Hello there,
i'm a computer science student, and i'm programming a simulator (in
Matlab) using the HS (hard sphere) model... but i've encountered a
"little" problem...
The problem poses when 3 particles collide at the same time,
example: system with particles A,B,Z
A collides with Z
B collides with Z
(A does *not* collide with B, although the solution i'm looking for
should also handle that situation)
So we have 2 collisions at the exact same instant... when this happens
the behauviour gets erratic!
What i would like to know, is the set of equations on how to deal with
this.
So far, i've been doing some vector math illustrated at Fosdick's book:
An Introduction to High-Performance Scientific Computing, and this
deals with only 1 collision at a time.
I was able to get 2 behaviors, depending on how i handle the
simultaneus collision(if i do it "in loco"; or if i use a temporary
variable), but both of them are incorrect.
In case (1): the vector's directions stay correct, but the linear
momentum of the system changes.
In case (2): the opposite happens.
I've scourged the web for info on this... but no luck...
The papers/articles/reports/etc i've found just ignore these
situations.
And i've also found a reply on this newsgroup explaining that multiple
collisions only happen at peculiar places, like the sun's core.
But i would still like to model this situations, because in case (2),
where the momentum stays constant (and this is the most important
aspect of the simulation), there are situations that one particle
overlaps the other one (i represent them like billard balls, so they
have a radius)... and this isn't good at all.
So what i ask is for someone, to mention/explain the equations for
solving multiple collisions in the HS model; and/or point me out some
websites/books (hopefully one of the books will be in the univ library)
I'm sorry if this is a "minor" question compared with the other
advanced subjects... but OTOH, if it's a simple question, the
explanation/answer should be simple as well (i hope)
Thank you very much for you time :-)
.
|
