I’m currently working on a project comparing the results of an identical model in OpenMC vs MCNP.
The only difference between my current OpenMC simulation and the legacy MCNP simulation is the nuclear data being used - the MCNP simulation was performed with ENDF/B-VI, and my OpenMC simulation used ENDF/B-VIII. In this arrangement, the OpenMC result is ~2-3% less than the MCNP value.
I’m currently working on (and struggling with) creating a HDF5 format of the ENDF/B-VI so that I can completely replicate the simulation. what kind of change do you think I could expect to see going from ENDF/B-VIII to B-VI?
and could anyone point me in the right direction for achieving the conversion from ENDF/ACE to HDF5?
any thoughts and guidance would be much appreciated.
Openmc-dev also has some scripts to convert an ENDF-formatted file or an ACE file to h5 file. I haven’t tried all of this script, but I think you can try to use some part of this script.
Before I dive into the rabbit hole myself, did either of you, @LewisM23 or @Philp18 , succeed in creating a working ENDF/B-VI library? If so, would you be willing to share it?
I’m also attempting to replicate results from a paper that used older MCNP simulations.
@villads Here is a ENDF/B-VI.6 library I just put together by converting ACE files from MCNP’s ENDF66 library:
The only caveat is that I couldn’t convert Be9 because it uses an ACE law that is not supported in OpenMC. Everything else seemed to go through though.