Data Types
interface	Finalize

interface	SolveODE

type	timedisc_base

Functions/Subroutines
subroutine	setoutput (this, Mesh, Physics, config, IO)

subroutine	acceptsolution (this, Mesh, Physics, Sources, Fluxes, time, dt, iter)

subroutine	checkdata (this, Mesh, Physics, Fluxes, pvar, cvar, checkdatabm)
	compute the RHS of the spatially discretized PDE More...

pure integer function	getorder (this)

Variables
integer, parameter, public	modified_euler = 1

integer, parameter, public	rk_fehlberg = 2

integer, parameter, public	cash_karp = 3

integer, parameter, public	dormand_prince = 4

integer, parameter, public	ssprk = 5

integer, parameter, public	dtcause_cfl = 0

integer, parameter, public	dtcause_erradj = -1

integer, parameter, public	dtcause_smallerr = -2

integer, parameter, public	dtcause_fileio = -4
	smallest ts due to fileio More...

integer, parameter, public	check_nothing = INT(b'0')

integer, parameter, public	check_all = NOT(CHECK_NOTHING)

integer, parameter, public	check_csound = INT(b'1')

integer, parameter, public	check_pmin = INT(b'10')

integer, parameter, public	check_rhomin = INT(b'100')

integer, parameter, public	check_invalid = INT(b'1000')

integer, parameter, public	check_tmin = INT(b'10000')

character(len=40), dimension(3), parameter	fargo_method = (/ "dynamic background velocity ", "user supplied fixed background velocity ", "shearing box " /)

subroutine	inittimedisc (this, Mesh, Physics, config, IO, ttype, tname)

subroutine	integrationstep (this, Mesh, Physics, Sources, Fluxes, iter, config, IO)

real function	adjusttimestep (this, maxerr, dtold)
	adjust the time step More...

real function	calctimestep (this, Mesh, Physics, Sources, Fluxes, time, dtcause)
	Determines the CFL time step and time step limits due to source terms. More...

subroutine	rejectsolution (this, Mesh, Physics, Sources, Fluxes, time, dt)

real function	computeerror (this, Mesh, Physics, cvar_high, cvar_low)

subroutine	computerhs (this, Mesh, Physics, Sources, Fluxes, time, dt, pvar, cvar, checkdatabm, rhs)
	compute the RHS of the spatially discretized PDE More...

pure real function	getcfl (this)

subroutine	fargoadvectionx (this, Fluxes, Mesh, Physics, Sources)
	Calculates the linear transport step in Fargo Scheme along x-axis [mignone2012] . More...

subroutine	fargoadvectiony (this, Fluxes, Mesh, Physics, Sources)
	Calculates the linear transport step in Fargo Scheme along y-axis [mignone2012] . More...

subroutine	calcbackgroundvelocity (this, Mesh, Physics, pvar, cvar, w)
	Calculates new background velocity for fargo advection. More...

real function, dimension(mesh%igmin:mesh%igmax, mesh%jgmin:mesh%jgmax, mesh%kgmin:mesh%kgmax, physics%vdim)	getcentrifugalvelocity (this, Mesh, Physics, Fluxes, Sources, dir_omega_, accel_, centrot)
	Compute velocity leading to a centrifugal acceleration with respect to some given axis of rotation which balances any local radial acceleration. If the acceleration is not given explicitly determine it from a complete evaluation of of the right hand side of the transport problem. More...

subroutine	finalize_base (this)

Detailed Description

Author: Tobias Illenseer; Björn Sperling; Manuel Jung; Jannes Klee

Function/Subroutine Documentation

◆ acceptsolution()

subroutine timedisc_base_mod::acceptsolution	(	class(timedisc_base), intent(inout)	this,
		class(mesh_base), intent(in)	Mesh,
		class(physics_base), intent(inout)	Physics,
		class(sources_base), pointer	Sources,
		class(fluxes_base), intent(inout)	Fluxes,
		real, intent(inout)	time,
		real, intent(inout)	dt,
		integer	iter
	)

Definition at line 858 of file timedisc_base.f90.

◆ adjusttimestep()

real function timedisc_base_mod::adjusttimestep	(	class(timedisc_base), intent(inout)	this,
		real, intent(in)	maxerr,
		real, intent(in)	dtold
	)

adjust the time step

This function implements adaptive step size control based on an error estimate.

see E. Hairer, Solving Ordinary Differential Equ. II, 2ed, Springer (2.43c)

Definition at line 706 of file timedisc_base.f90.

◆ calcbackgroundvelocity()

subroutine timedisc_base_mod::calcbackgroundvelocity	(	class(timedisc_base), intent(inout)	this,
		class(mesh_base), intent(in)	Mesh,
		class(physics_base), intent(inout)	Physics,
		class(marray_compound), intent(inout)	pvar,
		class(marray_compound), intent(inout)	cvar,
		real, dimension(mesh%igmin:mesh%igmax,mesh%kgmin:mesh%kgmax), intent(out)	w
	)

Calculates new background velocity for fargo advection.

Attention: Only works when velocity is shifted in second direction.

Definition at line 1761 of file timedisc_base.f90.

◆ calctimestep()

real function timedisc_base_mod::calctimestep	(	class(timedisc_base), intent(inout)	this,
		class(mesh_base), intent(in)	Mesh,
		class(physics_base), intent(inout)	Physics,
		class(sources_base), pointer	Sources,
		class(fluxes_base), intent(inout)	Fluxes,
		real, intent(in)	time,
		integer, intent(inout)	dtcause
	)

Determines the CFL time step and time step limits due to source terms.

Definition at line 765 of file timedisc_base.f90.

◆ checkdata()

subroutine timedisc_base_mod::checkdata	(	class(timedisc_base), intent(inout)	this,
		class(mesh_base), intent(in)	Mesh,
		class(physics_base), intent(inout)	Physics,
		class(fluxes_base), intent(in)	Fluxes,
		class(marray_compound), intent(inout)	pvar,
		class(marray_compound), intent(inout)	cvar,
		integer, intent(in)	checkdatabm
	)

compute the RHS of the spatially discretized PDE

Definition at line 1334 of file timedisc_base.f90.

◆ computeerror()

real function timedisc_base_mod::computeerror	(	class(timedisc_base), intent(inout)	this,
		class(mesh_base), intent(in)	Mesh,
		class(physics_base), intent(in)	Physics,
		class(marray_compound), intent(inout)	cvar_high,
		class(marray_compound), intent(inout)	cvar_low
	)

private

Definition at line 937 of file timedisc_base.f90.

◆ computerhs()

subroutine timedisc_base_mod::computerhs	(	class(timedisc_base), intent(inout)	this,
		class(mesh_base), intent(in)	Mesh,
		class(physics_base), intent(inout)	Physics,
		class(sources_base), pointer	Sources,
		class(fluxes_base), intent(inout)	Fluxes,
		real, intent(in)	time,
		real, intent(in)	dt,
		class(marray_compound), intent(inout)	pvar,
		class(marray_compound), intent(inout)	cvar,
		integer, intent(in)	checkdatabm,
		class(marray_compound), intent(inout)	rhs
	)

compute the RHS of the spatially discretized PDE

This subroutine updates all data on the right hand side (RHS) of the ordinary differential equation obtained via spatial discretization of the system of partial differential equations (PDE). It expects that the conservative variables (cvar) contain valid data inside the computational domain. On the basis of this data the following update steps are performed in the given order:

update of ghost cell data (this implies an update of the primitive variables (pvar) on the whole grid including ghost cells)
update of intercell numerical fluxes (this implies an update of all face states and additional data necessary to compute the fluxes)
update geometrical source terms
update external source terms (this implies an update of all auxiliary data arrays used for sources terms)

Todo:: Very hacky implementation of pluto style angular momentum conservation. A better implementation is needed, but we would need to restucture the timedisc module

Definition at line 993 of file timedisc_base.f90.

◆ fargoadvectionx()

subroutine timedisc_base_mod::fargoadvectionx	(	class(timedisc_base), intent(inout)	this,
		class(fluxes_base), intent(inout)	Fluxes,
		class(mesh_base), intent(in)	Mesh,
		class(physics_base), intent(inout)	Physics,
		class(sources_base), pointer	Sources
	)

Calculates the linear transport step in Fargo Scheme along x-axis [mignone2012] .

The linear transport calculation is done by operator splitting. This yields an equation

\[ \frac{\partial q}{\partial t} + \mathbf{w}\cdot \nabla q = 0, \]

which can be descritized as

\[ q_j^{n+1}=q_j^{n}+ \frac{1}{\Delta y} \left( \int_{y_{j+\frac{1}{2}}-w\Delta t}^{y_{j+\frac{1}{2}}} q_j^{n}(\xi)\mathrm{d}\xi - \int_{y_{j-\frac{1}{2}}-w\Delta t}^{y_{j-\frac{1}{2}}} q_j^{n}(\xi)\mathrm{d}\xi \right). \]

FargoAdvectionX does the advection step along the x-axis with \( \mathbf{w} = w \mathbf{e_x} \).

Definition at line 1449 of file timedisc_base.f90.

◆ fargoadvectiony()

subroutine timedisc_base_mod::fargoadvectiony	(	class(timedisc_base), intent(inout)	this,
		class(fluxes_base), intent(inout)	Fluxes,
		class(mesh_base), intent(in)	Mesh,
		class(physics_base), intent(inout)	Physics,
		class(sources_base), pointer	Sources
	)

Calculates the linear transport step in Fargo Scheme along y-axis [mignone2012] .

The linear transport calculation is done by operator splitting. This yields an equation

\[ \frac{\partial q}{\partial t} + \mathbf{w}\cdot \nabla q = 0, \]

which can be descritized as

\[ q_j^{n+1}=q_j^{n}+ \frac{1}{\Delta y} \left( \int_{y_{j+\frac{1}{2}}-w\Delta t}^{y_{j+\frac{1}{2}}} q_j^{n}(\xi)\mathrm{d}\xi - \int_{y_{j-\frac{1}{2}}-w\Delta t}^{y_{j-\frac{1}{2}}} q_j^{n}(\xi)\mathrm{d}\xi \right). \]

FargoAdvectionY does the advection step along the x-axis with \( \mathbf{w} = w \mathbf{e_y} \).

Definition at line 1612 of file timedisc_base.f90.

◆ finalize_base()

subroutine timedisc_base_mod::finalize_base ( class(timedisc_base) this )

private

Definition at line 2011 of file timedisc_base.f90.

◆ getcentrifugalvelocity()

real function, dimension(mesh%igmin:mesh%igmax,mesh%jgmin:mesh%jgmax,mesh%kgmin:mesh%kgmax,physics%vdim) timedisc_base_mod::getcentrifugalvelocity	(	class(timedisc_base), intent(inout)	this,
		class(mesh_base), intent(in)	Mesh,
		class(physics_base), intent(inout)	Physics,
		class(fluxes_base), intent(inout)	Fluxes,
		class(sources_base), pointer	Sources,
		real, dimension(3), intent(in), optional	dir_omega_,
		real, dimension(mesh%igmin:mesh%igmax,mesh%jgmin:mesh%jgmax,mesh%kgmin:mesh%kgmax,physics%vdim), intent(in), optional	accel_,
		real, dimension(3), intent(in), optional	centrot
	)

private

Compute velocity leading to a centrifugal acceleration with respect to some given axis of rotation which balances any local radial acceleration. If the acceleration is not given explicitly determine it from a complete evaluation of of the right hand side of the transport problem.

The algorithm follows these steps:

Get curvilinear components of the vectors pointing from the origin/center of rotation into each cell.
Get curvilinear components of the vectors pointing from the axis of rotation perpendicular to that axis into each cell, i. e.
\[ \vec{R} = \vec{r} - (\hat{e}_\Omega\cdot\vec{r})\hat{e}_\Omega \]
. and the components of the azimuthal unit vector in terms of the local orthnormal basis
\[ \hat{e}_\varphi = \hat{e}_\Omega \times \vec{r} / r \]
Determine the acceleration to balance.
Compute absolute value of azimuthal velocity using the balance law
\[ \Omega^2 R\hat{e}_R = -\vec{a} \Leftrightarrow v_\varphi = \sqrt{\Omega^2 R^2} = \sqrt{-R \hat{e}_R\cdot\vec{a}} \]
Compute components of \( v_\varphi \hat{e}_\varphi \) with respect to the local orthonormal basis.

Todo:: move code depending on physics into appropriate subroutines in physics modules

Definition at line 1828 of file timedisc_base.f90.

◆ getcfl()

pure real function timedisc_base_mod::getcfl ( class(timedisc_base), intent(in) this )

private

Definition at line 1421 of file timedisc_base.f90.

◆ getorder()

pure integer function timedisc_base_mod::getorder ( class(timedisc_base), intent(in) this )

Definition at line 1412 of file timedisc_base.f90.

◆ inittimedisc()

subroutine timedisc_base_mod::inittimedisc	(	class(timedisc_base), intent(inout)	this,
		class(mesh_base), intent(inout)	Mesh,
		class(physics_base), intent(in)	Physics,
		type(dict_typ), pointer	config,
		type(dict_typ), pointer	IO,
		integer, intent(in)	ttype,
		character(len=32), intent(in)	tname
	)

private

Definition at line 216 of file timedisc_base.f90.

◆ integrationstep()

subroutine timedisc_base_mod::integrationstep	(	class(timedisc_base), intent(inout)	this,
		class(mesh_base), intent(inout)	Mesh,
		class(physics_base), intent(inout)	Physics,
		class(sources_base), pointer	Sources,
		class(fluxes_base), intent(inout)	Fluxes,
		integer, intent(inout)	iter,
		type(dict_typ), pointer	config,
		type(dict_typ), pointer	IO
	)

private

Definition at line 630 of file timedisc_base.f90.

◆ rejectsolution()

subroutine timedisc_base_mod::rejectsolution	(	class(timedisc_base), intent(inout)	this,
		class(mesh_base), intent(in)	Mesh,
		class(physics_base), intent(inout)	Physics,
		class(sources_base), pointer	Sources,
		class(fluxes_base), intent(inout)	Fluxes,
		real, intent(in)	time,
		real, intent(inout)	dt
	)

private

Definition at line 906 of file timedisc_base.f90.

◆ setoutput()

subroutine timedisc_base_mod::setoutput	(	class(timedisc_base), intent(inout)	this,
		class(mesh_base), intent(in)	Mesh,
		class(physics_base), intent(in)	Physics,
		type(dict_typ), pointer	config,
		type(dict_typ), pointer	IO
	)

Definition at line 515 of file timedisc_base.f90.

Variable Documentation

◆ cash_karp

integer, parameter, public timedisc_base_mod::cash_karp = 3

Definition at line 180 of file timedisc_base.f90.

◆ check_all

integer, parameter, public timedisc_base_mod::check_all = NOT(CHECK_NOTHING)

Definition at line 190 of file timedisc_base.f90.

◆ check_csound

integer, parameter, public timedisc_base_mod::check_csound = INT(b'1')

Definition at line 191 of file timedisc_base.f90.

◆ check_invalid

integer, parameter, public timedisc_base_mod::check_invalid = INT(b'1000')

Definition at line 194 of file timedisc_base.f90.

◆ check_nothing

integer, parameter, public timedisc_base_mod::check_nothing = INT(b'0')

Definition at line 189 of file timedisc_base.f90.

◆ check_pmin

integer, parameter, public timedisc_base_mod::check_pmin = INT(b'10')

Definition at line 192 of file timedisc_base.f90.

◆ check_rhomin

integer, parameter, public timedisc_base_mod::check_rhomin = INT(b'100')

Definition at line 193 of file timedisc_base.f90.

◆ check_tmin

integer, parameter, public timedisc_base_mod::check_tmin = INT(b'10000')

Definition at line 195 of file timedisc_base.f90.

◆ dormand_prince

integer, parameter, public timedisc_base_mod::dormand_prince = 4

Definition at line 181 of file timedisc_base.f90.

◆ dtcause_cfl

integer, parameter, public timedisc_base_mod::dtcause_cfl = 0

Definition at line 184 of file timedisc_base.f90.

◆ dtcause_erradj

integer, parameter, public timedisc_base_mod::dtcause_erradj = -1

Definition at line 185 of file timedisc_base.f90.

◆ dtcause_fileio

integer, parameter, public timedisc_base_mod::dtcause_fileio = -4

smallest ts due to fileio

Definition at line 187 of file timedisc_base.f90.

◆ dtcause_smallerr

integer, parameter, public timedisc_base_mod::dtcause_smallerr = -2

Definition at line 186 of file timedisc_base.f90.

◆ fargo_method

character(len=40), dimension(3), parameter timedisc_base_mod::fargo_method = (/ "dynamic background velocity ", "user supplied fixed background velocity ", "shearing box " /)

private

Definition at line 197 of file timedisc_base.f90.

◆ modified_euler

integer, parameter, public timedisc_base_mod::modified_euler = 1

Definition at line 178 of file timedisc_base.f90.

◆ rk_fehlberg

integer, parameter, public timedisc_base_mod::rk_fehlberg = 2

Definition at line 179 of file timedisc_base.f90.

◆ ssprk

integer, parameter, public timedisc_base_mod::ssprk = 5

Definition at line 182 of file timedisc_base.f90.

Data Types

Functions/Subroutines

Variables

Detailed Description

Function/Subroutine Documentation

◆ acceptsolution()

◆ adjusttimestep()

◆ calcbackgroundvelocity()

◆ calctimestep()

◆ checkdata()

◆ computeerror()

◆ computerhs()

◆ fargoadvectionx()

◆ fargoadvectiony()

◆ finalize_base()

◆ getcentrifugalvelocity()

◆ getcfl()

◆ getorder()

◆ inittimedisc()

◆ integrationstep()

◆ rejectsolution()

◆ setoutput()

Variable Documentation

◆ cash_karp

◆ check_all

◆ check_csound

◆ check_invalid

◆ check_nothing

◆ check_pmin

◆ check_rhomin

◆ check_tmin

◆ dormand_prince

◆ dtcause_cfl

◆ dtcause_erradj

◆ dtcause_fileio

◆ dtcause_smallerr

◆ fargo_method

◆ modified_euler

◆ rk_fehlberg

◆ ssprk