[Mesa-users] Question about GR corrections for SMS

[Rob Farmer]

Hi
Please keep all replies on mesa-users so others can learn and/or help.

You should never need to alter the f90 source files (except your
run_star_extras.f90). Mesa has the concept of "hooks" which can change the
internal physics without needing to alter the mesa source code.

In this case look at the routine in star/other/other_cgrav.f90. If you
follow the instructions you can alter G however you want. This has the
advantage that it ensures that all occurrences of G have your corrections
applied.

You can then call your routine again from your run_star_extras.f90 in the
data_for_extra_profile_columns function to save the values to the profile
file.

Rob


On Tue, 23 Jun 2020 at 16:55, Nicholas Herrington <
Nicholas.Herrington at myport.ac.uk> wrote:

> Hi Rob
>
> Sorry for the very late reply, I haven't been able to access my work
> machine in a while. After coming back to it, I have not been able to
> replicate the 10^6 Msun run in version 11701 (only 500-800x10^5). I spoke
> to others in the field and they have said that without the GR corrections
> these stars don't necessarily have to reach 10^6 Msun.
>
> Since the 12778 doesn't have post-Newtonian corrections to the G, I have
> added the corrections to G within the subroutine associated to
> 'use_gr_factors' (within the hydro_reconstruct.f90). Instead of 'G =
> cgrav*gr_factors', I have added my own 'G = cgrav*Grel'. How can I output
> Grel and G in my profiles or history.data? Also is there a better way for
> me to correct G rather than changing the f90 files?
>
> Thanks
> Nick
>
> On Fri, 15 May 2020 at 18:35, Rob Farmer <r.j.farmer at uva.nl> wrote:
>
>> Hi,
>>
>> >Is there anything along the lines of the TOV approximation? Was
>> use_gr_factors intended for SMS?
>>
>> No only the changes to s% grav are implemented.
>>
>> > which is why I wonder if the new version has this.
>>
>> It is still there in 12778. But lots of other things will have changed
>> between versions, which may be stopping the star from getting to 10^6msun
>> (which is what i interpret your saying is the problem?). Or maybe you've
>> stumbled on a bug, it would be helpful if you could make a small test case
>> that shows this problem.
>>
>> Rob
>>
>>
>>
>> On Fri, 15 May 2020 at 18:59, Nicholas Herrington via Mesa-users <
>> mesa-users at lists.mesastar.org> wrote:
>>
>>> Hi,
>>>
>>> I am evolving some supermassive primordial stars (SMS) with accretion
>>> rates up to 1 Msun/yr. In the newest version of MESA is there general
>>> relativistic correction to G for SMS. I have found the control
>>> use_gr_factors with corrections to cgrav along the lines of
>>> (1-2GM/rc^2)^-0.5. Is there anything along the lines of the TOV
>>> approximation? Was use_gr_factors intended for SMS?
>>>
>>> I was running a SMS with version r11701 and with use_gr_factors off I
>>> get results exceeding 10^6 Msun which is expected, but with version r12778
>>> and the same inlist (use_gr_factors = .false.), the star reaches 10^5. This
>>> is close to the final mass of the same star from other stellar evolution
>>> codes with post-newtonian corrections to G, which is why I wonder if the
>>> new version has this.
>>>
>>> Thanks
>>> Nick
>>>
>>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.mesastar.org/pipermail/mesa-users/attachments/20200623/669c1a00/attachment.htm>