Why the water is flowing along the wall by VOF model?

michal.s · February 15, 2018, 07:59

Thank you, I will try in immediately.
Do you recommend some specific time step or solver options for VOF?
I was running one-phase simulations for this case with PV Coupling as Coupled, PRESTO! pressure discretization and pseudo transient.

usamaperwez · February 15, 2018, 10:34

Use variable time step (it is recommended for VOF), it will automatically try to adjust according to courant number.
PV Coupling as PISO
Gradient as Least Square Cell
Pressure as PRESTO
Momentum as Third order muscl
Volume Fraction as Geo-Reconstruct
Transient as First order implicit

In VOF model setting, activate Coupled level set also.

michal.s · February 15, 2018, 10:56

Thank you!
It looks like calculations goes a little bit better, however my time step is changing aroind 1e-07 / 1e-06.
I'll say what's the result after some calculations.
Looks like my mesh is too fine

usamaperwez · February 15, 2018, 10:58

Mesh aspect ratio should be 1.
It will take time since one of my simulation run at 1e-08 and it took 7 days to complete.

michal.s · February 16, 2018, 15:21

I tried all what you have suggested.
Worked a little bit better.
https://www.youtube.com/watch?v=DJwJ...ature=youtu.be
There is a short video with the results.

Still it doesn't behave properly.
Looks like the water stream is "laminar", like it was a vapour.

I also tried to do fast calculation using LES. With LES the flow was natural (as I wish to be).

Maybe I should manipulate with k-w settings?

usamaperwez · February 17, 2018, 02:08

another thing is that there is a collision. So you may need to activate collision model also. Just see the following collision tutorial on CFX using free surface model (same as VOF).
https://www.youtube.com/watch?v=KdFlwgLEnlo&t=16s

You want to generate splashing basically, i think so?
What about the wall contact angle? it is also very important in this situation.

michal.s · February 17, 2018, 07:02

I tried to modify the contact angle before, but didn't bring any changes (obviously that I have achieved before wasn't correct).

Thank you for the tutorial, that is moreless what I expect.
We are going to run an experiment, I am uploading a picture of what we had n the lab.
https://ibb.co/h4X0Ln

About collision in fluent - isn't in inside DPM?

I have to say usamaperwez I really appriciate all you suggestions and help! Without it I would be my nightmare

usamaperwez · February 27, 2018, 12:43

I rechecked again, there is no need of collision model.
Did your problem got solved?

usamaperwez · February 27, 2018, 12:54

which scheme are you using of VOF?
If it is explicit then try moving towards implicit !

michal.s · February 28, 2018, 03:57

Still doesn't work as I wanted to.
Using different turbulence models and settings the flow behaviour was changing. The realistic solution was with LES, however I would like to get it with k-w, k-E or RSM.

I also tried with steady state - sometimes it worked - stream shape became OK

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Centrifugal fan	j0hnny	CFX	13	October 1, 2019 14:55
Basic Nozzle-Expander Design	karmavatar	CFX	20	March 20, 2016 09:44
Low torque values on Screw Turbine	Shaun Waters	CFX	34	July 23, 2015 09:16
Displying interface of liquid water using VoF model	pchoopanya	FLUENT	2	March 15, 2013 17:42
Errors running allwmake in OpenFOAM141dev with WM_COMPILE_OPTION%3ddebug	unoder	OpenFOAM Installation	11	January 30, 2008 21:30

February 15, 2018, 07:59		#21
michal.s Member Michal Join Date: May 2017 Posts: 35 Rep Power: 9	Thank you, I will try in immediately. Do you recommend some specific time step or solver options for VOF? I was running one-phase simulations for this case with PV Coupling as Coupled, PRESTO! pressure discretization and pseudo transient.

February 15, 2018, 10:34		#22
usamaperwez Member usama Join Date: Jul 2016 Location: Pakistan Posts: 33 Rep Power: 10	Use variable time step (it is recommended for VOF), it will automatically try to adjust according to courant number. PV Coupling as PISO Gradient as Least Square Cell Pressure as PRESTO Momentum as Third order muscl Volume Fraction as Geo-Reconstruct Transient as First order implicit In VOF model setting, activate Coupled level set also.

February 15, 2018, 10:56		#23
michal.s Member Michal Join Date: May 2017 Posts: 35 Rep Power: 9	Thank you! It looks like calculations goes a little bit better, however my time step is changing aroind 1e-07 / 1e-06. I'll say what's the result after some calculations. Looks like my mesh is too fine

February 15, 2018, 10:58		#24
usamaperwez Member usama Join Date: Jul 2016 Location: Pakistan Posts: 33 Rep Power: 10	Mesh aspect ratio should be 1. It will take time since one of my simulation run at 1e-08 and it took 7 days to complete.

February 16, 2018, 15:21		#25
michal.s Member Michal Join Date: May 2017 Posts: 35 Rep Power: 9	I tried all what you have suggested. Worked a little bit better. https://www.youtube.com/watch?v=DJwJ...ature=youtu.be There is a short video with the results. Still it doesn't behave properly. Looks like the water stream is "laminar", like it was a vapour. I also tried to do fast calculation using LES. With LES the flow was natural (as I wish to be). Maybe I should manipulate with k-w settings?

February 17, 2018, 02:08		#26
usamaperwez Member usama Join Date: Jul 2016 Location: Pakistan Posts: 33 Rep Power: 10	another thing is that there is a collision. So you may need to activate collision model also. Just see the following collision tutorial on CFX using free surface model (same as VOF). https://www.youtube.com/watch?v=KdFlwgLEnlo&t=16s You want to generate splashing basically, i think so? What about the wall contact angle? it is also very important in this situation.

February 17, 2018, 07:02		#27
michal.s Member Michal Join Date: May 2017 Posts: 35 Rep Power: 9	I tried to modify the contact angle before, but didn't bring any changes (obviously that I have achieved before wasn't correct). Thank you for the tutorial, that is moreless what I expect. We are going to run an experiment, I am uploading a picture of what we had n the lab. https://ibb.co/h4X0Ln About collision in fluent - isn't in inside DPM? I have to say usamaperwez I really appriciate all you suggestions and help! Without it I would be my nightmare

February 27, 2018, 12:43		#28
usamaperwez Member usama Join Date: Jul 2016 Location: Pakistan Posts: 33 Rep Power: 10	I rechecked again, there is no need of collision model. Did your problem got solved?

February 27, 2018, 12:54		#29
usamaperwez Member usama Join Date: Jul 2016 Location: Pakistan Posts: 33 Rep Power: 10	which scheme are you using of VOF? If it is explicit then try moving towards implicit !

February 28, 2018, 03:57		#30
michal.s Member Michal Join Date: May 2017 Posts: 35 Rep Power: 9	Still doesn't work as I wanted to. Using different turbulence models and settings the flow behaviour was changing. The realistic solution was with LES, however I would like to get it with k-w, k-E or RSM. I also tried with steady state - sometimes it worked - stream shape became OK