[ET Trac] #2520: update gauge settings in TOV example to be more typical

David Boyer trac-noreply at einsteintoolkit.org
Wed Jul 17 16:25:09 CDT 2024


#2520: update gauge settings in TOV example to be more typical

 Reporter: Roland Haas
   Status: new
Milestone: ET_2023_11
  Version: development version
     Type: enhancement
 Priority: major
Component: 

Comment (by David Boyer):

@{557058:59e031ba-9bb5-4298-a472-7b99d0ae6f22} I attempted the changes of the parfile as mentioned in the ticket. I’m posting what I changed in the parfiles and the plots I got. I would like to hear your opinion on if it is okay, or if there is another fix you suggest.  
  
_**Original parfiles**_:

```
ML_BSSN::harmonicN           = 1      # these set:
ML_BSSN::harmonicF           = 1.0    # dot[alpha] = - alpha A
ML_BSSN::evolveA             = 1      # dot[A] = dot[trK]
ML_BSSN::evolveB             = 1      # these set: dot[beta] = dot[dot[beta]] = 0
ML_BSSN::ShiftGammaCoeff     = 0.0    # by choosing coeffs to zero the RHS
ML_BSSN::AlphaDriver         = 0.0
ML_BSSN::BetaDriver          = 0.0
ML_BSSN::advectLapse         = 0
ML_BSSN::advectShift         = 0
```

  
_**I Replaced with**_:

```
ML_BSSN::harmonicN           = 1      # 1+log
ML_BSSN::harmonicF           = 2.0    # 1+log
ML_BSSN::ShiftGammaCoeff     = 0.75
ML_BSSN::AlphaDriver         = 0.0
ML_BSSN::BetaDriver          = 0.0
ML_BSSN::advectLapse         = 1.0
ML_BSSN::advectShift         = 1.0
```

  
_**Summary of differences**_:  
I mainly had an issue with the given BetaDriver. **I did first attempt this by using what was directly stated in the latter half of the ticket \(no changes\), but whenever BetaDriver was turned on \(set to 1\), densities would begin to runaway with nans at around t=60M. Turning BetaDriver off seemed to fix this issue, but I’m not sure if it was okay to do.**   
  
On a smaller note, I also changed Lapse/ShiftAdvectionCoeff to advectLapse/Shift, as it was stated in the ML\_BSSN thorn that the former was the outdated name and was renamed to advectLapse/Shift. Same variable, just a new name. I don’t think this point really changes anything, just using the up-to-date variable names.  
  
_**Plots**_:  
I have included my plots here. They are VERY NEARLY the same plots as the current gallery example, but there seems to be slight differences, the most noticeable being slightly more runoff at the end of the low res run for density.  
  
Are these plots okay, or were they meant to be point-per-point exactly the same? And if BetaDriver needs to be set to 1, is there another variable that needs to be changed to avoid the runaway nans?

![](https://bitbucket.org/repo/8zBLME5/images/160932728-convergence_p.png)
![](https://bitbucket.org/repo/8zBLME5/images/3330105359-density_p.png)
‌

--
Ticket URL: https://bitbucket.org/einsteintoolkit/tickets/issues/2520/update-gauge-settings-in-tov-example-to-be
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.einsteintoolkit.org/pipermail/trac/attachments/20240717/423a4aab/attachment.htm>


More information about the Trac mailing list