CFD Online Discussion Forums

CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   CFX (https://www.cfd-online.com/Forums/cfx/)
-   -   Reproduce COMSOL simulation in CFX (https://www.cfd-online.com/Forums/cfx/126251-reproduce-comsol-simulation-cfx.html)

Antanas November 12, 2013 07:25

Reproduce COMSOL simulation in CFX
 
Hello!

I try to reproduce COMSOL's Airlift Loop Reactor example model in CFX and cannot obtain similar results (gas volume fractions, pressure and so on). So I have a couple of questions, maybe someone could answer:

1. Which multiphase model in CFX corresponds better to COMSOL's Bubbly Flow models (algebraic slip or something)? The assumptions of Bubbly Flow model are:
a) gas density is negligible compared to the liquid density;
b) the motion of gas bubbles relative to the liquid is determined by a balance between viscous drag and pressure forces;
c) the two phases share the same pressure field.

2. By default in COMSOL it's assumed that gas volume fraction is less than 0.1, therefore continuity equation has form div(v_liquid)=0. Is there such option in CFX?

3. It seems that pressure variable (which is used in governing equations) in COMSOL includes hydrostatic contribution (if gravity is added), while in CFX hydrostatic pressure is included in Absolute Pressure but not just pressure (if buoyancy is activated). Is it so and may it be the cause of the problem?

4. How can I implement COMSOL's Gas Flux and Gas outlet boundary condition in CFX? Should I use boundary source and degassing/deposition respectively?

ghorrocks November 12, 2013 17:55

This COMSOL model sounds like a standard inhomogeneous eularian flow model in CFX. But the devil is in the detail - what drag model? And corrections for things like pressure? Additional bubble mass from liquid dragged with the bubble?

You can only compare solutions between software after you have done a verification/validation exercise on both codes. It is meaningless to compare inaccurate simulations. Some tips are here: http://www.cfd-online.com/Wiki/Ansys..._inaccurate.3F

I have no idea what the gas flux and gas outlet boundary in COMSOL are. CFX has a degassing boundary, this may be similar.

I have

Antanas November 13, 2013 03:17

Quote:

Originally Posted by ghorrocks (Post 461713)
This COMSOL model sounds like a standard inhomogeneous eularian flow model in CFX. But the devil is in the detail - what drag model? And corrections for things like pressure? Additional bubble mass from liquid dragged with the bubble?

You can only compare solutions between software after you have done a verification/validation exercise on both codes. It is meaningless to compare inaccurate simulations. Some tips are here: http://www.cfd-online.com/Wiki/Ansys..._inaccurate.3F

I have no idea what the gas flux and gas outlet boundary in COMSOL are. CFX has a degassing boundary, this may be similar.

I have

The drag model used in COMSOL is Hadamard-Rybczynski. This is not present in CFX. Gas flux BC is -n(gas_vf*gas_density*gas_velocity)=Gas Flux, where gas_velocity = liquid_velocity + slip_velocity + drift_velocity. Gas outlet BC means no condition for gas at boundary and if slip condition is used for liquid at this boundary it seems that it similar to degassing in CFX.

ghorrocks November 13, 2013 06:52

You can specify your own drag coefficient either by CEL or a user fortran routine. So you can implement the model yourself if you wish.


All times are GMT -4. The time now is 08:51.