CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   FLUENT (http://www.cfd-online.com/Forums/fluent/)
-   -   How to create a tank which partially filled with fluid? (http://www.cfd-online.com/Forums/fluent/112493-how-create-tank-partially-filled-fluid.html)

ORYON January 30, 2013 00:06

How to create a tank which partially filled with fluid?
 
I'm conduct a filling simulation which is filling heavier density liquid into a tank which is partially filled with lighter density liquid. Let say the total height of the tank is 10m and partially filled with 3m water. Stratification will happen during filling.
My question is how to create the geometry of this tank and related setting by using Fluent 14.0?

RodriguezFatz January 30, 2013 03:14

Hi,
As far as I know, you need to name a body in your geometry / meshing tool. In this case you could create the tank with two different bodies - one "upper" and one "lower" body. Now set some named selections to make these bodies available in Fluent.
In Fluent - once you switched on the multiphase model and after initializing - you can patch the volume fraction of fluid i to zero and of fluid ii to 1 in the upper body and vice versa in the lower body.

What do you use for geometry and meshing?

ORYON January 30, 2013 03:49

I did try the method that u mentioned above, but the problem I faced is at time=0, the tank is filled with 3m heavier fluid but also the other 7m is already filled with lighter fluid. What I want is at initial stage the tank is filled with 3m heavier fluid only, the rest 7m is filled during filling with lighter fluid, so the total height of fluid inside the tank will increase during filling.

RodriguezFatz January 30, 2013 03:53

Remember: You can not have "nothing" inside a volume at any time. Also at t=0 you have to have some kind of fluid above 3m. It sounds like you have to introduce a third fluid (air?) to do what you want.

ORYON January 30, 2013 04:16

That means my problem involved 3 phases, am I right? By the way, how to made my simulation like the height of total fluid inside the tank increase during filling? what I did before is I named the top of the tank as pressure outlet,but it seem like didn't work.

RodriguezFatz January 30, 2013 04:37

Quote:

Originally Posted by ORYON (Post 404951)
That means my problem involved 3 phases, am I right?

It sounds like that.

Quote:

Originally Posted by ORYON (Post 404951)
By the way, how to made my simulation like the height of total fluid inside the tank increase during filling? what I did before is I named the top of the tank as pressure outlet,but it seem like didn't work.

Where does the second (filling) fluid come from?

ORYON January 30, 2013 06:12

Bottom filling-the inlet is at the bottom of the tank.

RodriguezFatz January 30, 2013 06:16

Then, pressure outlet for the top should be the right thing. What happened? BTW: You should set the backflow fraction of the two liquids to zero at the top to ensure that only air flows back from the top...

ORYON January 31, 2013 09:15

3 Attachment(s)
Hi, I faced some problem after my simulation and i attach my result and problem as attachment in this post. Generally my problem are:
1. how to set the inlet with filling rate 10m3/s?
2. why the path of filling doesn’t show?
3. Why the heel with mix and react with bog? What I want is the feed is mix with heel.

Thx

RodriguezFatz January 31, 2013 09:25

1) Is this something different to the things we discussed? I think we were talking about a three-phase flow.
2) Why do you use a pressure inlet? If you know the volume flux, use a velocity inlet and divide the 10m^3/s by the area of your inlet to get the velocity.
3) What do you mean by 2.
4) What do you insert through the inlet?

Can you create a picture of your geometry? I just see colored faces... Is there any symmetry?

ORYON January 31, 2013 09:49

1) what is the difference between three phase with my case?
3) By referring to figure 6, there is a horizontal path from inlet reach the surface of the heel( blue-green-yellow). But this kind of path didn't occur in my simulation(figure 5).
4) Lighter fluid(feed) is filling from the inlet to tank.

ORYON January 31, 2013 09:53

1 Attachment(s)
here is the geometry

RodriguezFatz January 31, 2013 10:04

1) Ok. Now, in your last post (picture) you say "this is a closed tank". But it isn't. It is open on the top since you defined it as a pressure outlet.
2) The inlet is just a small half-circle on the ground? What's the size?
3) Bottom and side-wall are defined as "walls" in fluent?

BTW: What is "heel"? My dictionary says it is some part of the foot. Nothing else mentioned there...

ORYON January 31, 2013 10:25

1 Attachment(s)
1) i think that is the key for my problem. For my case, i want to model a closed tank( LNG tank). What should i define for a closed tank?
2) the radius of inlet is 0.22m
3)Yes, the bottom and side wall are defined as wall.

"Heel" is refer to the fluid which is already inside the tank.
"feed" is refer to the fluid which is to be filled to the tank.
" BOG" -Boil of gas is the vapor form of fluid.

RodriguezFatz January 31, 2013 10:38

What happens to the gas at the top of the real tank during filling? Is it compressed?

to 2) -> then you have an inlet area of about 0.152m^2. To get your desired 10^m3/s you need an inlet velocity of 65.79m/s. No pressure inlet.

But what means "heel"?

ORYON January 31, 2013 11:00

1) Boil off Gas generally will be vented out and reliquefy back to Fluid.

"Heel" actually is a term used in Oil and gas industry. Heel in here is means the residual fluid( old LNG) inside the tank before filling the new LNG.

LNG- Liquefied Natural Gas

The objective of my simulation is to simulate the mixing behavior between 2 different density of LNG.

Here is the published paper:
https://www.google.com.my/url?sa=t&r...aNp6hvVBT052og

RodriguezFatz January 31, 2013 11:05

Ok, if it is vented out, then just use a pressure outlet for the vent, or leave the top as it is...
Now post some pictures after trying again!

ORYON January 31, 2013 11:11

ok, so the total volume of the Fluid( heel+Feed) will increase if i'm defined the top of the tank as pressure outlet?

besides that,by referring to figure 6, there is a horizontal path from inlet reach the surface of the heel( blue-green-yellow). But why this kind of path didn't occur in my simulation(figure 5).

RodriguezFatz January 31, 2013 11:22

Maybe because your pressure-inlet was defined incorrectly.

ORYON February 2, 2013 09:06

hi, after redefined the inlet as velocity inlet, the same problem also occur and the contour is similar to figure 2 which is volume faction of feed is 0 within the tank.That means that the feed is no filled into tank. Why this occur?


All times are GMT -4. The time now is 17:42.