CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   FLUENT (http://www.cfd-online.com/Forums/fluent/)
-   -   3 phase flow (wall boundary condition) (http://www.cfd-online.com/Forums/fluent/93097-3-phase-flow-wall-boundary-condition.html)

Mohsin October 5, 2011 00:25

3 phase flow (wall boundary condition)
 
Hello
I am simulating a spray dryer. The liquid (solution of N-Methyl-2-pyrrolidone (C5H9NO) and Coal) will be sprayed into the dryer. Hot N2 will also enter the domain in order to dry the solution and to provide swirl motion for proper mixing and heating.

Upon entering the domain, the solution heats up and N-Methyl-2-pyrrolidone (C5H9NO) should evaporate from the solution leaving behind solid coal particles (which will be collected at the bottom) whereas the N-Methyl-2-pyrrolidone (C5H9NO) vapors will leave the domain.

Could anyone please tell me how do I approach this problem? I am using DPM model in FLUENT and N2 Flow field have been simulated. My querries are:
1. In species transport how do i define the solution (1 phase, liquid) of N-Methyl-2-pyrrolidone (C5H9NO) and Coal?
2. What will be the wall boundary condition in this case? It could be “trapped” but the solution contains solid coal particles which should “reflect” also from the wall. How can I manage to use both of them? Do i need UDF for this? Or I just use "reflect" with Coefficient of restitution equal to 0.9 (as the liquid from the solution will have been evaporated until it strikes the wall).


I’ll be grateful for any suggestions..
Regards
Mohsin

mauricio October 6, 2011 07:22

spray
 
Hello Mohsin, i don't use Fluent yet so i don't know the names of the models, but assuming there's something equal to what i've found in cfx:

What i'd do in cfx is to create a mixture of solid coal and the N2, a gas fase of N2, and a binary mixture of liquid N2|gas N2 that represents the relationship of evaporation between them ( You'd have to introduce details to calculate the saturation pressure, in Cfx that can be handle by supplying antoine coefficients). I'm guessing the'res air in the tank as well besides the gas N2? Since the main interest is the solid particle trapping/interaction you might end up assuming it's just full of gas N2. Do mind the material properties you're using.

Not exactly sure how you'd go by it in fluent; there's a tutorial on a spray dryer on cfx that has solid particles and liquid around them.

About the trapping of particles, well... Do you have a way to determine when a particle is going to be trapped or going to reflect? for example a statistical probability or a relationship to its position or velocity? you could point the restitution coefficient to an if function; check with an expression what would be the behaviour of that particle and the set the appropiate restitution coefficient on that wall. I'm guessing that in fluent you'd use UDF for that if function, but in cfx there's an If function available without using user functions :P

Hope this helps a least a bit in this somewhat complex but interesting model :)


All times are GMT -4. The time now is 09:45.