[overture] Re: RigidBodyMotion: Light Particles
- From: Bill Henshaw <henshaw@xxxxxxxx>
- To: freelists <overture@xxxxxxxxxxxxx>
- Date: Fri, 02 Nov 2007 07:26:12 -0700
Hi Dominic,
It should be perfectly safe to increase the number of corrector steps
to any value. This will make the time-stepping more stable.
I will add a new option so that one can set the number of corrector
steps in the cmd file.
...Bill
#DOMINIC DENVER JOHN CHANDAR# wrote:
Hi Dr Henshaw,
I realize that there is the improved Euler's method you have
included in cg.v22 (Rigidbodymotion.C) . I havnt completely migrated to
v22 so i
use the new rigidbodymotion.C along with v.21. I find reasonable
comparison with experimental results in terms of average fall velocity
and angular velocity for falling objects when i use the improved eulers
method .... but i have to increase the number of corrector steps in
ims.C . ( I guess the default is 2 corrector steps). Does this have any
other consequences ? or is it correct to increase it blindly. I
understand that the relationship between forces and velocity is implicit
and we need more corrector steps, but i just want to make sure im on the
right track ...
btw.. in cgins too if i use the default number of corrector steps, it
blows up for low density bodies using improved Eulers method.
Regards,
Dominic
Other related posts:
