The CORSIKA fiducial sphere should be large enough to contain the entire mirror in all pointing directions. This has been checked by Konrad for the various telescope types, taking into account the mirror and axes offsets while doing so. Attached are a few plots demonstrating his calculations for MST, SCT and two cases for SST with different axes_offset values.
It would be good to develop a tool which calculates the size of the sphere instead of doing it manually each time. The attached examples can be used as guidance.
Since we don't expect frequent changes to these parameters, this issue could be low priority.
mst_sphere.pdf
sst_sphere2.pdf
sst_sphere.pdf
sct_sphere (1).pdf