CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   FLOW-3D (http://www.cfd-online.com/Forums/flow-3d/)
-   -   Air Entrainment Model (http://www.cfd-online.com/Forums/flow-3d/64559-air-entrainment-model.html)

atulverma May 14, 2009 09:18

Air Entrainment Model
 
Hi Everybody!
I read the technical note by "Hirt" on Air Entrainment it says the modeling must be done with turbulence.I have laminar filling(single fluid model) of the die but the gates are so designed that it sure to get the fluid overturning and thus get air entrainment.Accidentally before reading the note I switched on the model with the flow being laminar and got some results,I wonder what does that represent and I could see the voids but they were not represented as air entrained(I did put some value of surface tension),I understand that Flow3d uses the relation given in the note,but are voids not the air inside the fluid.Thus the representation is not clear and the limitations of the model should be put down clearly.
Ideally there should be no air entrainment during laminar filling but in my case it visible that air would be trapped inside and it is seen as void(complement of fluid fraction).My question is can I model air entrainment for the laminar flow I have mentioned.
Thanks
Atul

melissa May 14, 2009 12:41

Hi~
So the flow through the gates won't be turbulent? Are you modifying the effective viscosity to account for this? What is your reasoning for using laminar flow?
MC:)

atulverma May 15, 2009 06:44

hi melissa!
 
Hi melissa!
No, the flow through the gates are not turbulent the Reynolds no. is below transition.Overturning is little ahead where the flow goes down and due to certain velocity it splashes.Thus I switched on the model without knowing that it can be used only with the turbulent flow.
The effective velocity(no modification is done) does not change much though I have switched on the solidification model and up to coherency point fs=0.23 viscosity is a function of solid fraction but not much solidification is observed.
The reasoning for using the laminar flow is to get minimum air entrapment(ideally zero at the surface) as everything as continuum says will be in layer so no surface disturbances and thus no air entrainment this is used for low velocity filling of the die in squeeze casting.
Thanks
Atul

stefmascio May 15, 2009 07:48

since version 9.2 the "Air Entrainment" model was extended to work also with laminar flows, not only turbulent, that's why it worked and you got some results.
But are you looking for "small" air bubbles (smaller than the cell size)? Or big bubbles (bigger than the cell size)?
In the second case you will need to switch on the bubble model, and not the air entrainment one.

atulverma May 15, 2009 15:19

Hi stefmascio!
 
Thanks a relief! that Flow3d can model air entrainment in laminar flow.
I do not know the physics how the entrainment can be modeled,so can u please tell me some references or the formula which Flow3d uses for modeling the air entrainment in laminar flow,I think the manual or help does not contain this particular description.
Some basic questions.I am giving a surface tension value in the model and air density.
In the single fluid model when the free surfaces meet they generally capture void region in between them which represents nothing but the constant pressure zone,but this is not seen as air entrainment but seen only as complement to fluid fraction that is void,and void is not a conserved quantity or are they...? so what happens to them,if a void gets captured in the fluid and one is not using the bubble model to incorporate this?
Also once the air is entrained does it get distributed in the fluid or it remains on the surface,since i have not used a variable density model I do not think the effect of buoyancy is there(But if the density in the cell is of fluid +void,then it is automatically there,which should be)?
You have mentioned about bubbles,I am interested in getting the air entrainment during filling,does it require modeling bubble because I want look at the macro level so that at any time step I should get a clear picture of the entrained air and the region where it is.Of course I do not know the size of the bubble but suppose it is not small enough to be neglected then and I do not want to know the exact dynamics which way I should go...?
Thanks
Atul

melissa May 18, 2009 17:05

Hi~
There is information from the manual about air entrainment and also a technical note on our user's website>Resources>Technotes.
MC

stefmascio May 19, 2009 05:17

hello Atul,
if you don't use the bubble model, neither the air entreinment one, the air is treated always with a constant pressure (no matter if its volume decrease) so basically it is not conserved.
If you use the bubble model you can have bubbles shown as "void" regions, with their shapes and their correct dynamic (and a pressure evolving).
If you use the air entrainment model, the air will be modeled as a scalar quantity dispersed in the fluid, so a color, and it works well whenever your bubbles are quite small, like a foam.

Which model to use and which one not is up to you. Depending on the physics you are observing you can know if bubbles or small air entrained are relevant or not in your general dynamics, and on the output you want to observe. If a certain physics doesn't affect much the quantity you want to observe, then you can keep it switched off.


All times are GMT -4. The time now is 12:05.