I am producing results of fission rates as a function of burnup for a CANDU reactor, and I noticed that my fission rates were not beginning close to 0, but instead jumping up to relatively high values as soon as any of that nuclide was first produced. This can be seen in the following plots,
I a
This was completed for 5 hour intervals using predictor integrator, and for a total of 6.25 days. For pu-241 it appears to begin at 10e10. I am wondering if this is due to some kind of truncation by OpenMC, or as I am thinking, it’s due to the analog nature of OpenMC, and that these specific situations have such a low probability (for fission of Pu-241 for example, the factors that would be considered is nuclide account and flux over that fuel volume) that the approximation is not applicable except once a high enough counting statistics is reached. Is this the purpose of variance reduction techniques such as MAGIC and FW-CADIS, and is this something I should be concerned with doing when modelling a fuel CANDU reactor’s depletion cycle?