Help debugging a geometry-related error in my MCNP input deck

[MadGander]

TL;DR Summary

MCNP geometry error

I'm looking for someone to help troubleshoot my MCNP input deck. I'm getting a geometry related error most likely due to some sort of surface overlap. Haven't been able to identify the issue myself, so I'd appreciate a secondary check. Thanks!

 

[Alex A]

The input file is incomplete and is crashing for me. It seems to be a big lattice full of spheres. It's very verbose, with a lot of copy transpose.

I would suggest writing the lattice as a lattice. If you describe it I should be able to help.

 

[MadGander]

The input file is incomplete and is crashing for me. It seems to be a big lattice full of spheres. It's very verbose, with a lot of copy transpose.

I would suggest writing the lattice as a lattice. If you describe it I should be able to help.

Im trying to model a TRISO fuel compact using BCC lattice cells. I'm focusing specifically on the x-y plane to produce a cylindrical geometry. Depending on exact location, the periphery cells have some sort of pseudo-BCC structure to form the outer edge of the compact. Because of this, the periphery cells have certain void corners where no 1/8th particle slice exists. Appreciate the assistance.

 

[Alex A]

If this is, for example, a 17x17 lattice then we can model this. The spheres are tiny though, under 1mm and I can't imagine this amount and geometry reaching criticality. If this is a set problem and you have to solve it this way we can do it. If what you want is the answer from MCNP on a running reactor with this material as fuel I would suggest you homogenise. If you have a structure that is 10cm across, neutrons will not care much about the fine structure, only the quantities. Working out quantities of elements in bigger units and building the reactor from the big units is easier to model and needs less computer time.

So what are all the relevant parameters of the problem, what answers do you need and how do you want to solve it?

 

[MadGander]

Thread can be closed. I ended up ditching this route and am pursuing a different method.

 

[Alex A]

If it's something you can share, please do consider sharing it. All the best with the new method!