Mirror Cell Optimizer

Building a telescope mirror cell can get surprisingly technical. Mirror Cell Optimizer is designed to make that part a whole lot easier, with minimal input needed from the user. Insert your mirror diameter, focal length, desired secondary obstruction and material of the glass. After that, you have possibility to experiment with different cell designs, support points and see how the mirror would flex with the given configuration and pointing angle.

Software models how gravity affects the mirror from zenith all the way to the horizon. It supports everything from basic 3-point cells to high-count 24-point designs, making it useful for most amateur builds even with large mirrors. The software also includes a catalogue of edge support configurations to choose from. It is designed for Alt-Az designs, but future implementation might add equatorial mode. And the really practical bit? It doesn't just show you fancy deformation maps and RMS values from the perfect world where there is no cell flexure. It generates workshop drawings, complete with support point coordinates and dimensions for hole locations on all of the parts. You only choose what material you wish to use for the parts. Positioning of points and holes is done with few clicks.

So rather than juggling calculations, spreadsheets, and hand-drawn sketches, the goal is to go from “I have a mirror” → “Use the app to find out what kind of cell it needs” → “let's build it.” For telescope makers who'd rather spend their time under the stars than fighting the math, that's a pretty compelling idea. I however am not a software developer, or pretend to be one. For the app, I used AI to build the entire thing. The reality is that I haven't seen a single line of code building it which is amazing - and terrifying at the same time. Capabilities of these models are getting so good, that it seems almost unreal. But like with all AI produced content, it needs some real world intelligence to verify that information. AI can, and will, hallucinate and be perfectly happy with that result if the user doesn't intervene. That is exactly what I have done with this project. Bouncing between identical parameters in PLOP and this creation, making the AI actually state what it is modeling in an understandable form. This manual "feeding" of the model, ideas and practical solutions that actually work, are the base where the application is built on. The results are not identical to the good old PLOP, but they are getting very close. Close enough, that I feel that this project is ready to be shared with other amateur telescope makers to use as they wish for non commercial purposes. If you wish to do commercial things with it, I'm bribable with good eyepieces and pizza. Results Here are multiple mirror sizes and supports ranging from 3 points to 18 points ran in both applications. Green triangles are from PLOP, red from MCO. The position of the points is very close, also there might be some error in my overlapping of the circles manually. I believe this error is small enough that in the real world material selection to make supports will make more difference than slight offset from absolute perfection - if such thing exists. One would need to do full aperture interferometric tests of the mirror in it's cell to be certain regardless of any model in the workshop. Thus, this app has no intention to be the absolute truth, more of a general guide to make practical decisions with.

• 

• 

• 

Downloading the app is possible from my Drive from this link. I promise it doesnt contain internet asbestos or an Albanian ransomware. Currently it is tested in Windows 11 only and runs as an .exe file. Feel free to experiment with it, or even build a telescope and see how it performs. If you have ideas for improvement drop me a message on contact page and I just might tell AI to do something about it.