CFD Online Logo CFD Online URL
www.cfd-online.com
[Sponsors]
Home > Forums > CFX

oscillating airfoil interface problem

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

Reply
 
LinkBack Thread Tools Display Modes
Old   November 28, 2012, 15:42
Default oscillating airfoil interface problem
  #1
New Member
 
Join Date: Nov 2012
Location: Québec
Posts: 7
Rep Power: 5
etgau8 is on a distinguished road
Hi everyone,
I’m modeling a 2D oscillating airfoil NACA0025 in a pitching and heaving motion at a Re number of 500000. I’m running the simulation with different programs such as CFX, Fluent and StarCCM+. I got problem running my model with CFX. But I got good results with the other programs. The pitching motion is done with a sliding mesh and an interface transient rotor-stator. The heaving motion is set by a moving frame (variable inlet velocity function) and a Source term for the acceleration of the frame.
My first problem is that in CFX Solver, the monitoring of my force coefficient on the foil during the simulation gave me bad values (such as Cx<0). But in CFX Post, when I calculate the same expression I got the values that I’m looking for.
The second problem and the biger one is that I got noise at my interface (transient rotor-stator) as you can see in the pictures(the last one pitchingheaving04.png is at the end of the first cycle).
I tried to find a solution to my problems but I didn’t resolve them. I did try to use a mesh that has the same type of element on both side of the interface but the results are the same.
Thank you for your time and I hope that you can have an idea of a solution.

Etienne
Attached Images
File Type: png pitchingheaving03.PNG (38.5 KB, 86 views)
File Type: png pitchingheaving04.PNG (56.5 KB, 68 views)
etgau8 is offline   Reply With Quote

Old   November 28, 2012, 16:28
Default
  #2
Senior Member
 
Bruno
Join Date: Mar 2009
Location: Brazil
Posts: 272
Rep Power: 13
brunoc is on a distinguished road
Are you setting any type of domain rotation in the domain with the airfoil? If not, you don't need to use a Transient Rotor-Stator interface. Use a regular interface instead and use the 6-DOF solver built into CFX (it's called Rigid Body Solver) to control the domain motion. You should get good results from those.

There is a tutorial that shows how to use the 6-DOF solver. Take a look at it.
brunoc is offline   Reply With Quote

Old   November 28, 2012, 16:45
Default
  #3
New Member
 
Join Date: Nov 2012
Location: Québec
Posts: 7
Rep Power: 5
etgau8 is on a distinguished road
Hi Bruno, thank you for your quick reply!

Yes my domain containing the airfoil is rotating in a pitching motion, that's why I'm using a transient rotor-stator interface. I have used this type of interface in turbomachinery simulations in the past and didn't had any problems. I have aslo try to run a simulation only in pitching, without the heaving motion frame and without the momentum source term. Even in this case, I got a little bit of noise at the interface on my contours of pressure and vorticity. Any ideas what can be the problem?
etgau8 is offline   Reply With Quote

Old   November 28, 2012, 17:20
Default
  #4
Super Moderator
 
Glenn Horrocks
Join Date: Mar 2009
Location: Sydney, Australia
Posts: 12,287
Rep Power: 96
ghorrocks is a jewel in the roughghorrocks is a jewel in the roughghorrocks is a jewel in the rough
Try the Alternate Rotation Model for the rotating frame of reference. Worth a try.
ghorrocks is offline   Reply With Quote

Old   November 28, 2012, 18:13
Default
  #5
Senior Member
 
Chris DeGroot
Join Date: Nov 2011
Location: Canada
Posts: 388
Rep Power: 7
cdegroot is on a distinguished road
I assume those images were made in CFD Post... Are you using hybrid or conservative interface values? Sometimes things can look a little funny if you use conservative instead of hybrid. Just something to consider.
cdegroot is offline   Reply With Quote

Old   November 29, 2012, 15:40
Default
  #6
New Member
 
Join Date: Nov 2012
Location: Québec
Posts: 7
Rep Power: 5
etgau8 is on a distinguished road
Thank you all for your interest in my problem!

To Glen: I tried the alternate rotation model but it didn't solve the problem. I didn't see any changes in the physic fiels.

To Chris: The images are in CFX Post, and the setting is set to hybrid.

Any other ideas?

I post you a image of the pressure contour at a time in the cycle when I got important pressure signals at the interface.
Attached Images
File Type: png Pressure_contour.PNG (20.9 KB, 30 views)
etgau8 is offline   Reply With Quote

Old   November 29, 2012, 17:03
Default
  #7
Super Moderator
 
Glenn Horrocks
Join Date: Mar 2009
Location: Sydney, Australia
Posts: 12,287
Rep Power: 96
ghorrocks is a jewel in the roughghorrocks is a jewel in the roughghorrocks is a jewel in the rough
Quote:
I tried the alternate rotation model but it didn't solve the problem.
OK, worth a try.

Are you doing the rotation with a rotating frame of reference or a moving mesh?
ghorrocks is offline   Reply With Quote

Old   November 29, 2012, 19:27
Default
  #8
New Member
 
Join Date: Nov 2012
Location: Québec
Posts: 7
Rep Power: 5
etgau8 is on a distinguished road
Quote:
Originally Posted by ghorrocks View Post
OK, worth a try.

Are you doing the rotation with a rotating frame of reference or a moving mesh?
I'm using a moving mesh (sliding mesh) for the pitching motion and a reference frame for the heaving motion.

I'm posting a .txt file containing all my settings if you want to take a look.

Thank you!
Attached Files
File Type: txt NACA0025_V8_alternaterotmodel_002.txt (13.9 KB, 42 views)
etgau8 is offline   Reply With Quote

Old   November 29, 2012, 19:35
Default
  #9
Super Moderator
 
Glenn Horrocks
Join Date: Mar 2009
Location: Sydney, Australia
Posts: 12,287
Rep Power: 96
ghorrocks is a jewel in the roughghorrocks is a jewel in the roughghorrocks is a jewel in the rough
No, you are not using moving mesh. In CFX these are different things. Rotating frame of reference allows a domain to rotate. Moving mesh allows arbitrary mesh motion. You have the foil in a rotating frame of reference.

I am not certain of this but I suspect the domain acceleration is not accounted for in rotating frame of reference. This means the missing acceleration terms can cause spurious flows - like what you are seeing.

You might need to do this model using the moving mesh (the real moving mesh approach that is ), that is to specifically define the rotating motion. This will be a considerably slower simulation, but I suspect it will take care of the acceleration terms better.
ghorrocks is offline   Reply With Quote

Old   April 1, 2014, 14:04
Default
  #10
Member
 
Doug Hunsaker
Join Date: Mar 2009
Location: Logan, UT
Posts: 63
Rep Power: 9
doug is on a distinguished road
Etgau8,

Could you explain how you set this simulation up in StarCCM? I'm trying to do something very similar, but I'm not convinced that I've set it up properly. I'm using a User-Defined Vertex Option and specifying a linear displacement as a function of time for my overset region.

Thanks!
doug is offline   Reply With Quote

Old   April 1, 2014, 14:38
Default
  #11
New Member
 
Join Date: Nov 2012
Location: Québec
Posts: 7
Rep Power: 5
etgau8 is on a distinguished road
Quote:
Originally Posted by doug View Post
Etgau8,

Could you explain how you set this simulation up in StarCCM? I'm trying to do something very similar, but I'm not convinced that I've set it up properly. I'm using a User-Defined Vertex Option and specifying a linear displacement as a function of time for my overset region.

Thanks!
Hi doug,

I was using CFX and Fluent in the past, but now I am working with StarCCM+ since the set up of multiple moving bodies is quite easy using the Overset Mesh approach. You don't need to set up a User-Defined Vertex Option to specify a linear deplacement. These are steps that you can follow:

1- You need an overset region and a background region.

2- You have to create a new local coordinate system
(Tools/Coordinate System/Laboratory, right-click on local coordinate system to create a new Cartesian.

3- Create a motion
In Tools/Motions (right-click on motion->new Translation). Specify the Laboratory system as the Coordiante system and the Cartesian 1 system as the Managed coordinate system. The translation velocity can be define as follow [0, sin($Time), 0] for a vertical sinusoidal plunging motion.

4- Then, you have to specify the motion in the settings of your regions.
Regions/OversetRegion/PhysicsValues/Motion Specification, set motion to translation and the reference frame to Lab reference frame.
Regions/BackgroundRegion/PhysicsValues/Motion Specification, set motion to Stationary and the reference frame to Lab reference frame.

If you would like to add a rotating motion to get an oscillating body, (Translation/Superposing Motions) right click New and you set up the rotating motion. You will have to change the option in your overset region to Translation->Superposing Rotation.

Hope this helps!
etgau8 is offline   Reply With Quote

Reply

Tags
cfx airfoil interface

Thread Tools
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 On
Pingbacks are On
Refbacks are On


Similar Threads
Thread Thread Starter Forum Replies Last Post
Oscillating Airfoil - Dynamic meshing or user-defined velocity profile. DarrenC Main CFD Forum 5 July 19, 2010 22:33
Plunging Airfoil Problem jmo OpenFOAM Running, Solving & CFD 0 May 17, 2010 22:36
Interface problem Tomislav FLUENT 4 May 23, 2006 00:49
Oscillating airfoil problem ganesh Main CFD Forum 2 June 27, 2005 13:57
Convective Heat Transfer - Heat Exchanger Mark CFX 6 November 15, 2004 16:55


All times are GMT -4. The time now is 00:11.