[Mesa-users] Seeking help with an convergence problem in double_bh test suite

[Ruggero Valli]

Hi Levi,

I am sorry I hadn't noticed the run.out file. That's exactly what I was 
asking. So from the output it seems that the problem is not the rotation 
of the accretor, the trouble starts before it reaches critical rotation, 
at line 1453, where it says

>  retry: delta_mdot_hard_limit     146
The code is hitting an hard limit and this triggers a retry. In 
particular, here it is triggered by a delta_mdot that is too large. At 
the current step mdot is 2.110E-04, while in the previous step it was 
9.313E-05.

this gives

delta_mot=abs(mdot-mdot_old)/(delta_mdot_atol*Msun/secyer+delta_mdot_rtol*max(abs(mdot),abs(mdot_old)))

which is larger than 5, the mdot_hard_limit that you set in the star 
inlists, triggering a retry. So MESA will now try to recompute the step 
146 with a smaller timestep, which is usually a good idea.
Except that in this case, where for some reason I don't know, the next 
retry comes up with an even larger mdot, namely 2.334E-04, causing 
another retry. The situation gets rapidly out of hand until at line 2062 
mdot has become about 0.7 Msun/year (!!), and this huge amount of mass 
in one timestep is sufficient to bring the accretor above critical 
rotation, so that MESA will print

> w > w_crit: revise mdot and redo
and this goes on until the dt becomes smaller than the minimum dt 
allowed and MESA terminates.

This is what I could understand. As for why this is happening, I don't know.
*My suggestion is to try and relax the **delta_mdot_hard_limit and see 
if you are able to get through this.*

For debuggin purposes it can also be helpful to do terminal_interval=1, 
and print the summarizing informations about the stars and the binary 
every step, instead of every 10 steps.

I hope this helps,

Best,

Ruggero

On 7/16/23 14:45, Levi via Mesa-users wrote:
>
> Hello all,
>
>  I am currently working on creating a grid of massive binaries, and I 
> have encountered some difficulties during the runtime. In particular, 
> after mass transfer begins, M2 consistently reaches critical rotation 
> due to excessive accretion of angular momentum. And it unable to 
> effectively remove sufficient angular momentum, causing the model to 
> cease evolving.
>
> I have already attempted to address the issue by adjusting the 
> “implicit_mdot_boost” and “mdot_revise_factor” parameters to increase 
> mass and angular momentum loss. Regrettably, these modifications have 
> not had the desired effect.
>
> I have attached the model in the attachment, which can running by the 
> latest version of MESA-r23.05.1. Additionally, I have saved the 
> terminal output information in the “rn.out” file.
>
> I would greatly appreciate any guidance, suggestions, or solutions you 
> can provide.
>
> Best regards,
>
> Levi
>
>
>
> _______________________________________________
> mesa-users at lists.mesastar.org
> https://lists.mesastar.org/mailman/listinfo/mesa-users
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.mesastar.org/pipermail/mesa-users/attachments/20230716/24af9dab/attachment.htm>