Pb with fixed source (ERROR: The secondary particle...)

User Support
1

Hello,

I am running a fixed source calculation,
I am using my fuel as fixed neutron source.
I checked my keff, keff ~ 1.05895
I keep getting:

Simulating batch 1
ERROR: The secondary particle bank appears to be growing without bound. You are
likely running a subcritical multiplication problem with k-effective
close to or greater than one.

here is my source:

<settings>
    <run_mode>fixed source</run_mode>
    <particles>10000</particles>
    <batches>100</batches>
    <source type="independent" strength="1.0" particle="neutron">
      <space type="cartesian">
        <x type="uniform" parameters="-35.2 35.2"/>
        <y type="uniform" parameters="-35.2 35.2"/>
        <z type="uniform" parameters="-35.2 35.2"/>
      </space>
      <angle type="isotropic"/>
      <energy type="discrete">
        <parameters>[...]</parameters>
      </energy>
      <constraints>
        <domain_type>cell</domain_type>
        <domain_ids>1</domain_ids>
      </constraints>
    </source>
    <output>
      <path>results/depletion/Xe_Kr/d_34.40_co_0.0100/shielding/neutron/1.0000</path>
    </output>
    <temperature_method>interpolation</temperature_method>
  </settings>

any idea what I might doing wrong ?

(I posted the same question on slack, will be updating this if I get an answer on slack)

2

Hi @bam241. The problem you’re running into here is that your system is supercritical (keff > 1.0), which means that every source neutron is generating more than one follow-up neutron due to fission. This results in your neutron population growing with no bounds and overflowing the shared secondary bank (where created particles get stored in fixed source mode), which is what you’d expect as there is no steady-state solution to a fixed-source problem with multiplication greater than 1.0 (all solutions are time-dependant exponential growth).

The correct way to fix this depends on what you’re trying to simulate. If you want to run a shielding calculation outside of the fissile system, you’d need to tally a surface source on the outer part of the reactor in k-eigenvalue mode. Then, you can use this surface source for a fixed-source problem in the non-reactor part of the domain. You could also ensure the problem is subcritical (e.g. add control devices to your geometry to force keff to be less than 1.0) if modelling a fixed-source shutdown state. Finally, you can just run in k-eigenvalue mode.

3

thank you for taking the time !

Yes I was slowly getting to that conclusion.

This make sense… if my system is above critical, my neutron history on average never stops. So fixed source get into an infinite loop :slight_smile:

thank you again !