# Artificial high velocities at the interface using interFoam

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

 February 15, 2012, 12:32 #21 Senior Member   Niels Gjoel Jacobsen Join Date: Mar 2009 Location: Deltares, Delft, The Netherlands Posts: 1,631 Rep Power: 25 Hi Joris You could type Code: forAll(U, celli) { if (alpha1[celli] < 0.01) U[celli] = vector::zero; } However, you ought to be careful as you are threading onto an extremely dangerous path. What you are doing here is equivalent to be removing energy from the system. I do not have the solution to circumvent the high velocities, but I my case I have found the above approach to yield utterly wrong results. Best regards, Niels

 February 15, 2012, 14:10 #22 New Member   Join Date: May 2011 Posts: 15 Rep Power: 6 Thanks niels for your answer. I'll try that tomorrow. My case is the following : Water is flowing onto a slope by its own weigth. The bc are cyclic in the slope direction. The top boundary is set to athmosphere so the air is free to enter and leave the system. AT time 0 it gives : In reality, only a thin layer of air is moved by the flowing water, but it is quickly slow down upwards by the athmosphere. As I set up a athmosphere bc, I guess it is not too important to loose energy from the air field. Only the water field matter for me (the pressure and velocity of water would contribute almost totally to the water depth level). I think Shallow water equation are derived from Navier Stokes taking a free velocity bc for the water surface. So in my case, air should influate a minimum the water surface. (maybe I should decrease a lot air viscosity for that?)

 February 16, 2012, 03:51 #23 Senior Member   Niels Gjoel Jacobsen Join Date: Mar 2009 Location: Deltares, Delft, The Netherlands Posts: 1,631 Rep Power: 25 However, since you are removing energy from a system, where the Navier-Stokes equations are solved simultaneously for both air and water, then you indirectly removes energy from the water, since energy from the water will be used to re-accelerate the air. Work put into the air, which is then removed in subsequent time steps. The approach could be working for your case; you merely have to exhibit extreme caution. / Niels

 February 16, 2012, 04:14 #24 New Member   Join Date: May 2011 Posts: 15 Rep Power: 6 I see what you mean, I will be carefull though. I 'll post later my rersults, cheers joris

 May 5, 2014, 11:59 compile problem #25 New Member   Baek, Donghae Join Date: Jan 2013 Location: Seoul Posts: 24 Rep Power: 4 hello joris.hey I got same problem. did you solve it ??? please advise me

October 16, 2014, 17:18
To be sure
#26
New Member

Remi Carmi
Join Date: Jul 2014
Posts: 14
Rep Power: 3
Quote:
 Originally Posted by Bernhard I'd like to add something to this thread, since I encoutered similar problems for my case. This problems disappeared with reducing the density ratio in my case from 1000 to 10. Since I don't like the clipping in the air velocity I was able to get better results by applying the argument of Brackbill ( http://www.sciencedirect.com/science...2199919290240Y ) for surface tension, on the additional source term in the Navier-Stokes equations resutling for the reformulation of the pressure. Basically what I did, was replacing this (in pEqn.H and UEqn.H of interFoam) Code: - ghf*fvc::snGrad(rho) By Code: - ghf*fvc::snGrad(rho)*2.0*fvc::interpolate(alpha1) The argument by Brackbill is summarized as follows: as the interface thickness goes to zero, you can multiply the body force by a function that is 1 at the interface. Since \grad\rho is zero except at the interface and 2alpha1 is 1 there, I think this is justified (please correct me if I'm wrong) For me it solved the issues with the high air velocities, and this implementation is at least more justified than just clipping U.

I am facing a similar problem and I have tried all the solutions mentioned :

alpha1*div(rhophi,U) => well then the air flow is not transported and saturates where it is generated.
U air to 0 or a percentage of the solution value (that should work but it does not to violent a create instability)
I tried increasing the density : it is a bit better
I tried also to do higher rho and Brackbill.

What was your final solution Bernhard?
are you still doing the alpha1*div and the increasing rho plus blackbill?

Thanks
Remi

 November 7, 2014, 07:52 #27 New Member   Yuanchuan Liu Join Date: Oct 2012 Posts: 20 Rep Power: 4 Hi Remi, I have been bothered by this problem for quite a long time. Which of those solutions mentioned in the earlier posts works best for you? Cheers, Yuanchuan Liu

 November 7, 2014, 15:13 #28 New Member   Remi Carmi Join Date: Jul 2014 Posts: 14 Rep Power: 3 Hi Yuanchuan, I will increase the density of the air. But that depends on what you are trying to do? My problem was when putting an object near the interface and moving it. Then I have vorticity (shedding) generated by the object (chose your frame it happens if you put a flow or move the object). And when the vorticity reaches the interface it kind of generates fast flow in the easy to transport air. Making it thicker (just rho_water / rho_air =100 is still big) really reduce this not necessary transport. Killing the flow in air is definitely not a good idea. Now that I think about it, would making a viscosity gradient help? (high viscosity as you move up in the air to dissipate the energy or just higher air viscosity to damp the transport, it is kind of what people do just by making the grid more loose in the air? numerical viscosity is bigger!). Let me know what is your final call. Best Remi

 November 7, 2014, 15:44 #29 New Member   Yuanchuan Liu Join Date: Oct 2012 Posts: 20 Rep Power: 4 Hi Remi, If I understand you correctly, you offer two ways to mitigate the effects of air convection: 1. Increase air density so that it is not so easy for air to transport. 2. Increase air viscosity so that even if high velocity occurs it will damp out quickly. These two methods might work for my current case where a cylinder with a horinzontal thick plate oscillates vertically in xoz plane since air is not important here. Besides, I do not need to change the code, which is fine. However, if I later want to take into account the aerodynamic force acting on the object, altering the properties of the air phase might be improper as air becomes important in this case. Of course, all the methods mentioned in this thread will change the real physics and are actually based on the assumption that air does not play a major part in my cases. I am wondering if there is any way to solve this problem without the need to alter the original problem. Really curious about how commercial solvers deal with this problem. Anyway, thanks very much for sharing your thoughts with me. I will try them to see if they will make a difference. Cheers, Yuanchuan

 October 2, 2015, 04:49 a RANS in air only #30 New Member   Remi Carmi Join Date: Jul 2014 Posts: 14 Rep Power: 3 Hi all, I am still playing with this problem. I have recently tried a new idea. I activate turbulence in Air only. I created a new k-E RASModel where I use the scalar alpha1 to turn it on or off whether I am in air or water. This is quite convenient but might be wrong since I am in 2D. It however gives rather good results somehow in my case (at least qualitatively compares well with my experiments) I have a manually oscillating pincher near a free surface. Any other comments on your side? Best Rmi

 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 Saturn CFX 34 October 16, 2014 05:27 andersson.j OpenFOAM Running, Solving & CFD 0 February 8, 2011 11:43 Pankaj CFX 9 November 23, 2009 05:05 Andrea CFX 2 October 11, 2004 05:12 lego CFX 3 November 5, 2002 21:09

All times are GMT -4. The time now is 05:03.