Learning Forum

[cs_zy]

The channel is a 2D model.The inlet size is 750μm,314 grids, according to the standard particle release method, there will be 314 particle packages, if the particle diameter is 3μm, will the particle block the channel? Does it make physical sense to release particles in this way? Although I know that the DPM model does not consider the particle volume, I feel that the way of releasing particles is strange here. Should I change the way of releasing particles?

[DrAmine]

Particle is smaller than the channel but please ensure that the particle Size is smaller than the cell size. To account for flow blockage you might go fir Eulerian approaches ( or use bera feature in Fluent to include DPM blockage on the flow: simple/ weak way though)

[aitor.amatriain]

Apart from the comment of you should check the volume fraction of the flow. For values less than 0.1, the interactions between particles are not negligible and in that situation you should consider alternative approaches.

[cs_zy]

@Aitor Thanks.My volume fraction is less than 0.1, but my cell number is 314 at inlet. I can only chose standard particle release method because I use UDF to initial particle velocity. If the particle is 1 ╬╝m, small than the cell size. According to the standard particle release method,there will be 314 particle packages ,at least 314 particles, 314╬╝m. Actually 314 particles almost take part in 0.5 the width of channel which will have a great importance effect on the flow behavior in real life. So I mean in Fluent if a particle smaller than one cell but the total particle volume at releasing place take part in 0.5 the width of channel, dose this make sense? Actually if one cell at the inlet can represent less than one particle such as 0.1 particle, which can help to understand.
Thank you again @Aitor

[aitor.amatriain]

You can define an injection by means of an injection file so that you can control all of the parameters (number of particles, time step...)

[DrAmine]

Use file injection or udf to control number of streams. Or group. Or allow random injection at the surface and

[DrAmine]

Allow lower number of streams. Bear in mind more streams are required to get a better statistics