Free falling fluid in the tube

pechwang · May 4, 2015, 08:56

Hi everyone,

I used star-ccm+ to simulate a free falling fluid example. The geometry is very simple, it is a 3*3*50 cm tube. Front, back, left and right surfaces are walls and using wall boundary condition. Top and bottom surfaces are treated as inlet and outlet. Initially it is full of water. At time 0, the fluid falls and goes out through the bottom. Air enters the domain through top.

The boundary conditions that I used is shown below:

Top: pressure outlet, zero-gradient for pressure and [0.0,1.0] for phase fraction
bottom: pressure outlet, constant 0 pressure and [0.0, 1.0] for the phase fraction
Walls: no-slip

I have two issues in my simulations:

1. The residual is very large, which is large indicating it is not converged.
2. The time for fluid to exit the tube is about 0.58 sec. However it should be about 0.3 sec based on hand-calculation.

Can someone tell whether I have the right BCs for this problem? Can someone give advice how to make the simulation converged and achieve a reasonable free fall time?

tiatabasco · June 29, 2016, 05:37

Hi!

I m trying to do a similar simulation . But I have a problem in stopping the simulation when the liquid is drained out. How do I include a stopping crieteria?
I tried with with reports and monitors somehow the simulations goes on running endlessly

pechwang · June 29, 2016, 09:28

You can change the stopping time at system folder. Make changes in the controlDict file.

tiatabasco · July 13, 2016, 10:11

I tried do a similar Simulation and somehow air and liquid water forms a mixture. how can I avoid this. It st just a gravity driven flow and incompressible and inviscid?

Eike · August 1, 2016, 07:38

Quote:

Originally Posted by pechwang

Hi everyone,

[...]

The boundary conditions that I used is shown below:

Top: pressure outlet, zero-gradient for pressure and [0.0,1.0] for phase fraction
bottom: pressure outlet, constant 0 pressure and [0.0, 1.0] for the phase fraction
Walls: no-slip

[...]

Maybe you should get a volume (sphere or box) around the bottom an around the top of the tube to get the pressure outlet(s) away from the end of the tube. As for the inlet you could try stagnation inlet.

marmot · August 2, 2016, 06:00

I agree with your boundary conditions they should work.

Perhaps later to optimize the problem you might want to use dynamic time stepping or do it manually

In the beginning your time step could be larger but as air enters you might have to reduce, make sure the air/water line is not moving more than 1 or 2 cells per time step. I prefer with VOF using a lower time step and 5-7 outer or inner iteration per time step.

Is surface tension on, will this have an influence on the time to drain, maybe check this? If you are getting a mixture of air/water, try increasing the sharpening factor in the model parameters.

May 4, 2015, 08:56	Free falling fluid in the tube	#1
pechwang Member Pengchuan Wang Join Date: Nov 2012 Location: Michigan USA Posts: 58 Rep Power: 13	Hi everyone, I used star-ccm+ to simulate a free falling fluid example. The geometry is very simple, it is a 3350 cm tube. Front, back, left and right surfaces are walls and using wall boundary condition. Top and bottom surfaces are treated as inlet and outlet. Initially it is full of water. At time 0, the fluid falls and goes out through the bottom. Air enters the domain through top. The boundary conditions that I used is shown below: Top: pressure outlet, zero-gradient for pressure and [0.0,1.0] for phase fraction bottom: pressure outlet, constant 0 pressure and [0.0, 1.0] for the phase fraction Walls: no-slip I have two issues in my simulations: 1. The residual is very large, which is large indicating it is not converged. 2. The time for fluid to exit the tube is about 0.58 sec. However it should be about 0.3 sec based on hand-calculation. Can someone tell whether I have the right BCs for this problem? Can someone give advice how to make the simulation converged and achieve a reasonable free fall time?

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Free Falling Water Film Mass and Heat Transfer	Boreas	Main CFD Forum	2	January 12, 2015 02:38
free falling of a wedge into water	ghoroobeaakhar	CFX	6	April 23, 2014 12:28
Rigid body free falling in Fluid .	archeoptyrx	Main CFD Forum	0	May 24, 2013 16:41
using ghost fluid method in shock tube problem	Amir	Main CFD Forum	4	February 24, 2011 10:09
Is there free Fluid Structure Interaction solution	Junpeng	Main CFD Forum	4	November 25, 2008 10:14

June 29, 2016, 05:37		#2
tiatabasco New Member Teena T. Join Date: Mar 2013 Location: Germany Posts: 14 Rep Power: 13	Hi! I m trying to do a similar simulation . But I have a problem in stopping the simulation when the liquid is drained out. How do I include a stopping crieteria? I tried with with reports and monitors somehow the simulations goes on running endlessly

June 29, 2016, 09:28		#3
pechwang Member Pengchuan Wang Join Date: Nov 2012 Location: Michigan USA Posts: 58 Rep Power: 13	You can change the stopping time at system folder. Make changes in the controlDict file.

August 2, 2016, 06:00		#6
marmot Senior Member kevin alun Join Date: Sep 2011 Location: Germany Posts: 106 Rep Power: 14	I agree with your boundary conditions they should work. Perhaps later to optimize the problem you might want to use dynamic time stepping or do it manually In the beginning your time step could be larger but as air enters you might have to reduce, make sure the air/water line is not moving more than 1 or 2 cells per time step. I prefer with VOF using a lower time step and 5-7 outer or inner iteration per time step. Is surface tension on, will this have an influence on the time to drain, maybe check this? If you are getting a mixture of air/water, try increasing the sharpening factor in the model parameters.