[mesa-users] WD models with low temperature

Thu Apr 14 16:25:12 EDT 2016

On Apr 14, 2016, at 1:23 PM, chenhl wrote:

> Dear Bill,
> 
> Thanks a lot for your reply! 
> 
> These plots are very helpful for understanding.  Now I know what's wrong with my initial model.
> 
> I tried to make a new WD model without the He shell.  Then I succeeded to reduce its central temperature to 1.0e7 K.
> Thanks again for your help!
> 
> Best wishes,
> Hailiang 

That's good news!  Glad we've made some progress today. ;)

-Bill

> 
> 
> 
> On Thu, Apr 14, 2016 at 9:02 PM, Bill Paxton <paxton at kitp.ucsb.edu> wrote:
> 
> On Apr 14, 2016, at 11:03 AM, chenhl wrote:
> 
>> Dear Bill,
>> 
>> Thanks a lot for reply!
>> 
>> Actually I have noticed that difference. I have tried to run the calculation with which_atm_option = 'grey_and_kap' in the second inlist file. 
>> Indeed the code can keep going. But after one hour, the central temperature only change slightly.  Then I simply stop the calculation.
> 
> ok.  that's much different behavior that I'm seeing for the C/O wd that is used in the wd_cool test case.
> 
> let's take a look -- I'm now running with your wd1.mod to see what I find.
> 
> some things to try:
> 
> varcontrol_target = 5e-5
> 
> this is probably much lower than you really need -- at least for initial test runs.  i'm setting it to 1d-3 for now.
> 
> mesh_delta_coeff = 0.7
> 
> this is giving us over 2400 cells.  we can probably get by with a lot fewer.  i'm setting it to 1.2 for now.
> 
> which_atm_option = 'grey_and_kap'
> 
> the surface of your model has X=0, Y=0.92, Z = 0.08 -- so using opacities for the surface BC may not be necessary or useful.  let's go simpler:
> 
> which_atm_option = 'simple_photosphere'
> 
> 
> now with those changes we see something interesting -- lots of burning is still going on!
> 
> if you want to get to a cool center for your ONe wd, you have to get beyond the stage of doing He burning in the envelope.   That means waiting patiently while it does flashes, or stepping in and turning on a mass loss mechanism of some sort to remove the helium layer at the surface.
> 
> There is also still strong C burning happening just below the He layer.   That will also force smaller timesteps until it finishes.
> 
> The cooling of the center will have to wait until the burning completes.    The model you are starting with is still stuck at the tip of the AGB doing He shell flashes and C burning in thin layers near the surface.     That needs to finish before cooling at the center can get going.
> 
> LOOK AT PLOTS TO SEE WHAT IS HAPPENING IN THE MODEL!!!!    
> (sorry for the cap's -- sometimes I just can't stop myself from shouting.)
> 
> 
> 
> compare those to the plots for the C/O wd at the start of cooling in wd_cool -- the burning is complete in this case and the code can take timesteps of millions of years here vs under a year when the burning is still going on in your ONe model.
> 
> 
> 
> Cheers,
> Bill
> 
> 
> 
> 
> 
> 
> 
> 
> 
> 
>> 
>> In the former email, you said you have a check. I guess you use some CO WD models.  
>> Did you try to use some ONe WD models or the model I attached in the first email? 
>> 
>> Best wishes,
>> Hailiang 
>> 
>> On Thu, Apr 14, 2016 at 7:40 PM, Bill Paxton <paxton at kitp.ucsb.edu> wrote:
>> Hi,
>> 
>> Just as a check, I let wd_cool keep going as long as it could.  All was fine until the center temperature dropped below 1e6.  At that point we really do fall off the edge of the world in the current eos, and it gives up.  So it runs great for about 800 steps, then abruptly dies.  Here's the last good model:
>> 
>> 
>> bill
>> 
>> 
>> 
>> 
>> 
>> 
>> 
>> On Apr 14, 2016, at 6:14 AM, chenhl wrote:
>> 
>> > Dear all,
>> >
>> > I want to make some ONe WD models with central temperature Tc ~ 1.0e7 K.
>> > But I always fail to cool the WD down.
>> >
>> > I try to do this with two steps.
>> > First, I use make_o_ne_wd test case to make a WD model with a mass I need.
>> > Then I use wd_cool test case to reduce the central temperature.  I always fail at the second
>> > step with following error.
>> >
>> > ----------------------------------------------------------------------------
>> >                                                     dt    8.7052562366518861D-14
>> >                                      min_timestep_limit    1.0000000000000000D-13
>> >
>> >  stopping because of convergence problems dt < min_timestep_limit
>> >
>> >
>> >
>> >
>> >  terminated evolution: convergence problems
>> >
>> >
>> >                   runtime, retries, backups, steps         15.275467          2.000000       188        69      2693
>> >        2693   8.551489  1.065E+05 -12.382713   5.451730   1.253481   0.000000   0.000000   0.000000   0.919403   0.000000  28.486602   3126    188
>> >  -19.325921   7.496896  -0.246400   3.879481   5.942448 -99.000000   1.253481   0.000000   0.373003   0.080597   0.014044   6.309162     25     69
>> >  2.3535E+07  24.633374   4.565573   5.439943   7.114711  -6.300953   1.253481   0.030127   0.508866   0.000000  9.860E-01 -0.833E+02   convergence
>> >
>> > Fri Apr 15 18:24:37 CST 2016
>> >
>> > ******************************************************************
>> > failed to create cool.mod when running inlist_wd_cool2
>> > ******************************************************************
>> >
>> > The version of MESA I used is 4906. The inlist files I used are attached.
>> >
>> > Is there anyone have experience on this?  Thanks!
>> >
>> > Best wishes,
>> > Hailiang
>> >
>> >
>> >
>> >
>> >
>> > <inlist_wd_cool1><wd1.mod><inlist_make_o_ne_wd><inlist_wd_cool2>------------------------------------------------------------------------------
>> > Find and fix application performance issues faster with Applications Manager
>> > Applications Manager provides deep performance insights into multiple tiers of
>> > your business applications. It resolves application problems quickly and
>> > reduces your MTTR. Get your free trial!
>> > https://ad.doubleclick.net/ddm/clk/302982198;130105516;z_______________________________________________
>> > mesa-users mailing list
>> > mesa-users at lists.sourceforge.net
>> > https://lists.sourceforge.net/lists/listinfo/mesa-users
>> 
>> 
>> 
> 
> 
> 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.mesastar.org/pipermail/mesa-users/attachments/20160414/4a59f17c/attachment.html>