Re: Simulating a partial ring of particles using periodic boundary conditions

From: Pfenniger Daniel <>
Date: Mon, 28 Jan 2013 12:36:55 +0100

Hi Phil,

What you want to do is named "shearing box approximation".
As the referential is rotating the approximation must
include the Coriolis and centrifugal forces. Further,
the shearing due to differential rotation must be included,
unless solid rotation holds.
To use perdiodic BC and FFT's one has to consider an *afine*
periodic box with radial axes inclined wrt to radius changing
in time from -45 deg to +45 deg (see Huber & P 2001,
A&A 374, 465 for an implementation).

I don't think Gadget 2 has these features built-in :(


Phil Sutton wrote:
> Hi all,
> I am currently and have been modelling a complete ring of particles
> using vacuum boundary conditions with very good success. However I would
> like to investigate some long term structures and features within the
> ring and would therefore like to reduce the overall calculations down by
> reducing particle numbers. For example only model ľ of the ring and
> concentrate on certain features. I have not used periodic boundary
> conditions before but I canít see how place a box round the particles
> would work in my situation. Ideally I would have liked a box that co
> rotated with the particles, and I assume that GADGET2 uses a stationary
> box? Is it possible to use GADGET2 to simulate only a small section of a
> rotating ring? Would particles leaving the box on one side have the
> correct direction and position on the other side again? Also if
> particles did leave the box on one side and started again of the other
> side what would happen to a moon that was orbiting with the ring. For
> example ring particles are ok to start again on the other side, but a
> moon should not come around again until its faster or slower orbit
> allowed to meet back up again with the same part of the ring. So
> therefore would you need to treat particles differently when
> reintroducing them back in again?
> Thanks
> Phil
Received on 2013-01-28 12:39:02

This archive was generated by hypermail 2.3.0 : 2023-01-10 10:01:31 CET