Turbulent mixing of two liquids

zimon · June 20, 2016, 15:45

Hello dear FOAMers, I'm a student who has just started using openFOAM (and is new to this forum). I'd like to simulate the mixing of two incompressible fluids in a 3D mixer, into whom the fluids are continuously injected. The solver I'm looking for should be able to handle turbulence, and i'm mostly interested in a steady state solution (i.e. looking at the system after everything has stabilized). After having looked around I was thinking of using twoLiquidMixingFoam, but I know very little about it. Do you think that this could be a good choice? Secondly, is there a way to tell the solver to stop the simulation after a "steady" situation has been reached?
Thanks in advance!
Zimon

Coris · June 21, 2016, 15:36

Are the fluids miscible or not?

In the case they are, I am concerned about interpolated properties (like 50 %) -- what if eg viscosity is non-linearly dependent on concentration?

In case they aren't, I am concerned about interphase drag.

So no idea how this works in OF, documentation is really limited :/

zimon · June 21, 2016, 15:55

Hello Coris, thanks for your reply. Fluids should be miscible. I'm not truly interested in a specific set of properties for the fluids (they could be the same, like both water), but only to track their motion in the mixer. Probably it could just suffice for my purposes to use simpleFoam to determine if there are vortices but it would have been cool to see the different liquid phases moving and merging.
Unfortunately the only example in the tutorial for twoLiquidMixingFoam (lockExchange) doesn't use turbulence nor a continuous supply of the fluids but I guess the former can be switched on and the latter can be modeled by imposing the appropriate boundary conditions at the inlets/outlet. More complicate could be obtaining a steady state solution since it seems to be a transient solver (differently from simpleFoam). I wonder if interMixingFoam could work as well: in that case fluid properties should be automatically averaged in each cell (I think the same happens with twoLiquidMixingFoam but I'm not sure).
Thanks again for your interest!

Coris · June 21, 2016, 16:13

You can activate turbulence with turbulenceProperties: https://github.com/OpenFOAM/OpenFOAM...enceProperties

BC's have to be changed probably yes

interMixingFoam looks overkill..

Another hint: run the tutorial case first and see what the result is in paraView

Coris · June 21, 2016, 16:16

Ah and about the steady state: RA(N)S treats the flow in some pseudo-steady state, so calculating transient is equivalent to the false time stepping approach.

(You need URANS for an unsteady approach to RANS.)

zimon · June 21, 2016, 16:47

Hi Coris, thanks again for your quick reply. I' ve already run the tutorial case but sadly it only displays the mixing of two fluids that are in contact but with no pumping, in a "laminar" case. The results can be seen in this video I' ve found on Youtube: https://www.youtube.com/watch?v=XxWXu3PCDXY
Just one (more) question: why do you say that interMixingFoam is an overkill? I know it's devised for 3 fluids but I guess you can use it with just 2.
Thanks a lot for the other advices (and for the patience with a total noob

)!

Coris · June 22, 2016, 06:39

A regular inlet and outlet bc is straightforward, but I don't know how to specify which fluid enters at which point..

You can easily run into trouble trying to use something in OF that was not designed for it..

June 20, 2016, 15:45	Turbulent mixing of two liquids	#1
zimon New Member Join Date: Jun 2016 Posts: 3 Rep Power: 9	Hello dear FOAMers, I'm a student who has just started using openFOAM (and is new to this forum). I'd like to simulate the mixing of two incompressible fluids in a 3D mixer, into whom the fluids are continuously injected. The solver I'm looking for should be able to handle turbulence, and i'm mostly interested in a steady state solution (i.e. looking at the system after everything has stabilized). After having looked around I was thinking of using twoLiquidMixingFoam, but I know very little about it. Do you think that this could be a good choice? Secondly, is there a way to tell the solver to stop the simulation after a "steady" situation has been reached? Thanks in advance! Zimon

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Turbulent eddy diffusivity for mixing??	Ray	Main CFD Forum	3	November 5, 2020 11:57
Problem with divergence	TDK	FLUENT	13	December 14, 2018 06:00
Mixing two immiscible liquids in Kenics static mixers	longbk	Main CFD Forum	1	May 26, 2017 03:27
Mixing of miscible liquids	Farshid	FLUENT	2	May 8, 2003 16:22
turbulent mixing of two fluids	ulrich bieder	Main CFD Forum	2	October 9, 1999 16:15

June 21, 2016, 15:36		#2
Coris New Member Joris C. Join Date: Jan 2013 Posts: 29 Rep Power: 13	Are the fluids miscible or not? In the case they are, I am concerned about interpolated properties (like 50 %) -- what if eg viscosity is non-linearly dependent on concentration? In case they aren't, I am concerned about interphase drag. So no idea how this works in OF, documentation is really limited :/

June 21, 2016, 15:55		#3
zimon New Member Join Date: Jun 2016 Posts: 3 Rep Power: 9	Hello Coris, thanks for your reply. Fluids should be miscible. I'm not truly interested in a specific set of properties for the fluids (they could be the same, like both water), but only to track their motion in the mixer. Probably it could just suffice for my purposes to use simpleFoam to determine if there are vortices but it would have been cool to see the different liquid phases moving and merging. Unfortunately the only example in the tutorial for twoLiquidMixingFoam (lockExchange) doesn't use turbulence nor a continuous supply of the fluids but I guess the former can be switched on and the latter can be modeled by imposing the appropriate boundary conditions at the inlets/outlet. More complicate could be obtaining a steady state solution since it seems to be a transient solver (differently from simpleFoam). I wonder if interMixingFoam could work as well: in that case fluid properties should be automatically averaged in each cell (I think the same happens with twoLiquidMixingFoam but I'm not sure). Thanks again for your interest!

June 21, 2016, 16:13		#4
Coris New Member Joris C. Join Date: Jan 2013 Posts: 29 Rep Power: 13	You can activate turbulence with turbulenceProperties: https://github.com/OpenFOAM/OpenFOAM...enceProperties BC's have to be changed probably yes interMixingFoam looks overkill.. Another hint: run the tutorial case first and see what the result is in paraView

June 21, 2016, 16:16		#5
Coris New Member Joris C. Join Date: Jan 2013 Posts: 29 Rep Power: 13	Ah and about the steady state: RA(N)S treats the flow in some pseudo-steady state, so calculating transient is equivalent to the false time stepping approach. (You need URANS for an unsteady approach to RANS.)

June 21, 2016, 16:47		#6
zimon New Member Join Date: Jun 2016 Posts: 3 Rep Power: 9	Hi Coris, thanks again for your quick reply. I' ve already run the tutorial case but sadly it only displays the mixing of two fluids that are in contact but with no pumping, in a "laminar" case. The results can be seen in this video I' ve found on Youtube: https://www.youtube.com/watch?v=XxWXu3PCDXY Just one (more) question: why do you say that interMixingFoam is an overkill? I know it's devised for 3 fluids but I guess you can use it with just 2. Thanks a lot for the other advices (and for the patience with a total noob )!

June 22, 2016, 06:39		#7
Coris New Member Joris C. Join Date: Jan 2013 Posts: 29 Rep Power: 13	A regular inlet and outlet bc is straightforward, but I don't know how to specify which fluid enters at which point.. You can easily run into trouble trying to use something in OF that was not designed for it..