CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   CFX (http://www.cfd-online.com/Forums/cfx/)
-   -   multiphase Eulerian particles (http://www.cfd-online.com/Forums/cfx/26418-multiphase-eulerian-particles.html)

 Zbynek September 22, 2008 09:50

multiphase Eulerian particles

Dear all, I'm trying to simulate water droplets moving in air using Eulerian - Eulerian multiphase model of CFX 11. As a result I need to obtain water flux impacting on a wall immersed in air/water flow. I have defined air as a continuous fluid and water droplets as dispersed fluid particles. Solution converges well and result looks good inside the water/air flow field. The problem is on the wall surface. Water droplets are trapped on the surface and form some type of film with 100% water fraction which makes impossible to measure water flux impacting on the surface. As the simplest and physically correct solution of this problem seems to me to define different BC for each phase (wall for air and outlet for droplets). There is a DESGASING OUTLET BC option.It is nearly what I need but it defines free-slip wall for continuous phase which is unacceptable for my case. I have tried to define negative source on the wall to suck incoming water from the surface but to do it I need normal velocity of droplets impacting on surface. Since velocity on surface is forced to be zero or tangential (no-slip/free slip), I'm not able to calculate the flux.

Does anybody know how to define different BC for each phase e.g. no-slip wall for continuous phase and outlet for dispersed droplets?

or

Does anybody have idea how to measure flux of water droplets impacting on a surface?

thanks for your time, Regards Zbynek

 Bart September 22, 2008 11:39

Re: multiphase Eulerian particles

If you do lagrangian particle tracking, this will give you the flux of particles on a surface.

 Juan Martin September 23, 2008 16:51

Re: multiphase Eulerian particles

Hello. I think that measuring the flow in a parallel plane close to the wall would be enough to get the flux of water reaching the wall. Please tell me what do you think. I'm working also with multiphase. Bye.

 Bart September 24, 2008 02:59

Re: multiphase Eulerian particles

With Eulerian multiphase you are modelling hold-up. In contrast to Lagrangian modelling, the particles do have a volume. When the water accumulates near the wall and the water-film is thicker than your distance to your monitor plane you will 'measure' no flux. Also due to turbulent dispersion, small droplets might have problems getting to the wall. So they do not actually deposit on the wall but do get close. If the volume fraction of your water is low (<3-5%) I would seriously consider doing lagrangian particle tracking.

Bart

 Zbynek September 24, 2008 11:04

Re: multiphase Eulerian particles

Hi, I deal with aircraft moving in clouds. Droplets in clauds are usually very small 10-100 microns. Currently we use our in-house Lagrangian modul for the job, but it is not very efficient approach. If there is a big 3d domain and you are looking for water mass impacting on relatively small surface e.g. an antenna which is surrounded by complex flow it is quite difficult task using particle tracking. We suppose that Eulerian approach is more modern and efficient way how to obtain distribution of water impacting on surface. Moreover it gives you more useful information than particle tracking. That is why we want to move to Eulerian method. Zbynek

 Bart September 24, 2008 11:21

Re: multiphase Eulerian particles

Ok, makes sense I guess. But now you need to store and calculate two velocity fields with locally a high resolution. If you are looking for a small area of interest for deposition in a large 3D domain, you can also limit the area where you release your particles (in the Lagrangian method that is). Particles in the outskirts of your domain which are not likely to deposit are of no interest anyway.

Succes, Bart

 All times are GMT -4. The time now is 06:26.