continued cooldown occurs to final temperatures of 348*F (448 K) for seguence 5.25A and 259'F (399 K) for sequence 5.27A. The "B" members of each set represent manual isolation of the stuck valves, yielding minimum temperatures of 433*F (459 K) for sequence 5.25B, 399'F (476 K) for seguence 5.26B, and 339'F (443 K) for sequence 5.27B. The time required for isolation purposes was determined based on conversations with Calvert Cliffs-1 operational staff. A 15-minute period was assumed to be required to isolate one valve, a 20-sinute period to isolate two valves, etc. It should be noted that for the actual analysis of risk only the "A" cases were considered. The effects of isolation ("B" cases) were, however, determined for the purpose of consideration in the event that one of the "A" cases was identified as a dominant risk sequence.
o I O SEQ S16 g ,
.4 !Eq,.S.8,?,,
.t.!E9.522 Ii L . . . . . . ..
x SEQ 5 35, I o SEQ S y , i- \ * ,, ' 9 SEQ $ 28 A I X
#g' 8 _ SEg,S 21_q I
, , --.--. N I -
,--------.___ - I
-g. A
*"% g "g
'N- g .ad, s.:
oc _N----""~~---e I hl . n ni Ci - 4 9 .& M p H l F3 I E I I I I
=
w E. - TIME ( seconds ) Heure 4.47. Extrapeinsed dowaceaner tesaperaseres for reacter trip (Sequences 5.18, 5.19, 5.22, 5.35, 5.%,5.2I A and 5).
.. . os -
I O SEQ S2SA g, 3...SE,Q,,5 25,Q II + SEg~S26A
-' ~E 5EQ 5E55 I
# ~~ o SEQ 527h" I
/ ,
9 SEQ S b5
,/
I
/_
^ I I
./'/ I n
ad I \ ,
' ; M R
. . . c I P, a g ?
Ei o ' x .+_------.
- - ._. __________---------- gg-Rg _ ~-
-_______,._, g
}
_- 4 I E I I a g R, S e A A A A A = TIME ( seconds ) __ ~ Figure 4.48. F.xtrapelated dowacesser lengeratures for reactor trip (Segmences 5.25 A and B,5.26 A and B,5.27 A and B).
3 E o SEQ 518
/ t' ,'
.4...SEg. 9.!g.,
- l / ,
.t EE9. W .
l l , x SEQ 535,
! / ,a' o SEQ S y, ,
~
I ~ .:! N.
/
,/ oe '
- SEQ 521 A
-s- - SE
- g .5---
2t h. e n ^ * / . < dl * *
./ ab c ,
, * \. -
~
',.__/ l _. . x
. n a .
?
s ml
- p. H p
se e 4 alas sion 4deo ados adoo ideo emo TIME ( seconds ) Figure 4.49. Extrapolated dowycosmer pressures for reactor trip (Sequences 5.18,5.19,5.22,5.35,5.36, 5.21 A and B).
- _ - -- ._ - .~. -.
i o SEQ 525A
..e...Sg .s 29.e n -
o
.t,Sg ,g g X SEQ R 6,B
/ o SEQ 527A l 9 SEQ 527E
./. =
^
n
-l
.l 4
=_9 le
~
i
/ - - - - - - -n y N - ' ,p n Q . g 9 Mi A
w I n e sino adeo :Joo 4doo sino edeo de soon Tilde ( seconds ) Figure 4.50. Extrapolated downcemer pressures for reactor trip (Sequesices 5.25 A and B,5.26 A and B,5.27 A and B).
.am-
e e II, 1 0 SEQ 518
.4...EE9. 9.!!..
.t _ peg _5 g ,
X SFQ 5 35,, o SEQ 536 l 7 SEQ 5E} a , peg _5 gl_Q n h g . Me E i - g /* l ("3
!*~ ?
- b t 8
bg a "5 e
- 1=
I
\ \_ C 1 0 ; m ,
7 e 3d00 2 doe 3dOO 4dOO SdD0 edOS ? doe 3000 Tilde ( seconds )\
\
Figwe 4.51. Extrapolated dowacosner heat-transfer coefficients for reactor trip (Sequences 5.18, 5.19, 5.22,5.35,5.36,5.21 A and B).
- - ~ . .
tI "i o SEQ 525A i .4...s!1 AzSo 1 {
.t.H9 SMA l x SEQ 526B i o SEQ S pA.
i I $ V SEQ 5278 n : n A k x E I I g R l i:
. gh n 8 ?
i " a bl 8 a i: n* s i ( . 4 I I ! i I i 4 : 0 - : ..:. . . .g
, f,e aies an'ee edeo sino sies s ease 4
TIME ( seconds ) Figure 4.52. Extrapolated downcomier heat-transfer coefficients for reactor trip (Sequences 5.25 A and B,5.26 A and B,5.27 A and B). we +
e . CC-4.113 1 Parametrio cases of zero, one, or two NSIV failures are represented in the j above list. The MSIVs have profossd infisease on the course of M Y failure evente. Here the MSIVs are successful, the dowmooner temperature does not drop below 500*F (533 K) unless assisted by other oooldown mechanisms as shown in Figure 4.47 for seguemees 5.18 and 5.19. One MSIV failure leads to minimum temperatures of 400*F as la segmenee 5.22 (two TBVs open) to 450*F as in segsease 5.21A (one DV open, IANL transiest 7) . Figure 4.48 , i shows the response for one NY (seguemees 5.25A and B) failure coupled to the failure of both MSIVs to close. he pressure profiles for these seguences are presented la Figures 4.49 and j 4.50. The combination of system reheating and coatissed okarsing pump flow I cause is11 repressurization of most esses. Figures 4.51 and 4.52 present the heat-transfer ocef ficient profiles for l l I the above sequences. The main differences are in the timing of the RCP , i trips, which oosse later for the mild ADV and single DV cases. 2 I 4.6.2.6. Small-Break LOCA (10.016 ft ) l 2 The seguences assoaisted with the small-break LOCA (10.016 f t in size) are described in Table 3.14 la Chapter 3. The 17 segmences isolade isolatable and monisolatable breaks, NY and ADY failures,10/4 rusback failure and f ailure of operators to turn off charging pump flow af ter break isolation l
_ - - . _ _ - _ =__ - _ _ _ - _ _ _ _ _ _ - _ - _ _ - . - _ - _ _ _ - _ . _ . . - - _ - - . _ _ _ _ _ _ _ _ _ - _
*
- l 4
< CC-4.119 and to throttle AFW. He temperature, pressure, and heat-transfer coeffi-olent profiles for selected seguences are presented in Figures 4.53 - 4.55. . I Detailed discussion of the other seguemees is provided in Appendix J, Soo- i ; tion J.S.
i ne temperature profiles la Figure 4.53 show diversity la ostoome due to I j combinations of cooldown mechanisms. no warmest seguesee, 6.12, experi-i eased early SGIS and loss of NFW flow such that IFI flow and oeessional ADV activity were the only sources of cooling. A combination of NFW and EFI l flow provided cooldown for sequences 6.1 and 6.3 satil SGIS at arossd 2000 seconds. RFI sooling contisses out to 1.5 hours, at which time the break is isolated in seguence 6.1. ng mest ooolest transients are seguemees 6.7 l (LANL transiest 12) and 6.8, la which a stsek-open ADV augments RFI sool-I
! down to yield a flaal temperature of 300*F (421 K). Seguence 6.10, the 1
coldest seguense among those identified for this initiator, included two i 2 stsek-open ADVs augmenting IFI sooldown to yield a minissa temperature of I 253*F (396 E). I i l n e pressure profiles in Figure 4.54 basically follow that of LANL tran-s siest 12. He isolatable break oases deviate from transient 12 values ! after break isolation at 1.5 hours. Seguence 6.2 features f ailure to tara of f okarsing pump flow and so repressurises to the FORY or safety valve set { ! point. H e other isolation cases, sequenses 6.1 and 6.8, repressurise to
- Dependent on whether PORVs are isolated.
/
-/ I o SEQ si g
. A S.g. s z,.
Ii -
.t.Sg s g.
x SEQ 6"7 ] I ? .-- ...................... ,' 8 o SEQ 6 8_ _1..SEg,110 3
.... ... , .= _ Sg s 12 I
' ...,, j eI N ..... ... ,,_ .*
I;; v m an N m -
- N.N.D 'q ,
58 a ' g i X- 6 $ u o I I E I R R R 1, a A A A A A A = TIME ( seconds ) Figure 4.53. Extrapolated downcomier imperatures fdhnall. break LOCH 0.016 2 ft ).
b 4 8 I o SEQ s
..k...kbS..h.h..
iI I _t.!Eg 6 3,
, f x SEQ 67 l o SEQ 68_
l
! V SEQ 6 to
.9,@Eg 6[2 f
8 ft
,I -
! A
<k !
i "R , i =_$
~ ! _
. n
. . y A
a w M. n i - _J
=
u
, g s
1 $ H
'o dos ados ados odos odos edeo A saco TlWE ( seconds )
Figure 4.54. Extrapolated downcomer pressures for mail-break LOCA (40.016 ft 2).
l o . 4 i CC-4.122 ( x an/a aam
) mo31 H on g
_ a:- esm; e c e eje;e m.eje;
$ $bbb$'$ $kkb o<!+lxo>la! j l
1 1 a i j e -
- g 3 _
a i g, l. g5
.g2
. m E
= w
- e
.3
-I
- l
! s 4
.g 1 t
l
~
e l -g a m w e M M M M M $ ( 4 au un/nla ) m 03 1 H I
, - - . - - c-
4 e CC-4.123 i the HPI shut-off head pressure. However, if the system is water solid,
~
. i.e., if no steem voids are present, the reheating after break isolation would cause repressurization similar to seguence 6.2. i Figure 4.55 shows the heat-transfer coefficient profiles for these sequences. The behavior for transient 12 (seguence 6.7) is provided for 2 ! comparison with the assumed minimum value of 400 Btu /href t2 ,,F (2270 W/m ), 1 4.6.2.7. Small-Break LOCA (1.02 ft2 ) 2 j The sequences deallas with a monisolatable small primary break (0.02 ft ) is provided ir. Table 3.15 in Chapter 3. The eight seguences include vari-ons combinations of TBY and ADV failures, NW runback failure, sad failure i 4 of operators to throttle AFW. Figures 4.56 - 4.58 provide the temperature, i pressure, and heat-transfer coefficient profiles for seguences 7.1, 7.4, and 7.6. Seguences 7.2, 7.3, 7.7, and 7.8 correspond to sequence 7.1, which is equivalent to LANL transient 11. Sognence 7.5 is similar to segnance 7.4. Detailed disonssion of these sequences is provided in Appen-I' dix J, Section J.9. i The temperature profile in Figure 4.56 expresses the influence of HPI cool-ing alone (seguence 7.1), HPI sooling with one ADV open (seguence 7.4), and l HPI cooling with both ADVs open (seguence 7.6). 73V and MFW rumback j l
. failures only incur an early SGIS, whleh eliminates such cooldown neohan-isms, leaving only the HPI cooling neohanism. Therefore, these other cases
_m_ _,,-y__,__ , . _ . . ,, .4---m.....,.___-7_ , _ _ , - _ , _ . . _ _ . . - _ , _ _ _ - , _ , _ _ , _ , - , _ .
b 3 I o SEQ 71 g 8... S.Eq , 7 3, . g * ..
...SE9.76
= ,
'a g
,\. ,
~
g
+., ~ ~ . ,~ ,
w ' 6 _g
, 8 -
'~~.'-.,
g ,,
-..,,,~~.~. g
+,,.. .., .
..~~~..~..,,
n , . , .. ~... **a .s
.. e-a p
I
...'-s...--..... .., 5 6
E R R I A R, g & A A A A *
- TIME ( seconds )
Figure 4.56. Extrapolated dowacosmer tesaperatures for unau. break LOCA ( =0.02 ft 2).
I CC-4.125 ( YdM ) 2HASE384 j n n a a e e *
- g iIl?l 5E!=
n n:n. 0 4+ i 3 k j , e e e L I , $
' mJ 1
4 L.1 l
. a.
2 N
' N ! ! 'I l i
4 . 5 h. l g, s
^
i g M $35! W M 0 ( VISd ) IMd
. - - . , - - . , ,, .n- , - < - , -
+ e. t i CC-4.126
\
I . ( x mA -) J.nm I a-
- PR $~
e n se. nn g o< 4: 9n gg l 'k
'o ,
1
. i,o ,
i . O Ll ]= l v + 5
}
i l o
.)
i l d 1 l a cost esse ens amur ones e ( 4 W Wale ) J.GM 1 H
CC-4.127
?
l l altimately resemble sequence 7.1. The mininas temperature of sognence 7.G, the soolest seguemee is 253'F (3 M K). he pressure profile in Figure 4.57 is that of the IML transiest 11, which is applicable to all of the seguences for this faitiator. He heat-transfer coefficient profiles in Figure 4.58 imelade the assaned minimum value and also the prof 11e for LANL transient 11 for eenparison.
o -e > l l ESTIMATION OF PRESSURE. TEMPERAIURE, l AND REAT TRANSFER COEFFICIENT I J.1 Introdnation l The evaluation of the risks of pressurised thermal shook (PTS) entails l the complias of overooolias isoident event trees to fractured nochanics caloslation of the probability of ve ssel orack propaga tion. De link i between as event tree and state and the fracture neokasios calculation is
, i j the transient behavior of pressure (P), temperature (T), and heat transfer ;
i i l coefficient (h), in the reactor vessel dowsooner region. Dat is, the l P, T, h transiest profiles from the sequence defined by an event tree j end state beoose tapats for the fracture seehanies caloslation. Here are i j tems of millions of end states on overeoclias traastent event trees. Dae ; i J to the cost and sceplomity of thermal hydraalies and fracture seeksales l calculations, it is not praetleal to evaluate every end state separately. 1 i Therefore it beooses neeessary to a) redase by similarity groupias the j aanber of end states to be evelaated and b) redsoe the number of detailed thermal hydram11e calculations thronsk the use of less rigorous estimation t echnique s . i
! This appendis sanearises the approach used to stoap sequenses and estimate i
j P, T, k profiles for the Calvert Cliffs FTS study. Section J.2 deseribes I the estimation methodology developed for this study and the approsok aad rationale for sequence seospias. Sections J.3 throash J.9 present the i results of evalua tions for esek of the maj or initiatias events. Rose initiations inelade: I 4 4 J.1
. t s
- 1. Large main steam line break at hot zero power,
- 2. Small main steam line break at hot zero power,
- 3. Large main steam line break at full power,
- 4. Small main steen line break at full power,
- 5. Reactor trip,
- 6. Small break LOCA at full power, and
- 7. Medium break LOCA at full power.
Other initiators such as small break LOCA at hot zero power and' loss of main feedwater at full power (followed by anziliary feadwater overfeed) are not addressed in this appendiz. The P, T, h transient profile estimates presented here are based on TRAC-e PF1 calculations reported by Los Alsmos National Laboratory ( ). The detailed transient calculations are samarized in Chapter 4.0 of this PIS report. Computer tapes of IRAC plot output files for these calculated transients were also employed in the development of parameters applied to the temperature and pressure estimation procedures. The sole and extensive use of these TRAC calculations in sequence P, T, h profile estimation implies that the estimations reported here will be subj ect to the same modeling assumptions and code characteristics driving the uncertainties in the IRAC calculated results. Additional uncertainties
. introduced by the estimation procedure have not been fully evaluated. Such uncertainty was minimized by using the estimation procedure to duplicate portions of the transients calculated by 1RAC and thereby check the validity of assumed parsneters and extrapolation models.
3.2
e s The estimated P, T, and h profiles presented in this appendix represent a " single point" estimate of downconer conditions. That is, the estimated conditions are assumed to hold for the entire downconer region without any azimuthal or axial variations. The detailed TRAC calculations demonstrated both azimuthal and axial variation in fluid temperstures and heat transfer coefficients. To minimize the impact of the fracture mechanics calculations the cooldown model used in the estimation procedure was set up to yield the expected tempehature of the coldest sub-region of the downcomer rather than the overall average temperature for the whole downcomer region. J.2 Methodolorv J.2.1 General Approach Af ter an initial survey of the data resources and the sequences identified for estimation, the five-st ep process depicted in Figure J.1 was employed in the development of Calvert Cliffs-1 pressure, temperature, and heat transfer coefficient estimates. This approach allowed logical reduction of the number of cases to be evaluated and derived the greatest benefit from the information in the TRAC calculations. The first step involved the grouping of similar sequence s within each transient initiator table. An evaluation of the TRAC calculations fer the effects from different operating states provided the criteria for ; assignment of sequences into groups. Besides providing grouping criteria, step 2 developed the parameters for the cFM town (temperature) and coolant swell (pressure) models used on occasion for this study. To assure correct interpretation of conditions during seque nce s, the appropriate parameters J.3 l
~
e e were applied to the cooldown model to duplicate portions of sequence s j calculated by IRAC. These validation efforts took place in step 3. (See : Section J.2.3.) In step 4, the pressure, tempe ratur e, and heat transfer coefficients were estimated. Temperature could be estimated either by piecewise application of IRAC results or by calculation using the cooldown model. De me thod selection depended on the complexity of the sequence and the availability of applicable data from the TRAC calculations. Early portions of many evaluated sequence s had stated configurations identical to those of a particular TRAC calculation, so piecewise use of the TRAC results was applied. The cooldown model was then used to extrapolate the remainder ' of the transient out to 2 h. Certain mild (i.e., high temperature) transients were not explicitly extrapolated, since they were well removed from the threshold temperature E; g >10- ). nose mild sequence s were assigned the P, T, and h profile of the IRAC calculation or extrapola-at which vessel failure probability can be calculated (P tion most closely representing the anticipated response of the sequence. Pressure estimates were derived from observation of pressure trends in the TRAC calculations and by a pressure prediction model. De Calvert Cliffs-1 plant fe atures a low-head HPSI system which cannot repres surize the primary above the pump shut-off head of 1285 psia. The charging pumps can repressurize the primary up to the PORY setpoint (2400 psia), but does so at a very low rate due to low flow capacity, ne charging pumps were not throttled in any of the TRAC calculations. H erefore, there are ample cases available for evaluation of the contribution of the charging l J.4
)
)
o e pumps to system repressurization. De coolant swell model accounts for depressurization due to coolant expansion on reheating. This model is used for extrapolated cases where charging pump flow is throttled. Heat transfer coefficients were based on piecewise selection of TRAC data and result of modeling performed by Theophanous at Purdue University (see Section 4.4). In general, the TRAC calculations predicted relatively constant values while the r e'ac t or cooling pumps (RCP) are running and a step change to a lower but constant value af ter RCP trip and establishment of natural circulation. The heat transfer coefficients predicted by TRAC did not include correction for free convection effects. Therefore, the IRACS values were underpredicted for natural-circulation flow conditions. Theophanous found that the contribution of free convection to the downconer heat transfer coef ficient off set increases or decreases inforced convection such that a total value of 400 Btu /hr ft F2o(2270 w/m k) 2 ' was maintained over a wide range of natural circulation flow conditions. The extrapolations presented in this appendiz are a composite of TRAC calculated heat transfer coefficients for pre-RCP trip regimes and corrected estimates for natural circulation regimes. The completed estimatious were documented in step 5. This documentation comprises Sections J.3 through J.8 of this appendix. J.2.2 Sequence Grcuping When all PTS initiators and failure branches are se t up in event trees, several million end states result. To obtain a tractable yet representative set of PTS transients some method of sequence grouping is necessary. J.5
e e i l l Chapter 3 describes the construction of the event trees and process used to eliminate "non-contributing" states (i.e., component failures made l irrelevant by the action of other systems or components). De collapsed event trees from this process still contain in excess of 100,000 end states. Section 3.5 describes the screening process used to separate end states into a set of discrete sequences for evaluation and a set of residual sequences for which no further evaluation was performed. he discrete sequences include all events with estimated frequencies greater than 10 -7 per year. Sequences representing identical combinations of failures were collapsed to a single group and the corresponding frequencies were s amed. Sequence s with frequencies be twe en 10-7 and 10-8 per year were also examined for similarity with' the discrete se quence s and were collapsed toge ther with specific discrete sequences and where appropriate. His approach minimized the cumulative frequency of the residual. He resulting set of discrete sequences are found in Tables . AI t oge ther, sequences emerged from this grouping process. The grouping process of Chapter 3.0 were based on system configuration and event frequency. Further, grouping may occur based on the thermal hydraulic impact of configuration. De impact i of a particular component or system can be evaluated from observation and evalua tion of the effects of its operation or failure in the 11 TAC calculations. In this way the importance of failures or actions could be classified as dominant, minor, or inconsequential. Sequences with the same dominant features were grouped together for analysis. In later stages, the influence of minor events was evaluated to check the consistency of the groupings. This checking accounted for the thermal-hydraulic interaction I i 3.6
. e 1
i or feed-back dr.e to the combination of failures. Some sequence s were reassigned to other groups as a result of such checks. The sequence tables of Chapter 3. (Tables ) also contain scue families l of se que nce s. The members of a family differed from each other only in the timing of certain corrective actions (throttling of charging pump flow, throttling of AFW, etc.). The most severe se quence in the family would be evaluated to determine the extent of the cooldown. Where the "most severe" sequence of a family turned out to be very mild from a fracture mechanics standpoint, no further ef fort was spent on evaluating other members of the f amily. Where the most severe sequence did present a significant FIS risk, the other members of the family were evaluated separately or assigned to other groups,as appropriate. The groupings for each of the initiators are discussed in sections J.3 through J.9. J.2.3 Cooldown Model I The temperature response of a transient sequence is a function of the system's configuration during the se quence , including the timing of configuration changes (e.g. , RCP trip, MSIV, MFIV closure, AFAS, etc) . The
- sequences calculated at Los Alamos National Laboratory using TRAC represent
.l only 12 out of the millions of se quence s on the overcooling event trees. $ The cooldown model is a means to apply the information generated by the TRAC calculations to other sequences requiring temperature response estimation. The approach used in the cooldown model was to obtain separate mass-energy balances around the steen generators and the reactor vessel (i.e., balance J.7
of the primary cooling system) to predict the rate of temperature change. All pertinent cooling and heating mechanisms were included. However, to obtain these mas s-energy balances, the assumptions listed in Table J.1 were necessary to simplify the system to a two node model. The assumption of no steam generator heat transfer resistance will result in prediction of slightly lower primary temperatures than are reported by TRAC. He ! error is proportional to the rate of heat transfer. The error will be less than 10 0F for large steam line breaks (LANL Transient 3) and less F than 5 for single turbine bypass valve failures (LANL Transient 4A) under conditions where na tural loop circulation pr evail s. The assumption of thermal equilibrium in the steam generator secondary allows the use of l simple choke flow models to predict steam flow rate. Conditions close
; to thermal equilibrium are obtained by TRAC for steam generators during blowdown. Division of the reactor coolant system into only 2 nodes,
" smears out" the type of temperature lag information available from the finely noded model used by 1RAC. Herefore, the cooldown model will respond f aster to input parameter changes than will the IRAC model. Direct comparison of cooldown model extrapolated temperature response with TRAC calculations suggest that his effect is small for cases where natural loop circulation remains large (>500 lb/sec).
The final assumption involved in the cooldowr. model is that TRAC calculated mass flow data may be applied to the evaluation of other sequences. This assumption is necessary because the mass flow information necessary i to implement the cooldown model cannot be calculated from a simple 2 node model. Engineering j udgement is used to identify segments of the J.8 l .- _ _ , . - - ..__ _---_.- -- - .. .. .- .- _. . - . . -
O l . TRAC calculations relevant to the sequence being evaluated. Pertinent mass flow data are then extracted from the identified TRAC calculations for application to the cooldown model. The required parameters for the model are listed in the derivation of the model as described below. i Model Derivation and Characteristics The cooldown model consists of 2 simultane ous non-linear differential equations describing the mass-energy balance of a primary node (i.e., vessel, loop piping, and RC pumps) and a steam generator node as follows: d(MU) primary = Q7 Hgpg -sHg g + QD (t) + kCP + k ~ O sec II*II dt d(MU)SG =QHg -i ST HST + O sec (J.2) dt _- where A HHPI " Product of HPI mass flow and specific enthalpy at HPI nominal HPI temperature (THPI) vs. thermodynamic ref erence temperature (T,,g)
" bI Cp GHPI ~
ref in gH = product of primary leak flow (pressurizer surge line or break) and specific enthalpy at hot leg temperature (Tg) .
=A g Cp (TH - T,,f), (valid for liquid flow only)
Ag Hg = product of feedwater mass flow and specific enthalpy at feedwater temperature (Tg)
=a g Cp (Tg - T,,f)
A Hg = product of secondary steam flow and specific ST enthalpy for saturated steam at steam generator- conditions (Tgg)
=A ST (AHy ,p(TSG) + Cp (TSG - T,,f))
3.9
Q (t) = decay heat input as function of time D
= ANS Decay Heat Function for transients from full power
= constant value for transients from hot standby CP = Pump power deposited in coolant
= Heat transferred from vessel wall to coolant Q,,, = Heat transferred from primary to secondary In the absence of heat transfer resistance, Q,,, is limited only by transport of energy to steam generator by the hot leg flow (A ) '#
H Q,,, = AH Cp (TH -T SG I - The left had sides of equations J.1 and J.2 may be expanded by use of the chain rule d(MU) dU dM l
=M - + U -
dt dt dt where M = total mass U = specific energy = C (T - Tgg) du dT
=C -
dt dt dM
- =
fi=massflowacrosssystemboundaries. dt Substitution into the lef t hand sides of equations J.1 and J.2 d(MU) primary dTh P#I
- v H - T,,g ) %-y U.D dt dt d(MU)SG dT 3g
" MSG yC +C y (T3g - T,,f) (ig-sST) II 4) l J.10
then placing these expressions with their respective right hand sides yields Primary dT H Mp ,g Cy dt
+C y (TH - T,,,) (sHPI - SL ) " i i
*HPICp (THPI -T,,g)-sfP(TH-Tref) + O (t)D + QRCP + % - SH Cp (TH -TSG)
(J.5) Steam Generator dT SG M SG yC +C y (T 3g - T,,g) (sg-AST) " dt
&g Cp (Tg - T,,f) - AST (AH (TSG) y + Cp (T3g - T,,f) ) +AH Cp (TH -TSG)
, (J.6) For liquids C, may be assumed to be equal to Cp . Using this assumption and cancelling common terms yields Primary dT EHPI p(THPI - TH g(t)
- b - Yp H ~
H , , . kCP SG gy ) dt M C M C M C M C M C Pri p pri p pri p pri p pri p Steam Generator dT 3g sg C,(Tg -TSG) ~ SST AH,,(TSG) + AH (TH -T SG I
-- = (J.8) dt MSGCp MSGC , MSG Cp In this form, the thermodynamic ref erence state (T,,f) has been eliminated, leaving only the expressions for heating and cooling mechanisms.
Flow rates for HPI, leak, hot leg and feedwater are independent parameters extracted or estimated from 'IRAC calculations. Steam flow rate is a function of steam enthalpy and pressure, break (or valve) area, and flow J.11 l
l resistance. The estimation of steam flow is based on an isentropic choked flow model altered to account for these elements. The model is of the form AST = f (P,H) AkP (J.9) where l f(P,H) = choked isentropic mass flow (1b/hr/in2 -psia (upstream pressure)) as a function of pressure and mixture enthalpy. See ASkE , steam tables, 4th ed. , Figure 14. ' A = break (valve) size (in2) k = factor by which ef fective arca of break is reduced to compensate for flow resistances in lines and valves P = Pressure (psia) i By evaluating this expression for satur ated steam enthalpy at various t emperatures and taking a power curve fit against corresponding saturation temperatures the expression was converted to e AST = Ak (1.87045 x 10~4)gT 4.32991 (ib/sec) (J.10) which has an accuracy better than 13% be tween 200*F and 500 F upstream steam temperature. The choked flow condition holds over this range for TBV flows to the condenser, but becomes invalid at low temperatures for breaks to the atmosphere. - With the expression for stessa flow substituted into the cooldown equation for the steam generator, the total model becomes Vessel dIH &HPI C,(THPI - TH I + ED(t) + kCP + k ~ $H C,(TH -Tgg) (J.11) dt M Pri Cp with M =M priO + IEHPI ~ AL ) dt i J.12
--)
- _~ . . _ - . . _ . - _ _ . ...
l i Steam Generator
-11 4 32991 (J.12) dT ST ,
E FW C ,(Tpy-TSG)-AKzl.87045x10 T SG C,(TH -Tg) M dt SG Up with MSG " "SGO +- IEFT ~EST) dt which is a se t of simultaneous, non-linear differential equations which can be solved numerically to obtain the hot leg temperature (T H) and steam generator exit tempe ratur e . Downconer liquid temperature is obtained from the following equation AHHgg + sHPI HHPI + k CP + b T (#*13) DC " (AH+ERPI) Cp I with all quantities as defined above. This equation defines the downconer temperature in terms of the mixing of loop flow and HPI and the heating of the fluid by RCP power input and heat transfer from the vessel wall. This equation does not af fect the mass-energy balances (equations J.11 and J.12) described above but is used only to define the local fluid temperature in the downconer. 4 l Annlication of Cooldown Model The cooldown model calculates temperatures for the hotles, steen generator, and vessel downconer using only a 2 node energy balance. The Calvert i Cliffs 1 plant is equipped with 2 separate cooling loops which may be i I subj ected to asymmetric operating condition (e.g., one steam generator I blowing down while the other is isolated). Such situations require applica tion engineering j udgement to fit the existing conditions to the model. Judgement is also required to develop the required mass flow data for input to the model. , 1 J.13 l
- _ . _ . . . -_._ . . _ ,_, . . . . . . _ _ __. _ _ . . _ . . . _ . ~ . _ , - _ _ . . . , _ _ _ _ _ _ . _ _ _ . _ _ . . . _ _ _ . . , . . . . _ . . _ . . .
4 As described in the introduction to this section the general approach i for evaluating a particular sce nario is to first identify which of the I TRAC calculations most closely matches the description of the scenario. I Of ten the TRAC calculation and the evaluated scenario are identical out to some specified point in time or particular event (SGIS, RCP trip, etc.) where the evaluated sequence becomes different from the TRAC calculation. Temperatures and mass inventories of the primary system and the steam generator s are extracted from the TRAC calculation at this point to set up the initial conditions for the extrapolation of temperature by the cooldown model. Also, the of fective value area for the model choked flow calculations is selected so that the model will closely follow the steam i flow rends observed in the IRAC calculation. l
! The initial mass inventories in the primary and steam generators may be distributed in different ways to account for assymmetric loop operation.
For example, where~ a steam generator totally isolated from the rest of the primary (no heat transport possible) due to flow stagnation in that loop, the water utss and its energy content (temperature) is left out of the model, since it cannot influence temperature trends elsewhere. Should the loop flow be restored later, both the water mass and the energy would I be put back into the model where it can influence total system heating or cooldown. Another example is where one steam generator is undergoing cooling by blowdown with the other steam generator loosing heat to the i primary due to continued loop flow. In this case, the inventory of steam generator would be added to the primary mass since both are working toge ther to retard the cooldown of the system. Should any of these conditions change to symmetric or another form of assymmetric conditions, J.14
the extrapolation should be stopped for adj ustment of primary and steam generator node masses. . Other system state changes will require interruption of temperata:o extrapolation to alter input parameters. Some of these state changes and required changes in parameters are listed in Table J.2. Whenever one - of these state changes are encountered the current values of the hot leg and steam generator temperatures as calculated by the cooldown model are' applied as input to the next extrapolation segment. New values of primary l and steam generator mass inventories, total loop flow, HPI flow, primary , i leak (pressurizer surge line) flow, feedwater flow, feedwater temperature, heat input rate from wall heat transfer, decay heat factor, RCP heat, and secondary side break (valve) area are also supplied to match the new system ! state conditions. This process continue s until the entire 0-7200 second i period is evaluated. By estimating the temperature profile of a TRAC calculated transient, the validity of data interpretation relating to transient response can be checked. Where the extracted parameters are correct, the extrapolation will closely follow the TRAC calculation. For example, the times to SIAS and SGIS signals for 1 ft 2 main steam line breaks at Hot Zero Power (LANL Transient 1) and at Full Power (LANL Transient 2) as estimated by the cooldown model were not significantly different. An example of a l full 7200 second extrapolation is given in Figure J.3 which compares the cooldown model to the results for LANL Transient 12, PORY LOA with stuck i ! open ADY. This case was selected because it features a secondary side break which causes a general system cooldown coupled to localized cooling l l i J.1S e
, - , . . - , - . . ..-. _ --. - - . , , . , - w - , , , , . - ,, ,,,,a..,--. , , , - - - - , , - , , - , .. . , - - ,
4 due to significant HPI flow. The two TRAC curves represent downconer condition under the nozzles of the stagnated (B1) and flowing (A1) loops ) which represent the expected range of conditions. The cooldown model I l always assumes that all HPI flow is mixed with the flowing loop, thus i yielding a temperature lower than the average for the two loops. In this case the extrapolated temperature stays within 10 to 50*F of the calculated minimum temperature loop values. J.2.4 Pressure Estimation by Coolant Swell Model An overcooling event will cause the primary coolant to cooldown and contract, drawing water out of the pressurizer via the pressurizer surge line. As the water level drops in the pressurizer, the steam layer expands and the system pressure decreases. As the pressure decreases SIAS initiates charging pump flow and, if the pressure decreases to below 1285 psia, high pressure injection flow commences. These injection flows stabilize system pressure during the rapid cooldown portion of the event se que nce . If the injection flow volume is greater than the shrinkage rate or if the system enters a reheating mode, the pressurizer water level will increase, compressing the steam layer and increasing the pressure. The rate at )I which the pressure recovers is of importance because of the contribution of pressure in the fracture mechanics calculations. l To de termine the best alscrithm for estimation of pressure recovery rate, ' the TEAC calculations for Calvert Cliffs were examinea in detail. TRAC l and RELAP5 calculated PTS cases for the Oconee-1 and H.B. Robinson-2 plants l l l J.16 i ( I
. o were also examined. It was observed that the codes predict that the system 6
pressure varies linearly with pressurizer water level. Furthermore, the PORV set point pressure is reached when the pressurizer is on the verge of becoming water solid. The alternate theoretical model of the ideal adiabatic compression of the pressurizer steam layer yields non-linear pressure vs. pressurizer water level response and predicts an exceedingly fast repressurization to the PORV set point pressure. Clearly the ide al adiabatic compression model is not representative of repressurization rates predicted by TRAC and RELAPS. Therefore, the observed linear relationship between pressurizer level and system pressure was employed for this study. In most of the Calvert Cliffs sequences evaluated, the system pressure dropped below and then stabilized at the HPI pump shut off head of 1285 psia. At this point system cooldown me chanist.as had been isolated or
- corrected and the system has commenced reheating. Injection flow fron l the high pressure injection system had ceased and injection flow from the charging pumps may or may not be throttled depending on the specification of the sequence. The reheating of the coolant will cause the coolant volume 3
to swell and (with the charging pump flow) refill the pressurizer. The
; required increase in temperature to cause total refill of the pressurizer l
and therefore repressurization to the PORY set point may be de termined by the following equation ST X (Tg, 2400 psia) = (1 + ) y (T ,g 1285 psia) (J.14) V pri J.17 l
I 1 where Y (T,P) = specific volume of water at specified temperature and pressure Tg = Limiting average primary temperature at which coolant swell (and charging pump flow) volume equals available pressurizer steam volume Tg = Initial average primary temperature at start of system reheat VST = Available steam volume in pressurizer at start of reheat, and Vp ,g = Volume of primary system susceptible to reheating
=primaryvlumewighoutpresspizerorHPIlinevolume 3
= 9601 ft - 346 ft = 9255 ft This equation assumes that there are no primary steam voids outside of the pressurizer and that the pressurizer steam volume (Vg) is known at the beginning of repressurization. Table J.3 contains estimates of ef fective-steam volumes for the repressurization phases of the LANL transients.
These volumes represent the effective steam volume (i.e., the amount of volume change which results in attainment of PORV-set point pressure) and do not necessarily represent the actual steam volume in the pressurizer. Engineering judgement dictated the selection of V ST, for the estimation of repressurization rate. In evaluation of sequence s similar to a LANL transient, the corresponding value of V would be applied to equa tion ST J.14. In other cases generalized values reflecting the trends in Table J.3 were used. Hot zero power sequences were evaluated using a VST 'f 3 600 ft3 A value of 700 ft was applied to severe transients at full power and values between 1000 ft 3 and 1500 f 3t was applied to milder transients at full power. f For each sequence estimation, the steam volume (VST) and initial average system tempe rature (T g ) were applied to obtain the average temperature at which full repres suriz a tion is obtained. The sequence temperature J.18 ! l
.. - - - -__ - ..-. --_ __ . -- . . = - . - ._ --. . - _ - .
extrapolation is then exasined to obtain the time at which this temperature is achieved. If charging pump flow continued over this period, the accumulated volume over the interval is subtracted from V g and the final average temperature is recalculated. This is repeated until covergence is obtained. The resulting time represents the moment at which the PORV set point pres sur e is reached. Pressure be twe en the beginning of reheat and attainment of full pressure is obtained by linear interpolation. 1 Due to the assumptions involved in the coolant swell model, the prediction of repr es surization rate is imprecise. In most cases the uncertainty in the calculation would be conservatively bounded by the use of the I repr es suriz ation curves calculated by TRAC. Some mild transients may 1 ] repressurize f aster than the rates predicted by TRAC. J.2.5 Heat Transfer Coefficient Estimation As discussed in Section 4.4 corrections to downconer convection heat I transfer coefficient have been developed. The trade-offs between forced and free convection effects under natural circulation conditions resulted in a nearly constant heat transfer coefficient of 400 Btu /hr ft20F , This value was applied as a floor or lower limit for all evaluated cases following coast down of the RCP's. Therefore all evaluated cases had similar heat transfer coefficient profiles. i t i 3.19
; J.3 Larne Steam Line Break at Hot Zero Power J.3.1 Description of sequences The sequence s for large main steam line break at hot zero power are initiated by a 1 ft 2 (or larger) break in a main steam line down steam of the flow restricter end upstream of the MSIV. De system is initially j at steady state conditions and nominal steam generator levels for hot zero power. The decay heat level is 9.38 MW, corresponding to 100 hours after shutdown.
[ The seven specified sequences for this initiator are listed in Table J.4. The dif ference s in the sequence s relate to MSIV operation, isolation of AFW to the affected steam generator, HPI operation, and operator actions to turn of f the charging pumps and throttle AFW flow, i J.3.2 Bases for extrapolation LANL Transients 1, 3, 4, and 5 relate to large steam line breaks at hot 2 zero power. Transient i features a 1 ft break in steam line A. The only other assumed f ailures in this transient are failures to turn off charging pump flow and to throttle AFW st +22 inches in the intact steam generator B. nose conditions make Transient i equivalent to sequence 1.4 in Table i J.4. Transient 4 features a double-ended break in loop A coupled with system l i failures to isolate AFT to SGA and operator failures to turning of f the charging pumps and throttle AFW to both SGA and SGB. J.20
.-. .-- =- .- - . = ..
't i I , Transient 5 also assumes a donble ended steam line break in loop A and operator failure to turn off the charging pumps" upon reaching the HPI flow limiting pressure. For this case, however, the MSIV's f ail open and the ope rator is assumed to throttle AFT to both steam generators after 8 minutes. Transient 3 duplicates the events of Trans!ent 1 with the exception that only two RCP's tripped 30 s af ter SIAS. This transient does not correspond to any of the sequences in Table J.4, but it does illustrate the beneficial effects of not tripping all RCP's for a steam line break incident. For the purpo se s of extrapolation, Transients 1, 4, and 5 provide i temperature and pressure profile s directly applicable to first 1000-1400 s of the specified sequence s since there is no deviation of condition be twe en the sequences and corresponding LANL transients over this period. The stable loop flow and vessel wall heat transfer conditions later in Transients 1 and 5 and the condition pertinent to restoration of flow in loop B late in Transient 1 were employed generally in the evaluation of sequences of Table J.4. J.3.3 Results and Discussion J.3.3.1 Sequence 1.1 A. Basis - Transient 1 D i J.21
B. Departures from Basis The departures involve operator actions to turn off charging pumps when primary systems regain pressure to the HPI shut off head 8.9 MPa (1285 psia) and to throttle AFT to intact SGB on reaching + 22 inches level. C. Temperature extrapolation The specified departures from Transient 1 will not affect temperatures until after broken SGA dryout. Therefore Transient 1 was used out to 1400 s. By this time the charging pump and HPI flows have ceased. AFT continues to SGB until reaching +22 inches level at 1900 s. However, flow in Loop B will remain stagnant until the primary becomes warmer than SGB (379 F (446K) at AFW cutoff). The lowest temperature, 253 F (396K), occurs e at 1308 s. Extrapolation proceeded in two segments. Heat transfer from metal to water increased system temperatures to 379*F (446K) at about 4200
- s. There af ter Loop B flow limitad further temperature increases by heat transfer into SGB. Since AFW was throttled, general system cooldown was not experienced for the renainder cf the sequence. The temperature reached 388 F (470.7K) at 7200 s. The temperature profile is presented in Figure J.3.
i D. Pressure Extrapolation HPI flow ceases at around 1000 s in Transient 1. However, some additional charging pump' flow wonid be required to maintain pressure until SGA dryout is complete. Primary system pressures from Transient 1 were used out to 1400 after which .the coolant swell model was employed to predict J.22
~ . - - , . - - - - - - . . . . - - - . . . -
t-repressurization to the PORV set point, 2400 spia (16.6 MPa) at 6000 s. The pressure profile is presented in Figure J.3. 1 E. Heat transfer coefficient extrapolation l The heat transfer coefficient da ta from Transient 1 was used out to 500 l
\
l s at which time the value had declined to the final corrected value of 400 Btu /hr ft 2o F (2270 W/m 2 g), C.3.3.2 Sequence 1.2 A. Basis - Transient 1 and Sequence 1.1 B. Dep'artures from Basis i This sequence differs from sequence 1.1 only in the failure of the operator to throttle AFW to intact SGB. See Section J.3.3.1 for other details. C. Temperatures extrapolation This se quence will follow the temperature profile of sequence 1.1 out to the point at which primary flow in stagnated loop B is restored. The minimum temperature of 253*F (396K) at 1308 s was taken. The effect of unthrottled AFW on the stagnated loop B is to increase the secondary inventory and reduce the average temperature of the loop. Since the loop is isolated from the balance of the primary due to temperatur e induced I flow stagnation, the temperature reduction in the stagnated loop occurs concurrently with the reheating of the flowing loop and the balance of the primary. Based on data from Transient 1, the t emperatur es of the l flowing and stagnant loops becomes equal (3400 F) at around 3500 s. Since l l J.23
. o ! the now flowing loop B is still receiving AFW, it cools the primary and 1
causes stagnation of loop B. Le downconer temperature sags to and remains nearly constant at 340*F (444K) for the remainder of the transient. He temperature profile for sequence 1.2 is presented in Figure J.5. D. Pressure extrapolation The coolent swell model does not predict water solid primary condition for this se que nce , hence the PORY set point pressure is not achieved. The prediction of a final pressure of 2000 psia (13.8 MPa) is based on the 1 i fractional rise of primary average temperature relative to the temperature i i required for full repressurization. This method is imprecise as to timing. l The pressure could reach this maximum value as early as 3500 s compared to 7200 s as shown in Figure J.5. l E. Heat transfer coefficient i ! The downconer heat transfer coefficient data from Transient 1 was used ont I to 165 s at which time the value had declined to the final corrective value I ! of 400 Btu /hr ft 2 F (2269 W/m 2g), f ! J.3.3.3 Sequence 1.3 l A. Basis - Transient 1 1 B. Departures from Basis 4 The throttling of AFT to intact SGB in sequence 1.3 is the only departure from Transient 1 conditions. l l j J.24 i - . .. -- .- - __ . . __
t C. Temperature extrapolation The throttling of AFW to the intact and stagnated SGB occurs at about 2000 s at which time the average temperature in the steam generator is 379 F (466K). Primary flow in loop B will not be restored until the primary exceeds this t emperatur e. Transient 1 data were used to 3000 s. Af ter which extrapolation proceeded as sisting charging pump flow of 18.3 lb/s (8.3Eg/s), loop A flow of 540 lb/s (245 Kg/s) and wall heat treasfer varying linearly from 500 Btu /s (0.53 MW) to 300 Btu /s) 0.32 MW) over the period 3000 s to 7200 s. As seen in Figure 4.3 in Section 4.5.2.1, the 4 extrapolation closely follows Transient 1 to 4000 s and then proceeds to a final temperature of 346 F (447K) at 7200 s. The overall cooling effect of charging pump flow limited primary reheating such that loop B flow restoration does not occur. As in Transient 1, the minimum temperature of 253*F (396K) was observed at 1308 s. The temperature profile is presented in Figure J.6. D. Pressure extrapolation
- There are no mechanisms to cause the pressure profile for the se quence to vary from that of Transient 1. Therefore, Transient 1 data were used. See Figure J.6.
E. Heat transfer coef ficient The heat transfer coefficient conditions for seque nce 1.1 are equivalent 2 o F (2269 W/m 2g) g, to this sequence . The final value of 400 Btu /hr f t l l obtained by 165 s into the sequence. I J.25 l
- . l
_. ._ _ - .- . _ A
1
. . j i
l J.3.3.4 Sequence 1.4 The specified conditions for sequence 1.4 are identical to the case modeled in Transient 1. J.3.3.5 Sequence 1.5 A. Basis - Transient 4 B. Departures from Basis The operator action te turn of the charging pumps on attainment of HPI shut of f pressure is the only departure of sequence 1.5 from Transient 2. C. Temperature extrapolation The temperature profile for the sequence is as stened to be identical to Transient 4. See Figure J.7. D. Pressure extrapolation The pr es sure profile for the se quence is identical to that of Transient 4 out to 1125 s where the HPI flow limiting pressure of 1285 psia (8.9 MPa) is achieved and the operator is assumed to shut off the charging pumps. Since there is no significant heating of the system af ter this time, the pressure remains at this final value as depicted in Figure J.7. E. Heat transfer coefficient The heat transfer coefficient profile is identical to that of sequence 1.1 (Section J.3.3.1) . J.26
I J.3.3.6 Sequence 1.6 A. Basis - Transient 5 B. Departures from Basis These sequence s specify operator action to turn off charging pump flow on achievement of HPI shut off pressure. Transient 5 does not. In Transient 5 the operator kills AFW to both steam generators at 480 s. In sequence 1.6 the operator limits the AFW to 160 gal / min per SG. C. Temperature extrapolation There is suf ficient SG secondary water inventory present to cool the primary down near to 212*F (373K) given a large break and MSIV failure. Therefore, sequence 1.6 is essentially equivalent to Transient 5. Figure J.8 depicts the temperature data for Transient 5 to the end of the calculation, 3300 s, and the extrapolation to 212 F (373K) at 7200 s. D. Pressure extrapolation The pressure profile for Transient 5 was used to 1250 s, at which point the primary reaches the final pressure of 1285 psia (8.6 MPa) as shown in Figure J.8. E. Heat transfer coefficient i The heat transfer coefficient drops from its initial value of 3700 Btu /hr ft 2o F (2100 W/m2 K) after RCP trip at 62.2 s to a final value of 400 Btu /hr 2 l ft at 250 s. i i ( J.27 ? l c - I
- ~. - - -
. o J.3.3.7 Sequence 1.7 This se que nce features a large break in a steam line and failures to isolate feed flow to the broken SG, to limit repressurization by turning of f the charging pumps, and to throttle AFT to the intact stema generator.
This seque nce is as assigned the temperature pressure and heat transfer coefficient profiles of Transient 4 due to the close similarity of the transients. J.4 Small Steam Line Break at Hot Zero Power J.4.1 Description of Sequences, The sequences for the small steam line break at hot zero power are initiated by a TBV size break, 0.52 ft 2 (0.048 m ), located in a steam line down stre:m of the flow restricter and upstream of the MSIV. The system is initially at steady state condition and nominal steam generator levels for hot zero power. The decay heat level is 9.3 8 MW, corresponding to 100 hours after shutdown. The eight specified sequences for this initiator are listed in Table J.5. The differences in the se quence specifications involve MSIV operation, isolation of AFT to the affected steam generator, HPI operation, and i operator action to turn off the chsrging pumps and throttle AFW flow. J.4.2 Bases for Extrapolation LANL Transient 1 (1 ft 2 steam line break at HZP) comes closest to matching the conditions for the small (TBV-sized) steam line breaks at hot zero J.28
.._- ---_ -._- - . _ _ _ . - . - _ - _ - _ _ - -- - - . . - . _ - = . - . - _ . - .
. o e
power. The primary and secondary water inventory data, loope d flows prior to RCP trip and after RCP trip, and syst een average temperatures corresponding to SIAS and SGIS from Transient I were used to construct the i initial response of the plant to the small break at hot zero power. To check the accuracy of the projection assanptions, duplication of Transion: ) 1 was a t t empt ed. SGIS was predicted at 16 to 18 s, within 1 to 2 s of the Transient i value, and SIAS was predicted at 56 s versus 54 s for TRAC. 1 Next, the same flow and system testperature relationship were applied to the 0.52 ft 2 small break. The resulting predictions were SGIS at 55 s and SIAS at 90 s. The RCP's were assumed to be tripped at 120 s. Natural cirentation flow data from Transient 1 was used for predictions later in the sequences. 1 5 J 4.3 Results and discussion I J.4.3.1 Sequences 2.1 and 2.2 l A. Basis - Transient 1 1 i B. Departures from Basis l The flow and thermal data in Transient I were used in the extrapolation i l of these small break sequence s. Early response s of both of the named sequences are similar. Later responses are determined by associated system f ailures and operator actions. Neither of the sequences will repressurize completely due to operator action to shut of f the charging pumps. Sequence I 2.2 lacks operator action to throttle APW to SGB. 1 J.29 i-- .- _ , _ _ _ _ _ _ _ _ _ __ _ _ . _ _ l
l C. Temperature extrapolation Using Transient 1 data as discussed above, Sequence 2 .1 was extrapolated ' over the period 0 to 7200 s as shown in Figure J.9. Events along the course of the extrapolation include SGIS at 55 s, SIAS at 90 s, RCP trip j at 120 s with establishment of natural circulation assumed at 220 s. AFAS occurs at 360 s based on SGA inventory, but prior isolation of AFW to that SG directed all flow to SGB until throttling at 3000 s on level. The minimum temparature of 250*F (394K) occurs at SG dry out (2240 s). Temperature recovers af ter this point with flow in loop B starting at about 4000 s. The system reheats to a final temperature of 348*F (449K) . D. Pressure extrapolation Pressure is assumed to drop from 2250 psia (15.5 MPa) to 1740 psia (12.0 MPa) at 90 s (SIAS) and then to 1200 psia (8.3 MPa) at 180 s. Since the system cooldown is less severe than s in Transient 1 further depressurization was not assumed. The system is assumed to repressurize to 1285 psia (8.9 i MPa) by 1000 s at which point the charafss pumps are turned off. HPI flow maintains pressure at this level until SGA dryout. The coolant swell model s
\
predicted a final pressure of 2210' psia (15.2 MPa):.s3 i 70 s due to system reheating. (See Figure J.9.) ; l' E. Heat transfer coefficient The downcomer heat transfer coefficient 'fdr$ from 4230 Btu /hr ft 2 op 2 (24000 W/m K) at the tripping of the ICP'i (120- s) to a final value of 400 Btu /hr ft 2o F (2270 W/m2 K) b ' g, ' 250{s.
,. ,o ,
l.? r Y, / y.3 0 -
, > i A
1 p p ar---- - - , , , - . - - - , , - . . , , , -- - , - , , -
l J.4.3.2 Sequences 2.3 and 2.4 , I A. Basis - Transient 1 and extrapolations for Sequence 2.1 B. Departures from Basis The failure to turn off the charging pumps in sequence s 2.3 and 2.4 and subsequent early repressurization separates thes two sequence s from the group discussed in Section J.3.4.1. Sequence 2.3 include s throt tling of AFT to the intact sto me generator while sequence 2.4 does not. De f ailur e to throttle will results in colder intact loop t empe r atur e s , an earlier restoration of loop flow, and lower final temperatures. Herefore, sequence 2.4 was selected to represent the group. C. Temperature extrapolation The events for sequences 2.3 and 2.4 follow basically the same path as for se que nce 2.1 except for scue minor differences due to continued charging pump flow, he time of SGA dry out shifts 40 s later to 2280 s at which point the minimum temperature of 242*F (390 k) is obtained. Some reheating occurs until loop B flow resumes at 3500 s. He system then cools to its final value of 269'F (405 k) at 7200 s. Figure J.10 gives the temperatur e I profile for the group. D. Pressure extrapolation Due to similarities in cooldown rates and related conditions the pressure j profile for Transient I was assigned to sequence 2.4 (see Figure J.10) . J.31
O 4 E. Heat transfer coefficient The heat transfer coef ficient profile for sequence 2.4 is identical to that for sequence 2.1. J.4.3.3 Sequence 2.5 A. Basis - Transient 1 and Sequence 2.1
. B. Departures from Basis The failure to isolate AFW to the broken SGA in sequence 2.5 is the only departure from sequence 2.1.
C. Temperature extrapolation The initial response of Sequence 2 .1 applies for the first 360 s of the transient. Failure of AFW isolation to SGA assures continual cooldown throughout the duration of the se que nce . After 1000 s the continued flow of AFW adds to SGA inventory faster than water is lost to blowdown. The t i minimum temperature of 216'F (375 k) occurs at 7200 s as shown by Figure J.11. D. Pressure extrapolation The shut down of the charging pumps prevents any repressurization of the system above 1285 psia (8.9 MPa) since the system does not reheat. The pressure response is in Figure J.11. l J.3 2 l l
E. Heat transfer coefficient The downconer heat transfer coefficient response is the same as that for sequence 2.1 (Section J.3.4.1) . J.4.3.4 Sequence 2.6 A. Basis - Transient 1 1 B. Departurea from Basis l The main departures from Transient 1 in this sequence are the failure of both MSIVs to close and the operator action to turn of f the charging pumps. Since neither loop will stagnate and both loo'ps will be undergoing significant cooldowns, the data for the flowing loop in Transient I was
,v W applied to this sequence.
C. Temperature extrapolation This se quence begins to diverse from sequence 2.1 soon after the MSIV failure to close at 55 s. SIAS and RCP trip do not shift significantly from their respective 90 s and 120 s values in Sequence 2.1. However, feeding this small steamline break with both steam generators delays AFAS to 1250 s. AFT flow exceeds the break stream flow for the remainder of the transient, yielding a not increase in steam generator water inventories. As shown in Figure J.12 the minimum temperature of 215'F (375 k) is obtained at 7200 s. J.33
o . D. Pressure extrapolation The* pres sur e profile for sequence 2.6 is assumed to be identical to the profile for sequence 2.5 (see Figure J.12) . E. Heat transfer coefficient Due to similar timing of the RCP trip the profile for this case is assumed
' to be identical to that of sequence 2.1.
L J.4.3.5 Sequence 2.7 A. Basis - Transient 1, Sequence 2.6 B. Departures from Basis l In sequence 2.7 the charging pumps are not turned off, leading to full repressurization of the primary. His is the only difference be twe en sequence 2.7 and 2.6. C. Temperature extrapolation The cooldown due to sta an blowdown from the break will dominate the se quence . Therefore, the temperature profile of sequence 2.6 was assigned to this case. He profile is presented in Figure J.13. D. Pressure extrapolation Due to the similarities between this sequence and Transient 1, the pressure l profiles from Transient 1 was applied to this sequence as shown in Figure i
- 3.13.
l l J.34 l l
r i
- E. Heat transfer coefficient
! The heat transfer coef ficient profile for this sequence is assumed to be
- the same as that of sequence 2.1.
J J.4.3.6 Sequence 2.8 A. Basis - Transient 1, Sequence 2.6 B. Departures from Basis Sequence 2.8 differs from 2.6 only in the operator killing AFW at 300 s, in essence throttling the AFW system before it comes on. Like 2.6, sequence 2.8 dif fers from Transient 1 by the operator turning off the charging ptusp s. C. Temperature extrapolation This se quence is identical to se quence 2.6 out to 1250 s when AFAS is generated. Since AFT was killed previously by the operator, there is no AFW supplied to either SG. There : s still nearly 200,000 lb of water in the secondary system available for blowdown, through the break. The blowdown rate also decreases due to the decreasing temperature and pressures in the secondary system such that SG dryout does not occur prior to 7200 s. As shown in Figure J.14 the minium temperatur e of 224*F (380 k) occurs at the end of the period. The temperature difference between sequences 2.6 and 2.8 is due to the influence of AFW addition. I. 3.35
D. Pressure extrapolaticn The pressure profile (Figure J.14) for this seque nce is assumed to be identical to that of sequence 2.5 due to operator action to turn of f the charging pumps coupled with no reheating of the system. E. Heat transfer coefficient The heat transfer coefficient profile for this sequence is assumed to be the same as that of sequence 2.1. J.5 Large Steam Line Break at Full Power J . 5 .1 Description of Sequences The se que nce s for large steam line break at full power are initiated by a 1 ft (0.0929 a ) break in a steam line downstream of the flow restricter and upstream of the MSIV. The system is initially at steady state at full power. Both the reactor and the turbine s are assumed to trip coincident with the appe arance of the break. The system du ay heat function is assumed to be 1.0 times the ANS standard. The nine specified sequences for this initistcr are listed in Table J.6. The differences in sequence spe cifica tion involve MSIV operation, MFW runback af ter trip, ADV operation, HPI operation, and operator actions to turn off the charging pumps and throttle AFW flow. J.36
o . J.5.2 Bases for extrapolation LANL Transient 2 se rve s as the basis for evaluation of large steam line breaks at it.11 power. Transients 8 and 9 also provide information on SG [ overfeeds useful for evaluation of sequences 3.8 and 3.9. The similarities between late transient natural circulation flows in the basically different Transients 2, 8, and 9 also lends credence to the applicability of transient data in sequence extrapolation.
=
J.5.3 Results and Discussions J . 5 .3 .1 Sequences 3.1, 3.2, 3 .3 , and 3.4 As seen in Table J.6 sequences 3.1 through 3.4 feature all the conbinations of operator success or f ailure to turnoff the charging pumps upon attainment of HPI shut of f pressure and throttling of AFW to the intact SG. With both of these f ailures, sequence 3.4 is identical to the specifications for Transient 2 and is represented by that calculation. As noted above, sequences 3 .1, 3 .2, and 3.3 will differ slightly from Transient 2. Variations will be typically in the direction to reduce PTS risk, i.e., higher tempe ratures and lower pressures. Since Transient 2 itself does not represent any great risk due to its high temperatures, assignment of Transient 2 P, T, and h profiles to seque nce s 3 .1 through 3.4 does not represent any significant error. l J.37 i l \ l l
. -+ + w w --,,g. .-p+ - - - .--g-m-*s-w- - - - - - - - . - - - -
J . 5 .3 .2 Sequence 3.5 , A. Basis-Transients 2 B. Departures from Basis In this se que nce the AFW is not isolated to the broken steam generator. The operator is assumed to turn off the charging pumps and to throttle AFW to the intact steam generator on reaching +22 inch level indication. C. Temperature Extrapolation The tempe rature pro file for Transient 2 was used out to 400 s where the brokon SGA dries out. At SGA dryout the downcone r temperature is , 371 F (461K). (De tailed examina tion of the 'IRAC data for Transient 2 revealed that AFW was directed to the broken SGA rather than to the intact SGB during this period. This modeling error was corrected in the restarted TRAC calculation after 400 s. However, this modeling error matches the specification of sequence 3.5, thus enhancing the accuracy of the extrapola tion. ) After 400 s, the extrapolation continues with an AFW flow of 44.5 lb/s (20.2 Kg/s) to the af fected steam generator. The boiling and blowdown of this AFW flow is the sole cooldown mechanism for the remainder of the seque nce . Decay heat input exceeds the cooling capacity of this AFW flow out to 2000 s, with the system reheating to 411*F (483K) . Af ter 2000 s, the cooling exceeds decay heat and wall heat inputs such that the system cools to a minimum of 240'F (388K) at 7200 s as shown in Figure 3.15. 3.38
i 8 i D. Pressure extrapolation The pressure profile for this sequence follows that of Transient 3 rerun (3A) out to SGA dryout at 400 s. The limited cooling after this point allows the primary to repr es suriz e to the HPI flow limiting pressure of 1285 psia (8.9 MPa) by 640 s when the operator is assuesd to turn off the charging pumps. HPI pump action will prevent depressurization for the remainder of the sequence as shown in Figure J.15. E. Heat Transfer Coef ficient f The downconer heat transfer coefficient profile is assmed to be similar to that for Transient 2. The initial value of 4230 Btu /hr ft 2o F (24000 , 2 w/m K) holds out to the tripping of the RCPs at 45 s. De coefficient drops to the assumed minimum value of 400 Btu /hr ft2 F (2270 w/m 2g), J.5.3.3 Sequence 3.6 l A. Basis - Transient 2 - response to large steamline break at in11 power
)
l Transient 5 qualitative response to double MSIV failm e during large steam line break i B. Departure from Basis Sequence 3.6 features failure of both MTV's and throttling of AFT to a rate of 160 gal / min to each steam generator. The charging pumps are also i shut off upon attainment of HPI shut off pressure. De differences between Transient 2 and sequence 3.6 are assmaed to parallel the di (fer ence s between Transient 4 and Transient 5 relating magnitude of total loop flow s l , for flowing-stagnated loop conditions versus symmetric loop conditions. In J.39
-__ = _ - - - - . --- -
a e the case of Transients 4 and 5, roughly the same levels of total loop natural circulation flow are obtained. The same behavior is assumed for evaluation of Sequence 3.6. C. Temperature extrapolation Extrapola tion commenced af ter 40 s in Transient 2 with parameters cho sen to reflect the failure to isolate SGB, the sustained flow of AFW, and the symmetric nature of primary loop flows. The timing of basic events such as AFAS, SIAS, and SGIS did not change for the se que nce . He failure of the MSIV's effectually doubled the secondary water inventory available for blowdown. As shown in Figure J.16 the tempe rature decreases out to 1700 s and then stabilizes momentarily due to balancing of declining loop flow and temperature with wall heat transfer to yield relatively constant i downconer fluid temperature. This occurs although the cold les temperature is continuing to decline. About this same time a balance has been struck be tween the energy floe out the break- and the energy input by decay heat and wall heat transfer. The steam generator temperature, which determines the break energy tiow, declines slowly after this point in response to s
; declines in wall heat transfer and decay heat input . This steam generator 4
temperature response drives the downconer temperature for the rest of the transient. The steam generators do not dry out. The AFW flow overwhelms the break flow by 5500 s causing partial refill of both steam generators for the rest of the se quence . The minimum tempe ratur e of 226*F (381 k) ocents at 7200 s. J.40 l P % - - ---g ,-----y-v- - - --w .- sr+ e --
,y---.--,w- --
,----,-y----,-w , , - - , - , , , , , - , - ,--,.-w-,,,y ---ce w
o . D. Pressure extrapolation The bulk of the cooldown occurs prior to 600 s and the pressure response for Transient 2 is used to this point. In Transient 2 the HPI shut off pressure of 1285 psia (8.9 MPa) has been reached and .the charging pumps are assumed to be shut off at this point. The HPI system will maintain the primary at this final pressure for the remainder of the transient (see Figure J.16) . E. Heat transfer coefficient The heat transfer coef ficient holds its initial value, 4230 Bta/hr f t 2o p i (24000 W/m2 K), out to RCP trip at 45 s. RCP coast down and an initially very strong natural circulation flow delay to 600 s the attainment of the final value of 400 Btu /hr f t 2o F (2270 W/m 2g), J.5.3.4 Sequences 3.7 A. Basis - Transient 2 - response to large steenline break Transient 12 - response to stuck open ADV B. Departures from Basis I Sequence s 3.7 features operator action to turn off charging pamps and throttle AFW to the intact steen generator (SGB) . This seque nce also l features a stuck open ADV on the " intact" loop (SGB) which is not isolated. Transient 12 data were used to calibrate the cooldown model for ADV flow versus steen generator temperature. , J.41 l
C. Temperature extrapolation i The severe cooldown at the beginning of the transient was expected to overwhelm the cooldown du to the ADV. Therefore, Transient 2 data were l used out to 450 s when SGA has dried out. The minimum temperature of 399'F l ! (477 k) occurs at 300 s. Stagnation of loop B allows the ADV to cool SGB so that loop B flow begins at around 500 s. A balance between break energy flow, AFW heating load, and decay heat input is established by 1500 s and I persists through to 7200 s yielding a final temperature of 432'F (495 k) as shown in Figure J.17. Through AFW was refilling SGB through the later 1 stages of the transient, the +22 inch limit was not achieved, so AFW was i not throttled. D. Pressure extrapolation The pressure response for Transient 2 was used for this sequence out to i the attainment of HPI shut of f pressure at 500 s as shown in Figure J.17. t E. Beat transfer coefficient The heat transfer coef ficient profile is taken to be identical to that of sequence 3.6. i I.S .3.5 Sequences 3.8 and 3.9 A. Basis - Transient 2 Transient 8 - response to 2 loop MFW overfeed Transient 9 - response to 1 loop MFW overfeed i. J.42
_ - _ _ . . . . _ _ _ _ _ _ . _ __ __ _ __ . _ _ _ _ _ - - - _ _ _ . _ _ _ _ . _ . . _ _ _ . _ _ _ _ ._~ . I i B. Departures from Basis ! Sequences 3.8 and 3.9 include operator actions to turn off charging pumps J and to throttle AFW to the intact steam generator. Sequence 3.8 specifies
! a MFW overfeed to the broken steam generator. Sequence 3.9 specifies i overfeeds to bo th steam ge nerator s. The effect of the double overfeed ,
i { in 3.9 is masked by the stagnation of loop B and the generally minimal
- PTS impacts of overf eeds as evidenced by Transients 8 and 9. Therefore, i
! sequence 3.9 is grouped together with sequence 3.8 for evaluation. l C. Temperature extrapolation l The overfeed of the broken steam ge ne r ator (SGA) serve s to extend the i period for dryout and produce lower minime temperatures. De closure I of the NFIV spon SGIS at 44 s stops the overfeed, leaving SGA with an additional 80,000 lb (36400 kg) of water relative to the inventory for i Transient 2 at the same instant. Projection of temperature trends based l on Transient 2 loop flow data yielded a minimum temperature of 276'F (408 k) at 800 s, when SGA dries out. With the collapse of the cooldown i I mechanism, the downoomer is soon reheated to the hot leg temperature as the loop flow sweeps warm liquid into the downoomer. Following this initial rise, a slower rise due to general reheating of the system takes place over the rest of the se quence . By 5000 s the temperature rises to 534'F (552 l k) where the ADV system is used to control temperatur e, ne temperature profile is presented in Figure J.18. l ? r l l 1.43 l l l --., ._ _ _ . . _ . _ . . - . - _ _ - . . _ - - - _ - . - . . _ - _ , , . , , _ . - - . . _ - , _ _ _ . . . _ , - . _ _ - - . _ , . _ _ . - .
9 D. Pressure extrapolation
. I The pressure profile from Transient 2 is used out to 450 s due to the j similarity of conditions. Repr es suriz ation to 1285 psia (8.9 MPa) is assaned to be delayed to 950 s due to the rapid cooldown prior to SGA dryout. The coolant swell model predicts full repressurization to 2400 psia (16.6 MPa) by 1800 s as shown in Figure J.18. )
E. Heat transfer coefficient The downconer heat transfer coefficient profile for these se quence s is assimied to be identical to that of sequence 3.6. i 4 l 1 i ? i i 3.44
J.6 Small Steam Line Break at Full Power J.6.1 Description of Sequences l The sequences for small steam line break at full power are initiated by a TBV sized break, 0.52 ft2 (0.048 m2 ), located in a steam line downstream of the flow restricter and upstream of the MSIV. Le system is initially at steady state at in11 power. Both the reactor and the turbines are assumed to trip coincident to the appearance of the break. The system decay heat function is assumed to be 1.0 times the ANS standard. The 12 specified sequences for this initiator are listed in Table J.7. The dif ference s in se quence spe cifica tion involve MSIV operation, MFW runback j af ter trip, ADV operation, isolation of AFT to the affected steam generator, HPI operation, and ope rator action to turn off the charging pump s and throttle AFW flow. J.6.2 Basis for Extrapolation I LANL Transient 2 serves as the basis for evaluation of sequences involving small (TBV sized) steam line breaks. It is assumed that the loop flow data in this transient is applicable to the smaller break transients although cooldown rates are lower. For a 1 ft2 break the cooldown model predicted i timing for SIAS, SGIS, and AFAS very similar to the timing predicted by l l TRAC. With the Transient 2 loop flow trends and the TSV sized break, the cooldown model predicts SIAS at 25 s, SGIS at 85s and AFAS at 230 s for 1 loop blowdown or 410 s for blowdown of both SG (i.e. MSIV f ailure) . All of
, the sequence s use the same initial extrapolation but proceed to different end points based on specified conditions and actions.
l J.45
1 J.6.3 Results and Discussion J.6.3.1 Sequences 4.1 and 4.2 ^ A. Basis - Transient 2 B. Departures from Basis Sequences 4.1 and 4.2 require the operator to shut of f the charging pumps when the primary repressurizes to the HPI shut of f pressure. Sequence 4.1 also require ope rator action to throttle APW to the intact SG while sequence 4.2 does not. Sequence 4.2 represents the most severe overcooling sequence of the two and will be used to characterize the group. 1 C. Temperature Extrapolation i The initial extrapolation described in Section I.6.2 was extended to SGA dryout at 860 s at which point the miniaan temperature of 337'F (442 k) i was obtained. Displacement of cold fluid by hot les flow and reduction in HPI flow caused the downconer temperatur e to rebound. De primary becomes hotter than the SGB secondary at about 1300 s causing restoration
- of na tur al circulation flow in loop B. Drottling of AFW in sequence 4.1 would occur later and therefore wonid not affect the course of the transient. Under the infinance of decay heat the system reheats to 534'F (552 k) by 600 s. The temperature profile is presented in Figure J.19.
D. Pressure Extrapolation The pressure during the sequences will drop an initial vains of 2250 pois (15.5 MPa) to 1784 psia (12.3 MPa) at SIAS (25 s) and then to an assumed l 3.46 l
minimum of 1200 psia (8.3 NPa) at 60 s. He primary is assumed to repressurize to the HPI flow limiting pressure of 1285 psia (8.9 MPa) when SGA dries out (86 0 s). System reheating and coolant swell causes water solid conditions at 6500 s, at which point the pressure is at the PORY set l' point pressure 2400 psia (16.6 MPa) . The pressure profile is presented in Figure J.19. E. Heat Transfer Coefficient
" lie downconer heat transfer coefficient holds its initial value of 4230 2 2 Btu /hr ft F (24000 w/m k) until RCP trip at 55 s. By 250 s the coefficient has dropped to its final value of 400 Btu /hr f 2o t F (2270 w/m 2 k).
J.6.3.2 Sequences 4.3 and 4.4 A. Basis - Transient 2, Sequence 4.2 i
; B. Departures from Basis The sequences are similar to sequences 4.1 and 4.2 except that the charging pimps are not shut off, leading to complete repressurization early in the transient. Sequence 4.4 also features failure to throttle AFW to the intact SG and is chosen to represent this group.
l C. Temperature Extrapolation l These sequences are virtually identical to sequence 4.2 out to SGA dryout l ( 86 0 s). ne continued addition of charging pump and AFW flow supplies
)
excessive cooling to the system, limiting the final temperatur e to 4860F 3.47 1 . l I , _ _ _ _ _ _ _ - - ~ ,-_, _ _ _ - - _ . - _ _ - - - _ - - .--
- - . - . - . . _ - - . _-= - -_ _. .
o . l l u ($25 k) as shown in Figure J.20. n e minimum temperature of 337'F (442 k) occurs at 860 s, the same as in sequence 4.2. l D. Pressare Extrapolation l This sequence follows the same profils as sequence 4.2 out to 860 s where the operator is assmed to fail to shut off the charging pumps. De j combination of coolant swell and charging pump flow increases the primary pressure to 2400 psia (16.6 MPa) by 2000 as shown in Figure J.20. his i rate of repressurization is consistent with the rate of the more severs tr ansients. E. Beat Transfer Coefficient i j The downconer heat transfer coefficient profile for sequence 4.4 is assmed i to be the same as that for sequence 4.2. i i j J.6.3.3 Sequence 4.5 A. Basis - Transient 2. Sequence 4.2 B. Departures from Basis -l ] Sequence 4.5 features operator actions to turn off charging pumps and to throttle AFW to the intact SG. However this sequence also specifies ! f allara to isolate AFW to the affected steam generator. l C. Temperature Extrapolation no initial extrapolation for sequence 4.2 applies out to 230 s. At this I point AFAS initiates AFW flow to the broken SGA. ne addition of AFW flow 7 J 48 l i
to SGA inventory delays SGA dryout to 1300 s where the temperature is 314*F (430 K). Thereafter, break mass flow and system cooldown are limited by AFV flow to SGA. ne arrival of the break flow limitation is signalled by a small temperature increase to 1350 s (see Figure J.21) . The collective action of continued AFW flow and declining decay heat cause a cooldown to 281'F (411 k) at 7200 s) . D. Pressure Extrapolation The primary pressure declines to an assumed minimum of 1200 psia (8.3 MPa) by 60 s and is assumed to recover to the HPI shut of f pressure of 1285 psia i l (8.9 MPa) at 1300 s, where the operator kills the charging pumps. Despite continued but slow cooldown, the pressure is assaned to remain at 1285 psia (8.9 MPa) by occasional flow of HPI. (See Figure J.21) . - i i E. Heat Transfer Coefficient i The downconer heat transfer coef ficient profile is assumed to the same as that of Sequence 4.2. J.6.3.4 Sequence 4.6 A. Basis - Transient 2 B. Departures from Basis t In sequence 4.6 both MSIV's fail to close, resulting in blowdown from both steam generators. The operator is assmed to limit AFW flow to 160 gal / min per SG and turn of f the charging pumps. nose changes were extrapolated assm ing symmetric loop flows with total flo,s equivalent to total loop l J.49 I
- - - , , - - g ,, -._. - - , - , - . - - * - *- .y- , _ _ - , - -m , - - g- 9-- - - - ,~w +-e- --
flows from Transient 2. Under these conditions SIAS, and SGIS occur at 25 and .85 s, as they do in sequence 4.2. With, blowdown from both stems SG water inventory declines out to generators, AFAS is delayed to 410 s. There is a slight increase 5000 s where AFW begins to exceed break flow. , in SG water inventory out till the end of the sequence. he cooldown curve shown in Figure J.22 is dominated by the heat balance between decay heat and wall heat transfer inputs and losses due to AFW heating and break flow. The minimum temperature, 225*F (397 k) occurs at 7200 s. D. Pressure Extrapolation i The pressure declines to the assumed lowest value of 1200 psia (8.3 MPa) i by 60 s and is further assaned to rise to HPI shut of f pressure,1285 psia (8.9 MPa) by 1300 s, by which time the bulk of the cooldown and coolant shrinkage has concinded. The action of the HPI system will maintain pressure at this final value as indicated in Figure J.22. 1 1' E. Heat Transfer Coefficient
- I The downconer heat transfer coef ficient profile is assaned to be the same as that for sequence 4.2.
j J.6.3.5 Sequences 4.7, 4.9, and 4.10 l A. Basis - Transient 2 - Systen Response to Steam Line Break l B. Departures from Basis Sequence 4.7 features a MFW overfeed to the affected steam generator and ope rator actions to throttle AFW to the intact SG and to turn off the J.50 I
= _ _ ._ _ __ _ . _ _ - __. - . .. __
charging pumps. The loop flow data from Transient 2 and the overfeed flows and temperatures from Transient 9 were applied to these cases. In sequence 4.9 the AFW is not throttled to the intact steam generator. In sequence 4.10 both steam generators are overfed. However, as illustrated in Transients 8 and 9, NFW overfeeds to intact steam generators does not result in significant overcooling events and so, inclusion with sequence a 4.7 was assumed to be a valid representation. 1 C. Temperature Extrapolation The overfeeding of SGA concurrent with the opening of the smell steamline break results in SGIS being generated at 55 versus 85 s for sequences where NFW overfeed does not occur. Approximately 100,000 lb (45,400 kg) of excess water is fed to SGA. This added inventory extends SGA dryout to 1700 s, at which time the minimum temperature of 318"F (432*F) is obtained. The system reheats to 534*F (552 k) by 3800 a where it is maintained by ADV action on SGB. Figure J.23 presents the temperature profile for this sequence. D. Pressure Extrapolation The pressure drops to the assumed lowest value of 1200 psia (8.3 NPa) by 60 s. The charging pumps are shut off as the system reaches EPI flow i limiting pressure at 1600 s. The rapid reheating causes coolant swell and water solid conditions by 3500 s as predicted by the coolant swell model. The pressure profile is shown in Figure J.23. } 1.51 ' l
E. Heat Transfer Coefficient The downconer heat transfer coef ficient profile is assmused to be the same as that for sequence 4.2. J.6.3.6 Sequence 4.8 Sequence 4.8 differs from Segnance 4.7 only in the failure of the operator to turn off the charging pumps. Because decay heat wonid overwhelm any cooling due to charging pump flow, this failure was j udged to have i j negligible impact on temperature profile, he pressure profils is the same j as that for Sequence 4.7 except that the action of the charging pumps will be to advance the timing of repressurization to the PORY set point, 2400 psia (16.6 MPa), at 2400 s versus 3500 s for Sequence 4.7. H is difference j d is illustrated in Figure J.24. De downoomer heat transfer coefficient profile is assumed to be the same as that for Sequence 4.2. J.6.3.7 Sequence 4.11 i A. Basis - Transient 2 - response to steam line break Transient 12 - behavior of stuck open ADV I B. Departures from Basis i This sequence features operator action to turn of f charging pumps and to i throttle AFT to the intact stoma generator. His segnance also specifies a stuck open ADY in the line opposite the break. His vaine is not isolated. Since both steam generators are blowing down, provision was made l for throttling AFW to 160 gal / min per line if no SG differential pressure J.52
)
i J develops to initiate AFT isolation to the low pressure SG. Due to the disparity of " break" sizes, isolation is expected to occur. l C. Temperature Extrapolation The ope n ADV does not perturb the timing of SIAS relative to that of Sequence 4 .2 . SGIS is advanced to 80 s and AFAS is advanced to 140 s. 1 The signal to isolat, AFT to the broken loop is expected before AFAS is generated. Therefore all AFF is directed to the loop with the stuck open ADV. The broken steam generator, SGA, dries out at 1650 s, where the system reaches its minimum temperature of 297'F (420 k) . Decay heat input is greater than cooling due to ADV flow and AFW heating in SGB, so the downconer temperature begins to recover. SGB begins to refill also, but never reaches the level where APT is throttled. The final temperature is 394*F (474 k). The temperature profile is presented in Figure J.25. l D. Pressure Extrapointion , The primary pressure drops to its assumed minimum of 1200 psia (8.3 MPa) by 60 s and recover to HPI flow cut off pressure of 1285 psia (8.9 MPa) i i at SGA dryout, 1650 s. The charging pumps are turned off at that point. System reheating and coolant swell lead to water solid conditions and full I l repressurization to 2400 psia (16.6 MPa) at 4500 s as shown in Figure J.25.
- E. Heat Transfer Coefficient j The downconer heat transfer coef ficient profile is assumed to be the same as that of Sequence 4.2.
l 3.53
-. . . - - - . - . - - - . _ . _ . . . - , _ . . . . - --_. . -, . ~ . . - - - _ . . . . _ - - -
J.6.3.8 Sequence 4.12 A. Basis Sequence 4.11 ! B. Departures from Basis The only difference be tween Sequence 4.12 and 4.11 is the failure to turn of f the charging pumps. All other descriptions of the sequences in Section J.6.3.7 apply here also. C. Temperature Extrapolation Seque nce 4.11 and 4.12 are identical out to 1650 s where SGA dries out. The minimum t empe ratur e, as shown in Figure J.26, is 297*F (420 k) . The cooling mechanisms of ADV steam flow, AFW heating in SGB, and charging i pimip fJ ow combine to limit reheating relative to sequence 4.11. The final temperature is 363 F (457 K). His 31'F (17 C) difference is due to the incr emental contribution of charging piimp flow to a cooling regime dominated by ADV flow. As noted in Section J.6.3.6, charging pump flow alone will not significantly infinance cooldown. i i D. Pressure Extrapolation l The primary pressure response for this sequence will be the same as for seque nce 4.11 out to 1650 s. Hereaf ter charging pump flow and coolant swell due to reheating will cause water solid conditions and repressurization to the PORY set point, 2400 psia (16.6 MPa) at 2400 s. This response is shown is Figure J.26. J 54
. . i E. Heat Transfer Coefficient The downconer heat transfer coefficient profile for Sequence 4.12 is assumed to be the same as that for Sequence 4.2.
J.7 Reactor Trip at Full Power J.7.1 Description of Sequences
- The sequence s for reactor trip at in11 power are initiated by a reactor i
trip and simultaneous turbine trip. he cause of these trips is not ] i
- specified. These sequence s are characterized by sub sequent failures of components and systems as delineated in Table J.8. The system was at I
steady state at full power prior to the trip, ne decay heat function i j is assumed to be 1.0 times the ANS standard. A total of 42 sequences are identified for this initiator on Table J.8. The
- dif ference s in sequence specification involve MSIV operation, MFW rumback 4
after trip, various combinations of TBV and ADV failure, and operator actions to isolate stuck open ADY's and TBV's, turn of f charging pep s , and throttle AFT flow. C.7.2 Basis for Extrapolation The course of sequences for the reactor trip initiators will be determined I by the types of failures accompanying the initiator. The LANL calculations addressing these situations are: i Transient 2 - 1 ft 2mainsteam line break at full power, Transient 6 - turbine trip with turbine bypass valve stuck open. 1 J.55 l
O e l Transient 7 - turbine trip with one IBV and one MSIV stuck open. Transient 8 - main feedwater overfeed (both SG's), and Transient 9 - main feedwater overfeed to one SG. Transient 6 corresponds exactly to Sequence 5.18. Transient 9 corresponds closely to Sequences 5.2 and 5.3. All other sequences were either estimated e xplicitly or were assigned to a partionlar transient or extrapolated i sequence. ii.e types of cooldown mechanisms, such as saltiple TBV f ailures or ADV-TBV combins tions and mitigating f actors such as SG isolation or SG dryout serve as the basis for sequence assignment. Of the 42 sequence s
; in Table J.8 only 12 were evaluated explicitly. These are addressed separately.
J.7.3 Results and Discussion J.7.3.1 Sequence 5.19 A. Basis - Transient 6
.i B. Departures from Basis In this sequence two DV's f ail to close. ne NSIVs are operable. All 4
other conditions are similar to Transients 6 and 7. C. Temperature Extrapolation For two stnok open TBV's the cooldown model using Transient 6 flow ,
. I paraseters predicts SGIS just before 150 s. he minisse downooser I temperatur e is 498'F (532 k). kSIV closure ends the cooldown and the 0
system reheats above 535 F (552 k) by 600 s. See Figure J.27. 1.56 ' I 4 .
. ._. __ - __ _ - . _ _ . _ ~ . _ ,_ ~_. _. __ _ .
.- _. =- . . __. - . - _ .. - .- ___
D. Pressure Extrapolation i j Dae to similarities in conditions. pressare profile for Transient 6 was used j to 150 s where the pressure is 1700 psia (11.7 NPa). Fall repressarization to 2400 psia (16.6 MPa) is assumed at 600 s as shown in Figure J.27. E. Heat Transfer Coef ficient 1 The dowacomer heat transfer coefficient profile of Transient 6 was used as the heat transfer coef ficient from the TEAC calonistion remains above 420 Ben /hr ft *F (2380 w/m2 k) thron8hout the transient. J.7.3.2 Sequence 5.20 A. Basis - Transient 6
- 4 B. Departares from Basis
) In this sequence three TBV's fail to close. The MSIVs are operable. All j other conditions are simile.r to Transients 6 and 7.
1 C. Temperature Estrapolation j For three stack open TBV's and flow parameters for Transient 6 the l cooldown model predicts SGIS by 120 s with a dowmooner mialaan temperature of 499'F ($32 k) . ne system reheats to $30'F ($$0 k) by 600 s. The temperstare profile is presented in Figure J.28. D. Pressare Estrapolation I l The pressure profile from Transient 6 is need for this sequence (see Figure ! J.28). l l [ e 1.37 i
, . t E. Heat Transfer Coefficient The downcomer heat transfer coefficient profile of Transient 6 is used for this sequence.
I.7.3.3 Sequence 5.21B A. Basis - Transient 7 B. Departure from Basis In this sequence on TBV fails tu close and one MSIV fails to close. 600 s af ter the MSIV failure, the TBV is isolated. He operator is also assmeed to turn of f the charging pumps and throttle AFW at + 22 inches level. C. Temperaturo Extrapolation In Transient 7, a MSIV fails to close at 509 s. 600 s later (1109 s) the stuck open TLV is isolated. Transient 7 data are used to this point, where the minimum t empe rature of 479*F (521K) is obtained. The system reheats ai.rie 530'F (550 k) by 2300 s. (See Figure J.29). D. Pressure Extrapolation The pressure profile for TrtAsient 7 is used for this transient out to 1106 s where the pressure is - 1668 psia (11.5 MPa). ne pressure never
~
drops below the HPI flow cut eff pressure, so the requirement to turn off
,1 .
the charging pumps is ignored. Between system reheating and charging pump j flow, the system repressurizes to the PORVjset point pressure of 2400 psia (16.6 MPa) by 2500 s. ./ o ,
. ), / ? -J 58
.no m
O '
\
E. Heat Transfer Coefficient This se que nce uses the downconer heat transfer coefficient profile for Tran sient 4A out to 700 s where the value is held at the assumed minimum value of 400 Btu /hr f t2o F (2270 w/m k). J.7.3.4 Sequence 5.22 A. Basis - Transients 6 and 7, Sequence 5.19 B. Departures from Basis In sequence 5.22 two TBV's fail to close and one MSIV fails to close. The TBV's are isolated by ope ra tor action 900 s after the MSIV fails. The ope rator also turns off the charging pumps and throttles AFV to the intact SG. The larger break (two TBV's) suggested early SIAS and SGIS so Transient 6 and Sequence 5.19 data were used for the early pressure extrapolation. Transient 7 was applied to later stages of the event. C. Temperature Extrapolation The results of Sequence 5.19 were used out to the initiation of SGIS at 150 s. One MSIV fails to close resulting in assymmetric steam generator i pressure which isolates the APT line to the affected steam generator before AFAS occurs. i Therefore there is no AFW flow to the affected steam generator. Due to the size of the break, the affected steam generator dries out at 800 where the minimum temperature of 397'F (47 6 k) is obtained. Affected SG dryout occurs 150 s before the IBVs are assumed to be isolated, so this action is too late to have any ef fect. The primary J.59
1 l 0 0 reheats to above 530*F (550 k) by 2200 s. The temperature profile is presented in Figure J.30. D. Pressure Extrapolation As in Seq ence 5.19, the pressure is assumed to follow Transient 6 out to 150 s where the pressure is 1700 psia (11.7 MPa) . Continued cooldown due to the stuck MSIV allows the pressure to drop below 1285 psia (8.9 MPa) by 500 s. Further depressurization was not shown in Figure J.30 although it did occur. HPI flow rates for primary pressures as low as 1100 psia (7.6 MPa) were used in temperature extrapolation prior to affected SG dryout. At 800 s, the SG dryout is complete and the system has repres surized to the HPI flow cut off pressure, 1285 psi (8.9 MPa), and the ope rator turns off the charging pumps. The coolant swell model predicted water solid conditions and full repressurization to the PORV set point, 2400 psia (16.6 MPa) by 1800 s. E. Heat Transfer Coefficient The downconer heat transfer coef ficient profile of Transient 6 was used for this sequence. J.7.3.5 Sequence 5.25A A. Basis - Transient 7 B. Departures from Basis The only differences between this sequence and Transient 7 are the failure of both MSIV's to close the requirements that the operator turns of f the 3.60 e
I ( charging pumps on repres suriza tion to the HPI flow cut off pressure and reduction of AFW flow to 160 gal / min per stesa generator. C. Temperature Extrapolation The sequence follows the temperature profile of Transient 7 out to 570 s where SGIS fail s to cause closure of both MSIV's. The operator limits AFT to 160 spa per steam generator, at which flow, neither steam generator goes dry throughout the se que nce . Blowdown continues until 7200 s where the final minimum temperature is 348 F (449K) . The temperature profile is presented in Figure J.31. D. Pressure Extrapolation The transient is sufficiently mild that depressurization below the HPI cut off pressure doec not occur. The requirement to shut off charging pumps is ignored here and the system is allowed to fully repressurize. The pressure profile for Transient 7 was applied to this sequence as shown in Figure J.31. E. Heat Transient Coefficient I The downconer heat transfer coefficient profile is assumed to be the same as that for Transient 6, which never drops below a value of 420 Btu /hr ft 2o F (2380 W/m2 op ), J.7.3.6 Sequence 5.25B A. Basis - Transient 7 B. Departures from Basis J.61
1 In this se quence 1 TBV sticks open and both MSIV's fail to close. The oper ator isolates the TBV 600 s af ter SGIS. The opera tor is assumed to turn of f the charging pumps and limit AFW af ter MSIV failure to 160 gal / min , 1 per SG. l C. Temperature Extrapolation The temperature profile for Transient 7 is used for this sequence out to 570 s where SGIS occurs. Le failure of both MSIV's to close requires extrapolation after this point. The total loop flow rates from Transient 7 were assaeed to be evenly divided be tween loops A and B for the extrapolation. No assymmetry ef fects were assansd. The minimum temperature at TBV isolation, approximately 1100 s, was 433*F (4M k) . Thereaf ter, the e system reheats to above 530 F (550 k) by 2300 s sa shown in Figure J.32. D. Pressure Extrapolation The pressure profile for Transient 7 was assigned to this sequence due to similarity of conditions. E. Heat Transfer Coefficient The downconer heat transfer coefficient for this se quence is assumed to be the same as that for Sequence 5.21B, which assmes Transient 7 response 2o out to 700 s and assignment of the corrected minimum of 400 Btu /hr ft p 2 l (2270 w/m ) thereaf ter. 3.62 l l
i 4 J.7.3.7 Sequence 5.26A A. Basis - Transient 7 l B. Departures from Basis In this sequence, 2 TBV's fail to close and both MSIV's fail to close. The operator is assumed to limit AFT flow to 160 gal / min per SG, throttle AFW upon reaching +22 inch indicator level in the SG's, and turn off the charging pumps on repressurizing the primary to the HPI flow limiting pressure. It is assumed that the TBV's are not isolated over the course of the sequence. C. Temperature Extrapolation The tempe rature profile for Transient 6 or 7 is used out to 50 s where 2 TBV's fail to close. Loop flow data for transient was adapted for extrapolation throughout the se que nc e . Extrapolttiaa ;-:dicts SGIS generation at 150 s with failure of the MSIV's to close. AFAS occurs by 240 s. As shown in Figure J.33, the cooldown continues unchecked throughout the sequence. Steam generator secondary water inventory reaches its minimum at 3500 s. Af ter this point AFT flow exceeds blowdown flow and the stesa generators start to refill. However, the +22 inch level is not attained, so AFT is not throttled. The minimum tempe rature of 290 F (416K) occurs at 7200 s. i J.63
. . l l
D. Pressure Extrapolation l The pressure profil e for Transient 6 is used out to 150 : where the f ailure of the MSIS's to close causes continued depressurization to 1285 , psia (8.9 MPa) at 500 s. Some further depressurization would occur but is not included here (see Figure J.33). The charging pumps are assumed to be turned of f and the HPI is assumed to stabilize pressure at this final value. E. Heat Transfer Coefficient
- The downconer heat transfer coefficient profile of Transient 6 is used for this sequence.
J.7.3.8 Sequence 5.26B A. Basis - Transient 7, Sequence 5.22 B. Departures from Basis In this se que nc e , 2 'IEV's fail to close and both MSIV's fail to close. The operator is assumed to isolate the TBV's cyproximately 900 s after the MSIV failures. The operator is also assumed to turn off the charging pumps and limit AFW flow to 160 gal / min per steam generator. C. Temperature Extrapolation For this seque nc e the tempe rature profile for Sequence 5.22 was followed out to 150 s where SGIS occurs. The failure of both MSIV's allows the blowdown to continue until the stuck open TBV's are isolated (1050 s). The minimum tempe rature of 386*F (470 k) is obtained at 1100 s. Thereafter, J.64
the system reheats above 530 F (550 k) by 3500 s. The temperature profile l l 1s presented in Figure J.34. D. Pressure Extrapolation The pressure profile for this sequence is shown in Figure J.34. The data for Transient 7 is used out to 150 s. Thereafter, the pressure is assumed to drop to 1285 pain (8.9 MPa) by 500 s and there remain until TBV isolation at 1050 s. Pressure would actually drop below the stated value s for portions for the period prior to 1100 s, but such deviations could not j be explicitly de te rmined. As presented, the pressure profile represents an expected upper bound. The system will repressurize to the HPI flow cut off pressure, 1285 psia, and the operator will shut off the charging pumps. The reheating af ter 'IBV isolation will cause water solid conditions within the primary by 2700 s as predicted by the coolant swell model. The pressure will achieve its final value, 2400 psia (16.6 MPa) at this point. E. Heat Transfer Coefficient The downconer heat transfer coef ficient for Transient 6 was used for this sequence. J.7.3.9 Sequence 5.27A l l A. Basis - Transient 7 i B. Departures from Basis l l In this sequence 3 TBV's f ail to close and both MSIV's f ail to close. The ope rator is assumed to litit .JW flow to 160 gal / min per SG, to throttle i 1.65
AFW upon reaching +22 inch indicator level in the SG's, and to turn off cht.rging pumps upon repressurization to the HPI flow limiting pressure. It is assumed that the TBV's are not isolated throughout the sequence. C. Temperature Extrapolation The temperature profile for Transient 7 is used to 50 s where 3 TBV's fail to close. Extrapolation predicts SGIS at 120 s at which point both MSIV's fail to cl o se . AFAS occurs at 140 s. The steam generators do not dry out, but do experience a minimum in secondary water inventor be tween 3000 s and 3500 s. AFW flow exceeds blowdown flow for the remainder of the sequence but the level does not rise enough to require AFW throttling. As shown in Figure J.35, the system obtains a minimum downconer tempe rature of 259 F (399K) . D. Pressure Extrapolation f The pressure prof ile for this sequence is shown in Figure J.3 5. The profile fre_ Transient 6 is used out to 150 s followed by an assumed drop to the final pressure of 1285 psia (8.9 MPa) by 500 s. Early in the se que nc e , lower pressures than the predicted values are expected. E. Heat Transfer Coefficient The downconer heat transfer coefficient profile for Transient 6 is used for this sequence. I i J.66
1 J.7.3.10 Sequence 5.27B A. Basis - Transient 7, Sequence 5.20 B. Departures from Basis i In this sequence 3 TBV's are assumed to stick open following reactor trip and both MSIV's fail to close. Le ope ra tor is assumed to isolate the TBV's 1200 s after SGIS, to throttle AFW flow to 160 gal / min per steam generator, and to turn of f charging pump flow as the system repressurizes to the HPI flow cut off pressure. C. Temperature Extrapolation For this se quence the tempersture profile for Sequence 5.20 was used out to 120 s where SGIS occurs. The MSIV failures allow blowdown to continue until 1300 s when the operator is assumed to isolate the TBV's. A minimum t empe rature of 339 F (443 k) is obtained at isolation. The system then reheats above 530 F (550 k) by 4300 s. The temperature profiles for the seque nce is presented in Figure J.36. D. Pressure Extrapolation The pressure profile for the sequence is presented in Figure J.36. Transient 7 data were used out to 150 s where a pressure of 1190 psia (11.7 MPa) is obtained. hereaf ter the pressure is assumed to drop to 1285 psia (8.9 MPa) by 500 s and remain at this level until 1300 s, just after isolation of the TBV's. Actual pressures will be lower than I the assumed vaines over most of this period. Following TBV isolation the system reheats and eventually goes water solid. The coolant swell model J.67 l
predicts repressurization to the PORY se t point, 2400 psia (16.6 MPa), by 2700 s. E. Heat Transfer Coefficient The downconer heat transfer coef ficient profile of Transient 6 was selected f or this sequence. J.7.3.10 Sequence 5.35 A. Basis - Transients, 6, 7 - initial conditions, Transient 12-ADV flow behavior B. Departures from Basis In this-sequence one ADV fails to close and both MSIV's f ail to close. The operator is assmed to turn of f the charging pasps upon repressurization to the HPI flow cut of f pressure and to reduce AFW flow to 160 gal / min per steam generator. The loop flow data from Transient 6 wss adapted to this se que nce . Transient 12 ADV flow data were used to calibrate the choked flow function in the cooldown model. C. Temperature Extrapolation The seque nce follows the first 50 s of Transient 6. At this point the TBV's and 1 ADV close. The remaining stuck open ADV does not cause any significant cooldown or primary depressurization due to its small flow. By l 950 s. AFAS is obtained and AFW flow commences. Toge ther, the open ADV , l and the AFW flow provide suf ficient cooling to cause SIAS at 1400 s and RCP trip by 1430 s. SGIS occurs at about 1550 s but both MSIV's fail to I 1 3.68
close. The operator reduces AFT flows and turns off the charging pumps as required. The SG blowdown and AFT flow reduce downcomer temperature to 419 F (488K) at 7200 s as shown in Figure J.37. D. Pressure Extrapolation The pressure profile follows Transient 6 out to 50 s where all of the TBV's and 1 ADV close. The pressure drops from an initial value of 2283 psia (15.7 MPa) to 1970 psia (13.6 MPa) at 50 s. The pressure stays at this level until cooldown commences upon initiation of AFW at 950 s. SIAS at 1755 psia (12.1 MPa) occurs by 1400 s and the system is assumed to depressurize to a final pressure cf 1285 psia (8.9 MPa) by 1600 s. This behavior is shown in Figure J.37. The mildness of the transient suggests the depressurization to this level might not occur. E. Heat Transfer Coefficient The downcomer heat tran sf e r coefficient remains constant at about 5000 2 Btu /hr f t F (28,360 W/m K) (Transient 6 data) out to he trip of the RCP pumps af ter 1400 s. By 1500 s the heat transfer coefficient drops to its assumed minimum value of 400 Btu /hr f t *F (2270 W/m2 g), J.7.3.12 Sequence 5.36 A. Basis - Transients 6, 7 - initial conditions, Transient 12-ADV flow behavior l 1 J.69 l . . -- - . --
. e l
)
l I Departures from Basis B. In this se quence both ADV's fail open, thus prolonging steam generator blowdown throughout the sequence period. The operator is assumed to turn off the charging pumps upon repressutization to the HPI flow cut off pressure at:d to throttle AFW at +22 inches indicated steam generator level. Automatic isolation of AFW is not generated since SG blowdown is symmetric. C. Temperature Extrapolation The tempe rature profile follows Transient 6 out to 50 s where all TBV's close. AFAS initiates AFW at 700 s. SGIS at 1100 s cuts off the MFW systems. As shown in Figure J.38, steam gener ator blowdown leads to a minimum temperature of 347 F (448E) at 7200 s. D. Pressure Extrapolation This se quence is initially milder than Transient 6 or 7. It is expected that the charging pumps flow can maintain primary pressure above the HPI flow cut of f pressure. Although such selection is conse rva tive, the pressure profils of Transient 7 was applied to this sequence (see Figure J.38). E. Heat Transfer Coefficient l For this s e que nce , the downconer heat transfer coefficient profile of Transient 7 was used out to 700 s where the assumed minimum value of 400 f Btu /hr ft 2o F is reached. 3.70
J J.7.3.13 Remaining Sequences Reactor trip se quence s not explicitly evaluated in this section were assigned to the TRAC calculated transients or extrapolated sequences most closely aligned to the particular non-extrapolated sequences. The sorting procedure emphasized similarities in temperature profiles, sometimes at the expense of significant deviation in pressure profiles. Table J.9 summarizes the assignments of the reactor trip and supporting rationale. Seque nce s 5 .1, 5.6, 5.9, 5.10, 5.13, 5.14, 5.17, 5.18, 5.21A, and 5.28 s either do not suffer any overcooling at all or are mild events mitigated i by the MSIV and MFIV systems. These are all grouped with Transient 6, a very mild transient, even though the temperature and pressure responses will be different. Sequences 5.2 and 5.3 are most like the mild Transient
- 9. Sequences 5.4 and 5.5 are likewise similar to Transient 8. All of these transients are so mild that the fracture mechanics calculation would I
likely not distinguish between them. i j Seque nce s 5.7, 5.11, 5.15, 5.19 and 5.29 all feature failure of 2 TBV's with successful operation of the MSIV's and MFIV's. These sequence s i are represented by Sequence 5.19 (Section J.7.3.1) , . the most conservative of the group, which is still very mild (T,g,=500*F) by FM standards. Sequences 5.8, 5.12, 5.16, 5.20, and 5.30 featuring 3 stuck open TBV's are all likewise mitigated by SGIs and are represented by Sequence 5.0.0. i Sequence 5.21A features one stuck open TBV and f ailure of 1 MSIV to clo'se makes this sequence very similar to LANL Transient 7. Sequences 5.23 and i J.71 i
. . - - . -- - , , . , . - - . , , , . . - , n ..
~~e - . - , . . . --. - , , - - . . - - . - , , . ,,, , . - - - , - - .-
)
Table J.9 Reactor Trip Sequence Assiganents P,T,h Sequence Profile Source Caaments l 5.1 T-6 Normal trip sequence
, 5.2 T-9 Lower P 5.3 T-9 Exact match 5.4 T-8 Lower P 5.5 T-8 Exact match 5.6 T-6 Lower P 5.7 Seq. 5.19 SGIs ends cooldown, lower P 5.8 Seq. 5.20 SGIs ends cooldown, lower P 5.9 T-6 SGIs ends cooldown, lower P 5.10 T-6 SGIs ends cooldown 5.11 Seq. 5.19 SGIs ends cooldown 5.12 Seq. 5.20 SGIs ends cooldown 5.13 T-6 SGIs ends cooldown 5.14 T-6 SGIs ends cooldown, lower P 5.15 Seq. 5.19 SGIs ends cooldown, lower P
- 5.16 Seq. 5.20 SGIs ends cooldown, lower P 5.17 T-6 SGIs ends cocidown, lower P 5.18 T-6 Exact match 5.19 Sec. J.7.3.1 Extrapolated sequence 5.20 Jac. J.7.3.2 Extrapolated sequence l 5.21A , T-7 lower p 5.21B Sec. J.7.3.3 Extrapolated segnance 5.22 Sec. J.7.3.4 Extrapolated sequence 5.23 T-2 Higher T, lower P 5.24 T-2 Higher T, lower P 5.25A Sec. J.7.3.5 Extrapolated Sequence 5.25B Sec. J.7.3.6 . Extrapolated Sequence 5.26A Sec. J.7.3.7 Extrapolated Sequence -
l 5.26B Sec. J.7.3.8 Extrapolated Sequence i l
Table J.9 Reactor Trip Sequence Assignments (Continued) I i P,T,h Sequence Profile Source Comments j 5.27A Sec. J.7.3.9 Extrapolated Sequence 5.27B Sec. J 7.3.10 Extrapolated Sequence 5.28 T-6 SGIS ends Cooldown, lower P 5.29 Seq. 5.19 SGIS ends Cooldown, lower P 5.30 Seq. 5.20 SGIS ends Cooldown, lower P 5.31 Seq. 5.35 Higher T 5.32 Seq. 5.35 Higher T 5.33 Seq. 5.35 Higher T 5.34 Seq. 5.35 Higher T 5.35 Sec. J.7.3.11 Extrapolated sequence -- 5.36 Sec. J.7.3.12 Extrapolated sequence 5.37 T-2 Expected severity be tween T-2 and T-6 5.38 Seq. 3.6 ' Extected severity be tween Seq. 3.6 and T-6 i l I l l 1
5.24 which feature 3 and 4 TBV failures with failure of a single MSIV resemble Transient 2. Sequences 5.31, 5.32, 5.33, and 5.34 feature a stuck open ADV but all are milder than Sequence 5.35, to which they are assigned. Sequences 5.37 and 5.38 feature a failure to trip the turbine following a reactor .tr ip . In addition, 1 MSIV fails to close in 5.37 and both MSIV's fail in 5.3 8. If turbine overspeed or underspeed protection systems promptly de te ct and correct the situation, these events will not have significant PIS consequence. As a conservative bounding case, such events could ba considered equivalent to main steam line break cases Transient 2 and Sequence 3.6, respectively. J.8 Small Break LOCA J . 8 .1 Description of Sequences The sequences for small break LOCA at full power are initiated by a PORV-sized break, 0.016 ft (0.0015 m ) opening at the top of the pressurizer with the system ope rating at full power. The break may be either non-isc1 stable or isolatable as required by the sequence specifications. Prior to the appe arance of the break the system was at steady state at full power. The decay heat function following the trip was assumed to 'be 1.0 times the ANS standard. The 17 specified sequences for this initiator are listed in Table J.10. The differences is sequence specification involve ADV operation. TBV operation, J.72
0
- and operator action to isolate the break, turn off the charging ptusps where applicable, and throttle AFT flows.
J.8.2 Basis for Extrapolation The basic course of the sequence s will be directed by LANL Transient 12 which features 1 PORY and 1 ADV stuck open. Other LANL transients which address phenomena relevant to the required sequences are: ! Transient 6 - 1 stuck open TBV, Transient 7 - 1 stuck open TBV and 1 stuck open MSIV, P Transient 11 - medium break LOCA with intact SG's, Transient 8 - main feedwater overfeed (both SG), and Transient 9 - main feedwater overfeed to one SG. Transient 12 correspondr almost exactly with Sequence 6.7. All other sequences require explicit estimation or selective assignment of profiles. J.8.3 Results and Discussion J.8.3.1 Sequences 6.1, 6.5, 6.17 A. Basis - Transient 12 B. Departures from basis In sequence 6.1 a PORY f ail s open and all TBV's and ADY's operate properly. i The operator is assissed to isolate the break at 1.5 h (5400 s) to turn off the charging panps apa repressariaation to the HPI flow limiting pressure, and to throttle AFT on SG 1evel. The loop flow da ta from Transient 12 J.74
,, , u . , . _ , ,- - . - - , , , - - . , , - - . - ww-,,,-e, , , - - - w
were applied to this sequence (total Transient 12 loop flow was divided i equally among both loops) although the stuck open ADV in Transient 12 would tend to augment loop flow relative to these sequences. Sequence 6.5 fea tures and overf eed to one steam generator. Sequence 6.17 4 features a stuck open TBV. Both of these situations are corrected by SGIS such that the initial perturbation is lost by 2000 s such that these sequences are very similar to Sequence 6.1 late in the event. C. Temperature Extrapolation The temperature profile for Transient 12 was used out to 227 s where the ADV's are assumed to clo se . Afterwards cooldown continues on the basis e of HPI, charging pump and MFW flows absorbing heat from the system. SGIS occurs at 2200 s and eliminates the WW flow. AFW is never ini .' i n t e d . Decay heat has declined to the extent that HPI and charging pump flows alone can continue cooldown at a rate which accelerates with time (i.e. as decay heat decreases) out to isolation of the break (PORV) at 5400 s (1.5 h). The system is water solid, so isolation of the break causes immediate repressurization which eliminates HPI flow and causes the operator to turn off the charging pumps. The minimum temperature of 412'F (484 K) occurs at this point. The downconer temperature jumps to the SG tempe rature due to the loss of the localized cooling effects from HPI flow. General 1 reheating of the system results in a final temperature of $23 *F (546 K) . The temperature profile is presented in Figure J.39. , 4 l l l l l J.75
. 1 D. Pressure Extrspolation The pressure profile for Trsasient 12 is used out to 5400 : (1.5 h) as shown in Figure J.39. After isolation, the system is assumed to quickly j repres suriz e to the HPI flow limiting pressure, 1285 psia (8.9 MPa) where i
the operator turns off the charging pumps. The system is essentially water solid and will experience repressurization to the primiary safety valve setpoint pressure due to the swelling of coolant from system reheating (see Section J.8.3.2, Sequence 6.3) . Here it is assumed that the operator can manipulate coolant inventory to prevent such repressurization.
- E. Heat Transfer Coefficient i
The downconer heat transfer coef ficient profile of Transient 12 is used out to 450 s where the coefficient reaches the assumed minimum level of 400 Btu /hr ft F (2270 W/m 2g), J.8.3.2 Sequence 6.2 This se que nce is similar to Sequence 6 .1 in every particular except for the failure of the ope rator to turn off the charging pumps following isolation of the break. As shown in Figure J.40, the effect of this f ailure is to allow prompt repressurization to the primary safety valve set point pressurs, 2500 psia (17.2 MPa) . The tempe rature and heat transfer coef ficient profiles are essentially unchanged. s J.7 6
l l l J.8.3.3 Sequence s 6.3, 6.4, and 6.6 A. Basis - Transient 12 B. Departures from Basis These seque nce s f e ature a non-isolatable PORV sized break. Seque nce 6.3 features normal runback of NFW while sequences 6.4 and 6.6 feature overfeeds of one and both steam generators respectively. The overfeeds will have little impact upon the systen response late in the sequence s and so may i be grouped with Sequence 6.3. Transient 12 loop flows and wall heat flows are used in extrapolation of the sequences. i C. Tempe ra ture J.41 gives the tempe rature profile for the se que nce . Transient 12 data is used out to 227 : af ter which the failure of 1 ADV to close causes Transient 12 to become colder than this sequence. The combined effects of NFW flow to the system generators and increasing HPI flow provides enough cooling to cause continued decline in the downconer i temperature. By 2200 s the steam generators have cooled to below 500 F (533 K) where SGIS terminates NFW flow. Decay heating has decliaed to the j extent that cooling from HPI flow alone can continue the cooling, but at i a slower pace. There are no demands on the AFW systems in this sequence. The final downconer temperature is 375'F (464 K) . D. Pressure Extrapolation The pressure profile for Tranrient 12 was applied to this sequence in its entirety O to 7200 s. This profile is presented in Figure J 41. t i l J.77
o . j t E. Heat Transfer Coefficient The downconer heat transfer coef ficient profile of Transient 12 is used out to 450 s where the coefficient reaches the assumed minimum of 400 Bta/hr ft2oF (2270 W/m 2g), J.8.3.4 Sequence s 6.7, 6.9, and 6.13 A. Basis - Transient 12 B. Departures from Basis $ These se quence s feature a non-isolatable PORV-sized break and 1 stuck open ADV. The operator is assumed to throttle AFT to the intact steam generator, en action not assmed for Transient 12. Sequence 6.9 features MFY runbsck failure to one stcom gene rator and throttling of AFW and SG 1evel. Sequence 6.13 include s failure of a single TBV to close and throttling of AFT to level. C. Temperature Extrapolation These sequence s were treated as being equivalent even though Sequences
- 5.9 and 5.13 experience relatively strong cooldowns early on due to a MFY overfeed and a stuck open TBV respectively. SGIS will termina te cooldown from these sechanisms such that temperatures will still ' eb above 500'F (533 K). Some mild reheating of the system will occur until the Transient 12 cooldown mechanisms (i.e.,1 open ADV and HPI flow) can resume the system cooldown. *1his type of behavior is also demonstrated in Sequence 6.12 as discussed in Section J.8.3.7. Since the early cooldown mechanisms tend to resemble Transient 5 in the later stages they are grouped toge ther with J.78 l
1
1 Sequence 6 .7 . Sequence 6.7 itself is deemed equivalent to Transient 12. The only difference in specifications of Sequence 6.7 and Transient 12 is the throttling of AFW to the intact steam generator. Since, the intact loop is essentially stagnant for most of the se que nc e , there would be no discernable effect from the throttling of AFW. Therefore, the Transient 12 temperature profile was applied throughtout the event. The minimum temperature of 300 F (422 K) was obtained at 7200 s. (See Figure J.42) . C. Pressure Extrapolation Due to similarities of case specification, the pressure profile of Transient 12 was used throughout the event. The pressure at 7200 s, the time of the minimum temperature, was 944 psia (6.5 MPa) . (See Figure J.42). D. Heat Transfer Coefficient The downconer heat transfer coef ficient profile of Transient 12 was applied to this sequence out to 450 s where the coefficient reaches the assamed 2 minimum level of 400 Btu /hr f t 'F (2270 W/m K). J . 8.3 .5 Sequence 6.8 i Sequence 6.8 varies from 6.7 in that the primary break is isolated at 5400 s (1.5 h) with the charging pumps being turned off as the system repressurizes to the HPI flow limiting pressure. The temperature response 1 is taken to be the same as that of Transient 12. Scae localized increase
. in downconer temperature would be expected upon loss of HPI and charging pump flow. However, the open ADV is driving the cooldown at this point and 7.79 1
j will prevent the type of large temperature increases reported for Sequences 6.1 and 6.2. The pressure response follows that of Transient 12 out to 5400 s. Since the primary system is essentially water solid at the time the break (PORV) 4 is isolated, the pressure rapidly rises to the HPI flow limiting pressure of 1285 psia (8.9 MPa) and the operator turns off the charging pumps. The tempe rature and pressure profiles are given in Figure J.43. The downconer heat transfer coefficient profile is assumed to the same as that of Sequence 6.7. J.8.3.6 Sequence 6.10 A. Basis - Transient 12
- B. Departures from Basis In Sequence 6.10 both ADV's fail to close af ter reactor trip. The operator is assumed to throttle AFW on SG 1evel. With both ADV's open there will be sufficient cooling so that neither steam generator will stagnate. The affected loop flow from Transient 12 was applied to both loops in the se que nce . This assumed symmetry will not yield differential pressures in the steam generators and the associated isolation of AFW to the low pressure steam generator.
C. Temperature Extrapolation The se que nce follows the trends for Transient 227 s where both ADV's are now assumed to fail open. SGIS occurs by 500 s, cutting off NFW flows. 1 i The closing of the MSIV's, of course, does not influence blowdown from the 1 l l J.80 :
4 open ADV's. AFAS occurs at 1150 : with the operator limiting AFW to 160 gal / min per steam generator. Blowdown and shrinkage of the SG secondary inventory coupled with this reduced AFW flow prevent attainment of the
+22 in indicator level in the SG's by 7200 s. Higher AFW flow rates would result in AFT throttling within 7200 s but would not cause significantly lower tempe ratur e s . As shown in Figure J.44, the temperature declines to 252 F (395 K) by 7200 s.
D. Pressure Extrapolation The pressure profile for Transient 12 was applied fully to this sequence as shown in Figure J.44. E. Heat Transfer Coefficient The downconer heat transfer coefficient profile for Transient 12 was used out to 450 s where it decreases to the assumed minimum value of 400 Btu /hr ft 2o F (2270 W/m K). J.8.3.7 Sequences 6.11, 6.12, 6.14, 6.15, 6.16 A. Basis - Transients 11,12 B. Departures from Basis These sequenc a s feature f ailures from 1 to 4 TBV's to close after reactor trip following initiation of the transient. MSIV and MFIV closure at SGIS will terminate this cooldown mechanism early, leaving HPI and charging pump flows as the only cooldown mechanism. Sequence 6.12 also features l J.81
a WW runback failure to one SG and as a conse rvative case, was selected i to represent the group. l C. Temperature Extrapolation The combination of NFW overfeed and a stuck open TBV cause SGIS by 250 s. l The closure of the NSIV's and MFIV's terminate cooldown from the SG's. 1 Decay heat overwhelms the cooling due to HPI flow and the system reheats until the ADV's open at 600 s to limit temperature. The ADV's will cycle until 2500 s, when decay heat declines to the extent that HPI flow alone can continue the cooldown of the system. This early behavior is different than for sequences 6.1, 6.2, and 6.3 w' nere continued MFW flow with HPI flow prevented reheating during the first 2500 s. AFAS is not induced in this sequence and hence AFW does not enter the picture. HPI induced cooling yields a final temperature of 410*F (483 I) at 7200 s as shown in Figure J.45. Stagnation of both SG loops was not assumed to occur based on the trends of Transient 11. D. Pressure Extrapolation The pressure profile for Transient 12 was used for this group of sequences. The profile is shown in Figure J.45. E. Heat Transfer Coef ficient The downconer heat transfer coef ficient profile for Transient 12 was used out to 450 s where it drops to the as sumed minimum value of 400 Btu /hr f t 'F (2270 W/m2 g), J.82
t J.9 Nedium Break LOCA J.9.1 Description of Sequences 4 The sequences for medium break LOCA are initiated by a 2 inch diameter or 0.0218 ft (0.002 m2 ) break in the hot leg. The intent for choosing a break of this size is to obtain loop stagnation without total depressurization of the primary. The system is assumed to be at steady state at full power prior to the break. The decay heat function following reactor trip is assumed to be 1.0 times the ANS standard. The seven specified se que nc e s for this initiator are listed in Table J .11. The differences in sequence specification involve ADV operation TBV operation, MFW runback and operator action throttle AFW when applicable. J.9.2 Basis for Extrapola tion The basic course of the sequences will be directed by LANL Transient 11. Other LANL transients reflecting expected sequence phenomena include
- Transient 6 - 1 stuck open TBV, Transient 12 - small break LOCA with one stuck open ADV, Transient 8 - main feedwater overfeed (both SG's), and Transient 9 - main feedwater overf eed to one SG.
Transient 11 corresponds exactly to Sequence 7 .1. All other se que nce s [ require explicit estimation or assignment of profiles. l J.83 s
=, - -
-.m - -r-- ,,, ,,- ,w~ , , - - , , - - - - , , , - - ,
---r .- , - - , ,,,. - - --,- -
l 1 1 J.9.3.1 Transients 7.1, 7.2, 7.3, 7.7, and 7.8 A. Basis - Transient 11 B. Departures from Basis Sequence 7.1 corresponds exactly with Transient 11. Sequences 7.2 and 7.3 include overfeeds to one and both SG's respectively. Sequences 7.7 and 7.8 feature one and two stuck open TBV's respectively. I C. Temperature Extrapolations The early cooldown mechanisms (TBV, overfeed) of Sequences 7.2, 7.3, 7.7, and 7.8 are termina ted by SGIS . The tempe rature profiles for these case will recover scaewhat and will closely resemble Transient 11 in the ' later stages. The potential for these deviations to bring on stagnated conditions could not be assessed. The temperature profile of Transient 11 was assigned to this group of sequences. The profile is shown in Figure J .46 . The minimum temperature . D. Pressure Extrapolation i The pressure profile of Transient 11 is applicable to this group. The profils is given in Figure J.46. i E. Beat Transfer Coefficient The downconer heat transfer coefficient profile for this group follows the basic trends in Transient 11. The coefficient holds its initial value of 5040 Btu /hr f t 'F (28530 W/m 2 K) out to 64 s where the RCP's are tripped. l By 250 s the value has dropped to about 650 Btu /hr ft F2o(3690 W/m 2 g) l J.84 i
, - - , -- - - - - - ~ ,, ---r ,n, -
l t and the coefficient drops to its assumed minimum value of 400 Btu /hr ft "F (2270 w/m2 K) by 600 s." J.9.3.2 Sequences'i".yand7.5 , A. Basis - Transients 11, 12 :- -
--,,,s,
- y B. Departures from Baris
~s These se que nce s featuis .a non-isolatable medium size primary break and a I .
stuck open ADV. In Sequence 7.4 the operator is assure,d to thr,ottle AFT
- ,.<,a to the intact SG 'on level, in Sequence 7.5'the AFT. is not' throttled. ~
~
1 . . . . >
- i ?
Temperature Extrapolalion \' C. -
/
~
- ., m; -
Due to the similarity of conditions, the tampe rature profile of Transient 5 was applied til those se que nce s . , Tho, 'y.cofile ,1s shoda,' in Figuxe N.47. , The open ADV will .ausure cooling induced Joop flow 'in one loop so~ that { a- 4' s ~,,, - - ,' the total stagnation predict.ed late in TralD4, tat 11 will not occur. <The . final temperature for 'the sequence isr 300 F (47,2 K) .' - -
!, f ,
- . ,f .
D. Pressure Extrapolation. ~
; i ,
,. r '
f/
? / f,;_
The pressure profile for the sequsaces was ass.uned 'to be that of Transient
~ '
- .p i 11. The profile is shown in Fi gure J.47. The-final pressure is 512 psia i s,"
l * (3.5 MPa).
^
j E. Heat Transfer Coefficient ,y ,. (
>?
s The downcomar , hea't transfer coef ficient proflie for these sequence s is . . assumed to be the~ same as that for Sequence 7 .1 (Section J.9.3.1) . The
# g W 5 t 9 i 3.85 9, !
/ <
)
assumed minimum value of 400 Btu /hr f t F (2270 W/m K) is obtained by 600 s in the sequence. J.9.3.3 Sequence 7.6 A. Basis - Transient 11, Sequence 6.10 B. Departures from Basis In Sequence 7.6, the non-isolatable primary break is accompanied by the f ailure of both ADV's to close af ter reactor trip. The operator is assumed to throttle the AFW on SG 1evel reaching +22 inches.
~
C. Temperature Extrapolation The failure of the ADV's to close will keep with SG's cool relative to the core exit and will promote na tural circulation in both loops. Flow stagnation of the type in Transient 11 is not expected. The conditions in Sequences 7.6 and 6.10 are very similar so the temperature profile for Sequence 6.10 was applied to Sequence 7.6 (see Figure J.48) . The minimum temperature of 252 F (395 K) is obtained at 7200 s. D. Pressure Extrapolation The pressure profile for Transient 11 was assigned to Sequenca 7.6 due to similarity of conditions. The profile is presented in Figure J.48. E. Heat Transfer Coefficient-l l l The downconer heat transfer coefficient profile for this seque nce is assamed to be the same as that for Sequence 7 .1 (Section J.9.3.1) . The J.86
2 assumed minimum vaine of 400 Bta/hr f t F (2270 W/m I) is obtained by 600 s in the se que nc e . I I l l l J.87 l
Figure C.1. P, T, h Estimation Approach Resource Data 12 calculations by Step 2 Step 3 LANL (TRAC) Check consistency of Determine applicable includes TRAC cases, extract : ; parameters by steam line breaks % duplicating TRAC relevant parameters TBV failure results in PORV LOCA cooldown model Medium break LOCA
- MFW overfeed a EFW over feed Loss of load ORNL
_Specified Sequences y Total over 100 cases y Steam line breaks: Large break at hot zero power Step 5 Small break at hot zero power Step 4 3 .; Document Large break at full power Evaluate Small break at full power Group specified sequences by similarity A. Temperatures by piecewise selection of l Turbine trips: TRAC curves and use of Overfeeds cooldown model TRV,ADV failures ; B, Pressures by piecewise I Small break LOCA's selection of TRAC curves f or by coolant swell I calculations C. Heat Transfer Coefficient by piecewise selection of TRAC curves and corrected limiting value
)
9 Mt .Tu k Steam Generator Node Primary Node ] l I
- a. a. l I .
ST Tw Tu Tsa l l si.am l o.n.,sio, ex I I
o.a_, , j_ l
,l:
to 1 i
- 1. rc u - n
" l l
e l I g j Core 4 On 8 l OncP l 3
. I e U HM I HM 1
0,
=. l Aco Toe I_ _ ____ __ J FIGURE C.2 MASS AND ENERGY FLOWS FOR 2 NODE COOLDOWN MODEL O
I
I 00.89.30 64 0 9 fWir. 1964 Jne-p1NUDI 1$500 Ol5WLR WCR 4. 2 TEMPERATURES AND PRESSURES FOR S' g_UENCE f 1.1 I
,,-------------< o TEMP i
-+
- - -P --
.hES m
l I ,
,o ,
8 ' l ,,'
^ *
- u. , ,-
wE e kn
- ..- # H.
g ; X ,. , ui
* ,' a:
Ei
, ,/
4 E I / lma. 2 '/ s H E a N 6 o sae 6 6 4 4 4 em TIME ( seconds )
n3s~ ynuo. .D l g H I P? M ! E T
?
l o t d o 2 - 1 E . C . N . . E - i.o U . Q . E - S - l . I . O . , o. F . S . ) E . s R - d n U - o c S . e S E
.=(s R -
E 2 P .
. M I
. T 5
D ' R E W N - A A - o
.oo L
r . s 1 5 S ., l O E , R - 0 C U - 5 5 T - 1 A . R . .ona 3 E ~ 2 01 E e P ~ l a- M - 51 E . A T . 9 6 ., .= 98 , i, tW
/ :
f
/
9
- s, D
E N /. j .: e 's ;, 5 o o 5 0 I g E 5 9 y R, R E H I 2 0 w wyg 0 I f - _ O _
9 9 ao. 22. 52 ura e eviv, aset JNTNun , Issco osssrte vra e. 2 TEMPERATURES AND PRESSURES FOR SEQUENCE 1.3 1 ,__..____......______._..___................... o TEMP
.+
_ __P.R.E.S
,7 '
a ,/ e i n ,.e n . .- lag a Rs f . g
,- s.
R :
/ ,, - ,
l ,
. ' gm
~
BR i / R
'. ,'/
R R I sooo woo sono I . suco 4000 naoo naoo anno o TIME ( seconds ) .
O O 4 9 mt onor a
~ o, 088E i
h T.! r c.. e o +ll r a B B 8 8 N l O ' Z l W l
-I D
- 7 w.
l W . I U l C A -. l
.I I ^
~e j
1 l 1 ..-
- l. L.k e l to
~
- c. . 2
- l 9 Ql Z !.
h ! g i l g =. D l
- E-l '
<! l
%l a 's g m, E y Al l
9 31 . ; g Wi l bl . 8 o ' 3 c .g
~
'+ .,,'-
.. /
2 g - .- 3
* * '
- 0" op one occ out one a m i 4 s
(1) auturasm.t l L
e ( m ed ) 2ESS3Hd ces coat oost osot C. A e
$e.
a. e 0 +8'
+ 0 0
0 e e A l e
. h.2 U l o
Z ' 3.1 l 'I D l o- , 8 L.3 v: l, O i
.g A l t
0
= i O !.
0 ew l c: y L I. n b G Z l
'a s
l 4
~ $ l x l '.'
9 P e l e .s i 81 4.2 h A l
- :E 4 Er.1 i
9 l 6l *'
.8 R
., .. 5 2 4.' .
E e '.. ___ _ . ................**+ M M M M M M M M M Olt Mt a (3) EMMM t 1 i e
. -e (visd) 2HaSSIHd ooor oos coor cost
.b
- c. g.
d zl e s.. O +l 1
\. .f
~ .
N i. W i. U:
'l
- b. i.
D C" \ W M \.. 8 O s
'. 2 N .
^
$ \
4 xi '.
'. 3 D'
. 8 s-
@' '. s a .
- A{
C, \ 5 Z' \
< \ g
~
3 D. , .
~
e<.' . l W l i
.I h A ,
- 2 t.
m, 3
, ej :
i 0 3
' .I I ........... ---~
i s est ces os+ oc+ out one one one est oot s oos
* (3) 2Hn1YE3dM31 8
O s
9 1., . ...., . . . n. , m,n. no. .nm... . , _TEMPERAfURES AND PRESSURES _['OR_ SEQUENCE 2A a
, , . . _ _ _ _ _ _ _ _ _ . . . . . . . . _ . . . . . . . . . . . . . . . . . . . . . . . . . , o TEMP
,/ + ..
...PR.t3 I /
I
/
/
l l
- 3 I j
~ >
, n wg v,
, s.
l ,,- $ *8. M : ' R : ,- N
/ { Rh, h*Ri l
.. a
- l g :. .,* ..
I R
. R I
e a. 6 6 6 6 6 4 em TIME ( seconds )
9 9
^s8-i khg n X I
PM M E F T f. O t. 4 o 5 . 2 . e E _ C . N . .oo E . J_ t . Q . E . S . R_ . O . .. F .
. )
S . s E .
. d R n o
U ._ c S S
....(se E .
R . E 2 P . M I S . T D . B t N . U A . A L F
.=
5 5 S . 1 0 E . R . 0 C 5 U 5 3 T . A _ R .
. .=
e n E . te P ._ f 5
- M _
S E . EJ T _ t a . . ts r , m , 9 9 E. '
.I...I;*:, .
I I. I I I 3 I n I X R R 3 3 a A- M $@ c 8 fi
rtar a co.35.es wo e nas uses me-s sons , assco ansvta von o. : _ TEMPERATURES AND PRESSURES FOR SEQUENCE 2.6 s o TEMP
+
_ _ _ _P_R_E_S o I I O I . t
\
e 8 Im e S 1 : $2 s. 1, -
,,,_______________________________________________________________________4 I ___ w i nE~
k a I $ I R N o 6 6 6 6 6 6 6 ew. TIME ( seconds )
- 1. . , . . , . . ... . . m,n. , ,s. o m. .<. . . .
_TEhj_PER_A_TURES AND PRESSURES FOR SEQUENCE 2.7 . I ------------------I O TEMP
~
- .t.ff.9
+
I l E : S i e a 1 'f '
- i. ...-
ts : ,- Isa. a : .' p? hI l Ik
+ l I /
R t. *,- a I I I adoo .doo Woo sono so'uo adoo sino 4400 o TIME ( seconds )
O O ~ h I a= = = = g A pi b. m C +l t O g i g e c' l I. w. o; s l Z . . Ml D , CY w .I Dh s e l Z' e O i t u Ca. 8 g l 4 x l o l l V) l M l 'I 8
~
el A l 2 M g 9 Z; l h s Bli c:: 9 Di l
- 8 H<3 w; $ .
~ w i I A l 7 2 l g w i l
E-'l' . e l S l
...........--- + ,
m* M 096 M M M M M M M M M 8 (l) MMcDU1 s I
O
-, . . . . . . . .. ,,.. ,.4,,, . , , - ..,so .. ...
_TEhlPERATURES AND PRESSURES FQR SEQUENCE 3.5 I O TEMP
+
_ . . .P.R.f3 j l a i I l
^ ! -
ld8.
- Il
., + ._......____________________.,
l (d I
+
ge I [ n. J I I I R I. uk. 6 6 6 TlWE ( serids ) 6 6 6 m l l l l
;I l ,J ll1 g
e o ~ m . *8. N> E I i P 3-M - E t. T R. P. o +. a 4 6 . 3 . = E . G - N . E - U - Q - E - S - R - O . . F .
. )
S . . E . _ a
. n R .
o U - . c S -
- _(
S - F - R . m P . u 2
. n S
D . R E W N . A A . _ L P 5 1 S . f
# E .
R . c B C 5 U - S T - B A - R - E . N P - a 1 8 M - 9 8 1 E - T - 0 . 0 9 . . 3 . f R 9 M 9 ..J g 4 %* il; I I i e t. I
^w- @ I.
0 0 I f O L F A
i -t:'
- 5. 8 ,-
k I l I . PS M I E R. T P o +.. o-4
. 4 7 .
3 E . C N E _ 6 U _ Q . E .- S . R O . 6 F S . ) E . d s R .
. n o
U . S S n
.ce i..(s E _
's R _
. E a.
P . M I
. _ T D .
e W N _ A _ a L r _ 6 i s S . I D E R c . o U c s . s I T . A _ R .
. 4 r
o E . p as P i s M ._ e o E _ J T t . e t s . 4 t s u . r 9 D
. ,1 e e C
- a *
. :*e 1.e 5
II'I. o 3 0 I I I I a I R I 5 0 0 u-I T S L F l i
9
,0:
*k i +',
WD 9 mi, 39 9 A19-s 151000 ISits 83%yut vt. .3 ,2 Let 3 0. 53.37 . k TEMPERATURES AND PRESSURES FOR SEQUENCE 3.8 a o Time _+_ _ __PRis.
,/
- Ii> e ao e
E / o a l e I ', I
. / . _
. ./ 5.1
~I 3, l
,/ ~
/
g . ____ -
. ./
e I A l ,' a. I I I R
==.
. va.
R. TIME ( seconds )
.. i y
'n 5
at i es. se.at era e nar. sees no-sswas , assen essseta era s.: _ TEMPERATURES AND PR SE SURES FOR SEQUENCE 42 I o TEMP
',,+-------< & PRF3 y,
,,. ')
I
, 0 g
B
.-- i ,,
s I ;
',,. ~'
,, l38.
m ; y I :+ ~... A. ka: a. I I ll E R .
.o n
,, m e io
.xn so.
TIME ( seconds )
o L l
.u.n n. . m. n.. - r=, . no oim w. . . . .:
TEMPERATURES AND PRESSURES FOR SEQUENCE 4.4 I o igue
/
.t. PP,q Ii /
/
a E >
/
/
a
-o l
,/ r 3 : /
,/ ^
s l e' mI * ' l .5
; s.
/ -
M :
. . ./ ta I .f. .__- a i i. a.
! I R X t 1 e i 4 4 4 4 4 4 TIME ( s.conds )
- l L
. o A5a- gl
$ i R o n
o s P3 M f E R. T P o +.. ' d so
. d 5 .
4 . E . C . . N .
. i om E . e U .
Q . E . S . R .
. o O .
Ise F . s S .
)
E . s R . d n U o c S o e e s S d e( E . R .
. E 2
P . M I e . T a D . t v N . aL A .
. d s
o
-t . a s S og E .
R s a U . s e T . A .
. o R . d e
S: E ..
. a T
e P . e- M . s A E . T . t e ., In o ls , e s T , R , u f 3 . a . r . t i
. D.*::,
e
,6 3
3 . 8 I , I i I R H 5 8 ~ M g 0 4 T S E
O
~
$a- wihn. B E I o I k m e
P M 9. E F T f. o t. a 4 oo d
. 7 6 .
4 . E . C . o o N . d o E . U . Q . E . S .
. e R .
. d s
O . e F .
. )
S . s E . d n R . o c U . 0 e s h S 0 4( S . e E . E R .
. M P . 7 I
T
- a. D . -
N A d s o a-m s S i
. E .
R . o c U . m T . o A . Ine R . s m E . e P . n.
- M .
m E . T .
. o o
. d i -
m .
= .
= .
- ,ib
- ;,; e i
u < R R I I H I l I a I e N %4 a.i e i
^I ss-i d I s
o o o P . M S. E E. T R. P. o +..
- a s
. o
. d i
7 _ 4 _ E _ C . s N .
. d o
E o U ._ Q _ E . S . R ._ _ o O _
. i n
s F _ S .
)
E . d s R .
. n U . o c
S . e S . d o u s
. e(
E . R . E P . M e
. p I T
e D ", w N
- a t
A ,
, i o
n r '. s s s S t e E . o R ' c s U - s a T < , A ,
- o R ,
d n a a E , a m r P , a- M ~ e m E ~
~
~
T_ ~ s e
~
~
n e s a -
'ao n
ur . a . o n ._
;:.:-::{
s r i e
- a. I n s 5 R I R I
a z. on
- ww >I i
t at
.i . . .... w . m. n.. --. i-n . isse. oism n, ..,
TEMPERATURES AND PRESSURES FOR SEQUENCE 4.8 I f------------------------------------------------------< o TEMP
/ t fM9 R" /
l 3 I
- I l
3 ,: ,/
^
- ,/ ^
Is8. I ; / c l
!2 : -
ga j ... / a SR W gl, m. a I R R , H I . . . . . . . o scoo anno anno 4000 sono . coo m sono TIME ( seconds )
He 4 9 ( med ) su.m zud = =
= = g i
- n. -
;El s.
e O +
. O
-I am. 8 8
-r .
0 m e
' O Z l c' .8 g . s D
o' x] l'. v2 . I I P
~
l t
- z. +, i o '., e
's ,
W 4 W O. 's 2 e 's E
$ Z '.
s 4 '. .I A E '.
== g a .
9 9 D
- e -
e< 's
- x ,
M I
- 2. ',
M 9 O I e E-*
'e.
t t
.- l.
a
- i 9
6 000 m 008 OM 000 00E 80C ON ON 00f tot 5 (4) 28nnu2at21 as T rw
~S -
1
y, k I I m. e S. u r E. T R_ P_ - o +.. 3 6 2 1 4 _ E _ C _ 6, N _ E _ U _ Q _ E _ S _ R _ 6 O _ F _
. )
S _ d s E _ _ n R _ o c U _ e S S 6(s E _ E R _ M
, P _
I T D _
. N _
6 A _
. S . E _
R o U s T A /,8,-ll.# / 6 R ! t m E I,y./
,- P -
M mE - T_ _ _ k u I
!l:.;.;:p e 8 I I g I a I R '
I
. ^ %m
e PLDI B on . 21. 92 NED 9 MAY, 8984 m8-ETWW9 , I Sse.O DISWtn v[R g. 2 L . e TEMPERATURES AND PRESSURES FOR SEQUENCE 5.19 t I i
. .__.- O TEMP
=
,.....__ __..____________ ....__.____.....__ . ..__...____7.._____
s g _+_ _ _P_R_E_S . p
,e i. g , ;. ,
*^ .
a .(
- 4 E ,' , a l -
-t ,. .( , . .. .
s '3
>- 1
.('
~
s
'- rc g F, ,I ) s N ,,
5 1 , y .
' 3 w ,
- x s
,I $
'> h3g
'i - 3
- s.
pr +, . < 4 g 4
*g
\* f. . ,
t [n,_ A i j R h s 1 e s y
. * > +4
, k -
s E ,
' R .e .
X s I' R R , . . . . 000o sooo woo amo anno 3000 4000 o sooo TIME ( seconds ) eu
o e
*I 08.34.35 ED 9 MI, 8964 Jup-aIN040 , IS$CO Dil5Ftn vta 3.2 TEMPERATURES AND PRESSURES FOR SEQUENCE 520 I
j------------------------------------------------------------------ o TEMP
^ PRES gs / I
/ o l
-l .
l y i L l
/ I
' .,j -
~
I isg - I
- i. 4 g I a I R, ,e A A A A A A -=
TIME ( seconds )
O e PLOT 3 08.33.34 ut0 9 NRY, 4989 JDS-E TN0ll , ISSCO DissrLR vtn e.3 TEMPERATURES AND PRESSURES FOR SEQUENCE 5.21D I
,................__ ___.....___..................__.-d o TEMP
+
,',' -------PRES g, , ,
,- o I .
-l
~.
+
's+ ,o' .
{ 's , l A u..+.' n . v$ '$3 M v I lA ig p n-I I R g-I. ie, e e A TIME ( seconds ) e e A -
e root i on.ss. s uro s rm, sus Joe-simaa , issco osssrta vtn e.2 TEMPERATURES AND PRESSURES FOR SEQUENCE 5.22 I o TEMP
,..___.____............__......___.....__...........__...... H
.t _ EE.Q Ed /
o
/
/
k I / l / 3-t / v.' -
^
m , / - w el i l
/ l33.
's / -
M ', e s -- ~4 R l X X R edeo tdoo sono o 340a 4- adoo adoo odoo TIME ( seconds )
Q
\
'8 08.46.0o W[o 5 ftRV,1989 J0041 Holt , 31Sco O!5WLA Fa s.2 TEMPERATURES AND PRESSURES FOR SEQUENCE 5.25A I o TEup
_t _ PF,Eg l 1 *
/
k
$ I. .
'+, ,e
+
's, '
3 +, +,,-
-4... n
$3 a
~ .
M = -o a ai i l . R-I X H-7doo omo o so'ao an adoo adeo adoo edoo TIME ( seconds )
g e 8 08.37.00 W.D 9 rint, test mie-EIN083 , It s DISm na e. 2 TEMPERATURES AND PRESSURES FOR SEQUENCE 5.25B I
,----------------------------------------------------s o TEMP
, + PRES am j o i '
a : ' ' I
.'., f I +,
~-+--
,, " R
^ -
~
*g . l5t
, 4 4
R I l R R R R
- g g,, A A A * " '""
TIME ( seconds )
a e 108 8 08. 38. te NEo 9 MF, 1989 JS-RINoto , ISSco DISSPLA vtR 5. 3 __TEMPEIJATURES AND PRESSURES FOR SEQUE CN E 5.26A a o TEMP
+
. . . .P.R.E.S R
l l 5 t v.' e . 2e
'. 11 n
W '. ........_......__..........__........---.._____......4 8 Ei
=
R 0 iM.n 4 I I a nooo odoo 7doo sooo ads adoo :400 4d00 o TIME ( seconds )
e e PLOT 4 08.10.38 NED 9 nPT, 39st J06-RTN015 , l$5C0 DISWLR VCR S.3 TEMPERATURES AND PRESSURES FOR SEQUENCE 526B E
,------------------------------------------------ e O 7EMP
,a
+ . --
. . . P.R ES
. ,. a
,' ~
e' g s .-
- v. .'
^
i e'
'. R .s Il ,
R.
- v. .. . . . . . . , . -
N e, g lx n. hX X I 4 . E odoo 7doo sooo o sh adoo adoo 4400 edoo TIME ( seconds )
o
=
Of 3 08.92.31 WED 9 tmy, 49st .me-51Not t , 155CO DIlsrLA vtn 8. 2 TEMPERATURES AND PRESSURES FOR SEQUENCE 5.22A a o TEMP
.t _ W.5 Rd R
l s i
^
Y'. - II '. $$ E, 8! '..... ..._____________.__.._____________..__..______________..___._....+ w I SR p I C k a R I R 3 odoo edeo idoo sooo o doo doo doo edeo TIME ( seconds )
\
e e 1 06.49.07 u[o 9 my, 3909 Joe-stesons , 15 % O Ol$9LA VCR e.2 TEMPERATURES AND PRESSURES FOR SEQUENCE 5.27B 3
,--------------------------------------------------< o TEMP l -t ?BM g<
l a
,f R / l -
l
- l l q t l v.' ^
n . l s '. l $5 v5 . l 8.
'. l -
k! '. l
.--- ---+
s*. J. n R R . E sde adoo adeo 4doo adeo edoo 6 anoo o T!WE ( seconds )
- o. ......o. . . ... ,,,, . ...... . i m o om,u. .. .
TEMPERATURES AND PRESSURES FOR SEQUENCE 5.35 I o TEMP
+
_ _ _ P_ _R_E_S
-l
~,
I 4 a
^ l m Il I, is
- 8.
n . I R I R a adoo odoo odoo idoo sooo o adoo adeo adeo TIME ( seconds )
, , o,....., <o....... _.,,.,.,,m, oi,m. . . . .
_ TEMPERATURES AND PRESSURES FOR SEQUENCE 5.36 I _____________________________________________________ o nup
! _t_ fFM R- l I /
i e I : ,-
-]
,+ . ,'
's 3 e
+
'. 4___..+'*"
m ^ s .5 w$ 8. N! 1" l a. 4 H I R-R adoo odoo adoo 7d00 emo o no'uo adeo adeo TIME ( seconds ) i I
O O ftCT 4 08.58.45 GED 9 ruti, 1999 J08-E TN0 t$ , IS$co SIssrtR vtR s. 3 __ TEMPERATURES AND PRESSURES FOR S QE UENCE 6.1 I O TEMP
.t. ?.BM .
g; g R I i. I n e - s, '. l .M I 8.
'. ,..... ............+
,g i, ,
D , g
\
',,,,,+',......,...................................;.'
, 5 I
]
I' i R I o =
== =
A & .g., s . TIME ( seconds )
D
,, . . . . . . . . .. ... -- . . n= an= "a >
TEMPERATURES AND PRESSURES FOR SEQUENCE 6.2 R f o TEMP
.t. f.8M Er k
R : l I R a
^ : -
~g ; .
5.s .
. R.
. l -
M : GR I. I N lR
\
\ .
...,-~ ...................-.-- - -------*f -l' R
R R R R g,. 4 6 = " *
- 4 TIME ( seconds )
'\
\
)
LOI I 08.53.26 NES 9 rwty, tset m .gTN021 , f5 W OI55Pt.il VCA S.2 _ TEMPERATURES AND PRESSURES FOR SEQUENCE 6.3 R o TEMP
.t. f.@
Eu I l i 3 . A la8. 6
- I' l* -
R '. R
\
....+,.......,..s ~~........................................+............+
g m R R H t v s , soo. - g,. .s,. ,,,, o no= = TIME ( seconds )
AP ' s 3 08.56.17 BCD 9 my, 39st J08-5 Tra196 , ISSCO Olssrta VER g. 3
. TEMPERATURES AND PRESSURES FOR SE UENCE 6.0 s
o TEMP R, tfM
-l I .
a n e
~' ! ,-
t.
; is 1 8.
- g. I, ..__.__________..._4
- e k, e' p4e, R ,....+,,---
-s ;
....._________...........____4 I, R
R R R H . A o >=o = = .h .d., .,& . TIME ( seconds ) A
. D i - . . . . . . . ... ... -,-o. is - . ism < .. . _TEMPERATl]R.ES AND_ PRESSURES FOR SEQUENCE 6.10 s o TEMP
. t E-@
Ri R ; I I l I .
^ ^
l .
-ig i
1 la3, H 's i i
\ , - ,,
I 's ~* * '
,,,,,,,,,,,,,,,,,,,,,,__. ------ ----+------------+ n.
a I R I, s. A A A A A * ==* TIME - ( seconds )
F i oz. .o . . ... ... m i ., , ,, ,,,m,,., __ TEMPERATURES AND__ PRESSURES FOR SEQUENCE 6.12 a o TEMP
.t.fFM g,
g N M I . A B n
\
s. gw
-C=< 5 8
5. I - l- ;e. O
\
f*g ',...v'......'~~...
' ............_________...._....... ...... p__. ....... 4 4
R I H
. H o =
=
m m 6 6 s _ TIMF ( seconds ) t
(_ n O e e et o.= = o. c i g.
.-. o.
e O +l
+
e
=
8 I M + w . o l z . N
- l -l D
o'l a
- ./
m . l 2 l o ,
's Ni l s
^
~ + ,=
o l 8 m .' 8 d l -f"--
~ -
ama=t
.e I
- 3
- Q; [ E
= zi !
<l ! .g s .
m h ! 5 I 2l . l 8 DI ' E-- '
~.
Ei + -l t 1; 2 l { x . e :
~
E O f t l *+ l l t , 4 one oss cot est oot est coc ow one est oct y (1) zun:.vmanu
3 D 02.02.90 MED 9 riRI, ISet JD8-83Mn36 , ISSCO DISVLR TER S. 2 TEMPERATURES AND PRESSURES FOR SEQUENCE 7.4 s 0 TEMP
-.t. f.@
3, 3 : \ I l \ b 3-:
- n n ,
$ds,
-{ ,
I
.R-2 9
1
.I g
'-+ --..........+ ,,,__ ~...'~~..
$ ..+ ,,, ~~ . .
.. -+......+ g R
I. 4,0 4 4 4 TIME ( seconds ) 6 6 6 aan
b-Joe-Eino t t ISM:0 DISSPLA VER S. 2 B 02.01.37 niCD S net, 390s . _ TEMPERATURES AND PRESSURES FOR SEQUENCE 7.6 I o TEMP
.t fR_Q I
I R!, 3l l _ a.
- t. Ia8.
_l - - M R N n
'.-+
lji lR .....,,,,_ R + __,,,,_ . Mg
--t_..... 4 g y
R-A A A A .ao. I. is A A TIME ( seconds )
._____ - - - ____ _ _ _ _ _ _ _ .}}