[Mesa-users] rel_run_E_err Warning

[Josiah Schwab]

Hi Jen (& Erica),


> Thanks for this email - this really helps me to understand a bit better
> what this is actually telling me, although it's taken some time to process
> it. The thing that's still a bit surprising to me is that this is a very
> solar star - solar mass, solar composition, solar age. Gold tolerances are
> on. I wouldn't have thought we're in particularly ugly regions for the EOS,
> or any of the other input physics... but perhaps I haven't appreciated the
> lack of precision, even at solar conditions.
>

I also find this surprising.  I recently ran a set of 1M sun models from
pre MS to 1st thermal pulse and didn't encounter unexpected rel_run_E_err.

I ran the inlist Erica sent in the first message and didn't immediately see
these warnings.  Could you send a work directory where we can compile/run
and reproduce what you're seeing?  (As I write this, I see that Bill
requests the same.)

One note is that if you run a 1M MESA model with the eps_grav form out of
the box, you get poor energy conservation during the MS.  The deeper reason
is not numerical, rather it is a result of the default implementation of
eps_grav dropping the term associated with composition changes.

This assumption is well justified in terms of the total energy budget of
the star.  The energy associated with the change in mean molecular weight
is dwarfed by the energy release from nuclear reactions that cause it.
However, our reference point for checking energy conservation is not the
time-integrated radiated energy of the star, but is the instantaneous total
energy of the star (excluding rest mass).  Relative to that, the energy
change is not negligible.

If you turn on the eps_grav composition terms (setting
"include_composition_in_eps_grav = .true.), you see the improvement in
reported energy conservation.

This caveat is not relevant when using the "dedt form".

Josiah
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.mesastar.org/pipermail/mesa-users/attachments/20200507/12f9b803/attachment.htm>