$`d`$ is the distance from the disc element to the planet, and $`r_{\rm sm}`$ the
$`r_{\rm sm}`$ is called the
smoothing length:
smoothing length:
$`r_{\rm sm} = \epsilon R_H`$, where $`\epsilon`$ is a smoothing parameter and
$`r_{\rm sm} = \epsilon R_H`$, where $`\epsilon`$ is the smoothing parameter and
$`R_H`$ is the Hill radius of the planet.
$`R_H`$ is the Hill radius of the planet.
To select this choice for a simulation in the config file:
To select this choice for a simulation in the config file:
...
@@ -38,10 +37,11 @@ To select this choice for a simulation in the config file:
...
@@ -38,10 +37,11 @@ To select this choice for a simulation in the config file:
width="320" height="240">
width="320" height="240">
The plot shows the effect of smoothing the potential for various values of the smoothing parameter. In the 3 dimensional case we smooth the potential in order to avoid the singularity in the case
The plot shows the effect of smoothing the potential for various values of the smoothing parameter. In the 3 dimensional case we smooth the potential in order to avoid the singularity at
$`d -> 0`$. The choice of the smoothing length depends mainly on the resolution: as a rule of the thumb the minimum smoothing length should correspond to the size of about 4 grid cells.
$`d \rightarrow 0`$.
The choice of the smoothing length depends mainly on the resolution: as a rule of the thumb the minimum smoothing length should correspond to the size of about 4 grid cells.
@@ -59,26 +59,27 @@ To select this choice for a simulation in the config file:
...
@@ -59,26 +59,27 @@ To select this choice for a simulation in the config file:
## The potential due to indirect terms
## The potential due to indirect terms
We recall that we use a non inertial frame : the origin of the
We recall that we use a non inertial frame : the origin of the
system is centered on the primary body which is
system is centered on the primary body which is accelerated by the planet(s) and by the disk.
accelerated by the planet(s) and by the disk:
Therefore, we have to consider two indirect terms in the potential accounting for the acceleration of the primary induced respectively by the planet(s) and the disk:
with $`\vec r_p`$ and $`\vec r`$ indicating the distance between the
with $`\vec r_p`$ and $`\vec r`$ indicating the distance between the
planet and the star and a grid cell and the star
planet and the star and a disk element and the star. The second term is:
$`\tilde \Phi_d = \vec a \cdot \vec r `$
$`\tilde \Phi_d = \vec a \cdot \vec r `$
where $`\vec a`$ is the acceleration of the primary due to the whole disk.
where $`\vec a`$ is the acceleration of the primary due to the whole disk.
When doing simulations in production mode such terms must be included in the computations. However, it may be useful to control the contribution from indirect terms. At this purpose the indirect terms may be de activated in the config file with the flag IndirectForces in the Referential group :
When doing simulations in production mode such terms must be included in the computations. However, it may be useful to control the contribution from indirect terms. At this purpose the indirect terms may be de activated in the config file with the flag ``IndirectForces`` in the Referential group :