drawing 3D solid paraboloid of rotation

[labumm]

I am trying to draw an off-axis paraboloidal mirror. My strategy was to:

1) create a paraboloid of rotation (3D solid) then

2) orientent that and subtract it from the mirror blank (substrate).

Thus would creat a paraboloidal depression.

 

I am stuck at step 1.

 

I used the VBA routine posted by Joro to generate the parabola as a pline.

http://www.cadtutor.net/forum/showthread.php?t=6703

I can rotate the half parabola 360 deg to make a surface.

I can create a closed pline (half paraboloid + axis + truncation). I have tried turning the closed pline into a region, but that doesn't work.

The revolved surface created from the closed pline does not turn into a solid.

 

I hope there is a simple solution I am missing. Thanks in advance.

 

--Lloyd

[kencaz]

Probably the easiest way would be to revolve the full profile of the mirror to create the solid. Sounds, like that's what you were after, but maybe you left the profile open and would result only in a number of revolved surfaces.

 

after, creating your profile use "pedit" to join your polylines.

 

 

Also, you really would not have to create a surface to cut the blank, just make the mirror directly.

 

Hope I followed you correctly.

 

KC

[labumm]

Kencaz, thanks for the reply.

 

Humm. yes I did that this morning (closed pline w/ pedit) and it didnt work, but it worked just now. So thanks for encouraging me to try this particullar route agian.

 

Perhaps you can tell me how the pline spline interactes with the generated solid or rotation. A quadratic spline should be a perfect parabola, but is the resulting solid boundary generated from the spline equation--resolved to the accuracy of the floating point numbers. Or it is some interpolation from the original points of the pline?

 

If I used something like RayCAD to ray trace the optical system using this representation of a paraboloid, would the result be geometrucally accurate?

 

Thanks,

--Lloyd

[kencaz]

I am not sure how accurate the model would be optically... I don't really deal with tolerances below -+.0005. If you looking at grinding your own mirror, then I'm sure Autocad would be able to help you with the tooling.

 

I had never heard of RayCAD until you mentioned it, however, it seems though it works quite well with AutoCad and Inventor.

 

KC