[Eli]

Hi there!
I'm trying to simulate a simple "tank discharging" problem. There is a hole at the bottom of the tank as an outlet. I wonder what the best Boundary Condition is for outlet!
The hydrostatic pressure can be a good prediction, but it deponds on water height in the tank that is decreasing by time!
How can I set such a B.C that varies with time and need to be updated every timestep???
Thank you very much

[Tristan]

If the tank is discharging then you should use the ambient pressure that it is discharging to as the outlet boundary condition. The hydrostatic pressure doesn't make sense since the fluid is discharging and therefore not static.

Tristan

[Eli]

Thank you Tristan for your answer!
Well, somthing that makes the fluid discharge is the weight of it above the outlet, isn't it? Does CFX understand it when we use ambient pressure?

Thanks again

[Tristan]

I'd never noticed that there is no gravity checkbox available in CFX. Can you add a momentum source to introduce the force due to gravity?

Tristan

[Eli]

of course CFX has gravity option! when you activate Buoyancy Force, it means that you are including gravity in your simulation.
but the point is:

"the pressure calculated by the solver when buoyancy is activated, excludes

the hydrostatic pressure gradient"

anyway, can anyone help me set a boundary condition using some variables from transient result?

[AliTr]

I assume you have a tank full of water with an outlet at the bottom. and model is a multiphase Air/Water (buoyancy activated).
you need two boundaries.
First is an open boundary at the top (tank roof) set to Pres & Dirn, 0 Kpa. Air volume Fraction=1, Water Volume fraction=0. this will ensure air ingress when water leaves the tank.

Second boundary is similar to the fist one but at your outlet. This will ensure gravity discharge into atmospheric pressure.

Make sure that you have enough mesh refinement around the outlet, say at least 8 diagonal nodes.

[Eli]

Dear AliTr
Thanks for your respond
I'm going to check it ,and if it doesnt work out I'll come back sooon!

Eli

[Rui]

Hi,

Have a look at these:
http://www2.dem.uc.pt/antonio.gameir...SEAS_paper.pdf
http://www.wseas.us/e-library/confer...rs/498-704.pdf
http://www.ansys.com/magazine/issues...r-products.pdf
http://www.ansys.com/industries/cons...oiletvalve.htm

... and though it may sounds simple, the discharge of tank is a difficult numerical problem (transient, two-phases, free-surface), and very CPU time consuming. But I'm sure you'll find it by yourself.