CFD Online Discussion Forums

CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   ANSYS Meshing & Geometry (https://www.cfd-online.com/Forums/ansys-meshing/)
-   -   [ICEM] Could use some guidance with an O-Grid (https://www.cfd-online.com/Forums/ansys-meshing/203109-could-use-some-guidance-o-grid.html)

kepler123 June 18, 2018 10:20

Could use some guidance with an O-Grid
 
1 Attachment(s)
Hi everyone, it's now a week since I began using ICEM CFD, I've watched a lot of tutorials, etc. as there is no one at my university to guide me and I need to use it for my Ph.D. Also this is my first time meshing stuff.

I'm on a tight schedule and I need to complete the meshing by the end of this work week and therefore am asking for help from the experts out there. I have attached all the necessary files to run the project on ICEM CFD 18.1

My aim: Have a C section inlet flow field on a rotor blade (or wing) with a flat trailing edge.

What I have managed to do
:

I imported the geometry, created the flow field and divided the initial block into a few blocks to build a surface mesh on the blade. Watching a tutorial https://www.youtube.com/watch?v=bhpNHLfTGgg and https://www.youtube.com/watch?v=k3ntUDvIXJM I managed to understand the process of blocking, etc. I have even refined the mesh around the leading edge with an O-grid.

The problem:


I have plenty of blocks on the inlet which I would like to merge before creating a new O-grid (good idea?), because as of now I created an O-grid and I'm sure it's done wrong. Alternatively, I could do something like in the video above also for the inlet and not just the leading edge, but due to the many blocks I have due to the C section as opposed to the Rectangular section of the video I do not seem to find a way to get the O-grid to work right.

Also if you zoom into the leading edge, you will see that immediately in front of it, I have a H grid rather than an O-grid after my attempt to form an O-grid.

Steps I followed:
  1. Import of the geometry
  2. Created the flowfield
  3. Made a block around it
  4. Associated the points and edges
  5. Split block in the X-Z plane to be able to associate with the mid plane of the inlet and outlet (due to the C-inlet I had to do this) OR is there another way to do it?
  6. Split the block to surround the wing such that the wing ends up in a block of its own separated from the rest of the flow field.
  7. Had to split up the block the wing is in to associate the edges of the block with the leading edge and trailing edge.
  8. Associated all the necessary points and lines.
  9. Split the block in the vertical plane to create an o-grid on the leading edge (made sure all associations were right before I made the o-grid).
  10. Set up some decent pre-mesh params to get an idea of what the premesh looked like.
  11. Tried to merge blocks on the inlet of the flow field and failed, then selected all the visible edges and made an O-block which I am damn sure is wrong, but I tried it anyway.
  12. Uploaded the result here.
I think I should do this:
  1. After point 10, I should split the inlet fluid zone in a way similar to what I did for the leading edge and create an O-grid using faces?
If yes, how do I make sure that the O-grid does not also lead to skewed cells near the wall? Is there a way to ensure this does not happen or do I just play around with the vertex of the O-grid points until the mesh looks good? This is what I did on the wing as shown in the video, seems good. But I have a feeling this won't work on a larger scale as at some point many cells are going to converge and form some really skewed cells, just like you see right now on the top and bottom walls.


P.S: I know that the section is not symmetric about the mid point in the z direction, stupid me. But I don't see that affecting the process of meshing. I will re-do it once I get the general idea right.

Thanks for any help.

shereez234 June 19, 2018 17:04

Quote:

Originally Posted by kepler123 (Post 696340)
Hi everyone, it's now a week since I began using ICEM CFD, I've watched a lot of tutorials, etc. as there is no one at my university to guide me and I need to use it for my Ph.D. Also this is my first time meshing stuff.

I'm on a tight schedule and I need to complete the meshing by the end of this work week and therefore am asking for help from the experts out there. I have attached all the necessary files to run the project on ICEM CFD 18.1

My aim: Have a C section inlet flow field on a rotor blade (or wing) with a flat trailing edge.

What I have managed to do
:

I imported the geometry, created the flow field and divided the initial block into a few blocks to build a surface mesh on the blade. Watching a tutorial https://www.youtube.com/watch?v=bhpNHLfTGgg and https://www.youtube.com/watch?v=k3ntUDvIXJM I managed to understand the process of blocking, etc. I have even refined the mesh around the leading edge with an O-grid.

The problem:


I have plenty of blocks on the inlet which I would like to merge before creating a new O-grid (good idea?), because as of now I created an O-grid and I'm sure it's done wrong. Alternatively, I could do something like in the video above also for the inlet and not just the leading edge, but due to the many blocks I have due to the C section as opposed to the Rectangular section of the video I do not seem to find a way to get the O-grid to work right.

Also if you zoom into the leading edge, you will see that immediately in front of it, I have a H grid rather than an O-grid after my attempt to form an O-grid.

Steps I followed:
  1. Import of the geometry
  2. Created the flowfield
  3. Made a block around it
  4. Associated the points and edges
  5. Split block in the X-Z plane to be able to associate with the mid plane of the inlet and outlet (due to the C-inlet I had to do this) OR is there another way to do it?
  6. Split the block to surround the wing such that the wing ends up in a block of its own separated from the rest of the flow field.
  7. Had to split up the block the wing is in to associate the edges of the block with the leading edge and trailing edge.
  8. Associated all the necessary points and lines.
  9. Split the block in the vertical plane to create an o-grid on the leading edge (made sure all associations were right before I made the o-grid).
  10. Set up some decent pre-mesh params to get an idea of what the premesh looked like.
  11. Tried to merge blocks on the inlet of the flow field and failed, then selected all the visible edges and made an O-block which I am damn sure is wrong, but I tried it anyway.
  12. Uploaded the result here.
I think I should do this:
  1. After point 10, I should split the inlet fluid zone in a way similar to what I did for the leading edge and create an O-grid using faces?
If yes, how do I make sure that the O-grid does not also lead to skewed cells near the wall? Is there a way to ensure this does not happen or do I just play around with the vertex of the O-grid points until the mesh looks good? This is what I did on the wing as shown in the video, seems good. But I have a feeling this won't work on a larger scale as at some point many cells are going to converge and form some really skewed cells, just like you see right now on the top and bottom walls.


P.S: I know that the section is not symmetric about the mid point in the z direction, stupid me. But I don't see that affecting the process of meshing. I will re-do it once I get the general idea right.

Thanks for any help.




Hello there;



Glad you learned something from my video. Your struggle is something everyone faced at certain stages.


By a flat trailing edge I think you mean a sharp trailing edge. If you have a C section inlet with sharp trailing edge to follow the best thing to do is.



1) Create an O grid

2) Split the O grid from its diagonal edges.
3) Collapse the block right behind the sharp trailing edge.



That way you will have the skewness problem resolved.


By the way if you post some pictures it will be really useful and people will be more responsive as most people are too busy to download and view files. But I hope you get a resourceful help.


Keep well!


All times are GMT -4. The time now is 14:52.