[mesa-users] helium flash mlt bug

[Jakub Ostrowski]

Hi!

I'm writing to report a bug I found when I was trying to calculate a 1 M_S model evolution.

When mixing_length_alpha = 1.8 there is a problem with helium flash. As you can see in the attached HR diagram the behavior in the vicinity of the tip of giant branch is totally wrong. After this weird behavior is done, everything seems to look pretty normal.

The problem occurs independently of other parameters. I attached the inlist I used, but when I changed other parameters, like enabling overshooting or choosing different reaction rates, the problem was also present.

The behavior was correct when I set mixing_length_alpha = 2.0, as you can see in the second attached figure. It was also correct with 1.8 but in 2 M_S model.

I would also like to ask about the behavior after the helium flash. There is chaotic behavior in the HR diagram (you can see it in both figures) before the proper onset of helium burning. Do you know if there is some physical reason behind it or is it numerical problem?

I use MESA 4740 with latest sdk on OS X 10.8.2. The problem is also present on Ubuntu 12.10 (v4740, latest sdk). If you need more informations or files let me know.

Cheers,
Jakub


-------------- next part --------------
A non-text attachment was scrubbed...
Name: inlist_1m_mlt18
Type: application/octet-stream
Size: 13927 bytes
Desc: not available
URL: <https://lists.mesastar.org/pipermail/mesa-users/attachments/20130109/138a00b2/attachment.obj>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 1m-hr-mlt18.ps
Type: image/ps
Size: 76460 bytes
Desc: not available
URL: <https://lists.mesastar.org/pipermail/mesa-users/attachments/20130109/138a00b2/attachment.bin>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 1m-hr-mlt20.ps
Type: image/ps
Size: 77768 bytes
Desc: not available
URL: <https://lists.mesastar.org/pipermail/mesa-users/attachments/20130109/138a00b2/attachment-0001.bin>