ML17262A545
| ML17262A545 | |
| Person / Time | |
|---|---|
| Issue date: | 12/20/1978 |
| From: | Levine S Office of Nuclear Regulatory Research |
| To: | Harold Denton Office of Nuclear Reactor Regulation |
| References | |
| RIL-0042 | |
| Download: ML17262A545 (6) | |
Text
MEMORANDUM FOR:
FROM:
SUBJECT:
a UNITED STATES
.., NUCLEAR REGULATORY COMMISSIO.
WASHINGTON, D. C. 20555 DEC 2 O 197B Harold R. Denton, Director Office of Nuclear Reactor Regulation Saul Levine, Director Office of Nuclear Regulatory Research RESEARCH INFORMATION LETTER N0.42 : CRITICAL EXPERIMENT PROGRAM FOR NEUTRONICS CODE VERIFICATION This research infonnation letter transmits significant experimental results from a program of critical experiments which were perfonned for the NRC in the ZPR-9 facility at Argonne National Laboratory. The final report on this program will not be completed until September 1979, but the results from the comparison of experimental data to calculation using diffusion theory, transport theory and Monte Carlo methods completed thus far are significant and deserve early attention.
Introduction This program of distorted geometry critical experiments was planned and
.carried out to provide benchmarks for the validation of the neutronics part of codes used in safety analysis such as SIMMER.
A second objective is to validate the VIM Monte Carlo code for use as a secondary standard for validation of other neutronics methods.
Meltdown configurations in LMFBRs can be expected to have regions with high fuel concentrations giving extreme spectral changes and large regions of void giving rise to large streaming paths. Neutronics methods other than Monte Carlo have difficulty in calculating these configurations accurately. A series of experiments was needed to detennine the importance of these difficulties and to provide a basis for improving the accuracy and reliability of accident analysis methods.
As will be discussed later, this program has demonstrated that diffusion theory neutronics calculations would underpredict ramp rates which might occur in a meltdown and could be non-conservative.
1*-,*-,
I I
Harold R. Denton I I
Background
This program was initiated at ANL in FY 1975 to evaluate the ability of neutronics codes to calculate the reactivity of meltdown configurations in l.MFBRs.
At that time, there were no acceptable critical experiments
~
by which to test and calibrate the codes and data used to analyze meltdown accidents. The only experimental data available were from measurements made in ZPR-III Assemblies 27 and 28, ZEBRA assemblies BG and 12, and in the CRBR-EMC program in ZPPR.
The fonner measurements were perfonned 16 years ago in uranium fueled critical assemblies with significant compositional mismatches relative to present LMFBR designs.
VIM Monte Carlo calculations of these experiments have been made; however, the design of the experiments precludes a systematic validation effort (e.g., no unperturbed reference assembly was constructed, and only the eigenvalue measurements were made).
In addition, the loading records of these very complicated assemblies proved hard to decipher and were sometimes inconsistent.
The more recent CRBR-EMC measurements provided some prototypical results, but the small magnitude of the reactivity changes involved (< $2) made calculations using a Monte Carlo code extremely difficult. Thus, there was a clear need for critical experiments designed ~xplicitly to provide data for the validation of accident analysis methods including Monte Carlo methods, and a program to design such series of experiments was
- undertaken at ANL.
A preliminary set of experiments was designed an.d presented to the Fast Reactor Critical Experiment Review Group Meeting on September 21, 1976.
The Review Group made several worthwhile suggestions for changes and concluded that the revised program should proceed.
Arrangements were made with the Department of Energy (then ERDA) to use the ZPR-9 facility at ANL from July 1, 1977, to December 31, 1977.
On June 7, 1977, the program described in the enclosed document ZPR-TM-279 was presented to the RSR staff and approved.
The experimental portion of the program was completed on schedule in December 1977. Analysis of the data has continued through FY 1978 and will extend into FY 1979.
The enclosed report ZPR-TM-327 contains the major experimental results and a comparison of these results with analysis. The final report on the program will be completed at the end of FY 1979.
Harold Results A surrmary of the results from the program is given in the enclosed memofandum ZPR-TM-327.
The eight critical configurations studies are shown in Figure 1 of this report.
The analytic.al comparison with the experimental eigenvalues are shown in Figure 2.
The results of VIM Monte Carlo calculations on the critical experiments are given in Attachment A of this letter.
Examination of Figure 2 and Attachment A lead to the conclusion that diffusion theory calculations lead to non-conservative estimation of reactivity going from the reference to the slumped configurations.
The significance of the ramp rate at prompt critical iri an HCDA calculations has been pointed out by the NRR staff in NUREG-0122(1).
The parametric study on ramp rate shown in table 5.6 of that report shows that increasing ramp rate from $50/sec. to $100/sec. can increase the work energy from the HCDA by a factor of 3.
The reactivity change during the excursions reported in table 5.6 is of the order of $0.30 while the error in caleulating the reactivity change between the reference and slump-in critical configuration calculated by diffusion theory is over $2.0. It is apparent there is a potential for underestimating ramp rate at prompt critical and hence work energy in an HCDA calculation.
The tendency of the Sn transport and VIM Monte Carlo methods to overpredict reactivities and the difficulty in predicting material reactivity worth make one suspect that there are serious errors in the ENDF/B-IV cross section sets. It should be noted that the VIM calculations used a plate-by-plate geometry description.
As data on the detailed measurements are reduced, they will be reported in ZPR-TM memorandums.
Users who wish to perform a detailed neutronics calculation on these experiments, will need to have detailed drawings of the ZPR-9 reactor, fuel plates and loadi.ng patterns. This mater'ial as well as hom.ogenized number densities can be obtained from the Applied Physics Division of ANL through Dr. Dave Wade.
Provisions have been made to supply this information to ORNL and LASL safety analysis groups for code verification.
(1)
Meyer, J. F. L. Lois, J. F. Carter, and T. P. Speis "An Analysis and Evaluation of the Clinch River Breeder Reactor Core Djsruptive Accident Ene.rgetics 11 NUREG~Ol22, March 1977
I Harold Reconmendations and Future Plans Data reduction and preparation of the final report will be completed in FY 1979. A VIM Monte Carlo Calculation on Configuration 2 - Sodium Voided Test Zone will be made and the S cases not shown on Figure 2 of ZPR-TM-327 will be completed.
n Analysis to resolve cross section difficulties that cause differences between experiment and VIM Monte Carlo eigenvalues is needed.
The Applied Physics Division of ANL is the logical place for the work to be done.
The work should be supported by DOE rather than NRC since it is not really confirmatory research. Discussions with DOE on this problem are underway.
When and if the cross section difficulties are resolved, it would be desirable to prepare secondary benchmark VIM Monte Carlo calculations using a homogenized model of the experimental configuration. This will eliminate the need for plate-by-plate description of the geometry to verify a code.
Planning for these calculations will be carried out in FY 1979.
Any further work in this area is dependent upon Advanced Reactors Safety Research funding levels and the relative priority assigned to this problem.
~.ob::
Enclosures:
- 1. S. K. Bhattacharyya, et. al.,
"Program Description and Pre-analysis of the LMFBR Safety Related Critical Experiments on ZPR-9" ZPR-TM-279, June 7, 1977.
- 2. S. K. Bhattacharyya, et. al.,
"A Critical Experimental Study Office of Nuclear Regulatory Research of Integral Physics Parameters in Simulated LMFBR Meltdown Cores, 11 ZPR-TM-327, September 29, 1978.
!ATTACHMENT A N.
..::t
..::t n
fl)
='
"tJ
~
..::t
- N
..::t n
fl)
='
"tJ 11' iz:
k expt Reference
- 1. 00093 Na Void Zone 0.99696 Symm. Slump Out 0.98890 As~. Slump Out 0.9916 Symm. Slump In b
RSR Core Meltdown Criticals Program Configuration Eigenvalues and.C/E Values (in *parentheses)
Isotropic Benoist Gelbard S1+/Po S1+/Po Monte Carlo D's D's D's (DIF3D)
TWOTRAN 200 K hist 0.99270 0.98976 0.99093
- 1. 00232
- 1. 00328
- o. 9939 +/- o. 0015 (0.9918)
(0.9'888)
(0.9900)
(1. 0014)
(1. 0023)
(0.9930) 0.99094
. o. 98379 0.98577 1.00154 (0.9940) co. 9868)
(0.9888)
(1. 0046) 0.97620
- o. 97105 0.97388 1.00271
- o. 9964 +/- o. 0015 (O. 9872)
(0.9819)
(0.9848)
(1.0140) q.0076) 0.97875 (0. 9870) 0.99959 0.99462 0.99824 Ih/%6ka 975.61 971.86 969.57 971.17 968.18 Symm. Slump In 1.00115 0.98308 1.00823
- 1. 0041 +/- o. 0015. 965.18 (0. 9820)
(1. 0071)
(1.0029)
Asymm. Slump In 0.9862
- o. 97231
.969. 77 (O. 9859)
Asymm. Slump In 0.9920 0.98060 971.17 with Blanket (0.9885)
Collapse a From Gelbard D Diffusion Cale.
b
- This configuration was not physically constructed for operational safety reasons.
e e
Harold. Recomendations and Future PiL.
Data.l'&duct1on and_ preparation of the final.report will.be completed.1n FY 1~79. A VIM Monte Carlo C~lculat1on on Configuration 2 - Sooium Voided Test Zone will be made and th~ S cases not* shown on Figure 2 of ZPR"."TM-3~7 will be completed.
-n Analysis *to resolve cross section difficulties that cause differences between experiment and VIM Monte Carlo eigenvalues is needed. *rhe Ap~lied Physics Division.of AN,~ is the logical place for the work to be done. The work should be ~upported by DOE rather than NRC since it 1s not really confinnatory research; Discussions with DOE on this problem are underway.
When and 1f the cross section difficulties are resolved, *1t would be desirable to prepare secondary benchmark VIM Monte Carlo calculations usipg a homogenized model of 'the experimental configuration. This will eliminate the need for plate-by-plate description of the geometry to yerify a code. Planning for ~hese calculations will be carried out in
/FY 1979. Any further work in this area is dependent upon Advanced.
Reactors Safety Research fun~ing levels and the relative priority l
- assigned to thfs problem.
./*
/
- Saul Levine, Director
Enclosures:
- 1. S. K. Bhattacharyya, et. ~1.,
... Program Description and Pre-analysis of the LMFBR Safety.
Related Critical Experiments
. on ZPR-9" ZPR-TM-279, June 7, 1977.
- 2.
S~ K. Bhattacharyya, et. al.,
"A Critical Experimental Study Office of Nuclear Regulatory Research of Integral Physics Parameters in Simulated LMFBR Meltdown Cores,"
- *zPR-TM-327, September 29, 1978.
i DIST:
J Subj *. Chron'. _ Debbie: ROG Circ. Wood: ROG
. SEE PREVIOUS YE w
N J
~
... ~-~.:!l!lfiligL.. ll!ig.;~~/f.)WL.. ~-******** BES..,
....... BES..,lq.£3.... 8'S.!f,,......... J
...... _., WQQR/.~P.N.. ~!JR. J.S. **** ~~~
... ~r4JRLD:.......... ~. *SCROG NS........ BUDN~i~........... LE"Y.f[E......... 1
- ,..., i11.a1.2a...........
..~.... ll/.1.4/.7.a.... 12/.b.tia........ :i2;..,t.J.a........... :i.u... /.7.s......... J.21.J~.7.s......
~
~-**1* ~*
AA.ol.A u
ftftU**..n.m*:.... ~INTHIQ Ofl'PICC: t*H *&*a.~., **