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

Ansys 12.1 - Create Rotor/Stator Interface for Fluent

Register Blogs Community New Posts Updated Threads Search

Like Tree5Likes

Reply
 
LinkBack Thread Tools Search this Thread Display Modes
Old   November 20, 2010, 14:54
Default Ansys 12.1 - Create Rotor/Stator Interface for Fluent
  #1
opm
New Member
 
Join Date: Nov 2010
Posts: 2
Rep Power: 0
opm is on a distinguished road
Hi,

I try to setup a simple 2D sliding mesh problem (see appended picture of the mesh) with Workbench and fail at properly creating the interface between the rotating and stationary parts. I already searched through the documentation, this forum and the internet for the last two days and didn't find a solution. Maybe I just fail to recognize what the problem really is. So here is what I did so far:
  • I created a geometry in Catia consisting of two circular rotating areas with "blade-dummies" and a rectangular stationary area.
  • I formed a new part out of the three surface bodies in DM ("Form New Part" command) to get a conformal mesh.
  • I named all the boundaries and fluid zones. That's the step where I assume my problem starts. I tried to created interface-"Named Selections" for both surface edges at the connection between the stationary and the rotating part. Ansys than throws an error at me "The mesh file exporter does not support overlapping geometry in named selections. ..."
  • If I just leave it that way, start Fluent, setup the problem and start the calculation, the mesh degenerates after a while and I get negative cell volumes (see the two appended figures, before and after the calculation).
  • I tried to make a wall Named Selection and rename it to interior in Fluent like it is described in the "Multiple Rotating Reference Frame" tutorial but had the same results.
Does anyone know how to properly setup the problem in WB DM and Meshing to be able to use the "Mesh Interface" dialog in Fluent?

Thank you in advance.
opm
Attached Images
File Type: png SM_2D_Mesh.png (67.9 KB, 375 views)
File Type: png SM_2D_NamedSelections.png (38.9 KB, 264 views)
File Type: png SM_2D_Mesh_Case_01.png (66.4 KB, 227 views)
File Type: png SM_2D_Mesh_Case_01_after_calculation.png (69.1 KB, 209 views)
opm is offline   Reply With Quote

Old   November 22, 2010, 09:40
Default
  #2
opm
New Member
 
Join Date: Nov 2010
Posts: 2
Rep Power: 0
opm is on a distinguished road
Hi again,

I just tried something else. If i skip the following step

> [...]
> I formed a new part out of the three surface bodies in DM
> ("Form New Part" command) to get a conformal mesh.

> [...]

I am able to create Named Selections for both surface edges (rotating and stationary zones). Now I am able to use the interface dialog in Fluent without any further problems and the calculation is running like expected.

Now the question is how to get this working with a conformal mesh. Any ideas?

Greetings,
opm
opm is offline   Reply With Quote

Old   November 24, 2010, 13:04
Default Line Sizing.
  #3
Senior Member
 
PSYMN's Avatar
 
Simon Pereira
Join Date: Mar 2009
Location: Ann Arbor, MI
Posts: 2,663
Blog Entries: 1
Rep Power: 47
PSYMN has a spectacular aura aboutPSYMN has a spectacular aura about
You don't actually need a conformal mesh (shared nodes), you just want an aligned mesh. You can get this by setting a line sizing the same on both sides of the circle. It should work perfectly. (right click on mesh => Insert Sizing, then select the lines...

Another option that I have not tried yet for 2D curve sizing is the "Match" method. I would have used that if this were a 3D problem.
PSYMN is offline   Reply With Quote

Old   March 17, 2011, 10:13
Default
  #4
New Member
 
Andreas
Join Date: Feb 2010
Posts: 6
Rep Power: 16
tallknuseren is on a distinguished road
As PSYMN mentioned, you don't even need a conformal mesh to create a mesh interface, interfaces are for connecting domains who touch at non-conformal faces or edges. Create named selections for the touching geometry at both touching bodies, redefine those to "Interface" at boundary conditions, and create interfaces at "Mesh interface".

I'm struggling with rotor stator simulation also. My problem is that there is a jump in both pressure and velocity across the interfaces, but the mass flux seems to match. Unfortunately, I can't show the whole geometry of this problem, but I attach a screenshot of some part of it. Clearly, you can see the jump in velocity at the countour plot. I should mention, that the flow passes through one stator, further through a rotor and then into the last stator. So there is actually two interfaces.
Attached Images
File Type: jpg rotor_stator.jpg (44.5 KB, 376 views)
tallknuseren is offline   Reply With Quote

Old   March 26, 2011, 19:38
Default
  #5
Senior Member
 
PSYMN's Avatar
 
Simon Pereira
Join Date: Mar 2009
Location: Ann Arbor, MI
Posts: 2,663
Blog Entries: 1
Rep Power: 47
PSYMN has a spectacular aura aboutPSYMN has a spectacular aura about
That looks like more than a typical interface jump... Are you sure it is just an internal wall? How about the fluid regions? Do they all have the same properties (viscosity, etc?)
__________________
-----------------------------------------
Please help guide development at ANSYS by filling in these surveys

Public ANSYS ICEM CFD Users Survey

This second one is more general (Gambit, TGrid and ANSYS Meshing users welcome)...

CFD Online Users Survey
PSYMN is offline   Reply With Quote

Old   April 20, 2012, 10:11
Default
  #6
RR2
New Member
 
Join Date: Apr 2012
Posts: 6
Rep Power: 14
RR2 is on a distinguished road
I no this thread is a little bit old now, but i have the same problem as meantioned above. I am doing a 3D CFD analysis on a centrifugal pump with an open impeller. I have two domains, one containging the fluid in the impeller (its a little bit larger than the impeller diameter, only to avoid an interface were the blade ends), and one containing the rest. When defining the interface between them i got some problems.

I have tried both ways described above. When I used the "form new part" command in Designmodeler (to get a conformal mesh) and defined the two interface surfaces as named selections in ANSYS meshing i got an error (the same as the thread starter got). I then tried to use the mesh sizing command in Meshing, and since I am working in 3D I defined this as a face sizing. This didnt change the solution at all, i still got interface jump between the cell zones. You guys also meantioned the match command in Meshing, and I have tried it but don't understand how it works.

As you said above I dont need a conformal mesh to avoid this jump, so any help concering the other options would be appreciated!

In FLUENT I define the interface using the mesh interface command, with default settings (i.e. no periodic boundary condition or coupled wall), so maybe there is some problems there to?

Thanks!

/RR2
RR2 is offline   Reply With Quote

Old   April 23, 2012, 06:15
Default
  #7
RR2
New Member
 
Join Date: Apr 2012
Posts: 6
Rep Power: 14
RR2 is on a distinguished road
Anyone? Appreciate all input i can get!
RR2 is offline   Reply With Quote

Old   June 19, 2013, 03:33
Default
  #8
Member
 
Arthur Loginow
Join Date: Aug 2012
Posts: 99
Rep Power: 14
Maralady is on a distinguished road
RR2 I am having the same issues, somehow I think I am close to the solution, I really think that the problem comes from the solid model, If you want to send me a PM to share information, or if you already solved the problem, please your help will be appreciated, I have opened several threats with a similar subject but I don't get an answer from nobody, and I have not been able to find any documentation that says something about how to create the interfaces at the meshing level.
Maralady is offline   Reply With Quote

Old   April 12, 2016, 15:36
Default
  #9
Senior Member
 
Have a nice time!
Join Date: Feb 2016
Location: mech.eng.ahmedmansour@gmail.com
Posts: 291
Rep Power: 11
Ahmed Saeed Mansour is on a distinguished road
Quote:
Originally Posted by PSYMN View Post
You don't actually need a conformal mesh (shared nodes), you just want an aligned mesh. You can get this by setting a line sizing the same on both sides of the circle. It should work perfectly. (right click on mesh => Insert Sizing, then select the lines...

Another option that I have not tried yet for 2D curve sizing is the "Match" method. I would have used that if this were a 3D problem.
How can i define the interfaces for a sliding mesh 3D pump?
Ahmed Saeed Mansour is offline   Reply With Quote

Old   May 20, 2016, 23:05
Default
  #10
New Member
 
Fadi hajj
Join Date: Feb 2016
Posts: 24
Rep Power: 10
Fadih is on a distinguished road
hello guys,

I am trying to simulate a 3D vertical axis wind turbine, and I have been struggling with overlapping mesh at interfaces and can't seem to find a solution for them. I tried using match control but its not possible because ansys cannot match between 2 bodies even the face selected is shared.
does anyone have any clue how to solve the overlapping mesh.
I have attached pics of my model, the turbine rotates ( hence the big cylinder) also the blade rotate around their center relative to the turbine (hence the 3 inner cylinders)

Is their anything I can do to resolve this.
Please help guys.
Thank you
Attached Images
File Type: png 3.PNG (19.5 KB, 25 views)
File Type: png 4.PNG (31.6 KB, 27 views)
File Type: png 5.PNG (25.8 KB, 24 views)
File Type: png 6.PNG (17.6 KB, 20 views)
Fadih is offline   Reply With Quote

Old   May 21, 2016, 16:27
Default
  #11
Senior Member
 
Have a nice time!
Join Date: Feb 2016
Location: mech.eng.ahmedmansour@gmail.com
Posts: 291
Rep Power: 11
Ahmed Saeed Mansour is on a distinguished road
Dear Fadi, I got the same problem when I tried to generate a mesh for a 3D impeller of a pump. I had an overlapping mesh problem. After watching a specific video on youtube about sharing topology, the mesh had not any problem. Try to share topology of your 3D design because at the case of a 3D assembly, there is a chance for many meshing problems.
https://www.youtube.com/watch?v=4JhDn0V26l0

https://www.youtube.com/results?sear...hared+topology
Ahmed Saeed Mansour is offline   Reply With Quote

Old   May 21, 2016, 16:35
Default
  #12
New Member
 
Fadi hajj
Join Date: Feb 2016
Posts: 24
Rep Power: 10
Fadih is on a distinguished road
Dear ahmed,
thank you for replying, by sharing topology i am letting ansys know that these 2 are in contact without the need to define contact regions but would i still be able to rotate the fluid volumes? any idea on that end?
Fadih is offline   Reply With Quote

Old   May 21, 2016, 18:18
Default
  #13
Senior Member
 
Have a nice time!
Join Date: Feb 2016
Location: mech.eng.ahmedmansour@gmail.com
Posts: 291
Rep Power: 11
Ahmed Saeed Mansour is on a distinguished road
Try all the possible solutions because I didn't try the 3D before...So, try to share topology of all the parts of the turbine to become one part and the mesh should be around this part...after this, define the rotor domain and the stator domains..keep trying and I hope this works.
Ahmed Saeed Mansour is offline   Reply With Quote

Old   May 21, 2016, 18:38
Default
  #14
Senior Member
 
Have a nice time!
Join Date: Feb 2016
Location: mech.eng.ahmedmansour@gmail.com
Posts: 291
Rep Power: 11
Ahmed Saeed Mansour is on a distinguished road
Check out this tutorial.
http://www.mediafire.com/download/e8....5_L07_SMM.pdf
Ahmed Saeed Mansour is offline   Reply With Quote

Old   May 21, 2016, 18:56
Default
  #15
New Member
 
Fadi hajj
Join Date: Feb 2016
Posts: 24
Rep Power: 10
Fadih is on a distinguished road
thanks alot
Fadih is offline   Reply With Quote

Old   May 21, 2016, 19:00
Default
  #16
Senior Member
 
Have a nice time!
Join Date: Feb 2016
Location: mech.eng.ahmedmansour@gmail.com
Posts: 291
Rep Power: 11
Ahmed Saeed Mansour is on a distinguished road
Not at all, Is there any problem now?....you can check out the mixing tank tutorial on youtube. You can treat the turbine as a one part with a cylinder( rotating domain) and the box will be the static domain and no need for 3 cylinders around each airfoil.
Ahmed Saeed Mansour is offline   Reply With Quote

Old   May 21, 2016, 19:06
Default
  #17
New Member
 
Fadi hajj
Join Date: Feb 2016
Posts: 24
Rep Power: 10
Fadih is on a distinguished road
i have done the 2D case and works perfectly, its just the mesh for the 3D model is tricky, i have done shared topolgy as we speak now waiting for fluent to read the case to check if the mesj is overlapping
Fadih is offline   Reply With Quote

Old   May 21, 2016, 19:08
Default
  #18
New Member
 
Fadi hajj
Join Date: Feb 2016
Posts: 24
Rep Power: 10
Fadih is on a distinguished road
there is no overlap anymore i just need to se if fluent still understands that there are interfaces and its not just one part
Fadih is offline   Reply With Quote

Old   May 21, 2016, 20:50
Default
  #19
Senior Member
 
Have a nice time!
Join Date: Feb 2016
Location: mech.eng.ahmedmansour@gmail.com
Posts: 291
Rep Power: 11
Ahmed Saeed Mansour is on a distinguished road
Check out these tutorials

http://www.mediafire.com/download/k7...ry_14.5_v1.zip
Aeroelastico likes this.
Ahmed Saeed Mansour is offline   Reply With Quote

Old   May 21, 2016, 21:28
Default
  #20
New Member
 
Fadi hajj
Join Date: Feb 2016
Posts: 24
Rep Power: 10
Fadih is on a distinguished road
thank you for all your help, greatly appreciated
Fadih is offline   Reply With Quote

Reply


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
Evaporation model in ansys Fluent 12.1 oldisbest Fluent UDF and Scheme Programming 12 March 26, 2020 10:11
Coupled walls - Ansys Fluent 12.1 Geisel FLUENT 2 May 4, 2010 06:21
Actuator disk model audrich FLUENT 0 September 21, 2009 08:06
Where's the singularity/mesh flaw? audrich FLUENT 3 August 4, 2009 02:07
fluent add additional zones for the mesh file SSL FLUENT 2 January 26, 2008 12:55


All times are GMT -4. The time now is 10:53.