Mixing fans in a container. Advice needed (Fan Outlet??)

Ukarthur246 · June 3, 2018, 19:44

Hi all,
I have to attempt a simulation that is beyond my experience level, so I am trying to get other more experienced thoughts and ideas before I start.

The situation is a container with 2 inlets one being air and the other another gas, mixing in the container and then exiting via 2 axial fans blowing out of the container.

I need to check both fans are receiving the same amount of gas.

Heres my questions:
How would people model this??

Do I need to model the fans and use a sliding mesh method, or is there an alternative method??

If I just used one inlet (air) and had no mixing. Is there a way again to check each fan receives the same amount of air without modelling the fans??

All replies are appreciated

Gert-Jan · June 4, 2018, 04:49

You say you have 2 inlets. One with air, and one with gas. And you need to check if both fan will obtain the same amount of gas. But air is also a gas. What do you mean exactly. Can I assume your are speaking about air and natural gas and that you want the same fuel-air ration for both fans?

Nevertheless, it is difficult to say without knowing the geometry. The flow towards the fans can be very complicated. If a fan is in a corner and anti rotation baffles are absent, then you might get a strong swirl and the flow will be less than the other, because the fans will operate on a different working point. So, we need more info.

Also, what is present downstream the fans? Are the outlet pressures identical?

Ukarthur246 · June 4, 2018, 05:46

Thanks Gert-Jan

Yes you are correct in thinking one is air and the other natural gas and I want to ensure the same amount of fuel reaches each fan.

The geometry is simple at the moment with a square container with a circular air inlet at the bottom with known flow rate and the fuel inlet perpendicular to this half way up the container.

In simple terms there are no baffles, and the 2 fans are located symmetrically along the mid plane at the top edge at about a 45degree angle to the vertical i.e. facing away from the container.

They simply blow the fuel air mixture out into the atmosphere. Ive tried to show it below in a basic diagram.

2x Fan out
.........\\___//
.........|........|
.....IN=.......|
........ |__.__|
............ | |
IN air

Gert-Jan · June 4, 2018, 06:52

It all depend on the velocity distribution inside the container. If the inlet flow is significant, then a lot of recirculations will occur, which will infleunce the incoming flow of the fans. If velocities are small, you might get away away with ignoring the fans.

Therefore: you can start a calculation without the rotating fans. Just use 2 pressure outlets. Then, if you notice that the flow to these outlets won't be straight, then rerun the case with the real fans.

Remember as general rule of thumb: if the flow towards a fan, has a swirl in the same direction as the fan rotates, the amount of flow will be more than the operating point, but the head will be lower. And: if the flow towards a fan, swirls in the counter direction, the amount of flow will be less than the operating point, but the head will be higher. So if your setup is symmetrical and the fans are identical and rotate in the same direction, I think you will end up with both fans operating at different set points. Therefore it might be better to implement the fans anyway.

Ukarthur246 · June 5, 2018, 14:11

Would this need to be a transient multiphase simulation??
Also if I know the fan flow rates how does the simulation change??

Gert-Jan · June 5, 2018, 18:09

- No, it will be a transient multicomponent simulation. This is much easier than multiphase.
- I don't understand this question.....

Ukarthur246 · June 5, 2018, 18:29

Thanks for all your replies Gert- Jan I really appreciate it.

The second part I was getting at was that if I know the fans flow rate can I use this as an outflow BC instead of using sliding mesh and modelling the fans completely??

Multicomponent model?? As they are both gases can I simply have two inlets with air coming out of one and fuel the other?? This would simplify the problem massively.
Was I incorrectly assuming whenever two different fluids interact I have to use multiphase, even if they are both gases for example??

Gert-Jan · June 5, 2018, 18:45

Yes, centainly you can set a massflow at your outlet. But then your massflows are fixed and won't be an answer coming from your CFD-calculation. If you can live with that, then that is completely fine.

As long as you use air and natural gas, you can describe this using a multicomponent model in a single phase (=gas). You only use multiphase if you add a liquid or a solid to your air. Like oil or charcoal. Or if you have two inmiscible fluids like oil and water. But that is far beyond your problem. You are a lucky person.

Ukarthur246 · June 5, 2018, 18:54

So if Iam purely interested in the quantity of each component i.e. fuel and air I can set the outflow flow rate??

What if the second inlet was a steam or vapour type inlet can I still use a multicomponent model??

Also is a multicomponent model a model you have to choose like multiphase or species transport for example??

Gert-Jan · June 5, 2018, 19:06

1) Yes. Provided you can neglect the density differences between both streams in case they differ in composition.
2) As long as it is dry steam, then both are still gas, so no problem.
3) No. There is nothing to choose. You have to setup your own mutlicomponent mixture. But there is definitely a gas combustion example/tutorial. So why not use that as a basis? Can't get any better start than that.

Ukarthur246 · June 5, 2018, 19:20

Thank you so much you may have just saved the day, I felt like I was getting nowhere with this problem.

Just to check, setting up my own multicomponent mixture still alows two inlets with the two different gases??

Finally where are these tutorials the fluent theory guide or somewhere else??

Gert-Jan · June 6, 2018, 02:58

Tutorials can be found at the ANSYS customer support site.

Alternatively ask the Fluent Forum. Probably someone can share a pdf. Or the CFX forum if you want to use that software. If you want ot use fluent, I cannot help you any further. WIth CFX I can.

Regarding the inlet conditions. you have to define two components: Air and gas. Then on inlet 1 you set massfraction 1 and 0, and on the other 0 and 1.

June 3, 2018, 19:44	Mixing fans in a container. Advice needed (Fan Outlet??)	#1
Ukarthur246 New Member Arthur Join Date: Jun 2018 Posts: 8 Rep Power: 7	Hi all, I have to attempt a simulation that is beyond my experience level, so I am trying to get other more experienced thoughts and ideas before I start. The situation is a container with 2 inlets one being air and the other another gas, mixing in the container and then exiting via 2 axial fans blowing out of the container. I need to check both fans are receiving the same amount of gas. Heres my questions: How would people model this?? Do I need to model the fans and use a sliding mesh method, or is there an alternative method?? If I just used one inlet (air) and had no mixing. Is there a way again to check each fan receives the same amount of air without modelling the fans?? All replies are appreciated

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
outlet pressure Boundary settings -velocity streamline under ambient temp.conditions	Vishnu_bharathi	CFX	12	November 21, 2017 06:56
Mixing of fluids in a container with a rotor	crippled	ANSYS	0	April 17, 2014 06:49
Advice for fan cooling needed	CK	CFX	3	January 20, 2005 16:53
flow simulation across a small fan	jane luo	Main CFD Forum	15	April 12, 2004 17:49
advice needed!	Filip	Main CFD Forum	2	April 26, 2002 15:27

June 4, 2018, 04:49		#2
Gert-Jan Senior Member Gert-Jan Join Date: Oct 2012 Location: Europe Posts: 1,833 Rep Power: 27	You say you have 2 inlets. One with air, and one with gas. And you need to check if both fan will obtain the same amount of gas. But air is also a gas. What do you mean exactly. Can I assume your are speaking about air and natural gas and that you want the same fuel-air ration for both fans? Nevertheless, it is difficult to say without knowing the geometry. The flow towards the fans can be very complicated. If a fan is in a corner and anti rotation baffles are absent, then you might get a strong swirl and the flow will be less than the other, because the fans will operate on a different working point. So, we need more info. Also, what is present downstream the fans? Are the outlet pressures identical?

June 4, 2018, 05:46		#3
Ukarthur246 New Member Arthur Join Date: Jun 2018 Posts: 8 Rep Power: 7	Thanks Gert-Jan Yes you are correct in thinking one is air and the other natural gas and I want to ensure the same amount of fuel reaches each fan. The geometry is simple at the moment with a square container with a circular air inlet at the bottom with known flow rate and the fuel inlet perpendicular to this half way up the container. In simple terms there are no baffles, and the 2 fans are located symmetrically along the mid plane at the top edge at about a 45degree angle to the vertical i.e. facing away from the container. They simply blow the fuel air mixture out into the atmosphere. Ive tried to show it below in a basic diagram. 2x Fan out .........\\___// .........\|........\| .....IN=.......\| ........ \|__.__\| ............ \| \| IN air

June 4, 2018, 06:52		#4
Gert-Jan Senior Member Gert-Jan Join Date: Oct 2012 Location: Europe Posts: 1,833 Rep Power: 27	It all depend on the velocity distribution inside the container. If the inlet flow is significant, then a lot of recirculations will occur, which will infleunce the incoming flow of the fans. If velocities are small, you might get away away with ignoring the fans. Therefore: you can start a calculation without the rotating fans. Just use 2 pressure outlets. Then, if you notice that the flow to these outlets won't be straight, then rerun the case with the real fans. Remember as general rule of thumb: if the flow towards a fan, has a swirl in the same direction as the fan rotates, the amount of flow will be more than the operating point, but the head will be lower. And: if the flow towards a fan, swirls in the counter direction, the amount of flow will be less than the operating point, but the head will be higher. So if your setup is symmetrical and the fans are identical and rotate in the same direction, I think you will end up with both fans operating at different set points. Therefore it might be better to implement the fans anyway.

June 5, 2018, 14:11		#5
Ukarthur246 New Member Arthur Join Date: Jun 2018 Posts: 8 Rep Power: 7	Would this need to be a transient multiphase simulation?? Also if I know the fan flow rates how does the simulation change??

June 5, 2018, 18:09		#6
Gert-Jan Senior Member Gert-Jan Join Date: Oct 2012 Location: Europe Posts: 1,833 Rep Power: 27	- No, it will be a transient multicomponent simulation. This is much easier than multiphase. - I don't understand this question.....

June 5, 2018, 18:29		#7
Ukarthur246 New Member Arthur Join Date: Jun 2018 Posts: 8 Rep Power: 7	Thanks for all your replies Gert- Jan I really appreciate it. The second part I was getting at was that if I know the fans flow rate can I use this as an outflow BC instead of using sliding mesh and modelling the fans completely?? Multicomponent model?? As they are both gases can I simply have two inlets with air coming out of one and fuel the other?? This would simplify the problem massively. Was I incorrectly assuming whenever two different fluids interact I have to use multiphase, even if they are both gases for example??

June 5, 2018, 18:45		#8
Gert-Jan Senior Member Gert-Jan Join Date: Oct 2012 Location: Europe Posts: 1,833 Rep Power: 27	Yes, centainly you can set a massflow at your outlet. But then your massflows are fixed and won't be an answer coming from your CFD-calculation. If you can live with that, then that is completely fine. As long as you use air and natural gas, you can describe this using a multicomponent model in a single phase (=gas). You only use multiphase if you add a liquid or a solid to your air. Like oil or charcoal. Or if you have two inmiscible fluids like oil and water. But that is far beyond your problem. You are a lucky person.

June 5, 2018, 18:54		#9
Ukarthur246 New Member Arthur Join Date: Jun 2018 Posts: 8 Rep Power: 7	So if Iam purely interested in the quantity of each component i.e. fuel and air I can set the outflow flow rate?? What if the second inlet was a steam or vapour type inlet can I still use a multicomponent model?? Also is a multicomponent model a model you have to choose like multiphase or species transport for example??

June 5, 2018, 19:06		#10
Gert-Jan Senior Member Gert-Jan Join Date: Oct 2012 Location: Europe Posts: 1,833 Rep Power: 27	1) Yes. Provided you can neglect the density differences between both streams in case they differ in composition. 2) As long as it is dry steam, then both are still gas, so no problem. 3) No. There is nothing to choose. You have to setup your own mutlicomponent mixture. But there is definitely a gas combustion example/tutorial. So why not use that as a basis? Can't get any better start than that.

June 5, 2018, 19:20		#11
Ukarthur246 New Member Arthur Join Date: Jun 2018 Posts: 8 Rep Power: 7	Thank you so much you may have just saved the day, I felt like I was getting nowhere with this problem. Just to check, setting up my own multicomponent mixture still alows two inlets with the two different gases?? Finally where are these tutorials the fluent theory guide or somewhere else??

June 6, 2018, 02:58		#12
Gert-Jan Senior Member Gert-Jan Join Date: Oct 2012 Location: Europe Posts: 1,833 Rep Power: 27	Tutorials can be found at the ANSYS customer support site. Alternatively ask the Fluent Forum. Probably someone can share a pdf. Or the CFX forum if you want to use that software. If you want ot use fluent, I cannot help you any further. WIth CFX I can. Regarding the inlet conditions. you have to define two components: Air and gas. Then on inlet 1 you set massfraction 1 and 0, and on the other 0 and 1.