[Mesa-users] Relaxing Giant Star

[Francis Timmes]

hi brad,

maybe both models are correct within their assumptions.

besides hydrostatic equilibrium and the thermodynamics,
both of which could be large effects, what else might be different? 
does the 3d model have sufficient resolution in the outer regions 
to capture the atmosphere (exponential falloff) that the mesa model 
appears to show? does the input 3d model account for photon and 
neutrino energy losses? might the cumulative physics differences 
account for both models being "correct" in their respective frameworks?

from a stellar engineering perspective, maybe multiply the input entropy profile 
by a constant to get better agreement with the central region's rho & temperature profiles.
in some sense, this path "corrects" for the (unknown) initial differences in the thermodynamics.

fxt






> On Dec 16, 2022, at 2:59 PM, Bradley Munson via Mesa-users <mesa-users at lists.mesastar.org> wrote:
> 
> Hello All,
> 
> I am using MESA version 21.12.1 on macOS 11.2.2. 
> 
> I am attempting to use the MESA relaxation procedures to relax a giant star starting from ZAMS. I am relaxing the composition, angular momentum, and entropy (from density and temperature) from the input files I've included here. I've also included the inlist I have been working with.
> 
> The outcome is usually either the model crashes during the entropy relaxation process, or it relaxes to a structure that is very different from the input model. The input model comes from a 3D merger model that is spherically averaged, so I understand it will not be in HSE and the EoS is different, but the density and temperature of the MESA model is around an order of magnitude lower than the input model throughout the whole profile (and much lower in the outer 10% of the star by mass). I've attached a couple of plots showing the differences between the input model (dashed line) and MESA relaxed model (blue solid line).
> 
> <image.png>
> 
> We've tried turning on and off the gold standards during relaxation, changing the relaxation timescales and max dts, changing the EoS options, starting from pre-MS, changing the energy equation option, and interpolating the input data to get more data points, but nothing seems to help. Has anybody had success with relaxing massive stars in MESA from another code? Does anybody have any advice on what else we can try?
> 
> Any help will be much appreciated!
> 
> Brad<abund_alt.dat><am.dat><eosDT.dat><inlist_common><inlist_relax>_______________________________________________
> mesa-users at lists.mesastar.org
> https://lists.mesastar.org/mailman/listinfo/mesa-users