Difference k_eff transport and depletion version 0.14.0

Hello everyone,

I’ve encountered an anomaly while using OpenMC version 0.14.0 installed using the Mamba package in a Conda environment. Specifically, during a depletion analysis of a sodium core, I observed a discrepancy of 100 pcm in k_eff between running in depletion mode and transport mode. I haven’t made any changes to the model other than introducing the depletion module from OpenMC. I verified the model by transforming my model to serpent version 2.2.1 and the k_eff of the transport part matches almost perfectly with the serpent model for depletion. These are the multiplication factors obtained:

  • Transport OpenMC k_eff: 1.00823 +/- 0.00015
  • Depletion OpenMC k_eff: 1.00726 +/- 0.00021
  • Serpent depletion k_eff: 1.00818 +/- 0.00006

Upon checking the discussion forum, it seems that a similar issue was reported in version 0.13.0-dev and version 0.13.2, as discussed here and here.

If anyone has insights into the potential cause of this discrepancy, I would greatly appreciate your assistance.

Kind regards,
Daan

import openmc
import openmc.deplete

geometry = openmc.Geometry.from_xml()
settings = openmc.Settings.from_xml()
tallies = openmc.Tallies.from_xml()
model = openmc.model.Model(geometry=geometry, settings=settings, tallies=tallies)

operator = openmc.deplete.CoupledOperator(
model, '../chain_endfb71_sfr.xml')

dt = [1, 1, 4, 4, 10, 10, 10, 10, 10, 10, 10]
power = 65e6

cecm = openmc.deplete.CECMIntegrator(
operator, dt, power, timestep_units='d')
cecm.integrate()

Just noting that I also encountered this issue here.