Telescope Reviews: ZEMAX can do CAD drawings, too!

The 3D graphics I see posted here from Art, Chris, etc. are just beautiful. I only have AutoCad-LT, which has a great 3D wireframe sketchpad capability, but I still haven't figured out how to make solids from wireframes and render them.

So with a little extra bit of studying the ZEMAX manual, I figured out how to make ZEMAX draw shaded model graphics I hadn't thought it was possible to do. See the attached picture of the new lightweight tube I'm going to do for my 16" f/6 Newtonian.

Several methods are at work here:

1. I added the glass "ZERO_CTE_SOLID" to the user glass library. It has an index of refraction of one regardless of wavelength, a wavelength band of 1E-6 to 1E9 microns, and has zero CTE, dn/dt and density. It's just air with a label. This label makes it different from simple dummy surfaces, and tricks ZEMAX into drawing non-refracting solids for plates, focusers, etc. All the plates in this model are made of ZERO_CTE_SOLID. I haven't tried this trick with OSLO but I suspect it will work the same.

2. Use of user-defined aperture obstructions. The hexagonal plates are defined as aperture obstructions, but then a circular hole is specified right after the hexagon aperture, as in

POL 0 0 10.10362971 6 0
CIR 0 0 8.25 64

As long as successive aperture objects following the first object are totally enclosed within the first polygon and do not themselves overlap, ZEMAX treats these as holes in the obstruction. The hollow cylinders (such as the upper end cylinder) are done in the same way. This little fact is actually one of the few ZEMAX features that is not well documented in the manual. The only drawback is that in the current March 6 version, the maximum number of sides for circles, arcs and ellipses is 64. I have put in a suggestion to the ZEMAX folks to increase this to 360 to get smoother renderings.

The bottom plate between the hex plates, and the cylindrical struts between the rear mirror plates, are also drawn as user-defined obstructions, with no holes in them. They are obstructions clear of the optical path, thus they don't interfere with full-aperture raytracing.

(3) Use of surface color and transparency. The upper end cylinder is 30% transparent, for example.

(4) Use of an elliptical ZERO_CTE_SOLID surface with an in-plane surface offset to properly draw Newtonian diagonals.

(5) Use of "Skip Rays" and "Do Not Draw Surfaces" to allow ample use of dummy surfaces for model construction without having to see them or the rays to them.

This is a purely sequential file. You can also import IGES and STP files from another CAD program, make them a non-sequential object surface within a sequential file, and draw even more complex objects, such as the carbon fiber struts and ball joints I'm going to use on this new tube.

The whole thing can be done in strictly non-sequential mode, but the optical analysis functions such as MTF, Encircled Energy, etc. are not available in non-sequential mode. If optical analysis, optimization or tolerancing is needed, you have to stay in sequential mode and embed non-sequential objects within the surface list.

This JPEG image is kind of scrungy due to the wisely chosen 800x600 pixel limit on CloudyNights. The image I downsized this from is beautifully rendered.

Anyone that wants this example file and the supporting UDA files needed is welcome to them. Just PM me with your email address and I'll be glad to send them to you.

Mike

--------------------
56 mirrors, lenses, 16" f/6 Newt, 6" f/10 refractor, TOA-130S, Tinsley 5" f/15 Mak, 6" f/4 RFT, Coronado PST. Still to build: 24" f/10 Modified Dall-Kirkham, 10" f/26 Mak, 8" f/12 apo, spectrohelioscope, Herrig, Schupmann, and a new design you'll like.