[mesa-users] make_he_wd in 5596 hydro_failed_to_converge

Monique Windju mwindju at ucsc.edu
Thu Dec 19 15:42:25 EST 2013


Worth noting: I'm finding the 5596 build of make_he_wd to be very delicate.
I ran a test to try and figure out which controls were making it crash, and
after reverting to the original inlists that came with 5596 I was unable to
create a 0.15 solar mass helium white dwarf. I got the
"hydro_failed_to_converge" error during inlist 2, and given that the only
option I modified in these original inlists was setting "new_Y=0.97" in
inlist 3 I am worried about the overall stability of this new build of
make_he_wd.

I remember in an older mailing list email someone said
"hydro_failed_to_converge" might have something to do with a model evolving
for too long during certain inlists. I am currently running a test using
different termination ages and model numbers, but I do not expect this
route to give me a permanent fix to these issues.


On Thu, Dec 19, 2013 at 11:42 AM, Monique Windju <mwindju at ucsc.edu> wrote:

> Hi all,
>
> I've been using MESA for about 6 months, and I recently upgraded to
> version 5596 on OS X Mavericks. Before I upgraded, I was using 5271 on
> Ubuntu to build helium white dwarfs with pure hydrogen surface layers. When
> I try to build them in 5596, however, the models fail in inlists 2 and 3
> with the error:
>
> > stopping because of convergence problems -- too many retries
>
> > terminated evolution: hydro_failed_to_converge
>
> > termination code: max_number_retries
>
>
> Could someone explain why these errors occur or how I can fix this? The
> masses I have tried to reach are 0.10, 0.15, 0.20, and 0.25 solar masses. I
> think the main issue is the switch from using "h1_boundary_mass_limit" and
> "h1_boundary_limit" controls to "he_core_mass_limit" in inlist 1, since
> that's the only obvious change I can see between the two versions.
>
> I've attached my inlists from versions 5271 and 5596, as well as the
> abundance profiles from 5271 that I am trying to achieve. I have also
> attached the pgstar outputs for a 5596 model that failed during inlist 3,
> ordered by inlist (eg. Grid1 means the Grid Window output for inlist 1,
> Grid2 is inlist 2...)
>
>
> Thanks!
>
> Monique Windju
> Supercomputing Lab for Undergraduates
> Theoretical Astrophysics Santa Cruz
> University of California, Santa Cruz
> mwindju at ucsc.edu
>



-- 
Monique Windju
University of California, Santa Cruz
mwindju at ucsc.edu
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.mesastar.org/pipermail/mesa-users/attachments/20131219/6758826d/attachment.html>


More information about the Mesa-users mailing list