I don't actually know what you mean by elastic. But the diameter is being changed according to the velocity that I set at the outlet.

Also, what means "atmosphere"? Pressure outlet?
And: Does the fluid accelerate inside your domain? Due to gravity?

In all honesty; I can't really help you with this unless there's some context. I have no idea what you're trying to do. If you could give me some more information like: What situation are you trying to model? What information are you trying to obtain?

Yes, it can be considered like pressure outlet in this case. And yes, it is being driven down due to both gravity and the velocity that I set at the inlet.

I have already explained everything I guess, I try to explain more, I hope it helps ;)

http://i50.tinypic.com/352hpi9.jpg

According to the picture, melted polymer is flowing down with an initial fixed velocity to a domain filled by air. Due to gravity and the initial velocity, it flows down till it reaches the point where the fluid is not influenced by gravity anymore, but it is driven by two disks which are rotating by a specific speed ( just one of the disks has been depicted in this picture ). So, I want to set the velocity of the fluid both at inlet and outlet! (outlet velocity would be the velocity by which disks are pulling the polymer down!)
From the results that I gained in openFoam, at a specific outlet velocity, while the inlet velocity is kept fixed, ( when outlet velocity is 22 times bigger than inlet velocity ) the polymer flow is cut and this results in some malfunctions in the system which should be avoided!
I need to do the same simulation using Fluent now!
If any further information is needed, please let me know ;)

Bests,
Saba

Now it makes sense. The pressure outlet allows for mass conservation.

Ok, now, can't you use two velocity inlets? One with positive, one with negative velocity?

No, this approach doesn't work. I put it into practice, but the result was totally strange! Fluid doesn't exit the domain and tends to move upward when I set the velocity inlet with negative velocity at outlet!

You are right. If you set the velocity at an outlet in Fluent, it will set constant velocity along the whole outlet. I guess, this causes the strange behavior. I don't think you can have a variable fraction of polymer/air. Maybe with an UDF.

Hello Saba,
Recognize me?:)
sup dude? Seems you've been working on multiphase flow.
I read some of the replies. Why are you interested to set the outlet as velocity? some guys mentioned the elasticity. Is the fluid elastic or non-Newtonian?

Hey man, of course!
Good to see you here!
Yeah, I'm supposed to simulate a two-phase flow that I fully described in my previous posts. ;)
According to the explanation of the problem ( here ) and (here), we are just aware of velocity of fluid at inlet and outlet and nothing else!
In principle, the fluid is non-Newtonian. But just to start, I am considering it as a Newtonian fluid. I need to replace it by non-Newtonian fluid in further steps, anyway.

Cheers,
Saba

I see.
There is some issues I have in mind if I understand it well. You are simulating a two-phase flow, one liquid and another air, is that correct? your domain first filled with air and then the liquid will enter. When you initial your solution with a mixture, the software fills the whole domain with it. But if you are considering the domain first filled of air and then the liquid enters, it needs some udf to do that.
What's the disk location? Couldn't you just set the outlet boundary as a moving wall to simulate the moving disk? I was not able to find any picture here.
And what's the Re number? You're considering compressibility?