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

when I change the position of the arfiole cfx not converge?

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

Like Tree4Likes
  • 1 Post By ghorrocks
  • 1 Post By Lance
  • 1 Post By Lance
  • 1 Post By ghorrocks

Reply
 
LinkBack Thread Tools Display Modes
Old   October 12, 2013, 07:31
Default when I change the position of the arfiole cfx not converge?
  #1
Member
 
wadoud
Join Date: Nov 2010
Posts: 50
Rep Power: 5
wadoud31 is on a distinguished road
hi evry one
I want to study the influence of several vanned distributor opening, so I changed the position of the airfoile, for the first case I got the solution, but when I changed the angle for the other cases the cfx did not converge, although I kept the same mesh (hex type , bloking the same) and the same boundary conditions.
Here the picture
best regards
thank you.
Attached Images
File Type: jpg BLADE.jpg (26.0 KB, 13 views)
wadoud31 is offline   Reply With Quote

Old   October 13, 2013, 01:25
Default
  #2
Super Moderator
 
Glenn Horrocks
Join Date: Mar 2009
Location: Sydney, Australia
Posts: 9,242
Rep Power: 74
ghorrocks has a spectacular aura aboutghorrocks has a spectacular aura aboutghorrocks has a spectacular aura about
Did you generate new meshes for each position, or use something else?

It could be the first condition is attached flow and the simulation converges easily, but the others are separated flow and convergence is harder in this case it crashes. If this is the case you will need to improve mesh quality, smaller timesteps, double precision or a better initial condition.
wadoud31 likes this.
ghorrocks is offline   Reply With Quote

Old   October 14, 2013, 05:26
Default
  #3
Member
 
wadoud
Join Date: Nov 2010
Posts: 50
Rep Power: 5
wadoud31 is on a distinguished road
thank you for reply
for the first case with a 13 angle I got the solution for several boundary conditions after 1500 itirations, but when I changed the angle of the blade either 15,18,20 or 25, after 3000 itiritions the residual target is higher than 0.001 and the convergence critiria is 0.00001
to remedy this
first I kept the same blocking and I associated the edges and vertices, after I checked the quality of the mesh.
second I used a new hex mesh (new blocking).
I kept the same boundary conditions.
but the problem persists.
if you have any ideas or solutions to this problem
I am very grateful for your help
wadoud31 is offline   Reply With Quote

Old   October 14, 2013, 05:40
Default
  #4
Senior Member
 
Lance
Join Date: Mar 2009
Posts: 472
Rep Power: 10
Lance is on a distinguished road
So it doesnt converge when you increase the angle of attack? You probably get separation, and as Glenn said:
Quote:
Originally Posted by ghorrocks View Post
It could be the first condition is attached flow and the simulation converges easily, but the others are separated flow and convergence is harder in this case it crashes. If this is the case you will need to improve mesh quality, smaller timesteps, double precision or a better initial condition.
wadoud31 likes this.
Lance is offline   Reply With Quote

Old   October 14, 2013, 05:53
Default
  #5
Member
 
wadoud
Join Date: Nov 2010
Posts: 50
Rep Power: 5
wadoud31 is on a distinguished road
thank you for reply
how can I eliminate this problem to obtain the solution in the different positions of the blades
best regards
wadoud31 is offline   Reply With Quote

Old   October 14, 2013, 05:56
Default
  #6
Senior Member
 
Lance
Join Date: Mar 2009
Posts: 472
Rep Power: 10
Lance is on a distinguished road
i'll quote it again.
Quote:
Originally Posted by ghorrocks View Post
If this is the case you will need to improve mesh quality, smaller timesteps, double precision or a better initial condition.
wadoud31 likes this.
Lance is offline   Reply With Quote

Old   October 17, 2013, 05:25
Default
  #7
Member
 
wadoud
Join Date: Nov 2010
Posts: 50
Rep Power: 5
wadoud31 is on a distinguished road
thank you for reply and I am very grateful;
I took your advice into consideration so I changed the mesh, timestep, double precision, I kept the same boundary conditions as the first case, but the problem persists. if you need more details I will disponoble.
best regards
wadoud31 is offline   Reply With Quote

Old   October 17, 2013, 05:34
Default
  #8
Super Moderator
 
Glenn Horrocks
Join Date: Mar 2009
Location: Sydney, Australia
Posts: 9,242
Rep Power: 74
ghorrocks has a spectacular aura aboutghorrocks has a spectacular aura aboutghorrocks has a spectacular aura about
Have a look in the literature for what other people have done for these simulations. Have a look at the meshes they use, the timesteps they need and the turbulence models they assumed.

If it is still not converging you need an even better quality mesh and/or even smaller timesteps.

The FAQ has more details: http://www.cfd-online.com/Wiki/Ansys...gence_criteria
ghorrocks is offline   Reply With Quote

Old   October 19, 2013, 05:19
Default
  #9
Member
 
wadoud
Join Date: Nov 2010
Posts: 50
Rep Power: 5
wadoud31 is on a distinguished road
hi
thank you for your help and kindness
here some details
boundary conditions. (inlet): total pressure: 1.6258 bar, total temperature: 333K
(output of the rotor): Static pressure: 0
(wall) adiabatic
turbulence model used: RNGK-e
physical timescale: 3e-06 [s]
I used a hex mesh
-the height of the first cell was 0.1.
to remedy this
I kept the same boundary conditions
I have increased the number of nodes to 687375
-physical timescale becomes 1e-08 [s]
-the height of the first cell becomes 0.05
-Double-precision
-I check the quality of the mesh (see picture)
I am very grateful for your help
Attached Images
File Type: jpg cfd.JPG (29.6 KB, 7 views)
wadoud31 is offline   Reply With Quote

Old   October 20, 2013, 20:16
Default
  #10
Super Moderator
 
Glenn Horrocks
Join Date: Mar 2009
Location: Sydney, Australia
Posts: 9,242
Rep Power: 74
ghorrocks has a spectacular aura aboutghorrocks has a spectacular aura aboutghorrocks has a spectacular aura about
You are reporting just one mesh quality metric - and not a very important one at that.

The important ones are aspect ratio, volume ratio, face angle.

My post #8 still lists the important things for you to consider.
ghorrocks is offline   Reply With Quote

Old   October 23, 2013, 05:18
Default
  #11
Member
 
wadoud
Join Date: Nov 2010
Posts: 50
Rep Power: 5
wadoud31 is on a distinguished road
thank you for your quick and accurate response.
concerning the quality of the mesh I attached a picture and I hope that you'll clear up the thing
Attached Images
File Type: jpg mesh.jpg (87.4 KB, 2 views)
wadoud31 is offline   Reply With Quote

Old   October 23, 2013, 07:11
Default
  #12
Super Moderator
 
Glenn Horrocks
Join Date: Mar 2009
Location: Sydney, Australia
Posts: 9,242
Rep Power: 74
ghorrocks has a spectacular aura aboutghorrocks has a spectacular aura aboutghorrocks has a spectacular aura about
Are you doing this steady state or transient? Separated airfoil flow is hard to model steady, so you might need a transient approach, and there is a chance you will need DES/SAS/LES.
wadoud31 likes this.
ghorrocks is offline   Reply With Quote

Old   October 26, 2013, 04:39
Default
  #13
Member
 
wadoud
Join Date: Nov 2010
Posts: 50
Rep Power: 5
wadoud31 is on a distinguished road
hi
thank you for your help and kindnes
i used steady state;
best regards
wadoud31 is offline   Reply With Quote

Reply

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
Comparison of fluent and CFX for turbomachinery Far CFX 49 April 3, 2012 21:08
Pros and Cons for CFX, CFdesign, COMSOL Val Main CFD Forum 3 June 10, 2011 02:20
ATTENTION! Reliability problems in CFX 5.7 Joseph CFX 14 April 20, 2010 15:45
Phase Change in CFX derrek CFX 0 June 9, 2003 18:27
CFX 4.4 installation problem Pandu Sattvika CFX 1 December 1, 2001 04:07


All times are GMT -4. The time now is 07:53.