Table and mixture EOS #96
Conversation
…temperature or density. Useful for ICs
|
@pdmullen should be ready for review |
adamdempsey90
left a comment
adamdempsey90
left a comment
There was a problem hiding this comment.
Still need docs
| w_b = eos_d.BulkModulusFromDensityInternalEnergy(w_d, w_s, lambda); | ||
| w_t = eos_d.TemperatureFromDensityInternalEnergy(w_d, w_s, lambda); |
There was a problem hiding this comment.
If we are storing/tracking temperature now, should we change other EOS calls to *FromDensityTemperature? I like working in (rho, T) space since rho and T are guaranteed to be positive unlike sie (for general EOS).
There was a problem hiding this comment.
I don't really want to carry temperature around. It's only here so we can output it, but there might be other ways to do that. But the proper formulation is density internal energy, so I'd prefer to keep that as is.
src/pgen/disk.hpp
Outdated
| Real rtp = xcyl[0] + dxr; | ||
| Real rtm = xcyl[0] - dxr; | ||
|
|
||
| // this is a mess |
There was a problem hiding this comment.
What's going on inside here?
There was a problem hiding this comment.
I need dPhi/dR in cylindrical coordinates, so I am approximating the gradient in whatever coordinate system we are running in and dotting it into the rhat unit vector.
| v(0, gas::prim::density(0), k, j, i) = dens; | ||
| v(0, gas::prim::sie(0), k, j, i) = pars.g_pres / ((gamma - 1.0) * dens); | ||
| // input density and pressure | ||
| v(0, gas::prim::sie(0), k, j, i) = ArtemisUtils::EofPR(eos, pres, dens); |
There was a problem hiding this comment.
Should we be setting temperature as a primitive variable instead of sie now?
There was a problem hiding this comment.
No, I think we should still keep it sie
|
|
||
| } // namespace ArtemisEOS | ||
|
|
||
| #endif // UTILS_EOS_IDEAL_H_HE_HPP_ |
There was a problem hiding this comment.
This is a lot of work...
| Real &wl_ivz = wl(ivz, i); | ||
| Real &wl_ipr = wl(IPR, i); | ||
| Real &wl_ise = wl(ISE, i); | ||
| Real &wl_ibl = wl(IBL, i); |
There was a problem hiding this comment.
I thought for normal hllc (i.e., not ghllc) that we were going to stay with the original reconstructions such that we maintain thermodynamic consistency for ideal gases.
There was a problem hiding this comment.
I think it would help me a little more if you explicitly spelled out how to do that (or pushed some changes that do that).
There was a problem hiding this comment.
I have added an extra step during reconstruction that makes all of the reconstructions self-consistent. I have chosen density and pressure as the ground truth. I chose pressure so that I can investigate hydrostatic reconstructions in the near future.
…this for to make the recon consistent for the gas
Background
This adds support for the ideal gas H/He mixture from D'Angelo+2013 and more general tabular equations of state using spiner.
There were some minor updates to the Riemann solvers and I've added both temperature and bulk modulus as allocated variables (the latter is needed for the solvers and the former is for outputs since the mapping from pressure and density is no longer trivial).
I've added a shock tube example demonstrating that this works. Testing is still ongoing. Shearing sheet seems to work.
There is likely some cleanup required in the EOS file itself, too.
Description of Changes
Checklist