Tidal deformation #773
Tidal deformation #773
Conversation
# Conflicts: # star/private/hydro_rotation.f90
matthiasfabry
added
enhancement
|
|
||
| ! Initial eccentricity of the system | ||
| ! ~~~~~~~~~~~~~~~~~~~~ |
There was a problem hiding this comment.
I'm pretty sure Sphinx will emit a warning if the underline isn't as long as the heading.
| ! ~~~~~~~~~~~~~~~~~~~~ | |
| ! ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ |
| @@ -3,6 +3,6 @@ | |||
| # goal of QUIET is to make it easier to spot warnings and/or circular dependency | |||
| # errors. | |||
|
|
|||
| QUIET = yes | |||
| QUIET = no | |||
There was a problem hiding this comment.
I think we want yes in production.
| QUIET = no | |
| QUIET = yes |
|
|
||
| write (*, 1) 'no implementation for other_tidal_deformation_function' | ||
| ! must set f_switch | ||
| ierr = -1 |
There was a problem hiding this comment.
Though it's quite common in the blank other_* routines and I don't expect anyone to change anything here, just a comment that ifort will warn that f_switch has no value here.
| real(dp), parameter :: nudge = 1d-4 | ||
| real(dp), pointer :: xvals(:), yvals(:), yvals_gtr_than_1(:), fpfunc1d(:), ftfunc1d(:), & | ||
| irotfunc1d(:), otherrfunc1d(:), afunc1d(:) | ||
| logical :: inter_ok = .false., dbg = .true. |
There was a problem hiding this comment.
I'm guessing we don't generally want the debug output on, otherwise it isn't really debug output.
| logical :: inter_ok = .false., dbg = .true. | |
| logical :: inter_ok = .false., dbg = .false. |
| inter_ok = .true. | ||
| end if |
There was a problem hiding this comment.
I won't have a chance to look at this closely for a bit but do we want to inter_ok to .true. inside the debug block? Should it not be set to .true. in normal operation? I'd expect
| inter_ok = .true. | |
| end if | |
| end if | |
| inter_ok = .true. |
| if (dbg) then | ||
| write(*, 1) 'starting interpolator setup' | ||
| end if |
There was a problem hiding this comment.
Though it's cosmetic, we can just have
| if (dbg) then | |
| write(*, 1) 'starting interpolator setup' | |
| end if | |
| if (dbg) write(*, 1) 'starting interpolator setup' |
I imagine there was more in this if block while you were working on this.
| call star_ptr(id, s, ierr) | ||
| if (ierr /= 0) return | ||
| call binary_ptr(s% binary_id, b, ierr) | ||
| if (ierr /=0 ) return |
There was a problem hiding this comment.
| if (ierr /=0 ) return | |
| if (ierr /= 0) return |
|
|
||
| i_rot = 0d0 | ||
| if (s% simple_i_rot_flag) then | ||
| i_rot = (2d0/3d0)*r00*r00 |
There was a problem hiding this comment.
| i_rot = (2d0/3d0)*r00*r00 | |
| i_rot = two_thirds*r00*r00 |
| use auto_diff | ||
|
|
There was a problem hiding this comment.
What do we need from auto_diff here?
| @@ -243,44 +299,94 @@ subroutine check_omega(s, str) | |||
| end subroutine check_omega | |||
|
|
|||
| subroutine update1_i_rot_from_xh(s, k) | |||
| use auto_diff | |||
There was a problem hiding this comment.
As above, what do we need from auto_diff here?
|
Not a blocker for me but would it be possible to use this in the test suite somewhere, be that in a new test or tweaking an existing one? |
Following on #637, this (rather large) overhaul of
hydro_rotation.f90includes support for computing the stellar structure corrections of tidally distorted stars. This is especially useful when modeling contact binaries.The setup is as follows:
I used procedure hooks (akin to
binary_other_torquefor tidal J_dot calculations) to communicate the binary calculated fp, ft, and i_rot values for the stellar layers.Then these values are "blended" with the single-rotating-star corrections according to the synchronicity of the component to the orbit. I implemented a user hook
binary_other_tidal_deformation_switch_functionif one does not agree with this treatment.The user can still use the
other_fp_ftandother_i_rothooks to bypass all default calculations (and blending).