An Adjusted Cross Section Library for DFBR

Peter, J. Collins

PNC TN9410 97-034, 35 Pages, 1997/04

PNC-TN9410-97-034.pdf:1.07MB

While in the Reactor Physics Research Section of the Advanced Technology Division at OEC, I participated in the project to construct a data library for the demonstration fast breeder reactor (DFBR). This library would be produced using a combination of evaluated differential cross sections together with integral experimental data for fast reactors, so as to assure sufficiently accurate calculations for the DFBR designs. I had much experience of the design and use of experiments for the large-size cores at ZPPR under the title JUPITER which was performed under the USDOE/PNC joint agreement. My contribution here was mainly in extension of the experimental database to include the very-hard spectrum fast criticals from the Los Alamos National Laboratory (LANL). The data for these cores are described. Our work at ANLW with the GMADJ code, which is similar in effect to the ABLE code that we use at PNC, showed why many experiments are important in this project as well as those in the more obvious Pu/U oxide conventional cores which are of current interest for the DFBR. This point was not appreciated at PNC and is discussed here. The data from the fast spectrum critical experiments made at Los Alamos are described together with information that I have been able to find concerning the uncertainties. The main interest is these experiments has been for prediction of criticality. Consequently, the full covariance information that we would like has not been published. However, the uncertainty in the fuel content is, by far, the major contributor to the uncertainty. The LANL experiments have been a principal leg of the data testing for fast reactors for all versions of ENDF/B in the US. For our work, they provide measurements at Mev energies which are not available from the experiments in the softer-spectrum of the LMFBR.

Investigation of the impact of the difference of probability table calculation method

Tada, Kenichi

no journal, , 

JAEA started to develop the new nuclear data processing system FRENDY. FRENDY can generate the ACE (A Compact ENDF) file without NJOY module and it is implemented not only NJOY's nuclear data processing method but also FRENDY original method. In this presentation, the impact of the difference of nuclear data processing method on the k-effective value is investigated.