FVY-86-104, Forwards Responses to Remaining 11 Questions Re NRC Review of RELAP5YA Computer Program for LWR Sys Thermal Hydraulic Analysis
| ML20213D831 | |
| Person / Time | |
|---|---|
| Site: | Vermont Yankee, Yankee Rowe, Maine Yankee, 05000000 |
| Issue date: | 11/04/1986 |
| From: | Capstick R YANKEE ATOMIC ELECTRIC CO. |
| To: | Rooney V Office of Nuclear Reactor Regulation |
| References | |
| FVY-86-104, FYR-86-107, MN-86-138, NUDOCS 8611120260 | |
| Download: ML20213D831 (75) | |
Text
_
. Telmhons (617) 87N100 f ,h TWX 710-380 7619 YANKEE ATOMIC ELECTRIC COMPANL~ ~__.
- 1671 Worcester Road, Framingham, Massachusetts 01701
- sYANKEE -.
November 4,1986 FVY 86-104 FYR 86-107 MN 86-138 United States Nuclear Regulatory Commission Washington, DC 20555 Attention: Mr. Vernon L. Rooney BWR Project Directorate No. 2 l Division of BWR Licensing References * (a) License No. DPR-3 (Docket No. 50-29)
(b) License No. DPR-28 (Docket No. 50-271)
(c) License No. DPR-36 (Docket No. 50-309)
(d) Letter, YAEC to NRC, FVY 83-4, dated January 14, 1983 (e) Letter, INEL to NRC, dated September 4, 1986 l (f) Letter, YAEC to NRC, FVY 86-94, dated October 16, 1986
Subject:
Response to Additional NRC Questions on the RELAPSYA Computer Code
Dear Sir:
By letter dated October 14, 1986 [ Reference (f)], information was provided responding to 14 of 25 questions requiring resolution pertaining to your review of the Yankee Atomic Electric Company computer program for LWR System Thermal Hydraulic Analysis (RELAP5YA). We herein provide, as Enclosure 1 to this letter, information responding to the remaining questions.
We trust that the enclosure satisfies the informational requirements associated with your review of the referenced submittal; however, should you have any additional questions or require future information, please contact us.
"..ery truly yours, YANKEE ATOMIC ELE RIC COMPANY fYo R. W. Ca'pstick Licensing Engineer Vermont Yankee Project RWC/bil Enclosure 8611120260 861104 l
L
!1
[I
!I RESPONSE TO 11 NRC QUESTIONS ON RELAP5YA ig October 31,1986 l5
.I I
I I
I t I
I I
I If the calculated high power assembly fluid mass results of Q1.7 Reference Q-6 for LBLOCA-EA (Fig. 4.1-14) and LBLOCA-EB (Figure 4.2-14) are compared, LBLOCA-EB shows an unusually large (50 lbm) water influx to the bundle at about 90-100 seconds not calculated in LBLOCA-EA. Considering the response similarities of LOCA EA and EB, why does LOCA-EA not show a similar influx response?
I (Note: Fven the conservative BE case, LBLOCA-BA (Reference Q-7), does not show this influx even though it's at a lower bundle power at this time.)
A1.7 The mass surge in Case EB occurs between 90 and 130 seconds.
The Peak Clad Temperature (PCT) occurred at 78 seconds.
Therefore, the mass surge did not cause the temperatures to turn around and had no impact on the PCT. Furthermore, nearly all the liquid correspeading to the influx penetrated only into I the topmost nodes of the assembly.
illustrate this point.
Figures 1.7-1 and 1.7-2 Between 90 and 130 seconds, only Volume 162090000 shows a significant drop in void fraction (to about 0.3). In the same time frame, Volume 162080000 shows a slight decrease to 0.85, while all other elevations remain essentially unchanged. Since the temperatures of concern occur in Volumes 162050000 and 162060000, the influx does not affect the volumes where the relevant clad temperatures are located.
The mass influx is caused by small differences in the hydrodynamic characteristics of the Side Entry Orifices (SEO),
as well as the leakage paths to the bypass for the three channels: high power, central average, and low power. Since Cases EA and EB have different hydrodynamic histories, it is not necessary that Cases EA and EB show the same hot channel inventory history.
{
Figure 1.7-3 illustrat the influx mechanism. The figure is a I snapshot at 100 secondo. The horizontal axis corresponds to elevations above the center of the lower plenum topmost cell I
I (Volume 012). The vertical axis shows the pressure at each elevation. There is one curve for the bypass region and one I for each of the three core channels: central average, low power, and high power. Figure 1.7-4 indicates the central average channel has very low upward flows. Therefore, the near I linear pressure increase from the upper plenum (16.83 ft), to the center of the central average bundle inlet piece (1.842 ft) shown in Figure 1.7-3, corresponds mostly to the hydrostatic head of vapor. At this time (100 seconds), the central average SEO is in counter current flow. Hence, there is a small pressure increase across the SE0 into the lower plenum (0.0 ft I elevation). The bypass curve nearly overlaps the central average curve from the upper plenum down to the 12-ft elevation. This indicates the presence of nearly stagnant i vapor in the upper regions of the bypass as well. From 12 ft to 1.842 ft the bypass pressure increases almost linearly indicating a hydrostatic head of liquid. Figure 1.7-3 shows a significant pressure drop from the 1.842 ft elevation to the lower plenum for the bypass curve. Figures 1.7-5 through 1.7-7 show this hydrostatic head will drive flow from the bypass into I the inlet pieces of the three core regions, at a similar rate per bundle. The resistances between each region and the bypass are not identical. Neither are the SEO resistances and flow areas of the three regions. The flow from the bypass to the inlet pieces causes downflow in the SEO of the high and low power assemblies. This causes the significant pressure drop from the 1.842 ft to the 0.0 ft elevation for these regions.
However, flow from the bypass does not disrupt the CCFL l condition in the central average assembly at this time. This l
l is because the hydraulic characteristics (areas and flow resistances) are different from the high and low power assemblies. Above the inlet piece, all three regions are i nearly dry and stagnant up to Node 8 at the 13.5 ft elevation.
This is indicated by the nearly flat slope of the pressure curves between the 13.5 ft and 1.842 ft elevations. Clearly.
l the pressure at the outlet of all channels must be that of the upper plenum. The mechanism which allows this to occur in the l
lB i
I t high and low power channels is liquid penetration into the
- -= Lupmv.t cello iv fe w a hydrostatic hcad. This implic: th:t l the CCFL condition for these channels has moved to the top of 1 Node 7. The hydraulic resistances along leakage paths from the bypass to central average bundle and across its SEO are such that the SEO remains in countercurrent flow at this time (100 seconds). Therefore, there is only a very small pressure increase from the SEO to the lower plenum. This means there is no driving force for liquid to penetrate into the topmost cells I of this region at this time. This is because the small pressure drop associated with a low vapor upflow provides the mechanism to bring the bundle inlet pressure to the upper i plenum pressure value.
I I
I I
I lI l8 I
ll 1
.I
I VERMONT YANKEE NSSS LICENSING MODEL
.I CASC EB: LARGE BREAK LOCA APPENDIX K RESULTS DEG RECIRC LOOP OISCHARGE PIPE BREAK (2 X 3.64 FT2) i 3 i
,blD i q I
g l c I '
p .
il E,a; o
J' adV l -
h { ,
5 g
il' i
g.- , r j-l a-o VOIOG 16'040000 2 ,
a -oVOIDG 162030000 5 *--* VOIDG 162020000
+--+ V010G 162010000 1 Eo d.
1 o.o e5.0 tic.o 16'0.0 240.0 20 TIME (SEC)
I Figure 1.7-1: High Power Region Void Fractions (Nodes 1 to 4)
I e.
' ' 5i k MY k:il h ' '~ ' ' i hi"
- u l
E. ,,
i f f i
- i .; <
, 3 ,
ll -
I ae
~ .h l l::
," i i n
I g o l d z ( o e
E o t d h 3 gy n. u...L I
- c. C i, h
..!:,[:
g g o , a g g i u :
N :ll 1 * $- e-oVOIDG 162090000 "o--
n- " : - o ---
a-o V010G 162080000
- - aVOIOG 162070000
- VOIDG 162060000 1 "
m VOIDG 162050000 o.o so.o 120.0 180.0 240.0 300.0 i
I TIME (SEC)
Figdre 1.7-2: High Power Region Void Fractions (nodes 5 to 9)
11 (1l(lll1 l e
m
(
0 1
_ _ 8 1
M E g G - 0 R .
R i 6 m i E
V R
- 1 RRE EW a LWO l m SROP SRP RT H 0
i4 F
.T i
PNWG 1 (
YEOl m BCLh a" x M 0 U o . x .N a i X- ,'
b e 2 1 E L
P m N i 0E 0
.R i 1 W
0 m N-X .
b I N i08 .E W i V O n s o
- _ B i t
- _ n a W' -
b 0
.N l E
v e
i x_ + i 6 O s
I u
. s M .
T R V r
e 0V e r
4 .E u M ,
L s s
," E e r
+ P 3
M 0 7
, 2 1 e
r M u N
0 0
i F
g M - - ~
aQ ndm od.
.nm t n4m t o4D m t m C&nO~
(
$aDn xQ
( (
W W
i t l!llll
VERMD*4T YAr*EE t4SSS LICEt45tt4G _t SCEL CASE EG LARGE ORCAK LOCA PPPENDIX K RESULTS DEG REC!RC LOOP OISCHARGE PtFE_EREPK f2X3.64 FT21 l
- o s:
72 o
C*
I M 2
S e.
2 o-cLOWER TICPLATE o--o VPPER TIEPLRTE E 3 u.
E O- I -
g E :
W : ,
@ r- ,
i' l $-U:
e 0 l b.,
x \. s s !.
- ,- 7 "Mi =
.* 2 C
I ; 30.0 60.0 35.0 INO.0 1 *.>0. 0 0.0 TIME ISEC1 Figure 1.7-4: Upper and Lower Tieplate Flows for The Average Power Assembly I o
~
LJ t.J E ' '
l' l
l
~
l
.J ,. /'I
[W; W 5 * * - i
,y jj Y^%
n? i I i e
O i i
M l o
l $Y~
E l L
I ~) o 0*
c E =.
k 0.0 30.0 60.0 30.0 120.0 150.0 TIME ISEC1 I Figure 1.7-5: Bypass to Low Power Assembly Leakage Flow.
l l VERMONT YAr4XEE NSSS LICEr4 SING PCCEL CASE ES: LARGE ERERK LOCA FPPD. Ole K RESULTS
\
CEG RECIRC LOOP DISCHARGE _ FIFE BREPX f2X3.64 FT21 kl I c-U lg*
d bbd i i lI( ,k
' ~ j lg 3, I f 3 'NyctAd W WW , ,
I
' c, '
.s
$y i l Ei
! M j t p -
5
t I
i ~
5
- a. !, ~
$ l x l wi$
l l
,I 1 %,
ec 4
I O. 0 30.0 60.0 30.0 :23.0 isa.C IIPE (SEC1 Figure 1.7-6: Bypass to Average Power Assembly Leakage Flow lI ,
f U
C' I g e-d 8
d-t E U ,
E u
I j( i n l {3lpN I
wm,,,
I t
O l
$ o., - l ,, J F" E' !
5 E
E 'i T' . .
0.0 30.0 60.0 90.0 120.0 150.0 TIME (SEC1 Figure 1.7-7: Bypass to Higher Power Assembly Leakage Fbw i
E Ql.11 The following general question pertains to Question Q.VII.S.
- Gin:: int:rph::: ir:g =: del: :c re enly - - ~ ~-- ~^ *-- ~~ - ' ' ~~ -
of 729-1464 kg/s-m2 (FRIGG Loop) and 0-100 kg/s-m (GE Level Swell Tests), provide code assessment, or justification for omitting assessment, for the mass flux range of 100-729 kg/s-m and for void fraction transitions between zero and one.
l (Note: This unassessed mass flux range is approximately 507. of 2
I the total 0-1464 kg/s-m flux range presented.)
A1.11 The FRIGG loop experiments (Reference 1.11-1) simulated by RELA 25YA in Reference 1.11-2 provided the most direct assessment of the RELAP5YA interphase drag models since the void fraction was measured independently (by gamma-ray attenuation) at well-defined steady-state system conditions.
These experiments were conducted in a 4.375m long, 36-rod bundle, with uniform axial and radial power distribution. The lowest mass flux in these experiments was 729 kg/s-m ,2 a
In the General Electric vessel top blowdovn tests (Reference 1.11-3), differential pressure over a 2-ft length was measured at seven separate axial segments in the vessel during a blowdown transient. In these tests, a two-phase level transient is observed due to the combined effect of flashing (due to depressurization) and gravity separation of liquid from vapor. In the test simulated by RELAP5YA (described in I
Reference 1.11-2), the vessel mass fluxes in the two-phase 2
region belcw the level were typically about 20 kg/s-m , with 2
a maximum of about 100 kg/s-m at the beginning of the transient. These flows are low enough that pressure differences in the vessel are mainly due to static pressure differences caused by density variations. Thus, the average void fraction could be derived with reasonable accuracy in the axial segment over which the DP measurement was made.
I I
I
) For assessing the interphase drag models in RELAPSYA, the most desirable experiments would be those in which the void fraction is experimentally determined with reasonable accuracy for well-defined system conditions. Such is the case in the FRIGG loop experiments and the General Electric level swell experiments. Hence, these tests were identified in f
Reference 1.11-2 as most appropriate for directly assessing the RELAP5YA interphase drag models. However, Reference 1.11-2 contains other assessment cases where the RELAP5YA interphase drag models have a significant impact on the results. One case is the TLTA Boil-Off Test 6441/6. In the simulation of this test, the void fractions in the bundle were calculated very I well by RELAPSYA. The mass flux in this test was low, similar to the General Electric level swell tests. However, both the General Electric level swell test and the TLTA boil-off test do cover the entire void fraction range, from zero to unity.
I We have searched for additional data for assessing the RELAP5YA interphase drag models in the mass flux range of 100-729 kg/s-m . The only data we have found so far are from I
the FRIGG Ivop. These data are from the series of tests conducted with nonuniform radial and axial power profiles in the test bundle (Reference 1.11-4). The lowest mass flux in 2
these tests was 472 kg/s-m . Described below are the results of the RELAP5YA simulation of two of these tests at mass fluxes 2
of 472 kg/s-m and 483 kg/s-m . Both tests were conducted at a pressure of about 30 bar.
I The FRIGG loop experiments with nonuniform radial and axial I power profiles were conducted in the same facility as the previous uniform power profile tests described in Reference 1.11-2. The test section used in these tests (FT-36c) was slightly different than the test section used in the previous tests (FT-36a). Figure 1.11-1 shows the FRIGG loop facility and Figure 1.11-2 shows test section FT-36c along with instrumentation location. A cross-section of the test bundle is shown in Figure 1.11-3. The relevant geometrical I
- p. 1 E
information for Test Bundle FT-36c is shown in Table 1.11.1, I along with the previous test section. As can be seen, the only differences between Test Section FT-36c and FT-36a are that FT-36c is slightly shorter (4.365m) than FT-36a (4.375m), and that the radial and axial power profiles are nonuniform. The radial pov.er profile is indicated in Table 1.11.1. The axial power profile is shown in Figure 1.11-4.
I A typical test consisted of measuring void fractions at seven axial locations using gamma densitometry for a fixed pressure, I flow, inlet subcooling, and bundle power. The results are presented in Reference 1.11.4 as plots of mean void fraction versus the dynamic equilibrium quality. The mean void fraction at each axial location is the weighted mean of void fractions measured over four separate radial zones.
The REl.AP5YA model of these tests is similar to that described in Reference 1.11-2 except that twenty equal length nodes (each 0.21825m long) were used to represent the test bundle. The I axial power profile shown in Figure 1.11-4 was appropriately integrated and averaged to yield the relative power in each axial node. Two tests were simulated with this RELAP5YA model. In each simulation, the calculation procedure was to specify the cutlet pressure, inlet flow, inlet subcooling, and total bundle power as per the test conditions, and to allow the 1
calculation to reach steady-state conditions. The void t
fractions at each of the 20 nodes at steady-state conditions constitute the results of the calculation. The dynamic
' equilibrium quality at each node is calculated separately via an energy balance. The void-quality relationship from the l
calculation is then compared to the test data. Figure 1.11-5 l shows this comparison for Run 613120 at a mass flux of 483 l kg/s-m . Figure 1.11-6 shows a similar comparison for Run 613124 at a mass flux of 472 kg/s-m2 and higher bundle power. As can be seen, the RELAP5YA calculations compare very I
well to the data, i
I l5
I The RELAP5YA simulations of FRIGG Tests 613120 and 613124 indicate the adequacy of the interphase drag models at mass fluxes of about 475 kg/s-m2 over a void fraction range of 40%-95%. This mass flux is approximately in the middle of the mass flux range of 100-729 kg/s-m where previous assessment was lacking. There are no void fraction data from the FRIGG 2
loop experiments at mass fluxes lower than about 475 kg/s-m .
I
~
References I 1.11-1 Nylund, O., et al., " Hydrodynamic and Heat Transfer Measurements on a Full-Scale Simulated 36-Rod Marviken Fuel Element with Uniform Heat Flux Distribution," FRIGG Loop Project, Report FRIGG-2, R4-447/RTL-1007, Atomenergi, I Stockholm, Sweden, 1968.
1.11-2 Fernandez, R. T., et al., "RELAP5YA - A Computer Program for LWR Thermal Hydraulic Analysis, Volume III: Code Assessment,"
Report YAEC-1300P, Yankee Atomic Electric Company, Framingham, Massachusetts, October 1982.
I 1.11-3 Slifer, B. C., and J. E. Hench, " Loss-of-Coolant Accident and Emergency Core Cooling Models for General Electric Boiling i Water Reactors," NED0-10329, April 1971.
1.11-4 Nylund, O., et al , " Hydrodynamic and Heat Transtei Measurements on a Full-Scale Simulated 36-Rod BHWR Fuel Element with Nonuniform Axial and Radial Heat Flux Distribution," FRIGG Loop Project, Report FRIGG-4, R4-502/RL-1253, Atomenergi, Stockholm, Sweden, 1969.
I I
I I
l I
E TABLE 1.11.1 Geometry Data for FRIGG Loop Test Bundles FRIGG FRIGG FT-36a FT-36c Number of heated rods 36 36 Heated length, mm 4375 4365 Radial heat flux distribution Uniform Nonuniforml Axial heat flux distribution Uniform Nonuniforml Heated rod, OD, mm 13.8 13.8 Unheated center rod, OD, m 20 20 Shroud, ID, mm 159.5 159.5 Equivalent diameter, mm 26.9 26.9 Heated equivalent diameter, mm 36.6 36.6 Number of spacers 8 8 Chimney height, mm 1540 1550 j Operating pressure, bars Variable Variable Inlet subcooling, OC Variable Variable i Inlet throttling, velocity heads Variable Variable HO 2 HO Coolant 2 1 Relative radial heat flux distribution of FT-36c: ,
i I FT-36c 1
g 6 inner rods 12 interjacent rods 0.854 0.926 l
E 18 peripheral rods 1.097 I 2 Relative axial heat flux Gstribution (same for the three circles):
Figure 4.
see I
l I
E l
I c 2 L 10119
- m . ,. .., @ _.
h STEAM SEPARATCN I ;**
, it L 8199 .
L 8619 CONDENSER h 7155 I L 7399
'"" ~
I i L sies 4 -2 n ,
(? '
- r: ,e 1915
- L 5601 P
o,,, ,,
i 52 ,,,,, ,,,
regwintion of ggg gg j
- -'a st
- 4 meat s t
subcootsng g 54:0 2300 C Nb U end I L4199
-( :
ll Length .I ( N N 2 55 .,Q n D..
1.agth
.9 Is .t ' , , , ,
l I / 227 m: =
I ,,,,v.
?tST SEC?iCN F t . 3 8e (1-=*;
t ,i t ..t., , -
Mt L e Levet indication pressure top I Me Moss flow meter .
Po Pressure tap - ' =
I its i
31 4
~
L...e no .*
a no ,,, i . , ,, n
,5 d'*
I Critic oiet. vi met.e N 488
- :wie w: a _'
I ti.e.ricai n.ai.e tenir var store up) '- -
]
- Dimensions in
_/ , o I Figure 1.11-1 FRIGG Loop Facility I
I
E 1 I
I me (7 esi 7!.M.ill-
- y=
. . - .a. u
--e--
m I ,. -
.a.-
)W 10 J'"y"' .-
\
E . e. . ,7
..,i ,
- - = = 3e32 (3915)
I Twt.no Os estee -5003 (4005) T (f:9c)
-::m 723 724 t.r -- td - dees (4800)' 7 4?ac t47*7)
P22 I Esw of elseter - 4436 (4415) 3s.a e- -mM(436:s End of h g.g (435tl ma P21 T21 T22 4274 (a2G) at:0 teW) g7 aog2 (830) l E se e - 3mz (3?se) N m n. m := c:3m le 3%B (' age)
I ss e - szss (32321 IW 3
ma t is sio: tsais s
F- -17 25 L:rv11 5 Se.e.e - 2710 (27c2)
,a M6114 ft? 2469 (2462) 64 2383 (2 M) l s - nn tm, ,
., _(_,
~ '= i = >
me _ , . , , , , , ;3 l . _ , _ , - _ , , . . . _ , _ ,
s, - icS. i_ _
2 ,2 _ , _ ,
f
_ g9 I N1 414 (a13)
I i
- ' No fic tit 17e tinj s- .f
- . . . , , i g
1_ ,z-sc
- e. 1. u-2 mec z._ s.
3 I
I
E.
I l i
I Unhegfed center rod 1
j
-- &, l I 'I
' 's ,
' /'
/ i)i
\
@_, - /
I ! g@@ 78bk( /'" ' D sg. ~
8
, 22, GW 0 so T
~
Bernout detecter no (FT.36e)
I I a -.1. u-3 c ... ec 1, ,, ,z.se,, ,z.,,, ,,, ,, ,,,
1 I I
I I
g g
g 5 _
4 5 _
- - 'I k:. 32A oy 4 c
2 6 g 3 e,
5 3
3 bl 4 T o_ F o 4 2 f 0 o g
9 6 5 r 9
o o i
t
- 2 5 u 1 3 b 4 i S .
o.
1 3 a r
t s
s i D
2 4
3 e x E '6 1
- 0 3
t a
n l u
1 i F d
r t o
2 o a c e M 8 1
6 1
2 12 5 t h
s H
1 - - l t
e a n i
/ 2 A x
8 1
e 2
2 i v
t
- - a
- 2 5
7 l
R e
1 n
5 a 6 1 e
- H 1
5 -
/i t
, 4 1
1 1
- o. 2 1 9
/ 8 0
e r
u g
6 i 9 2 5 F o 6 4 o f.
6 0
5 e.
o 42 l0 5
0 4 2 o o
6 0 o 0
- 3[j I
- 5.A
' i'
I I
5 FT GG RUN 6 '. 3 :. 20 P=30.5 BAR G=483 KG/S-M2.
I 9 o=2980 KW orsus=s.1 oEG.c.
O i i i i o
I. - .
l l
I r-o s .
C- LJ Z CO ,C' u -
l t a .
,v/
I 1oi c.-
e m'
- --- DRT6 i
R0_RPS ISG Lu !
i i
.._. l I _ - .
~
l l
i
- . J
[}',
___. = - -
I c ,
= !
G -- . i,1 ,
c+_. m. i e i
I L_ ! If !
v_- ! ,
5 5 Y, I,
. m l I l !
I (-). p
- r. - p) nL. c e
,a p) , a p c. [ ; , U J. 9rV , L)i
- s. a s.
g J Y \! , E 0B '1 DUFiL I T't , %
ne-e 1.u-s m o m c.,_1. m mcca enuo g
I I
I
'I .
.5 FR GG RUN 6:.3124 P=30.5 BAR G=472 KG/S-M2.
I o 0=4495 KW DTSUB=5.9 DEG,C, O I I I l C
5 -
I. O
. % _ 3 O
I
~
' ' A_;
<d CO f / l O .6' !
C I
O j U ,
l -
I % =b -
!Z.
0 L-c...n gg7A I _.
O ! REL9P5 ." E :'
.__- _j l __
I i iy I Q I
' '/ , :
j -
,-; O q, I I '
(~)
3
~~
E O i
I
' l l I I 0.0 10.0 20.u a0.0 40.0 g JYN. E DB 'i . DUFiL I ~ Y , %
Figure 1.11-6 RELAP5YA Comparison to FRIGG Run 613124 iI I
I
I Q1.13 YAEC stated in Q/A.I.1 that the impact of t.he interphase mass transfer models in RELAP5YA has been shown to be in the conservative direction. This state.nent was based on code I assessment work presented in Volume III of YAEC-1300P. The work referred to in Volume III was an EM calculation for a TLTA large break LOCA. While the calculated results are conservative with respect to the PCT, it is not clear the interphase mass transfer models will have a conservative impact for BWR/PWR SBLOCAs. Provide additional information to support the assertion that the interphase mass transfer models in RELAP5YA have a conservative impact on SBLOCA.
I A1.13 The following information provides support for our position that the RELAP5YA interphase mass transfer models together with other RELAP5YA models and Appendix K requirements provide conservative LOCA-ECCS licensing analysis results.
Our goal in performing LOCA-ECCS licensing analyses is to ensure that we produce conservative results that comply with 10CFR50.46 criteria. The parameters and criteria are the following:
- 1. Peak Cladding Temperature (PCT) 1 2200 F locally.
- 2. Maximum Cladding Oxidation (MCO) 1 17% localty.
I 3. Maximum Hydrogen Generation (MHG) 1 1% core wide.
- 4. Core geometry remains amenable to cooling.
I 5. Long term cooling is achieved.
A conservative prediction of maximum cladding temperature and core liquid inventory for each LOCA case generally ensures that we have not underpredicted the paramet.ers that demonstrate compliance with 10CFR50.46 criteria.
I I
I We achieve this goal by using the following:
I 1. A qualified method, approved for licensing analyses, that includes conservative models required by Appendix K.
l
- 2. Conservative input for certain parameters that includes those required by Appendix K and those additionally I selected by the user.
- 3. Sensitivity studies, for example, break spectrum and I burnup studies.
Thus, the combined features of the licensing analysis method ensure conservative results.
I Numerous phenomena occur and interact during a postulated LOCA. RELAP5YA contains many models to simu' late the dominant I phenomena and their interactions. Where possible, these models have been assessed against separate effect test data to assess I their adequacy. However, direct separate effect assessment of certain models, such as interphase mass transfer, is not possible due to the lack of relevant detailed data. Also, separate effect model assessment does not always ensure that model interactions will be reasonable or conservative. Thus, we also rely upon assessment of the overall analysis method against integral system effect tests that simulate LWR LOCAs.
I This latter approach is the context within which the interphase mass transfer models were assessed. The following sections I describe the BWR and PWR integral system effect tests used for this assessment. The major objective of these simulations is to verify that RELAP5YA does not overpredict the core liquid inventory nor underpredict the peak cladding temperature in these LOCA tests. Each test and the corresponding RELAP5YA calculation involves significant interphase mass transfer.
Thus, the simulation of these LOCA tests provides the basis for I determining whether the RELAP5YA interphase mass transfer i
I
i
'5 1
models together with other models and input parameters provide I conservative LOCA predictions. Table 1.13-1 cross refer _ences the attached figures to their respective document.
I BWR LOCA Assessment RELAP5YA has been assessed against three TLTA blowdown tests for BWR LOCA applications. These tests, the type of RELAP5YA analysis (BE or EM) and pertinent results for the pressure, bundle mass inventory, and cladding temperatures are summarized below.
I TLTA Test 6426/1 simulated a large break LOCA without ECC injection to examine the blowdown and severe core heatup characteristics. A RELAP5YA best-estimate (BE) analysis was performed and compared to test data in Section 5.2.4 of Reference 1.13-1. Figure 1.13-1 shows that the predicted pressure compares well with the data. Figure 1.13-2 shows that RELAP5YA underpredicted the bundle mass history. Figure '.13-3 shows that the predicted cladding temperature is always higher than the measured temperature at the peak power location half I way up the bundle. Figure 1.13-4 shows that the predicted cladding temperature at about 75 percent up the bundle is f higher than the data until about 200 seconds where they cross.
However, the following information should be noted:
I a. The axial location fer this predicted temperature is five inches above the measurement location, and is, therefore, in a lower axial power zone.
I b. The prediction used an average rod with a local radial peaking factor of 1.000, whereas the peaking factors for the two rods in this figure are 1.056 and 1.013.
- c. The predicted cladding te..serature corresponds to the outer surface, whereas the measurement locations are near the inner surface and would naturally be hotter.
I J
I
- d. TLTA Test 6425/2 shows that the bundle was quenched by 150 seconds when ECC injection is allowed. )
l Thus, during the period of interest, this RELAP5YA BE analysis followed the trends, but conservatively predicted the bundle mass and cladding temperatures.
TLTA Test 6425/2 simulated a large break LOCA with ECC injection to examine the blowdown and core response with ECC.
A RELAP5YA BE analysis was performed and compared to test data I in Section 5.2.3.4 of Reference 1.13-1. Subsequently, this case was reanalyzed and reported in Appendix A.IV.1 of Reference 1.13-2. Figure 1.13-5 compares the calculated and measured system pressure for this test. The calculated pressure agrees with the data until LPCS and LPCI injection began at about 64 and 75 seconds. Thereafter, the calculated pressure declines below the test data until it finally stabilizes near the end of the test.
I Earlier we had speculated that the underpredicted pressure was caused by excessive condensation due to the cold ECC injected into the steam-filled regions. Recently we have critically reviewed this aspect of the assessment, and believe the underprediction resulted from not modeling certain internal structures. This is based upon the following information:
I a. TLTA Tests 6425/2 and 6426/1 were essentially identical tests except the latter did not have ECC injection, I
- b. Figure 1.13-6 presents an overlay of the measured pressures for these two tests. The comparison shows that the pressure history after ECC injection began is higher than without ECC injection. This suggests that additional net vaporization occurred after ECC injection began in Test 6425/2 rather than condensation.
I I ,
I
- c. Figure 1.13-7 shows'that the calculated preesure history for TLTA Tests 6425/2 and 6426/1 were essentially the same with and without ECC injection. Thus, we now believe that I excessive condensation is not the cause for the underpredicted pressure.
- d. The RELAP5YA TLTA model for these tests contains passive heat structures for the vessel and the metal-ceramic channel wall. However, it did not model the upper and lower electrode plates, upper plenum, bypass tubes, control rod guide tubes, and pretest heaters in the lower plenum since we had thought the stored energy in these I internal structures was less cignificant. Recently, we performed a hand analysis using the system I depressurization model in Appendix C of Reference 1.13-3.
This analysis approximately accounts for the difference stored energy release from these structures would have on the depressurization rate. The results from this analysis indicate that the extra stored energy would account for the difference between our RELAP5YA calculation and th.e measured pressure histories. Finally, we note that all I passive heat structures are represented in our VY NSSS model.
!I Figure 1.13-8 shows that the RELAP5YA BE analysis generally underpredicted the bundle mass for TLTA Test 6425/2.
Figures 1.13-9 and 1.13-10 show that RELAPSYA generally overpredicted the cladding temperatures in the bundle.
A RELAP5YA EM calculation was also performed for Test 6425/2 I and is described in Section 5.2.3.5 of Reference 1.13-1. The comparison of the calculated and measured pressure histories is very similar to Figure 1.13-5. Figure 1.13-11 shows that the calculated bundle mass history is lower than the test data.
Figures 1.13-12 and 1.13-13 show that the calculated cladding temperatures were substantially higher than the test data for this EM analysis.
'I .
I TLTA Test 6432/1 simulated a small break LOCA with partial ECCS, i.e., LPCS and LPCI were available and HPCS was unavailable. A RELAP5YA BE analysis was performed for this I test. The results are compared to test data in Section 5.2.5.4 of Reference 1.13-1. Additional information is presented in Reference 1.13-2. Figure 1.13-14 shows that the calculated pressure followed the measured value reasonably well, including the ADS period, until the ECCS began injection. Then the predicted pressure becomes lower than the data for the same reasons described earlier, i.e., the lack of certain heat structures in the model. We observe that actuation of the ADS I
essentially converts the small break into an equivalent larger break LOCA that depressurizes more rapidly. Figure 1.13-15 compares the calculated bundle mass history to the test data.
The two parameters show good agreement until the ECC injection began at about 450 seconds. The drainage characteristics, and perhaps the more rapid depressurization, calculated by the code then depletes the bundle mass, whereas it maintained a constant value in the test until 500 seconds. Figure 1.13-16 compares the calculated and measured cladding temperatures at the high power elevation in the bundle. The cladding was well cooled I throughout the test, whereas the calculation showed two brief heat-up periods when the bundle mass was depleted. Thus, this RELAP5YA BE calculation provided a somewhat conservative prediction of the bundle mass and maximum cladding temperature histories. A corresponding EM analysis was not performed for this test. However, we believe it would yield more conservative results because the additional power (102 percent of initial power and 120 percent of decay heat) would have resulted in a more severe core heatup.
I Based upon the above assessment cases, we draw the following conclusions about RELAP5YA for BWR LOCA applications:
- 1. Each BWR assessment case involves significant interphase mass tranrfer. This includes vaporization due to depressurization and heat transfer from heated rods, the I
I i I
channel, and vessel walls. This also includes l condensation during periods of ECC injection.
l
- 2. No direct measurement of interphase mass transfer rates are available for comparison to the calculations. l However, the impact of the interphase mass transfer models can be inferred by comparing other calculated parameters to test data. This includes system pressure, bundle mass inventory, and cladding temperatures.
I 3. The system pressure in these TLTA tests was generally predicted well during blowdown until ECC injection began.
Thereafter, tha predicted pressure was somewhat lower than the test data. be now have evidence (described previously) that indicates this resulted from not including certain TLTA internal structures in the RELAP5YA model. All internal structures are represented in the Vermont Yankee NSSS model.
I 4. RELAP5YA generally predicted the trends for the bundle mass inventory in these tests; the predicted magnitudes were generally reasonable or lower than the test data.
- 5. The RELAP5YA Bh analyses generally predicted the trends l for the cladding temperatures, but overpredicted the peak cladding temperature durir.g the tests. The EM analysis for Test 6425/2 demonstrated that additional conservatism is obtained when using the Appendix K power and heat I transfer requirements.
- 6. Therefore, we conclude that the BWR LOCA licensing method (which includes the interphase mass transfer models) will provide conservative licensing analysis results.
I I
I 1
PWR SBLOCA Assessment RELAPSYA has now been assessed against three LOFT and one I Semiscale SBLOCA tests for PWR SBLOCA applications. These tests, the type of RELAPSYA analysis (BE or EM), and pertinent I results for the pressure, primary system inventory or vessel liquid level, and cladding temperatures are summarized below.
LOFT Tests L3-6 and L8-1 were two sequential SBLOCA tests that examined the system and core response with the Reactor Coolant Pumps (RCPs) operating (Test L3-6) followed by RCP termination (Test L8-1). Test L3-6 occurred from 0 to 2371 seconds. Test I L8-1 began at 2371 seconds when the RCPs were tripped off, experienced a core heatup as the two-phase level drcpped I through the core, and was terminated after 2464 seconds when the accumulator and two HPIS flows were initiated. A RELAP5YA BE analysis was originally performed and compared to test data in Sections 5.3.3 and 5.3.4 of Reference 1.13-1. A revised BE analysis was submitted in response to NRC Questions IV-4 and IV.9 in Reference 1.13-4. The revised analysis reflects our I current PWR SBLOCA guideline to set the ECC temperature at 200 F to limit the pressure decrease due to condensation following ECC injection.
I Figures 1.13-17 and 1.13-18 compare the calculated pressures from the original and revised analyses to the L3-6 and L8-1 test data, respectively. Only HPIS-A operated from 0 to 2428 seconds when it was terminated. Both the original and revised calculated pressures followed the data trend well during the L3-6 test period. The revised analysis shows I somewhat less underprediction beyond 1100 seconds, and is about 28 psia lower than the data at 2370 seconds. The revised analysis continues to follow the data more closely during the L8-1 test. This includes the rapid pressure decrease near the end of the calculation. This decrease was caused by initiating injection from the two HPISs (at higher than scale.d flow rates for LOFT) and the accumulator. This comparison provides I _ J
evidence that the PWR SBLOCA guideline on ECC injection I . temperatures can suppress excessive condensation for PWR applications. Figure 1.13-19 compares the calculated primary system inventory to test data since information on the core liquid inventory was not available. The comparison indicates that this inventory was predicted reasonably well, although son.ewhat lower than the data. Figure 1.13-20 compares calculated cladding temperatures to the peak hot rod temperature measured in the L8-1 test. The original and revised peak temperatures calculated during the core heat-up period are about 50 F higher than measured. Subsequently, the core was rapidly cooled by ECO injection in both calculations and the test. Thus, these RELAP5YA BE analyses followed the trends, but conservatively predicted the vessel inventory and peak cladding temperature.
LOFT Test L3-1 was a SELOCA test with the RCPs tripped at the start of the accident and accumulator injection allowed at the normal setpoint during system depressurization. A RELAP5YA BE I analysis was performed and presented in response to Question V.1 (see Appendix V.1-1 in Reference 1.13-4). Further information on the RELAP5YA assessment for this test is presented in response to Questions Q1.19 and Q2.1 of this do c umei.t . We believe most parameters were calculated reasonably well based on comparisons in the cited reference.
No core liquid level nor primary system inventory test data are available for comparison. The calculated system pressure was about 100 psia higher than measured at the time the accumulator I actually began injection. This delayed the calculated accumulator injection time by about 100 seconds and led to a somewhat slower injection rate as indicated by the accumulator pressure and liquid level comparisons (Figures V.1-6 and V.1-7 in Reference 1.13-4). The break flow rate was predicted reasonably well throughout the test, but tended to be somewhat higher than the test data between 350 seconds and I 1100 seconds. Therefore, we believe the primary system inventory might have been underpredicted. The core was well
I cooled in both the test and the calculation. However, since the calculated primary system pressure was somewhat higher than in the test, then the rods remained at a higher calculated I saturation temperature than in the test. Thus, we believe this RELAP5YA BE analysis also follows the trends, but tends to yield slightly conservative values for the vessel inventory end cladding temperatures.
I Semiscale Test S-LH-1 was a SBLOCA test that experienced core uncovery twice during the t'est. Recently, both a BE and EM analysis were performed with RELAP5YA for this test. These analyses and assessment are discussed in Appendix B at the end I of this document. The results show that each analysis underpredicted the vessel liquid level during the core heat-up periods (e.g., see Figure 4.10). The peak cladding temperature was underpredicted by about 40 K in the BE analysis, whereas, it was overpredicted by 67 K in the EM analysis.
Based upon the above assessment cases, we draw the following conclusions about RELAPSYA for PWR SBLOCA applications:
I 1. Each PWR assessment case involves significant interpMse mass transfer. This includes vaporization due to depressurization and heat transfer from heated rods and steam generator tubes; for the LOFT tests, it also includes the pipes, reactor vessel, and internal structures. This also includes condensation during per!ods of ECC injection.
- 2. No direct measurement of interphase mass transfer rates I are available for comparison to the calculations.
However, the impact of the interphase mass transfer models can be inferred by comparing other calculated parameters to test data. This includes system pressure, primary system inventory or vessel liquid level where available, and cladding temperatures.
I o
I
- 3. The primary system pressure was generally predicted well wnen using the PWR SBLOCA guideline on raising the ECC temperature to 200 F. This was particularly evident in I the L8-1 assessment when a large, atypical amount of ECC was injected at about 2464 seconds. This technique suppressed what appears to be an excessive amount of condensation due to ECC injection in these PWR simulations.
I 4. RELAP5YA generally predicted the trends for the primary system inventory or vessel liquid level where test data were available; the predicted magnitudes were generally reasonable or lower than the test data. We believe that the interphase drag models play an important role in calculating steam-liquid separation phenomena and the mass distribution within the primary system. Our experience, for ext.mple with S-LH-1, suggests that RELAP5YA tends to underpredict the core liquid inventory due to excessive ECCS bypass in the downcomer annulus.
I 5. The RELAP5YA BE analyses generally provided reasonable predictions of the LOFT cladding temperatures. The BE I analysis underpredicted the Semiscale peak cladding temperature by about 40 K. Using the EM power and heat transfer assumptions led to an overprediction of 67 K.
Thus, ve believe that these assumptions will compensate for certain shortcomings in the code for PWR SBLOCA licensing analyses.
I 6. Therefore, we conclude that the PWR SBLOCA licensing method (which includes the interphase mass transfer models) will provide conservative PWR SBLOCA licensing analysis results.
I I ,
References 1.13-1 Fernandez, R. T., R. K. Sundaram, J. Ghaus, A. Husain, I J. N. Loomis. L. Schor, R. C. Harvey, and R. Habert, "RELAP5YA
- A Computer Program for Light-Water Reactor System Thermal-Hydraulic Analysis, Volume III: Code Assessment,"
Yankee Atomic Electric Company Report, YAEC-1300P, October 1982. (Proprietary) 1.13-2 Letter, R. W. Capstick (YAEC) to D. R. Muller (NRC), " Response I to Additicinal NRC Questions on the RELAP5YA Computer Code,"
FVY 85-122, NMY 85-206, FYR 85-129, December 31, 1985.
1.13-3 Lee, L. S., G. L. Sozzi, and S. A. Allison, "BWR Large Break Simulation Tests, Volume 1: Experimental Results and Analysis, NUREG/CR-2229, General Electric Company, July 1982.
E 1.13-4 Letter, G. Papanic (YAEC) to J. A. Zwolinski (NRC), " Response to NRC Questions on the 'tELAP5YA Computer Program," FYR 85-121, November 1, 1985.
1.13-5 Lee, L. S., G. L. Sozzi, and S. A. Allison, "BWR Large Break Simulation Tests - BWR Blowdown / Emergency Core Cooling Program, Volume 2: Appendices I-N," .WREG/CR-2229, General Electric Company, July 1982.
I I
I I
I 1
I TABLE 1.13-1 Cross Reference for Figures I Current Figure Number Original Figure Number Original Reference 1.13-1 5.2-25 1.13-1 1.13-2 5.2-31 1.13-1 1.13-3 5.2-29 1.13-1 I 1.13-4 1.13-5 5.2-30 A.IV.1-9 J-5 and L-5 1.13-1 1.13-2 1.13-5 1.13-6 I 1.13-7 1.13-8 1.13-9 A.IV.1-9 and 5.2-25 A.IV.1-13 A.IV.1-23 1.13-2 and 1.13-1 1.?1-2 1.13-2 1.13-10 A.IV.1-24 1.13-2 1.13-11 5.2-23 1.13-1 1.13-12 5.2-19 1.13-1 1.13-13 5.2-20 1.13-1 1.13-14 5.2-36 1.13-1 1.13-15 5.2-41 1.13-1 1.13-16 IV.10-2 1.13-2 1.13-4 I 1.13-17 1.13-18 1.13-19 IV.4-2 V.9-1 IV.4-3 1.13-4 1.13-4 1.13-20 IV.4-4 1.13-4 l
I lI 1
l l
I
!I
'I I ,
M -
0 1
3 ENN MEE OLL .
DPP HRR t s
AEE e EWP T -
iOP SLU A
. T L
T 0 n 3 i 2
e 24 . s l
n 1
PPP o p
s e
R N o n
U eo ri R = ut 1 23 sa sl
) eu c r c e Pl M S
( mC a
6 ' 0 e 2 5 E t A 4 1 M sY yS I
6 T SP A,
T f l S oE R
E n T oo A st Y i S r1 P a/
p6 A.
I m2 E o4 C6 y 0 :
7 1 3
1 m # 1 e
r u
A g T i A F D
0
?c m_ 1 0 0 0 0 0 0 0 2 8 4 1
92 8 d R 3:an.
1
- l;i I! l i' I u
0 1
3 1
d*
E -
L u D AJ N
U M B b-t s
e T
M A T
~ 0 L r 3 T A 2 S n i
S I
A b- -
y M - r o
- t n a so N ii U l-I I t a
R )
sl 1 su ac Ml M
6 b d 0 5
(
c e
)S l dA eC nY a
2 1 E uS d M BP 6 I A T f L oE T R S n E oo st T i r1
~ a/
p6 m2 o4
- C6
_ d A ^ 0 2 T 7
.M A D
- 3 1
A = _
1 Y _
5 e P r M A L
u g
_ E i
_ R F
[ni e
_ ' 0
- 1
_M 0 0 0 0 5 1
M
_. adv n e
u
g g
0 1
3 a 2 1
0M P
. E 3 T
e T A
n W i S >
a N
" n 1 o 7i 1 E 0 t 7 L 3 La 2 El u
m I L
ER I
E
)
t c al a
eC 7W r 0O uA m N O U P SL t Y aS rP R ' eA pI A mE I 2 T eR m A D
)
c e
T dt a
o S
6 2
g ' 0 5
(
E l 1 C/
6 t
1 1 M f2 I o4 6 T 6 n
T ot
) ss S T i e M E I
A rT a
pA A mT T oL A CT M )
D A 0 :
1 3 Y -
5 3 M P. 11 j 0 7
1 L 7 1 E
R L e M
E
( M r i
u g
F M
M q 0
- 1 0 f 0 0 0 0 0 2 E 0 4
1 0
1 6
_ C v gEdwh E
E
!! I l: ,< l I i !
M M
- 0 1
3 3
M 61 5
P 00ME .
l T ,
F M PT R
S n
i S ,
M "
7 n 0 o 77 1i 00 0 t 1 1 3 La 2 El M LL EER , t c u
1 E al 69W a A eC 03O T r M N OO SSL A D
uA tY a5 U PP R
rP eA pL 1 23 mE M A T )
c T
eR o
A e dt D S a 6
2 g 8 0
( l 1 C/
M 4 -
5 1
E M f 2 6
6 =
I o4 T 6 T n ot S ss M E i e rT T A a Y pA
, 5 P
mT oL N' A L
E CT R
M ,
4 3
0 7 1
_ 1
.M e
r u
g A) i Y "0 S
F M P A
1 LI E
R 2 1 s W 1 L
E A _
0
( ' ,- ' 1 0 0 0 0
_ M'
_ 0 4
1 0
0 1
0 6
0 2
M n8
=
M m
1 l l
- ;l ;
il l l l1
lj M
M :5EiS.5 a e
s s
a o, M _ 3 GNN NEE OLL M DPP .
NRR NEE EWP T OP M SLU
$. 4 I ; m M t 2
a 2 PPP .
M N s
e r
U u R s
. s i23 _.
_ e r
P M me i
t t 5 - s a s 2 e y 4
wl i u S M 6' - e i :
T 5 T h ? -
M S
E T ,
A-f LI y -
w 1 1
3 e
I r N u g
hC :_ O. i F
M M
W l s
- j A
- =
Y S
A P Y A 5 L P E M A L
R E
R M Lg JL o s
m =-
s - -
M e
- -53t S.5 8 M
M M
M 1, l1 ll 1lll .1l : li 1 lll
I I
3_3.
g . ., , _3. .
i i i ..
~
I WZZ rWW OJJ I O G. C. mn E22 CC W W W30 I >-- O 0.
mm $
I
- ~ ,
U "N7 m a 111 c
E
(
-~m n
- n I 3
=2 5 - -
i,;:
I \ Y T '"
2-xx o
f %g c0 I 3, n_
E W e N '
.i0 n*
l N I $$ ~ $$
u-sw 6" g
I _
n 3
- a
~
4
~
- c I ,
l R
~
R
(**9 EWnS$3W4 03WASV3M H
I I
I
M M M M M M M M M M M M M M M M M M M 00 TEST 6t4 25 RUN 2 1 P1 STEAM DOMG
- g. t 2 P2 LOWER PL CN
~"
RELAPSYA -
m p
- h3 P4 UPPER PLEN N
A f
I 9 i' =
b i ta i $ l d
1 400 Ijjh 5 -
R5YA TLTA 6425/2 Cale
~ -
I
R5YA TLTA 6426/1 Calc
? STLTA 6425/2 Data
- a ,
REIAPSYA - C - *
---p_._________
0 -
-10 70 150 230 310 i TIME (Seconds)
Figure 1.13-7: Comparison of RELAP5YA System Pressures i for TI.TA 6425/2 and TI.TA 6426/1 I
I I
I ~
g - __
I ? . _
+
l m, r a q I m y
- h-a 'C I _ * * '* * *
- w=== . .e
~
w E ,
i%
N 5
m "g
_M .-
C e h !jy
.3 I $-- _ 7
- 2
- Y .
< k E - < -.
4 e
a
< E t g
f 5
- a. .
I 3
I E
- I 7 ^
.,i 3 I MI SEvn I
i I
I 1
I lr ~,_..,_,,
I .
i l ccecm-
- c. c. c. a.
b .:
I EM3WE g
EEEEE I
~
,dddddx "
8
' 95sE58 s 5 l 5EEEEE' 5 gataca y . 4 I .~ n = m e a-3 E
I g
t a
2 w
?
J - s
-1 E
{
n E w [
') E q " G 8 a .
- 3 E ._
.G Ch E ..
"G@@@@@@GT " ~
2
@@@@@@@@ l
.M &
@@@@
- C E @@@@@@@@
l #eeeeeeeg a
f 4f fwNvw.m3A I
E I
E I
I E . ~,..w,_,.
I o.
n s*xmm . ? .5
- -. o - ,_
I t t t t >-
"\
.EEEE I
d oooo5 ) ! E
$22$d )
I ~ddddeu C- @5 3
't ZEE 88' I gacem
- m m -r m o
a p'
9 e
s 8
- - t I! $ 5 _
s 1 l 0
, -I ;
I G w
5 u b G
o l s l "@@@@@@@@
g J
3 a
I ,
c I @ @ @ @ @ @ @@ @@
Q@@@@@@Q ,
I R
9 f
is.l sunavu,m st I
I - - . _-
M M
I I 0
1 3
M 1
E L
M D N
U B
M ,
t s
e T
0 A M 3 2
T L
S T S n 2 A A i Y
M N A T
5 P
y n r o A A o i N D L t s
t a
U l E R i l M R l l u 1 c
_ s l s a A. M a C
_ ) M M 5 m 0
c e l e E 2 5 S d A v 1 ( n 4 . u "
6 E B "
M M I f >
l I
T o E S n R
L A o o I s t M - i r
a /
2 p 5 m 2 o 4 C 6 M :
1 1
0 3 M 7 1 1
e r
u M i g
F I
M a M 0 1
0 0 5 0 _
1 1 _
M E$
M M
1 ! ! ,. Ii f i!l ; ' l
I I
I e n
fJ "J F4 M =
I -= tu a ?. ?. 1 C C C C lll: E ,
. . . . .y I
- - C - - --
- A 'A L *e A C. C.
- L 1. g Di
~
N ZZ a >
z l cc C
6 S5566 J \ >
~
"3 j
\ ' "N
~~
= Jaaaa l
N u w w _: x a af a ^ W-I 6 C: - C'3 = 2
- 7.7 O - ra 7 'A O
$$$00d
7 7 7 "o
>+8 u
G.
yC g<
y m Em I
un
- ?a n = x x b$
3~4 W >
"5 15 8 !=
I e N
~
s N
c
~
S W
5 1"
U"
" 0
< {
I E
- $ ~ g{
g a
%~
\ I$
- @@@@@@@@* r e I
I @@@@@@@@@
s
^
m
~
~
- s I '
+J b C
9 g
.eeeeeeee,s
@ @ @ @ @ @@ @G j
= u I
4 8 8 8 8 n
C @
4 (3o) 3.MW3OU1 I
I I o
~
T I O_ V A m .,.=s z ]
50 ...wcon '
I f
Mo- <w '
IJ= u. .a. u.
LLCW ko g C
I .2u32 mmzz Z
, je I
C000y ; O o N N *U N .,J n : -
____~
~ og aawa w u y u e:
- a5
~ m-omy I --=eo COCO
~
"o
- w~
au G.
oc
$m m m m .s -
gzecz s<
I -mm=m s +'
0"
- f Ma en m- 9 im I e m
Ed-y e
r
~
a 1"
e
~,
f w 5 c~
I -
e
~i r
0 t !;
I d is 5 is I *@@@@@@@@* '
as 0
l
[ ^
A
]
o o
I @@@@@@@@@
@@@@@@@e ,
t s
3
+e e e e e e e e,, </
q .
g l E f I" .' W O
2 o
- s. s
e s.
n (30) E01W. IOU 1 I
I
M M
M 0 0
0 1
M D
M E T
A N A I T TM L SR T M '
EE TT
% 0 i n
' 8 e 5
e J. 5W, 7
s n
M - L on po si l
- et I
J R a l
_U . eu M h E f' rc ul C sa S sC A e T
L
) rA M T E M
c e
PY S
1 O
D
" S
(
mP e A.
I 0 t l N M C 0 E sE U A S P 5 M yR M R
/
2 3
E T
S CL P
L I
T S
f t o
6 4 1 P
o 1
T n/
S E A o2 s3 T Y 5 i 4 P r6 A a L pt A S E ms M Y 5 NE R oe
- CT P P :
A L
O 4 S 1 1 E O, 0 -
R A 5 3 M -
i 2 1 E
'l 1 1 N
I c
r L O ME E u M A VS E LO o- i g
N N I O YB FO E P
M SD OC V E E EEL R NN SRA UV YR OS LE PS L RVSEO M , AOERR EGRPT 0
0 0 0 0 0 0 0 M 2 1
8 4 M Q4EV U8m v
M i
g E
0 0
- - - 0 1
E A
T L
0 T I
5 7 n E u in o
yi rt A oa tl su E i c l
l l a
sC s
aA
. E C S
MY eP S
A ) l A T P c dL L j e nE it T 0 S i E
l I
0 ( h t
5 o 1 E f t N M o U
R
- I T n/
1 o2 E /
2 3
4 s3 i 4
_ 6 _ r6 T
A a S pt Y ms
_ E
.E T S
S P
A CT oe N
L :
E E 5 P R 1 O -
s a 3 o T 0 1 A 1 5 .
2 1 A e r
A Y u T
A 3~ S P i g
D _ A F
_ L E
R T--
0 0 0 0 o
-- 5 0 5 1 1 3O v mpx li' , ;', ' i ', ,l I !i! . I1ij' ; l Ij) .: 1 i l
I I
I I I !
i i i -l l
r e
iEii5*
EEEEE>! /
I W a=arIs --
4 JI
,s
=====1 Ti I :"
j didids 1 2
%4 g ;;
E!ES Eg 2.
I e y eeeEea
- ~ e . ..
g 2
c
- 3 3
- 53 5.
9 l ~
2
- !!3 s3:
\ - 3es I -
Rs; OE4 met oHe enu I "@@@@@@@@*
aas
{U8
@@@@@@@@ 8il I @@@@@@@@
]~"
T
!l @@ @@ @@ @@@ @@ @@ @ @ 9 2
@@@@@@@O $
I #GOGOOOOQ Z I
,I i ,
l
! ! i i u.ii navvimu I
I
e E \
I . . . . . . . . . .
a af f, I
v-aE=s
- T ec ~
I U
~
hh t
~
So
~ E
.5
. e
.g aE I $
5%
e*
s?
-l 30 I C-
- E5 o U N I. a- ~8 3il R
o 55i
- n =2 l -
g i
l
.l - -N ~
5.5 EI uo o
~
~Y I -
-i i$
o 5 I . .g d
. .g l
l
.d O*EE5 0*DiXE 0*0be! 0*0ds! 0*CE61 0*Cbt! 0*0but 0*be 0*d31 0*Cb6 0*dz 0*O (WISd) 380SS38d 03710H d00110WINI II L
I t i M .
M ,
M
- - - - - b - . W 5
3 S
. N3 M 0 1 3 -
. N2 8 L
M O.
Ts F n o Do I i
. d3 t
_ na il M - _-
0 0
ec u
rl 5
4 ua 2 sC MRT A
D .
0 0
5 e
s eA rY PS P
3 m eIA.
1 MLT- _- t E 0 sR 8 -
y 0
4 ) Sd
. 4 e 2 Cys E ri F
S av MOL O
T
- 0 0
3 4
2
(
E riR M Pdn I
me T f a N 0 MI0 S
R A
0 2
2 4
ig o
na on si l
P 0 ri MC ar 1
r 0 0 .
1 pO 4 m 2 oo R 0 Ct Y
S - 0 :
MP 0
a.
L 0
2 8
1 E L 0 3 R DA 1 0
EN e .
SI 3 1 N-1IG 2 M _, 8ER VI 0
e r
r LRO u
.o s g TRR 3 2
i F
FYY O55
,- LRR 0
. =-
v
- a 0 3
7 v 2
- e f
.M 0
5
. . 3
_ 2
- e 3 5
M u ~ . . -
3 2
- 0 9w o42 adN 90 a R~ 9g 9 R" *o E_mS u=aMm=a_ c.w' *E n_ 8Hz" M
M l
l, 4I '\I< i1: , ,;I
,1,l1; ill ,lli! 4i -
! 1 l l M
M
-M 5 i :-i !iii5i5i:-=!E E E iE s):iF 5: Fihp59'0
- $c=
3 2
- i M ' '
d b
'i O
' ; - i W '
DA EN L ;
d 6
3 SI ;
' b L
'i 6IG 8
' VI : i T F s
' 3ER :
On LRO ' Lo i
TRR : d nt FYY I ' b ia R *i O55 s yu l
W T R M LRR .
= i r c ol t i.
OE' '
nC T' M e W
8S' 3Y' L S '
i I
h '
6 b
i 4
vA nY I 5 m
P e A.
Y I.
TR' F
' t F sf 1 O R'M' 6) y L I b "..d 2 C r] '
e i
ys O R'P' (ES ri T ' av
(
me N E' '
W O H' S T' '
6 E b IM iR r
Pd n
I O T f a R IN' '
i i o A '
1 n1 W P S' M
O S'
' on ni CM' R
'o d
i g ri ar RL
'i b pO W YR' S T' P O m
oo Ct AT' L ' 'O :
E 'd
- W R
'i b
1 3
' 1
'0 1 W 'e
' 6
'4 e r
u g
' i
- F W 'i 11 'd
'O k
W '
E 5 _ l'ij EiEi2=i15i5 5 ai :i5 E:1: 0 i i i. : o
- i3 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 D 0 0 0 0 0 0 0 0 0 0 0 0 0
- 5 4 3 2 1 3 2 1 0 9 8 *1 6 1 1 1 1 m -red ~ meE a$OW I 1 (I ll l l-
I I
I .
n a e a a e a a e l
- g .1g - s i sgg 4g I Eli .
5 3 ine c I o i/ j 3
Ea o
.g *f
), ~
?8 I ==4 f k$
6 -
- < , 33 3 -
s c.:
g _N o. T I)' O o 5 WhO>
g- ! "j I w
- Og l i- .
a
-i j c O ~g u u to eo 9 .
s a
~. y
- d u.
o
- l l g l 0*0321 0* DOS 0*be C* dot O*d[B 0 *ms 0*$N 0*dtE 0*dE 0*dIt 0*O (J) 3801983dW31 SN100 0 0 930-BSS 300N l
I
In the answer to Q.IV.9 on the verification of the ability of I Q1.15 REl.AP5YA to calculate the various modes of natural circulation (N/C), YAEC provided the results of RE1.APSYA calculation for Semiscale Test S-NC-2. REl.AP5YA did a reasonable job of calculating the N/C flow rate at the 30 kW core power level, but overpredicted the flow rate at the 100 kW core power level. Provide additional information explaining why the flow rate at 100 kW was not calculated well. Also, YAEC stated oscillations in the calculated N/C flow were attributed to unsteady calculation of the void distribution. Provide I additional information to clarify what is meant by the phenomenon, what causes this in the calculation, and its potential impact on SBLOCA calculations.
A1,15 We believe the reviewers are referring to NRC Question Q.VI.9 rather than Q.IV.9.
The 100 kW case was re-examined in order to address the two questions above. A slight error in one elevation change I involving a four-inch discrepancy was discovered and fixed.
Also, the feedwater flow rate to the steam generator secondary side was adjusted to maintain a nearly constant inventory, as in the experiment. The 100 kW cases were then reanalyzed with the most recent version of RELAP5YA.
Figures 1.15-1 through 1.15-3 cortpare the new calculations to experimental data for three major parameters (primary pressure, hot leg temperature, and downcomer flow rate) as functions of I primary system inventory.
The calculated results are nearly the same as before. Although the calculations follow the data trends, the calculated N/C flows are still higher than the experimental data at inventories below 90 percent. The oscillation observed at 65 percent inventory in the previous analysis now occurs at 60 percent. However, the oscillation amplitude is slightly less than before.
I
I I The Issue of High Calculated Mass Flows The calculated flows appear to be " shifted to the left" when compared to the measured mass flows shown in Figure 1.15-1.
The calculated pressure (and therefore, saturation temperature) is higher than the data at these lower inventories also.
The S-NC-2 Quick Look Report (Reference 1.15-1) discusses flow sensitivity to operation of the external heaters. The external heaters were used to simulate adiabatic conditions in the I Semiscale apparatus. Figure 1.15-4 (Figure 12 of Reference 1.15-1) shows the ef fect of these heaters on flow for the experiments at 30 kW, 60 kW and 100 kW.
Quoting from Reference 1.15-1: " Figure 12 indicates two values for loop mass flow at about 79 percent inventory for the 30 kW case and 77 percent inventory for the 60 kW and 100 kW cases.
The higher flow rate points in each case represent the natural I circulation flow rate with all external heaters on, and the lower points represent cases with cold leg and pump suction heaters off. In addition, all points below the 77 percent and 79 percent mass inventories represent cases with cold leg and pump suction external heaters off."
For the 77 percent inventory 100 kW cases, the measured flow is about twice as high with the external heaters turned on as it is with them turned off. Similar sensitivity was seen at the other two power levels.
Boiling was visually observed in the cold leg at the 77 percent inventory. The cold leg (CL) and pump suction (PS) heaters were shut off at the onset of boiling in the pump suction.
This meant all cases below 77 percent were run without the heaters. "After a few minutes, the visual evidence of boiling dissipated and the test proceeded. "When external heaters were turned off the effect on the loop mass flow rate was a sudden decreased flow ..." (Reference 1.15-1).
The large impact of heater power upon experimental flow rates 3 implies a high degree of uncertainty in the heat transfer boundary conditions for this experiment and, consequently, a large uncertainty in the flows. The uncertainty is especially important, since the PS and CL heaters are supplied with 20.8 kW while the vessel and hot leg heaters are supplied with 11.2 kU. The external heater power or core heat losses represent a significant fraction of the 100 kW core power.
The RELAPSYA heat transfer model also assumes there is no heat exchange with the ambient environment. Unfortunately, it appears that this assumption became invalid for core inventories of 77 percent and lower. The higher calculated pressures and flows imply that the effective heat transfer area I for removing heat from the primary system is smaller in the calculation than in the experiment.
4 It could well be that heat losses in the experiment increase the effective heat transfer area, thereby violating the adiabatic assumption. Considering the great sensitivity of flow rate to external heater power, I the uncertainty in the heat transfer boundary conditions could very well account for differences between the calculated and measured flows.
The Issue of Oscillatory Mass Flows The calculated primary system flow rates began to oscillate during the final stage of a 100-second drain time from the 65 percent to the 60 percent inventory case. Thereafter, the I system continued to oscillate with a 27-second period for 500 seconds. This oscillatory behavior was not calculated for the 65 percent nor the 57.5 percent cases.
The harmonic oscillation is caused by liquid velocity differences within the steam generator tubes for this void fraction range. The upside liquid velocities are lower than the downside values due to the action of gravity on the opposite velocity directions (similar to a roller coaster).
t I
Ihis causes some liquid to accumulate in the upside, and a decreases the liquid fraction in the downside. Eventually, this condensate is carried over the U-bend to the downside.
This causes a temporal imbalance in the density distribution around the primary loop and increases the flow rates until the imbalance ceases. The process repeats in a periodic manner since the downside liquid drains faster than the upside liquid reaches the U-bend. The 27-second period is approximately equal to the upside tube length divided by the average upside liquid velocity.
The calculated oscillatory behavior may be a consequence of lumping the six intact loop steam generator tubes together with one average height. This steam generato: actually has three groups of two tubes with heights of about 337, 365, and 391 inches. Thus, each group could have a different upside liquid transport time that would lead to less coherent flow behavior and dampen this oscillatory tendency. Perhaps this explains the lack of an observable oscillation in the S-NC-2 I test results.
We also examined the capability of the S-NC-2 test procedure to i
detect potential oscillations. Three procedural limitations became apparent, any of which coulo have prevented detection of an oscillation similar to that calculated.
I First, data points are only taken every 40 seconds, while the calculated oscillation has a period of 27 seconds. Second, the I selected inventories for steady-state cases are spaced five percent apart in the vicinity of the calculated oscillation.
The large oscillation calculated at 60 percent was not calculated at 57.5 percent nor 65 percent. Third, the average test time per inventory case was relatively short. Based upon Reference 1.15-1, the 100 kW test series lasted 100 minutes.
I
E c During this period, 18 inventories were tested. Therefore, the I average test time per inventory is 333 seconds. This includes drainage time, stabilization, and data acquisition time.
Possibly, the oscillation did not have time to develop.
The calculated oscillation occurred at an inventory sufficiently high to have a negligible impact on PWR SBLOCAs.
The Semiscale bundle was calculated to be well cooled by nucleate boiling at the 65 percent, 60 percent, and 57.5 percent inventory cases. The authors of Reference 1.15-2 I concluded utat adequate core cooling can be sustained if the system mass inventory during small break transients is about 35 percent or more. We agree with this statement based upon our pump trip studies for Maine Yankee and the Yankee plant at Rowe (References 1.15-3 and 1.15-4).
References 1.15-1 Loomis, G. G., K. Soda, C. P. Fineman, " Quick Look Report for I Semiscale MOD-2A Test S-NC-2," EGG-SEMI-5507. EGLG Idaho, Inc.
1.15-2 Loomis, G. G., K. Soda, "Results of the Semiscale MOD-2A Natural Circulation Experiments," NUREG/CR-2335, EG&G Idaho, Inc.
1.15-3 Schor, L., S. Haq, A. Husain, J. Ghaus, " Justification of Reactor Coolant Pump Operation During Small Break LOCA Transients - Maine Yankee," YAEC-1423, April 1984, I 1.15-4 Loomis, J., J. Phillips, A. Husain, " Reactor Coolant Pump Operation During Small Break LOCA Transients at the Yankee Nuclear Power Station," YAEC-1437, July 1984.
I I
i I
I
! SSN02 - 100KW - MRSS PLOW
- 1. 2 -
i*1- LEGEND l 1 g, l'EfrE?acuum 0.9 -
I -
(n 0.8 -
s E
N 0.7 -
I c.D d 0.6 -
0.5 -
l o d
r 0.4 -
8 Es n I (n (n
(C 0.3 -
0.2 - E b
l 01-0.0 -
, a a
-0.1 -
-0. 2 -
I -0.3 50 55 00 65 70 75 80 85 90 95 100
% INVENTORY
! rigure 1.15-1: oowncomer nass riourate I
I I
I I
I !
I I
I SSNC2 - 100KW - PRIMRRY PPESSURE 13.0 12.5 -
I 12.0 -
LEGEND f-- " " - m
.1 5 -
f 11.0-5 10.5 -
a g 10.0-
, g g 9.5 -
3 $ 9.0 -
E 8.5 -
y 0.0 - g E 7.5 -
5 o
7.0 -
- : : : _ m 6.5 - , , a a m 8 6.0 -
5.5 -
5.0 , , , , , , , , ,
50 55 60 85 70 75 80 85 90 95 100
% INVENTORY g
,1 1.1,.m ee _ , e,. _ ,,. _ ,.
I I
I I
I I
i l 620.0 SSNC2 - 100KW - HOT LEG TEf1P 615.0 -
LEGEND 610.0 - [:hmmm 605.0 -
d 600.0 -
g 595.0 - a 5 590.0 -
sur E 585.0-580.0 -
% 575.0 -
cp 570.0 - a 565.0 - ,
52.0- ,
a , , ,
I " " "
555.0 -
a a 550.0 - e 545.0 -
50 Sb 6b 6b 7b 7b 6b 6b 9b 9b 100 l % INVENTORY Figure 1.15-3: flot Leg Temperature I
I I
I I
W W W W M M m ma g g g g g i
i i
i O- 3 5 3 gss8agI I I 5g5 II 5 g5 5 5 5 giI s ga33 5-gl I EgaI eag3 BS 5 5 a -5 0.3 5- A 5 l j O Case 1 (30 kil) g A j AB E E A Case 2 (60 kW) i 4
j 0.7 5- a case 3 (100 ku) a E E A =
A -E Ao 0.6 E- A o A
=
m O E
=
i a E
=
a -E s A O i ;
. 0.5 E- BGB E l
5
= a O F
=
w A =
i
! 3, 04 h A S
=
s 5
d 0.3
=
=
E-- 0 %g0% =
l = a =
m = O =
', m E a A -E E! 0.2 E- =
=
~
=
~
0 =
O1 E
E a A A O -
!' m eeIeeet leaenleae:
a ead ki e d i.- edie iI a t$ I e 0.0 -
70 75 80 85 90 95 100 i
55 60 65 i
MASS INVENTORY (% of Total) i
! TEST NC2B Figure 1.15-4: Effect of Mass Inventory on 1.oop Mass Flowrate 1
1 i,
4 i
f
l I i I Ql.17 Q.VII.9 Q.VII.27, and Q.VII.30 requested YAEC to compare the range of data in the different types of assessment calculations (core level, CHF/ post-CHF, and reflood) to those expected g during SBLOCAs. The YAEC response only provided the range of assessment data and did not provide a comparison to the expected SBLOCA conditions. Provide this comparison for both BWRs and PWRs for our review. At a minimum, it should compare system pressure, power levels, mass fluxes, reflood rates, linear heat generation rates, qualities, inlet subcooling, and I
outlet superheat as appropriate for each type of test.
In addition, for Q.VII.9, the tests used for the core level assessment all had pressures below 650 psi. Are there expected SBLOCA conditions where a core level may form at pressures above 650 psi? If so, provide assessment calculations or justification for omitting assessment for the core level at these pressures.
A1.17 I
The ranges of thermo-hydraulic parameters encountered during SBLOCAs are very much dependent on the size and location of the break and the type of plant analyzed.
Three cases were chosen to illustrate typical conditions calculated by RELAP5YA for SBLOCAs at the Maine Yankee, Yankec at Rowe, and Vermont Yankee plants.
I For the Maine Yankee plant, we are presenting results for a 2
I 0.05 ft break in the cold leg. The pumps are tripped two minutes after HPSI actuation. Details for this break are preseated in Reference 1.17-1. Figures 1.17-1 through 1.17-9 present system pressure, core power, core inlet and outlet liquid velocities, core qualities and void fractions, lower and upper plenum temperatures, and reactor vessel collapsed liquid level. Similar thermo-hydraulic parameters for a one inch ID cold leg break (Reference 1.17-2) are presented in Figures I
I 1.17-10 through 1.17-17 for the Yankee Plant at Rowe. The main reactor coolant pumps for this break are tripped at 92 seconds in the transient when the primary system pressure is 1,250 psia.
2 Results for a 0.05 ft break in the recirculation loop discharge pipe at the Vermont Yankee plant are presented in Appendix A.l.9. Additional parameters are shown in Figures 1.17-18 to 1.17-22.
For the second part of this question, please see the assessment I of Test S-LH-1 in Appendix B.
References 1.17-1 Schor, L., S. Haq, A. Husain, J. Ghaus, " Justification of Reactor Coolant Pump Operation During Small Break LOCA Transients, Maine Yankee," YAEC-1423, April 1984 1.17-2 Loomis, J. N., J. H. Phillips, A. Husain, " Reactor Coolant Pump I Operation During Small Break LOCA Transients at the Yankee Nuclear Power Station," YAEC-1437, July 1984.
I I
I I
I I
I I
MY EM CASE S12 g 0.0S FT2 BREAK EM 2 MIN TRIP 3
O I 8 I -
- g. . _.
L t
I o s P 90B010000 .
Ei-o--eP 610010000 I e-i3
),
A P primary o
I d, .
I tc 1
i
~
= :
P secondary k
I .0 21b0.0 #*
0.0 7 Int (SEC)
Figure 1.17-1 Primary and Secondary System Pressures I
- 1. b o.
78 1 -
o
\ i-I no
{ !!-
s l k r u - l I
' a:
I o
L I d c.o 700. H00.0 2@.0 e.0 ==o TIME (SEC) j Figure 1.17-2 Reactor Power I
I MY EM CASE S12 0.0S FT2 BRERK EM 2 MIN TRfP a
E #
I j. "
ll El
, js- % w lpj f id) # d i '
e I s'@
d m-e VELFJ OUTLET I o g-e-eVELFJ INLET t
l i
I ' O.0 7d0.0 1500.0 TIME (SEC) 2160 ae00.0 350c.0 Figure 1.17-3 Core Inlet and Outlet Liquid Velocities I
I u xv I
'O d' N '"-" I a-oMFLOWJ OUTLET I Q e -eNFLOWJ INLET C
l $- -
i 1: m b
I e. . ,
- . . ', s .
ti I 2150.0 2000.0 3500.0 O.0 7d0.0 1400.0 TIME (SEC)
Figure 1.17-4 Core Inlet and Outlet Flow Rates
NY EN CASE S12 0.05 FT2 BREAX EM 2 MIN TRIP I '
l(
i
. , t ,
s i g- ,
y . .. .- . ..
f n g .
,,f o
y .
i . .- . . .
{ .
b I,, r g , .
.. S., . . '. _
I l g l e- 100010000
! e eOUALS 100040000 l..ln
- EQUALS
'[ *QUALS 100050000 g
I y- '
O Li h .
,c i
4,.
I, IC 9
O
. n -
j i[t :
0.0 700.0 t 400.0 2100.0 3000.0 3500.0
. 3
"' 'S' '
E Figure 1.17-5 Quality in the Core 1
I I
3 ' "
$ T r Ib 1
N
, ,j ... . .
.- e a
l 3 -- I ,-
i i g l .
u i .
- , f u a,, o ,7 g ..
. . - . . a I.p ? .
j, q p41" <
[ :, k,i I .
a <
mm __
=
Figure 1.17-6 Core Void Fraction
MY EM LhSE S12 0.05 FT2 BREAK EM 2 MIN TRZP o
g t .,
I ;
,1 Slit I -
e - oTEMP 23010000
.. o S i e-eTEMP 720010000 I W W
I h g5 inlet I 5-J-- ~-
p i o I 0.0 700.0 it00.0 2100.0 2800.0 3500.0 TIME (SEC1 Figure 1.17-7 Core Inlet Temperature I o I
n
!]'
o -i il
- u. Outlet e - eTEMP 46010000 t
. i: t j
I ;-- @- e- cTEMP 720010000 ii g i .
3 ,
if 0,, ,I ,!
. !. . .. .. . , g .- -
4 e i 4
)
s , ,
c d j ,
p ..
a< i .i .- -- -
u -\ - - - - -
a l D i
/
sat LL _ y_
..a
~
A l i
I i 0.0 700.0 1400.0 21d0.0 2850.0 3500.0 TIME (SEC)
Figure 1.17-8 Core Outlet Temperature
I I
I MY EM CASE S12 l 0.05 FT2 BREAK EM 2 MIN TRIP
, , , , , . .--' " ^ ^
M .
i I .
C. .
k
- o E N i:i a .
O .
a e?,
a I ~
I -
g s
'y -
o , -
o 0 l 400 g00 1200 1600 2000 2400 2000 3200 3600 TIME (SEC) rigure 1 17-9 Reactor Vessel Collapsed Liquid Level I
I I .
I
- I
YR PTS Et1 CASE 1X4 ONE IN. 80 OL BRERK i 1
I i I is _
r I o m- eP 906010000 ~ ~ ~ ~ ~ ~ ~
e- eP 45010000
. S!l--
i I R u
9 'k P primary -
I !
P seconc ary I 9
$ M0 saa.o 1200.0 1800.0 o.O do.o TIME (SEC)
Figure 1.17-10 Primary and Secondary System Pressures I bo x d I
I :j I -
I E
- a. \
o c4 I.
I o 0.0 400.0 Edo.0 TIME (SED) 1200.0 1U.0 M.0 Figure 1.17-11 Reactor Pouer
YR PTS EM CASE IX4 ONE IN. 10 DL BREAK a
d E Di 0- -
I -
8 a
us E '......... ."; r;T ^-22? g,g -j -
I 3 .
, i_
I b
. . . . . . . . . . [
s-eVELFJ INLET I o o
e-e VELFJ OUTLET 4
o Y 000.0 1200.0 1600.0 2000.0 0.0 400.0 TIME (SEC)
I Figure 1.17-12 Core Inlet and outlet Liquid Velocities I
E ux-I
, i.
I o-e NFLOWJ INLET e-e MFLOWJ OUTLET 3
s I
S q g ; _ ___ .p; ,o
~
7_
3 I o E
I ;-
0.0 400.0 800.0 TIME (SEC) 1200.0 1800.0 2000.0 Figure 1.17-13 Core Inlet and Outlet Flow Rates l
YR PTS EM CASE 1X4 ONE fN. TO OL BREAK I
E l I _
d- --->
c I U m - eOUALS 100010000 o--c CURLS 100020000
- -* QUALS 100030000 l
- 4 '
M l i j d- 4< li.
y m. ,,, t, a k , .~ _ m . a N' $ I 5 d. _ _ . . .
0.0 400.0 800.0 1200.0 1800.0 2000.0 TIME (SEC)
I Figure 1.17-14 Quality in the Core I
I e
- ap' o i
,i c'
c 4 n ;[l
} .., [ .. .
ll I 0 - eVO10G 100010000 J
oo .
2* 100020000 I + -V010G
- e VO100030000 IDG y d' *- ,
E l
- g. - o _
I ; _ p {g !g -
I 0.0 500.0 tMdT' #@
000.0 TIME (SCO) 3200.0 1800.0 2000.0
, Figure 1.17-15 Core Void Fraction
YR PTS EM CASE 1X4 ONE IN. 10 DL BREAK I -
grv ,
I o Acq, g e-4\
E d 5
p
. I I ! l Y6 g.
a o
a 0.0 4d0.0 800.0 1250.0 1800.0 20C0.0 TIME (SEC)
I Figure 1.17-16 Core Collapsed Liquid Level I
o N
E g.
~
E \
OUTLET c h--- INLET g iihJ :
, sat , inlet f
,e- d'l' ( -1,.gfg d-
~
sat outle t e i I N*
0.0 4do.O 800.0 1350.0 1650.0 2000.0 TIME (SEC)
Figure 1.17-17 Inlet and Outlet Subcooling
i VERMONT YANKEE NSSS LICENSING MODEL CASE EWa SMALL BREAK LOCA RPPENDXX K RESULTS l
! RECIRC LOOP DISCHARGE PIPE BREAK (0.05 FT2)
{ Dx~
l i
, s 8 s:
b_
N e,o , ,
!S f
- g
- - AM .b=m gge[hd ,
s I o a
0.0 90.J 180.0 270.0 380.0 450.0 TIME (SEC)
I Figure 1.17-18 Core Inlet Mass Flow Rate I o.
I I a U
s
~
.o e--cLOWER TIEPLRTE I ,g$
d o--e UPPER TIEPLATE i
,1 o
i I d gi i n.!
y- -r -
f..- _-[ i i .J ei d E
I "o g g --
ei I o
' O.0 95.0 18'0.0 2/0.0 360.0 450.0 I TIME (SEC)
Figure 1.17-19 Average Power Assembly
- - _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ . _ _ _ _ . _ . _ . _ _ _ Liquid Velocitles l
VERMONT YANKEE NSSS LICENSING MODEL CASE EW SMALL BREAK LOCA APPENDIX K RESULTS RECIRC LOOP OISCHARGE PIPE BREAK (0.0S F'T2)
I a I u, E'
I .
a - cLOWER TIEPLATE e--cVPPER TIEPLATE g jj tyt; o _ .. , - q --
xr vvp em N .
i o - - -
9 s
I :e, gF I e o
I 0.0 90.0 180.0 TIME (SEC) 27'0.0 360.0 450.0 Figure 1.17-20 High Power Assembly Liquid Velocities I
I ,
R
- .- - - -i -
N
>9,m
'a
_\ _
' il :t 4 5,.h h
gkN o - . . - -. n.-- ,.
so I bY l n c
b h-.- a- oBYPRSS INLET ki e--e LOW POWER E
E h
as o 270.0 380.0 450.0 0.0 e0.0 180.0 TIME (SEC)
I Figure 1.17-21 Bypass and Low Power Assembly Liquid Velocities
l I
E l
I I
VERMONT YANKEE NSSS LICENSING MODEL CASE EW: SMALL DREAK LOCA APPENDIX K RESULTS RECIRC LOOP DISCHARGE PIPE BREAK (0.0S FT2)
I c2 .
8 ee g
3.......
l I !i5 e
o g- -
s g r-eUPPER PLENUM I .J 9 o --o LOWER PLENUM %
- g. .
I a se e 2-o ss I 0.0 3d.0 180.0 TIME (SEC) 270.0 380.0 450.0 Figure 1.17-22 Upper and Lower Plena Temperatures I
I I
I I
I
I Q1.18 In A.VII.18 YAEC stated that THTF Test 3.09.10I was chosen for I assessment of RELAP5YA because it was thought to represent a condition that might be encountered during a SBLOCA, a slow boil-off transient resulting in a two-phase mixture level at low flow conditions. However, based on statements in Volume III of YAEC-1300P (Section 5.1.5) and in response to the initial questions, it is clear there is considerable uncertainty in how the test was run and as a result there is considerable uncertainty on how the test should be simulated.
I Because of these uncertainties, the code / data comparison have little value for code assessment. Provide the results of calculations of other experiments to assess RELAP5YA's ability I to calculate the two-phase mixture level during low flow conditions.
I A1.18 Using hindsight, we agree that THTF Test 3.09.10I was a poor choice for code assessment because of the uncertainties identified above. However, the two-phase mixture level is I rarely available in test reports. When it is presented, it inevitably involves subjective interpretation of the test data by the analyst. We believe it is more meaningful to compare I predicted values of collapsed liquid levels, differential pressures, or void fractions to reported data since these parameters are more readily available and more directly derived from instrumentation signals. The following identifies two experiments with low flow conditions which have been used to assess the RELAP5YA code.
I TLTA Boil-Off Test 6441/6 had very low flow rates in the downcomer, bypass, and bundle regions. Figures 5.2-47 through 5.2-54 in Reference 1.18-1 compare predicted and measured values of the downcomer and bypass differential pressures and the bundle void fractions for this test. The predicted results agree satisfactorily with the test data.
I I
I i l
I Appendix B presents new RELAP5YA code assessment results for I Semiscale Small Break Test S-LH-1. This test predominantly involves low flow rates within the primary system. The assessment includes several comparisons of predicted and measured liquid levels within the system.
Reference l.18-1 Fernandez, R. T., R. K. Sundaram, J. Ghaus, A. Husain, I J. N. Loomis, L. Schor, R. C. Harvey, and R. Habert, "RELAP5YA
- A Computer Program for Light-Water Reactor System Thermal-Hydraulic Analysis, Volume III: Code Assessment,"
Yankee Atomic Electric Company Report, YAEC-1300P, Volume III, October 1982. (Proprietary)
I lI I
I I
I -
I I
I I
I
I Q1.19 In its answer to Q.IX.6 YAEC stated that the stratified choked I flow model was assessed in the LOFT L3-1 calculation discussed in response to Q.V.I. A review of that analysis did not find a discussion of the stratified choked flow model. Provide additional information on how the stratified choked flow model was assessed by comparison to LOFT L3-1 test data. Also, YAEC will use the Moody model in licensing calculations when the break junction void fraction is greater than 0.05. With this, model break mass flux is determined from the upstream pressure I and enthalpy. However, will the transition from nonstratified to stratified flow in the volume upstream of the break have any affect on the composition of the break flow? For example, the RELAPS/ MODI and RELAP5YA horizontal stratified choked flow model attempts to account for vapor pull through or liquid entrainment by adjusting the junction vapor or liquid fraction (see Section 2.2.1.3 of the RELAPS/ MOD 1 manual, Volume 1). Is something similar done in RELAPSYA when the Moody model is used?
I A1.19 The horizontal stratified choked flow model in RELAP5YA and RELAPS/ MODI is described in Section 2.2.1.3, Appendix A, Reference 1.19-1. This model is used for both the standard RELAP5 and the Moody two-phase critical flow models when horizontally stratified choked flow conditions occur. The model is activated when the following conditions exist:
- 1. The junction is connected to a horizontal upstream volume.
I 2. The junction area is less than or equal to the flow area of the upstream volume.
- 3. Cocurrent two-phase flow existed in the junction at the previous time step.
I 4. The stratified flow regime exists in the upstream volume, i.e., G is less than 200 kg/m sec.
I I
I
~
- 5. The vapor velocity in the upstream volume is less than the limiting vapor velocity for relatively smooth horizontal I stratified flow given by the Taitel-Dukler formula.
When activated, the model computes a modified junction void .
l The final fraction, "g j, i.e., it is not donored.
junction void f raction uses a weighted value of this and the old time step value. Finally, the junction liquid fraction is computed:
a"+1 c", + 0,1 (a* , _ an) ,
83 83 8J 8J (1.19-1)
I a"+1 , t _ a n+1 (1.19-2) f3 SJ Thus, this model attempts to account for vapor pull through or liquid entrainment by adjusting both the junction vapor and liquid fractions when horizontal stratified flow exists.
I The horizontal stratified choked flow was assessed using the LOFT L3-1 test data and RELAP5YA calculation submitted in response to Question Q.V.1 (References 1.19-2, 1.19-3). The comparison of calculated and measured results presented in Figures V.1-1 to V.1-5 provided the basis for concluding that
! this model is satisfactory. A discussion of stratification i phenomena is presented below.
I First, we looked for evidence that the fluid was horizontally 3 stratified in the cold legs during the L3-1 test.
Figure 1.19-1 shows the intact loop cold leg density profile indicated by the three gamma densitometer beams. This data clearly shows the two-phase stratified flow beyond 100 seconds. This figure also shows that the RELAP5YA average density at this location became two-phase at about 110 seconds and became predominantly steam by about 600 seconds. The code I calculated a 100-second delay in the accumulator injection I
I actuation and a lower injection rate. Thus, the calculated density remains lower than that indicated by the center beam.
I Figure 1.19-2 shows the broken loop cold leg density indicated by the center beam of the gamma densitometer. Unfortunately, the top and bottom beam detectors failed. However, the center beam density is very similar to the center beam density for the intact loop cold leg. From this, we infer thet the broken loop probably contained two-phase stratified flow also. This is located upstream of the break orifice. This figure shows that RELAP5YA calculated an average density similar to the measured center beam density until 634 seconds when accumulator I injection began. Beyond 734 seconds, some injected accumulator coolant reaches this location, but less than indicated by the center beam density.
Figure 1.19-3 shows the break flow rate derived by EG&G personnel for the L3-1 test and break flow rate calculated by RELAP5YA. The derived flow rate declines as the system pressure and cold leg density decrease between 0 and 634 seconds when accumulator injection began in the test. The I derived flow rate continues to decline between 634 to 800 seconds even though the cold leg density increased during this period. We infer from this that liquid stratified in the pipe l upstream of the break orifice, but the orifice continued to pass predominantly steam. Furthermore, the mass flux obtained by dividing the break flow rate by the pipe area is 2
substantially below 200 kg/see m . Finally, hand calculations indicate the vapor velocity in the upstream pipe are less than the transition velocity given by the I Taitel-Dukler formula in Reference 1.19-1. The RELAP5YA calculated break flow rate compares reasonably well with the derived flow rate for the test. The calculated flow rate is
! somewhat higher than the derived values between 300 and 1,100 seconds due to the higher system pressure. During the l
two-phase break flow period from about 170 seconds on, the horizontal stratified flow model was used in the RELAP5YA 1I
.I
I calculation. Evidence for this is given in Figure 1.19-4 which shows that the void fraction at the break junction is different from that for the upstream pipe volume.
I Based upon the above information, we conclude that the horizontal stratification model works reasonably well.
References 1.19-1 Fernandez, R. T., R. K. Sundaram, J. Ghaus, A. Husain, J. N. Loomis, L. Schor, R. C. Harvey, and R. Habert, "RELAP5YA
- A Computer Program for Light-Water Reactor System I Thermal-Hydraulic Analysis, Volume I: Code Description,"
Yankee Atomic Electric Company Report, YAEC-1300P, Volume I, October 1982. (Proprietary) 1.19-2 Bayless, P. D., J. B. Marlow, and R. H. Averill, " Experiment Data Report for LOFT Nuclear Small Break Experiment L3-1,"
NUREG/CR-1145, January 1980.
1.19-3 Letter, G. Papanic (YAEC) to J. A. Zwolinski (NRC), " Response I to NRC Questions on the RELAP5YA Computer Program," FYR 85-121, November 1, 1985.
l I
I I
I I
lI
O _
PoO-
- Nh
. . h _
g f C ME I
f
. 1 3 _
}
g ab _
a 5 M i n
g b
5 5
y
. t 0 i 0 s 5 n 4 e l
D g a g en R
. bLo 0 dt l a i
g T A
D .
5 0
2, ol Cuc pl g
1 3
L T
R000 R8c 111 bd '
b 0
0 0
1 1
0 1
oa o, C t Y cS aP A
F O
Y000 5 PPP Cc -
a 5) 0 W {0.cIE oo t A, n
F t.
mL O sC i- - S f R
T CCCc b
RDDo t N .
e: M(E M o1 n
O -
- 5. I n -
S - doT i3 W I R
A P
e:= e 0
rI a
pT mF M 0 o O, b
O d c1 C .
7 M R Y
O.
o 5 1 d :
P .
- O 1 A R 3 0 L Y L 9 E E R 5 P
f r
d 0 5
1 1
A o L L 0 e _
E 0 r _
R d4 u E . g b'
t a i 0 F d
o _
3 _
E ;
l 0
s
[2 E
I
- d1 0
0 0
' O E i, 4' -
o3[ ,
- g. . ,
- 9s a ,N .
o N@ R:
E
' .ji i:',
I .
I I
. . . . . . . 0.0015.0 g
- g j ~ -
% .l
$ .g i
.E .
_ s 9
3
.g 8
- # I I:
3 g
r < -a i I 3. e
~
.i 9
fa<
h
\
t$
I S 3- -R j E $ M 5: '
o ~@ t~ o
.gE 5-l _
{ .
iC 'sLv U -h $$
h -
I
- {- e -+ ':
ll
= ..
N - G- ~
- - - e -
. F % .g q h
=-
w:
___+ ->
u e
I .g
- _ N- -
T E o
~
- v-
~
- .i l I oa lC-N e
alt oh ok ok ok ck oit a hi os oo A
O
\
l l
(E13/W97) 111SN30 S31 0100 dD01 N3NDES 1
.g .
I 000148 I , , , , , -g I my '1s =
C xm_m,
c
~ wo 9
[ -o d t
-I l
- %3 O %
S mE to k
s a ; o c h-
~
j ~s - %. $
I o J 3 om do 33 5$
o I H z
~
~$ {n j {
o
- c. -
o I en _ u _d y 86 8-E q s
I c O. T Q_ _o S 2'_- _
O c e 3
_o -
I x_
o_
c I h3 E
3
_o O
I e O ~i i i i I 0*09 i
0*0S 0*0F Q'OE 0*05 0'01 _-O'O (03S/W9T 3198 M0ld W389 I
m M M M M M M M M M C I I i l I , I I I Z
o j
, ge.t y l{\f I l
f-4 o o_
- r j
'"I f fEl_
[ j '
'I!
M .l ii
- Upstream Liquid Below Break O to S Liquid Entrainment at Break m
O O l g ] Break Orifice Region O - - '
!~
Q- 9 le ~
CO~ f Upstream Liquid Above Break
> Vapor Entrainment at Brea'e
- Break Void Fraction _
O_ DO Ups trear.: Void Fraction t
O O00 i i i i i i i i 0 200 400 600 800 1000. 1200 1400 1600 1800 TIME Figure 1.19-4: Break and Upstream Void Fractions
I Ql.20 Attached is a summary of the updates included in RELAP5/ MODI in going from Cycle 18, the code version used to develop RELAP5YA, to Cycle 29, the final version of RELAP5/MODl. For each update, state whether it was (1) incorporated as written by RELAP5 development, (2) incorporated, but in a modified form, or (3) not included in RELAP5YA. For each update in Category 2, briefly discuss why and how the update was modified. If an update was not included (Category 3) briefly discuss why it was not incorporated into RELAP5YA.
I A1.20 The update summary referenced in the question has been reproduced in a simplified form and attached as Table 1.20.1.
Each item has been given a number that contains the updated cycle number to the left of the decimal and a consecutive number within that update to the right of the decimal. The column on the right indicates the following:
I 1. The update was incorporated As Written (AW).
- 2. The update was incorporated in Modified Form (MF).
- 3. The update was Not Incorporated (NI).
I The following information provides a brief discussion of items in Category 2 (MF) and Category 3 (NI).
I 19.1 MF We use local variable TTEMP rather than QUAD 1 since the parameter is a temperature, not quality.
19.2 MF We use (AJ.GT.l.0) rather than (AJ.GT.0.9) to retain the stratification model for a broader range.
19.5 NI We currently do not use this feature.
19.6 NI This has not been a problem for us.
I
l 19.8 NI We do not use this feature.
19.14 MF We use slightly modified coding to improve readability, but the equations are the same. l l
I 20.1 NI This update is subsequently modified in a later-update.
20.3 NI This is not an error correction, but rather an attempt to " improve" the code. We are not convinced.
this change is desirable.
20.4 NI Same as above.
20.5 MF We have incorporated the three lines that restore pure donoring and remove the punt logic in JPROP.
I We are not convinced the other changes are desirable.
20.6 NI The proposed Flip-Flop logic contains errors. We have experimented with several alternative Flip-Flop algorithms and have not found one that is entirely satisfactory.
I 20.7 NI Same as above.
20.12 NI This update contains an error. We do not believe this update is warranted.
20.13 MF We have incorporated the elevation checker including the correction in Item 28.1.
21.1 MF We calculate CHF using modified Biasi and Zuber as documented in Volume I, YAEC-1300P.
21.3 MF Our logic is similar, but also contains a check that the internal transition boiling lockout flag IEM2 is off ( 0 ).
21.5 NI YAEC has not detected a problem. We use the rewet/ quench model and the EM logic for licensing analyses.
l I
21.6 NI This has not been a problem for us and does not affect calculations.
21.7 NI We do not use this feature.
21.9 NI This change is subsequently deleted by Item 27.1.
21.10 MF We use a different solution to obtain FRICTF that is well behaved.
21.12 NI We have not had a problem with this.
21.13 NI We do not need this feature.
22.1 NI We have not had a problem with these.
24.1 I
NI We have not changed the environmental library. The typo error in HTADV was created in Item 20.12 which we did not incorporate. This is not an error correction, but a printout of diagnostic information when an error occurs.
I 26.2 MF We have addet other diagnostic information to detect this error when it occurs.
27.1 MF Our version of RELAP5YA already has the correction I to fix the negative indefinite in IJPROP. Since we did not incorporate VHR STUFFER per Item 21.9, then we have no need to remove it.
I
I 29.1 MF RELAP5YA has a modification to VOLVEL that has worked satisfactorily. We have adopted the change as written to minimize the truncation error in the I convected terms and to eliminate the possible indefinite in the accumulator. We have not used inertial valves.
I E
R I
I I
I I
I I
I I .
"/
e
I
~
I ,
TABLE 1.20-1 REL AP S /M001 UP D ATE S
'l BEGINNING OF UPDATE 5 TC CYCLE 014 TO CREATE CYCLE 019 - 23NCV82' U.S. ss3001.
~
Sh8Achf1N3 Ob$hk$hI STATE.
gp 19.2 FIX NgGATjVE SQUARE ROOT UNDER STRATIFICATION 508kbukNEIMODIFIECs JPROP. ,
U.S. s83002.
^
Rh!Ch I O Ekb SUBROUTINES MODIFIED IRKIN, RKIN.
^
$h8R0 N D ED IPUMP.
INfuk 0 H T C E suiR0 URINES n0DI IED INf Car .
g NI 19.6 jgMOV) jjS{0F UNIT TRIP CCMPCNENT BEING INITIALIZED.
SUBR60 TINES M6DIFIECs RCONVR.
I 19.7 PR T OF MCMEhi 0F INERTIA IN BRITISH UNITS IN SudRbuTIMES MbOIh$k6RVALVE.
AW I "8 !PS!!!!!!ioff!!!!5.is'fuYlRi8"!BRl!!",WM 81nH
^*5"'"5 ""'" ""'
" l E868 !!f91ilfhE8?'S SUBROUTINES MODIFIED: RKIN. RRKIN.
" ' '"^"" l N AW 19.9 g MI5gGgyR15ks TO AN LRROR DIAGNOSTIC IN j SUIkobilNES MODIIIEDs RTMDJs RTMOV. l I6iR!ui1EV!'EisfFIEci%IIN. .
u.S. . 3003.
.g "" ^"
4 tj'hnerm%mmestiami'm"a i
fu!R0dINESiisiliEos TSTATE.
g t
1 I
i I
l REL AP5/M001 UPDATES (CONT 8 0)
I ~
U.S. 883004.
I SUIRN O Eb f50b . *
^"
l 2 19.14 15 Ribilsis980i"Ill!U fHJ'"^"-
S CORRgCTIONS TO WALL FRICTION, FORM LOSS ANO DISSIP ATION MF kh8RCUTkNISM00!FIECs VEXPLT.
END OF UPDATES TO CYCLE 018 TO CREATE CYCLE 019 I 8EGINNING OF UPDATES TC CYCLE 019 TO CREATE CYCLE C20 - 18FE883 U.S. 883005.
"""5 "'
2 iP!!N '
N!!Pis52id !'t!"i!ItHi"""' "Sc "5 SU8R00 INES n0DIFIECs VEXPLT.
I 20.2 CORRECTION TG RBRNCH FOR VELCCITIES WHEN USING ERITISM AV IUhRbOTINESN00!FEC RSRNCH.
3 u. S.. 820c,.
ur 20.3 r pgg;pg;gyTgogc;ggg,gggSgggggt,,uDn455EuCRAND y
SUBROUTINES MODIFIEDs SCNREQ.
" ao ' "'
io!!O!!!Eis958Illi!! Sun.
MF Hg20.3 , p, NovE ENEn.v CORRECTION WsEN ountS, QUALA peggSTg3P.00!FIED:
Sb8ROUTkN EQFINLs IJPROP, JPROP. j ur l120.q- ggpgg iT ;ggp ggEggg3 g >tgypigg,g "<o","""> y THE n NE n$i m vR. s i e g m i
20a g3;gggg,;04ggg;,,7ag,yg,,mmn mmmen
^W lg 20.8 ;;;;gg,jgg;gggi;g;(xg;,gEanNt.
I I.
I I
~
I RELAP5/MCD1 UPDATES (CONT'D) ,
I ^w 2od E8hi!il ill! Ell!'cilkIEi" """-
I 20.12
!!ilMTInts'iablElID" it!M"" '"-
^w l 20.22 ^w
!!k!MTInis'!5b! Ell 0" !!!5n" " 'ccu"-
U.S. 883006.
x1 20.12 'ggg,;g w up; gggg nigugg,gggnan a unE-nu l SUBR00kINEkM001FIEbs IONT AL, HTADVs HT1 TOPS TEAN.
U.S. 883007.
I
- kb8AC ES 00hkE! ffbkN [f.
END OF UPDATES TO CYCLE 019 TO CREATE CYCLE 020 SEGINNING CF LPDATES TC CYCLE 020 TO CREATE CYCLE C21 - 29 MAR 83 U.S. 883008.
I TH 208 bI ski SU8A0011NES MODIFIED
!OR NAhUN CHFCAL. -
^
! MbONDEN.
I bT kN hE NORkUTINk3PODIFfEC au8R T0h!
21.3 Nr gggggg,NggTggggggg; sggggggs rH mn DNs.
I
^"
21.4
{g ggxgg Cgo y, g 3 gjl 5N!P " """ ^" "*
f 22 5 5"" c^'c"*" "- s l 181 !Mdli"!!!!!!E!!'"NillH" U.S. 883009.
NOT RfM AT E NUMBER ME O! h5 g NI los!6)A,ai ,001, u0 a 1Nx ,,, OUT,UT.
l 21 7 U.s. 6,010.
e,nauitsam" ICOMPN, a' is"""""" "" ai" s1 SUBROLT1his AOLIFIEtt RVALVE, VALVE.
2'
- tlimmiN!; s:Diszi!"9smanime""> """""> im I
I I
RELAP5/M001 UPDATES (CONTeo3 I U.S. 883011.
NI 21.9 UPDATES TC IMPLEMENT VHR STUFFER FOR 2-PHASE MASS kb$AU71NESMODIFIED: STATE.
l
^"" """ '" """""-
21.1o
!5idOHilS aidi!!! "9 "M"^"'
" '"c" " c """'"- ^"
!!idMiAIS aid lil! Alii!!"
l u.S. 83012.
" 5 ' 5'"2 " ""' c" "'
21.12 I '
ggg'lulfil 2: YAlf1d"8E'11I0" '
SUBROUTINES MODIF E08 ICOMPM.
l iBl!0 Hill!'usililbl5 Ildl^ciRIN"illlU flAllM"- l shD CF UFDATES TO CYCLE 020 TO CREATE CYCLE 021 I BEGINNING CF UFCATES TC CYCLE 021 TO CREATE CYCLE C22 - 05JUL83 U.S. 883013.
hh$ CEf ALNi f 0 f 0 Rfh5 0 bkfMGbROR.
SUBRCUTINES M00IFIID8 ICOMPN, PLOTS, RVAL VE, VALVE.
END OF UPDATES TO CYCLE 021 TO CREATE CYCLE 022 SEGINNING OF UPDATES TO CYCLE 022 TO CREATE CYCLE C23 - 01SEP83 U.S. s83015.
!hf $6 f P OkfN.
SU8kOh!NESMODIFIEDs RPUMP.
, U.S. s83014.
^"
l 23.2 f5idMYNf19hht!!is,IN!M" "" ^" " "-c"c""-
l ENO CF UFDATES TO CYCL E 022 TO CREATE CYCLE 023 I
I
C .
RELAPS/MCD1 UPOATES (CONT'03
[ .
SEGINNING OF UPDATES TC CYCLE 023 TO CREATE CYCLE 024 = 02NOV83
{ U.,.
8 00,.
c> pUSqA sU pNE3 rwFrEmswsme-a M60!qF HTADVEDsna "
a END OF UPDATES TO CYCLE 023 TO CREATE CYCLE 024 BEGINNING OF UPDATES TC CYCLE 024 TO CREATE CYCLE 025 - 15NCV83 U.S. 884002.
UBRC NE D N N
{' END OF UPDATES TO CYCLE 024 TO CREATE CYCLE 025 BiGINNING OF UPDATES TO CYCLE 025 TO CREATE CYCLE C26 - 07FE884 U.S. s84004 Ak' 26.1 RCR YIhG TO DELETE A NONEXISTENT HTORO-5U8R00kIMh5NG01FIEC RCDELT.
[ U.3. 84003.
NE A A R ESbR E ER f Tk V . G kUG I bT b SU8ROUTkN!5M001 fib 8ACUMsICONhMa REShe=$$ ATE.
END OF UPDATE 3 TO CYCLE 025 TO CREATE CYCLE 026
( 6dGINNING OF UPDATES TO CYCLE 026 TO CREATE CYCLE 027 - 26APR84 U.S. 884005.
CP FOR NEGATIVE INDEFINITE. MF 27.1 gP {Tgg IJPROP,'INVJT, STATE.
SUBROUTINES MODIFIECS END CF UFD ATES TO CYC' L 026 TO CREATE CYCLE 027
I l -
RELAPS/M001 UPDATES (CONT'0) ,
g I SEGINNING GF UPDATES TO CYCLE 027 TO CREATE CYCLE 028 - 19JUL84 u.S. 84008.
h0kNE N $N kN O J5 CT! W 11xEt Hu8'Ii5 h08RbbflNESN0h5FhC IELVTMs VEXPLT.
END OF UPDATES TO CYCLE 027 TO CREATE CYCLE 02 B EGINNING OF UPDATES TO CYCLE 028 TO CREATE CYCLE 029 - 18SEP84 U.S. e64007.
Okk! 'T hh . $$ THE FV FhG.
Vf(VfAHROAT IN b STATES VALVE, VEXPLT, VO LV E L .
SU6RCUTINES MC01FIECs I cNO OF UPD ATES TO CYCLE 028 TO CREATE CYCLE 029 I
I I
E I
I I' .
s l l
1
I
- 2. PWR-Related Questions (Reference Q.8)
Q2.1 In Appendix V.I-1, which presents the LOFT L3-1 additional assessment calculation, Figure V.1-1 shows that RELAP5YA predicts a lower depressurization rate than the test. The I lower depressurization rate was believed to be due to a lower break flow quality and subsequent lower energy removal at the break. Please elaborate further on this assumed lower break flow quality dependence. (Include, if possible, comparisons to L3-1 break flow quality data or void fraction.)
A2.1 This question was deleted from the set transmitted in Reference 2.1-1. However, we have reviewed the LOFT L3-1 tsst data and our RELAP5YA calculation with respect to this I question. First, no test data exists for the break flow quality or void fraction. Therefore, we have considered the potential causes for the lower calculated depressurization rate between 350 and 500 seconds. The sources of heat, namely the core decay heat and the passive heat structures, were adequately modeled. The calculated decay heat compared well with an independent calculation using the 1979 ANS Standard decay heat model as well as the decay heat curve presented in Figure 21 of Reference 2.1-2. During this period of time, it I appears that the steam generator does not act as a major heat sink in the test nor the calculation. HPSI is the only ECC injected in this time frame and it is reasonably predicted.
The only remaining significant parameter that would affect the depressurization rate is the total enthalpy leaving the Primary l System. Perhaps there was an additional leakage path from the 1 l
Primary System that we have not accounted for, but we do not have any direct evidence of this. That leaves us to conclude that the calculated break flow quality during this period was I somewhat lower than in the test.
I I
I
I References Letter, E. M. McKenna (NRC) to G. Papanic (YAEC), " Request for I 2.1-1 Additional Information - TMI Action Plan, Item IIK.3.30, Small Break LOCA Outline," NYR 86-225, October 3, 1986.
I 2.1-2 Bayless, P. D., J. E. Marlow and R. H. Averill, " Experiment Data Report for LOFT Nuclear Small Break Experiment L3-1,"
NUREG/CR-1145, January 1980.
I I
I I
I I
I I
I I
- I I l lI I ,
l
I I l Q2.4 YAEC stated (Q.II.14) that for CE plants the two cold legs / loop would be modeled separately. The nodalization diagram given in Figure IV.16.1 for Maine Yankee, a CE plant, does not show the
-I two cold legs modeled separately. Clarify what nodalization is used to.model the Maine Yankee cold legs.
I A2.4 We believe the reviewers meant to refer to Question Q.VI .14 rather than Q.II.14, and Figure VI.15.1 rather than IV.16.1.
Our response to Question Q.VI.14 stated that if we were to apply the code to a CE plant with two cold legs per loop, the separate cold legs in each loop would be modeled separately.
However, although Maine Yankee is a CE plant, it has three I loops with one cold leg in each. Therefore, the nodalization diagram shown in Figure VI.15.1 reflects one cold leg per locp.
I I
I E
I I
I I
I I
I I - - - -- - - - -
I Q2.5 In response to Q.IX.2 to 4, YAEC provided some information on how they would model a PORV break. In its answer, YAEC stated the PORV passes steam because the pressurizer spray is terminated by decreasing system pressure. During PORV break calculation does the pressurizer level ever increase so that I the PORV would would pass a two-phase mixture? If so, what pressurizer nodalization is needed to accurately track the pressurizer level? Would more than two pressurizer nodes as discussed in Q/A.II.12, be used in this type of calculation?
Has YAEC reviewed the PORV flow data generated by the EPRI Safety and Relief Valve Test Program to assess the PORV flow calculated with RELAPSYA and the modeling techniques discussed in the answer to Q.IX.2 to 4? Discuss what impact the I differencebetweenthecalculatedandac{ualPORVflowwould have on licensing calculations. Will a break area spectrum study be performed for pressurizer breaks? If not, will YAEC complete a discharge coefficient sensitivity study to assess the effect of uncertainties in the break flow on the calculated system response?
A2.5 The following information clarifies our response to Questions Q.IX.2 to Q.IX.4 (Reference 2.5-1).
I For PWR SBLOCAs, the Power-Operated Relief Valves (PORVs) will not actuate due to their high pressure setpoints and the generally decreasing primary system pressure. If a SBLOCA is coupled with a stuck open PORV (single failure criterion), then the PORV will pass steam due to diminished primary system inventory and pressure. These cases are within the 10CFR50.46 definition of LOCAs.
If a stuck open PORV is postulated without a pipe break, then 5 the pressurizer level will increase and the PORV can pass a two-phase mixture. We would use several nodes to model the pressurizer (see Answer 2.2) and the RELAP5YA standard critical flow models to simulate this plant transient. YAEC has reviewed information from the EPRI Safety and Relief Valve Test l
I
I I Program to assess the flow calculated by RELAP5YA (References 2.5-2 and 2.5-3). Table 2.5.1 summarizes assessment results from a water and steam test for Dresser Valve 31533VX-30 which I is the same type of valve used on the Maine Yankee pressurizer. The discharge coefficients of 0.825 and 0.645 that reconcile the calculated and measured flow rates for water and steam, respectively, lie within the well-established range of 0.6 to 1.0 (see Answer A.IX.1, Reference 2.5-1).
YAEC is interested in stuck open PORV events from the viewpoint of plant operational safety, e.g., to review emergency I
procedures and to assist with operator training. Thus, we are interested in more realistic system analyses rather than the licensing analysis approach. YAEC has conducted a stuck open PORV analysis for the Yankee plant at Rowe and the Maine Yankee plant in the past. The PORV models used discharge coefficients of unity to maximize the fluid loss. In each case, the PORV did pass steam and a two-phase mixture during the transient.
The energy removed through the steam generators and the PORV was more than sufficient to remove the core decay heat.
However, the PORV throat areas are sufficiently small that I sufficient coolant was retained to prevent any significant core heatup. For each plant, the results indicated that a stuck open PORV was a relatively benign transient from a LOCA-ECCS and core cooling perspective. We believe that using smaller discharge coefficients would not have significantly altered these results or conclusions. Therefore, we believe that I neither a break spectrum nor a discharge coefficient study is warranted. We would emphasize a more realistic, yet limiting, scenario.
I I
I 1 I l l
I TABLE 2.5.1 Calculated and Measured Flow Rates for Dresser PORV 31533VX-30 l I Inlet Measured Calculated Effective Temperature Flow Rate Flow Rate Discharge Test Pr re 791 Number (psia) ( F) Fluid (1bm/hr) (1bm/hr) Coefficient 16-DR-6W 2360 636 Water 331,200 401,076 0.825 20-DR-1S 2296 TSAT Steam 129,600 201,027 0.645 References 2.5-1 Letter, G. Papanic (YAEC) to J. A. Zwolinski (NRC) " Response to NRC Questions on the RELAP5YA Computer Program," FYR 85-121, November 1, 1985.
Singh, A., (Editor), " Safety and Relief Valves.in Light Water I 2.5-2 Reactors," NP-4306 SR, EPRI, December 1985.
2.5-3 , "EPRI/Wylie Power-Operated Relief Valve Phase III Test Report; Volume 3: Summary of Phase III Testing of the Dresser Relief Valve," NP-2670 LD, EPRI, October 1982.
I I
I I
I I
I
I I APPENDIX B I
I I Assessment of RELAP5YA Against Semiscale Small Break Test S-LH-1
1.0 INTRODUCTION
I The assessment of RELAP5YA for. licensing analysis of PWR SBLOCAs was presented in References 1 through 7. Although the assessment covered the spectrum of phenomena expected in PWR SBLOCAs, there was no direct assessment against an integral test designed primarily for PWR SBLOCAs in which core I uncovery/ recovery was encountered. The Semiscale Small Break Test S-LH-1 provides these phenomena. Hence, RELAPSYA has been assessed against this test. In the assessment presented, two calculations have been carried out.
The first is a Best-Estimate (BE) calculation, while the second is more similar to an Evaluation Modef (EM) calculation.
I
2.0 BACKGROUND
Semiscale Test S-LH-1 (Reference 8) simulated a 57. centerline break in a PWR cold leg. The phenomena observed in these tests and a post-test analysis of the test using the RELAP5/ MOD 2 computer code (Reference 9) have been presented in Reference 8. The test was conducted in the Semiscale Mod-2C I Facility (Reference 10).
The input deck used for the RELAP5YA simulation of this test was derived from the input deck used in the RELAP5/ MOD 2 analysis of the facility (Reference 11) and an earlier RELAP5/ MOD 1 model (Reference 12). Various changes to the input were necessary because of the different input requirements of RELAPS/ MOD 2 and RELAPSYA. Some additional changes have also been made to reduce the computational costs. The changes are described in the next section.
I I
I 3.0 RELAP5YA BE Model for Test S-LH-1 Figure 3-1 shows the nodalization used in the RELAP5/ MOD 2 analysis of I Test S-LH-1 (Reference 8). The changes required to convert tro RELAP5/ MOD 2 input to appropriate RELAP5YA input were as follows:
- a. Conversion of cross-flow junctions to normal junctions because RELAP5YA does not have this feature.
- b. Corrections and modifications to various control systems. These control systems accessed features not available in RELAPSYA. For example, in RELAP5YA, input for heat structures cannot be modified I during restarts.
Several additional changes were also made from the perspective of modelling philosophy and computational cost. These are explained below.
- c. Only heat structures simulating the heater rods in the core and the steam generator tubes have been retained. All other heat I
structures were removed. Essentially, this implies no heat loss from the system. This is considered reasonable because in the test, guard heaters were employed (and modelled in RELAP5/ MOD 2) which kept heat losses to a minimum (about 57. of core heat). This change was made primarily to decrease computational time.
I d. The number of radial mesh points in the heater rods were reduced I from 18 to 7. Again, we Lc11 eve this would result in substantial savings in computational cost without overly compromising the accuracy of the simulation.
I e. The number of hydrodynamic nodes in the core was increased from 6 to 12 to provide better resolution of the core thermal-hydraulic response. The change results in core nodes of 1.0 ft length compared to the 2.0 ft length used in the RELAP5/ MOD 2 analysis.
I I
- f. The temperature of all ECC water was raised to 200 F. This is consistent with the RELAP5YA method outlined in Reference 6.
I g. The model for the aecondary side of the steam generators was derived from an earlier RELAPS/ MOD 1 model of the semiscale I facility. This final model is similar to the models expected to be used in PWR analyses using RELAP5YA.
All the above changes are consistent with the modelling cor.cepts that will be used in RELAP5YA analysis of PWR SBLOCAs. The one difference is that in the plant model, heat losses and passive metal mass will be included.
These are not expected to have a significant impact in Semiscale Test S-LH-1.
The sequence of events for Test S-LH-1 and the test results have been I outlined in Reference 8. The results of the RELAP5YA simulations of the test are presented in the next section.
I I
I I
I I
I :
I i I
I
I 4.0 RELAPSYA BE SIMULATION OF TEST S-LH-1 Figure 4-1 shows the system pressure comparison. Figures 4-2 and 4-3 show the break mass flow rate and the fluid density upstream of the break, respectively. The pressure prediction is seen to be reasonable from I Figure 4-1. Figures 4-2 and 4-3 show that the fluid conditions and flow rate at the break are reasonably calculated in the early period. However, after about 500 seconds, RELAP5YA tends to push liquid to the break location causing the density fluctuation seen in Figure 4-3. This has a significant impact on I the calculated results and is further discussed below.
High Pressure Safety Injection (HPSI) began shortly af ter the break was initiated. Figure 4-4 shows the HPSI flow rates to the intact and broken loops.
The drainage from the pressurizer and the upper head is shown in Figures 4-5 and 4-6, respectively. The liquid fraction in the bypass line that connects the downcomer to the upper head is shown in Figure 4-7. The bypass is calculated to clear at about 180 seconds. This occurred at about 240 seconds in the test.
I When the bypass line cleared, steam from the upper head was able to vent through to the break. Until this happened, however, the steam was I blocked due to the presence of liquid in the steam generator tubes and the pump suction pipes. The calculated collapsed liquid levels in the upside and downside of the broken loop steam generator are shown in Figure 4-8. As can be seen, the liquid in the upside of the tubes remained until about 200 to 250 seconds. Figure 4-9 shows that liquid in the upside tubes of the intact loop steam generator also remained until about 270 seconds. In the test, liquid was indeed held up in the steam generators; however, this lasted only until I about 140 seconds.
I After the steam generator tubes drained in the test, liquid that collected in the pump suction pipes continued to block steam flow. A manometric depression of liquid level in the vessel resulted. In the test, the intact loop seal cleared at about 180 seconds and the broken loop seal cleared later at about 280 seconds. Before the intact loop seal cleared, the I
I vessel level had dropped to about 40% of the core height and caused a core heatup. When the intact loop seal cleared, the core level recovered to the I
top of the core. Subsequently, when the broken loop seal cleared at 280 seconds, the vessel level rose some more. The calculated levels in the 1 broken loop seal and intact loop seal are shown in Figures 4-10 and 4-11, I respectively. In the period before 300 seconds, both loop seals were calculated to be plugged much later than in the test. This resulted in a substantially lower liquid level in the core as is seen in Figure 4-12. The intact loop seal was calculated to clear at about 300 seconds and led to core recovery. The broken loop seal never cleared in the calculation. Thus, the l vessel level did not recover to the same height as in the test.
I After the loop seals cleared in the test, the liquid in the vessel boiled off. The steam readily flowed to the break location. In the calculation, this period began with less liquid in the vessel than observed in the test because the broken loop seal was still plugged. Due to continued steam blockage, the calculated vessel level decreased faster than in the test.
Essentially, RELAP5YA distributed the fluid in the system more towards the loops than in the vessel. This can be seen in the comparison of liquid I
levels in the downcomer shown in Figure 4-13 during the period from 200 to 500 seconds.
At about 500 seconds in the test, the system pressure reached the accumulator setpoint. The accumulator was actuated and the vessel level recovered by the end of the test at about 1,000 seconds. In the calculation,
} the accumulator was actuated at about 475 seconds and began core recovery.
However, liquid was calculated to enter the loops and plug the break, resulting in a decrease of the depressurization rate. This, in turn, caused l
the accumulators to shut off until the pressure dropped further. This caused lg W core recovery to be arrested and the vessel and downcomer to remain in a depleted condition for a long period of time. This can be seen in Figures 4-12 and 4-13. The accumulator flows in the intact and broken loops are compared in Figures 4-14 and 4-15, respectively, and show the calculated intermittent accunulator flows.
I I
l 5
There were two core heatup periods in the test that are a consequence of the vessel level response, shown in Figure 4-12. The first was due to the l
manometric core depression before loop seal clearing, and was short-lived.
The second was due to the boil-off and was terminated shortly after accumulator injection began. The highest clad temperatures at five axial I locations in the core are shown in Figures 4-16 through 4-20. The peak clad temperature occurred at the 228 cm elevation (Figure 4-18). As can be seen in Figure 4-18, the calculated peak clad temperature is about 40 K below that observed in the test.
I I
I I
I I
l I 1
I I
I I
LI I
I 5.0 RELAPSYA EM SIMULATION OF TEST S-LH-1 I In order to assess the impact of certain licensing assumptions used in PWR applications of RELAP5YA, a calculation was performed by increasing the steady-state core power by 2% and the decay power by 20% (after an initial I 10 seconds to allow fission heat to decay). These were the only EM features activated in this calculation. Other EM features, such as the Moody Critical Flow Option and the heat transfer regime lockouts have not been employed in this calculation. The Moody model is not expected to impact the results of PWR SBLOCA analysis other than shifting the limiting break size to a lower value. The heat transfer lockout flags were not activated since we do not I plan to use them for PWR SBLOCA analyses per our meeting with the NRC at Bethesda, Maryland on June 22, 1984. The objective was to see if RELAPSYA, with these PWR SBLOCA EM features, would yield conservative results with I respect to core liquid inventory and peak clad temperature.
The calculated results were very similar to the previous BE calculation from the perspective of hydraulics. The hydraulic behavior, including pressurizer and upper head drain, liquid hold up in the steam generators, loop seal plugging, and initial core level depression were almost identical. The I broken loop seal continued to remain full of liquid. The accumulator behavior was also nimilar. The only major difference, due to increased core power, was in the core thermal response as shown in Figures 5-1 through 5-5. These EM results show the calculated peak clad temperature is about 67 K higher than observed in the test.
I I
I I
I I
I
6.0 CONCLUSION
S The assessment of RELAP5YA against Semiscale Test S-LH-1 indicate the I following:
- a. All major thermal-hydraulic phenomena observed in the test were calculc*rd to occur in about the same sequence, but the timing of some events was different.
t .' The RELAP5YA calculation generally tended to redistribute the liquid in the system towards the loops rather than retaining it in the vessel. This resulted in an extremely conservative calculation I of core liquid levels.
- c. The core thermal response was calculated reasonably, but comparison to data indicated overprediction at some periods and underprediction at other periods. In the BE calculation, the peak clad temperature was underpredicted by about 40 K. In the EM I calculation, the peak clad temperature was overpredicted by about 67 K.
- d. The net result of this assessment is that with EM assumptions, RELAP5YA is expected to provide a conservative calculation of the core liquid inventory and peak cladding temperature for PWR SBLOCAs.
The comparisons to the S-LH-1 test data have shown that RELAP5YA may be overly conservative, especially in the calculation of core hydraulics.
I I
I I
I REFERENCES
- 1. Fernandez, R. T., et al., "RELAP5YA - A Computer Program for LWR System I 2.
Thermal-Hydraulic Analysis," YAEC-1300P, October 1982.
Letter, J. A. Kay (YAEC) to J. A. Zwolinski (NRC) on Transmittal of Responses to 33 NRC Questions on RELAP5YA, March 1, 1985.
- 3. Letter, J. A. Kay (YAEC) to J. A. Zwolinski (NRO) on Transmittal of Responses to 26 NRC Questions on RELAP5YA, April 30, 1985.
- 4. Letter, G. Papanic (YAEC) to J. A. Zwolinski (NRC) on Transmittal of Responses to 80 NRC Questions on RELAP5YA, July 1, 1985.
- 5. Letter, G. Papanic (YAEC) to J. A. Zwolinski (NRC) on Transmittal of Responses to 15 NRC Questions on RELAP5YA, August 15, 1985.
- 6. Letter, G. Papanic (YAEC) to J. A. Zwolinski (NRC) on Transmittal of Responses to 43 NRC Questions on RELAP5YA, November 1, 1985.
- 7. Letter, R. W. Capstick (YAEC) to D. R. Muller (NRC) on Transmittal of Responses to 39 Additional BWR Questions on RELAP5YA, December 31, 1985.
I 8. Loomis, G. G., and J. E. Streit, "Results of Semiscale MOD-2C Small Break (5%) Loss-of-Coolant Accident Experiments S-LH-1 and S-LH-2,"
hTREG/CR-4438 or EGG-2424, November 1985.
- 9. Ransom, V., et al., "RELAP5/ MOD 2 Code Manual," EGG-SAAM-6377, April 1984
- 10. Beucher, T. J., and J. R. Wolf, " Experimental Operating Specifications for Semiscale MOD-2C Feedwater and Steam Line Break Experiment Series,"
EGG-SEMI-6625, May 1984.
I 11. Personal Communication, G. G. Loomis (EG&G) to M. A. Landerman (ITI) and R. K. Sundaram (YAEC), June 1986.
"RELAP5 Standard Model Description for the Semiscale I 12. Leonard, M. T.,
MOD-2A System," EGG-SEMI-5692, December 1981.
I I
I I i I
E E E E E E E E E E E E E E E E I c'~__] [E=_J Le" J [ l*~9 l
"" t_ a" .
C.as S "" ewI_ e._
c ._ ]
- t. :
G u. ,.4 cues A
[ l rwi LM f,l C+':
[ c ro. _j I PRES $URilER I s / co .2 L#2'c,.. wcs ..' ,
.] .
[,,.~. - -
u
"---l 7 m s _.
2 ,.
L L ,
c,n l E cies 4' 34 )-
1 3
cno' 3
}-
. 3 3' CS30 .3 2- cAs en. 1 1 ,
6 E ._
cm ~
2 2
=
3
- CV92
[_ ,
. _t._ 2 ' ' *
- r-~i c .,', a
- _c"*-
c'az c'** , Y am 9 7- p- ,,,
oil em [: r :1 ciesis ; c, ,,[. [. l c]cu3 EIS' 4 t i crio f 6 ^'S[g i i c.uo C3'O f: C 'C' l E CSw7 as y# .>
.l C230 l . - .Q- % # A 3%
~
f ( CIBS 1 W
g5 cao s l-- c_m ' T cm )> ,
\ * ..} *
.{.. *j'l ' f. CtJe J- 1- g [
- M c202
.[ 'i */ ],'. f
. g Gt C 'r.,2 [,
['{ C.C r% -
m Ir -
-- it ,- W -t_
e ~1 i . _ ,
g373 (;g,3 2 CMt C263 c261
+-.
n 3 sF r
, l cvoo l **
csoo -
l
- caso
~
c1% *
~ . 2. . i.
.I _ -. , ;, '*'
- ts t ,
, q ,,
A J1 *$ J4 90
.S io g I A
. . 4 c,,o 0080 C340 ,
_ m I -
C.S.'O
- ClM V c420 ' --
c130 e .
C430 EfQ<EN t.OOP l INT AC T t OOP l_. ciaa
]
. w o e.,os w s w etat h hv.
- w.c. Amy 9 s > Sof E Mes et A TE RS Figure 3.1
- Nodalization Diagram for RELAPS Calculations
s I
I i , i i , i -
o.
o o
o I o a
o ~o o.
o I no o
O n
i o o
d o
I a- o a o LL -
e a >
_g 30 i ~. j\ s -
3g
=a i
U om #8 I d- et IJm i 3, -
c, c
o WM 18 u O. UC *S I H to U
g n
r
_ oo cc 1.D c e-8*
ou mW 22, H Td a cc x :=
IU _J d-c-
du oH
=* u-og a
a- 26 Io to 5
c__ _
c-
- c. C - t*
oU gr me ta E O ~
85 IU to
~
oH
-8m e
l '
_a o.
o 2 N
=
w I .
l o
.l
' o d
(l 1
w i ' i i ' i '
g 41 0*SI g 21 0 01 gz O.S Sg 0*0 (gam 38nss38d l
lI
amm . sum uma mum umm uma man ums um sum SEMISCALE TEST S-LH71 en BREAK hASS' FLOW RATE N, i e i i i i i i i
~
1 o
w .-
i N~
\ CD Mn
-m t
W~
H Co !
MO _
MF6RK _
~ i M MFLOWJ 375000000 2 ,
i O 1
- _] in
- k ". _ _
i o
w p
it
.i.
l [O r .- r i.
i O
- t
. .-t !
w 7:. y j y ,
s ,.
x -
Wm g .-
l r ,' ., _
Q i 03 d d j o
< a.g i
o . i ac ,-
-50.0 0.0 Sb.0 INO.0 150.0 2NO.0 2NO.0 3NO.0 3N0.0 4NO.0 450.0 500.0 TIME RFTER BRERK (S) i Figure 4.2: Comparison of Measured and Calculated l (RSAG18G, BE) Break Mass Flow Rates l.
1 .
- E3 #* Dee I .
I l o.
Q o
I _
G Q o.
I .
=:
- MC O
a h _oo' 3 I ~
CD
,e m
9 I
-g 3o
.l
?
~ N3
- C
- bN O LO -A I_J i Z
u_
O o
-T U-
_o o" dae Wy 30 m o --~ "
o C *$
I f-~
LO y
C 3)~ o
~
O C t
- Ce U
CD
'8 Et L.
W
=e If~u r c
o_ m c
No to q, oe
.tt - o' W oH
==
e o sa2
__J Q:
H CD I *
- Lt. g ",
IgC]
[o C
- o. C ' =g oU u
ae x
a '
EE uh g I E U,
v U
c-C -
e
- o. b 7,*
I *,.
~ <
-3
- o o
3 I
~
s
..g.
- e o I o
~~
i o *
. o o
I '
0.*000I
. 0*009 0*009 0*N0k 0*N02 0*0 meM) HISN 30 OInB Ny3yg l ,
I ,
1 I. m .e_ ,, _. .
I I I I i 1 I o.
o o
I op o I
i -
i
(
- o. .
,) ~ o o
I o
> cn i i a.
- - 0 w o
3 e
- O se d o =
p' .
I _d ?
e
, o n a
t <
~ m v
- ti om? -
-8
- 3
,f 1 o x at
- c., c p I ..
.ou, o z ma C $"
' c2 I o. x ious
\. o U mu o s a=v=
+ n_ c x: -
I .
i _d o
u a m.
i: g r 5:
I \
_b oH um
~a8 OO O oe I
8 OO ,' N U :::
OO .
OO b, o :;
88 OO : _o a
I o
- Lt1 .
1 1
88th
"ri ~ ~ ="-+ #
~
. . . . .o'o o.
I :
~
au-O hhhh .
~
a.
o 1
n , :, o I 0 r t --
0 0 kid 0 cid a Old*0 90N'O 90N'O FON*O EON'000*b I ( S /9'>I) 3188 MO~lJ SSUW I
M M M m M M M M M M M M M M M M M M M l,
RELPRSYR CLRCULRTION OF SEMISCALE TEST S. :
RSRGISC o "RESSURIZER LEVEL ,
t o' i i i e i e 8
9 -
]-
w E
C)
"9 -
CNTRLVAR 000000006
__J O -
U M LPRZ w
3 o .
88- ^'
i o l O :
_J o O-o l
J,
~~...._.. .
g j i
i 1 9 !
o_ I I I I f I I I '
-50.0 0.0 50.0 100.0 150.0 200.0 250.0 300.0 TIME RFTER BRERK IS)
Figure 4.5: Cociparison of Measured and Calculated (RSAG18G,BE)
Pressurizer Levels
I :
'I I O.
o I
l I i f I-O
!' W
'i W O 8 -
I -
g
-S, .
8 o l -
e m
_a e T
I w au r
_a J O
a M
y IZ _] - _o"
?D O
be eE ca (q
g _3 L.) MM " ao I mH H
~-
m a T 0%
,N O u IE g.a=:
U C EE IH U
.C L-)
O&
o*U CH E 3-wn J M ~W 8 W *y I. C O E-
_d u o[ u-N 03 - Er a5 IorLO 96 G
-8 4 I
~
- t' -
R b I _
. _o W
6 O
I 0 0Sk i
0 7 O'00k 0*0SE i
0'00E i
0'052 i i 0*002 0'OSI 0
-T l 4 (W3) 73A37 OU3H 83ddn I
I - . .
.4db.-
~ m.o. - . - .
- ~ . .
I O. '
, i t i O 1
o T. : :
O C.
I -
~
8t O,
l r
o +- '
- i O .
CD '
O I ?
_oO N
a m
8 U
1 b
- QW m
b I1J p c-
-@m, S
s 3
i C oo e y- =
Co cr I Lg.
- t L
oo S
C,y a
s t
D
_ oo
-- lor ,on e2 8 a.
I,.CC (D CD ay
-M
- x so
^
y O_
- gb. oH ICW
) oo v L_
>> c ' Q*
~
=
g . -
m_ b ct.r 5 C D I"E U O
L p o. H 6
CD l g -
, -@ s g -
./ m ;
c_ g I _
../ -
-8 h I O.
h O.
O I g.1 01 9'O g
90 j
VO E0 00 O
g N011088J 010011 SSUdl8 s I '
I
- . . . . . . 1 8 1 i
o.
l l o l l ,....-..-
o a.
+..,, ,a o
E ', ' * .
o.
I C
_o o
CD I -
e z- *
=
o d
om I --*
(f)
- II J
\
C-
~
~
3 i.' . o dM e
LG 88 oc I !--
J gg *oe .g
=
LO O, c co e
_o m I.O_ U **
xx EE de o&
U E-a G
J '] er 9e
_ oo C-cOu Du sI , CO U e u
g a E <
IW E LO y
ca H
O-9~
o E-o n
U i 9 CD , <
-I mg o_ '
o d 8 o .ae
! OJ A
I. r i
O o
'.- ~
E *
'~~ 2 $-
o.
o I O'0001 I
I 0*009 I
0*009 I
0 00F i
O*002 l
0*O 0*00E-S13A31 OIn0Il 38n1 6518 l .
CWD]
E lI
I 1 I
I I , , , , ,
e s
I E8 o 88 d I 88 o 88 QQ o
!! -No 5 ad c EE o I 7 o
y-ao "m
a r o o- -
y
-J o b o_
_ox o a- a E
C y e J .a e-
,G o I L W c. C2 j
u -
o o
. c._
o- oe a I "y J W J
O L_
W H
G m
c _ o c.
I o t.n W_
u J U 8
~
E u ob $e IwLO a
o-
-@ n H
C.D
- tn a
- J o
- _o m
=
3,.
l
.........- s
! O.
, -g
,E c
. o I r 1 i i l i o
l 0*0001 0'009 0~009 0~005 0*002 0'O 0 002-I (WD) S13A31 OIn0Il 3801 SSll I
1 I
I I O.
C i ! i I l I
O
...__ O.
, , . . . . . . . . . . . . . . yn bb I -
s.
88 88 88
-8m .
...a, I _
': ' 88
_b o
a y
...r
9E QQ Q I O C
- C- Q
, a ~ (.O IA -
I z
1 C
O CO C "" ~
N I
- q:. g
- . o y #
Q K 3 LO -
J " o. O ca Iyu e
d O_
o
~
E-
_C oe U
"i n
sai
[._ m J 3 og ITo LO w
a-U d
c T
C-y y
y
.e Ioco w-C es o
t<>
C I CD J
~~- ~ .. '
. . . -g o.
1,,
g I . . -
m o.
[
~
. O C. O I . . . . . - .
c ..
c.
I O 1 i i I i i I 0*00E 0*002 0*001 0*0 0*001- 0*002-0*00E- ;
g (WO) S13A31 0InDIl Sdl8 I
I
E .
I I , , , , ,
9 s
l es o e o
88 d I -
gg oo OS O
o CD o
I _
E m E.
[]
_a o
NY I W C-au E
o.
_om N
I C g I I
_J O
O o.
~O M m I m
Q- oC Q H
m a $, o*m .
LJ Q ,
a,
~
I c_ !
~ 'eu[ -
N ~3
u C w
++ $ o. g E-- '"..
C.: d-
-@ C. 8 W > / e u G o
E O. E , .D-I J &,
o (j-) O_ -
' oH ! ..
_.) _ .
J' b O I
e 9 .
~
_oo to C
N y I -
u o k) ., _ oo' I ,.
s
~
o d
I 0 00E A
y 0'002 0 001 i
0'O i i i O'001- 0'002-O*00E-(WD) S13A31 0I0011 Sdll I
I I . . . - - . - -
e I
I ,
o.
o I t I
o I s, o
I O *
- O o 8 o
. O cn .
O O
I
- s.
- E d R3 8 nis ?
I - -
35
_a o.
oe t
-m gg e-e >
bb IZl d l
0
~
o 13
_oM a c, $-
I ~C
'1 y
J
-g 8 C, u
C 13 u
5?
5 g ==
Ius y
O_
3
_a
- LJ o
z H
u Q
_J ._j O U_. 8 "-
I$LO y
D-
_b T 55 E o e<
.E g- E OE IEo co _
_aHo M
1 N
I '
i
- .i _a g
. ~ . . 0 X o.
c .
I -
[
7 o
~
C.
I 0'001- 0'002-i 0 00E-i 0 00F-i 1 O'005- 0 009-o l (WD) 13A31 OIn0Il 13SS3A I
I --
r
see O--
- , egeheh Me . emto e e e gm e - ee. ph*Meen he .
I I-I O.
I 's
'}
O
~
I 8 !.22:... -
_o o .
O w.
O C : .,
O :.. .
I E (D >
Y.' b,
_ Ob O
O hk ,;I-b5 -
o 3h
~
- g_ .p
]
.~ " cn as I?uJ i
U a ~
o*
- ao ve e3 c
q J- _
oc eu I (y[q 3 C c C
U O
m$
00 as o ~
o .
g ;o S a- o o e H
-a om u O L I g_) r C) 0- - @C "or e -
US
$5 o es Iur(f) o
.' _ o E-*
m a-E .-
R I T %
s:
O.
I
=
.===
'*****, C I 0'O 0~001-I 0*002-i i 0*00E-
).
1 O'00k- 0 00S-f l [WO) 13A37 83WOONM00 I
I
--.o. m m.e . m. .....m . ~ ~ - . . . - .
I I
O.
I I I y , ,. -
i !
O O
O I -
eo' .
.r.;;
s.
~ . ,
~a Q
O m
I p
O x
O CD I -
O O
O e
c cm aJ 3 sa N
.a- g*
II ,.
_b' OD u
0 3-g O em I (d I
.O g
L (D y g c y
@g g
I h o- -
La.
~ ~
ww .a c-
~c *n H -) Oy ww D .
r o L.3 o so OH IUC c.3 3
D-
'a O
C v L_ - o e O C g
0$
uo
_ $ L.] E.5 IOr
~ _.) -
Rr -
85 e
(,o O. H I
8 O
-8N 9 et O o O w I _
o a
n v
_O O
=
m
=
o, I -
33 ca OL cr
_O O
O I
C O.
l l l l
E0'O 00'0 I Ol*0 80'0 90*O (S/SM) M0lJ 801910Wn009 ll FO*O I
M M M M M M M M M M M M W M M M M M SEMISCHLE TEST 5-LH-1 o BL flCC'UMUL)lTOR FLOW i i i i i e i 7 i i e o ,
! (D co N 9_ _
co 0
- M F Z O@ -
- d d-I f1CFLBL
% NFLOWJ 530010000 Q
H <<
9_ _
r .,
o R, '
E .~ l '
D < i O J am . .
_J %
G3 l
I I I I I I I i 1 1 I
-100.0 0.0 100.0 200.0 300.0 400.0 500.0 600.0 700.0 800.0 900.0 1000.0 TIME RFTER BRERK [S1 :
Figure 4.15: Comparison of Hessured and Calculated (RSAG18G, BE) BL Accumulator Flows
num num nun aun num uma sum num aus um num uma sum uma sum uma suus SEMISCALE TEST S-LH-1 o CLRD TEMPERATURE. EL 115 CM.
O i i e i i i i i i m .
l 8- -
i M*
m tij TC2115
& o. a-u HTTEMP 1500C0307 O 8-HN o--o HTTEMP 150000407 -
T T
LaJ O_ o
- I- -
LaJ O Oa HmL ,,
Q C
go .
C O toh-
. ..g o .
i i i i i !
I i i i
-100.0 0 J 100.0 200.0 300.0 400.0 500.0 600.0 700.0 -800.0 900.0 1000.0 I TIME RFTER BRERK (S) j r
j Figure 4.16 : Comparison of Measured and Calculated (RSAG18G, BE) Clad Temperatures, EL 115 CM
mun em uma en aus sum um ums um aus uma uma sus aus sua num num SEMISCHLE TEST S-LH-1 9 CLRD TEMPERATURE EL 181 CM.
o i i i i i i i i i CD 9
o -
o-M*
m TC3181 0-0HTTEMP 150000607 U
M a.
D -
H @-s C -
x '\
W '
Q_ o g
E
_J 9 -
c.] g _
m _
tn 8
I I I I I I I I I 0.0 100.0 200.0 300.0 400.0 500.0 600.0 700.0 800 a 900.0 1000.0
- TIME AFTER BREAK [S)
Figure 4.17: Comparison of Measured and Calculated (RSAG18G, BE) Clad Temperatures EL 181 CM
I O.
I I I I O O
O
~~
I
+
l C.
0 0
! O 1 :.
m I -
e O.
O Q
5 j
I :
O.
mn 3 ".s
.ae tal d
I A
", o- !
-8~
w 3.
UD a
- u
! O
_g y j.$g, I
v; -
a_ /,,,. ............. ........ ...... cc ma m d N ,*.
- *oz G.A .o
- 3 w '
J O. g ea Iy ,
F y
"2~
- a '
O OZ
- D c
EU wa
-4
- ,,- ..,. O o sa y - m M H c I d tu LJ C.
z.
LJ
..*-.,}
t O y' OC O
gg
,,, m u-e0 h U I 2 y
D G
w J-O.E OH 00 m
u : O I
~
ss : m l
OO r ,
N CO *
- OO
- e OO O. u l
\ 00 : a 3 00
- to 1
pp
- O een
- N m.
CD CL 0.
NEE O.
N t- U. U [] O m s a I. C.3 6-* l-e t~ Z Z I ~
- r1 < I I ,,,,,,,. O,
. O I ) 38f11983dW31 0910 I
I
- m. . e me.,% e.... .
- e I .
I I' s g
l i i i i I oo e a o .
- -8 5 o .,
I V O _o 9
o-
%a
';'d oa
- g b d I1 _J i
n] o Y
v5 eC C
1 C d- _8 C et o w a.
Il -
(/)
(.tj RQ C
EE G - CC _a ec IH LJ c-
.H
.C o
L 1 H
TU e
_) CC 9 L_ og IC C.3 07 sLJ CL -
r a C'
-@ u Oco
$G
- o. <
- y r 8e r H o, .- u-Iu07 Q-
_o H S
o C ss 1 I 0 J g@
88 88 i
_d o i y
~
ce R I
~
o Geh o _d SWW I
E'EM n .,
o g _d 0*006 0 009 0'00Z 0*009 0'005 0*00k
(>ll 38fll883dW310970 l
I A lI
uma uma sum uma num aus em uma uma amm ' ems sus sua muu.
SEMISCRLE TET S-LH-1 l
o I.:LRD TEMP $RATURE EL 351 CM. i i i i i i i i [
o i g -- '
m -
g I
O. _
O~
l
^8 M
- TB3351 o-oHTTEMP 150001107 i
- --a HTTEMP 150001207 Wo D
H @x-T W
Q_ o r -
W83 H fO
= C -
O C
a -
oo_ .. __
8 8 i i 1 l I I I l l 0.0 100.0 200.0 300.0 400.0 500.0 600.0 700 0 800.0 900.0 1000.0 TIME AFTER BREAK (S)
Figure 4.20: Comparison of Measured and Calculated !
(RSAG18G, BE) Clad Temperatures EL 351 CM
4 I 9 I I I o
O h
I ss OO s(
_o
- i
- M9 :
S I oo Oc mm 3
o
_ &~ _o g an c. c.
- o o
-EE
-WW w M
NHH OHH "Ir _a a id I~E _ no u
- b
^
jg i O co 'k. m d c.
o A tn O o v doa 8
_.] O O
l M
{
Eb QG J
LIJ U y @'
I .G u o
-8 mQZ 8s av
% ee
% M
? EU U
I 'U j
E_
Z
_o QH o 2:
u v L_ e "3 g C' Eel IOoH E:
o H- -8 U
- =
UG E o MI kh um c- g IaW J O_ .
_O O "
E I
N 8
_o 0 a
- k.
une
?
r* O n -
I 0*009 0 004 t
0*009 Y
0*005 l
0*00k 1
( >l ) 3801983dW31 0910 I
I - - - -. .. . * - . .
E E E E E E E W M M M e m m e e e e i
4 i
SEMISCALE TEST S-LH 1 i
i o CLAD TEMP $RATORE ELI 181 CM.I I i 1 I I I I I '
C3 o
m o - TU3181 i g- c HTTE1 150 607 _.
_g M*
Uo Oc -
a g- E .
s s T i Cr -
y .
e.
o_ o !
i .
r y 8-se C----------.............f__ _
f g- . . . ;
-J O g_
to o
? I I I I I I I I I i 300 400 500 600 700 800 900 1000 0 10 0 200 TlME FlFTER BREAK f S1 Figure 5.2: Comparison of Measured and Calculated (RSAG18G, EM) Clad Temperatures at EL 181 CM
- t I
m m m m m m m m M i
SEMISCALE TEST S-LF1-1 l 9 CLfl0 TEf1PERATURE EL 228 N o i I I i i i i i i 8
l -- lt3228 9 o-o HTTEMP 150000707 g_ *-* HTTEMP 150 00807 _
M*
M T o. !
1 ao e i i
1 HR C o
- & .'t 1 o w
- i l
O_
r ao { ! <
- wa*
L, A i 1 m -- . . . . . . . . L- :
a -
c J9 '._.. ---------_
o a_ . , , , , _
m 9
8 e_ l I I I I I I I I O 100 200 300 400 500 600 ' 700 800 i 900 1000 TIME RFTER BRERK (S)
Figure 5.3: Comparison of Measured and Calculated I (R5AG18G, EM) Clad Temperatures at EL 228 CM
" " "' EN' " ein mum nun um umu um seu muu mun uma sus
. SEMISCALE TEST S-LH-1 _ ___ __
~ '
o ' CLAD TEMP 5RATORE EL i 253 i CM. i i d i e i i L 8 -
i o TB5253 -
6- o-oHTTEMP 150000807 '
ba# -
o a-a HTTEMP 150000907 M*
m a 0N o g N9 . t W. a o g- _
Hs
. C l z ,
U i 1o g' -
e
- E- -
W 8- i i 1' Hw o -
i C
_J 9 -
. - ---.. i c_) a _ ._,, _
m o, '
8e I I l l l l 3 I I 0 100 200 300 1 400 500 600 700 809 900 1000 4 TIME RFTER' BREAK IS) ?
Figure 5.4: Comparison of Measured and Calculated 1 l
t (RSAG18G, EM) Clad Temperatures at EL 253 CM i
.i
uma e ums
.i SEMISCALE TEST S-LH-1 i
! o CLRD TEMPERATORE EL 351 Cri.
l d i i i i i i i i e i 8 i
i o
1 d -
^M 8_ o-oHTIEMP TB3351
'i 150001107 a-aHTTEMP 150001207 W
& o.
->o_
o -.~...
a-x U '
Q_ o N
Ef Uo[ ,i C3 a- -
J ~ '
O o.
L . .~ .. . .
8
! O. .
8 i i , i i i i i i 0 100 200 300 400 500 600 700 4 800 900 i 1000 TIME R.FTER BRERK [S) .
Figure 5.5: Comparison of Measured and Calculated (RSAG18G, EM) Clad Temperatures at EL 351 CM
-_ --- -