Questions about output units and Postprocessors Block in Moltres neutronics-only simulation of MSRE

[gamma-spectrum]

Dear all,

I have Moltres (version 2025.09.02) within a MOOSE installation (version 2025.09.04) and I'm currently following the tutorial Introduction to the Moltres Input File Syntax in order to reproduce the Molten Salt Reactor Experiment (MSRE) results. My aim is to understand the units of output variables and the information given by variables in Postprocessors block.

But first, I'm curious about the meaning of nts in the input file nts.i. It means neutronics or something like that?

I have launched my compiled version of Moltres following the instructions of the web page

cd projects/moltres/tutorial/eigenvalue
../../moltres-opt -i nts.i

and I have obtained the Exodus output file nts_exodus.e. Also, I have opened that file with Paraview tool and I'm able to access to the computed variables (group1, group2 for neutron fluxes of fast and thermal groups and pre1, pre2, pre3, pre4, pre5, pre6 for precursor concentration) and the information generated by Postprocessors block (within a section called Data Arrays).

As it is said in post Discrepancies in MSRE simulation results #329, this is a neutronics-only eigenvalue simulation and the eigenvalue simulation is scaled down by default to be equivalent to having 1 source neutron.

Keeping this in mind, and considering that in nts.i it is used as units cm for distance and s for time, I deduce that for neutron fluxes (group1, group2) units are 1/cm2 per source neutron. This is correct?

For precursor concentration, if I assume that the 2D model has a thickness (third dimension) of 1 cm, I deduce that for precursor concentrations (pre1, pre2, pre3, pre4, pre5, pre6) units are s/cm3 per source neutron. This is correct?

In the case of Postprocessors block, variable powernorm represents the fission heat but because no energy unit is set in nts.i I assume that energy units should be MeV because it is the energy unit used in the group constant file xsdata.json. Following this reasoning, powernorm units should be MeV (per source neutron). This is correct or it is actually used other units?

Last but not least, I have some questions about variables set in Postprocessors block regarding group1, group2 variables. Block group1max gives the maximum value for group1 and this can be easily checked looking for the maximum and minimum values of group1 in Paraview.

For blocks group1norm and group1diff I have checked the information in the Moose and Moltres syntax webpages and in both cases the information says that their objective is to compute the volume integral of a variable. If it is a volume integral, units are 1/cm2 * cm3 per source particle? Because in my results the quantities of group1norm and group1diff are different (169.87 and 0.145, respectively) I guess that each variable has a different meaning. Which are the difference between the two?

Kind regards,
Juan García

[smpark7]

The units for neutron flux and precursor concentration are cm-2 s-1 and cm-3 s-1. Note that the length unit follows whichever unit your cross section file values are provided in. For power, Moltres converts MeV to J.

group1norm and group1diff are used for numerical convergence purposes. group1norm computes the L2 norm of the group1 variable field. group1diff computes the L2 norm of the difference between the current and previous timestep solution. Eigenvalue simulations in moose take 1 "timestep" from the initial guess to the final solution. So you get the difference between the initial and final solutions