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

study of wings (stall)

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

Reply
 
LinkBack Thread Tools Display Modes
Old   June 21, 2014, 13:10
Arrow study of wings (stall)
  #1
New Member
 
Daniel Ocaña
Join Date: Jun 2014
Posts: 9
Rep Power: 2
danuco is on a distinguished road
Hi all,
I am currently doing my final degree project to finish my studies in aeroespace engineering. I have to study several wings in ANSYS-FLUENT aerodynamically and to verify the obtained results with experimental results.
I am having a lot of problems because in my simulations, the stall appears before that it has to be. For example, the stall appears to 12º angle of attack in the experimental data and it appears to 8º angle of attack in my simulation.
any advice about what can be happening??
It is urgent, I have to submit my project in less than a month!

Sorry for my English
danuco is offline   Reply With Quote

Old   June 21, 2014, 15:27
Default
  #2
New Member
 
Nick Prendergast
Join Date: Dec 2013
Posts: 27
Rep Power: 2
neprendo is on a distinguished road
Hi

I might be able to help, but I have a couple questions.
1. what turbulence model and wall treatment settings are you using.
2. what is the Y+ value for your simulations.
can you also upload a picture of the mesh you are using

Nick
neprendo is offline   Reply With Quote

Old   June 21, 2014, 19:27
Default
  #3
New Member
 
Daniel Ocaña
Join Date: Jun 2014
Posts: 9
Rep Power: 2
danuco is on a distinguished road
Quote:
Originally Posted by neprendo View Post
Hi

I might be able to help, but I have a couple questions.
1. what turbulence model and wall treatment settings are you using.
2. what is the Y+ value for your simulations.
can you also upload a picture of the mesh you are using

Nick
I am using k-e realizable with enhanced wall treatment with pressure gradient effects, I have used the model constants by default.
I have done an inflation in the wing surfaces with a first layer thickness of 0.005 mm, I have achieved y+<1 in the simulations.
I have attached the mesh pictures.
Thanks in advance
Attached Images
File Type: jpg 1.jpg (89.0 KB, 19 views)
File Type: jpg 2.jpg (102.9 KB, 16 views)
File Type: jpg 3.jpg (89.5 KB, 14 views)
File Type: jpg 4.jpg (51.1 KB, 10 views)
danuco is offline   Reply With Quote

Old   June 22, 2014, 03:32
Default
  #4
New Member
 
Nick Prendergast
Join Date: Dec 2013
Posts: 27
Rep Power: 2
neprendo is on a distinguished road
You're already doing what I would have suggested. The only thing I could say is that while you have convergence you might have converged on the wrong answer, your mesh seems fine, so there could be something slightly wrong in your solver settings. Have you tried using a reynold's stress model?
neprendo is offline   Reply With Quote

Old   June 22, 2014, 06:58
Default
  #5
New Member
 
Daniel Ocaña
Join Date: Jun 2014
Posts: 9
Rep Power: 2
danuco is on a distinguished road
Quote:
Originally Posted by neprendo View Post
You're already doing what I would have suggested. The only thing I could say is that while you have convergence you might have converged on the wrong answer, your mesh seems fine, so there could be something slightly wrong in your solver settings. Have you tried using a reynold's stress model?
No, I have never used this turbulence model.
Right now, I am running a case with this model but I am very unexperienced in it, What configuration do you recommend me? I am using the configuration that you can see in the attached image with K or Turbulent Intensity in the Reynolds-Stress Method in inlet and outlet (B.C)
In the inlet and outlet boundary conditions, I am using 0.05% turbulent intensity and 0.001m turbulent length scale, Could theses values be correct?
Many Thanks!!
Attached Images
File Type: jpg 5.jpg (79.0 KB, 16 views)
danuco is offline   Reply With Quote

Old   June 22, 2014, 11:14
Default
  #6
New Member
 
Nick Prendergast
Join Date: Dec 2013
Posts: 27
Rep Power: 2
neprendo is on a distinguished road
ye the default settings should be ok, I only suggest this model as a way to check your K-e results have converged on the right answer. The reynolds stress model is the most accurate RANS model available but its also the most computationally expensive, It might be worth running in your case. If you are wondering about any of the settings in it, I recommend checking out the Ansys fluent theory guide. It's a monster but it will explain everything much better than myself
neprendo is offline   Reply With Quote

Old   June 22, 2014, 13:19
Default
  #7
New Member
 
Daniel Ocaña
Join Date: Jun 2014
Posts: 9
Rep Power: 2
danuco is on a distinguished road
Quote:
Originally Posted by neprendo View Post
ye the default settings should be ok, I only suggest this model as a way to check your K-e results have converged on the right answer. The reynolds stress model is the most accurate RANS model available but its also the most computationally expensive, It might be worth running in your case. If you are wondering about any of the settings in it, I recommend checking out the Ansys fluent theory guide. It's a monster but it will explain everything much better than myself
I have tried to apply this model but around of 100 iterations the solution diverge.... I have used very low solution controls.... What can the reason be?
danuco is offline   Reply With Quote

Old   June 22, 2014, 13:35
Default
  #8
New Member
 
Nick Prendergast
Join Date: Dec 2013
Posts: 27
Rep Power: 2
neprendo is on a distinguished road
I don't know, try decreasing the turbulent length scale to roughly %5 of your channel height if it isn't already
neprendo is offline   Reply With Quote

Old   June 22, 2014, 13:57
Default
  #9
Member
 
Haris Hameed
Join Date: Oct 2009
Posts: 37
Rep Power: 7
harishameed33 is on a distinguished road
hi

your mesh is not fine enough near the wall to capture the stall... you extruded the layers but these layers suddenly meet the coarser unstructured mesh. turbulence model type will effect the results but not that much. your difference is very large, so something is wrong in your computational model. i suggest you refine your mesh near the wall and make it coarse in the farfield.... even SA model will produce good results
harishameed33 is offline   Reply With Quote

Old   June 22, 2014, 14:07
Default
  #10
Member
 
Haris Hameed
Join Date: Oct 2009
Posts: 37
Rep Power: 7
harishameed33 is on a distinguished road
here is some model mesh for you

hope this will be helpfull
Attached Images
File Type: jpg 4u.jpg (89.6 KB, 26 views)
harishameed33 is offline   Reply With Quote

Old   June 22, 2014, 14:30
Default
  #11
New Member
 
Daniel Ocaña
Join Date: Jun 2014
Posts: 9
Rep Power: 2
danuco is on a distinguished road
Quote:
Originally Posted by harishameed33 View Post
here is some model mesh for you

hope this will be helpfull
Many thanks for your advises, really thyy are a great help.
I will try to get that kind of mesh but I don´t know if I will be able to get it, strutured meshes 3D in Ansys Meshing are really difficult to achieve
danuco is offline   Reply With Quote

Old   June 23, 2014, 07:30
Default
  #12
New Member
 
Daniel Ocaña
Join Date: Jun 2014
Posts: 9
Rep Power: 2
danuco is on a distinguished road
please, can anyone give me any advices to get an suitable mesh in an easy way in Anys Meshing????
danuco is offline   Reply With Quote

Old   June 23, 2014, 09:02
Default
  #13
Member
 
Haris Hameed
Join Date: Oct 2009
Posts: 37
Rep Power: 7
harishameed33 is on a distinguished road
are you using ICEM form meshing???? is it a compulsion??? there are other meshing tools like gridgen, gmabit and pointwise that helps to create hybrid-unstructured mesh of good quality....
harishameed33 is offline   Reply With Quote

Old   June 23, 2014, 09:48
Default
  #14
New Member
 
Daniel Ocaña
Join Date: Jun 2014
Posts: 9
Rep Power: 2
danuco is on a distinguished road
Quote:
Originally Posted by harishameed33 View Post
are you using ICEM form meshing???? is it a compulsion??? there are other meshing tools like gridgen, gmabit and pointwise that helps to create hybrid-unstructured mesh of good quality....
No, I am using the meshing tool available in the workbench, What meshing software do you recommend me?
danuco is offline   Reply With Quote

Old   June 23, 2014, 11:55
Default
  #15
Member
 
Haris Hameed
Join Date: Oct 2009
Posts: 37
Rep Power: 7
harishameed33 is on a distinguished road
i have no idea regarding the meshing tool of workbench for CFD.... i dont know how much control it gives to the user....
all the other meshing tools are equally good.... each has its own pros n cons...

i use pointwise to create structured and hybrid-unstructured mesh....its T-Rex function is very good for viscous meshing. but i dont know weather it will be easy for you or not. since you have to do this work urgently...
harishameed33 is offline   Reply With Quote

Old   June 24, 2014, 11:12
Default
  #16
Senior Member
 
Gonzalo
Join Date: Mar 2011
Location: Argentina
Posts: 102
Rep Power: 6
gfoam is on a distinguished road
Hi, have you tried using SST-kw turbulence model? It is a goog model to predict adverse pressure gradient boundary layers and the stall phenomena. But, are you sure thar the result in the paper or experimental results you have are corrected for blockage effects and are for free stream conditions? Regards.
Gonzalo
gfoam is offline   Reply With Quote

Old   June 24, 2014, 18:35
Default
  #17
Senior Member
 
Lefteris
Join Date: Oct 2011
Location: UK, Greece
Posts: 186
Rep Power: 5
Aeronautics El. K. is on a distinguished road
Haris and Gonzalo are both right in their comments. However, I find it strange that in your CFD you get the stall earlier rather than later from what the experimental data indicate.
How many points do you have in your boundary layer mesh?
Also, the mesh at the trailing edge region doesn't seem to be very good.
__________________
Lefteris

Aeronautics El. K. is offline   Reply With Quote

Old   June 25, 2014, 04:27
Default
  #18
New Member
 
Daniel Ocaña
Join Date: Jun 2014
Posts: 9
Rep Power: 2
danuco is on a distinguished road
Quote:
Originally Posted by Aeronautics El. K. View Post
Haris and Gonzalo are both right in their comments. However, I find it strange that in your CFD you get the stall earlier rather than later from what the experimental data indicate.
How many points do you have in your boundary layer mesh?
Also, the mesh at the trailing edge region doesn't seem to be very good.
yes, definitely my problem is the mesh, I am learning ICEM , it seems me imposible to get a good strutured mesh in Meshing....
danuco 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
Study of wings in Ansys Fluent danuco ANSYS 0 June 21, 2014 12:41
Prediction of stall point cfdeng CFX 1 February 21, 2013 16:30
static stall atrh Main CFD Forum 1 March 9, 2004 06:07
Looking for experiments in dynamic stall Anton Lyaskin Main CFD Forum 0 February 10, 2003 03:57
stall yin Main CFD Forum 1 May 10, 2000 03:03


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