[Mesa-users] issues with overshooting: invalid location for overshoot boundary

[Temmink, K.D. (Karel)]

Dear all,

For completeness, I wanted to add that I also encountered this potential bug in version r21.12.1.
I have also commented this in the github link.

-Karel
________________________________
From: Mesa-users <mesa-users-bounces at lists.mesastar.org> on behalf of Temmink, K.D. (Karel) via Mesa-users <mesa-users at lists.mesastar.org>
Sent: 19 May 2022 14:23
To: Evan Bauer <evan.bauer.astro at gmail.com>
Cc: mesa-users <mesa-users at lists.mesastar.org>
Subject: Re: [Mesa-users] issues with overshooting: invalid location for overshoot boundary

Dear Evan,

thank you so much for your reply!
I have commented my issue on github.
It's reassuring I'm not the only one running into this issue and that it's being looked at!

Thanks again and with kind regards,

Karel


________________________________
From: Evan Bauer <evan.bauer.astro at gmail.com>
Sent: 18 May 2022 01:01
To: Temmink, K.D. (Karel) <karel.temmink at ru.nl>
Cc: mesa-users <mesa-users at lists.mesastar.org>
Subject: Re: [Mesa-users] issues with overshooting: invalid location for overshoot boundary

Hi Karel,

Thanks for this report. There is an open issue on github describing something similar:
https://github.com/MESAHub/mesa/issues/400<https://urldefense.com/v3/__https://github.com/MESAHub/mesa/issues/400__;!!HJOPV4FYYWzcc1jazlU!_J_o21P4-mOu2bYOUdwgxbSUtPFqqlhBEsXkuvOpb0AWDQ8OqsBQax_aJYYVQaHUAavDXHkOO6wiRwpfV-8D9rG28ejZ$>

So you may be running into a legitimate bug that we need to fix. I'd recommend posting this report as a comment on that github issue so that we can start thinking about escalating the severity and pushing to solve the problem if it really is a bug.

Cheers,
Evan


On May 16, 2022, at 2:15 PM, Temmink, K.D. (Karel) via Mesa-users <mesa-users at lists.mesastar.org<mailto:mesa-users at lists.mesastar.org>> wrote:

Dear mesa-users,

I've recently upgraded to the latest version of mesa, mesa-r22.05.1, and ran into some issues.
In acquainting myself with this new version, I tried to simulate the evolution of low- to intermediate mass single stars (1-8Msun at ZAMS).

Unfortunately, I ran into numerical problems for every simulation I've ran so far with this version.

For example, near the end of the He-flash, when the burning flame has just reached the centre, mesa exits with the error:

 k, k_a, k_b        1643        1643        1643
 s%top_conv_bdy(i)= T
 D(k)   1.6625067809834432E-002
 s%D_mix(k+1)   0.0000000000000000
 s%overshoot_D_min   100.00000000000000
 Invalid location for overshoot boundary: cz_bdy_dq, dq=  -2.1741526243966756        3.6151445011785386E-004
 set_mixing_info failed in call to add_overshooting
 update_vars: set_hydro_vars returned ierr          -1
 set_some_vars: update_vars returned ierr          -1
 prepare_for_new_step: set_vars_if_needed ierr
 do_step_part1: prepare_for_new_step
terminated evolution: nonzero_ierr
​
I have tried changing a few inlist settings related to the meshing (such as mesh_dlog_pp_dlogP_extra), which, perhaps unsurprisingly, did not seem to help.
Because the error message mentioned overshooting, I also removed all settings related to overshooting from my inlists. In that case, the error did not appear and the simulation got through the entire He-flash without issues.

However, I do want to include the effects of overshooting in the calculations, so leaving them out of the inlists is not an option for me.
I was wondering if anyone else has encountered this error too, or could advise me on how to get past this. I would be extremely grateful for any help.
I have attached my inlists and run_star_extras.f90 that should reproduce this problem. I have also included a photo to restart from, since the issue only appears at timestep number 14748.

Thank you in advance for your time and effort.
With kind regards,

Karel Temmink

Astrophysics PhD candidate (stellar and binary evolution) at Radboud University, Nijmegen, The Netherlands


<inlist_project><my_history_columns.list><my_profile_columns.list><run_star_extras.f90><x500>_______________________________________________
mesa-users at lists.mesastar.org<mailto:mesa-users at lists.mesastar.org>
https://lists.mesastar.org/mailman/listinfo/mesa-users<https://urldefense.com/v3/__https://lists.mesastar.org/mailman/listinfo/mesa-users__;!!HJOPV4FYYWzcc1jazlU!_J_o21P4-mOu2bYOUdwgxbSUtPFqqlhBEsXkuvOpb0AWDQ8OqsBQax_aJYYVQaHUAavDXHkOO6wiRwpfV-8D9gD9x3pG$>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.mesastar.org/pipermail/mesa-users/attachments/20220520/c0f3f894/attachment.htm>