Discussion Closed This discussion was created more than 6 months ago and has been closed. To start a new discussion with a link back to this one, click here.

Modeling a Partially Full Mixer

Audrey Siebert-Timmer

Please login with a confirmed email address before reporting spam

Hello,

We want to model a front load washer partially filled with water as shown in the attached Cross Section 3 screen shot.

We formed our base model off of the schematic pump 1 example in the model library of COMSOL 3.4 (see attached mixer model screen shot). The problem with this model is that the mixer is entirely full of water, and the model we are trying to create (the front load washer) is only partially full of water.

Is it possible to model a partially full mixer?


3 Replies Last Post 12 gen 2012, 02:11 GMT-5
Ivar KJELBERG COMSOL Multiphysics(r) fan, retired, former "Senior Expert" at CSEM SA (CH)

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 18 dic 2011, 05:22 GMT-5
Hi

I would propose to llok in the direction aof an ALE section formthe partially filled water part (but do you need to model the air above ?) or perhaps try a level set method, check the doc, I must admit it's not something I have tried, but probably there are other out there with bettrer or more precise ideas (check also the sloshing tank model in the model library)

--
Good luck
Ivar
Hi I would propose to llok in the direction aof an ALE section formthe partially filled water part (but do you need to model the air above ?) or perhaps try a level set method, check the doc, I must admit it's not something I have tried, but probably there are other out there with bettrer or more precise ideas (check also the sloshing tank model in the model library) -- Good luck Ivar

Audrey Siebert-Timmer

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 12 gen 2012, 00:15 GMT-5
Thanks for your reply Ivar,

We looked at the sloshing tank example initially as we are not concerned with motion of the air in the system (we would assume that the water behaves as one cohesive unit). However, we ran into difficulties in modeling the water sloshing in a closed system. First, it was difficult to define the boundary of the system to be circular using tangential and normal coordinates. Second we found that if we increased the max displacement angle (phi) the sloshing tank example would not converge. As we want the fluid to rotate in the closed system this caused some concern.

Returning to the drawing board we looked at the Schematic Pump #2 example in the Chemical Engineering Module. This is simply exactly what we want to model but without and inlets and outlets and with the pump to be half full. However, I could not find any examples of mixing two fluids or water and air, hence, we are still stumped.

On a side note we tried to run the Schematic Pump #2 by just changing the impeller geometry and the model would not converge. The same boundary conditions are applied and the outer geometry remains relatively the same (the inlet and outlet is removed ... which converged with the original impeller) it is just the impeller has more blades and is larger as shown in the picture attached. Does geometry have that much of an effect on the model converging?

Thanks for your reply Ivar, We looked at the sloshing tank example initially as we are not concerned with motion of the air in the system (we would assume that the water behaves as one cohesive unit). However, we ran into difficulties in modeling the water sloshing in a closed system. First, it was difficult to define the boundary of the system to be circular using tangential and normal coordinates. Second we found that if we increased the max displacement angle (phi) the sloshing tank example would not converge. As we want the fluid to rotate in the closed system this caused some concern. Returning to the drawing board we looked at the Schematic Pump #2 example in the Chemical Engineering Module. This is simply exactly what we want to model but without and inlets and outlets and with the pump to be half full. However, I could not find any examples of mixing two fluids or water and air, hence, we are still stumped. On a side note we tried to run the Schematic Pump #2 by just changing the impeller geometry and the model would not converge. The same boundary conditions are applied and the outer geometry remains relatively the same (the inlet and outlet is removed ... which converged with the original impeller) it is just the impeller has more blades and is larger as shown in the picture attached. Does geometry have that much of an effect on the model converging?


Ivar KJELBERG COMSOL Multiphysics(r) fan, retired, former "Senior Expert" at CSEM SA (CH)

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 12 gen 2012, 02:11 GMT-5
Hi

I would suspect you have a mesh density issue, as you must have turbulent flow too. This type of modelling is not really my speciality, there are other more experienced users out here, I haven't done much (yet) in this domain.

Anyhow have you tried to get some help from "support" they could also guide you, or go to one of the COMSOL CFD courses and give them the problem to show as demo ;)

--
Good luck
Ivar
Hi I would suspect you have a mesh density issue, as you must have turbulent flow too. This type of modelling is not really my speciality, there are other more experienced users out here, I haven't done much (yet) in this domain. Anyhow have you tried to get some help from "support" they could also guide you, or go to one of the COMSOL CFD courses and give them the problem to show as demo ;) -- Good luck Ivar

Note that while COMSOL employees may participate in the discussion forum, COMSOL® software users who are on-subscription should submit their questions via the Support Center for a more comprehensive response from the Technical Support team.