ML20073L296
| ML20073L296 | |
| Person / Time | |
|---|---|
| Site: | 05200004 |
| Issue date: | 10/05/1994 |
| From: | Marriott P GENERAL ELECTRIC CO. |
| To: | Borchardt R NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM), Office of Nuclear Reactor Regulation |
| References | |
| MFN-128-94, NUDOCS 9410130049 | |
| Download: ML20073L296 (13) | |
Text
m-i GENuclear Energy
, P. W. Marriott. Manager GeneralEketric Company AdvancedPlmt Technobgies 175 Curtner Avenue. MC 781 Sm Jose. CA 9S125-1014 4069256948(phone) 408925 t193(tacsimik)
October 5,1994 MFN No.128-94 Docket No. STN 52-004 Document Control Desk U. S. Nuclear Regulatory Commission Washington,1). C.
20555 Attention:
Richard W. Ilorchardt, Director Standardization Project Directorate
Subject:
NRC Requests for Additional Information (RAls) on the Simplified Boiling Water Reactor (SBWR) Design
Reference:
Transmittal of Requests for Additional Information (RAls)
Regarding the SilWR Design, Letter from M. Malloy to P. W. Marriott dated April 8,1994.
The Reference letter requested additional information regarding SilWR fuel bundle testing in the ATIAS facility. In fulfillment of this reouest, GE is-submitting Attachment I to this letter which transmits the responses to RAIs 900.63 and 900.64.
Sincerely, h
3 P. W. 5 rriott, Manager O Advanced Plant Technologies k J. M/C 781, (408) 925-6948
,, " Responses to NRC RAls" cc:
P. A. Bochnert (w/l copy of Attachment)
M. Malloy (w/2 copies of Attachment)
F. W. Hasselberg (w/l copy of Attachment) 120 C 4 /)
.pt(D 9410130049 941005 PDR ADOCK 05200004 A
Fuel Performance Testing RAI Number: 900.63 Question:
In a letter dated October 8,1993 (MFN No.164-93), GE Nuclear Energy (GE) responded to the staff's request for additional information (RAI) Q900.1 regarding fuel performance testing. Contrary to the position stated in Q900.1 that the staff considers the ATLAS tests to be part of the testing program for SBWR design certification, GE's response asserts that this testing is not part of the certification testing program.
The staff reaffirms that it considers the ATIAS tests to be part of the testing program required by 10 CFR 50.47(b)(2) for SBWR design certification and requires GE to revise its documentation on the SBWR testing program to include the ATLAS tests.
GE Response:
10CFR50.47(b) requires that :
The performance of each safety feature of the design has been demonstrated through either analysis. aoprontiate test orocrams. exnerience. or a combination thereof: and that sufficient data exist on the safety features to assess analytical tools used for safety analysis. GE believes that there is a suflicient data and analytical basis to predict critical power for the SBWR fuel bundle within i5% This assertion is substantiated in response to the next RAI (900.64). This value of the uncertainty will be used in the MCPR analysis for SBWR transients and accidents.
SBWR unique fuel performance testing will be performed as part of the detailed fuel design process, and will be performed after SBWR fuel is ordered. This should allow the uncertainty in MCPR prediction to be reduced to approximately 3% consistent with GEXL predictions of other fuel designs.
.I_
RAI Number: 900.64 Question:
GE's response to RAI Q900.1 (MFN No.1f493, October 8,1993) does not include the details necessary for the stalT to perform an evaluation. Please provide, in much greater detail, the following information:
Detailed information to support GE's conclusion that the code and the model a.
are applicable over the range of geometric (e.g., fuel length, pitch, diameter, spacer configuration) and the thermal-hydraulic (e.g., mass flow rate, power profile) parameters representative of the SBWR fuel design.
- b. Provide details of the planned testing, including the test specification, te<,t matrix, and test schedule.
Provide details of the analyses planned in conjunction with the fud c.
performance test program, including documentation for the thermal-hydraulic correlations to be validated for use in SBWR safety analyses.
- d. Provide verification that the ATLAS facility is able to match approximately the appropriate thermal-hydraulic conditions for SBWR fuel. In particular, how will ATLAS simulate the natural circulation flow behavior that will exist in the SBWR?
GE Response: (a, b, c)
The SBWR fuel bundle is of the GE 8X8E and/or GE 8X8EB design. The GE8 design is an 8x8 lattice with two water rods. This design has been licensed by the NRC (Ref. 900.64-1) and has been operated extensively in operating reactors
(>l million rods). The SBWR fuel will have identical fuel rods, but of a shorter length (9 ft vs 12 ft fuel column). The fuel bundles will have identical lower tieplate, spacers and upper tieplate. Thus, the lattice configuration and local pressure drop c apmmnts are identical. The only factors that need to be considered are the durence in length and the range of flow parameters during normal operation.
The bundle critical power is predicted using the GEXLO2 correlation. The GEXLO2 correlation has a data base of 1516 points. The critical power is predicted with a mean value of 0.9998 and a standard deviation of 2.68E The range of the data base is :
6 2
2 0.1 to 1.50x10 lb/ft.hr (130-1950 Kg/m -s),
Mass Flux :
Pressure: 8001400 psi (56-84 bar),
Inlet Subcooling: 0 to 100 F (0-55 K),
Axial power profiles: 888 points cosine and 628 inlet-peaked, Local (rod-to-rod) peaking : various; uniform to 1.6 corner to 1.4 interior.
These data are for a 12 ft heated length.
1 -
l
4 I
l The SBWR operating conditions of l
6 2
2 0.48 -0.63 x10 lb/ft -br (650 -850 Kg/m -s),
mass flux:
pressure: 1040 psi ( 71.7 bar),
j subcooling: 30 -47 F (l7-26 K) are well within this data base. The only remaining issue is the effect of length.
t This is addressed next.
First, it should be recognized that the boiling length type of correlation is well suited to predict the effects oflength and axial power shape. This was one of the main reasons it was chosen to replace the local conditions based CHF correlation. Figure 900.641 shows data from various 6 ft and 12 ft lengths correlated by the boiling length correlation. The GEXL0l correlation (which j
preceded GEXLO2 and is of similar form) included in its data base 84 data points with a 16 rod bundle with a 6 foot heated length.
l Second, it is well known that the plot of critical power vs. length flattens out with increasing length. Figure 900.64-2 shows data taken at Columbia University for various lengths. At SBWR mass fluxes, thc difference in critical power between a 9 ft lag bundle and a 12 ft long bund'e is less than 10%. The error in predicting thi3 efTect is clearly a small fraction of this absolute difference.
(
Third, GEXLO2 has been compared against predictions from a mechanistically based, accurate subchannel analysis code (COBRAG) for the SBWR bundle.
COBRAG is a multi-field two-phase flow model. The vapor, liquid film (s) and f
droplets in a subchannel are treated as separate fields. Boiling transition is l
talculated mechanistically based on liquid film dryout. Excellent comparisons A
have been obtained with a wide range of critical power data in full scale rod bundles. Figure 900.64-3 shows predictions of data from 8x8,9x9, and 10x 10 bundles. Figure 900.644 shows predictions for different axial power shapes.
Figure 900.64-5 shows a range of spacer types. Figure 900.64-6 shows a range of mass fluxes. Almost all data are predicted within 5%, with a standard deviation of less than 3%. COBRA predictions were also made of the Columbia data with a 6 ft heated length, with excellent results (see Figure 900.64-7). Thus, COBRA provides a basic and completely independent means of predicting BWR bundle critical power. In Figure 900.64-8, critical power calculated by GEXLO2 for a range of SBWR conditions for the SBWR bundle is compared with critical j
power calculated by COBRAG. The standard deviation of the difference j
between the two independent methods is less than 3%. When this is combined with the standard deviation of the error in COBRAG predictions (< 3%), an j
RMS value of 4.2% is obtained. GE proposes to use an uncertainty value (1 l
sigma) of 5% for the prediction of critica' power in the SBWR bundle.
l l
i 1
l 3-l J
~
1 GE Response: (d)
The ATLAS test facility can be configured to test a shorter length bundle. This would be done using the same setup which exists at present.
The difference between forced and natural circulation conditions, or 'hard' vs.
' soft' inlet conditions, is important when considering thermal-hydraulic stability. If the stability threshold for the bundle were to be reached before critical heat flux in the SBWR, the ATIAS results would be nonconservative, as the ATLAS tests are performed with 'hard' or forced flow inlet conditions. In actuality, the critical power threshold is reached well before the stability threshold and the ATLAS type of measurements are valid. This is demonstrated below.
The critical power for the SBWR bundle at rated conditions is of the order of 5.7 NIW. At this power level, calculations were performed with the stability code (FABLE). The decay ratio is in the range of 0.16 to 0.36 based on axial peaking (nominal vs. design basis shape). The core decay ratio at the scram power level is less than 0.5. Thus, there is a large margin to the onset of either channel or corewide (and therefore, regional) oscillation. These large margins to instability have also been independently confirmed by calculations made byJ.
Starch-Leuba at ORNL Thus, the use of 'hard' inlet conditions for determination of critical power is justified.
References:
900.64-1 Letter, NIFN-082-85, C. O. Thomas (NRC) toJ. S. Charnley,
" Acceptance for Referencing of Licensing Topical Report NEDE-24011-PA-6, Amendment 10, ' General Electric Standard Application for Reactor Fuel',"
Niay 28,1985.
0.36 l
V 0.32 g
~
E V
q E
V 1
d 0.28 D
O V
.A Y
5 EV 5
0.24 FREQN ANNULUS DjTA G/1(f' = 0.540 lblH-M SYM00L AXIAL HEAT FLUX SHAPE g
7 12 h UNIFORM g
O 6 h uNiFonu E
12 h COSINE d
4 h COSINE M
g OUTLET PEAK 0.20 T
INLET PEAK A
O 6 h HALF COSINE d
80 in. HALF COSIN E Q
0.16 O
1.0 2.0 3.0 4.0 SA S.O 7.0 8.0 g,0 J
BOILING LENGTH (h) j Figure 900.64-1 criticd Oudity n Boiling Lengt. Freon 114 Annulus Dats 5-
~
Critical Power for 9 ft. channel Critical Power vs. Length 3.0
^ 2.5-S G='l360;[
9 2.0-f
/
o 1.5--
/
g-OO d
1.0 g
=
b 0.5
=
C 2
4 6
8 10 12 Heated Length (ft)
s 8
8 x
s m
N, x Q
N
\\
x x
s
.o x
o s
's.AN 8
m x
s s
m x +'ss
'a xx e
g s4sN a
x s
e 0
N 'y'N 8
\\
T 8 0
x e
f N
x t '\\
Q<
N I
Q.
m
\\\\\\
c o 8
O c
s o
o 5
\\
e N
o q
e 8
88e w
G eea o
ans
\\\\
0 23s ss\\s
..+
Nh 8 w
o o
o o
o m
ho 8
8 8
8 8
F a
m e
s w
w q
(m)ovusoa mO
'g Figure 900.64 3
'tO.
COBRAG Critical Power Comparisons-Axial Power Shapes 10500 o Top Peaked
,/
,/
/
a Cosine f
,/
21
+ Bottom Peaked
+5% /
/
9500-
/
,4 /
a
/W
/
g
=
?
/
/ /
i /
/y' p/
t S
/
-5%
6 8500-
/ *'t /p a<
/
' /
/
'/
O 7500-e
/,l 6500-
/
/
W/
55SS'0 6500 7500 8500 9500 10500 0
sep 00 y
0 T
0 1
r
', E e
c
/
j a
', 'g e/
0 0
p f
/,
0 y O '/ -
S
/
9 t
/
5 n
/ '9'/
e
/
%/ a r
0
/'
0 e
/ j *,/
6 f
)
G f
+
5 w 8
/f /
i k
A D
(
A R
A
',/
T
/,
A B
s 0
/,,/'f/
D 0 O
n 67 C
os
//X i
/
ra 0
p 0
/
0 m
6 l
e l
t e
aeC
//
o rle u t
gr
///
i r
C ge n EFU
- s+
r
/' 0
/ b e
1-0 0
0 0
0 k
w b
0 0
0 0
0 o
0 0
0 0
0 0
9 8
7 6
P 1
ng Oh@U iac 3* 3 g *e C 1
i 2
t i
rC
.?
i1
\\1l i\\
i i'
COBRAG T
Critical Power Comparisons-Various Mass Fluxes 10500
/fI 2
- 678 kg/m f3 a 1119 kg/m f3
/
/4 2
2 y
f
+ 1560 kg/m f3 A1832 kg/m f3
/
pf 2
g
/
,/
y
/q4', -
~
y b8500'
,l./ & s 5,
/
/
$<75oo-
,y 6500-l
,/
/
, DATA (kw),
55SbOO 6500 7500 8500 9500 10500 O
4 5?l
/
e -SX
/x
/
O
/
/
j asm 7
m
/
O
/
/
/
u s
/
/
/
X/
^f#
/
/
/
m
/
/
y o
/
/
/
1500
/
.E
/ /
5
/ /
u
/ /
//
//
1500 2000 2000 3000 Critical Power (kw)- Data Critical Power Data for 6 Feet 4x4 Lattice Bundle -
Columbia University Test Data Versus COBRA-G Figure 900.64-7 11 -
m
.cer
.../
4
+5%
i sooo
/
/
/
-5s
,/
j
/
A/
- /
e 3
8000
/
/
/*
+ '+
a
/ ly,
/
/s m
'+
/
u
/
5
/
+,
ow
/
'iii
/
8
/
/
c cooo u
/*
/
/
/
'/
/
/
/
/
%-= s aooo 7000 sono sooo Critical Power (kw)-COBRAG Critical Power Data for 9 Feet GE8 Fuel Bundle -
GEXLO2 Correlation Versus. COBRAG Figure 900.64-8
- 12
,