[Mesa-users] Differences between MESA versions

[Josiah Schwab]

Hi James,

I've attached the inlists
>

Thanks.  Looks like you're pretty much just taking the defaults, in which
case I can say that I think the r15140 defaults are generally better than
past versions.

For your r12115 weirdness, this looks like the EOS.  The 'eosDT2' option
(which has since been removed:
https://lists.mesastar.org/pipermail/mesa-users/2020-March/011168.html)
didn't perform well down in substellar conditions.

If I turn those removed EOSes off (and return to the same eosDT that would
have been the default in r8845), by adding the following options in controls

      use_eosDT2 = .false.
      use_eosELM = .false.

then I get a much more sensible track (see attached plot).  So that solves
part of your mystery (and I would recommend against using r12115).


Working with r15140, you should probably also think a bit about which
energy equation you want to solve.  The default is a total energy equation,
but that doesn't always perform well in degenerate conditions (requires
smaller timesteps to get good results).

I'd be curious to see a comparison of a model in r15140 with the defaults
and one with the additional controls

always_use_dedt_form_of_energy_eqn = .false.
always_use_eps_grav_form_of_energy_eqn = .true.
use_time_centered_eps_grav = .true.

That last flag is part of an ongoing experiment, so please don't use it
without writing back and showing me the results of that comparison.

Josiah
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.mesastar.org/pipermail/mesa-users/attachments/20210301/51ed1189/attachment.htm>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: plot.png
Type: image/png
Size: 22279 bytes
Desc: not available
URL: <https://lists.mesastar.org/pipermail/mesa-users/attachments/20210301/51ed1189/attachment.png>