[Mesa-users] ppisn: evolve >150 Ms massive stars in 15140

[Jing-Ze Ma]

Hi Rob,


Thanks for your help! I tried setting convergence_ignore_equL_residuals = .true. but the evolution is still going very slowly.


It turns out that


 okay_to_reduce_gradT_excess = .true.
 gradT_excess_lambda1 = -1
 change_mass_years_for_dt = 1d-2
 mesh_delta_coeff = 0.5d0


would do the trick for a 250 Msun model in 15140.


But for more massive models (e.g., 300 Msun) it fails again. During relax_mass from 90 Msun to my specified initial_mass, I noticed that Tcntr and Tmax are not exactly the same, which might be what's keeping timestep small. I may try go back to the version in marchant 2019 if I want more massive models in the future.


Cheers,
Jing-Ze


-----Original Messages-----
From:"Rob Farmer" <r.j.farmer at uva.nl>
Sent Time:2021-05-25 20:07:50 (Tuesday)
To: "马竟泽" <mjz18 at mails.tsinghua.edu.cn>
Cc: mesa-users <mesa-users at lists.mesastar.org>
Subject: Re: [Mesa-users] ppisn: evolve >150 Ms massive stars in 15140


Hi,


Try setting:
convergence_ignore_equL_residuals = .true.
in your controls inlist to help speed up the evolution.



>I suppose that either this is a new problem from new version, or this has been solved in Marchant 2019 paper but not written in ppisn test_suite. Any suggestion would be deeply appreciated.


test cases  /= published versions. The test case needs to be fast and work well for the initial conditions it is set for, while hopefully that means it works well for similar systems but that is not guaranteed. If you really need 250msun models then i suggest using the inlists and mesa version specified in marchant 2019.


Rob





On Fri, 21 May 2021 at 17:07, Jing-Ze Ma via Mesa-users <mesa-users at lists.mesastar.org> wrote:

Hi mesa users,


I am having trouble evolving massive stars with mass above 150 Ms past He main sequence in ppisn test_suite. During relax_to_radiative_core or helium burning, I keep getting retries like


 retry: max residual jumped -- give up in solver
 retry: logT < hydro_mtx_min_allowed_logT
 retry: logRho > hydro_mtx_max_allowed_logRho


which slow things down and make it hard to get past He main sequence stage or even enter the main sequence stage. I am keeping everything as defaults in ppisn test_suite except with rotation_flag turned off. I am not sure whether this is caused by different mesa versions or not, since I am using 15140.




A possible solution I am experimenting is to add


 okay_to_reduce_gradT_excess = .true.
 gradT_excess_lambda1 = -1


and then turn off okay_to_reduce_gradT_excess when hydro is on, which would evolve a 150 Ms to the end.


By further adding


 change_mass_years_for_dt = 1d-2


I manage to evolve a 200 Ms to the end.
But for a 250 Ms model, I have not found a satisfying solution yet. Tweaking pre_ms_T_c or mesh_delta_coeff or turning off conv_vel_flag is not helping much.


I suppose that either this is a new problem from new version, or this has been solved in Marchant 2019 paper but not written in ppisn test_suite. Any suggestion would be deeply appreciated.


Thanks in advance!


Cheers,
Jing-Ze
 
_______________________________________________
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/20210528/34dbdd29/attachment.htm>