CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   FLUENT (http://www.cfd-online.com/Forums/fluent/)
-   -   Setting velocity at outlet (http://www.cfd-online.com/Forums/fluent/110355-setting-velocity-outlet.html)

saba_saeb December 9, 2012 13:01

Setting velocity at outlet
 
Hi all,

I am supposed to simulate a two-phase flow where a fluid enters from the top of a domain initially filled by air and exits the domain from an outlet located at the bottom. I am interested to change the fluid velocity where it is going out of the domain. According to available boundary conditions in fluent, it's not possible to set the velocity at the outlet.
I had a look to previous posts and topics in the forum, but didn't find anything related or useful. This question has been asked before, but no one has replied to any of them yet.
Also, according to the tutorial,
"In special instances, a velocity inlet may be used in FLUENT to define the flow velocity at flow exits. (The scalar inputs are not used in such cases.) In such cases you must ensure that overall continuity is maintained in the domain."
I tried the suggested approach, but didn't work for me.
Any help would be much appreciated.

Cheers,
Saba

syavash December 9, 2012 18:11

Quote:

Originally Posted by saba_saeb (Post 396531)
Hi all,

I am supposed to simulate a two-phase flow where a fluid enters from the top of a domain initially filled by air and exits the domain from an outlet located at the bottom. I am interested to change the fluid velocity where it is going out of the domain. According to available boundary conditions in fluent, it's not possible to set the velocity at the outlet.
I had a look to previous posts and topics in the forum, but didn't find anything related or useful. This question has been asked before, but no one has replied to any of them yet.
Also, according to the tutorial,
"In special instances, a velocity inlet may be used in FLUENT to define the flow velocity at flow exits. (The scalar inputs are not used in such cases.) In such cases you must ensure that overall continuity is maintained in the domain."
I tried the suggested approach, but didn't work for me.
Any help would be much appreciated.

Cheers,
Saba

You may change velocity magnitude at outlet by varying exit area.

saba_saeb December 9, 2012 18:20

Quote:

Originally Posted by syavash (Post 396554)
You may change velocity magnitude at outlet by varying exit area.

Hi Syavash and tnx for ur reply.
I might face 2 problems doing that.
1. In my case, a polymer is flowing down the inlet which has the diameter of 4 mm. By making outlet smaller, say 1 mm, which is surrounded by wall ( can I replace wall by anything else? ), a great fraction of polymer would hit the wall and remain there ( since the polymer is quite viscose) and doesn't exit the domain at all which makes the simulation far away from the real case!
2. I want to extend the simulation later such a way that I would be able to change the velocity of the fluid at the outlet as time goes by.

Cheers,
Saba

syavash December 9, 2012 19:01

Quote:

Originally Posted by saba_saeb (Post 396555)
Hi Syavash and tnx for ur reply.
I might face 2 problems doing that.
1. In my case, a polymer is flowing down the inlet which has the diameter of 4 mm. By making outlet smaller, say 1 mm, which is surrounded by wall ( can I replace wall by anything else? ), a great fraction of polymer would hit the wall and remain there ( since the polymer is quite viscose) and doesn't exit the domain at all which makes the simulation far away from the real case!
2. I want to extend the simulation later such a way that I would be able to change the velocity of the fluid at the outlet as time goes by.

Cheers,
Saba

Hmm..., I see. What kind of B.C. is implemented at outlet?!

saba_saeb December 9, 2012 19:10

Quote:

Originally Posted by syavash (Post 396557)
Hmm..., I see. What kind of B.C. is implemented at outlet?!

I set pressure-outlet just to see whether it works or not, and it did! But it doesn't satisfy what I'm looking for! I need to set the velocity at the outlet and pressure-outlet doesn't do that, clearly!

syavash December 9, 2012 19:16

Quote:

Originally Posted by saba_saeb (Post 396558)
I set pressure-outlet just to see whether it works or not, and it did! But it doesn't satisfy what I'm looking for! I need to set the velocity at the outlet and pressure-outlet doesn't do that, clearly!

Have you tried checking option "target mass flow rate" ?!
I should be reminding have you considered continuity satisfaction?!

saba_saeb December 10, 2012 02:25

Quote:

Originally Posted by syavash (Post 396559)
Have you tried checking option "target mass flow rate" ?!
I should be reminding have you considered continuity satisfaction?!

Thank you again for ur reply.
I tried to implement outlet-vent where "target mass flow" is available, but since I'm using VOF solver, it is not available! When I make VOF az the solver, this option disappears and it's not possible to set it as boundary condition.

How should I consider continuity?

Cheers,
Saba

RodriguezFatz December 10, 2012 06:18

Hi Saba,

Why don't you just set (in Fluent) a "velocity inlet" boundary condition to your outlet and a "pressure outlet" bc to your inlet? Then type a negative velocity at your "velocity inlet" to ensure that the fluid flows out of the domain.
You will get warnings of reversed flow at your pressure outlet, but that doesn't matter, because you actually want the fluid to enter there.

saba_saeb December 10, 2012 09:19

Quote:

Originally Posted by RodriguezFatz (Post 396617)
Hi Saba,

Why don't you just set (in Fluent) a "velocity inlet" boundary condition to your outlet and a "pressure outlet" bc to your inlet? Then type a negative velocity at your "velocity inlet" to ensure that the fluid flows out of the domain.
You will get warnings of reversed flow at your pressure outlet, but that doesn't matter, because you actually want the fluid to enter there.

Hi Rodriguez,

It doesn't solve my problem since I need to specify the velocity of the fluid at inlet as well as outlet. However, as you suggested, I set velocity-inlet boundary condition to both inlet and outlet while putting negative sign for outlet, but the results are just strange! It seems the fluid tends to move back to the domain instead of exiting it at the outlet!

Bests,
Saba

syavash December 10, 2012 09:27

Quote:

Originally Posted by saba_saeb (Post 396652)
Hi Rodriguez,

It doesn't solve my problem since I need to specify the velocity of the fluid at inlet as well as outlet. However, as you suggested, I set velocity-inlet boundary condition to both inlet and outlet while putting negative sign for outlet, but the results are just strange! It seems the fluid tends to move back to the domain instead of exiting it at the outlet!

Bests,
Saba

Dear Saba,

Try to adjust the outlet gauge pressure (pressure outlet B.C.) in order to get desired flow rate, and feed response back.

saba_saeb December 10, 2012 09:32

Quote:

Originally Posted by syavash (Post 396653)
Dear Saba,

Try to adjust the outlet gauge pressure (pressure outlet B.C.) in order to get desired flow rate, and feed response back.

Thanks Syavash,

It seems the only way to set velocity at outlet is to play with pressure instead of setting the exact desired velocity, as you just mentioned. I will have a try using pressure-outlet and will let you know about the results.

Bests,
Saba

RodriguezFatz December 10, 2012 09:33

I don't get you. Do you want to specify the velocity at both the inlet and the outlet?

saba_saeb December 10, 2012 09:34

Quote:

Originally Posted by RodriguezFatz (Post 396655)
I don't get you. Do you want to specify the velocity at both the inlet and the outlet?

Yeah, exactly!

RodriguezFatz December 10, 2012 09:36

So you use compressible air or what?

jamesg3373 December 10, 2012 09:40

You can't specify the velocity at both the inlet and the outlet. One has to be a pressure BC (inlet or outlet). This is because you need to conserve mass. So unless there is some degree of compression this isn't possible.

What exactly is the nature of the problem? Have you got a diagram? Is there gravity?

saba_saeb December 10, 2012 09:55

Quote:

Originally Posted by jamesg3373 (Post 396660)
You can't specify the velocity at both the inlet and the outlet. One has to be a pressure BC (inlet or outlet). This is because you need to conserve mass. So unless there is some degree of compression this isn't possible.

What exactly is the nature of the problem? Have you got a diagram? Is there gravity?

There wouldn't be any problem if I set velocity at both boundaries I guess, since both air and water are exiting the domain at outlet. This lets the fluid diameter to be changed in order to satisfy continuity.
I did quite similar simulation using openFoam and setting velocity at both inlet and outlet didn't make any problem at all!

Bests,
Saba

RodriguezFatz December 10, 2012 09:57

What do you mean by "fluid diameter"?

jamesg3373 December 10, 2012 10:04

Mass flow rate is kg/s.

What this means in terms of dimension is that the amount of fluid passing though your BC is determined by the velocity of the fluid, the density of the fluid and the area of the BC.

Your total mass flow rate into a system should be the same as the total mass flow rate out of the system. The only way I can see your velocity at each BC being different is if the area of the BCs are different or if the density of the fluid has changed.

If you want more help on this model you're going to have to provide more information in the way of diagrams and explanation. At the minute it's very difficult to imagine the problem you're trying to solve.

saba_saeb December 10, 2012 10:07

Quote:

Originally Posted by RodriguezFatz (Post 396668)
What do you mean by "fluid diameter"?

Here is a picture which might help ;)

http://i50.tinypic.com/2iizf9d.png

By fluid diameter, I mean the polymer thickness or the diameter by which fluid is exiting the domain!
According to the picture, fluid is coming down the inlet which has a diameter of 4 mm. As polymer gets closer to exit, it gets narrower. I have to mention that outlet diameter has been set to 2 mm. But, it is clear that "fluid diameter" is much less than 2 mm at outlet and air is exiting the domain through outlet as well as a consequence. So continuity can be satisfied since the fluid is wholly free to adjust its diameter or "thickness"!
I hope the picture makes it clearer! ;)

Bests,
Saba

jamesg3373 December 10, 2012 10:10

So what you're saying is that you want to model the outlet as elastic? So the diameter changes according to the pressure at the outlet?

EDIT: I've just reread your last post and in hindsight I don't think that's what you meant, but I'm still struggling to understand the problem. Is there a real life situation you can liken it to?


All times are GMT -4. The time now is 17:18.