CFD Online Logo CFD Online URL
www.cfd-online.com
[Sponsors]
Home > Forums > Software User Forums > ANSYS > FLUENT

MRF, 3d Centrufugal Fan Analysis in FLUENT

Register Blogs Members List Search Today's Posts Mark Forums Read

Like Tree3Likes
  • 1 Post By jpando
  • 1 Post By jpando
  • 1 Post By jpando

Reply
 
LinkBack Thread Tools Search this Thread Display Modes
Old   December 16, 2014, 13:30
Cool MRF, 3d Centrufugal Fan Analysis in FLUENT
  #1
New Member
 
alireza esmaily
Join Date: Dec 2014
Posts: 18
Rep Power: 11
kaktus10 is on a distinguished road
Hi Friends
Now i'm working on a centrifugal fan.(same as photos)
I've a lot of questions. i need help.
I know i should use MRF method (and Steady-State), but it doesn't work right (= I can't reach to right solution).

my questions:
1- meshing? (in detail pls.)
2- cell zones? (in detail pls.)
3- boundary area?
4- solution methods?
5-...

however, i tried to solve a 2d,mrf,fan problem, but i couldn't.



kaktus10 is offline   Reply With Quote

Old   December 16, 2014, 16:14
Default
  #2
New Member
 
Join Date: May 2012
Posts: 26
Rep Power: 13
jpando is on a distinguished road
Quote:
Originally Posted by kaktus10 View Post
Hi Friends
Now i'm working on a centrifugal fan.(same as photos)
I've a lot of questions. i need help.
I know i should use MRF method (and Steady-State), but it doesn't work right (= I can't reach to right solution).

my questions:
1- meshing? (in detail pls.)
2- cell zones? (in detail pls.)
3- boundary area?
4- solution methods?
5-...

however, i tried to solve a 2d,mrf,fan problem, but i couldn't.
First off, MRF is known to be the least accurate method for a fan. It is accepted for general engineering problems and is typically used as the quicked way to get a solution. If you have time, I suggest you use a sliding mesh approach. It is more accurate but you will have to sacrifice more computational time.

2D models are ill-advised unless you have symmetric flow. Even then, they don't always solve correctly.

If you are only looking at the flow, then use a coarse enough mesh to obtain a y+ close to 1 on your fan surface for enhanced wall functions.
kaktus10 likes this.
jpando is offline   Reply With Quote

Old   December 16, 2014, 16:31
Default
  #3
New Member
 
alireza esmaily
Join Date: Dec 2014
Posts: 18
Rep Power: 11
kaktus10 is on a distinguished road
Now i'm using mrf method.
There is a problem. The interfaces between two zone dosen't work right? What operation should I do?
Thanks
kaktus10 is offline   Reply With Quote

Old   December 16, 2014, 16:57
Default
  #4
New Member
 
Join Date: May 2012
Posts: 26
Rep Power: 13
jpando is on a distinguished road
Quote:
Originally Posted by kaktus10 View Post
Now i'm using mrf method.
There is a problem. The interfaces between two zone dosen't work right? What operation should I do?
Thanks
The interface between the zones should be cylindrical. You have the make sure your mesh follows the cylinder surface very closely. Also If the two regions contrain different size cells, there is a maximum ratio of 5:1 for a fluent interface. When you create the interface you also have to select the smaller size surface as your first interface.

Since you are looking for steady state values, you can initialize your sliding mesh with mrf results if you go that route.
kaktus10 likes this.
jpando is offline   Reply With Quote

Old   December 17, 2014, 00:55
Question
  #5
New Member
 
alireza esmaily
Join Date: Dec 2014
Posts: 18
Rep Power: 11
kaktus10 is on a distinguished road
Quote:
Originally Posted by jpando View Post
The interface between the zones should be cylindrical. You have the make sure your mesh follows the cylinder surface very closely. Also If the two regions contrain different size cells, there is a maximum ratio of 5:1 for a fluent interface. When you create the interface you also have to select the smaller size surface as your first interface.

Since you are looking for steady state values, you can initialize your sliding mesh with mrf results if you go that route.
Thanks jpando

I'll solve a very simple problem at first. please see my result.
1- what's your mean about ratio? (mesh ratio at interfaces or size of cell zones?)
2- OK -> smaller size is interface1 and bigger is interface2.
3- How can i use the sliding mesh with mrf? (what it means?)
Thanks
kaktus10 is offline   Reply With Quote

Old   December 17, 2014, 00:57
Default
  #6
New Member
 
alireza esmaily
Join Date: Dec 2014
Posts: 18
Rep Power: 11
kaktus10 is on a distinguished road
In this result ,the dissconection between two zone is very clear.
There is no inlet & outlet. Inner zone has Frame Rotation. Outer zone is stationary area. fluid is Air.
Rotational Velocity is -2500rpm.

kaktus10 is offline   Reply With Quote

Old   December 17, 2014, 01:03
Default
  #7
New Member
 
alireza esmaily
Join Date: Dec 2014
Posts: 18
Rep Power: 11
kaktus10 is on a distinguished road
In this result ,the dissconection between two zone is very clear.
kaktus10 is offline   Reply With Quote

Old   December 17, 2014, 05:57
Default
  #8
New Member
 
alireza esmaily
Join Date: Dec 2014
Posts: 18
Rep Power: 11
kaktus10 is on a distinguished road
kaktus10 is offline   Reply With Quote

Old   December 17, 2014, 05:57
Default
  #9
New Member
 
alireza esmaily
Join Date: Dec 2014
Posts: 18
Rep Power: 11
kaktus10 is on a distinguished road
Please Help
kaktus10 is offline   Reply With Quote

Old   December 17, 2014, 09:39
Default
  #10
New Member
 
Join Date: May 2012
Posts: 26
Rep Power: 13
jpando is on a distinguished road
Unfortunately I am unable to see you pictures. I think it has someting to do with the network I am on.

The ratio I am talking about is the cell size ratio on the surface of the interface. See section 5.4.2. Non-Conformal Interface Algorithm in the fluent user's guide. They give very good details on what is required for an interface.

So the sliding mesh is setup when you have two separate zones. The rotating cell zone is set with frame motion for MRF and mesh motion for sliding. Mesh motion requires a transient calculation to move the mesh. Start off with very small time steps and increase until each time step turns the cell zone about 1 degree (0.017453293 radians). You can increase the step size up to about 2 degrees if the model is behaving well at 1 degree.

From what you are saying,
"In this result ,the dissconection between two zone is very clear.
There is no inlet & outlet. Inner zone has Frame Rotation."

It sounds like the interface is not set as an "interface" in the boundary conditions. I can only guess since I don't see the picture.
kaktus10 likes this.
jpando is offline   Reply With Quote


Old   December 18, 2014, 09:37
Default
  #12
New Member
 
Join Date: May 2012
Posts: 26
Rep Power: 13
jpando is on a distinguished road
From what I see, the biggest issue is that you are trying to do this problem in 2-dimensions. There is no symmetry in this problem where you can ignore the third dimension. I recommend rebuilding your mesh in 3D and considering all the geometry.
jpando is offline   Reply With Quote

Old   December 22, 2014, 03:11
Default
  #13
New Member
 
alireza esmaily
Join Date: Dec 2014
Posts: 18
Rep Power: 11
kaktus10 is on a distinguished road
Hi friends,
I solve the problem in 2d and now i'have some problems:

1-At the output in center of the output line there is reverse flow.
2-Flow Rate at the output(outlet) is very low.
3-Flow Rate at the input(inlet) is Zero!!!

photo of solution:
https://www.dropbox.com/s/jfuchg3mqz...sults.PNG?dl=0
kaktus10 is offline   Reply With Quote

Old   December 22, 2014, 03:27
Default
  #14
New Member
 
alireza esmaily
Join Date: Dec 2014
Posts: 18
Rep Power: 11
kaktus10 is on a distinguished road
ok my friends, i found the problem.

the input velocity magnitude in boundary conditions was zero. Amazing mistake!
kaktus10 is offline   Reply With Quote

Reply

Thread Tools Search this Thread
Search this Thread:

Advanced Search
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are On
Refbacks are On


Similar Threads
Thread Thread Starter Forum Replies Last Post
how to do analysis for centrifugal air fan using MRF? mrtkrs FLUENT 0 October 13, 2014 04:52
MRF Solver fan simulation - problem sivakumar FLUENT 0 February 15, 2013 04:26
2D Fluent CFD Analysis lamboram FLUENT 0 July 12, 2011 14:02
Fan Modeling in Fluent L. Hamid FLUENT 0 May 24, 2011 03:22
flow simulation across a small fan jane luo Main CFD Forum 15 April 12, 2004 18:49


All times are GMT -4. The time now is 14:29.