Eulerian CFD simulation of rise of a spherical bubble in a continuous denser medium

nomi.tahir · March 7, 2016, 08:47

I tried solving the rise of a spherical bubble of air in continuous fluid ( water) using VOF method but it diverges . Although I found a tutorial dealing with ink drop falling in water which was solved using VOF. What I dont get is that if an ink droplet simulation in water can be dealt using VOF then why is my solution diverging when I am doing it for rising bubble.
I did it with mixture model and it works . So kindly let me know why possibly my case diverging using VOF model but it's working with mixture model. Thanks for your help ! Kindly let me know what could be the parameters due to which mixture model predicting better results

CeesH · March 7, 2016, 12:55

Can you provide a bit more information? Grid size, setup, timestep size, that kind of details?

nomi.tahir · March 7, 2016, 16:57

delta y is 0.038 m and delta x 0.0135 m . A column of dimension X=0.5m and Y=1.5 m. Delta t is 0.01 sec. The scheme is implicit first order. What else you want ?

CeesH · March 7, 2016, 17:21

have you tried with an explicit scheme and low Courant number (<0.25)?

nomi.tahir · March 8, 2016, 09:31

I did it with explicit. I lowered the time step as low as 2e-08 sec . But the bubble detachment didnt take place then . It took what 3 plus more days and the bubble growth was great but the bubble didnt detach . So I stopped it as I realized may be it wont detach . What do you say on this ?

CeesH · March 8, 2016, 09:40

Hm, I'm currently running some 2D droplets with VOF, and they are sinking just fine. But you mention growth and detachment; how do you model these things? Does a rising bubble work if you already start with it being detached?

For bubble detachment, the surface tension force will of course be very important. How is this included in your simulations (and how is it in mixture - since that did work)

Best,
C

nomi.tahir · March 8, 2016, 09:47

A rising bubble with lower time step of 0.01 sec is rising . Air is given at inlet with a velocity described by UDF and after a while when the bubble grows the bubble detaches but then soon after it begins to rise the continuity diverges . What I cant understand is that if falling drops can be simulated through VOF model just as you mentioned then why aint rising bubbles detachment and continuity convergence an issue here ?

CeesH · March 8, 2016, 10:37

Rising bubbles should go just as well; the sign of gravity should have no influence whatsoever. But if I understand correctly, the issue only starts as soon as the droplet detaches and actually starts rising? It sounds to me like the timestep is too big, which becomes noticeable as soon as the droplet starts rising. Maybe you can do variable timestepping with a set courant criterion?

March 7, 2016, 08:47	Eulerian CFD simulation of rise of a spherical bubble in a continuous denser medium	#1
nomi.tahir New Member Join Date: Mar 2016 Posts: 11 Rep Power: 10	I tried solving the rise of a spherical bubble of air in continuous fluid ( water) using VOF method but it diverges . Although I found a tutorial dealing with ink drop falling in water which was solved using VOF. What I dont get is that if an ink droplet simulation in water can be dealt using VOF then why is my solution diverging when I am doing it for rising bubble. I did it with mixture model and it works . So kindly let me know why possibly my case diverging using VOF model but it's working with mixture model. Thanks for your help ! Kindly let me know what could be the parameters due to which mixture model predicting better results

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
CFD Salary	CFD	Main CFD Forum	17	January 3, 2017 18:09
CFD Design...The CFD Future	John C. Chien	Main CFD Forum	20	November 20, 2015 00:40
Solar Radiation in OpenFOAM	plainstyle	OpenFOAM Running, Solving & CFD	15	July 8, 2014 05:43
Where do we go from here? CFD in 2001	John C. Chien	Main CFD Forum	36	January 24, 2001 22:10
Reactor simulation using CFD takes to the Internet	Mark Brennan	Main CFD Forum	0	April 16, 1999 06:57

March 7, 2016, 12:55		#2
CeesH Senior Member Cees Haringa Join Date: May 2013 Location: Delft Posts: 607 Rep Power: 0	Can you provide a bit more information? Grid size, setup, timestep size, that kind of details?

March 7, 2016, 16:57		#3
nomi.tahir New Member Join Date: Mar 2016 Posts: 11 Rep Power: 10	delta y is 0.038 m and delta x 0.0135 m . A column of dimension X=0.5m and Y=1.5 m. Delta t is 0.01 sec. The scheme is implicit first order. What else you want ?

March 7, 2016, 17:21		#4
CeesH Senior Member Cees Haringa Join Date: May 2013 Location: Delft Posts: 607 Rep Power: 0	have you tried with an explicit scheme and low Courant number (<0.25)?

March 8, 2016, 09:31		#5
nomi.tahir New Member Join Date: Mar 2016 Posts: 11 Rep Power: 10	I did it with explicit. I lowered the time step as low as 2e-08 sec . But the bubble detachment didnt take place then . It took what 3 plus more days and the bubble growth was great but the bubble didnt detach . So I stopped it as I realized may be it wont detach . What do you say on this ?

March 8, 2016, 09:40		#6
CeesH Senior Member Cees Haringa Join Date: May 2013 Location: Delft Posts: 607 Rep Power: 0	Hm, I'm currently running some 2D droplets with VOF, and they are sinking just fine. But you mention growth and detachment; how do you model these things? Does a rising bubble work if you already start with it being detached? For bubble detachment, the surface tension force will of course be very important. How is this included in your simulations (and how is it in mixture - since that did work) Best, C

March 8, 2016, 09:47		#7
nomi.tahir New Member Join Date: Mar 2016 Posts: 11 Rep Power: 10	A rising bubble with lower time step of 0.01 sec is rising . Air is given at inlet with a velocity described by UDF and after a while when the bubble grows the bubble detaches but then soon after it begins to rise the continuity diverges . What I cant understand is that if falling drops can be simulated through VOF model just as you mentioned then why aint rising bubbles detachment and continuity convergence an issue here ?

March 8, 2016, 10:37		#8
CeesH Senior Member Cees Haringa Join Date: May 2013 Location: Delft Posts: 607 Rep Power: 0	Rising bubbles should go just as well; the sign of gravity should have no influence whatsoever. But if I understand correctly, the issue only starts as soon as the droplet detaches and actually starts rising? It sounds to me like the timestep is too big, which becomes noticeable as soon as the droplet starts rising. Maybe you can do variable timestepping with a set courant criterion?