Re: [mesa-users] AGB convergence problem

[Héctor MR <hec...@pi...>]

  Hi Laura,

  In general, convergence problems in MESA are difficult to diagnose.

  I have been creating WDs for a long time and have frequently found the
problem you mention: my stars were getting stuck in the middle of the AGB
(mostly for high initial masses and metallicities), going back and forth
without being able to get to higher effective temperatures.

  After some useful interactions in this mailing list, I found that
increasing the AGB mass loss, as you say, could help circumvent this (via
Blocker_scaling_factor from r8118 on ahead). I do not fully understand why
this should be a problem for your purposes; you can always implement your
own stopping condition via extras_finish_step or extras_check_model under
run_star_extras.f (http://mesa.sourceforge.net/run_star_extras.html#toc-2).
On the other hand, I would not recommend you to modify the default value of
min_timestep_limit or your results might get noisy (in case you were
thinking about it).

  When you say "the model seems to get stuck somewhere possibly near the
AGB tip", have you actually checked it with an HR diagram using PGPLOT? I
have found that this is one of the most powerful debugging tools in MESA
and it could help you see what is occurring to your star in a better way.

  Hope this helps.


  --
  Héctor