[A_Keen]

Hello everyone,

I'm somewhat new to FLUENT and have been using a steady state, massless dpm group injection in a 2D axisymmetric, laminar (Re < 7 in my whole domain), steady simulation. My issue is that my resultant pathlines overlap in the "Poiseuille flow" region, and I don't understand why they seem to bunch up.

The left borders are inlets, the right an outlet, the bottom is the axisymmetric boundary and the rest are walls:

The underlying mesh is largely quad with inflation at the walls. At the inlet (left side) there is ~55 cells in the y-direction.


I have exported the position information from the simulation for post-processing in Matlab. I have tried injecting many different numbers of particles (15 - 500) along the line x=.002 m. When I plot the 300 pathline case, it looks like the following example:

Zooming in on a smaller region looks like this:



Evidently, the pathlines start in independent regions but they end up with a spacing smaller than a micron.

Looking at the number of discrete regions where the pathlines end, it is greater than the number of cells across, so it doesn't seem to me like the limitation is directly mesh-related. From my understanding, I thought interpolation was done between the different cell points to propagate the particle, so I didn't expect these groups to bunch up so much. Also, when I look at streamlines in CFD-Post this behaviour does not occur. The streamlines remain straight in this region without "bunching up", as I expected. From my understanding, a steady laminar simulation of massless DPM should be equivalent to streamlines, so why is this not the case? Have I misunderstood?

Question summary: How does the massless DPM work such that it is causing this phenomena?

Thanks for the help