Multiphase Flow Simulation using VOF Model in ANSYS Fluent | 09 | Implementing the CFD Basics

[@bassiakosjr]

Hello there! Does anyone know how do we proceed after we built the mesh to actually solving the case? Like other below, I'm facing some trouble seeing the "part-air" and "part-water" in the cell boundary conditions. Unlike in previous tutorial, there was no explicit definition of named selections therein. Shouldn't we introduce named selections in order to see them back in the cell boundary conditions and boundary conditions of the solver. Thanks

[@amanbakshi7698]

Can you show me the simulation for the case where the droplet(half droplet) is attached to the base of a 2d horizontal rectangular channel with a parabolic velocity profile at the inlet using the same multiphase model

[@subratadutta7710]

Sir can you please tell that which model should we use to simulate a rotary atomizer? like VOF for the jet incoming and dpm for the atomization outside of the atomizer?

[@ciafaler]

DId you use boolean operation to substract droplet from air? Or they are overlapped?

[@RicardoPorteladaSilva]

Hi Tanmay Agrawal, your videos are very helpful! Thank you!.

[@ahsaniqbal1356]

is this possible to find the attachment of particles to the bubble?I am actually new to modelling.

[@ecemay5540]

Hello! I dont have 2 cell zones so I cant define water volume fraction as 1 because I can click only interior surface zones, not water part zone. What can I do to fix my problem.

[@subratadutta7710]

I also have the same problem, did you get the solution?

[@deepaksteaphen8030]

Iam.using vof for air water multiphase flow and i have prblm with pressure drop in my case constant pressure drop .how can in maintain constant pressure drop . Vof is suitable for this case pls reply

[@ivannepomnyashchikh7945]

Also, for the boundary conditions you need to choose no-slip boundary condition at the surface of the droplet. Otherwise, ANSYS will have to use something by default - and personally I have no idea what the defaults are. Apart from that, you need to specify boundary conditions at all the boundaries of your domain - usually we specify atmospheric pressure. This is called far field pressure boundaries. Gravity does not make the part of the boundary conditions. It is just one of the forces acting on the droplet and hence defines the equations we are going to solve. But equations themselves are useless without boundary conditions. Equations - PDEs in our case - dictate the law according to which boundary conditions propagate in the domain. Therefore, boundary conditions are crucial. Moreover, it seems very strange to me that your droplet has such a pear-like shape and that the color of the volume fraction inside the droplet is not uniform. The latter indicates that you have air in your water droplet which can't seem to be physical. To say nothing that analysis of the size of the flow domain (may be you could've chosen a smaller one and reduced the simulation time) and numerical uncertainties is necessary.

[@arunkumar1576]

How you obtained those 2 parts and 2 bodies?? I am unable to get it. I dont know how you got that geometry. Kindly help me get this geometry.

[@arunkumar1576]

Ok finally I got the geometry. If we don't select 'add frozen' for the first surface then we won't be able to get 2 parts 2 bodies.

[@vinayakbadiger5502]

hello sir, I am new to cfd but able to have result of mixing tank with impellers with 1 fluid, but now I want to mix water and air in a tank.so how can I do it pls help me. thank you

[@GinnGerr132]

Great video. I wonder which model should be used if the droplet is micron sized and the number concentration is large. Does the mixture model work for that？ Thanks！

[@TanmayAgrawal7]

Not sure actually! Perhaps try the DPM?

[@GinnGerr132]

​@@TanmayAgrawal7 Yeah, DPM should be accurate, but is more costly.

[@pulsedflowsolutions9058]

To learn how to simulate a 3D water droplet in Ansys Fluent, click on the following link: kzbin.info/www/bejne/g3-Vq4Cle5iBr6c

[@raju891]

Sir i am facing problem in 3D VOF method sir i created two separate volume water drop in side the outside volume. In between wall has created ? so i cant able to run simulation? please tell me solution for this

[@anshumansinha5874]

How's the cell zone showing 2 parts? Didn't you make 1 part in your geometry?

[@AbdulrahmanElmen]

Hello sir, I am trying to simulate dropley evaporation using vof, please tell me where to contact you to hopefully get your help Thank you

[@Aao_Baat_Krenge]

I have one question what if i create a rectangle and then split the circle out of it using tool , is it coreect

[@TanmayAgrawal7]

Akash Gangrade Hi Akash. There are definitely multiple ways to create the geometry. In the end, you want two bodies. In case if you cut a circle out of the rectangle, then it might be a bit tricky to move it (to simulate the fall).

[@TanmayAgrawal7]

Yes that's definitely a smarter approach. In this, I patched the entire geometry rather than defining a region using adapt as it's easier to do it.

[@anshumansinha5874]

Shouldn't this be a case of dynamic mesh! Since you've defined fine meshing for a circular region while meshing. But while running this transient simulation the water will fall down with time, So shouldn't the mesh also move with that second phase ( here, water)?

Good tutorial, thanks

[@ankiitbhu]

Hello I have not understood how you assigned the water to 'circle' . In your video there were already air , water as you mentioned ( 11th minute of your video) that you have already assigned water,air in 'other simulation'. So please can you tell how you assigned the phases to subparts in detail or some more hint .

[@GauravKumar-pg2ht]

Hello, same confusion. Were you able to resolve this?

[@akshaysreenivasan1829]

How is the solid surface at the base modeled and simulated?

[@TanmayAgrawal7]

You can assign that boundary as wall.

[@Mohamed.wael7]

Thank you so much for this video. I have a question about the mesh, why are you using this mesh with a very large domain. I mean can’t you reduce the height to have a more compact mesh to save computational time?

[@ivannepomnyashchikh7945]

I think your question needs to be reimposed. In every CFD simulation of an external flow we need to explore the size of the domain and achieve independence of the simulation relatively to the size of the domain (i.e. if I increase the domain even further my results must not change). How are we doing it? The criterion is that the domain boundaries must not interfere with the flow field. So, first we pick any domain, then do the simulation, then visualize pressure (or velocity) field, if we see that pressure (or velocity) contours are clearly intersected by the domain boundaries we move the domain boundaries a little bit further. The procedure should be repeated until there is no clear intersection of the contours by the domain boundaries. This will be the minimum possible domain - gives proper results with the minimum possible simulation time.

[@Mohamed.wael7]

Ivan Nepomnyashchikh Thank you so much. I understand. Do we need to do the same for an internal flow. I mean if we are simulating a long pipe with a droplet inside it. Can I take only a small portion of this pipe to decrease the mesh size. In this case is boundary independence necessary as well. I have heard that VOF works well if the secondary phase constitutes less than 10% of the domain is that true. Is that true! ( BTW I have tried periodic boundary and it introduced some errors)

[@ivannepomnyashchikh7945]

@@Mohamed.wael7 Sure, but for internal flow criterion for domain size independence is the absence of circulating (backwards) flow at the outlet. Once again you start with an arbitrary domain, launch the simulation, during the simulation FLUENT will be informing you in the text command line if you have flow recirculation at the outlet. If you have it - you results are unphysical. You need to move your outlet boundary a little bit further. Repeat the procedure until your domain stops interfering with the flow. That will be your minimum possible domain. The common rule it that you want to have fully developed flow at the outlet. One thing I need to mention is that picking the right domain size is a highly tedious procedure. Sometimes moving the boundaries don't help to prevent recirculation. In this case you need to decrease your mesh size. So playing with it is a nightmare. But that is the first must-have step to prove that your simulation is not junk but worth something. Periodic boundary conditions are for periodic walls. E.g. you have a flow over a ribbed wall. Ribs - that is what is called periodic wall (in this case). I don't know much about VOF. I myself have just started playing with it.

[@Mohamed.wael7]

Ivan Nepomnyashchikh Thank you so much. Your detailed explanation is very much appreciated.

[@sandeepgupta-uh4nq]

i am trying to model the flow of air past a cylinder in vicinity of the water bodies. and m having problem in creating two fluid domain. would be glad if u cld help

[@TanmayAgrawal7]

You need to have different cell zones of different materials I suppose.

[@anshumansinha5874]

After making 1 part in the geometry, I'm unable to get 2 zones in the cell zone condition!

[@TanmayAgrawal7]

Is your multiphase model on? Did you follow the exact procedure?

[@anshumansinha5874]

@@TanmayAgrawal7 Yes I did! Do you expect to find two parts in the cell zone after creating a single part? Shall I name both these parts in the meshing, 1 as air and other as water!

[@ahmadrkarami2159]

hi, tanx for video i try to simulate this problem and used both way to do it (adapt region and like what u did) but my droplet dont fall, and i dont know why gravity tick is on and using ansys 18

[@TanmayAgrawal7]

Are you using AND patching the correct volume fraction for water-phase? If problem persists, send me your model at [e-mail]

[@atulprakash332]

Hello Tanmay, it was an interesting to watch your simulation. I was wondering if you incorporated contact angle Boundary condition in your case setup? If yes, could elaborate little about the model which you used for contact angle? Thanks!

[@TanmayAgrawal7]

Hi Atul, I can't remember but maybe I didn't incorporate it.

[@ivannepomnyashchikh7945]

Thank you very much for the video! I am following it to simulate a single bubble rise in a vertical upward 3-D water flow. I can't properly put the bubble into my pipe in design modeler. You skipped that step in your video - so I was wondering if you could cover the topic in your tutorial please. Briefly: I have a 3-D case. I ended up having one part and two bodies. But when I go to Mesh and Setup I can see only one body. It turns out that both of your bodies are marked as "Surface body". But in my case, one body is specified as "Body" and another body is specified as "Surface body". And I can't seem to get them both being "Surface body". Thank you in advance. Ivan

[@ivannepomnyashchikh7945]

I've figured out how to do it. See the seventh reply in this post www.cfd-online.com/Forums/cfx/24537-how-model-sphere-designmodel-ansys.html Briefly: for a ball (i.e. a body) to be created in design modeler its sketch needs to be closed, otherwise a sphere will be created (i.e. a surface). Another thing that one needs to pay attention to is "Add material" and "Add frozen" when creating bodies. I have no idea which one to choose - but it seems to be important. There is a discussion on Quora about "Add frozen".

[@studypurpose7804]

thank you! Could you please share a tutorial for airborne dust simulation in underground tunnel of 3m dia and 100 m length ?

[@ashutoshkumarsaurabh72]

Sir, please upload simulation of the heat pipe

[@MrAmoslemi]

This is an incomplete tutorial without adding surface tension to see the real behavior of droplet fall! Did you try to add surface tension to it? when I add surface tension, the droplet doesn't move?

[@fesh503]

What is the value of surface tension you are adding?

[@asharmd2001]

Excellent video. Is it possible to make solid - liquid flow like a slurry flow in a pipe. Thanks.

[@TanmayAgrawal7]

It definitely is. Please refer to my newer video about granular multiphase model which is similar to what you need,

[@asharmd2001]

Bro, one basic question...How do you ' Time Step Size' and 'Number of Time Steps' required for a particular problem..because I have seen each person has own way to calculate, do you have any formula or equation to calculate the required ' Time Step Size' and 'Number of Time Steps' for this problem?...second.I mentioned about using laptop to run the simulation..can you give me the laptop specification? Thanks..

[@TanmayAgrawal7]

The time-step size is restricted by the stability criterion such as the CFL number. If the problem is simple geometry based, it can be readily calculated, otherwise you have to experiment. Start from a slightly larger time-step and if that diverges, go small. Max. number of time steps is simply the way of defining upto what value of physical time you want to run the simulation. Total time = Time step size * Total number of steps. This simulation was done on a Surface Pro 4. i5 6th gen. 8GB RAM.

[@Ya_baba_tajuddin]

Can you make video for steam bubble in water

[@diptasaha3841]

thank u so much, for this tutorial

[@vineettiwari5027]

please make a tutorial of water droplet falling in liquid pool

[@alviahmmed326]

It's really had to see for the high resolution.

[@shahinorrahmanb8728]

your video is awsome..........great helpfull. could you please upload a tutarial for multiphase flow occuring by pressure drop like flow of refrigerant in a refrigeration circuit. thanks in advance.

[@gembalisaichaitanya6179]

sir,suppose say droplet phase is water and air is surrounding phase,there are two droplets of water one above each other seperated by distance ,how to give different intial velocities to both water droplets?,please try to give your contact number sir,i already mailed you but there is no response

[@chandanmahata5241]

excellent job......can you please help me to create a geometry of stirred tank reactor

[@pradippatil507]

please make tutorial for partical tracking in dpm

[@TanmayAgrawal7]

I will try, thanks!

[@pradippatil507]

Tanmay Agrawal is UDF is necessary for practical tracking or it can directly excuted with creating injection ?

[@TanmayAgrawal7]

I think you can use the DPM model, it should do the job.

[@sreeramsarmabharatula9201]

Please please do a vedio on flow boiling....please ..i m waiting 4r u r reply...

[@kilasuelika6955]

Your computer screen must have very high resolution.

[@ateeqfastian]

good tutorial. can anyone please share how to obtain this type of geometry. As i m having unable to obtain this geometry. when ever i put droplet on rectangular region i creates a hollow circle in a rectangle. thanks

[@TanmayAgrawal7]

Make sure you have two different geometries/components. If both of the sketches are in the same geometry, it would cause trouble.

[@prashantchaudhary5898]

hello did you get how to obtain the geometry? I am facing the same issue, please help me out.