Reflective and periodic boundary conditions with DAGMC geometry

User Support
1

Hello, I just wanted to see if anybody else had this problem.

I’m currently modeling a 90-degree sector model (1st quadrant) for a tokamak, that is bound by x=0 to 1000, y=0 to 111, and z=±1000 and I have the -x and -y planes (so x=0 and y=0) planes as reflective boundary conditions. I wanted to make them periodic so I changed them to period, but I still go the similar results.

The geometry file looks like this:

<geometry>
  <cell fill="1" id="9999" region="9995 -9993 9994 -9992 9999 -9998" universe="2" />
  <dagmc_universe filename="dagmc.h5m" id="1" />
  <surface boundary="vacuum" coeffs="1000" id="9992" type="y-plane" />
  <surface boundary="vacuum" coeffs="1000" id="9993" type="x-plane" />
  <surface boundary="reflective" coeffs="0" id="9994" type="y-plane" />
  <surface boundary="reflective" coeffs="0" id="9995" type="x-plane" />
  <surface boundary="vacuum" coeffs="1000" id="9998" type="z-plane" />
  <surface boundary="vacuum" coeffs="1000" id="9999" type="z-plane" />
</geometry>

I tested a similar concept for a DAGMC geometry of Jezebel (so just a sphere with r=6.3849) and set the +y and -y surfaces as periodic, ran the case, and then I ran another case with +y and -y as reflective. For those cases:
vacuum all around: 0.99917 +/- 0.00103
periodic: 1.12174 +/- 0.00011
reflective: 1.12186 +/- 0.00014

This only gets more confusing for me when I try this on CSG, where a sphere is surrounded by void box:

<geometry>
        <cell id="1" material="1" region="-1" universe="1"/>
  <cell id="2" material="void" region="1 9995 -9993 9994 -9992 9999 -9998" universe="1"/>
  <surface boundary="vacuum" coeffs="0.0 0.0 0.0 6.3849" id="1" type="sphere" />
  <surface boundary="periodic" coeffs="6.3849" id="9992" type="y-plane" />
  <surface boundary="vacuum" coeffs="6.3849" id="9993" type="x-plane" />
  <surface boundary="periodic" coeffs="-6.3849" id="9994" type="y-plane" />
  <surface boundary="vacuum" coeffs="-6.3849" id="9995" type="x-plane" />
  <surface boundary="vacuum" coeffs="6.3849" id="9998" type="z-plane" />
  <surface boundary="vacuum" coeffs="-6.3849" id="9999" type="z-plane" />
</geometry>

then, regardless of the boundary conditions of the x,y,z planes, I get the same value for keff, which is 1.

It feels like I am not understanding period boundary conditions correctly, can someone help?

Thanks in advance!