[Mesa-users] Convergence problems for low mass planets

[Isaac Malsky]

Hi Dr. Farmer,

Without the restriction on max retries the model still fails, due to what I
suspect is how MESA is modeling the H/He envelope. I've attached the output
without max_number_retries set so low, and another run where I set flux to
1d7 erg/cm^2/s to get the model to fail quicker.

At the slightly higher flux the model seems to run indefinitely (25000+
models) with dt stable at around 5e2.

Thanks for the help!

Best,
Isaac

On Wed, May 1, 2019 at 11:46 AM Rob Farmer <r.j.farmer at uva.nl> wrote:

> Hi
> Your model is failing due to max_number_retries this option is only really
> useful for developers when we are testing the code. Try removing it and
> seeing if you model gets further
>
> Rob
>
> On Mon, 29 Apr 2019 at 21:32, Isaac Malsky via Mesa-users <
> mesa-users at lists.mesastar.org> wrote:
>
>> Hi everyone,
>>
>>
>> I'm trying to trace back the cause of convergence issues, and I'd greatly
>> appreciate any help/advice. I'm running a suite of sub-Neptune mass, highly
>> irradiated planets, and having problems evolving planets with ~10%
>> envelopes past 5 billion years.
>>
>>
>> The specific model that I'm evolving here:
>>
>>
>> M_p = 22.5 earth mass
>>
>> Y = .24
>>
>> Z = .02
>>
>> core mass fraction = .9
>>
>> flux = 1d8 erg/cm^2/s
>>
>> column depth = 1d1 cm^2/g
>>
>>
>> This model fails to evolve for the full 10 billion year evolution.
>> However, increasing irradiation_flux by a factor of 10 allows the planet to
>> evolve perfectly. The model fails to evolve past ~5.4e9 years, and the
>> final hydro_call_number before failure is 212. Also for reference, I'm
>> using mesa-r10398.
>>
>>
>> I've included the .mod file to load in, and the inlist to run, as well as
>> the terminal output. Also, I've attached the logs generated through the
>> debugging outlined in:
>>
>> http://wmwolf.github.io/projects/mesa_debugging/
>>
>>
>> I'd appreciate any insight about which routine is pushing the timestep so
>> low, as I would normally expect models like this to evolve.
>>
>>
>> Thanks for the help!
>>
>>
>> Best,
>>
>> Isaac
>> _______________________________________________
>> 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/20190501/56915ec3/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: Output.zip
Type: application/zip
Size: 1216139 bytes
Desc: not available
URL: <https://lists.mesastar.org/pipermail/mesa-users/attachments/20190501/56915ec3/attachment.zip>