Modelling free falling water into a air filled duct?

sandmike_83 · August 23, 2010, 04:35

Hello to everybody. I'am trying to represent using Ansys CFX a 10 m long duct initially filled with air. The duct has only one way in or out, a hole at the top through which water is introduced.

I was trying to simulate it using the surface model, with homogeneous model,transient, double precision,

I get two strange results:

1. When monitoring the water volume fraction in the duct I noticed that In the first time step (just when the water is invading the access hole) all the domain achieves a water volume fraction of 0.5. To me this sounds like a numerical problem, but I don't know if anyone has experienced something like this?

2. The velocity at which the water falls is more or less the same as the inlet velocity (2 m/s). Shouldn't it be close to the free falling speed of a waterfall (around 10 m/s)

Any advise is welcome
Thanks for your help

MS

ghorrocks · August 23, 2010, 08:43

Can you post some images?

sandmike_83 · August 23, 2010, 09:19

some pictures below, hope the quality is enough.
The velocity at the inlet is 2 m/s.
the duct diameter is around 2"
The duct lenght is around 10 m
The Mesh is not very refined, but enough for a first approximation

I am starting to believe that the first problem is related somehow with the given initial conditions. Seems to me that the solver is imposing a 0.5 volume fraction for the entire domain, ignoring the initial conditions file.

Now, for the second, no clue so far

Thanks for the possible help Mr Horrocks

ghorrocks · August 23, 2010, 19:30

Your first question is an initial conditions problem. Just define the IC as VF=0.

The second question is dodgy. Firstly, the flow will take some time to accelerate to terminal velocity. You will probably have to run it for longer. Secondly, do you know what terminal velocity is? It is a function of drop size, fluid density, drop shape and other parameters - and it is also a function of the surrounding air flow. Your flow is clearly different to your reference waterfall. Additionally, a waterfall will entrain an air gust with it so the water droplets see reduced drag and this will increase the terminal velocity.

sandmike_83 · August 24, 2010, 04:27

Ok, I will try changing the initial conditions to see what I get.

Yes, I did some calculations to determine terminal speed of one water dropplet , and I got that result, somehow similar to that of a waterfall. Terminal velocity is also achieved in a very small lenght, but maybe in this case due to the trapped air, it will require longer distances.

Thanks for your help

MS

August 23, 2010, 04:35	Modelling free falling water into a air filled duct?	#1
sandmike_83 New Member Michael Sanders Join Date: Aug 2010 Posts: 4 Rep Power: 16	Hello to everybody. I'am trying to represent using Ansys CFX a 10 m long duct initially filled with air. The duct has only one way in or out, a hole at the top through which water is introduced. I was trying to simulate it using the surface model, with homogeneous model,transient, double precision, I get two strange results: 1. When monitoring the water volume fraction in the duct I noticed that In the first time step (just when the water is invading the access hole) all the domain achieves a water volume fraction of 0.5. To me this sounds like a numerical problem, but I don't know if anyone has experienced something like this? 2. The velocity at which the water falls is more or less the same as the inlet velocity (2 m/s). Shouldn't it be close to the free falling speed of a waterfall (around 10 m/s) Any advise is welcome Thanks for your help MS

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Setfields inoutlet and water and air patches	erik023	OpenFOAM Pre-Processing	1	September 29, 2008 11:05
Modelling of bubble merging submerge on water	Carlos	Main CFD Forum	0	October 26, 2005 10:30
moving water from a chamber using air	Matt Mason	CFX	0	December 2, 2003 20:46
Terrible Mistake In Fluid Dynamics History	Abhi	Main CFD Forum	12	July 8, 2002 10:11
Asymmetric water free jet study	Norberto Parreira	Main CFD Forum	7	June 25, 2001 09:23

August 23, 2010, 08:43		#2
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,871 Rep Power: 144	Can you post some images?

August 23, 2010, 19:30		#4
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,871 Rep Power: 144	Your first question is an initial conditions problem. Just define the IC as VF=0. The second question is dodgy. Firstly, the flow will take some time to accelerate to terminal velocity. You will probably have to run it for longer. Secondly, do you know what terminal velocity is? It is a function of drop size, fluid density, drop shape and other parameters - and it is also a function of the surrounding air flow. Your flow is clearly different to your reference waterfall. Additionally, a waterfall will entrain an air gust with it so the water droplets see reduced drag and this will increase the terminal velocity.

August 24, 2010, 04:27		#5
sandmike_83 New Member Michael Sanders Join Date: Aug 2010 Posts: 4 Rep Power: 16	Ok, I will try changing the initial conditions to see what I get. Yes, I did some calculations to determine terminal speed of one water dropplet , and I got that result, somehow similar to that of a waterfall. Terminal velocity is also achieved in a very small lenght, but maybe in this case due to the trapped air, it will require longer distances. Thanks for your help MS