[paulzhang]

I want to revolve a domain to a block comprising of its path. Is there any way to make it in Pointwise?

Please shed a light.

[jchawner]

Hi Paul,

Yes, you can revolve a domain around an axis to create a block. It's part of the Extrude suite of tools. In the Create menu look for Extrude and then choose Rotate from the pull-right menu. You then define the axis about which to rotate, the degrees to rotate, and how many steps to take.

The extrude suite also includes tools for extruding along a line (Translate), along a curve of any shape (Path), and simply orthogonally relative to the domain's original shape (Normal).

Hope this helps.

[paulzhang]

Hi John. I really appreciate it your help. I tried that once.
Let me explain what I want and what I did at first. I need to draw a sphere–biconic with a nose radius of 0.01905 m (0.75 in.) and total length of 0.1488 m. First of all I draw a 2-D geometry, made the grid( Unstructured grid in the nose and structured in the tail). Then I selected the Rotate in the menu. And then step is one, degree is 360, axis selection curve and I selected one line at the bottom. In the end I MLB run. Some error comes. Then I tried to use 90 degrees. But the result is what the attachment shows. There are only the initial and end positions saved, while the path is not the one.

Originally Posted by jchawner

Hi Paul,

Yes, you can revolve a domain around an axis to create a block. It's part of the Extrude suite of tools. In the Create menu look for Extrude and then choose Rotate from the pull-right menu. You then define the axis about which to rotate, the degrees to rotate, and how many steps to take.

The extrude suite also includes tools for extruding along a line (Translate), along a curve of any shape (Path), and simply orthogonally relative to the domain's original shape (Normal).

Hope this helps.

[jchawner]

I think the problem is where you write "step is one." The number of steps should be the number of grid points you want in the extruded direction. If you say just one step then it does the entire angle sweep (90 degrees) in one step and then connects to the original. So that's a straight line. You need to set the number of steps to something like 21 or something.

But I also want to point out that you have a singularity (pole) in the surface mesh that lies on the axis of rotation. I'm not certain what types of cells will result. I recommend that you change the mesh topology to get that pole off the axis.

Best Regards

[paulzhang]

Exactly John. It works.

Originally Posted by jchawner

I think the problem is where you write "step is one." The number of steps should be the number of grid points you want in the extruded direction. If you say just one step then it does the entire angle sweep (90 degrees) in one step and then connects to the original. So that's a straight line. You need to set the number of steps to something like 21 or something.

But I also want to point out that you have a singularity (pole) in the surface mesh that lies on the axis of rotation. I'm not certain what types of cells will result. I recommend that you change the mesh topology to get that pole off the axis.

Best Regards

[paulzhang]

Hey John. I may have a rather wrong understanding of the term Pole. I thought it is used in substitute of line. For example in my case, I want to assign unstructured grid in the nose region. But the region is fan-shaped. There is no way to make that except to put a pole there right?

Originally Posted by jchawner

I think the problem is where you write "step is one." The number of steps should be the number of grid points you want in the extruded direction. If you say just one step then it does the entire angle sweep (90 degrees) in one step and then connects to the original. So that's a straight line. You need to set the number of steps to something like 21 or something.

But I also want to point out that you have a singularity (pole) in the surface mesh that lies on the axis of rotation. I'm not certain what types of cells will result. I recommend that you change the mesh topology to get that pole off the axis.

Best Regards