surface box normal

Hi All

I'm guessing there's a quick answer to this, but I can't find it...

How do I extract the normal of a plane from a surface box component? Or, otherwise said, How do I get the normal of a surface at the same u,v location as the surface box origins?

I've tried 'plane components' and 'evaluate surface'. Either one would seem to work, but the difficulty is recomposing the x,y,z vectors of the plane components into the correct vector (I got some different results but not the correct one), and/or inputting the correct u,v for the evaluate surface.

Any suggestions?

Thanks!

Rob

Replies to This Discussion

Permalink Reply by Rob Henderson on January 14, 2010 at 7:14am

Any suggestions?

I'm still getting all the planes with +z vectors when they should be responding to the curvature of the surface.

Permalink Reply by Luis Fraguada on January 14, 2010 at 5:43pm

If you provide an image, its usually easier to help...is this something like what you need?

Permalink Reply by Rob Henderson on January 15, 2010 at 3:40am

Yep! Much thanks, Luis. The 2d slider was the missing piece.

Below are a couple screen shots of what I got to work. Just a part of a larger def.

It still seems like a lot of work to arrive at a value that is already in the srfbox component: the vector giving the surface box its height. Is there not a simpler way to extract that value?

Permalink Reply by Luis Fraguada on January 15, 2010 at 4:11am

Well, you could get the normal from the center of some SrfIsotrims (the face of the srf box that lies on the surface)

Permalink Reply by Rob Henderson on January 15, 2010 at 5:07am

Nice, Luis, and that's a bit cleaner.

Still feels like a bit of a work around, though. And when you work through the srfbox you can locate the origin of the vector at the top of the box, which is equal to the H variable. For panelization that could also be useful. I think this is a question for David... David, is there a direct way to get the H value--vector with amplitude--out of the srfbox component?

If you're using the srfboxes, then the next question is how to equalize the number of vectors/orgins and surface boxes. Check out the image below and the attached def (use any two crvs to make the loft). The exploded brep should be giving an F value of 7200 (900 boxes * 8 faces/box), but it only gives 900. And that causes problems for the circle component.

Attachments: