Learning Forum

[bergl]

Hi,

i am modelling particles falling through a 3D-geometry with a gas flow from the bottom using DDPM (ANSYS 2021 R2). I want to remove the particles at a certain height using a UDF:

DEFINE_DPM_SCALAR_UPDATE(remove_particles, cell, thread, initialize, tp)
{

real y_lim_out=1.1;                
if (TP_POS(tp)[1]        Message("            PARTICLE ID %d REMOVED FROM DOMAIN\n", tp->part_id);
        tp->stream_index = -1;
        }
}

This works well when looking at the particle tracks (left): there are no particles below the specified height (red line). However, when looking at the contours of vol. fraction of the solid phase (right), it seems like FLUENT is still showing a vol. fraction >0 below the specified height.

According to the Theory Guide 2021R2 (14.5.23. Dense Discrete Phase Model), the solution for the solid phase, such as volume fraction or velocity field, is taken from the Lagrangian tracking solution. Why is there a solid vol. fraction at the very bottom where all particles have been removed?

Thanks a lot in advance,

Lukas

[Prashanth]

I think you are just removing the particles. You would need remove the particles well and dump/transfer the mass/sources when the logic is satisfied.

Instead of:

tp->stream_index = -1

Try:

TP_MASS(tp)=0.0;

TP_DIAM(tp)=0.0; 

 MARK_TP(tp,P_FL_REMOVED);

[bergl]

Hi,

Thank you very much for your input. I modified my UDF as suggested. 

I now use "MARK_TP(tp,P_FL_REMOVED)" which should do the same as "tp->stream_index = -1", right? Furthermore, i set the particle mass and diameter to 0 before the particle is removed.

However, the results are very similar as before.The particles are removed when looking at the particle tracks but the contour plot shows a high vol. fraction for the solid phase. In my case the particles should leave the geometry at the bottom. I tried to use the escape BC for the particles at the bottom (which is a velocity inlet with a rather small inlet velocity), however this lead to a bubbling fluidized bed that continously build up during the transient simulation (some particles where removed at the bottom, but not all of them). Therefore, I tried to remove the particles via a UDF a few centimeters above the bottom. However, now i get these strange results regarding the solid vol. fraction and furthermore, it seems to lead to a similar problem where a bubbling bed is forming while particles should just fall down due to gravity and be removed from the domain. Any more ideas what I am doing wrong?

I use these settings for the solid phase:

The solid-solid restitution coefficient is 0.8 and a specularity coefficient of 0.1 is used at the walls. The parcels are injected in the upper part using a surface injection while each parcel represents 500 particles (d=100 my, density=3000 kg/m3). The ratio of mesh-size to parcel-size should be OK.