Discussion Groups

Simulation CFD

Reply
New Member
AndrewWatt444
Posts: 1
Registered: ‎03-26-2013

Massed Particle Trace Time Step Size

306 Views, 1 Replies
03-26-2013 09:35 AM

I'm simulating a cyclone separator and adding various massed particles to analyze the filtration.  However, I can't find a good explanation for the "Time step size (sec):" option in the mass dialog box.  I notice that changing it greatly changes the paths the particles take.  For a steady-state problem like mine, what does this variable do, and how should I set it?

Please use plain text.
*Expert Elite*
OmkarJ
Posts: 437
Registered: ‎10-02-2012

Re: Massed Particle Trace Time Step Size

03-27-2013 02:07 AM in reply to: AndrewWatt444

Here is a guess. For typical massless traces, the Langrangian equation is solveed only for particle position. It can be done using the global velocity field that is already calculated during solution. In case of massed particle traces, the velocities of particle at any point may/will be different than the local velocities of fluid. Hence the equation for local particle velocity is solved, based on Newton's second law, i.e. Rate of change of momentum=Buoyant and drag forces etc.

 

Now, I pressume that the particle velocity equations may have been solved explicitly, with timestep size we specify in that box. Essentially, since it's explicit methodology, the value of timestep will have an influence on accurcy of velocity. Typically, the timesteps should be chosen such that particle will cross one cell distance during the timestep, a concept perhaps equivalent to particle cell courant number.

 

Appropriate timestep can be: Typical cell size/mean velocity at the cell. 

 

Though, this all is guesswork. More experienced folks may have more thoughts to offer.

 

OJ

Please use plain text.