Instable interface to solve VOF problem

 Register Blogs Members List Search Today's Posts Mark Forums Read

March 8, 2017, 01:43
Instable interface to solve VOF problem
#1
New Member

Jemyung Cha
Join Date: Jul 2009
Posts: 11
Rep Power: 10
Hello, CFD users.

I'm trying to solve some VOF problem.

Fig. 1
There are small air bubble attached on the cavity and water injected from left into microchannel. As time goes by, two immiscible fluid is shaped like by expectation.

Fig. 2
Very stong velocity fluctuation generated at the interface between air bubble and water like second picture.
Inlet velocity is 0.0175 m/s but maximum velocity is 0.869 m/s. I think it is non-physical and unexceptable results.

From some literature, this kind of result caused by big difference in viscosity of two fluids.

Question.
Do you have same phenomenon when calculating VOF, especially small scale problem?
Is there a some CFD tip to enhance instability using some damping technique in Ansys Fluent?

Attached Images
 pic1.png (27.5 KB, 11 views) pic2.png (110.0 KB, 9 views) pic3.jpg (146.4 KB, 9 views)

Last edited by cicatrix; March 8, 2017 at 05:09.

 March 8, 2017, 03:13 #2 Senior Member   Lucky Tran Join Date: Apr 2011 Location: Orlando, FL USA Posts: 1,982 Rep Power: 26 Is this a transient simulation? Have you tried a smaller time-step? Have you tried using a more dissipative discretization scheme? Like 1st order upwind instead of 2nd order? The benefits of second order scheme are not very beneficial when they are not accurate, in which case it makes much more sense to switch to a low-order more scheme that is more accurate. Can you really prove though that the instability is caused by the interface? Does not have to be a real physical proof, but at least from the numerical result? The VOF method is (extremely) diffusive, and this certainly can enhance instabilities and cause them to grow. You would need some way to introduce numerical dissipation to get damping. If it's not a real effect, then it's a battle between numerical diffusion vs numerical dissipation.

 March 8, 2017, 05:08 #3 New Member   Jemyung Cha Join Date: Jul 2009 Posts: 11 Rep Power: 10 Yes, this is transient simulation with very small time step (1E-8 second). I will follow your opinion and tests about 1st order scheme. Numerically - I did some tests with other commercial code (CFD-ACE+). The result I got from it with damping technique and fluctuation is disappeard. (attached figure 3) Physically- I think it makes sense. When I drop a small amount of water on a surface and record with high speed camera, surface(interface) of water droplet continuously move until its equilibrium state. I'm trying to search some similar phenomena and numerical diffusion/dissipation issue. Feel free to say to me any idea. Thanks.

March 8, 2017, 10:41
#4
Senior Member

Lucky Tran
Join Date: Apr 2011
Location: Orlando, FL USA
Posts: 1,982
Rep Power: 26
Quote:
 Originally Posted by cicatrix I think it makes sense. When I drop a small amount of water on a surface and record with high speed camera, surface(interface) of water droplet continuously move until its equilibrium state.
But does that correspond to the effect you see here? I mean in your simulation, the instability you are seeing is the same surface waves or something else? Your surface appears smooth and there does not appear to be surface waves. For example is it just some bad mesh that causes numerical oscillations? What is the origin of your instability within the simulation?

If you drop water on a surface, you see the surface waves move around. However, surface waves also do not spontaneously accelerate to say the speed of light..

 March 9, 2017, 01:22 #5 New Member   Jemyung Cha Join Date: Jul 2009 Posts: 11 Rep Power: 10 I'm not sure we are talking about same thing. I just want to express that there is a unstable interface when doing simulation water droplet and air bubble on a surface. In an engineering sense, air bubble is not moving but also acts like some wall with slip condition. I mean I have a numerical problem when using VOF model. Numerical oscillation is not disappeared even if I used very fine mesh. I guess that numerical solver cannot handle this kind of simulation properly. So I am looking for a someone who has a same experience about that.

 Tags microchannel, vof

 Thread Tools Display Modes Linear Mode

 Posting Rules You may not post new threads You may not post replies You may not post attachments You may not edit your posts BB code is On Smilies are On [IMG] code is On HTML code is OffTrackbacks are On Pingbacks are On Refbacks are On Forum Rules

 Similar Threads Thread Thread Starter Forum Replies Last Post Abo Anas CFX 26 December 13, 2016 11:17 Sanyo CFX 17 August 15, 2015 06:20 volo87 CFX 5 June 14, 2013 17:44 Ramsey FLUENT 1 February 16, 2011 14:16 Mikhail Main CFD Forum 40 September 9, 1999 09:11

All times are GMT -4. The time now is 04:23.