ML20010D612
| ML20010D612 | |
| Person / Time | |
|---|---|
| Issue date: | 06/17/1981 |
| From: | Advisory Committee on Reactor Safeguards |
| To: | Advisory Committee on Reactor Safeguards |
| References | |
| ACRS-1854, NUDOCS 8108280370 | |
| Download: ML20010D612 (7) | |
Text
_
Wd-M[Y MINUlts f~..
N-PO8 8f6/fj g
REACT 0D. FUELS /ECCS SUBCOMMITTEE MEETING
.( [/n"/el
,g MY 5,1981, WASHINGTON, D.C.
b s
The React 6r Fuels /ECCS Subcommittee held a meeting on May 5,1981 to review regulatory treatment of DNB, and DNB correlation, in use or under develop-ment by CE, B&W, and Westinghouse.
PRINCIPAL ATTENDEES:
"J
~
P. Shewmon, Chairman, Reactor Fuels W. Johnston, NRR 7 +}
M. Plesset, Chairman, ECCS L. PhillIps, NRR fQ hg f
M. Carbon, Subcommittee member R. Van Houten, RES iLN H. Etherington, Subcommittee member R. Marshall, CE
/g Bggg\\F 3 W. Mathis, Subcommittee member D. Lawrence, CE pN$ 1 Q.j P. Boehnert, ACRS staff G. Geissler, B&W 6-D. Bessette, ACRS staff K. Kneidel, B&W
' 9 co' I. Catton, Subcommittee consultant D. Farnsworth, B&W
'h T. Theofanous, Subcommittee consultant D.Dzenis, Westinghouse./X
j Z. Zudans, Subcommittee consultant J. Lee, Subcommittee consul *, ant R. Shumway, Subcommittee consultant A. Acosta, Subcommittee consultant Attached is a list of documents considered by the Subcommittee.
No written or oral presentations were received from members of the public.
LICENSING TREATMENT OF DNB L. Phillips, NRR, summarized the regulatory treatment of departure from nucleate boiling (DNB).
10 CFR 50 states that fuel failures are not allowed for nonnal operation and anticipated operational occurrences ( A00).
A00 events are a well-defined set of transients that the vendors are required to analyze. They are considered to be events that may occur on the order of once per year, to once per plant lifetime. One criterion to be met for A00s is that there must be 95/95 confidence / probability that the most limiting rod in the core does not enter DNB.
The departure from nucleate boiling ration (DNBR) is defined as:
DNBR = critical heat fiux actual heat flux
[
8108280370 810617 4
Reactor Fuels /ECCS '
Dr. Catton indicated that during full power normal operation, the DNBR for the most limiting point in the core is greater than 2.
For the most limiting A00, the minimum DNBR must be greater than some value (e.g.,1.3 for the W-3 correlation).
The DNBR limit (e.g,1.3) is the 95/95 confidence / probability value for the CHF data base, in other words, when the minimum DNB ratio is 1.3 there would be a 5% probability, to a 95% confidence level, that the limiting rod would enter DNB.
i The old method of perfonning a licensing evaluation for an A.00 to assure nonviolation of the DNBR limit, involved assuming each pertinent parameter (e.g., reactor power, flow) was a worst case value. Methods now under NRC review involve statistical combination of uncertainties. A second change that is under review is to use an open channel model rather than a c1csed channel model. Implementation of these changes would allow a power increase of 10%, if no other licensing criteria (e.g., LOCA) were limiting.
Mr. Phillips indicated that high burnup fuel will have low reactivity and j
low power, and would be, therefore, less likely to enter DNB.
FUEL BEHAVIOR FOR OPERATION BEYON'D DNB R. Van Houten, RES, summarized experimental information on survivability of fuel that operated beyond DNB.
i There was some discussion on whether CHF curves developed from steady-state i
tests would be correct for transient conditions.
Drs. Catton and Theofanous l
i indicated that for the purposes of A00s, the conditions are in effect quast-static.
1
Reactor Fuels /ECCS r
2 Nuclear plants have nominal rod heat flux ~es of 200,000 to 250,000 btu /hr/ft,
2 and peak 'fieat fluxes of 450,000 to 600,000 btt/hr/ft.
Figure 1 (attached) shows the survivability of fuel rods operated beyond DNB.
Figure 2 shows the survival / failure curve in relation to transients, including ATWS (not including LOCAs).
Dr. Carbon noted that there seems to be a large margin between DNB and fuel failures.
CE PRESENTATION ON DNB R. Marshall discue,ed the CE-1 DNB correlation and the TORC code. The correlation was developed from data from the Columbia Universit; facility.
The TORC code.was derived from COBRA-I'l-C. The Columbia facility employs a 5x5 electrically heated bundle, with length up to full-length and unifonn and nonuniform power profiles.
Previously, CE has used the W-3 correlation and the COSMO code.
The DNBR limit determined by CE is 1.13.
The NRC in their review, however, decided that 14x14 and 16x16 bundle data were two different populations. The smaller number of data points in each population resulted in a larger standard deviation.
Hence, a DNBR limit of 1.19 was imposed. CE, however, has per-formed statisticci tests that show the two data sets belong to the same popul ation.
CE had CHF tests performed at Winfrith, U.K., and the data were compared against the CE-1 correlation. The CE-1 correlation accurately predicted the Winfrith data.
b*
Reactor Fuels /ECCS,
B&W PRESEMTATION ON DNB G. Geissler, K. Kneidel, and D. Farnsworth summarized the B&W CHF correla-tion. B&W performed CHF testing at their Alliance Research Center. The CHF facility has a 10 MW power supply and cost $3 million to construct. An acoustic monitor was used to determine the onset of DNB.
B&W employs the B&W-2 (old) and BWC (new) correlations. The LYNX-2 code, which is derived from COBRA, is used to model themal-hydraulic conditions.
The minimum DNBR foi nomal operation for B&W plants was said.to be about
~
2.3.
WESTING 4OUSE PRESENTATION ON DNBR D. Dzenis summarized the Westinghouse CHF correlations. He indicated that nucleate boiling does not occur in operating plants.
It was also noted that LOCAs are separate in a regulatory sense from anticipated transients.
The data base for the W-3 correlation was single channel tests with an unheated shroud. The WRB-1 correlation is based on considerable number of l
tests of rod bundles, with the testing perfomed at Columbia University.
l Use of WRB-1 results in a DNBR limit of 1.17.
The data from various bundle configurations are consistent and match the WRB-1 correlation. Currently, no kestinghouse plants are DNB-lhited, all are LOCA-limited.
Appl ication of the new analysis techniques (i.e., statistical combination of uncertain-ties) provides about a 12% power margin compared to the old analysis.
Use of WRB-1 compared to W-3 would provide some additional benefit, on the 1
l order of 8%.
w
1.
BWC Correlation of Critical Heat Flux in 17x17 Geometry Rod Bundles: BAW-10143P Jan 1980 2.
Lynx 1 Reactor Fuel Assembly Thereal Hydraulic Analysis Code: BAW-10129 Oct 1976 3.
Critical Heat Flux Correlation for CE Fuel Assemblies: Uniform' Axial Power Distribution CENPD-162-P-A Sep 1976 4.
Critical Heat Flux Correlation for CE Fuel Assemblies with Standard Spacer Grids:
Nonuniform Axial Power Distribution CENPD<207-P Jun 1976 5.
TORC Code CENPD-161-P Jul 1975 6.
WRB-1 Correlation for Predicting Critical Heat Flux in Rod Bundles with Mixing Vane Grids WCAP-8762 Jul 1976 7.
Improved Thermal Design Procedure WCAP-8567-P Jul 1975 8.
NRC Presentation on Licensing Treatment of DNB - 5 slides 9.
Survivability of Fuel Operated' Beyond DNB - 15 slides
- 12. Westinghouse Presentation on DNB - 24 slides a
--n.
i.
hyaI l
,1 Comparison of in-Reactor Post-DNB Survival / Failure Data for Zircaloy Cladding with FBRB/ANL Zircaloy Ductile-Brittle Boundary Curve 3140*F 2000K y
i A
N N
L A Y\\ 8 s
270]*F 1800K
's A
a N\\ sy A
gk
\\
6 A\\
- g,,
a 2420*F 1800K a
"Vi h
U 5 1400K
%,5 E
D E
c%
O 9\\g
-5 1200K N
N 0
N
\\
I w
1340*F 1000K A PBF/PWR Y NSRR g
O CR/BWR Q
C Ductile: (No Red Failure) g A PBF/PWR VN5RR O
930,F 800K -
O cR/BWR O
Q SGHWR/BWR O GETR/BWR O Halden/BWR I
I I
I I
620*F 600K l
1 10 100 1000 10,000 100,000 Time After DNB (Seconds) l t
~
....*e,
.-=
e
-~
w.-
g-,
m
.m-
---g-w w
y
' SAFE' EQUIVALENT CLAD TEMPERATURE (S)
(NO CLAD FAILURE BY OXIDATION) 2420*F 1600K g
N
'N
\\
Probable Brittle y
N N
Clad Failure Ductile I
Cladding
\\
\\
s E
After Event N,
i s
g N
i y
s i
s
~
u
\\
s E
9 N
N, vD N
N, P
~5 1300K (1880*F) 3 l,,
i a-gic
\\
\\
w F
o s
I
~
' Safe'
~l8
\\
\\
1700*F 1200K Equivalent Sl $
Ductile N
\\
i Temperature E
(l18 i
Cladding Region After Event wl I
I I
i 102 103 104 105 Seconds i
I I
I I
l 1 Min 10 Min 1 Hr 8 Hrs 24 Hrs Log Time at Temperature
~