[mesa-users] Z=0.04 and MESA r9793

[RICHARD H D TOWNSEND]

> On Jun 2, 2017, at 10:03 AM, Josiah Schwab <jwschwab at ucsc.edu> wrote:
>
> Hi Philip,
>
>> I want to use MESA r9793 to evolve a 1.3Msol, Y=0.32, Z=0.04 star from the
>> zero age main sequence. I find strange behaviour near the base of the giant
>> branch for this high metallicity star: a jump in L and Teff, as shown in
>> the attached HR diagram (z0.04...pdf, red track, jump between the crosses).
>> The same model evolved with r8845 (black track) does not show this jump.
>>
>> I assume this is related to changes to the opacity and/or equation of
>> state, but I haven't been able to solve it. Do you have any ideas?
>
> As you suspected, this an EOS issue.  There are OPAL tables for Z = 0,
> 0.02, and 0.04.  For higher Z, we switch to HELM.
>
> The new release has controls that expose some of the blending options to
> the user (as opposed to hardcoding them).
>
> Try running with the options in star_job
>
>    set_HELM_OPAL_Zs = .true.
>    Z_all_OPAL = 1.0
>    Z_all_HELM = 1.0
>
> that should keep you on the OPAL tables (i.e. you'll never blend into
> HELM).
>
> What I don't understand (and will have to investigate) is why this
> particular case behaves so badly.  There was not supposed to have been a
> significant change in the default behavior.
>

As an addendum to this: I can’t get a metal rich, ~ 1Msun pre-main sequence model to converge unless I add the stuff above to stay on the OPAL tables. Example inlist attached.

So, it looks like there are some not-insigificant breakages from the higher-Z switch.

cheers,

Rich




-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.mesastar.org/pipermail/mesa-users/attachments/20170605/5e04a2d9/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: inlist_project
Type: application/octet-stream
Size: 935 bytes
Desc: inlist_project
URL: <https://lists.mesastar.org/pipermail/mesa-users/attachments/20170605/5e04a2d9/attachment.obj>