Dears,

I am a PhD student and the topic of my research is CFD of multiphase flows (dust - air mixtures) with chemical reactions. Because there seem not not be many of us in the world, I have a question: does anybody of you do the same? or perhaps not the same but something similar. I just would like to have a contact with anybody I do not know yet and who would like to exchange some information.

Best wishes, Pawel Kosinski pkos@helios.itc.pw.edu.pl

HI THere,

I simulate dust + gas (in 2D), in the Epstein regime, i.e. when the size of the (assumed spherically symmetric) dust particles is smaller than the mean free path in the gas. I also assumed that the density of the dust is much larger than the density of the gas. So the dust is just dragged by the flow, and the flow is not affected by the dust. For given value of the drag parameter the dust can concentrate in (anti)cyclonic vortices, etc.. I do solve for the Navier-Stokes Equations for the gas and I solve equations for individual dust particles (samples - a few 10,000 is enough). The forces acting on the dust particles are just the gravity, the drag and additional ones such centrifugal force and Coriolis force. I do not do any chemical reaction.

If this is related to what you are doing, then I'll be glad to try to help.

Patrick

I worked on hetrogenous reactions which is a reaction between solid and aques flow. Do you have any specific question ?. I will be gald to answer it.

Dear Mr. Patrick We have had some useful discussions earlier. As far as the particle effect on air flow. I found water droplet (around 300 micrometers)do affect the air flow ( 3 m/s, Re 70000). Infact the problem i face is when the air velocity is 1m/s. If i include the droplet momentum source in the air eqn. The solution was not converging. I used underrelaxation but no hope. O.K this was for counter-current flow. As far as the co-current flow, my simulation was fine with no problems and matches well with the exp.data any insight please regards rasuku

For sure your two-phase flow is in the Tailor regime, when the particles are actually macroscopic, rather than microscopic. Also they do affect the flow.

Concerning the convergence of your simulations, it is difficult to say anything since all this might well be due to the temporal integration scheme you are using, or the size of the time step..., etc.. or some other parameter of your problem. Check also that the drag parameter is a function of the difference of velocities between the particle and the local flow, so that when they are moving in opposite directions, you actually need to add their absolute values (it is proportional to the square of the relative velocity of the particle to the flow).

When particles and flow are all moving in the same direction then the relative velocity is smaller than in the case where they are moving in opposite directions. So your problem is related to the relative velcoity reaching numerically a critical values. You need to check how is the convergence (or divergence) when you change the time step size and other parameters of your problem in order to track the source of your numerical instability. This is true for any numerical problems or divergence.

PG

The Drag calculation etc. were right, also the reduction in time did'nt help. The problem, occures only when the droplet momentum source is taken into air (air V=1 m/s) eqn. (in other words when we consider effect of particle momentum on air flow - TWO-WAY coupling). At other higher air velocities and similar particle loading the soln. converged. Presuming that, for the cell size or Peclet number considered (especially at particle injection location) , the droplet source was too much to handle!! ( Will appreciate if some one can explain this phenomena better or even if they share their similar experience). I increased the grid size and found that the problem was well taken care off (ofcourse with a little trade off on grid indepenancy error). Any comments on my approach, is there any other thing which i have overlooked. Thankyou rasuku as i understand is when the momentum source of droplet is