Segmentation fault after a few time steps with ANEOS

[mikinakajima]

Scott and I have been trying to implement ANEOS, but the code experiences a segmentation fault after a few time steps. The EoS interpolation seems to be working fine, so we are confused. The fault is happening in PTCL::CalcPressure(sph_system) (main.cpp).

We also tested with a refined EoS data table (This is tested under eos_test branch) but we are still experiencing this issue.

We will try to have a better look at this..

[NatsukiHosono]

Hmm, did you try a debugger (e.g., gdb) tp specify the line which crashes the run?

[mikinakajima]

Thank you Natsuki! I think I identified the problem.. I find that the pressure difference between the iron core and mantle at the core-mantle boundary was too large with ANEOS (mantle had a much larger density than it should be ... like 8 g/cc with the default planetary radius). Actually the mantle pressure at CMB was much larger than the core pressure there. Tillotson and ANEOS seem to have very different rho-P relationships (which we should probably check). We increase the planetary radius and now the code seems working fine.