Discrepancies in MSRE simulation results

[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.

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.

On the one hand, I have opened that file with ParaView version 6.0.1 and chase the second time (Time: 1) for data representation. I have obtained the maximum value of neutron fluxes for fast (group1) and thermal (group2) groups as well as the precursor concentration of the six groups (prep1 to prep6).

On the other hand, I have obtained the same information from the figures of Lindsay et al., (2018) article. This article reproduces the MSRE results with Moltres. Although the model that I have simulated and the article one are not exactly the same (temperatures and sizes are a little bit different) I assumed that similar results should be obtained. Also, although taking from the figures the maximum values is not much accurate, for the purpose of this comparison it is enough.

Next two figures correspond with neutron flux fast (group1) group. First one belongs to my results and units are in 1/cm2/s (units deduced from the input definition of mesh and salt velocity units).

Second one is Fig. 3 of Lindsay et al., (2018) article and units are 1/cm2/s.

As it can be seen, the shape of both colour maps are similar but the maximum values of the first one is 2.503E-04 1/cm2/s while the second one is 8.000E+13 1/cm2/s. For completeness, I put a table with rest of my computed values and those of the article (including the units)

	group1 (1/cm2/s)	group2 (1/cm2/s)	pre1 (1/cm3)	pre6 (1/cm3)
My results	2.503E-04	1.199E-04	7.125E-09	9.075E-10
Article	8.000E+13	3.000E+13	1.800E+09	1.600E+08

I have computed the ratio between these values and the obtained value is around 4e-18. It could be missing some scale factor to obtain the actual values?

Kind regards,
Juan García

[smpark7]

The models are very different. The model in the paper is the steady-state solution of a coupled neutronics-TH simulation. The tutorial is a neutronics-only eigenvalue simulation. The flux in the eigenvalue simulation is scaled down by default to be equivalent to having 1 source neutron.

The model for the paper can be found in moltres/problems/2017_annals_pub_msre_compare.