95% External Source Sites Rejected - TRISO Model

User Support
1

I’ve run into an issue while modeling a generic HTGR fuel assembly from the INL virtual test bed.

image
image1244×689 139 KB

My source is described as a box (openmc.stats.Box()) bounding the hexagonal assembly, only allowing source points in fissionable cells (the TRISO particles).

I’m getting the following error:

ERROR: More than 95% of external source sites sampled were rejected. Please check your external source's spatial definition.

I believe this is because, by volume, the fuel in the TRISO particles is occupying less than 5% of the spatial sampling region defined by openmc.stats.Box().

What is a possible way to define the source that could get around this? I’ve explored options to define the source as a union of cylinders enclosing the TRISO fuel compacts, but because of the nature of the fuel arrangement, this is quite tedious, and I haven’t found a sleek way to do it.

Any suggestions?

2

Because you have to converge the source anyway for a k-eigenvalue calculation, it’s not that important to constrain source sites to fissionable materials. Thus, an easy fix is to simply omit the use of only_fissionable=True:

spatial_dist = openmc.stats.Box(lower_left, upper_right)

By the second batch, all your source sites will already be in fissionable materials, so relative to using the only_fissionable constraint, this means at worst you need 1 extra batch to converge your source.

4 Likes
3

Appreciate the note. This worked for me. Thank you!