Refilling a Tank

Whyman · July 7, 2009, 10:21

Hi everybody!

I would like to propose you a problem that i'm trying to solve, but for now unsuccessfully.
I have a tank and i want to simulate the first refilling phase of it. The tank is pressurized at 3 bar of any gas, like air or helium: at the bottom of the tank there is the inlet surface where i want to introduce a liquid, with a decided flow rate and pressure (the pressure is 1.5 higher than the gas pressure). At the top instead there is another outlet surface where the gas is sucked in order to hold the tank pressure on.
At first sight it would be recommended a compressible multiphase solver, but for now it hasn't been developed yet.
I have tried to run the incompressible multiphase solver but it crashes almost immediately.
Do you know if this simulation is possible in OpenFoam?
Have you got any suggestions to give?

Thank you all
Bye

girogirozakk · July 8, 2009, 00:28

Hi , stefano.

you can find interFoam's sample "fill water" in my URL.
that case may help you ?
http://giropenfoam.web.fc2.com/sample/interFoam/index.html

Bye

Giro

Whyman · July 8, 2009, 04:58

Hi Giro,

thank you very much for your help. It's exactly what i needed. I'm trying to run the simulation right now and i hope to obtain the right results.

Maybe you can answer me to another question. Do you know if it's possible to simulate a spin velocity applied to the domain? I mean like to simulate a rotation velocity. For now i can only apply a costant acceleration, but i would like to see what happens when the domain is rotating with a constant velocity.

Thank you
Bye
Stefano

eugene · July 8, 2009, 06:19

What is wrong with compressibleLESInterFoam?

Whyman · July 8, 2009, 06:31

Nothing wrong, but i would like to use the RANS formulation in order to make the simulation quicker.

eugene · July 8, 2009, 07:44

In that case
- copy the LES version somewhere,
- rename it and replace all instances of "LESModel" with "RASModel" and all instances of "nuSgs2 with "nut".
- Change Make/file to rename the executable and FOAM_APPBIN to FOAM_USER_APPBIN
- Change Make/options and replace LES with RAS here as well. (Except LESdeltas which can be removed)
- Recompile.
- Done.

girogirozakk · July 9, 2009, 03:21

> Do you know if it's possible to simulate a spin velocity applied to the domain? I mean like to simulate a rotation velocity.

now , "inlet" boundary that is constructed with so many "face" , please replace to some inlet boundary(each face) that have controled velocity what you want.
if you write some program(script) for setup , it is not so tiresome.

other approach ,
if i try that type boundary with other approach, may be i make tube-type mesh for inlet.
that tube have spiral wall .
if that tube have enough length and spiral wall height , i can get spin velocity on the real inlet point(may be).
(may be tiresome to adjust...)

that is not so good?

Giro

July 7, 2009, 10:21	Refilling a Tank	#1
Whyman Member Stefano Join Date: Jul 2009 Posts: 36 Rep Power: 16	Hi everybody! I would like to propose you a problem that i'm trying to solve, but for now unsuccessfully. I have a tank and i want to simulate the first refilling phase of it. The tank is pressurized at 3 bar of any gas, like air or helium: at the bottom of the tank there is the inlet surface where i want to introduce a liquid, with a decided flow rate and pressure (the pressure is 1.5 higher than the gas pressure). At the top instead there is another outlet surface where the gas is sucked in order to hold the tank pressure on. At first sight it would be recommended a compressible multiphase solver, but for now it hasn't been developed yet. I have tried to run the incompressible multiphase solver but it crashes almost immediately. Do you know if this simulation is possible in OpenFoam? Have you got any suggestions to give? Thank you all Bye

July 8, 2009, 04:58	It works	#3
Whyman Member Stefano Join Date: Jul 2009 Posts: 36 Rep Power: 16	Hi Giro, thank you very much for your help. It's exactly what i needed. I'm trying to run the simulation right now and i hope to obtain the right results. Maybe you can answer me to another question. Do you know if it's possible to simulate a spin velocity applied to the domain? I mean like to simulate a rotation velocity. For now i can only apply a costant acceleration, but i would like to see what happens when the domain is rotating with a constant velocity. Thank you Bye Stefano

July 8, 2009, 06:31	It works	#5
Whyman Member Stefano Join Date: Jul 2009 Posts: 36 Rep Power: 16	Nothing wrong, but i would like to use the RANS formulation in order to make the simulation quicker.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
how to maintain constant presure in a tank?	JB	FLUENT	5	July 31, 2008 05:12
Settling tank	John Grimm	FLUENT	1	March 20, 2005 08:20
tank leakage	H	FLUENT	0	October 4, 2004 20:28
Boiling and tank fill questions	Mir	CFX	1	April 28, 2002 17:45
Flow patterns in a stirred tank	Glenn Price	Main CFD Forum	13	March 20, 1999 10:16

July 8, 2009, 00:28		#2
girogirozakk New Member Giro Join Date: Mar 2009 Posts: 13 Rep Power: 17	Hi , stefano. you can find interFoam's sample "fill water" in my URL. that case may help you ? http://giropenfoam.web.fc2.com/sample/interFoam/index.html Bye Giro

July 8, 2009, 06:19		#4
eugene Senior Member Eugene de Villiers Join Date: Mar 2009 Posts: 725 Rep Power: 21	What is wrong with compressibleLESInterFoam?

July 8, 2009, 07:44		#6
eugene Senior Member Eugene de Villiers Join Date: Mar 2009 Posts: 725 Rep Power: 21	In that case - copy the LES version somewhere, - rename it and replace all instances of "LESModel" with "RASModel" and all instances of "nuSgs2 with "nut". - Change Make/file to rename the executable and FOAM_APPBIN to FOAM_USER_APPBIN - Change Make/options and replace LES with RAS here as well. (Except LESdeltas which can be removed) - Recompile. - Done.

July 9, 2009, 03:21		#7
girogirozakk New Member Giro Join Date: Mar 2009 Posts: 13 Rep Power: 17	> Do you know if it's possible to simulate a spin velocity applied to the domain? I mean like to simulate a rotation velocity. now , "inlet" boundary that is constructed with so many "face" , please replace to some inlet boundary(each face) that have controled velocity what you want. if you write some program(script) for setup , it is not so tiresome. other approach , if i try that type boundary with other approach, may be i make tube-type mesh for inlet. that tube have spiral wall . if that tube have enough length and spiral wall height , i can get spin velocity on the real inlet point(may be). (may be tiresome to adjust...) that is not so good? Giro