ML20236S968
| ML20236S968 | |
| Person / Time | |
|---|---|
| Site: | 05200003 |
| Issue date: | 06/30/1998 |
| From: | Campe K, Kudrick J Office of Nuclear Reactor Regulation |
| To: | |
| References | |
| NUREG-1632, NUDOCS 9807270348 | |
| Download: ML20236S968 (139) | |
Text
NUREG-1632 Evaluation of AP600 Containment Thermal-Hydraulic Performance U.S. Nuclear Regulatory Commission Omce of Nuclear Reactor Regulation 1
K. M. Campe l
J. A. Kudrick I
f
?
f ?
),
l
%...6)
~
8[ fhhk O O
3
I i
a AVAILABILITY NOTICE l
Availability of Reference Materials Cited in NRC Publications Most documents cited in NRC pub'ications will be available from one of the following sources:
1.
The NRC Public Document Room, 2120 L Street, NW., Lower Level, Washington, DC 20555-0001 t
2.
The Superintendent of Documents, U.S. Government Printing Office, P. O. Box 37082 Washington, DC 20402-9328 3.
The National Technical Information Service, Springfield, VA 22161-0002 Although the listing that follows represents the majority of documents cited in NRC publica-3 uons, it is not intended to be exhaustive.
l I
Referenced documents available for inspection and copying for a fee from the NRC Public l
Document Room include NRC correspor'dence and internal NRC memoranda; NRC bulletins, circulars, information notices, inspection and investigation notices; licensee event reports; vendor reports and correspondence; Commission papers; and applicant and licensee docu-ments and correspondence.
The following documents in the NUREG series are available for purchase from the Government Printing Office: formal NRC staff and contractor reports, NRC-sponsored conference pro-ceedings, international agreement reports, grantee reports, and NRC booklets and bro-chures. Also available are regulatory guides NRC regulations in the Code of Federal Regula-tions, and Nuclear Regulatory Commission Issuances.
Documents available from the National Technical Information Service include NUREG-series reports and technical reports prepared by other Federal agencies and reports prepared by the Atomic Energy Commission, forerunner agency to the Nuclear Regulatory Commission.
Documents available from public and special technical libraries include all open literature items, such as books, journal articles, and transactions. Federal Register notices, Federal and State legislation, and congressional reports can usually be obtained from these libraries.
Documents such as theses, dissertations, foreign reports and translations, and non-NRC con-ference proceedings are available for purchase from the organization sponsoring the publica-tion cited.
Single copies of NRC draft reports are available free, to the extent of supply, upon written request to the Office of Administration, Distribution and Mail Services Sectio', U.S. Nuclear n
Regulatcry Commission, Washington DO 20555-0001.
l Copies of industry codes and standards used in a substantive manr.er in the NRC regulatory process are maintained at the NRC Library, Two White Flint North,11545 Rockville Pike, Rock-ville, MD 20852-2738, for use by the public. Codes and standards are usually copyrighted and may be purchased from the originating organization or, if they are American National Standards, from the American National Standards institute,1430 Broadway, New York, NY 10018-3308.
NUREG-1632 Evaluation of AP600 l
Containment Thermal-Hydraulic Performance l
Manuscript Completed: May 1998 Date Published: June 1998 K. M. Campe J. A. Kudrick Division of Systems Safety and Analysis Omce of Nuclear Reactor Regulation U.S. Nuclear Regulatory Commission W:shington, DC 20555-0001 p u.uq i
I j
\\...../
I
ABSTRACT The thermal-hydraulic performance of the AP600 containment with respect to selected design-basis accidents (DBAs) was evaluated using the CONTEMPT and CONTAIN codes. The
. AP600 containment design includes a passive cooling system (PCS) in the form of gravity driven water flowing on the external surface of a steel containment shell. This design feature cannot be modeled directly within the CONTEMPT code. The CONTAIN code was used to estimate the PCS heat transfer coefficients that were provided as input into the CONTEMPT code. In order to establish the validity of this approach, a comparison was made of the two codes. The results show fair agreement in terms of containment pressure and temperature response to selected DBAs.
' Confirmatory analyses were made using the CONTAIN code with the intent of verifying the Westinghouse analyses that were performed using the.W_ GOTHIC code. The results indicate that the Westinghouse pressure and temperature estimates for the AP600 appear to be reasonable and are within the applicable acceptance criteria of the Standard Review Plan (SRP).
The CONTAIN code also was used to conduct a series of sensitivity analyses. The purpose of these analyses was to assess the relative importance of the key AP600 containment design features and operating conditions. One aspect of the sensitivity analyses involved the consideration of selected limiting assumptions regarding the principal modes of heat transfer with respect to the containment shell and the internal heat sinks.
i f
i iii NUREG-1632
CONTENTS Page Ab s tract................................................................... iii Abbrevi ati ons.............................................................. x 1.0 Staff DBA Analysis Using the CONTEMPT Code.............................. I 1.1 CONTEMPT Model of the AP600................................... I 1.2 PCS Heat Transfer Coefficients...................................... 2 1.3 DB A Analysis Results...........................
............... 3 2.0 CONTAIN/ CONTEMPT Comparison..................................... 10 3.0 Confirmatory Analysis Using the CONTAIN Code........................... 10 3.1 AP600 B ase Case Model.......................................... 10 3.2 DB A Analysis Results............................................ 13 3.2.1 Double-ended Cold-Ieg (DECLG) Break....................... 13 3.2.2 Main Steam Line Break (MSLB).............................. 17 3.3 Comparison with Westinghouse EGOTHIC Results.................... 26 3.3.1 Westinghouse EGOTHIC Analyses........................... 26 3.3.1.1 Evaluation Model.................................... 26 3.3.1.2 DB A Analysis Results................................ 31 3.3.2 CONTAIN/_WGOTHIC Comparison........................... 31 4.0 Sensitivity Analysis.................................................... 32 4.1 Sensitivity Matrix.............................................. 34 4.2 Sensitivity Results............................................. 35 4.2.1 PCS Wetted Area (DECLG)................................ 35 4.2.2 PCS Wetted Area (MSLB)................................. 41 l
4.2.3 PCS Flow Rate (DECLG)................................. 41 4.2.4 PCS Flow Rate (MSLB).................................. 72 4.2.5 PCS Water Temperature (DECLG).............
........... 72 1
4.2.6 PCS Water Temperature (MSLB)...............
... 72 l
4.2.7 PCS S tart Time (DECLG).................................
72 4.2.8 PCS Start Time (MSLB)................................... 89 4.2.9 Heat Sink Area (DECLG).................................. 89 4.2.10 Heat Sink Area (MSLB).................................. 93 t
4.2.11 Liner-Concrete Air Gap (DECLG).......................... 98 4.2.12 Liner-Concrete Air Gap (MSLB)............
................ 98 4.2.13 Initial Relative Humidity (DECLG)......................... 98
(
v NUREG-1632 l
___A
L CONTENTS (continued)
Page 4.2.14 Initial Relative Humidity (MSLB)............................ 108 Other Sensitivity Analyses........................................ 108 4.3 4.3.1' Sensitivity to Evolutionary Changes in Mass and Energy Source
^
Terms................................................... 109 Sensitivity to Limiting Heat Sink and PCS Assumptions.......... 109 4.3.2 Sensitivity Analysis Summary..................................... 116 4.4 5.0
' Summary and Conclusions.............................................. 122 References..........................................................
6.0 Appendix A: AP600 Sensitivity Analysis Case Identification Nomenclature............ 125 List of Figures 1.2-1 Average PCS film coefficients (LT D-B).............................. 4 1.2-2 Average PCS ambient air temperature (Cl-D-B)........................ 5 1.3-1 Containment pressure (LT-D-B)..................................... 6 1.3-2 Containment temperature (LT-D-B).................................. 7 1.3 Containment pressum (LT-M-B)....................................... 8 1.3-4' Contamment temperature (LT-M-B)................................. 9 2.0-1 Containment pressure (Cl-LT-D-B)..........................,...... 11 2.0-2 Containment temperature (Cl-LT-D-B).............................. 12 3.2.1-1 Containment pressure (Cl-D-B).................................... 14 3.2.1 Containment temperature (Cl-D-B)................................ 15 3.2.1-3 Source energy rate (Cl-D-B)....................................... 16 3.2.1-4 PCS source flow rate (Cl-D-B)..................................... 18 3.2.1 Energy transfer partitioned by heat structure (C1-D-B)................... 19 3.2.1-6 Containment shell cumulative energy transfer (Cl-D-B)................. 20 3.2.1-7 Cumulative energy transfer across containment outside surface 14aw (Cl-D-B)................................... 21 3.2.1-8 Percentage wet / dry energy transfer rates (Cl-D-B)...................... 22 3.2.1-9 Cumulative energy transfer across vertical and dome portions of containment inside surface (Cl-D-B)............................... 23 3.2.2-1 Containment pressure (Cl-M-B).................................... 24 3.2.2 Containment temperature (Cl-M-B)................................. 25 3.2.2-3' MSLB mass and specific enthalpy source (Cl-M-B)...............
... 27 3.2.2-4 Cumulative energy transfer partitioned by heat structure (Cl-M-B)......... 28 3.2.2-5 Cumulative energy transfer across containment outside surface (C l-M-B)........................................ 29 1
NUREG-1632 vi
l l
t l
1 CONTENTS (continued) 1 l
List of Figures (continued)
Page 3.2.2-6 Cumulative energy transfer across vertical and dome portions of containment inside surface (Cl-M-B)............................ 30 3.3.2-1 Containment pressure (C1-D-EM).................................. 33 4.2.1-1 Containment pressure (Cl-D-PCS-A)..............................
36 4.2.1-2 Containment temperature (Cl-D-PCS-A).......................... 37 4.2.1-3 Cumulative energy transfer (Cl-D-PCS-A)..........
................ 38 4.2.1-4 Cumulative energy transfer across containment outside surface (Cl-D-B)...................................... 39 4.2.1-5 Cumulative energy transfer (Cl-D-PCS-A)............................ 40 4.2.2-1 Containment pressure (Cl-M-PCS-A).
............................. 42 4.2.2-1 Containment temperatum (Cl-M-PCS-A)............................ 43 4.2.23 Cumulative energy transfer (Cl-M-PCS-A)
......................... 44 4.2.3-1 Containment pressure (Cl-D-PCS-F)............................. 45 4.2.3-2 Containment temperature (Cl-D-PCS-F)............................ 46 4.2.3-3 Containment shell surface map................................... 47 4.2.3-4 Containment temperatums (Cl-D-PCS-F): PCS at design flow rate......... 48 4.2.3-5 Containment temperatures (Cl-D.PCS-F): PCS at 50% of design flow rate................................................ 50 4.2.3-6 Containment temperatums (Cl-D-PCS-F): PCS at 40% of design flow rate............................................... 51 4.2.3-7 Containment temperatures (Cl-D-PCS-F): PCS at 25% of design flow rate............................................... 5 2 4.2.3-8 Containment temperatures (Cl-D-PCS-F): PCS at 10% of design flow rate........................................... 53 4.2.3-9 PCS coverage profiles (Cl-D-PCS-F).............................. 54 4.2.3-10 Dryout progmssion for full PCS flow............................... 55 4.2.3-11 Dryout progression for 50% PCS flow rate.........................
56 4.2.3-12 Dryout progression for 40% PCS flow rate
......................... 57 4.2.3-13 Dryout progression for 25% PCS flow rate
......................... 58 4.2.3-14 Dryout progression for 10% PCS flow rate.
....................... 59 4.2.3-15 PCS runoff at 100% design flow rate (Cl-D-PCS-F).................... 60 4.2.3-16 PCS runoff at 75% design flow rate (Cl-D-PCS-F)..................... 61 4.2.3-17 PCS mnoff at 70% design flow rate (Cl-D-PCS-F).................. 62 4.2.3-18 PCS runoff at 60% design flow rate (Cl-D-PCS-F)................... 63 4.2.3-19 Containment response to PCS flow rate variation (Cl-D-PCS-F).......... 65 4.2.3-21 Containment response to PCS flow rate variation (Cl-D-PCS-F):
surface 14 aw................................................. 66 4.2.3-22 Containment response to PCS flow rate variation (Cl-D-PCS-F):
j surface 17w.
...... 67 vii NUREG-1632
CONTENTS (continued)
List of Figures (continued)
Page 4.2.3-23 Cumulative energy transfer (Cl-D-PCS-F): shell...................... 68 4.2.3-24 Cumulative energy transfer (Cl-D-PCS-F): concrete.................... 69 4.2.3-25 Cumulative energy transfer (C1-D-PCS-F): steel...................... 70 4.2.3-26 Cumulative energy transfer (Cl-D-PCS-F): vertical wall................. 73 4.2.3-27 Cumulative energy transfer (Cl-D-PCS-F): dome...................... 74 4.2.4-1 Containment pressure (Cl-M-PCS-F)................................ 75 4.2.4-2 Containment temperature (Cl-M-PCS-F)............................. 76 4.2.4-3 Cumulative energy transfer (Cl-M-PCS-F): shell............
........ 77 4.2.4-4 Cumulative energy transfer (Cl-M-PCS-F): concrete.................... 78 4.2.4-5 Cumulative energy transfer (Cl-M-PCS-F): steel....................... 79 4.2.4-6 Cumulative energy transfer (Cl-M-PCS-F): vertical wall................. 80 4.2.4-7 Cumulative energy transfer (Cl-M-PCS-F): dome...................... 81 4.2.5-1 Containment pressure (C l-D-PCS-T)................................ 82 4.2.5-2 Containment temperature (Cl-D-PCS-T)............................. 83 4.2.5-3 Containment outer surface 14aw temperature (Cl-D-PCS-T).............. 84 4.2.6-1 Containment pressure (Cl-M-PCS-T)................................ 85 4.2.6-2 Containment temperature (Cl-M-PCS-T)............................. 86 4.2.7-1 Containment pressure (Cl-D-PCS-TM).............................. 87 4.2.7-2 Containment temperature (Cl-D-PCS-TM)........................... 88 4.2.8-1 Containment pressure (Cl-M-PCS-TM).............................. 90 4.2.8-2 Containment temperature (Cl-M-PCS-TM)......................... 91 4.2.8-3 Cumulative energy transfer (Cl-M-PCS-TM).......................... 92 4.2.9-1 Containment pressure (Cl-D-HS-A)................................. 93 4.2.9-2 Containment temprature (Cl-D-HS-A).............................. 94 4.2.9-3 Cumulative energy transfer (Cl-D-HS-A): steel........................ 95 4.2.9-4 Cumulative energy transfer (Cl-D-HS-A): concrete.................... 96 4.2.9-5 Cumulative energy transfer (Cl-D-HS-A): shell.......................
97 4.2.10-1 Containment pressure (C l-M-HS-A)................................. 99 4.2.10-2 Containment temperature (Cl-M-HS-A)........................... 100 4.2.10-3 Cumulative energy transfer (Cl-M-HS-A): steel....................... 101 4.2.10-4 Cumulative energy transfer (Cl-M-HS-A): concrete................... 102 4.2.10-5 Cumulative energy transfer (Cl-M-HS-A): shell......
............. 103 4.2.11-1 Containment pressure (Cl-D-HS-G)................................ 104 4.2.11-2 Containment temperature (Cl-D-HS-G)............................ 105 4.2.13-1 Containment pressure (Cl-D-I-H)................................ 106 4.2.13-2 Containment temperature (Cl-D-I-H).............................. 107 4.3.1-1 Mass release sources (Cl-D-B)....................................I10 4.3.1-2 Containment pressure sensitivity to mass source rate (Cl-D-B)........... 111 4.3.1-3 Containment temperature sensitivity to mass source rate (Cl-D-B)........ 112 l
1 NUREG-1632 viii i
l A
i CONTENTS (continued)
List of Figures (continued) l Page 4.3.1-4 Mass release sources (Cl-M-B).................................... 113 4.3.1 Containment pressure sensitivity to mass source rate (C1-M-B)........... 114 4.3.1-6 Containment temperature sensitivity to mass source rate (Cl-M-B)........ 115 4.3.2-1 Containment pressure (Cl-D-B)................................... 117 4.3.2-2 Containment temperature (Cl-D-B)................................ 118 4.4-1 Containment pressure sensitivities.............................
.... I 19 4.4-2 Containment temperature sensitivities............................... 120
)
List of Tables 1.1-1.
CONTEMPT input parameters for the AP600 containment model.......... 1 3.1-1 CONTAIN input parameters for the AP600 containment model............ 13 3.3.2-1 CONTAIN mput modifications representative of the EM................. 32 j
4.1-1 Sensitivity matrix................................................ 34
{
4.2.3-6 Containment temperatures (Cl-D-PCS-F):
PCS at 40% of design flow rate................................... 51 4.2.13-1 Containment atmospheric condition sensitivity to initial relative humidity.............................................. 108 1
4.4-1 Containment sensitivity ranking................................... 121 ix NUREG-1632
)
l i
l ABBREVIATIONS l
1 ADS automatic depressurization system I
1 CMT core makeup tank l
CVCS chemical and volume control system j
DBA-design-basis accident
-i
. DECLG double-ended cold-leg (break)
ECCS emergency core cooling system EM Evaluation Model 1
l IRWST in-containment refueling water storage tank.
LOCA loss-of-coolant accident.
l l
MSLB main steamline baak l
PCS-passive cooling system PWR pressurized-water reaactor.
l l
SG steam generator SRP Standard Review Plan l
9 i
4
-l NUREG-1632 x
EVALUATION OF AP600 CONTAINMENT THERMAL HYDRAULIC PERFORMANCE 1.0 Staff DBA Analysis Using the CONTEMPT Code i
In accordance with Standard Review Plan (SRP) Section 6.2.1.1.A, the staff may elect to perform l
a confirmatory analysis of a pressurized-water.cactor (PWR) dry containment. The purpose of l
the analysis is to confirm the applicant's predictions of the containment response to design-basis accidents. The analysis involves the use of the CONTEMPT code (version CONTEMPT-LT/28, NUREG/CR-0255) to evaluate loss-of-coolant and main steam and feedwater line breaks. This section describes how the AP600 was modeled for the analysis of DECLG (double-ended cold-leg break) and MSLB (main steam line break) events using the CONTEMPT code.
i I.1 CONTEMPT Model of the AP600 Westinghouse AP600 containment design data (WCAP-14407, Revision 1) were used to constmet a 20-heat sink,1-celllumped parameter representation of the AP600 containment. The i
choice of 20 heat sinks and a single cell was dictated by the input limitations of the CONTEMPT code. The total containment free gas volume, modeled as a single cell, consists of the main free l
volume bounded by the containment shell, as well as the reactor cavity, accumulator cavities, steam generator (SG) rooms, core makeup tank (CMT), chemical and volume control system (CVCS), and in-containment refueling water storage tank (IRWST). The key input parameters are summarized in Table 1.1-1.
i i
i Table 1.1-1. CONTEMPT input parameters for the AP600 containment model.
4 Design / Operation Feature Input Value l
Containment volume (m')
50,466 i
Initial containment pressure (KPa) 108.3 Initial containment temperature (K) 322.0 Initial atmospheric temperature (K) 319.3 Initial containment rel. humidity (%)
0.0 2
Liquid pool area (m )
83.8 Approximate steel heat sink mass (kg) 1,602,000(mise. steel) 1,644,000(cont. sh. ell)
Approximate concrete heat sink area (m )
6,500 2
l 1
The key CONTEMPT code features related to licensing-basis analyses of heat transfer are l
Tagami heat transfer coefficient, used in analyses of loss-of-coolant accidents (LOCAs) such as the DECLG while in the condensing mode.
Uchida heat transfer coefficient, used in analyses of MSLBs while in the condensing mode.
The condensing mode heat transfer is assumed to be diven by the difference between steam saturation temperature and the surface temperature of the heat sink. An 8%
revaporization factor is applied to account for reevaporation of condensed superheated steam during the blowdown phase.
When not in the condensing heat transfer mode, natural convection heat transfer coefficients are used for LOCAs and MSLBs. In this case, the heat transfer is driven by the temperature difference between the bulk vapor temperature and the temperature of the heat sink surface.
Conventional PWR containments rely on active emergency core cooling systems (ECCSs) and containment sprays for controlling the pressure and temperature inside the containment. The CONTEMPT code has provisions for modeling the performance parameters and actuation times for these systems. The AP600 containment design relies on a passive cooling system (PCS) for energy removal from inside the containment as a means of limiting the containment pressure and temperature. This consists of gravity-driven cooling water applied to the extemal surface of the containment shell.
The CONTEMPT code has no direct means of calculating and tracking PCS water film flow and heat transfer behavior on the outside surface of the containment. However, the code permits user specification of a time-dependent heat transfer coefficient with respect to the surface of a selected heat sink. In view of this, an estimate was made of the PCS heat transfer coefficients using the CONTAIN code (NUREG/CR-6.o3). The coefficients were obtained as described in the next section.
1.2 PCS Heat Transfer Coefficients The approach used was to run the CONTAIN code for the same single-cell model of the AP600 used in the CONTEMPT code. The output includes time-dependent heat transfer rates across the AP600 containment shell as well as the corresponding ambient air temperatures. Average heat transfer coefficients were obtained for wet and dry portions of the containment surface on the basis of the respective surface areas and corresponding temperature differences.
Specifically, the heat transfer coefficient h(t) for the containment shell was obtained as an area-and temperature-weighted average of the individual cell coefficients h,(t):
l 1
l NUREG-1632 2
A m
[ A (t) A, (Ut)ffe - T(t),,,,)
j h(t) =
[ A, (71t)f,g - T(t),,,,)
where A,,,
= total shell surface area J,
= shell surface area within celli T(t),,,,,,
T(r),,,,
= PCS film-to-air temperature difference for cell i.
Similarly, the average temperature 7,,(t) of the ambient air near the shell was obtained as an area-weighted average of the individual cell air temperatures T,,ft):
A 0r n
[ A, T,,}t)
T,(t) =
g
[ A, Figures 1.2-1 and 1.2-2 show the resulting plots for 5(t) and 7,(t) respectively, for a DECLG g
event. They were used to generate containment outer surface boundary conditions for use in the
{
CONTEMPT input specification.
1.3 DBA Analysis Results CONTEMPT calculations were made for DECLG and MSLB events using the previously described single-cell model of the AP600 containment. Figures 1.3-1 through 1.3-4 show the resulting AP600 containment response.
The pressure response for the DECLG event is shown in Figure 1.3-1. It has the characteristic of a double peak, corresponding to the two distinct phases of internal energy release - initial blowdown of primary coolant through the pipe break and a subsequent phase stemming from reflood steam injection. Both peaks are relatively close to the design pressure. The first peak is about 88% of the design pressure and the second peak is at about 92%. The shape and maximum value of the second peak is determined by the action of the PCS on energy removal through the containment shell. The CONTEMPT code models this through the PCS heat transfer coefficient described earlier. Separate calculations assuming no PCS actuation show complete absence of a local maximum,i.e., the pressure continues to rise beyond the design pressure. The corresponding. temperature response also shows a double peak, as seen in Figure 1.3-2.
l Figure 1.3-3 shows the pressure response for the MSLB event. The peak pressure is very close l
to the design pressure (about 98% of design pressure). The corresponding temperature response, 3
!lj!li 1
11iI1llill1iIilllllil1l1l)ili l
,ll l) li oooon
.. i 00
. 0 0
\\
1 Y*
)B D
T-b 00 L
. 0
(
1 s
t
)
n
/
c e
i e
c i
s f
(
feo e
c m
m iT l
if S
00 C
1 P
e gare vA 1
-2 1
.g o
i F
t
~
1 0
g M
0 0
0 0
0 4
3 1
^Ml"EN3.l l
ZC
- .Cu
^
i' i
!l
lllll jl I
llII il l ;~
jl i
l
)B
-D l
C
(
s y
e r
ru tarep
)
m c
e e
t s
r
(
ia t
e n
m ie i
b T
ma a,
SC P
e garevA 2-2 1
e g
c iF 0
s 0
5 8
J 3
2 3
J 3
g h.*k I
u 3@9&M l
i
(ojsd) a;nsso>d n
n n
n n
Q s
a s
a 3
-8 5
l l-I I
i I
.I g
o o.
A 3
d
-sl 2
I E
o g
n.l n
a, a,
e u
S l
l fl n
il E
g u
F u
a l
'I l
g 6
)
ob E
.h l
El 1
l I
)
I l.
3 3
2 (Dd) L0000'0 x a;nsseJd NUREG-1632 6
I
I I<I 1I1
\\,lll1l\\<Iii-l 1
'I\\
{jl{l'1lii!l li!!!l,I l,
^u. * *u3 O W o E eH o
4 4
4 4
4 4
e 4
e 4
4 4
4 7
s N'
4 3
2 0
3 a
7 6
5 4
1 2
2 2
2 2
2 2
1 1
1 1
1 1
m
~
oo l
h i
)B-D
-T L
(
e A,
ru t
are
)
p
^
c m
e e
s t
(
tn
^
e e
m m
^
n i
i T
a tn g
o C
2
-3 1
.8 iF 4-0 0
o o
0 c
0 0
o 1
0 u
7 e
5 4
3 4
4 n
s 3
a 3
3 3
^Mv Dss o gEEeH 4
2C b Caewto lii llt i
t
\\
(D!sd) a;ncsedd n
n n
n Q
Q i --
i i
7
~h l
I I
I d
I 1
n o
c?
!ii
?
H R
d 2
e gl 9
5
$-l c
+I.
E N
i u
p-
.g "a
l i f 6
4n
~
.EP l
u I
2 I
I
\\
~
i l,
Q R
a S
(Dd) [0000'0 x a;nsss>d NUREG-1632 8
(J,) a;nto;adwei R,
R R
R R,
8' R
an
- g i
'l q
1 n
a?2 l
E'-d e
. l 2
a 8.
mu E
3.
2
~
u M
E
=
A 3e 8
0
.T ea g
1 1
i i
i
\\
t 1
j 1
R I
(x) santo;sdwel 1
1 l
4 l
shown in Figure 1.3-4, has a peak value of about 458 K. It should be noted here that internal mixing and stratification effects in the case of a MSLB can be very dependent on the type of steam injection and its location (see AP600 SSAR and Tills,1996). Hence, the CONTEMPT results for the MSLB are subject to the uncertainties associated with injection effects.
2.0 CONTAIN/ CONTEMPT Comparison The AP600 containment response to DECLG and MSLB events, calculated using the CONTEMPT and CONTAIN codes, shows general agreement with respect to the gross features of the pressure and temperature profiles. In panicular, the DECLG pressure and temperature response to the initial blowdown phase is in good agreement, as indicated in Figures 2.0-1 and 2.0-2. However, some differences are seen in the peak pressure and temperature values corresponding to the reflood phase. Specifically, CC,NTEMPT peak pressure is about 17 KPa 1
(2.5 psi) higher than that for CONTAIN.
l As can be seen from Figure 2.0-1, the CONTEMPT pressure appears to stan deviating noticeably from CONTAIN at about 76 seconds. Examination of the energy flowing into the internal heat
,j i
sinks (including the containment shell) indicates that the rate of energy transfer starts to deviate between the two codes at about that time. Specifically, the energy transfer rate in CONTAIN is about 1.2 times that in CONTEMPT at 76 seconds. This ratio goes up to about 1.6 at 200 I
seconds and approaches 2.0 by about 500 seconds.
The preceding observations suggest that the deviations could result from differences in modeling of heat transfer to the heat stme'.ures. Specifically, surface heat transfer correlations, heat structure masses, surface areas, and material properties are possible sources of the deviation between the two codes.
3.0 CONFIRMATORY ANALYSIS USING THE CONTAIN CODE 3.1 AP600 Base Case Model The AP600 containment model used in the CONTAIN analyses is based on a version that was developed for earlier studies done by SANDIA. A full description of the original model is presented in the AP600 SSAR. The original model used a multi-cell representation of the containment atmosphere. The multi-cell configuration permitted representation ofindividual subcompartments and also presented a means of estimating the effects of containment atmosphere stratification.
In a separate study (Vijaykumar and Khatib-Rahbar,1997), a comparison was made between CONTEMPT and CONTAIN codes. For the comparison study, a new model was constructed by making two principal modifications: (a) collapse of the multi-cell geometry to a single-cell representation of the AP600 containment atmosphere and (b) reduction of the total number of heat structures from 176 to 20 (10 intemal heat sinks,10 containment shell pieces). These modifications were dictated by the CONTEMPT limitations of a single-cell containment NUREG-1632 10
l l
l (u!sd) a;nssa>d i-l 9
9 9
9 9
9 3
3 4
4 g
s l
g I
I 1
I 1
. l i
.9 8?
o C
l 6
l nl-4 l
2 y
sl 1
e a
+
e
't 4.l
}
S E.
s
- l e
E a
o l
g A
3 gl
- g j
dW a
C l
l l
1.
=
l
.I I
i l
l.
9 9
9 9
J n
n (Dd) L0000'0 x a;nssaJd 1
11 NUREG-1632 i
i (3,) a;nto;edwel 1
a a
a 3
3 A
x
-8 A
n n
n n
E h
5 g
A 4
B g
2 a
2 S
h l
??
x E
E m
.e5 8
8 u
dq N
e6 it 2
f f
f i
9 9
t B
i 5
R R
E R
R R
R
(>l) a;nto;adwal l
NUREG-1632 12
atmosphere and a maximum of 20 heat stmetures inside the containment. A similar study, using a single-cell 31 heat sink model (17 internal heat sinks,14 containment pieces) was made by the staff. In this case, the heat sinks were modeled on the basis of design data presented by Vijaykumar and Khatib-Rhabar.
Staff analyses using the single-cell model yielded a peak blowdown pressure that was about 6%
higher than the multi-cell results. The reflood peak pressure was slightly lower (by about 1%)
for the single-cell model. Overall, however, the results were reasonably close. Similar results were obtained by Vijaykumar and Khatib-Rhabar. They also found that the reduced number of heat sinks had a negligible effect on containment pressure. Hence, a single-cell reduced heat sink model was judged to be adequate for performing the sensitivity studies. The heat sink specifications as well as other CONTAIN input items are summarized in Table 3.1-1.
Table 3.1-1. CONTAIN input parameters for the AP600 containment model.
Design / Operation Feature Model Description Contamment volume (m')
50,466 I
Initial containment pressure (KPa) 108.3 Initial containment temperature (K) 322 Initial atmospheric temperature (K) 319.3 Initial PCS temperature (K) 322 Initial containment rel. humidity (%)
0.00 PCS start time (sec) 660 Initial (max) PCS flow rate (kg/sec) 13.8 Heat sink (steel) man (kg) 1,749,490 (misc. steel) i 1,706,320 (cont. shell)
{
Heat sink (concrete) areas (m )
1,526 3.2 DBA Analysis Results 3.2.1 Double-ended Cold-Leg (DECLG) Break l
For the base-case DECLG break, the containment pressure and temperature responses (Figures j
3.2.1-1 and 3.2.12, respectively) are similar to what had been estimated in earlier analyses. The l
principal features are the pressure and temperature peaks corresponding to the early (blowdown) and late (long-term cooling) phases of the DECLG event, separated by a transition period (reflood). The maximum pressure is associated with the second peak and is 363 KPa (52.7 psia).
f This is about 88% of the design pressure of 414 Kpa (60.0 psia). Figure 3.2.1-3 shows the source i
13 NUREG-1632 l
(ojsd) e>nsss>d n
n n
n n
n a
s E
M 38
,g I
I I
l
)
.l.
6 9
8 6
d al 1
2 l !
l E
9 W [
O f,I l
c
,l E
e a
1:
3 1 E b
~
4 N
d I
.e w
I I
]
l I
I.
3 3
2 S
(Dd) [QQQQ*Q X GJnSSGJd
- NUREG-1632 14
l
- l; lI!ii ililll{\\f!>ll:ll 1I 3
lji1 3i11lI 1!lil' Ak'v eu3EtEE*
4 M
4 4
4 4
4 4
0 9
4 4
4 4
[y0 7
5 4
3 i
7 6
5 4
2 2
2 2
m 2
1 1
1 1
1 0
0 0
1 000 n
- 0 1
)B-D-l T
C
(
e 0
r 0
u a
0 t
1 a
re
)
p c
m e
e s
t
(
tn e
e m
mn i
i T
a tn 0
o s 0 C
1 2
1 2
3
.g iF a
0 1
i 0
0 0
M 0
0 0
0 0
0 0
7 s
5 4
3 4
4 3
3 3
3 3
3
^MV
- oe@=aEe-u ZC$a'.0w 1
l\\'
ll 1
l
i\\i\\i\\!
,i i'
1i\\
oooo
~
c i
T
~
ooo
. o t
t e
m o
p
&m
)B e
e
.t o
D s
o
. c l
C i
(
)
e c
t e
a r
s y
(
gr e
en m
~
e iT e
c
[
r o
u o
o s
S
-1i N
3-1 2
3
.g iF r
o e
e i
m e
r r
r~
r f
f f
Y f
"o o
0 I
i i
1 i
t s
1 eoew%3 ou $uecJu ZC $ 04mt"
. a
\\I<
!!!I\\;ll l
llt
.l l
energy profile used in the analyses. The PCS source flow rate profile is shown in Figure 3.2.1-4.
Figure 3.2.1-5 shows the panitioning of energy transfer with respect to heat sinks. Up to about the time of PCS actuation the dominant transfer mode is energy storage in the steel structures inside the containment and the steel containment shell. Energy storage in concrete is less pronounced. However, it continues to absorb a significant amount of heat energy for a longer time period than steel. Internal steel appears to be saturated in the neighborhood of 2000 seconds, whereas storage in concrete continues well beyond 10,000 seconds.
Energy transfer and storage with respect to the steel containment shell differs from what is seen for the internal steel stnictures. This is due to the presence of a heat transfer path through the outer shell surface into the extemal environment. Figure 3.2.1-6 illustrates the combined effects of energy storage and transfer with respect to the PCS shell. The graph is in reference to a circumferential ponion of the PCS shell nearest the operating floor (see surface 14aw in Fig.4.2.3-3). As can be seen, energy transfer into the steel shell begins almost immediately.
However, most of the energy goes into heat storage within the shell, so that significant heat rejection to the outside does not appear until about 50 to 100 seconds into the event. The figure also shows the effect of the PCS beyond 660 seconds. Due to the greatly enhanced heat transfer rate via the PCS cooling water, the heat rejection to the outside begins to approach the energy being transferred to the shell from the inside. Beyond 10,000 seconds,it appears that an equilibrium is reached between the energy source inside the containment and extern-J energy rejection via the PCS.
Figure 3.2.1-7 provides a more detailed description of the PCS cooling effect on the energy 1
transfer. Here the total energy transfer across the shell surface 14aw has been resolved into l
" wet" and " dry" components. Until PCS actuation at 660 seconds, this surface, as well as the l
entire shell outer surface, is dry and all of the energy transfer to the outside is by air convection.
However, after the arrival of PCS water, nearly all of the energy transfer is across the wetted portion (surface 14aw). Figure 3.2.1-8 illustrates this on a percentage basis. As can be seen, the wetted ponion eventually is responsible for about 90% of the total energy transfer across the shell surface.
l Another aspect of energy transfer across the shell is the panitioning between the dome and venical portions of the shell (Figure 3.2.1-9). Until PCS actuation, the differences between energy transfer across the dome and vertical portions of the shell primarily are due to the difference in the heat transfer areas. After PCS actuation, the degree of energy partitioning between dome and venical wall changes. This change is attributable to wetted area differences l
between the dome and venical portions, which, due to enhanced heat transfer, accentuate the energy partitioning.
3.2.2 Main Steam Line Break (MSLB)
The pressure and temperature response to an MSLB is characterized by a single peak (Figures 3.2.2-1 and 3.2.2-2, respectively). The temperature peak is 441 K (334 "F) at about 41 seconds, 17 NUREG-1632
l i
5 5
U
- i o
3
- 7 is
~N e
ve o
8 E
=
E A
5 ci m
es ti:
I S
E 3
a
=~
(cas/6 ) ajo; mog M
NUREG-1632 18
1
)ces/J( etar refsnart ygrenE 5
i 4
S
=
I i
\\
l
\\i
-l s
i
\\l b
O i
h E
cn i'\\
a i
/'
n
\\ ,s \\
\\
,,, : H S
ig g
e
~
\\
!,1 "8
n
\\
,'l
.E 7
e
\\
2
\\
3
?
+,.
i s
gl l
1-n g
S\\
.8 p,
,E. l j
R q
\\
m
\\
\\
f
/
h s N
I E
i
\\
/
s
\\'
}
h
'fa s
g
\\
i s '
s, N\\
3
s, N
'\\s s
-y,,
5 5
4 4
4
)J(refsnart ygrene evitalumuC 16 NlfI130-19EZ
(n;g) X6;au3 b
b b
b o
g--
g-._-..
g....--
u--..--
g
? "
a 1
.e s
-O E
G e
q 8
y v
un
.E v
y 0
E E
i:
8=
8 iso a
Ec
\\.
e 1
e p
E.
ej
- o m
bk)
C t.
f.
f.
1.
"e e
'e i
D
'8 (r) X6;au3 NUREG-1632 20
j e,
'0 f
e
,0 1
i i
,0 0
_00 1
y 0
r 0
d 0
0 t
1 e
w 00 0
.. 1
/
l-I i
I l
I i
l
)
c la es to
(
t e
m iT 00
. 1 0
. 1 1
"c "0
v v
g o
i i
n I
1 m3xFEw M
g9
.i h
N
~
a 6
N B
1 5
eu l
4 L
)
~
N 9
v 8
e u
E D
1:
3o
. 8 g,
a Cu e
f.
4 E.
m e
_:~
tt 8
O S
8 g
S R
itsys ssoJ3D JG)SuoJ) A6;aus logo) 10 %
~
1 NUREG-1632 22
l t
l)
!I Il I
3 mwWCW e
'0
'0 9
0 g0 i
1 1
1 i
00
)
_ 0 B
0
. 1 D
l C
(
eca fr u
s e
0 d
0 i
. 0 s
0 n
1 i
tne m
n ia t
n oc f
o s
00 n
0 o
~
1 i
t r
o
)
c p
e e
s m
(
o d
e d
m n
i a
T l
ac 0
u 0
r i
t o
1 w
ev lac s
it s
ro o
w rc a
e m
re o
f s
snar t
y 0
g 1
r en e
ev i
ta lu muC 9
r r
1 1
p 2
0 0
0 1
1 1
1 1
3 n
AeeCW g
i F
4W zC$P $
t
l l
l' (ojsd) a;nsse>d S
2 E
E
}!~
I I
I I
)
l 1
e
?
E o
s e
al e
f c.
n 8
U E
+
a
=
n
,l 5
6 a
I a
i n.
m l
I
- 8 l
l l
II I
g o
o (Dd) L0000'0 x a;nssedd NUREG-1632 24
(.d.) 83nto>adwal 0
I n
m.
22 U
v em I
E g
5 n
u o
e s
g E
a 1:
E Ou g
a A
1, n
m
~
i e
iE
- e i
t I
t 5
8 R
R
(>l) a;nlo;adwal I
25 NUREG-1632
1 whereas the pressure peaks later, reaching 367 KPa (53.2 psia) at about 335 seconds. The time displacement is due to the presence of superheat, as can be seen from the saturation temperature l
curve shown in Figure 3.2.2-2. It should be noted that both temperature and pressure peaks occur before PCS actuation. Hence, PCS cooling has no influence on the temperature and pressure maxima for this event. This aspect is discussed in more detail in the description of PCS sensitivity results. Figure 3.2.2-3 shows the source mass and specific energy profile for the MSLB. The PCS profile is the same as for the DECLG event (Figure 3.2.1-4).
The energy partitioning among the heat structures is shown in Figure 3.2.2-4, and the results are similar to those for the DECLG event. One difference is noticeable, however, with respect to the intemal concrete and steel heat sinks. As seen in Figure 3.2.2-4, heat storage in steel reaches a maximum in the neighborhood of 1000 seconds, followed by a decrease. This indicates that after about 1000 seconds stored energy is being retumed to the containment atmosphere. Similarly, for concrete, a maximum is reached at about 7000 seconds, with subsequent heat flow reversal.
The dry versus wet heat transfer across an outer surface of the shell (Figure 3.2.2-5) shows characteristics similar to those observed for the DECLG event. The energy transfer split between the dome and the vertical portion of the shell (Figure 3.2.2-6) is similar to what was observed for the DECLG event.
3.3 Comparison with Westinghouse EGOTHIC Results 3.3.1 Westinghouse EGOTHIC Analyses 3.3.1.1 Evaluation Model In support of the AP600 containment safety evaluation, Westinghouse analyzed a number of postulated design-basis events involving the release of primary system mass and energy into the containment gas volume. The analyses were rnade using the Westinghouse-GOTHIC (WGOTHIC) computer code and a model of the AP600 containment referred to as the Evaluation Model (EM).
The main feature of the EM is a lumped parameter nodal representation of the containment and its internal volumes. Another key feature is the use of a special structure called a " clime" for modeling the mass and heat transfer of the passive cooling system (PCS) on the containment shell. Principal design aspects incorporated into the EM are initial conditions, structural steel and concrete heat sinks, inter-compartment flow-path specifications, mass and energy release profiles, and PCS characteristics (i.e., actuation times, flow rates, coverage fraction).
Some of the main assumptions inherent in the EM are the use of free convection on all inner containment surfaces, the neglecting of heat transfer to floor surfaces (on the premise that this bounds cell circulation and stratification effects), and the absence of convective and condensation heat transfer modes within dead-ended compartments after 30 seconds.
NUREG-1632 26
(6 /r) Ad oy)ua o!J!aads M
i s
i d
1 s
I m.
2 s
r-l u
o 1
e:s s
o e
x
-ne l
c m
o W
e 1
l m
c
. d" e
e m
i 1
E E
1:
m a
i 5
m a
d I
-h 2
z M
l 4
m.
l me e
I i2 t
l
- t..
t..
n_.
t.---
Io s
3
-s s
d s
b L
L k~
(cas/6 ) ago; sson M
l 27 NUREG-1632 L
ll 11l!ll!/il1i!)l;illl!jl!ll!llllll}l m5.i*A@gecj n
s L
9 q
'o v
v0 r
1 s
t t
i00
~~
_ 0 0
_ 1
)
~
B
-M x~
. 0
(
0 l
00 C
1
~
e r
u t
~
c u
r ts ta e
h y
~
b 0
~
0 d
0 e
~
1 n
o i
)
t
/
i c
tr e
a s
p
(
/
r e
e f
s m
n
/
a i
r T
t y
l 0
g 0
r 1
en e
/
e v
i
/
t a
l e
u w
/
m a
u r
l C
s 4
i
/
0 2
u r
1
/
2 i
3
's
/
g r !
i
/
F
't
/
l '#
i "e
9 9
'0 "0
1 1
1 1
u q*Ao,ueCW ZCgOL.S" o@
^3$ xe&Cls.
t f
'0 f
- 0 i
1 I
1 100
- 0
- 0 1
)
B M
l C
0
(
00 e
- 0 c
1 a
frus e
l
, e d
n i
s t
u o
tn e
m 0
0 n
0 i
1 a
t n
)
o c
c Y
e s
s s
(
or c
e a
m r
e i
f T
sn a
0 r
0 t
^
1 y
^
g
^
ren
^
eev i
t s
a e
l hs u
m la t
u e
C.
l 0
1 5
^
^
-2 2
3
.g i
F r
r r
r 1
8 "0
'0 0
0 1
1 1
1 1
1 m3 xP* w mc 7CC9 ll
00000
. 1 0
0
. 0 y
. 0 r
. 1 d
/
tew
\\l l
)
00
.. 0
. 1 f
MI l
I I
ces
(
la e
to m
t iT 0
. 0
. 1
. 0
. 1 1
d e
'0 "0
0 1
1 1
1 1
nmw $ eO ZC$9EM a
l il i'
l}
3.3.1.2 DBA Analysis Results Using WGOTHIC, Westinghouse analyzed a spectmm of pipe breaks representative of postulated LOCA and MSLB events (WCAP-14407, Chapter 6.2).
LOCA ever.ts were characterized by postulated double-ended guillotine pipe breaks occurring in
. either a hot or a cold leg of the reactor coolant system. As described in WCAP-14407, mass and energy releases beyond approximately 10 seconds were modeled in four phases: blowdown, refill, reflood, and post-reflood.
The blowdown phase covers the time from accident initiation to pressure equalization between the broken primary loop and the containment atmosphere.
Following blowdown, the refill phase covers the time it takes the passive core cooling system to refill the reactor vessel lower plenum. Relatively little mass and energy release was estimated for this phase by Westinghouse. As a conservatism, the refill phase was modeled by Westinghouse using the higher mass and energy release rates occurring at a later time, when the location of the steam release is postulated to change from the break to the automatic depressurization system (ADS) Stage 4 valves.
The reflood phase covers the time from initiation of core flooding to complete core quenching.
After core quenching, the post-reflood phase starts, characterized by energy flow to the reactor primary system coolant from reactor coolant system metal, core decay heat, and the steam generators.
With respect to LOCA pipe breaks, the EM for the DECLG produced the highest containment peak pressure (44.0 psig). Figures 1.2-1 and 1.2-2 show the time-dependent containment pressure and temperature response for the DECIE event.
Westinghouse also considered a spectrum of steam-line breaks, as described in WCAP-14407.
The highest MSLB pressures and temperatures calculated using the EM correspond to 102%
(43.6 psig) and 30% (44.8 psig) power levels. Figures 1.3-1 and 1.3-2 show the time-dependent pressure and temperature response for the MSLB event.
3.3.2 CONTAIN/_WGOTHIC Comparison As indicated in Section 3.3.1, Westinghouse performed a series of analyses of the AP600 design using the WGOTHIC code. 'A review of the Westinghouse analyses and the code itself indicates
- that the pressure and temperature estimates for the AP600 appear to be reasonable and to provide sufficient indication that the design meets the applicable acceptance criteria described in SRP Section 6.2.1.1 A.
In order to gain additional confidence in the Westinghouse analyses, the staff performed selected 31 NUREG-1632 L_
confirmatory analyses of the AP600 using the CONTAIN code. The multi-cell CONTAIN base-case model of the AP600 was modified in order to represent the Westinghouse Evaluation Model. The modifications involved using the WGOTHIC initial and boundary-related conditions for the Evaluation Model (including the most recent representations of the mass and energy source rates, as well as PCS flow rates and actuation time) within the CONTAIN input. The specific items modified in order to represent the Evaluation Model are summarized in Table 3.3.2-1. The rest of the CONTAIN input was the same as the base case model described previously.
Table 3.3.2-1. CONTAIN input modifications representative of the EM.
Input Parameter Base Case EM Mass and energy source Original profile (Tills,1996)
Latest values (WCAP-14407, Rev.1)
PCS flow rate Original values (Tills,1996)
Latest values (WCAP-14407, Rev.1 PCS actuation time 660 sec 337 sec Cor' / aent surface heat 1.0 0.73 trm ser multiplier (inside shell)
Containment surface heat 1.0 0.84 transfer multiplier (outside shell)
Heat transfer to heat sinks in On Off after 30 seconds dead-end compartments Figure 3.3.2<1 shows a direct comparison of the WGOTHIC Evaluation Model pressure curve with the pressure response calculated by CONTAIN. The results indicate fairly good agreement up to the reflood peak.
4.0 Sensitivity Analysis One aspect of design certification of the Westinghouse AP600 containment is the relative importance of the various containment design features and operating conditions. Specifically, in order to determine the direction and focus of design-basis accident (DBA) analyses, it is helpful to identify those variables that have a significant effect on the thermal-hydraulic behavior of the containment and to understand the underlying reasons for their importance. A set of sensitivity analyses was performed with the CONTAIN computer cc.de in order to estimate the impact of various design parameters on the response of the AP 600 containment to design-basis accidents l
(DBAs). The base-case input model for these analyses has been described in Section 3.1. The t
NUREG-1632 32
(o!sd) a;nssa;d 9
9 9
9 9
9 l
a x
3 g
a sg 3
I 3
4 4
1 I
^
E I
l
.. i 2
ol IG d
4 al 6
E a.l
- 8 o e g
s 2
+
e 2
!\\
3
?
il n
s E
.E e
P S
I"I Oa A
5
}l
.9 u.,
l o
l
~
I I
I.
x a
.a o
a 4
4 a
(Dd) L0000'0 x a;nssa>d 33 NUREG-1632
results of the sensitivity analyses are presented in this report primarily in the form of f
containment pressure and temperature profiles, as well as energy transfer into heat sinks and l
across the containment shell.
4.1 SensitivityMatrix As indicated in Section 3.1, th'e parameters in Table 3.1-1 were used to define a base-case input.
The base case input represents the nominal design values and initial conditions for the AP600.
Selected variations of these parameters were used to gauge the sensitivity of containment pressure and temperature. These are summarized in Table 4.1-1.
Table 4.1-1. Sensitivity matrix Parameter Base Case Sensitivity Values PCS wetted area 81% of total surface area 59% of total surface area (DECLG) 28% of total surface area 21% of total surface area l0% f Lotal syf_ ace area 2.
(MSLB) 59% of total surface area 28% of total surface area PCS flow rate 13.8 kg/sec (max) 50% of base case 10% of base case PCS water temperature 322 K 273 K PCS start time 660 sec 10 sec 1000 see 2
Heat sink area Misc. steel: 23,752 m 50% of base case 2
Mont. shell: 5,372 m 2
Concrete:
1,526 m Liner-concrete air gap 5 mils 40 mils,100 mils Initial relative humidity 0%
100 %
Both DECLG and MSLB sequences were analyzed using the sensitivity values in Table 4.1-1. In order to help identify individual cases, a systematic nomenclature was adopted for the sensitivity calculations. For example, the identification name Cl-D-PCS-T. PRESS represents a pressure (PRESS) curve obtained from a CONTAIN 1.2 calculation (Cl) of the DECLG sequence (D) involving a PCS water (PCS) temperature (T) variation. A more detailed description of the naming convention is given in Appendix A to this report.
NUREG-1632 34
4.2 Sensitivity Results 1
4.2.1 PCS Wetted Area (DECLG)
One factor with respect to the effectiveness of the PCS in cooling the AP600 containment shell is the extent of wetting of the shell surface. In the base-case model the assumed PCS coverage is about 81% of the total shell surface. This value is approximately the extent of water coverage
. postulated by Westinghouse on the basis of water coverage tests. It is reasonable to expect that small deviations from this nominal value can be made independently of the PCS source flow rate.
That is, water coverage deviating by less than about 20% from the base case can be used in conjunction with the design PCS flow rates in order to estimate the sensitivity of containment thermal response to water coverage. CONTAIN calculations for areas above 81% do not show any noticeable changes in the containment pressure and temperature responses. This suggests that the heat transfer across the containment shell is rate limited.12sser amounts of coverage lead to reduced energy transfer to the outside, hence raising the reflood pressure and temperature peaks, as shown in Figures 4.2.1-1 and 4.2.1-2 respectively. The reduced heat transfer across the shell and the resulting temperature rise inside the containment lead to increased beat storage in internal heat sinks, as shown in Figure 4.2.1-3. As can be seen, both concrete and internal steel l
show increased energy storage as the PCS wetted flow area is reduced to 21%.
l l
Also shown in Figures 4.2.1-1 and 4.2.1-2 are containment response curvec for water coverage I
areas less than.50% of base case (specifically,28% or less of totel containment area). What should be noted here is that the reduced wetted areas were specified independently of flow rate considerations. That is, even for extremely small wetted areas, full-flow conditions were maintained. In reality, small wetted areas could only be maintained at reduced flow-rates.
Hence, the reflood peak pressures and temperatures would show higher increases than those shown in Figures 4.2.1-1 and 4.2.1-2, respectively. For example,10% coverage would be associated with a substantially reduced PCS flow rate. This in turn would lead to a relatively early dryout and consequently significantly higher reflood maximum pressures and temperatures.
Figure 4.2.1-4 shows the effect of PCS wetted area reduction on the split between wet and dry j
heat transfer across the shell. Specifically, energy transfer across the containment outside surface j
is shown for both wet and dry portions. As expected, reduced PCS area coverage leads to a smaller total energy transfer across the shell, as well as a smaller fraction going across the wetted portion.
Figum 4.2.1-5 shows the energy transfer split between the containment dome and vertical wall.
Although the total energy transferred is less when the PCS wetted area is reduced to 21%, the portion going across the dome is seen to increase. This " reversal" is a consequence of the i
geometry of wetted area reduction. The reduction was restricted to the vertical wall area. Hence, the wetted area of the dome did not change. The reduced total wet area led to less total energy transfer, raising the inside temperature, which in tum led to a higher energy transfer rate across the dome.
1 I
35 NUREG-1632 i
1 I
l
l
(
1 1
(D!sd) a>nsss>d n
n n
n n
n 3
3 4
4 a
2 -
5 E
I E
5 l
I s
E 5
E.
~
l
. l-l
'k l
l o!
9 s
I 5
0 l
- I a
g o
al a
?
I E
E 4 1 1
3 o
e E
eI E
.E 1
u A
E
- 8 0
4 El n.
e 4a l
tt l
9 l
l I
I f
I f
7 3
2 5
(Dd) L0000'0 x a>nssaJd NUREG-1632 36
iII ill
!ll\\
1 l1 m oV e'"32eoEG-
) -
l 4
4 4
4 4
4 4
4 4
4 4
4 4
7 9
5 4
3 2
1 0
,4 d
8 7
6 5
4 2
2 2
2 2
2 2
2 i
1 1
1 1
1
~
~
1 Z
1 M
x 2
i a
00 0
. 0 1
)
A
-SC P
o D
e r
l F
o C
to
(
t 0
o t
0 e
_ 0 ru f
1 o
t a
E
)
r o.
c e
s e
p e
s m
r
(
e d
t e
t e
n t
t e
m e
w m
i T
n i
a 0
t 0
n 1
o C
2 1
2 4
ig F
o t
i 0
0 0
0 0
o 0
g 0
0 0
7 e
5 1
4 4
3 3
3 s
3 mMv ey 0yoEe-i t
2 Ch'.Oo
(nag)X6;au3 i
hs
,g e=
1 3
\\
\\
\\
l 1
\\
\\
l E
5 e
1
\\
l I
\\\\
a
\\
n l
E
\\'
e
\\
1 l
?
\\
9 l
0 1
. 8 r
i a
9
)
s9 e
8 E
o i:
/l 1
=
i f j
~
o
?
w$
l 13 f.
f.
t, g,
k (r) X6;eu3 NUREG-1632 38
jIlllll
)l lil;\\
^3wv x8eCw v
9 v
'0 Q-t 1
t 1
/
/
)B
/
/
D k
/
/
l C
i
/
/
(
^
/
e
/
ca
/
f
/
/
u r
f s
f
/
e
/
d i
f s
/
/'
tuo
/
0 t
//
0 n
0 e
1 m
l i
I I
l
)
n i
c a
e tn s
o
(
c e
s s
m o
r c
i T
a r
0 e
f
- 0 s
1 n
ar t
y g
ren e
e i
W#e e
v t.
y y
t o
i t
. e p
t a
o l
r o
u e
v
~
0 m
o 1
o c
c u
C S
s
~
C c
P 4
e M
m 1
5 2
~
4 ig F
w v
9 v'
0 0
0 t
1 t
1 1
1 n3 x8$w 2C@9Ed
r _.- -
i i
(n;g)X5;su3 1
b b
t bg
,g
\\
I l
\\
s q
f e
\\ \\
g g
?
O w
\\ \\ \\
j a
2 U
l I
l
. 8 9
e E
I n
8 b
3 I
5 S
E
?>
E
.2:
1:
w
- Ei I
1 9
.3
)
9_
ci t
3 4
m it
.\\
b b
b b
'o (r) X6;au3 NUREG-1632 40
l l
)
l As noted in the discussion of the base case, the containment peak pressure and temperature for l
the MSLB occur before PCS actuation and hence are not affected. For times beyond about 1000 seconds, the sensitivity to area reduction is as shown in Figures 4.2.2-1 and 4.2.2-2, respectively.
The sensitivity of energy transfer split between dome and venical wall is similar to what was observed for the DECLG event, as shown in Figure 4.2.2-3.
I i
l 4.2.3 PCS Flow Rate (DECLG) l Perhaps the most important aspect of the PCS heat transfer is the PCS flow rate. Flow rates in the range from 100% to 10% of the base case were mn with CONTAIN in order to gauge the effect on containment thermal performance. Figures 4.2.3-1 and 4.2.3-2 show the pressure and temperature response. As can be seen, an insufficient supply of PCS water easily can lead to containment pressures be' yond the design value of 60 psia. A key effect of reduced flow rates is the increased chance of local surface dryout, which in tum has a large effect on the local surface heat transfer coefficient. If the PCS supply is reduced to such a degree that the resulting dryout is significant, containment pressure can be elevated easily up to and beyond design pressure.
To get a better understanding of the relationship between PCS flow rate and containment response, a more detailed study was made of the PCS. Figure 4.2.3-3 shows a schematic of the outer surface of the containment. The shaded areas (surfaces 18w,17w,16w,15aw,15bw, 14aw, and 14bw) were modeled as wettable surfaces in the CONTAIN input. The PCS flow is initiated on surface 18w (top of containment dome) and is allowed to flow down across each of the surfaces until it reaches the bottom surface (14bw). Gravity is the motive force for the downward flow and surface heat transfer determines water removal by evaporation. The balance between these two effects determines the progress of PCS flow across the surfaces. If sufficient PCS flow is available, all of the surfaces will be wetted. Otherwise, evaporation will remove enough water so that dryout will occur at some point in the downward flow.
One indicator of the preser e of PCS water on a surface is the surface temperature. The temperature of an initially dry surface drops sharply as soon as it is exposed to the PCS water.
By examining the temperature profile for each surface, it is possible to identify the specific time when the PCS water arrives on the surface. Figure 4.2.3-4 shows a family of temperature profiles for the DECLG base case, each temperature curve corresponding to a specific surface For convenience, the figure also includes the containment atmospheric temperature profile.
area.
Temperatures without any break-points belong to surfaces that remain dry throughout the event.
Surfaces that get wetted have temperature break-points signaling the arrival of PCS water. For example, temperatures for the top two surfaces (18w and 17w) drop almost immediately after PCS actuation. However, as indicated by the break-points for the rest of the surfaces, the PCS flow does not immediately reach the bottom surface. The wetting of the next s'urface in the sequcnce (16w) occurs at about 30 seconds after actuation. The bottom surface is wetted only after some 720 seconds.
l 41 NUREG-1632
_ ___ _J
l4 lll1llll1!slil!
I:
l1i gr;,O.uj.* a.
s 5
5 s
3 5
4 40 M
2 10 4
M 4
es 0
0 0
^1 0
e.
00 u.
0
^
^
1 sce
)
a s
A is o
p a
SC 0
P
^
0 6
M-
=
0 l
0 C
a
' 0
^
(
P 1
^
e
^
5
)
r c
u 0
s s
e e
^
s
+
n s
e e
e
^
(
r 4
=
p 1
s t
^
c e
n 4
P e
s m
m
^
=
e i
n s
T e
a i
r t
u 0
n s
0 o
^
1 se C
^
r p
1 n
2 g
2 is 4
e
^
D
.g iF 0
1
^
^
^
^
^
~
1 0
0 0
4
^Oa.v oo8,o *
.'5
- o.2 1
2C$o4g" rw
l (J ) a;n;o;adwai l
l 0
R R,
8 R
c l
.E i
g I
1 8 E
I I
?
m i
j M
E g
a B
g
^e 2 e3E e E Es P E 0
n8
.s5 5u 1
e4 e's ei?
=E o
A I
f 3
9 R
R (x,) a;nio;adwal l
43 NUREG-1632
3.O x[eCw
'0
'0 e
'00 1
3 10 1
0 1
- 0
- 0 1
o e
r r
o e
d
~
e e.
t t
t
~~
e.
t..
0 S
s 0
C c
a_ 0 P
p 0
s 1
2 z
9 e
5 2
)A-S C
P
-M 0
l 0
- 0 C
e 1
(
r e
c
) f cs en s a r
( t y
egmren iT e e 0
v i
0 n
t 1
a lu mu l
C e
m h
s o
3 c
d 2
lo 2
t 4
g i
0 F
- e. 1 1
y f
e, 8
"0 0
I 0
1 1
nO sgCw 1
ZC 95U g
lt
l 1
l l
l (o!sd) aansso>d n
n n
n n
n 3
3 4
4 K
I i
1!
l Q
5 N
l l
I C
l d
M l
E 6
I I
i y
. g s
5 8
y gl v
a E-a i
o.l N
.E Ec Sl
"~
'a 6
8 m
i O
t
- a. i an i n
o l
+
ei 4
4
.. -*o 4
D-
?
1 I
l
.f-El l.
1 a
a (Dd) LOOOO*0 x e;nsseJd 45 NUREG-1632
Il' j <j.i i1 j!Iili il
- l mL v ?3"EeoE,H A
4 4
4 4
4 8
7 6
3 4
4 4
4 4
4 4
40 7
6 4
3 2
1 2
2 2
2 2
2 2
1 1
1 i
1 00 0
1 O"
i sae.e for K
2m 5
e r
.o a
, M n
g i
1 sn e
f o
)
F s
e S
s C
P D
l C
r
(
e e
lo ru r
)e t
w d
a a
o e
r l
c e
f 8'
p a
e g
s m
le M (**
e e
d t
f tn o
U e
5 m
1 n
ia tnoC 2-3 2
4
.g iF 0
o 3
0 0
0 0
0 o
4 3
1 0
9 8
7 e
3 3
4 4
3 3
3 3
nMv eu3 Ou&E*-
i s
n ze 0 oU uc
l l
l top I
18w 18d i:
e
\\-
Eo
~0 17w 17d
\\xk dry
{16wl 16d f'
/
15bw!
15bd j
3 15aw 15ad g
W>
14bw 14bd 14aw.
14bd Fig. 4.2.3-3. Containment shell surface map 47 NUREG-1632
(4) a;n}o;adwal 2
a a
E
- 8 x
s x
s x
n n
n n
n n
n n
- Q-Oww a
l E
a
,1 s
'5 o
ka
.I 1
x 2
m d
u A
fl e
g
/~
C
)
H i !
E' T-@
)
t i
x_
2 4
s o
S 9
g g
E a
- =
a eL 8
e g
E U
ecn EOU
.T 9
Qv 6
II 1
9 f
f I
i 1
i R
R R
R R
R (3) a;nlo;adwal NUREG-1632 48
Similar temperature plots for other PCS flow rates are shown in Figures 4.2.3-5 through 4.2.3-8.
Collectively, these temperature plots are an indication of the relationship between extent of dryout and PCS flow rate. Figure 4.2.3-9 summarizes this in a plot of percent of area wetting versus time.
A visual representation of wetting profiles is shown in Figures 4.2.3-10 through 4.2.3-14. In each of these figures, the shaded portions of the containment surface indicate the time-dependent progress of PCS wetting. Also shown is a scale profile of the PCS source flow rate and the relative timing of each wetting stage with respect to it. It can be seen that as the PCS source flow is reduced, dryout occurs at increasingly higher elevations. At 25% of PCS design flow rate or less, the wetting is limited to the containment dome area.
The information presented in Figures 4.2.3-4 through 4.2.3-14 focuses on the starting times for PCS wetting, as signaled by surface temperature break-points. Another variable, namely PCS water runoff, can be used to estimate the time it takes to establish excess PCS flow for each wetted surface. Figures 4.2.3-15 through 4.2.3-18 show PCS runoff (from the lowest surface, 14aw) for PCS flow rates ranging from 100% to 60% of design flow rate. Also shown for reference is the corresponding PCS source flow rate profile. As in the case of the surface temperature analysis, in each figure it is possible to estimate the runoff start time by noting the -
disdnct appearance of runoff flow.
Figure 4.2.3-19 shows the start times for wetting and runoff as a function of PCS flow rate l
profile. At design flow rate, wetting of the bottom surface starts at about 1400 seconds. For about 1200 seconds, all of the PCS water arriving on the bottom surface is evaporated. Wetting becomes established fully at about 2600 seconds, when the rate ofincoming PCS water exceeds the evaporation rate and runoff begins.
As the source flow rate is reduced, the time to establish full or equilibrium coverage (i.e., supply rate matching evaporation rate) of the bottom surface increases. The reason for the a measurably l
longer time to establish full wetting is that a greater fraction of the PCS water is removed via l
. evaporation. Slightly lesser PCS source flow rates (e.g.,80% of design) may still be sufficient to J
l exceed the evaporation rate and establish full design-basis wetting of the bottom surface. Flow l
rates as low as 75% of design do not appear to impact containment pressure and temperature significantly, as shown in Figure 4.2.3-20. There may be sufficient capacity remaining for energy storage within the steel and concrete heat sinks to compensate for the reduced heat l
transfer across the shell (a more detailed discussion of this aspect is included later in this section, in reference to Figures 4.2.3-23 through 4.2.3-25). However, continued reduction of the PCS flow rate eventually leads to dryout and a consequent rise in the containment pressure and l
Ultimately, the time for establishing full design-basis water coverage approaches infinity. That
- is, the PCS source flow rate is insufficient to overcome the evaporation rate and the water never reaches the bottom surface. This appears to occur at about 50% of design flow rate.
49 NUREG-1632
lilll l
i)llljlilllJ C v 83 Ot ( e&
30 4
4 4
4 mi1 4
4 4
4 4
4 4
g 8
7 5
4 7
6 5
4 3
2 1
1 1
g 30 2
2 2
2 2
2 2
00
. 0 1
w e
4 1
1 e
ta r
wo l
lf n
00 g
0 i
N
\\
. 0 s
e 1
d N
No f
e o
w M
5 w
0 1
o 5
w e
n t
t5 n
a w
w g
1 S
i o
o s
C e
l t
d P
W 0
0
)
-(
5 0
F 0
e 5
1 S
7 W
)
C 1
c P
S e
CP s
D
(
l l
e C
(
m s
i e
T ru t
0 a
0 r
1 ep m
e t
o t
v n
e m
e)W" n
ia tn e.
oC 0
/.
5 1
3 to 2
e 4
d i
s g
n i
iF 1
0 0
0 0
0 0
a m
0 9
8 7
s 5
4 3
3 3
3 3
3 3
3 1
4
^$,
05
- E ?
ZC;cmOLo g
l 1!ll
,l1
Il l
l
{b gt30 eaE, 4
M 5
4 3
2 1
4 4
4 4
4 4
8" 4
4 4
4 s0 7
7 6
5 4
2 I
2 2
2 2
2 i
1 1
1 1
t00
~
~
00
- 1 h
t e
ta
, w r
b w
,4 4
o 1
3 l
l f
n 0
ig 0
s 0
0 e
^
d 1
fo e
lor 0
w 4
7 o
1 n
t
\\
/
a n
w g
S 8
is C
o 0
)
1 e
d P
f E
0 F
0 0
1 S
4
)C te P
c SC eD P
s -
(l l
C e(
mes irT u tar 0
- 0 ep 1
m
,c e
t t
n e
)
m e.
la c
i, lo n
i
(
a tnoC W
1 0
6
^
3
^
2 4
O g
i O
F 1
0 y
0 0
0 0
0 0
9 8
7 e
4 3
1 4
3 3
3 3
3 3
m6 W$.O W"E*H 1
u~
ZCM[T9_ h
ll ijlli1 i
I1jjIjIlll1!!l1l)
)1 m&v 820 c EeH 4
4 4
4 S'
4 4
4 4
4 4
4 4
u 7
s s
4 0
7 6
s 4
3 2
1 I
1 1
i 1
,0 2
2 1
2 2
2 2
2 00
~
. 0 1
e t
w a
1 r
N
. 0 6
w 1,
e o
b l
5 f
0 1
0 n
w 0
g o
5 i
1 s
1 e
e e
d b
r 4
f w
1 o
o w
n e
4 o
5 1
i.
2 ta ro S
x C
5 0
P w
2 0
8 1
. 0 t
)
e 1
F s
c
)
S e
c C
l e
P s
(
D e
l m
C
(
n s
- n ee u
c 0
t 0
a ff 1
r o
ru e
s p
d )l m
e
- o e
g,c t
g o
t
, (l ne mn ia tno 0
C
'r 1
e 7-3 2
4 to e
g d
iF isn i
~
0 0
0 6
0 M
0 3
7 6
5 8
1 3
3 3
3 3
4 2v 830 OEeF WmO M
illl;
,i l
l
l
!l!
l lIll:Jl)lij l
lIl
^L,v 0 3{EeH
-a 4
4 4
4 4
4 4
4 4
4 4
4 4
m0 7
6 5
4 3
2 1
9 8
7 6
5 4
0 2
2 2
2 2
2 2
1 1
1 1
1 1
0
- 0 0
- 1 l
t e
a r
J w
0 o
l 0
f 0
e
- 0 n
1 l
ig I'
e o
s r
e I
8 1
w d
lo fo f
n g
i e
s 0
6 e
1 d
1
=
f t
6 o
a 5
E S
1 0
0 1
0 C
w o
' 0 P
5 f
o 1
1
)
e S
b C
) F 4
P c
eS 1
l w
s C e
( P 4
/
- 0 1
eD -
ml C
iT
(
s 0
e r
u 1
tarep m
)
e W
t t
"(
ne mn i
e.
a t
e.
n 0
4 o
1 C
8 fo e
3 id 2
sn 4
I
.g i
F 1
0 0
g 0
0 0
0 0
0 0
8 7
6 5
4 3
1 4
4 3
3 3
3 3
3 Mv g$ 0 b
ZC m2 h
.I_
i i
I l
- \\
ss 5l eE i
a a g,
i*
s o
,o o
1 gm g
.5 if 8
1 c.
i n.
i
' m o o H
l O N -
o n.
in e
lu?
e i
u 0
I g
6 I
i__________.
u.
L_
m 8
6 3
e ec.
O v
E y
P:
0
=
8 8
E D
.8 o-4 m
etx
-2 I
8 O
8 2
8 8
N oa;o pa))am alq!ssod 1010) lo %
~
54 NUREG-1632
0 - -
I 0,
1 0
0 1
0 0
0, 0
1 0
0 3
0, 8
1 3, -
1 e
ta 6
8 r
9 w
o 0
l 0
0 f
8 0
n 4
1 6 - -
g 7
is 1
9
_~
e 6 -
~
d 0
6 6
0 0
e 0
1 m -
id 1
e s
l e
o a
t c
it d
a r
r e
v W
O f
0 1
)s d
)
=n s
o d
tc n
e
= o s
c
(
t es
(
w Zdb9
0 j
0, 0
0 1
0 1
2 0
7 0
8, I
I 0,
4 1
0 1
8 3
1,
~ -
0 1
0 0
5
~
7 0
0 i
0, e
1 ta r
w o
l f
n 0
i0 g
1 2
i 9
7 s
e d
0 I
6 0
l 6
5 0
e I
1 d
e is m
la o
c it d
re v
0 - _
I 1
e ta r
)
)
s s
w d
=
d n
n o
o o
t f
c
= c e
e s
t s
(
(
5M9 Dw i
0 0
r 0,
0 1
I 0
0 1
c es
/g k
1 2
0 0
I 0,
0 1
0 e
0 a
I 0,
t 1
r w
o
~
l 8
f 1
9 n
g e
l 0
i e
0 s
s
/
1 o
e k
5 d
0 5
Z 7
8 0
1 6
Z 0
6 l
4 e
d l
0 1
e is m
l o
a d
ic tre v
e t
a Z
i 1 r
0 W
O f
)s
)s d
d n
n o
o
= c c
e
= e s
t
(
s t
(
- Ch9
'2
0 0
0, i
0, 0
0 0
0 1
c 1
5 1
9 k
3 1
0 0
i 0,
0 1
0 0
e i
0, t
1 a
r w
o l
f c
n e
s 0
/
g 9
i k
i0 s
4 1
e 3
d 5
8 6
0 5
_ cl_
6 1
6 2
e I
0 is d
1 l
e a
t itc a
r r
e Sw v
Co Pf l
I 1
0
)s d
)
n s
d oc n
=
e o
= c t
s
(
es t
(
zCh 9-@w a
l l
0 0
9 e\\
l 0,
0 7
0 1
e 0
s
/
g 1
k 1
5 0
0 0
l 0,
0 1
0 0
e l
0, ta 1
r w
1 5
o 9
lf n
c e
0 g
/g l
0 s
i k
s 4
1 1
e d
0 6
~
0 6
l 1
d l
0 e
i 1
s l
a i
e c
d tr t
e v
a r
w o
f 0
l 1
)
)
s s
d d
n n
o o
= c c
e
= e t
s s
(
t
(
ZCb9
il!
i j
i 4!!
,1 ll{
1l l1
- illl, 0000
- 0
- 1
^
/
[
j
)
0 F-00 S
0 C
^
I
^
1 P
D l
C
(
f f
e o
ta c
r u
w r
o 0
H 0
' 0 n
^
^
1 g
i
^
s
)
e C
d G
S
(
00 e
1 w
m ta o
I l
f f
I f
e on c
0 u
- 0 u
r 1
r o
S s
C P
^
5
^
1
-3 2
4
.g iF 0
1
^
1 4
3 2
1 0
o e
7 s
5 4
3 2
1 0
1 1
1 1
1 U ENmb 3 0' s'o UOE n
na 1
CbO' et 8
o lll(
l1
1lli il' 1'
!4 0000 0
. 1 l
0 0
. 0
)
0 F
1 S
C f
P f
o n
D u
r l
C
(
e tar 0
0 w
0 1
o lf
)
n c
g e
i s s
e
(
d e
w m7 5
o i
l T
f ta e
f c
0 f
r u
0 o
o 1
n m
s SC P
6 1
3 2
4 0
ig 1
^
F
^
i 2
1 O
s 8
7 8
5 4
3 2
1 0
1 1
nue*N&x.v 20' 5mOg uOQ
)
e-ZC 4Ou
!1 l
l
,i l1 i
il l]
1l1){
0000
. 0 1
t
_0 00 0
)
F 1
-SC f
P fo n
D u
r l
C
(
e ta 0
r 0
0 w
1 o
lf
)c n
N e
g i
s s
(
e d
e m
0 w
i 7
lo T
t f
a f
e 0
f c
1 r
0 o
u n
o m
s SC P
7 1
-3 2
4 0
g
.. 1 iF 1
2 1
0 a
7 6
5 4
3 2
1 1
1 1
v
)
oWiNc 5' E E uOQ n
ZCCm95n 8
l l
.h r
. l g
,m a
E, e
a' O
O E
I n
o e
no m
m v
o
- C3 I
8 e
g 3
e i
b
=
6 e!<
- 2 ea ir l
l 0
I I
E g
a g
e t
l c.'
=
e
=
=
~
.n (cas/6 ) alo; ssow SDd M
63 NUREG-1632 i
I
\\ \\!l\\1 lIll\\
4!
~
m
)
1 i
o
. s
- /
o
. a
)F-S 5
C P
D l
e C
l
(
4o s
r e
m w
i t
o t
l r
f a
ts yn f
I I
l 1
I l
l l
I l
1 I
1 g
f i
o s
n e
u d
/rg f
n o
i t
te w
laco l
E S
C P
9 1
-3 M
2 4
.g iF 0
1 0
0 o
cco s
2* v eE t$ =oC5 Ei9cO t
n
0 0
0 0
0 0
5 0
5 0
5 4
4 3
3 00 1
09 08 07 06 e
tor w
0 o
lf 5
SCP 0 %
4 0
3 02 0
1 0
0 0
0 0
0 0
0 0
7 6
5 4
3 2
0 1
gb e ogod x
.'5y,*
+ c.EC5" coo 8
$mo*
0 0
0 0
0 0
5 0
5 0
5 4
4 3
3 0 0
1 0
e.
9 m,
s2 0
t 8
J l
l l
j I '
I
_ 0 7
)
tuo ye v
. 0
. 6
.n l
l l
i a
{
b nio n.
n 0 (
5 i
t 0
O 4
l f
SC P
0 O
3 O
0 O
=
2 0
C
=
1 0
0 0
0 0
0 0
0 0
0 7
6 5
4 3
2 1
n5 =E." E 2 g s1
~
zCg9sww R
5 0
5 4
4 3
3 0
0 1
0 9
oo o.o e
0 i.
8 e
i ooo.
,i o
o s
o 2-i i
07
)
0 w6 0
5 m
04 I
l 1
l l
l I
i 03 0
_ 2 w
7 1
e 0
co 1
i I
I l
I l
l l
l I
f ws 0
0 0
0 0
0 0
0 0
0 0
0 0
9 8
7 6
5 4
3 2
0 1
1 g v g c. m $ e u H4 e"
zcEo
IlI i>
\\
{1 1!IlI
- l I!
I 3mvxD'Ew gm e
e q'
'o g
t i
i i
/
'O s, h on sp o
, c p
i l
le
/
h
/ '#g s
F) s *.
SC C
P
-D-n l
o
- e. c C
i
(
r e
)
f c
s e
na s
r
(
t y
e gr m
e i
n T
e e
o v
i n c t
i a
lu m
u C
3 l
2-3 E
2 e
hs 4
.g o
l i
n F
i r
i t
f f
g "o
o O
i i
g t
i i
m Cw zcg9 h
l i
Il m2ovxwbeCw o
'0 Y
e0 1
1 1
1 S0 0
. 0 0
1
~
sc 0
e 0
e
. 0 e
t t
l0 0
e r
w 1
rc o
l n
f o
S C
c P
E F) 0 1
S C
P D
0 0
1 0
C 1
(
r
) e cf s
e n s a
( r t
e y g
mr e
iT ne e
0 v
0 i
1 t
a l
u m
l uC 4
2 3
2 4
l 0
g 1
iF 1
8 f
0 0
0 1
i 1
1 1
m3 X&uoCw me ZCh2 $t o
!l l
Iilll(
il
- l l
,58xE.]
"0
'0
'0 0
00 10 1
1 1
1 00
- 0
-' 1 e
to r
/
wo 0
l e
f 0
to S
- 0 0
r C
1 w
P l
lo
,5 ee f
0 t
S 5
s C
P
)
E F-0 S
1 C
P-D-0 l
0 C
' 0 1
(
re
)fcs ena sr (t
y eg r men iT eev 0
i
' 0 t
1 a
lu muC 52 3
2 4
.g 0
i le 1
F e
ts
^
r r
1
'0 "0
"0 "0
'0 0
1 1
1 1
1 1
n x E' E w ZC PEY 8
1 l
Similar to the analysis of the shell surface temperature described earlier, the temperature drop across the shell also can be used to interpret the presence of PCS water on a surface. For example, Figures 4.2.3-21 and 4.2.3-22 show the approximate onset of dryout for the lowest surface (14aw) and a surface on the dome portion of the shell (17w). In each figure the temperature drop (AT) across the shell is plotted as a function of PCS flow rate for three different times after PCS actuation (also shown, for reference, is the containment atmosphere temperature).
j The curve for t = 1000 seconds in Figure 4.2.3-21 is seen to be relatively constant throughout the entire PCS flow-rate range. This indicates that, independently of PCS flow rate, the PCS water is not able to reach this far down the shell surface at 1000 seconds. Recalling that at 1000 seconds the containment is under reflood conditions, it is apparent that there is sufficient energy transfer across the shell to evaporate all of the PCS water before it arrives at this surface. Conversely, at 10,000 seconds and beyond, the source energy input is diminished to the extent that the PCS flow rate is sufficient to overcome the evaporation rate and achieve wetting all the way to the bottom
)
surface. This condition is indicated by the sharp reduction in the temperature difference between
{
inside and outside shell surfaces for flow rates above 60%. A similar set of curves is shown in Figure 4.2.3-22 for a surface located on the dome ponion of the shell (17w). Since this surface is relatively close to the PCS source, except for extremely low flow rates (i.e.,10% or less), there is no corresponding permanent dry zone. That is, wet conditions prevail for almost the entire flow-I rate range.
This relationship between PCS flow rate and the capacity to achieve fully wetted conditions for maximum heat transfer across the shell illustrates the importance of the " excess" capacity inherent in the PCS source design flow rate. Source flow rates just below 50% are likely to lead to the onset of permanent dryout staning near the bottom of the venical shell wall. Continued i
reduction in PCS source flow rate would extend the dryout zones to increasingly higher elevations, leading to rising containment pressures and temperatures.
Figure 4.2.3-23 shows base case,50%, and 10% of design flow-rate profiles for cumulative energy transfer across the containment shell. The curves for the reduced flow rates illustrate the expected result that, as PCS flow rate is reduced, the total amount of energy transferred through the shell to the outside environment also is reduced. However, as inferred earlier in this section, j
reduced energy transfer to the outside does not mean that all ofit goes toward raising containment pressure and temperature. As shown in Figures 4.2.3-24 and 4.2.3-25, some of the energy is picked up by internal heat sinks (concrete and steel, respectively). Nevenheless, the amount ofincreased heat storage is insufficient to prevent pressure and temperature rise inside the containment (see Figures 4.2.3-1 and 4.2.3-2).
It is noteworthy that for the 50% case in Figure 4.2.3-23, the total energy transfer beyond the reflood peak (i.e., > 25,000 seconds) begins to approach the base case. What this indicates is that at this point the source energy input rate is low enough so that the PCS flow rate, even at 50% of design capacity, is sufficient to prevent dryout.
71 NUREG-1632
\\
Another interesting feature associated with PCS flow-rate reduction is the observed energy transfer shift between the vertical and Jame ponions of the containment shell. As noted before, Figure 4.2.3-23 shows the energy transfer across the entire shell. Energy transfer across the i
venical wall follows a similar trend, as indicated in Figure 4.2.3-26. However, at 50% PCS flow rate, the dome portion of the shell experiences a higher rate of energy removal, as shown in Figure 4.2.3-27. This is because the dome continues to remain fully wetted (see Figure 4.2.3-11) for the 50% case. Reduced heat transfer across the vertical wall (primarily due to dryout) raises the containment temperature (and hence the temperature difference across the dome portion of the shell), to such a degree that a higher energy-removal rate occurs at the dome. Continued reduction of PCS flow rate, however, leads to elevated dryouts (see Figure 4.2.3-16 for 10% PCS flow rate) and a net drop in the energy transferred across the dome.
As indicated in the discussion on PCS wetted area sensitivity, the MSLB pressure and temperature peaks occur beforo PCS actuation and, hence, are not affected its presence. The degree to which post-peak pressure and temperature are impacted by PCS flow follows a similar trend to what was observed for the DECLG case. Figures 4.2.4-1 and 4.2.4 2 show the pressure and temperature profiles for base case,50%, and 10% PCS flow rates. Figures 4.2.4-3 through 4.2.4-5 illustrate energy transfer through the shell, as well as energy storage within the internal concrete and steel heat sinks. The split between vertical wall and dome energy transfer is shown in Figures 4.2.4-6 and 4.2.4-7.
4.2.5 PCS Water Temperature (DECLG)
The sensitivity of containment pressure and temperature to the initial PCS water temperature is relatively minor, as seen in Figures 4.2.5-1 and 4.2.5-2, respectively. This is to be expected, since the bulk of energy removal by the PCS is through evaporation. The only significant effect of the PCS water temperature is the timing of local dryout. For the base case, for example, the bottom surface of the vertical shell wall (14aw) experiences a substantial delay in PCS water-arrival time when compared to that for 273 K PCS water. The delay is about 1500 seconds, as can be seen in Figure 4.2.5-3. The reason for this is that the colder water does not reach the boiling point nearly as soon. Hence, dryout is delayed and the vertical descent of PCS wetting proceeds at a quicker pace for the 273 K case.
4.2.6 PCS Water Temperature (MSLB)
For the MSLB, similar to the previous PCS sensitivity cases (i.e., wetted area, flow rate), the PCS water temperature has a minor effect on containment pressure and temperature, as indicated by Figures 4.2.6-1 and 4.2.6-2.
4.2.7 PCS Start Time (DECLG)
The significance of PCS actuation time is illustrated in Figures 4.2.7-1 (pressure) and 4.2.7-2 NUREG-1632 72
i!ilIi!'I!lil)I)il llll!i!
iii
!I m3iv A?eCW H
9
'0 e
e
- g 1
1 f
t n
7 9
/
f l
l a
w la c
i t
/
b r
ev
/
s F) a g
-S 0
C 08 P
-D
)
c l
e C
(
s r
(
e f
e s
n m
ar i
t T
yg 0
r 0
e
^
1 n
e
^
ev i
ta l
u m
u l
C 6
2 n
e 0
3 m
1 2
s 4
c i
t g
i F
f "o
9
'o 0
o 1
1 t
1 t
t mD A@eCw w"
gCpmoL@w
il l
1 8 x E' E a L
9 f0 1
1 i0 1
0 7
_ 0
. 0 1
p e
to r
0 e
w 0
. 0 m
_N lo 0
o e
f 1
S d
f C
P
)F
/
E e
D t
S a
I
/
C r
w P
o l
D f
S C
l P
C m
(
0 5
r 5
0 0
. 0 fe 1
sn
)
a c
r t
e y
s g
(
re e
ne m
e i
v T
i ta l
0 u
. 0 m
1 uC 72
-3 2
~
4
.g e
i m
F o
0 d
, 1 1
f v
g 8
"0 t
0 i
1 1
x E' E "
ZC e5o t
'l 1
1 l
(o!sd) a;nsseJd n
n n
n n
n 3
a 4
m a
1 l
4 I
I i
a i
ex
- g I
1 s
?j
~
Q Q
o)
E h
8 al i2 I
U o
1 u
~
8 7
t E
E v
c.
4l e
5 u
E E
e 1=
-Q B
El 8
5 u
el 4
i ei c8
=
LE I
- 2 I
l l.
?
a 9
(Dd) L0000'0 x a;nsseJd 75 NUREG-1632
r l
(3,) aanlo;edwei l
R E
6 Eo.l 5
/
.5
_ H e
2
- e E
Y E
?
m E
2 E
S I
e o
5 9
Ek e
a E
s t=
u Ee 8
3 8
U Y
1 9ci
~e w
- 9 E
I(x,) a;nto;edwal NUREG-1632 76
lI I
1l li lj!illll1!l l
i l1 ill l
g.mV NueCw 9
0
'0
'00 1
1 1
1 100
- 0
- 0
- 1 to r
w on S
0 C
0 5
- 0 P
0 1
0 t
l 5
o l
r e
h o
s n
)
s F
c P
S s
C s
o P
e t
o
-M c
n 0
e 0
l e
0 C
b 1
(
)
r c
e f
e s
s n
(
a r
t e
y m
gr i
e T
n e
0 e
0 v
1 i
ta l
u m
u C
3
-4 2
4 l
0 g
1 i
F
^
l
^
1 f
'0 0
0
'0 I
1 1
1 1
9v >@ ecw vu Zhy'.Cw
4 l,lil\\lIl
.ll1
- 4iI;i41lli IiI l
tli lI l
ovxouecw o
d
'O 9
t r0 i0 1
0
_ 0 0
. 1 ed
]
y eo n
0 S
0 0
C 0
P 1
e s
te 0
r s
cn o
c
- )F
-S e
C i
m P-w 0
M 0
o
. 0 h
1 l
SC C
P
)
(
c r
s e
e 0
s f
1 s
(
na e
r t
m yg iT ren 3
e 0
1 ev i
ta lu muC 4.-4 1
2 4
0 g
iF l
1 g
e "0
0 0
l
~1 1
1 vxo'ecw zcgmOLO" a.
ll l-1ill
)lllli1;l l
l l
l ili l1 IiiI I
ll1lIi!.I lll1
^]Ov XGbCW f
'o
'0
'0 00 i
t 1
1 100
- 0 0
1
%_+.g e
te r
w 0
on 00 0
S
^
C
^
1 P
^
l E
e D
e e
t I
s s
o c
F) es a
e S
b to C
t P
w lo M
f 0
S' 0
C 0
l
^
C 1
P
^
^
(
E0
)
r
^
5 c
e f
e s
s n
(
a r
t e
y m
gr i
e T
ne 0
e 0
v 1
i ta lu muC 5-4 l
2 4
M 0
g 1
i
^
F
^
l r
r r
1 "0
0 0
0 1
1 1
1 1
1 n7v h0weCW 4
3 4
ZC5O'. h l)
lllii Il)I!
l-l l
3cv xmLDCa n
L f
- '0 e,0o i
1 I
1 0
9 0
s 0
1 e
te r
r o
n s
0 C
0 P
0 l
0 l
s e
t 1
a e
a w
Ns r
s l
o a
i c
l i
N S
t C
r P
ev 50
- )
1 F-S 0
C 0
P 0
1
=
M
)
ce C
(
s r
(
e f
e sn m
a r
i t
T yg o
r o
e t
ne ev i
ta lu m
l uC 6
~
4 o
2 i
4
.g iF l
r r
i "o
f v
e c
o I
i i
t i
m3 gu*Cla t
ZC0no.En oO o
?t 1lt l,llt ll
\\l l\\
Il1 ll
)l1 ii) l 3.m" XFeCW f
'0
'o fo i
1 f
ioo
. o
. o
. t ed r
e w
lo o
0 r
H 0
h hs 0
S P
- 0 C
1 e
S s
m C
o o
P d
so
- )
t F
SC P-M 0
0 l
- 0 C
1
(
r
)
e c
f e
sn s
a
(
r t
e yg m
re iT ne e
oo v
i t
ta lu muC 7-4 2
4 e
g m
. o iF i
o s
r r
'o "0
"O 80 e
1 t
t 1
im3,c'$w n
gx@' 8o
1lll ii l>
^9.mQve'3mm8 Q
s 5
5 5
5 5
4 40 M
2 10 e
4 4
s 5
4 0
0
'0 1
x
)e s
o mc es e
0 i
0 s D 0
,c 0
1 o_
)
T-is
[p S
C P
0 K
6, D
3
=,_
7 2
0 l
0 o
S 0
C 1
(
PL S
e C
)
r P
c u
0 e
s
+
s s
e t
^
t
(
rp 1
t e
n 4_
m m
e
=.
iT n
e i
r a
~ 0 n
t 0
=
o 1
C
^
r P
1 a_
5 9s 2
i s
4 E
.g iF 0
1 s!
1 0
0 0
4 3
1 2a.
OOOO.O x eu$me..
s zCh2aw M
l_
t (ll' tjl 1Illl,j1 il Q
o$ ou*QEoH 4
4 4
4 4
4 4
4 4
e 0
9 s
7' 4
7 6
5 4
3 2
y 4
3" 1
5 2
2 2
2 2
2 2
2 1
t 1
g 14 M,_
,t
)e K s e 2 c 2
e 3 s a S b(
SC 0
P 00 3
0 1
)T-SC P-D 0
0 l
t 0
C 1
(
)
e c
ru e
t s
ar
(
ep e
m m
e t
iT tne 0
m I
0 n
1 ia tnoC 2
5 2
4
.g i
I 0
F 1
~
1 0
j 0
0 o
0 0
1 g
9 7
s 5
4 4
3 3
s 3
3 3
m 83 OuQEgH t
=W Z
9aO l
lll
l1!iil}f;
\\
Ill)l i
)-
{!!
l
!l1lil{llIIll!l
~
g*v,u 3 o' E [ s d
4 4
4 4
s 4
4 4
u 4
4 4
4 o
s s
7 6
3 2
1 7
s 9
1 2
2 2
2 2
2 2
2 i
i1 1
i
).
n..c me
^
mo
)
en T-(
o m
b S
o C
t P
-D l
C
(
e ru tarep b
m x
e i
t 3
w 72
)c a
e 4
1 m
s
(
e ca e
fr m
u i
s T
re tu b
o i
t
~
ne mn i
^
a tn
~
oC 3-5 s
2 i
4
^
.g iF i
M m
=
o.
8 o
0 0
s 1
0 3
s 4
4 L
^E ege8eCEO fxmP5g T
(o!sd) a;nssaJd n
n n
n n
n 3
2 4
A a
's
.g l
l
~
l 2ia i
j~
i
\\.,1 I
"~
- ?
6 oil E
(
o 2
.I 2
o
- I E
l a.
1 e
g m
a 7l E
a S
4l e
s E
U e
1=
l
- 8 3
u i
c 4
i
.P N
21 d
t eo LL.
I 2
l I;
I.
L 9,
4 n
o o.
9 (Dd) LOOOO*0 x a;nsse>d 85 NUREG-1632
(3,) a;nlo;sdwal 2
l I
n
. i v
n P
E AO cr2 3
_ g e
o
?
e 8.
O 5
e E
8 i:
Ee
. s
'5 g
U ei4 l
N<
.eh
- 9 f
9 5
9 R
R (y,) a;nlo;edwal NUREG-1632 86
i'
,j iii l
!IjlllIllIlJliIl Ia m9.ma u3my'k 5
5 5
5 5
5 4
4 4
4 6
5 M
M 3
2 i
.=
=
oo o
0 0
i 0
0 1
e e_ m
)
ga M
s
^
c T-
=
=
S
)
C s
M,,
. eo P
c
=_
e D
s b
a o
b l
C P
(
L i
(
5
)
e 0
c r
o c
u
+
s e
s t
o s
s t
(
e i
t.
9 pr 4
s t
e n
m e
m i
e S
T n
=r C
ia P
n t
o n
i
=
o r
C
^
a_
1 e
-7 w_
2 4
.g i
F o
i i
0 o.
o.
4 3
2 mOLv O OO'O x
- onne a ec C
gxmO'.Cw
ll11)ll l
!I!iI li m1 v og O.cg. -
a
.4 I
d 4
4 4
4 e
4 4
4 4
4 4
3 2
1 s
a 7
s 5
4 n
4s 1
i 1
1 2
2
'2 2
2 2
,i1
~
~ i
.os ooo i
0 o
. 0
. 0 1
)M SC T-P S
C ee P
s 0
D 8 )e 6 s 0
l o
o C
Gc
. 0
(
M s s 1
e t
e r
o r
e (b
)
u c
ta S
e r
C s
e P
p
(
m e
e t
m e
t e
i n
s T
e 0
m 1
n n
9
._ o ia i
s t
t n
ro o
t C
s S
C 2
P
-7 2
4
^
.g iF
. o t
^
^
i o
0 m
0 0
0 o
e 5
1 0
0 8
a 3
4 4
3 3
mMv * :'~E# EeH s
y95w m*
(temperature). The delay of PCS actuation to 1000 seconds (base-case actuation is at 660 seconds) produces a significantly higher maximum pressure and temperature in the reflood peak, I
as seen in Figure 4.2.7-1. Conversely, earlier PCS actuation leads to a drop in maximum pressure and temperature, as seen in Figure 4.2.7-2. Specifically, the maximum pressure change is about 12 KPa ( l.7 psi) when changing PCS actuation time in either direction from the base-cas. Similarly, the maximum temperature change is about 1.7 K ( 3 *F).
4.2.8 PCS Start Time (MS.LB) l For the MSLB, the effect of PCS actuation time on containment pressure and temperature is not significant, especially in reference to the peak values. Figures 4.2.8-1 and 4.2.8-2 show the pressure and temperature response curves for PCS actuation times of 10, 660 (base case), and 1000 seconds. At first glance, however, one may question why the peak values are unaffected by PCS actuation at 10 seconds. Since the actuation time is well before the pressure peak (at ~335
. seconds), one may expect that there would be opportunity for the PCS to reduce the peak. For l
example, Figure 4.2.8-3 shows the outside surface heat flux for two containment shell surfaces l
(14aw and 16w). It can be seen that for the base case the PCS-enhanced heat flux occurs well past the peak pressure. For the 10-second actuation time, however, enhanced heat transfer is l
evident before the time when peak pressure is reached.
The reason that PCS start time has little or no effect on the peak value can be inferred to be related to the containment internal heat transfer conditions. For the MSLB, the containment temperature is in the superheat range up to the time of maximum pressure (see Figure 4.2.2-10).
Under superheat conditions, the iremal heat transfer on the inside shell surface is rate limited by l
convective heat transfer. Hence, heat transfer conditions on the outside surface have little effect on removing energy from inside the containment. Indeed, the absence of sensitivity effects in the previously described MSLB analyses of PCS area, flow rate, and water temperature is, to a large degree, due to the rate-limited heat transfer inside the containment, brought about by superheat i
steam conditions.
4.2.9 Heat Sink Area (DECLG)
The importance of intemal heat sinks in affecting containment pressure and temperature is l
illustrated in Figures 4.2.9-1 and 4.2.9-2. Two specific aspects are c'<ident as one reduces the heat transfer area of the heat sinks. As can be seen in Figure 4.2.9-1, a 50% heat sink area reduction leads to a containment pressure exceeding the design limit of 60 psia. The containment l
temperature maximum also increases significantly, as shown in Figure 4.2.9-2. In addition, the
[
. results in both figures indicate that these effects become noticeable relatively early in the DECLG I
event (i.e., within about 10 seconds).
Figures 4.2.9-3 through 4.2.9-5 show the energy storage partitioning among the internal heat j
sinks as well as energy transfer to the shell. It can be seen that as the internal heat sink area is reduced, the amount of energy storage is proportionally less. Also, there is a small increase of 1
89 NUREG-1632 j
(ojsd) a;nssedd e
o e
n n
9 s
a s
E a
I I
l 1
I 1
i
- l l
u j
B E
el e
e m
O s
i 1
- E B
e n.1 7
8 X
N 1
+
-l v
-a y
4 e
o el 8
E
.s b
E E
=
8 o
M I
h 1
5 sw El g
l
- 2 I
I l,
3 a
R S
(Dd) L0000'0 x a;nsse>d NUREG-1632 90
i I
i l.
1 i
(3,) a;n)o;edwal 0
E R
8 R
B8 6
m
.g i
- h y
g E
W m
E 8
s
)
E i
e 2
I
- 0 e
9 8.
E M
O o
W C
E U
1=
.e 8
s g
. 8 8
u 4
m N<
eb iE
)
- o l
?
9 R
R l
(>I,) a;n)o;edwal 91 NUREG-1632
1 l
ili lli
- 1 jll1 1
l 0000
^ 0
^ 1 4
i 0
i 0
0
. 0 1
)M T-S C
[e P
[$.
0 0
M 0
1 l
)
C h
c
(
e r
s e
f
(
sn e
ar m
t i
y T
gr e
[e 0
n 0
e 1
e 6
v i
p ta lu muC 3
-8 2
0 4
1
.g iF r
1 d
e O
0 I
s i
1
'q" xb.CJ A
L zcg9s eO
(
<l1j lll
!1ll, l;
1l lll l
1 l
l
i-(o!sd) aanssaid i
n n
n n
n n
3 2
3 4
x l
l l
t
.l I
s l
1 i
Iy l
l N
l
..l 1
o 5,
i i
Q l
9 B
o.!
g 8
7 R
+
C3 "5.
i Z.l m
- l e
g rxi B
/
E E
/
- 8 o
u u
5 l
eb II I
e l
l l.
~
e 9
9 9
m n
(od) L0000'0 x a;nsss>d 1
i 93 NUREG-1632
1
){l\\,
{li!!I!
i; C.v eu3~2eoEeF 4
4 4
4 4
4 4
4 e
4 4
4 3M 4
3 2
1 0
s s
7 s
5 4
M 4
4 1
2 2
2 2
2 2
2 2
i i
1 i
1 1
7 s
- 1 00 s
. 0 i
0 1
)A
-S I
I
-D lC
(
e i
h' ru i
t ar
)
e e
m c
p s
e
(
t t
e ne m
m s
o i
c n
T a
i e
t o
n n
oC da h
2-m 9
os 2
4
.g iF 0
1 0
o 0
0 0
o M.
a 0
s 5
4 5
8 e
1 3
3 3
3 3
3 4
mxv cu3 E co E,r c. oleo 2
\\ll\\ fit!l'
.l{
tt i
l
(cas/r) ago; Jajsuo;; X6;au3 2
e
'e h
i
'el
~.
i 1
11 l
i
]
5
_o 5
I h
5 m9 9
\\
9 E
e
\\
/!
7 Y
e e
\\
/
/
3 I
e
\\
/
/
E E
I P-l
\\
/ /
- s i
y 2
\\
/
/
u
\\
./ $
+
m N
/
d 1
\\
l f
l
{
N
~
x
\\
k N
N.
N 1
.\\.
'a (r) X6;au3 95 NUREG-1632 i
I (oes/c) a;o; ;ejsuo;; X5;eu3 e
b e
i b
e g
l 2
E 8
8 ex h
\\
q
\\
I B
\\
i 4
E E
]
5 8
w
~
s 3
a ea u
f a
ei e*
- e e
- a..
t..
t..
t..
t-k k
k k
- g k
g (r) X6;eu3 NUREG-1632 96
l l
l l
(cas/c) ago; ;ajsuo;; X6;au3 9
p o
o o
b b
oe
, g 4
l
=
\\s
?"
\\a\\
a<
=
Q O
I r
a
?
5 I
eO eo E
l=
?
i
=
8 4
5$
s a
11 H
5 9
m
's 6v ea 1
it 2
i s
s
_e o
o a
(p) X6;au3 i
97 NUREG-1632 i_________________
m
energy entering the shell. However, the net result is higher containment pressure and temperature, as indicated earlier.
I 4.2.10 HeatSink Area (MSLB)
For the MSLB, heat sink area reduction also has a significant effect on containment pressure and temperature. A 50% heat sink area reduction (see Figure 4.2.10-1) increases the pressure maximum from 366 KPa (53.2 psia) to 408 KPa (59.3 psia, nearly the design pressure limit), a pressure increase of 42 KPa (about 6 psi). Similarly, containment temperature maximum
- (Figure 4.2.10-2) increases from 441.2 K (334.3 *F) to 450.0 K (350.2 *F), a difference of 8.8 K (15.8 'F).
Figures 4.2.10-3 through 4.2.10-5 show the energy partitioning among internal heat sinks and the containment shell. The trends are similar to those for the DECLG. Energy storage for internal heat sinks is reduced, while a slight increase is seen for the shell..
4.2.11 Liner-Concrete Air Gap (DECLG)
Increasing the liner-concrete air gap for internal heat sinks can have a significant effect on containment pressure and temperature, as seen in Figures 4.2.11-1 and 4.2.11-2, respectively.
The air gap was modeled as a purely conducting heat transfer path, in series with the concrete liner and any coatings that are present on a liner surface. The base-case gap of 5 mils was increased to 40 and 100 mils. The peak pressure is increased by about 30 KPa (4.3 psi) in going from base-case to 100-mil air gap. The corresponding increase in maximum temperature is about 3.7 K (7.4 *F). The difference in the peak pressure and temperature between a 40-mil and a 100-mil air gap indicates the further increases in air gap would produce diminishing changes in pressure and temperature. However, it must be kept in mind that the air-gap model is limited to pure conduction. Hence, for larger air gaps the model neglects any convective heat transfer that may be present within the air gap. Inclusion of ccavection would ' end to diminish the effect of t
the air gap shown in Figures 4.2.11-1 and 4.2.11-2.
4.2.12 Liner-Concrete Air Gap (MSLB)
Containment sensitivity to liner gap variation was not analyzed for the MSLB.
4.2.13 Initial Relative Humidity (DECLG)
Figures 4.2.13 1 and 4.2.13-2 show the effect of initial containment atmosphere relative humidity on pressure and temperature. The base case has no initial water vapor in the containment atmosphere (zero relative humidity). As seen in Figure 4.2.13-1, assuming an initial relative humidity of 100% (saturation) leads to lower pressure throughout the DECLG event. The containment temperature for the 100% relative humidity case (Figure 4.2.13-2) shows an increase of about 10 *F initially (at about 1 second), diminishing throughout the initial blowdown and a portion of the reflood phase. At about 5800 seconds there is a cross-over, and for the rest of NUREG-1632 98
(ojsd) a;nsss>d t
i i
n n
n n
n n
s a
4 4
a
- s
- g i
l r
i l
i I
/
-l
- i l
/
~
3l 1
/
l
]
e
/
=
8 N
/
s al
/
. l u
c, o.l
/
g q
Y
=
.4ls' l
O E
4l N
n o
5 u
2 E
E
\\
1:
i l
\\
- 8 8
l u
\\
N k
El N
i n
N
.99 I
. e I
I l,
e, e
e e
n 3
(Dd) L0000'0 x a;nssaJd l
99 NUREG-1632
___________________ _ _ _ - A
(3,) a;nto;adwal l
/
l
~
l
/
?
1
/
52 h
N
/
G l
/
E l
/
/
8 8.
S E3 f
l E
1jl Em a
~
/-
.5 s
s
/
0 c4
\\
5
\\
l
'E s
2 9
f (y) a;nlo;adwal NUREG-1632 100
(oes/r) alo; Jejsuo;) X6;su3 "e
b b
b d
g 3'
I' i-g-
/
/
- l Ii i
i \\!
5 1
1 i
IN
- s 6
I
. g 3
\\
E
^
/
8 e
g S
g G) o
/
E e
\\
p g
//
}
\\
f
\\
/
/ /
E 6
p
\\
/#
x s
I N
5 i
N
(
, e 3
N
\\
ss s
5 I*
I-t.
t, "e
b b
b b
b (r) X6;eu3 101 NUREG-1632
(cas/p) ajo; Jajsuo;; X6;su3
%g
'e "s
'e k
k
'e
'I a
s I
- B e
8
}
l
/
2 b
/
\\
/
3
.\\
/
l !
5
\\
/
T E
i2
\\
I e
u E
g
\\
\\%
lf 5
e 8
5
\\
\\\\
I l!
1 i
y d
~
\\
1 lg N'
g x
l 1,
\\
\\
.s 2
S N
\\
N
\\
N N
=
b b
s s
g g
(r) X6;aug NUREG-1632 102
(
(oes/r) alo; Jejsuo;) X6;eu3 l
9 9
9 2
I
'S "S
M l'
l
5-j-
g--
i e
5
..2 N
=2d S
e e
O E
e
?
E E"
1=
S o
b I
- 8 k
bo
% \\
=
t
~
4 I
9 I
n n
w ti
- 2 II I.
t..
f.,
e,,
g,,
"e p
g b
b d
b (r) X6;eu3 103 NUREG-1632
l (ojsd) a;nssaJd n
n n
n n
n a
a E
A as g
I
\\
l I
//
l 1
l 1
h G
l mx 4
q 8
1 O
E s!
t S
1 o
p a
gl-9 q
a
}
E
+
3 a
- !I 4,
1 A
E E
- 1 N
1:
6
~
=
I i
5-Rn bl I
a l
I I
l.
I E
2 S
(Dd) 10000'0 x a;nssedd NUREG-1632 104
1 (ojsd) a;nssa>d n
n n
n n
n a
a 4
4 x
l I
I l
i E
3 l
1 g
.gl 4
o 5
E al e
cl I
l !
2 E
?
Y a
e nl E
.E P
5e i
1 0
e4 N
l ve II I
e I
I i
I l
I-i l
O O
O O
l d
O (Dd) L0000'0 x a;nssa>d l
105 NUREG-1632 j
(ojsd) OJnssaJd n
n e
9 0
0 I
i I
7
[h I
l l
5 i
g
.I 1
a
=
a ol 4
si 2
nl 2
j
_ g B
E I
3 E
S 4
ul
.E 5
~
u
$l H
d E
5 8n El
.s
~
w I
2 l
l l
l m.
0 a
a 3
(od)10000'0 x a;nssa;d NUREG-1632 106
i l
(J,) a;nto;adwal g
R 2
n n
M h
a a
s a
x n
n e
g l
1JN
.i.
O.-
a i
1 6
e 8
O e
1 2
8 e
E E
?.
2 w
8 g
i e
.5 E
5 P
8 u
4 4
C ei i
i ce j
Lt
- e j
Q l
1 i
i I
I I
E
(>{} a;nlo;adwal i
107 NUREG-1632
the event the temperature for the 100% relative humidity case is slightly below the base case.
The reasons for these results may be inferred by examining some of the containment atmosphere properties for the two cases. Table 4.2.13-1 lists containment atmosphere properties calculated at 20 seconds into the event. It can be seen that most of the gas properties (such as internal energy, density, and specific heat capacity) differ by no more than a few percent. However, there is a significant difference with respect to the atmospheric composition. The initial inventory of non-condensibles (air)inside the containment is reduced by about 11% when a 100% relative humidity atmosphere is specified. A reduced non-condensible inventory leads to a lower estimate of the containment pressure. In addition, the calculated specific heat capacity of the containment atmosphere is about 2.5% higher for the 100% relative humidity case.
4.2.14 Initial Relative Humidity (MSLB)
Containment sensitivity to initial relative humidity was not analyzed for the MSLB.
4.3 Other Sensitivity Analyses In addition to the set of parameters described in Table 4.1-1, several other plant design features were examined with respect to the impact on the containment pressures and temperatures.
Since the original submittal of the AP600 design, Westinghouse has made several changes to the estimated design-basis mass and energy source terms for postulated LOCA and MSLB events.
Section 4.3.1 (below) describes the effect of source-term changes on containment pressure and Table 4.2.13-1. Containment atmospheric condition sensitivity to initial relative humidity.
Containment Parameter 0% Relative Humidity 100% Relative Humidity (Base Case)
Pressure (KPa) 355.9 352.4 Temperature (K) 396.79 398.27 Internal energy (J) 1.58E+11 1.65E+11 Gas density 2.38 2.31 Sp. heat capacity (J/kg/K) 1.09E+03 1.12E+03 N (kg) 4.516E+04 4.029E+04 2
0 (kg) 1.371E+04 1.223E+04 2
H 0,(kg) 6.133E+04 6.396E+04 2
H 0,,, (kg) 1.685E+03 1.690E+03 2
NUREG-1632 108
temperature. A set of analyses also was performed in order to gauge the impact oflimiting heat-sink and PCS configuration assumptions, as described in Section 4.3.2. Specifically, calculations were made for totally adiabatic heat transfer (i.e., no heat transfer to any of the heat sinks of the containment shell), heat transfer limited solely to internal heat sinks (i.e., no heat transfer to the containment shell), and heat transfer limited to transfer through the containment shell (no heat transfer to internal heat sinks). Variations to these calculations were made with respect to the assumed presence or absence of the PCS.
4.3.1 Sensitivity to Evolutionary Changes in Mass and Energy Source Terms Throughout the course of safety review of the AP600 design, Westinghouse has made several revisions to the estimated mass and energy source term associated with design-basis events (e.g.,
LOCA, MSLB). The nominal effect of the revisions has been an increase in the peak calculated I
containment pressures and temperatures. In order to gauge the magnitude of this, a comparison was made of the containment response to the current source term (submitted in 1997) with that obtained earlier (approximately 1995) using a previously submitted source term.
i Figure 4.3.1-1 shows the DECLG mass release rate source terms for 1995 and 1997 versions.
l The principal change is the shift of the reflood mass release rate to an earlier time for the 1997 l
source. The 1995 source had a deadband (zero release) of about 40 seconds separating primary blowdown and reflood. The cumulative mass release is the same for both sources. The net effect of the source change on containment pressure and temperature is shown in Figures 4.3.1-2 and 4.3.1-3, respectively. Specifically, eliminating the deadband results in a peak pressure increase l
of about 10 KPa (1.8 psi). The corresponding temperature change is about 1.1 K (1.7 'F).
I A similar assessment of mass source change sensitivity was made for an MSLB at 30% power.
The 1995 and 1997 mass release rates for the MSLB are shown in Figure 4.3.1-4. The 1997 source shows a later time for blowdown termination (by about 240 seconds). The effect of this l
on containment pressure and temperature is shown in Figures 4.3.1-5 and 4.3.1-6, resp 6ctively.
l l
' The peak pressure increase for the 1997 mass release rate is about 23 KPa (3.7 psi). As seen in Figure 4.3.1-6, there is no significant change in the peak temperature.
4.3.2 Sensitivity to Limiting Heat Sink and PCS Assumptions The sensitivity analyses described in Section 4.2 provide a means of estimating containment pressure and temperature response to nominal changes in various design parameters. In addition, analyses were made of containment response to limiting assumptions with respect to the principal heat transfer modes (i.e., heat sinks and PCS). The usefulness of this type of information is in terms of acquiring a better measure of the margins available with respect to containment peak pressures and temperatures.
Specifically, analyses were made of containment response to the following limiting cases for a DECLG event:
109 NUREG-1632
ll\\!l
!i l
!\\I i'
\\
\\
\\
\\
\\
\\
~
~
5
.l#
k i
i
~
~
)B
~
-D k )c lC i
/
e
(
s s
(
e
<t
^
c ru e
o m
s
&; i e
l l
l T
s l
3 a
{
l i
l l
l e
ler s
sa M
1 1
3 4
.g iF 7
d r'
r r'
r r
,o
.t 8-f 1
1 f
'o
'-o 0
i 1
1 i
t 1
i 1
mSmNF"eowO2
$i9 0I 5
s J
(o!sd) a;nssaJd n
n n
n n
n 3
3 4
4 M
- s
,g I
~
l 2
l 4
U l
i
/
0 l
/
8 8e O
M
'S I
g il
~
s s
T.$
~
sE l
\\
O :-5 E
el
\\
EN
?
P I
4I I
8 E
a n
/
E el 5
a asI a
6
.b !
d bl
?
1 v
eb l
l I.
- (od) L 0000'0"x a;nssa;d
~
l11 NUREG-1632 I
\\\\\\\\;!
\\ll il
- i lil1ll
^p* c'5Oueag8g 30 4
4 4
4 4
0"
,4 8
7 6
5 4
4 4
4 4
4 4
1 6
5 3
2 1
2 2
2 2
2 2
2 i
1 1
1 1
1 300
~
00 1
)B-D l
0 C
00
(
0 e
1 tar ec ruos s
N sa m
~
o 0
t 0
y 0
t 1
i v
i
)
t i
c s
e n
/
s es
(
e r
e u
m ta i
r T
ep m
0 i
0 e
1 t
tn N
m e
n m
7 ia 9
t 9
n 1
o 5
C 9
9 3
1 1
0 3
1 4
.g iF
~
1 g
0 0
M 0
0 0
o 0
0 0
E 5
4 1
0 9
e 7
5 4
3 3
4 4
4 4
4 3
3 3
^b g'3~O eaEeH f#O out u
J
!\\\\ \\ !s
\\l
,(i
\\
l I
I llIl
(
ll.lll!
1!
ll l
l!
, g
- i
,v t
7 s
9 e
9 o
1 t
,e
. i
.t
. i
. 'o )c t
e s
(
I I
I I
l e
m
. 'o i T
. t
- a. 'o t
~
q
- a. #
1
- a. 0 1
r r
r r
r r
r d
o P
c
'0
'0
'0 v
4 s.
t i
1 1
1 1
i o
0 o
t 1
t mu N=cv $.Os..n o 2 a.
i
=w 2hm9
iIIl\\ill)!
l l
l m9E* e=* E -
Q s.
3 s.
5 5
o0 4
4 4
u 4
3 2
t0 0
s 0
0 1
)B 0
M 0
. 0 0
l 1
C
(
7 e
t 9
\\
a 8
r 1
\\
ec
\\
ru
\\
os
\\
s a
/
s i
0 a
s
\\
0
. 0 m
p
/e 1
0 g
o t
)
0 5
c y
6 9
t 9
e
\\
i s
v 1
=
i
(
t i
o s
P T
n e
e s
5
/
m 0
e i
r
+
T u
/
e s
4 0
s 4
. 0 e
1 r
1 p
tn
=
e e
m
/
r n
u i
s a
s t
e n
r o
p C
n g
5 i
s D
0 1
e
/
3 1
4
/
g iF 7
1 3
2 4^On.v "oo8;c
- eame 5
.n ZC 95M i
0 0
0 0
0 g
m2 0
0 0
0 0
m0 0
0 5
4 3
2 1
1 1
10 3
4 3
2 9
8 70 1
3 3
3 3
3 2
2 2
2 0
~
00 1
0 0
i0
\\
0 1
\\
^
\\
\\
~
\\
\\
\\
\\
00
'0
- 1
\\
)
C O
\\
S
(
e m
i T
5" f
i 0 1
/
/
/
[
f
~
1 0
0 0
4 4
4 10 m
M 0
0 0
5 4
3 5
73 5
3 4
^
- 2gi E.W
_G ZCn
completely adiabatic (zero heat transfer) heat transfer limited to internal heat sinks heat transfer limited to dry containment shell heat transfer limited to containment shell + PCS actuation heat transfer limited to internal heat sinks and dry containment shell base-ase (internal heat sinks, containment shell, PCS)
Figures 4.3.2-1 and 4.3.2-2 show the containment pressure and temperature response, respectively, to these limiting cases. As one would expect, the adiabatic assumption produces the most severe response. For this case, the pressure exceeds the design value as early as about 200 seconds. The effect of either the intemal heat sinks or the containment shell is to delay the design limit crossover by less than a factor of two (-200 to 340 seconds). The effectiveness of the internal heat sinks in energy removal is roughly comparable to that of the dry containment shell. Another observation to be made is that the design-basis PCS profile is insufficient by itself (i.e., without intemal heat sinks) to maintain the peak pressure below the design limit.
4.4 Sensitivity Analysis Summary The CONTAIN code was used to perform a thermal-hydraulic analysis of the AP600 containment. Westinghouse design data were used to model thermal hydraulic aspects of the AP 600 containment, including steel and concrete heat structures, mass and energy break flows, and initial conditions. The containment was modeled as a single-cell volume. A base-case analysis of the AP600 for DECLG and MSLB events yields characteristic containment pressure and temperature responses reported in related studies (Tills,1996; Vijaykumar and Khatib-Rhabar, 1997). The pressure maximum for both events is near the design pressure. Specifically, the maximum pressures calculated are 88.2% and 88.7% of the design pressure for the DECLG and MSLB events, respectively. The corresponding maximum temperatures were calculated to be 398 K (257 'F) for the DECLG event and 441 K (334 F) for the MSLB.
In order to gauge the containment pressures and temperature response to nominal changes in containment design parameters, a number of sensitivity analyses were made. Specifically, changes were made to PCS characteristics (wetted area, flow rate, initial temperature, actuation time), intemal heat sink areas, liner-concrete air gap, and initial relative humidity. The results for a DECLG event, described in Section 4.3, are summarized in Figures 4.4-1 and 4.4-2. These figures illustrate the approximate sensitivity associated with each of the parameters listed in the sensitivity matrix. The estimates are based on a linear fit of the results.
As a specific example, Figure 4.4-1 indicates tnat the containment pressure sensitivity to PCS flow rate is about -0.08 psi per 1% change in PCS flow rate. That is, the peak containment pressure decreases by about 0.08 psi if the PCS flow rate is reduced by 1%. Examination of the rest of the sensitivities in Figure 4.4-1 indicates that the sensitivities can be grouped into three categories, as indicated in Table 4.4-1. The table indicates that the most important parameters to containment pressure are heat sink area and PCS flow rate.
NUREG-1632 116
l Ill iii i'
l 41 n9Ev
$e 5
5 5
5 4
4 4
4 5
s s
5 s
5 5
3,0 4
3 2
1 4
m 4
4 4
4 4
1 1
1 1
9 s
5 4
3 2
,0 0
~
~
. 00 1
+
ee hs 00
/
\\
. 0
\\
0
/
1
/
/
/
/
/
/
/
/'
/ -
)
c B
it o
b S
D o
f C
d
~
P l
o
+
0 C
0
/
0
(
s f
h 1
e e
r
/
/
s
)
u s
k
+
c s
lan s
e
/
a, e
s ni s k
s r
r V
el n
(
p o '*
i, t
i s
s i*w t
o p
n h
loo e
e 0
e 0y m
m h
n 6
i i
T a
m t
=
n o- /
0 P
. 0 oC 1
hy 5-
~~
1 8
a 0
2 Wc
+
~
3 w
e 4
4 Wd 4-s 1
.g n
i
.d o F
=
e-ru s-0 s
1 e
rp g_
n is e
D 1
0 9
8 7
s 5
3 2
1 7
"oo8g x e 3Ee O Z
O' 8t J
(3,) a;nto;edwei E
E E
E E
E E
E E
Es
.g f
E 1
1 k
- l
\\
s
\\
\\
\\
-m
\\
d
\\
.\\ \\
j
\\
R 5
N s
ea
\\
E
.{
}f
\\
\\
E E
g
\\\\
1:
e
+
u f
- a
+
- 8 3
E
\\g u
.{
4 jf ll 1
]
N m
I (1
4 t
.s 9.
w
- e I
I f
f I
I 1
9 R
f f
R R
R R
R R
R 5 }9 6
3
(>{
a;nlo;edwai I
f NUREG-1632 118
lll E
R E
U R
S U
S S
E S
o.
R E
P R
s P
s O
XA S
M E
D E
S A
C 5
4 ESM l
i
/
l I
0 I
I 4
l g
l I
l ii iI Ir i i'
.l l
,I lI l
l8 Il l
5 3
s e
0 i
3 i
t v
i t
i s
)
n a
e P
s 5
(
e r
2 1
u 0
s s
0 e
0 r
p 0
t o
0 n
h 0
e a
m s
c e
2 X
a e c m
in 0
m r
s e a
u =
r n
s e
n g
0 i
5 u = m m 0 0 g
n a
0 o
r i
1 s
m u
a o
1 a
o t
o s
n 1
i s
o e
C 5
r 1
P 1
4 4
.g i
o.
F 1
e m
s s
o o
e e
e s
c c
o s
c o
o s
e c
c e
s e
o s
n e
e 5
t a
e s
n e
o o
o m
b o.
o A
E E
P E
s T
A A
R n
A E
C A
i R
R A
R D
T w
E R
K I
o A
T A
N T
T l
I R
E S
F S
EN W
S S
T U
S C
C A
C P
P EN P
3 2C h' Q
l!
ijlIi Ill!1I ll i
1 ERU TAR 0
E
, 1 P
4 M
E T
XA M
I E
l SAC E
I i
S l
AB 00 I
l l
4 g
{
l Ia
{
l I
Il 1l i4 0
93 s
)
e i
K t
i
(
v i
t e
i r
s u
n 0t es 03 a r
e r
e u
p ta m
re e
p T
m e
c c e
e b
e e m
t s s a
m M
- w "
1 0
o t
0 n
E N "
=
5 5
S 1
a 0
e m
7 1
1 3
n ia t
noC 2-4 4
0 6
g 3
i e
e F
se s
e s
o o
m o
o c
c c
c s
e s
e n
e o
o o
e o
h t
e b
n
~
0 53 A
E E
T A
P E
W A
E I
4G M
T R
RA T
w R
D R
O NK I
E A
T A
T L
I T
R E
S F
S E
W S
S T
N L
S C
C A
I P
P E
C H
P ZCbOh jo
Table 4.4-1. Containment sensitivity ranking.
High Medium Low Heat sink area Liner air gap PCS source temperature PCS flow rate PCS wetted area Initial relative humidity PCS actuation time PCS flow rate, in particular, plays an important role in the effectiveness of the PCS to limit themaximum containment pressure in the reflood phase of the DECLG event. Specifically, the results indicate that an ample PCS flow rate is necessary so that the PCS water can reach to the bottom of the containment wall. The AP600 design flow rate has an inherent excess flow rate capacity of about 50%. That is, initiation of dryout on the vertical wall does not occur until the PCS flow rate is reduced by 50%. Flow rates less than 50% of design are insufficient to wet the full height of the containment wall. The initiation of dryout can limit severely PCS effectiveness in energy removal.
Containment pressure and temperature response also is highly dependent on the amount of heat-sink storage capacity available inside the containment.. For example, a 50% reduction in the total heat sink area for intemal steel and concrete is more than sufficient to raise the maximum containment pressure to 4.1 KPa (61.3 psia), above the design limit. Hence, panicular attention has to be devoted to accurate representation of the structural design parameters (mass and area) in the thermal hydraulic modeN of heat sinks.
. Medium sensitivities are ass
,.ed $vith liner air gap, PCS wetted area, and PCS actuation time.
Minor sensitivity is observeo a a respect to PCS source temperature and relative humidity.
The sensitivity ana yses lead to the following observations:
(1) Heat sink capacity plays an important role in reducing containment pressure, even as far as 10,000 seconds into the DECLG event.
(2) Sufficient PCS flow rate is vital to the effectiveness of the PCS in energy removal and, hence, controlling containment pressure.
(3) Panicular attention has to be paid to modeling heat sink stmetural parameters (mass, area, air gap).
i l
(4) PCS actuation time can play a significant role in controlling the magnitude of the second
. pressure and temperature peak for the DECLG event.
In addition to nominal sensitivity analyses, evaluations were made of containment pressure and temperature response to changes in several other plant design features. Specifically, estimates were made of the effect of a revision in the mass release rates, as well as the effect oflimiting i
l 121 NUREG-1632
assumptions with respect to heat transfer to heat sinks and the PCS.
The impact of a revised mass release rate change in 1997 for the DECLG event was less than 2 psi in peak pressure and less than 2 *F in peak temperature. For the MSLB, the peak pressure increase was less than 4 psi. The corresponding temperature change was insignificant.
The evaluation oflimiting heat-transfer assumptions indicates that the internal heat sinks and the dry containment shell are approximately comparable in their heat-transfer capacity. The analyses also support the view that the design-basis PCS is insufficient by itself (i.e., without the thermal storage benefit of the intemal heat sinks and the containment shell) in maintaining the pressure below the containment design limit.
5.0 Summary and Conclusions Staff safety evaluation of the AP600 containment design included thermal-hydraulic containment analysis of selected DBAs. Normally, the CONTEMPT code is used by the staff to estimate containment pressure and temperature response. CONTEMPT is a lumped parameter code that is well suited for conventional containment thermal-hydraulic analysis. In conventional containment designs, the typical accident transients consist of a rapid high-energy blowdown that is terminated by the actuation and use of active safety systems (i.e., coolant pumps and sprays).
i Although the AP600 is similar to conventional large, dry containments, it has some notable differences. It relies on external cooling of the containment with the PCS, and it does not use active safety systems. The accident transients are comparatively long (measured in hours rather than minutes), and may provide opportunities for establishing flow patterns that are not necessarily representative of a well-mixed containment atmosphere. Funhermore, CONTEMirl is not capable of direct modeling of the extemally applied PCS. Hence, it was necessary to use the CONTAIN code to supplement CONTEMPT calculations.
The CONTAIN code is a multi-cell lumped parameter code equipped with a film tracking algorithm that is well suited to modeling the PCS. To establish continuity of methodology, a comparison was made of the two codes. The results for a DECLG indicate that the two codes yield comparable results for a single-cell representation of the AP600 containment atmosphere.
Furthermore, reasonable agreement was obtained with the CONTAIN code between single-and multi-cell models of the AP600. On this basis, the CONTAIN code was used to do selected confirmatory analysis of the AP600 Evaluation Model. The results show good agreement with Westinghouse }yGOTHIC results. Hence, providing that the assumption of a well-mixed containment atmosphere is valid, the results obtained by Westinghouse using the WGOTHIC code appear to be reasonable.
In order to gauge the relative imponance of some of the key AP600 containment design parameters, a set of sensitivity calculations was made using the CONTAIN code. The sensitivity analyses indicate that the containment heat-sink areas and PCS flow rate are the most controlling parameters with respect to containment pressure and temperature.
NUREG-1632 122
j l
- 6.0 References AP600 SSAR, AP600 StandardSafety Analysis Report, Revision 22, Docket No.52-003, Westinghouse Electric Corporation, Pittsburgh, Pennsylvania,1998.
' NUREGICR-0255, D.W. Hargroves et al., CONTEMPT-LT/28-A Computer Programfor Predicting Containment Pressure-Temperature Response to a Loss-of-Coolant-Accident, U.S.
Nuclear Regulatory Commission,1979.
NUREGICR-6533. K. K.^ Murata et al., Code Manualfor CONTAIN 2.0: A Computer Codefor
- Nuclear Reactor Containment Analysis, U.S. Nuclear Regulatory Commission,1997.
Tills, J., User Guidance on the CONTAIN Codefor AdvancedLight Water Reactors, SAND %-
0947, April 1996, Paprietary,.
Availability: DOE Sandia National Laboratories, Albuquerque, New Mexico 87185-0739 (upon request and pending approval by cogniznant DOE Departmental Element).
Vijaykumar, R., and M. Khatib-Rahbar, CONTAIN Applications to AP600 Design Basis.
Containment Analyses, ERl/NRC 97-201, March 1997, Proprietary.
Availability: Energy Research Inc., P.O. Box 2034, Rockville, Maryland 20847-2034 (upon request and pending approval oforiginator).
WCAP-14407, J. Woodcock et al., WGOTHIC Application to AP600, Revision 1, July,1997, Proprietary.
Availability: Westinghouse Electric Corporation, Energy Systems Business Unit, P.O.
Box 355, Pittsburgh, Pennsylvania 15230-0355 (upon request abd pending approval of cognizant Westinghouse official) l l
123 NUREG-1632 l
APPENDIX A AP600 SENSITIVITY ANALYSIS CASE IDENTIFICATION NOMENCLATURE The following is a case labeling scheme for the AP600 containment sensitivity analyses. It is a four-tier approach that can be expanded with additional ID letters as the need arises.
Tier 1: Code ID C.1 CONTAIN 1.2 C.2 CONTAIN 2.0 LT CONTEMPT Tier 2: Accident Sequence ID D
DECLG M
MSLB Tier 3: System / Component Analysis ID B
Base case PCS Passive Cooling System HS Heat Sink I
Initial Conditions EM Westinghouse Evaluation Model(EM)
Tier 4: Principal Attribute / Variable ID A
Area M
Mass V
Volume F
Flow rate T
Temperature P
Pressure H
Relative humidity TM Time G
Metal-concrete air gap 125 NUREG-1632
Specific examples illustrating the above scheme are given below.
1
(-
Frarnnies Cl-D-B Base case DECLG with CONTAIN 1.2 C2-D-PCS-F PCS flow rate sensitivity for a DECLG sequence with CONTAIN 2.0 LT-M-HS-G Heat sink metal-concrete air gap sensitivity for a MSLB case with CONTEMPT Cl-LT-M-HS-A CONTAIN 1.2/ CONTEMPT code comparison of heat sink area sensitivity for an MSLB sequence I
1 l
NUREG-1632 126
NRC FORM 335 u.s. NuCL.EA;i REGULATORY COMWsSION
- 1. REPORT NUM1ER rt49)
NRCM 1102, (Asagned by N;T., Add Vol., supp., Rev.,
and Addendum Numbers, it ast, '
32oi. 32o2 Cl!L!OGRAPHIC DATA SHEET (see manaons on u,e re r=e>
- 2. TITLE AND SUBTITLE NUREG-1632 j
Eviluation of AP600 Containment Thermal-Hydraulic Performance 3.
oATE REPORT PUBUSHED l
MONTH YEAR June 1998
- 5. AUTHOR (S)
- 6. TYPE OF REPORT K.M. Campe and J.D. Kudrick
.i.g
- 7. PERIOD COVERED (sociusae Dwes)
- 8. PERFORMING ORGANIZATION. NAME AND ADDRESS (#Nac, prowde Dwson. oke or Espion & S Nucerar Repusatory comnasaan, and madno ad*ess: #cznirects prowde name and maeng eatness.)
Division of Systems Safety and Analysis Office of Nuclear Reactor Regulation U.S. Nuclear Regulatory Commission Washington, D.C. 20555-0001
- 9. SPONSORING ORGANIZATION. NAME AND ADDRESS (#Nac. &pe *Same as abovoi #contrador, prowde AIRc Dwaron, oke or Asgson. u.S Nucieer Augusatory comernsson and maeng ad@ues)
Sime as above
- 10. SUPPLEMENT ARY NOTES
- 11. ABSTRACT (200mordsormas)
The thermal-hydraulic performance of the AP600 containment with respect to selected design-basis accidents (DBAs) was sviluated using the CONTEMPT and CONTAIN codes. The AP600 containment design includes a passive cooling system (PCS) in the form of gravity driven water flowing on the external surface of a steel containment shell. This design feature cennot be modeled directly within the CONTEMPT code. The CONTAIN code was used to estimate the PCS heat transfer coefficients that were provided as input into the CON rdMPT code. The results show fair agreement in terms of containment prs 2sure and temperature response to selected DBAs.
Confirmatory analyses were made using the CONTAIN code with the intent of venfying the Westinghouse analyses that were performed using the WGOTHIC code. The results indicate that the Westinghouse pressure and temperature estimates for the AP600 appear to be reasonable and are within the applicable acceptance cnteria of the Standard Review Plan.
The CONTAIN code also was used to conduct a senes of sensitivity analyses. The purpose of these analyses was to assess th3 r-tative importance of the key AP600 containment design features and operating conditions. One aspect of the sensitiW s
anityses involved the consideration of selected limiting assumptions regarding the principal modes of heat transfer with r&spect to the containment shell and the intemal heat sinks.
- 12. KEY WORDS/DESCRIPTORS (Dat nords orparases that m# esmar soseeners m ecsong re report.)
- 13. AVAN.ABlutY STA1LMENT unlimited, AP600 14 SECURITY CLASSIFICATION (Trus Page)
CONTEMPT unclassified (Trus Report)
CONTAIN unclassified
- 15. NUMBER Of PAGES 16 PftlCE NRC F OHM 336 (249)
Norm uras em y W by Eine FeoorW Forms, M
l O_S Federal Recycling Program
RE G.
A E D FC O NDR AN N N TA Srf S
U M E
L G R
A I
A E
CT P
5 5
0 H T 2
9 N 1 E A M C
X E D
I G N A A N 1 A M
1 N
O IG 6 TE 8 OAR 4IMU N 4CRN O 50O 7T 1
FREG 5 CNO6N 5 RIP I
5N-NH 0 VSFS 2SIPWA 1UDT2W N
O IS 1
S 0 0
M0 0
M0 33
-5 O5 E
S C5 SS E
0 SU TY2 EN E AR 8
T TO C SA S T D UVI B R DA E L N, LP AI R TUO C
N G T I O I
L R H A
A S N
E E A P
LW CUN i
l