[HuifengGong]

Hi all,

I'm doing a simulation for a droplet laden jet flow using LES&DPM, and I'm quite confused about the Mean DPM Diameter distribution along axial direction, as shown in the attached image. There is a distinct drop of diameter not far from the nozzle exit (nozzle exit is located at z=0.05), which I think is unusual and not correct. Does anybody know what happened to my simulation?

My case is about a droplet laden jet from a nozzle tube (10mm in diameter), in which the droplets are carried through a nozzle with air, generating a turbulent jet flow when emanating from the nozzle tube. The continous phase was first simulated using LES, where a velocity UDF was specified at the nozzle exit according to the experimental data. After the continous phase was statistically convergent, the dispersed phase was added at the nozzle exit using DPM model, where an injection file was used to specify the droplet size & velocity distribution at the nozzle exit according to the experimental data. 361 positions of the nozzle exit were determined in the injection file (5circles with 72 positions in each circle, and 1 position at the center), and each position was specified a identical velocity and an droplet size distribution (12 different droplet size bins).

Thanks a lot!

Huifeng

[mome]

hi, not an expert on this, but a few ideas that may be helpful
- is that constantly happening through your timesteps?
- are there any other related quantities that do not fit your expectations, like Weber no, velocity or particle mass?
- is it a breakup thing? what breakup model are you using, is the diameter evolution consistent with it, so is the model adequate?
good luck

[HuifengGong]

Originally Posted by mome

hi, not an expert on this, but a few ideas that may be helpful
- is that constantly happening through your timesteps?
- are there any other related quantities that do not fit your expectations, like Weber no, velocity or particle mass?
- is it a breakup thing? what breakup model are you using, is the diameter evolution consistent with it, so is the model adequate?
good luck

Hi mome, Thanks for your reply.

This Mean DPM Diameter is a statistical result, which is averaged during a few flow times after this droplet-laden jet is statistically convergent. I examined the axial DPM diameter at each timestep after convergence, and this trend of sharp drop in diameter constantly happened. So, I'm afraid this is not random, and there must be certain mistakes in my simulations to generate the unusual results.

I just focus on the velocity and the droplet size distribution. The mean velocities (both axial and radial) agree with the experimental data well, while the droplet size distribution is strange at the centerline. It seems that the small droplets concerntrate at that sharp drop area, causing a distinct decrease in mean diameter.

This case is just a droplet-laden flow, and there are no breakup and atomization processes. I just define a droplet size distribution (12 size bins) at the nozzle exit.

Is it possible that the mistake lies in the boundary settings, especially the droplet injection settings? I think my continous phase simulation with LES is OK, the problem may be appeared after the droplets are added. Is there anything wrong with my injection file settings (361 positions at the nozzle exit with a droplet size distribution at each position)?