New general anelastic non-dimensional subroutine (reference_type = 5) #450
Conversation
illorenzo7
force-pushed
the
reftype5-clean
branch
from
a67ba15
to
4178449
Compare
| Endif | ||
|
|
||
| ! Determine how user wants to specify Ra, Chi_A_Ra, and B_visc. | ||
| ! (The modified versions or not). |
There was a problem hiding this comment.
We should have a branch for a model without rotation.
There was a problem hiding this comment.
Actually I think now it's fine. At most, we could have the code throw a warning if they set rotation = .false but decide to set a non-zero Modified_Rayleigh_Number (say). But we don't do anything like that for e.g. the Magnetic_Prandtl_Number (the user can set that with magnetism=.false. and no warning will be thrown).
src/Physics/PDE_Coefficients.F90
Outdated
|
|
||
| denom1 = (1.0d0 - exp(-poly_Nrho/poly_n))/(1.d0 - Aspect_Ratio) | ||
| denom2 = -(Aspect_Ratio - exp(-poly_Nrho/poly_n))/(Aspect_Ratio*Shell_Depth) | ||
| nsquared = (rmin/radius)**3 / (denom1 + denom2*radius) |
There was a problem hiding this comment.
Could we explicily write dsdr in terms of dlnrho and dlnP or dlnT and then multiply by the appropriate amplitude control parameter?
| Call Integrate_in_radius(nsquared,norm) | ||
| norm = four_pi*norm/shell_volume | ||
| nsquared = nsquared/norm | ||
|
|
There was a problem hiding this comment.
Can you compute dissipation number in a similar way to the new way of calculating dsdr?
There was a problem hiding this comment.
I think what I just did is the cleanest way to do it (no need to integrate temperature and gravity a second time).
Implements to subroutine
Polytropic_Reference_ND_General()
in PDE_Coefficients. Note that my other pull request (#447) is included here. This new functionality allows user to specify simulation using Prandtl number, Rayleigh number (flux or otherwise), Ekman number, magnetic Prandtl number, and four number characterizing the polytrop ice: specific-heat ratio (default 5/3), poly tropic index, number of density scale heights, and shell aspect ratio.
User can choose between non-dimensionalizing the reference state using the profile values at the inner radius, outer radius, or volume-averaged over the shell (default).
User can non-dimensionalize time using either the rotation rate (time = 1/Omega_0) or viscous diffusion time.
User can non-dimensionalize the length scale using an arbitrary value, with default being the shell depth.