[Mesa-users] Relaxing Entropy from Profile Pairs?

Sun Apr 16 19:16:12 UTC 2023

Hello Charles,

I believe you're asking how MESA decides where to put zones?

See section 6.5 on 'Mesh adjustment' in Paxton et al. 2011
<https://ui.adsabs.harvard.edu/abs/2011ApJS..192....3P/abstract>.
-------
MESA is an adaptive mesh evolution code. The default number of zones and
where they are placed is determined by a wide range of individual mesh
functions which apply weights to various areas in the star (e.g. to resolve
temperature, pressure, and composition gradients among many other things)
in the controls_defaults, see mesh_adjustment
<https://docs.mesastar.org/en/release-r22.11.1/reference/controls.html#mesh-adjustment>.
A user can adjust the mesh resolution by changing any one of these many
controls, or use global mesh adjustment parameters like "mesh_delta_coeff"
which is a multiplicative factor on all mesh functions, and you can even
set a floor using "max_dq" which ensures that for a single zone dq = max_dq
so the minimum number of zones is 1/max_dq. The size of the mesh is
adaptively changed on each timestep, so the zones are not fixed to a
specific value at any point in the evolution.

-EbF

On Sat, Apr 15, 2023 at 11:08 AM Charles Gibson <gibson01 at allegheny.edu>
wrote:

> Thank you so much! This helped me out a ton. I was able to successfully
> relax these data from both a MESA run and a StarSmasher run.
>
> I don't think that I was too sure of what I was asking for with q values.
> I should have asked how MESA determines how many shells should be in a
> model, and how it divides those shells. It doesn't look like they are
> divided into equal mass or equal radius. Is there a table or equation that
> I can use to see how these shells may be spaced?
>
> Thank you again,
> Charlie
>
> On Sun, Apr 2, 2023 at 3:50 PM Ebraheem Farag <ekfarag at asu.edu> wrote:
>
>> Hello Charles,
>>
>> See : get_entropy_for_relax_from_eos
>> <https://urldefense.com/v3/__https://docs.mesastar.org/en/release-r22.11.1/reference/star_job.html*get-entropy-for-relax-from-eos__;Iw!!IKRxdwAv5BmarQ!dz_yB_S8baDh9yvA33gF86O0SivMIKTJFqfSiA5Ec7lulPHY0QDLYJleCFxwM2aG2emuYqKMM7_E_0iRWiaSlw$>
>>
>> Try setting the following in your &starjob section of your inlist:
>>
>> get_entropy_for_relax_from_eos = 'eosDT'
>> -------------------------------------------
>> "when generating the number of shells and each of their q values (how
>> much mass is located outside the shell), is there a specific formula that
>> determines how I should determine those values?"
>>
>> see below also from from star_job.defaults,
>> get_entropy_for_relax_from_eos
>> <https://urldefense.com/v3/__https://docs.mesastar.org/en/release-r22.11.1/reference/star_job.html*get-entropy-for-relax-from-eos__;Iw!!IKRxdwAv5BmarQ!dz_yB_S8baDh9yvA33gF86O0SivMIKTJFqfSiA5Ec7lulPHY0QDLYJleCFxwM2aG2emuYqKMM7_E_0iRWiaSlw$> (in
>> mesa-r22.11.1), please verify against the r10398 defaults:
>>
>> When any of the eos* options is used, then each line in the input file
>> must contain three columns instead of two, specifying the values of the two
>> thermodynamic variables used in the order specified above. So, for example,
>> when using ‘eosDT’ the format of the input file is
>>
>> 1st line: num_pointsthen 1 line for for each point where define desired entropyxq entropyxq = fraction of xmstar exterior to the pointwhere xmstar = mstar - M_centerentropy = specific entropy in units of erg/gr/K
>>
>> -------------------------------------------
>>
>> I hope this helps!
>> -EbF
>>
>>
>> On Sun, Apr 2, 2023 at 10:48 AM Charles Gibson via Mesa-users <
>> mesa-users at lists.mesastar.org> wrote:
>>
>>> Hello,
>>>
>>> I am an undergraduate student working on taking a star evolved in MESA
>>> from an SPH code (StarSmasher) and inputting it back into MESA to observe
>>> its evolution. I have been working in the directory
>>>
>>> star/test_suite/relax_composition_j_entropy
>>>
>>> to ensure that I can accurately recreate a model from a previous mesa
>>> run. So far, I have succeeded in replicating the star's evolution through
>>> this method.
>>>
>>> However, the StarSmasher simulations that I am using do not supply
>>> direct values for entropy. I read in Appendix B of this paper
>>> <https://urldefense.com/v3/__https://iopscience.iop.org/article/10.3847/1538-4365/aaa5a8*apjsaaa5a8app2__;Iw!!IKRxdwAv5BmarQ!eAVuCwomn0wLGPlCi_4SqSvQrROsX2Y_TuBprfGga0bhSO6wXIJlmJc-W0u6zBGNWJ1gv92vrZBdqqf020zJGhiz$> that
>>> I can supply different pairs of profiles including (rho, T), (P_gas, T),
>>> and (rho, e) in the entropy's place. These are all quantities that I can
>>> easily obtain.
>>>
>>> When I run the relax model using (rho, T) in place of entropy, the run
>>> fails when processing the entropy file. I have it formatted the same as the
>>> entropy file, but with three values on each line instead of two. Is there a
>>> variable in the inlist that I have to alter to indicate that I am changing
>>> the type of supplied entropy data? If it helps, I am using v10398.
>>>
>>> Additionally, as a side question, when generating the number of shells
>>> and each of their q values (how much mass is located outside the shell), is
>>> there a specific formula that determines how I should determine those
>>> values? I haven't gotten to the point where I am extracting values from
>>> StarSmasher yet, but once I do, I want to make sure that I determine the
>>> values so that MESA can effectively read them.
>>>
>>> Thank you,
>>> Charlie
>>> _______________________________________________
>>> mesa-users at lists.mesastar.org
>>>
>>> https://urldefense.com/v3/__https://lists.mesastar.org/mailman/listinfo/mesa-users__;!!IKRxdwAv5BmarQ!eAVuCwomn0wLGPlCi_4SqSvQrROsX2Y_TuBprfGga0bhSO6wXIJlmJc-W0u6zBGNWJ1gv92vrZBdqqf02-SGqssJ$
>>>
>>>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.mesastar.org/pipermail/mesa-users/attachments/20230416/5f19b612/attachment.htm>