[PSYMN]

That was a pretty specific request. How many people do you think would even have that sort of info? You may need to find them rather than assuming they will catch this thread.

[PSYMN]

OK, I took some lunch time to take care of this one...

A number of things are wrong with this model.

1) your blocking is not periodic... The verts on one side must have an exact periodic match on the other side. This can't really be fixed without starting over because your actual topology is not periodic...

2) Some edges that should be associated to curves are not. That was easily found and fixed...



3) You had no Ogrid. I guess this is what you were actually asking for help with.

This is also easy to fix... You go to the Ogrid tool, select all the blocks in the model, then Select all the faces except the face along the airfoil...

Wadoud_02_OgridSetup.jpg
Apply.

Wadoud_03_Ogridedges.jpg

Then I setup the Ogrid edge with 10 elements and a spacing2 of 0.0001 and copy to parallel... Generate Premesh...

Wadoud_03_OgridPremesh.jpg

But that doesn't help much if the model needs to be periodic. It would be easier to start over than fix it. So what are your next steps?


-----------------------------------------
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]

Since you are just creating a 2.5D model with no tip gap (would require flow over the end of the blade), your best bet is to just model this as a 2D model (less than half the work). Don't worry about CFX needing the solid elements, just output it as a 2D fluent.msh file and CFX will thicken it on its own during import.

Also, it is often difficult to make such a very curved geometry periodic with good quality, but things get easier if you recreate the geometry to model between blades rather than around the blade. This is because you won't need to make the sides of the blade periodic , but only the upstream and down stream flows.

To do this, I would start by deleting all your surfaces and curves except for the profile of the one side... Then create mid points along each of the curves. Some people copy the side curves to just shift them over. Others create new curves. I like to do a mixture. Here I have copied over the side curve, but I don't like how it comes out of the front of the blade...

Wadoud_04_2D_01.jpg

So I create some N points along the central curve (I used N=8 points), and recreate my own using the first 7 points along the curve, the mid point of the leading edge and the mid point of the inlet... I wanted it to meet normal to the inlet (faster convergence) so I decided to take a point from the top of the arc, project it to the inlet and create another segment.

Wadoud_04_2D_03.jpg

Then I trimmed the curve chain and copied it along with the airfoil using the inlet end points as my vector.

Wadoud_04_2D_04.jpg

Then clean up (trim airfoil, move inlet and outlet, etc.)

Then create a surface from the curves...

Wadoud_04_2D_05.jpg

No you try to block it...

[wadoud31]

thank you for your effort, I'll do the same blocking and after I try to master the périodicité, thank you

[wadoud31]

Message: |
| A negative ELEMENT volume has been detected. This is a fatal |
| error and execution will be terminated. The location of the first |
| negative volume is reported below. |
| Volume : -0.3019E-12 |
| Location : ( 0.61703E-03, -0.17941E-02, 0.83333E-03)

[PSYMN]

I don't know where this error came from or what you did to get it... I suggest your just try again... Perhaps post an image of your blocking scheme...

Often I may even print out (or usually just sketch out) the geometry first and draw in the blocking I hope to achieve, then I think about how to get there.

Did you start with 2D Planar blocking (instead of 3D)? What blocking topology did you try on this model? At what point in the process did it complain with that error?

[wadoud31]

thank you once again, here is the blocking

[PSYMN]

The biggest problem I see with that blocking is that it CAN NOT be made periodic, which is usually a problem. Is this the blocking that gave the above error? I don't see any inverted blocks, so perhaps your projections are wrong... You would need to look at the premesh... Or check the quality and find the bad premesh elements by clicking on the low bars of the histogram...

[wadoud31]

I had this message when i use the SST model; thank you

[PSYMN]

You should be checking quality while still in ICEM CFD. Check quality, determinant, min angle, etc. If you find it while in ICEM CFD Hexa, you can just move a vertex, split an edge or adjust an edge distribution to sort out your problem...

[wadoud31]

thank you for reply;how I can refine the walls to satisfy the condition of Y + in SST without problems?

[wadoud31]

hi all, I try to solve the problem of negative elements but I can't, here are some pictures of my Geometry; thank

[wadoud31]

here the blocking

[PSYMN]

I can't tell from this where you problems are... Turn off your premesh so you can see the bad elements. Put a scan plane thru the problem area to get a better idea what is going on... Then adjust the blocking or projections or edge curvature or what ever it takes to unravel things so that the mesh comes out correctly. Your topology is fine, you just need to fit it to the geometry better.

Simon

[wadoud31]

simon, thank you very much for your help, I am very grateful and I followed your advice.

[venkat_aero2007]

Hi
I am trying to mesh a propeller blade. I tried to mesh the propeller blade just like the wing. The propeller geometry is shown in the figure 1. Firstly I created a block and made it periodic. Then I made a two horizontal split (along y-axis), one in the front of the blade and other behind the blade. And I also made two vertical split (along x-axis), one in the suction side and other in the pressure side. Then I inserted a O-block and merged the vertex in the trailing edge and converted into C-block. Then I associated the edges to the curves of the airfoil section of the propeller blade. Figure 2 shows the block details. I made several split along the span of the propeller blade and associated the edges as I said previously. I got a nice mesh with good quality.

Actually I want to do optimization so I am going to automatize the meshing process. When I change the twist angle along the span of the blade, I get high skew cells near the blade region as shown in the figure 3. I am going to change the twist angle continuously during the optimization process. So this blocking strategy doesn't work.

I heard of shifted periodic blocking strategy will fine for the highly twisted blade. I have seen several posts about the shifted periodic block but I really don’t know how to do it and I don’t know how I can apply the shifted periodic for my propeller blade.

I need a blocking like figure 4. But I dont know how to get that. Please help me to construct such blocking strategy. I have attached the tin and block file. Please help me.

https://www.dropbox.com/sh/jkl3dqoaf...ont_domain.tin
https://www.dropbox.com/sh/jkl3dqoaf...ont_domain.blk

https://www.dropbox.com/sh/jkl3dqoaf...ont_domain.prj