boundary condition for air-water container

Shamoon Jamshed · January 1, 2022, 08:38

Hello friends,

I have a container filled with water. The air is filled at the top at 5 bar. The air is pushing water from the outlet just below the tank. What boundary condition should I used at outlet? Pressure outlet, outlflow or mass flow inlet with negative vector. In fact, for the third one, I should use positive vector since the downward direction is +ve x axis.

Regards,

SJ

LuckyTran · January 1, 2022, 17:50

It depends... what process you are trying to simulate. If this is like a tank filled with air-water at 5bar and I poke a hole in it and want to see how fast it drains, then I would use a pressure outlet. Because when I poke the hole in the bottom of the tank, it exposes it to the ambient pressure and the pressure difference drives the flow. If I have a tank hooked up to a vacuum pump then I might still use a pressure outlet if I knew the vaccuum pressure, but I might also use an outflow or massflow outlet.

With regards to the negative vector, if you specify a vector then you specify the vector. Negative magnitude doesn't really mean anything for a vector. If you specify the correct vector then you have the correct vector. If you put a negative magnitude, then you have to put in the opposite direction.

The reason for the negative business with a massflow inlet is because you specify a massflow rate in kg/s with an implicit assumption of an inward surface normal. If you specify your own direction vector then you don't have to play with this negative business.

Shamoon Jamshed · January 1, 2022, 21:23

Thanks Lucky
The thing is that I modeled a 3d sector piece (30 deg). I want to have achieved a mass flow rate at outlet = 26 kg/s. But with pressure outlet I am getting 0.006 kg/s (very very low). So that is why I want to put implicitly a mass flow boundary with this value. I think the second approach will still not give me correct value.

Shamoon Jamshed · January 1, 2022, 21:26

Thanks Lucky
The thing is that I modeled a 3d sector piece (30 deg). I want to have achieved a mass flow rate at outlet = 26 kg/s. But with pressure outlet I am getting 0.006 kg/s (very very low). So that is why I want to put implicitly a mass flow boundary with this value. I think the second approach will still not give me correct value.

LuckyTran · January 1, 2022, 21:51

Back to the example of a tank with a hole in it. The rate that it drains depends on the pressure head available in the tank and the size of the hole. If you're not getting enough mass flow then your hole isn't big enough.

Like I said, it just depends on what you are trying to simulate. Just saying I have a tank with water-air in it is an incomplete description of the problem. You've given us a set of partial differential equations and half the boundary conditions and asking what the other half of the BC's should be. Well, there is no general answer to that because you are the one that determines what the BCs are. Now if you give me complete physical description of the problem, I can assist you with what the remaining BCs ought to be. However, you've only given a tank with water in it and nothing else. It could be anything! Heck, I could put a pump and force massflow into the tank if I wanted. It's up to you to specify these things.

Now if you tell me you have a tank with water in it and want water to flow out of it at 26 kg/s (we can of course pretend that the hand of god somehow forces the water in the tank to magically flow out at this rate) then I will tell you to use pressure outlet with target mass flow rate option as a BC. Because if you use any other BC, then the flowrate will be not be imposed but free to be whatever it ought to be.

Shamoon Jamshed · January 2, 2022, 02:14

Ok, actually in real I have helium and kerosene. The kerosene is being pressureized through helium from top which have a pressure of 5 bar. Now I am using VoF model in fluent to see the vortex phenomenon from drain. Of course there will be strong vortices appreaing in the bottom, so we will design a vortex separator. For that we need to run a 3d simulation (even without any vortex breaker), to see the vortex effect and intensity. The information is for pressure of helium and the drain mass flow outlet, that the kerosene is being going out at 26 kg/s. The outlet drain is 200 mm.

Regards,

SJ

Shamoon Jamshed · January 2, 2022, 02:22

And also please let me know if the Fluent should show the tank fully occupied with Helium? Initially, He is 3 percent of the tank volume.

LuckyTran · January 2, 2022, 04:14

But this drain vents to atmosphere or what!? It's no fun guessing at what you mean.

If you use a massflow outlet then of course everything will be drained from the tank eventually.

200mm is not an area, I don't know what that means. I guess I have to just assume it makes sense.

Shamoon Jamshed · January 2, 2022, 05:17

the outlet is a pipe to which a pump is connected. The pipe is 6 m long.(from the outlet of the tank to the pump inlet). 200 mm is the dia of the outlet. What do you mean by I see the mass flow outlet? If I put it intentially , then of course the whole kerosene will vanish from the tank but, for the pressure outlet as boundary condition, the mass flow is not coming right, that is the trouble. from your answer, I think I should implicitly force mass flow boundary condition at the tank outlet since the flow is not draining into the atmosphere

Shamoon Jamshed · January 2, 2022, 05:43

Also, please let me know which fluid should be primary and which should be secondary, since some tutorials take water (being liquid) as primary but some tutorials dont bother

LuckyTran · January 3, 2022, 01:55

Well if you have a pump then it would actually make sense to impose some kind of flowrate at the outlet. And if you did use a pressure outlet, then you need to put the correct value for the outlet pressure.

It doesn't matter which is the primary or secondary. There's numerical precision reasons for one over the other (truncation errors) but either should work. If you are not convinced then just simply switch them and see for yourself. It takes no effort to do this.

Shamoon Jamshed · January 3, 2022, 08:00

Now somethings are making sense, this kind of problem is a bit new. What I did now that I impose a mass-flow-outlet boundary at outlet with phase -1 (kerosene). And a pressure inlet at the inlet with Vof of phase -2 (helium) as 1. Rest 0. Operating pressure is now at 5 bar. I am not imposing any operating density or gravity. Are these now a complete setup? I hope so. what do you say?

January 1, 2022, 08:38	boundary condition for air-water container	#1
Shamoon Jamshed Senior Member Shamoon Jamshed Join Date: Apr 2009 Location: Karachi Posts: 377 Rep Power: 18	Hello friends, I have a container filled with water. The air is filled at the top at 5 bar. The air is pushing water from the outlet just below the tank. What boundary condition should I used at outlet? Pressure outlet, outlflow or mass flow inlet with negative vector. In fact, for the third one, I should use positive vector since the downward direction is +ve x axis. Regards, SJ

January 1, 2022, 21:51		#5
LuckyTran Senior Member Lucky Join Date: Apr 2011 Location: Orlando, FL USA Posts: 5,674 Rep Power: 66	Back to the example of a tank with a hole in it. The rate that it drains depends on the pressure head available in the tank and the size of the hole. If you're not getting enough mass flow then your hole isn't big enough. Like I said, it just depends on what you are trying to simulate. Just saying I have a tank with water-air in it is an incomplete description of the problem. You've given us a set of partial differential equations and half the boundary conditions and asking what the other half of the BC's should be. Well, there is no general answer to that because you are the one that determines what the BCs are. Now if you give me complete physical description of the problem, I can assist you with what the remaining BCs ought to be. However, you've only given a tank with water in it and nothing else. It could be anything! Heck, I could put a pump and force massflow into the tank if I wanted. It's up to you to specify these things. Now if you tell me you have a tank with water in it and want water to flow out of it at 26 kg/s (we can of course pretend that the hand of god somehow forces the water in the tank to magically flow out at this rate) then I will tell you to use pressure outlet with target mass flow rate option as a BC. Because if you use any other BC, then the flowrate will be not be imposed but free to be whatever it ought to be. Last edited by LuckyTran; January 2, 2022 at 04:09.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Multiphase flow - incorrect velocity on inlet	Mike_Tom	CFX	6	September 29, 2016 01:27
several fields modified by single boundary condition	schröder	OpenFOAM Programming & Development	3	April 21, 2015 05:09
Radiation interface	hinca	CFX	15	January 26, 2014 17:11
Question about heat transfer coefficient setting for CFX	Anna Tian	CFX	1	June 16, 2013 06:28
inlet velocity boundary condition	murali	CFX	5	August 3, 2012 08:56

January 1, 2022, 17:50		#2
LuckyTran Senior Member Lucky Join Date: Apr 2011 Location: Orlando, FL USA Posts: 5,674 Rep Power: 66	It depends... what process you are trying to simulate. If this is like a tank filled with air-water at 5bar and I poke a hole in it and want to see how fast it drains, then I would use a pressure outlet. Because when I poke the hole in the bottom of the tank, it exposes it to the ambient pressure and the pressure difference drives the flow. If I have a tank hooked up to a vacuum pump then I might still use a pressure outlet if I knew the vaccuum pressure, but I might also use an outflow or massflow outlet. With regards to the negative vector, if you specify a vector then you specify the vector. Negative magnitude doesn't really mean anything for a vector. If you specify the correct vector then you have the correct vector. If you put a negative magnitude, then you have to put in the opposite direction. The reason for the negative business with a massflow inlet is because you specify a massflow rate in kg/s with an implicit assumption of an inward surface normal. If you specify your own direction vector then you don't have to play with this negative business.

January 1, 2022, 21:23		#3
Shamoon Jamshed Senior Member Shamoon Jamshed Join Date: Apr 2009 Location: Karachi Posts: 377 Rep Power: 18	Thanks Lucky The thing is that I modeled a 3d sector piece (30 deg). I want to have achieved a mass flow rate at outlet = 26 kg/s. But with pressure outlet I am getting 0.006 kg/s (very very low). So that is why I want to put implicitly a mass flow boundary with this value. I think the second approach will still not give me correct value.

January 1, 2022, 21:26		#4
Shamoon Jamshed Senior Member Shamoon Jamshed Join Date: Apr 2009 Location: Karachi Posts: 377 Rep Power: 18	Thanks Lucky The thing is that I modeled a 3d sector piece (30 deg). I want to have achieved a mass flow rate at outlet = 26 kg/s. But with pressure outlet I am getting 0.006 kg/s (very very low). So that is why I want to put implicitly a mass flow boundary with this value. I think the second approach will still not give me correct value.

January 2, 2022, 02:14		#6
Shamoon Jamshed Senior Member Shamoon Jamshed Join Date: Apr 2009 Location: Karachi Posts: 377 Rep Power: 18	Ok, actually in real I have helium and kerosene. The kerosene is being pressureized through helium from top which have a pressure of 5 bar. Now I am using VoF model in fluent to see the vortex phenomenon from drain. Of course there will be strong vortices appreaing in the bottom, so we will design a vortex separator. For that we need to run a 3d simulation (even without any vortex breaker), to see the vortex effect and intensity. The information is for pressure of helium and the drain mass flow outlet, that the kerosene is being going out at 26 kg/s. The outlet drain is 200 mm. Regards, SJ

January 2, 2022, 02:22		#7
Shamoon Jamshed Senior Member Shamoon Jamshed Join Date: Apr 2009 Location: Karachi Posts: 377 Rep Power: 18	And also please let me know if the Fluent should show the tank fully occupied with Helium? Initially, He is 3 percent of the tank volume.

January 2, 2022, 04:14		#8
LuckyTran Senior Member Lucky Join Date: Apr 2011 Location: Orlando, FL USA Posts: 5,674 Rep Power: 66	But this drain vents to atmosphere or what!? It's no fun guessing at what you mean. If you use a massflow outlet then of course everything will be drained from the tank eventually. 200mm is not an area, I don't know what that means. I guess I have to just assume it makes sense.

January 2, 2022, 05:17		#9
Shamoon Jamshed Senior Member Shamoon Jamshed Join Date: Apr 2009 Location: Karachi Posts: 377 Rep Power: 18	the outlet is a pipe to which a pump is connected. The pipe is 6 m long.(from the outlet of the tank to the pump inlet). 200 mm is the dia of the outlet. What do you mean by I see the mass flow outlet? If I put it intentially , then of course the whole kerosene will vanish from the tank but, for the pressure outlet as boundary condition, the mass flow is not coming right, that is the trouble. from your answer, I think I should implicitly force mass flow boundary condition at the tank outlet since the flow is not draining into the atmosphere

January 2, 2022, 05:43		#10
Shamoon Jamshed Senior Member Shamoon Jamshed Join Date: Apr 2009 Location: Karachi Posts: 377 Rep Power: 18	Also, please let me know which fluid should be primary and which should be secondary, since some tutorials take water (being liquid) as primary but some tutorials dont bother

January 3, 2022, 01:55		#11
LuckyTran Senior Member Lucky Join Date: Apr 2011 Location: Orlando, FL USA Posts: 5,674 Rep Power: 66	Well if you have a pump then it would actually make sense to impose some kind of flowrate at the outlet. And if you did use a pressure outlet, then you need to put the correct value for the outlet pressure. It doesn't matter which is the primary or secondary. There's numerical precision reasons for one over the other (truncation errors) but either should work. If you are not convinced then just simply switch them and see for yourself. It takes no effort to do this.

January 3, 2022, 08:00		#12
Shamoon Jamshed Senior Member Shamoon Jamshed Join Date: Apr 2009 Location: Karachi Posts: 377 Rep Power: 18	Now somethings are making sense, this kind of problem is a bit new. What I did now that I impose a mass-flow-outlet boundary at outlet with phase -1 (kerosene). And a pressure inlet at the inlet with Vof of phase -2 (helium) as 1. Rest 0. Operating pressure is now at 5 bar. I am not imposing any operating density or gravity. Are these now a complete setup? I hope so. what do you say?