Doubt about group constant file suffix

[js-jixu]

Hi, everyone.

I want to do a IAEA2D benchmark by Moltres, which needs to modify group constants. Then I want to correspond suffix of files of group constants to their physical meaning. Now, I've determined some suffixes but still had some undetermined ones.

I want to know the meanings of those group constant file suffix which aren't found in MoltresPhysicalKernels: FISSE, FLUX, NUBAR, CHI_T.

Any answers are deeply apperciated!

[smpark7]

FISSE = Energy produced per fission (~200MeV), usually represented with greek letters epsilon (ε) or kappa (κ)

FLUX = Group flux integrated over the whole domain. Moltres does not use this in any simulations. We'll probably remove this in future.

NUBAR = Average neutrons produced per fission, represented with the greek letter nu (ν) with a bar (straight line) above it to signify that it is an average value. I think Moltres does not use this either because it uses NSF (nu sigma_f) instead.

CHI_T = Total fission probability spectrum (prompt + delayed) represented with the greek letter (χ)

[smpark7]

The moltres/problems/2021-cnrs-benchmark/phase-0/nts.i input file should give you a good idea of how to set up your input file. It involves a six-group neutronics eigenvalue problem in a 200cm-by-200cm static molten salt domain. The InversePowerMethod executioner is suitable for calculating the multiplication factor eigenvalue. Note that this input file uses the [Nt] action to automatically initialize the required variables, kernels, and BCs for the six-group neutron diffusion equation.

Since the IAEA2D benchmark does not model precursors, you'll want to disable precursors by removing the [Precursors] section and setting account_delayed=false in [GlobalParams]. Moltres will use CHI_T for the fission spectrum χ1 and χ2. However, Moltres will still require you to provide CHI_P values by default. You can set CHI_P to any arbitrary value to avoid getting an error.

[js-jixu]

Thank you, Sun Myung.

I wrote a new input file today referring to the input file you mentioned, but I encountered some problems with the results. I put it under a new discussion so that our discussion can be referenced by later beginners like me.

[js-jixu]

Hi, Sun Myung.

I wonder if Moltres will use CHIP when sss2_input=1 ， and CHIT can be given arbitrarily (because Moltres does not use it for computation). Correspondingly, when sss2_input=0 ， Moltres will use CHI_T , and CHI_P can be given arbitrarily. Can I understand that if I use OpenMC to calculate the neutron cross-section, then I need to provide CHIP for the calculation, and CHIT to prevent Moltres from reporting an error (in fact, CHIT does not participate in the calculation); if I use Serpent2 to calculate the neutron cross-section, I need to provide CHI_T used for calculation, CHI_P used to prevent Moltres from reporting errors (in fact, CHI_P does not participate in calculation)? Is this understanding correct?

I would also like to ask you about the role of the scaling parameter. In this case, does the unit of neutron flux change to 10^3/(cm^2·s) when you use scaling=1e3 ?

[smpark7]

Hi Xu. That's not correct. Whether Moltres uses CHI_T or CHI_P depends on the account_delayed parameter. If account_delayed=true, then Moltres uses both CHI_P and CHI_D for prompt and delayed neutrons, respectively. Otherwise, Moltres uses CHI_T which is the flux-averaged total. CHI_T is usually sufficient for steady-state calculations.

sss2_input determines how Moltres reads the multigroup scattering cross sections. This originated from differences in SCALE/NEWT and Serpent 2 output files. SCALE/NEWT provides scattering cross sections ordered by the in-scattering group, i.e. 1->1, 2->1, 3->1, ..., n->1, 1->2, ..., n->n while Serpent 2 provides scattering cross sections ordered by the out-scattering group, i.e. 1->1, 1->2, 1->3, ..., 1->n, 2->1, ..., n->n. As you may have guessed, the parameter is named after sss2, the name of the Serpent 2 executable. OpenMC scattering cross sections follow the same format as Serpent 2 as long as you follow the steps outlined in the tutorial/godiva tutorial. So you'll want to leave sss2_input=true if you use OpenMC group constants.

[js-jixu]

Thanks a lot, Sun Myung. Your reply solved my doubts about the Moltres group constant suffix