[mesa-users] binary evolution(timestep controls)

[吴优]

Hi all,

I am currently working on binary simulations with MESA (version 7624),and I have some problems with the binary timestep controls.

In my run,I start with a 5M_sun star vs a 1.4M_sun star,and set 1.4M_sun as the accretor star(just taking it as a point mass).

Then set the initial condition to make the RLOF occur on main sequence stage of the 5M_sun star. I know it's possiable that the mass transfer is unstable,So l set the  "max_implicit_abs_mdot=1d-4",and limit accretion into the donor using mdot_edd.

However,the timestep of the model have some problems, which will occur shortly after the start of mass transfer, The value of the timesteps will drop to Logdt~-3 very fast, and the model return some errors like"hydro_newton_step failded to converge","hydro_newton_step returned ierr". as you can seen in my attached figure "masstransfer.png".

I have tried to change some parameters of timestep controls,but it dose not seem to work in my model.Ideally i expect that the timestep remain unchanged or will not drop too much.

Is there any way to achieve that? 

Many thanks!

You Wu


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.mesastar.org/pipermail/mesa-users/attachments/20160526/d35a39cb/attachment.html>