I am using the 21-May-08 build of Lammps with patches applied up to
4-Aug-08. I am having a problem with the fix rigid. Below is a
simple test example of three granular atoms lined up vertically with a
wall below the bottom atom. I want to make the top two atoms rigid
and use that rigid body to exert a force on the bottom atom. This
input file does work if you don't include the line "fix rigd top
rigid single" so it is not something else causing the crash. Once the
rigid fix is included, the input file is read with no errors, but it
won't run. Instead, I get the error: "p4_error: interrupt SIGSEGV:
11."
I am using the 21-May-08 build of Lammps with patches applied up to
4-Aug-08. I am having a problem with the fix rigid. Below is a
simple test example of three granular atoms lined up vertically with a
wall below the bottom atom. I want to make the top two atoms rigid
and use that rigid body to exert a force on the bottom atom. This
input file does work if you don't include the line "fix rigd top
rigid single" so it is not something else causing the crash. Once the
fix rigid requires atoms to have explicit per-type masses to compute
the moments of inertia. granular particles don't have those set and
one cannot set them from the input. you'd have to modify the code
somehow to make it work.
Axel's correct. Iadded an error message to check for this (8Sep08 patch).
You could probably use a hybrid atom style and define both per-atom
masses (as granular does) and per-type masses, assuming the system
is mono-disperse, and get it to work.