CFD Online Discussion Forums

CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   CFX (https://www.cfd-online.com/Forums/cfx/)
-   -   Particle Collection Efficiency (https://www.cfd-online.com/Forums/cfx/73366-particle-collection-efficiency.html)

rval March 5, 2010 15:27

Particle Collection Efficiency
 
I want to determine particle (say dust) collection efficiency in a fibrous filter media. I have previously run particle tracking simulations using the lagrangian approach with one-way coupling between particle and fluid (air) and coefficient of restitution equal to 1. However, based on the theories of particle deposition, I want to include drag force as well as brownian diffusion. I am therefore debating between lagrangian and euler-euler approach. I have not used the euler-euler approach before, so any pointers will be immensely helpful.

Thanks

ghorrocks March 6, 2010 05:58

To confirm - you want to have a filter which removes particles from the air, the particles have a Brownian motion applied to them and as the particles accumulate in the filter you want to increase the flow resistance of the filter?

rval March 6, 2010 23:36

Thanks Glenn.

Yes, you have summarized my problem well. I however, want to start with a steady state simulation and then move on to the more complicated problem of increasing flow resistance with increasing particle accumulation.

I read your comments in the thread "accumulating dust on a filter cartridge" and I did understand the general idea about running a series of steady state simulation.

My first hurdle is to run a steady state simulation.

ghorrocks March 7, 2010 05:52

Yes, that thread is my recommendations. Be aware that if you put random Brownian motion on the particles it will be difficult to achieve steady state convergence, for exactly the same reasons as turbulence dispersion.

Just a thought - if your particles are so small that Brownian motion is important then won't the particles just follow the streamlines, except for a small amount of dispersion from the Brownian motion? In that case a Eularian approach would make sense and you model the Brownian motion with a diffusivity. Whether this approach is valid will depend on many issues - I won't list them as hopefully you know when a Eularian approach makes sense.


All times are GMT -4. The time now is 19:10.