[mesa-users] element diffusion during core he burning phase

[Bill Paxton]

Ahhh ---- your old run_star_extras turns on rotation.   And the smoothing from rotational mixing
is enough to make things very easy for element diffusion.  Without the rotation, diffusion
has a much more difficult job, and it isn't a surprise that it takes a long time to do it.

You might check the new sdb test case in the star test_suite to get an example for comparison.
Are you runtimes much worse than you see in the sdb test?  If so, compare the inlists to
see if you can find parameters to change that will improve things without compromising
the results too much.

good luck,
Bill





On Apr 30, 2013, at 11:16 AM, Ehsan Moravveji wrote:

> Hi Bill,
> I am sure it is not because of run_star_extras. I use this test with a standard run_star_extras.
> So, it must be a platform issue.
> A guess: with 4930, can i still  use the previous mesasdk, rather than the most recent?
> I think I did not recently update my mesasdk.
> 
> thanks for checking this, btw.
> Best regards,
> 
> Ehsan Moravveji
> 
> 
> 
> On Apr 30, 2013, at 8:12 PM, Bill Paxton wrote:
> 
>> Hi,
>> 
>> It runs without problem on my mac --- but I'm still on mac osx 10.6.8 rather than 10.7.
>> 
>> Do you have a modified run_star_extras that I should be using?
>> 
>> -B
>> 
>> 
>> 
>> 
>> 
>> 
>> 
>> On Apr 29, 2013, at 11:33 PM, Ehsan Moravveji wrote:
>> 
>>> Thanks Bill,
>>> I have reattached the inlist and also the model at the step number 1500. Wait for about 100 steps when center he4 drops below 0.90, and it happens.
>>> i hope it helps.
>>> Ehsan.
>>> 
>>> 
>>> On Tue, Apr 30, 2013 at 12:34 AM, Bill Paxton <paxton at kitp.ucsb.edu> wrote:
>>> Hi Ehsan,
>>> 
>>> I'd like to check this with my current development version, but I can't use the photo from the older version.
>>> Please save a model from about the same time as the photo and send that.
>>> (photos are specific to a particular version; saved models are not.)
>>> 
>>> Thanks,
>>> Bill
>>> 
>>> 
>>> 
>>> 
>>> On Apr 29, 2013, at 6:41 AM, Ehsan Moravveji wrote:
>>> 
>>> > Dear all,
>>> > This is a going to be a feedback. With version 4930 + mesasdk on Lion 10.7, I faced a difficulty during the runtime.
>>> >
>>> > When the 5 Msun star (inlist is attached) reaches the core He burning phase, the progress in the step number on the terminal goes very slow. I mention that lg_dt_yr is reasonably large. This happens when core he drops below ~0.90, and the model number is almost 1599. I have also attached a restart photo.
>>> >
>>> > Commenting out some lines of the inlist, I finally found the guy in charge:
>>> >         do_element_diffusion = .true.
>>> > without diffusion, the run progresses as fast as before.
>>> > I wonder if the result would be the same on different platforms.
>>> >
>>> > I though that the captain of the ship and the rest of the crew might be interested!
>>> >
>>> > Best regards,
>>> > Ehsan Moravveji
>>> >
>>> >
>>> > <inlist_test>
>>> > <x1500>------------------------------------------------------------------------------
>>> > Try New Relic Now & We'll Send You this Cool Shirt
>>> > New Relic is the only SaaS-based application performance monitoring service
>>> > that delivers powerful full stack analytics. Optimize and monitor your
>>> > browser, app, & servers with just a few lines of code. Try New Relic
>>> > and get this awesome Nerd Life shirt! http://p.sf.net/sfu/newrelic_d2d_apr_______________________________________________
>>> > mesa-users mailing list
>>> > mesa-users at lists.sourceforge.net
>>> > https://lists.sourceforge.net/lists/listinfo/mesa-users
>>> 
>>> 
>>> 
>>> 
>>> -- 
>>> Kind regards
>>> 
>>> Ehsan Moravveji
>>> http://moravveji.webs.com/
>>> <inlist_test><mod1500-diffusion-on>
>> 
> 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.mesastar.org/pipermail/mesa-users/attachments/20130430/de2c76e2/attachment.html>