Monthly Archives: March 2025

optics

Foucault Knife-Edge Test

My first optics post was more about 3D printing than optics but this one is for real. I have been wanting to see how good the optics in my telescope are so I went about building a Foucault Knife Edge Tester.

A schematic on the Wikipedia article. https://en.wikipedia.org/wiki/Foucault_knife-edge_test#/media/File:Foucault-Test_1.png This file is licensed under the Creative Commons Attribution-Share Alike 3.0 Unported license.
What my setup looks like. The glow of the red light source can be seen under the camera, the telescope with mirror is at the far end of the table and the knife edge is sticking up in front of the camera.
The heart of the tests, simply a light that goes and a boxcutter blade taped in front. I also added some parchment paper to diffuse the light, and the internet recommended a this configuration where you use a two knife edges to make a slit for the light, I am not sure I could tell the difference. I used one long blade for one side of the slit and the knife edge proper for convenience. This was simply taped to box and a used books to get things aligned well enough. The returning dot just needs to fall some where on the blade and the camera placed behind. It was pretty finicky, a linear stage would be very nice.

It is a simple device which has a point source of light that propagates out spherically and converges back to a small volume after hitting the mirror. Now I say a “small volume” because depending on the shape of the mirror it won’t focus back to an exact point which is how you learn about the mirror! The rays of light hitting the out side of the mirror may converge earlier or perhaps later than other parts of the mirror. The knife edge lets us “cut off” the bundles of light from different parts of the mirror to see how different parts of the mirror are converging light and thus measure its shape.

In my case this didn’t turn out to be that visually interesting in the tester. My telescope happens to be what Astronomy Garage with Reflector has dubbed Bowden Bird Jones/ FC (See a great video on it here) meaning it is a spherical mirror with a sub-aperture corrector in the focuser tube. You can see that corrector in the focuser tube here.

The corrector inside the focuser tube, the reflection is the easiest way to see it.

Because the mirror is spherical (and reasonably good one at that) it does return to nearly a point, meaning that the knife edge either cuts off the light from the edge if it is before of after the focus, or if it is close to the focus generally just cuts it off entirely causing the light to blinking on and off with the slightest touch. This is one of the more interesting results I got:

The image taken in the camera near the focus. We can see the damage around the edges and some scratches on the mirror. Also maybe some texture in the middle of the mirror.

Mostly I saw things like:

Image where the knife edge is in front of the focus point. It is pretty need to see the knife come in from the other side of the image from the knife is physically, a cool consequence of the converging light.

Also to the naked eye, I had a very hard time see much of anything.

I should say my interpretation of the results as a spherical mirror is my understanding given my belief the mirror is spherical given the telescope design and from what I think makes sense. It would be nice to see a parabolic mirror for comparison.

Future Experiments in this Area

  1. I would like to buy a parabolic mirror for the telescope. This would allow me to have a different shape to compare to, but also I am interested in seeing how it compares to the spherical mirror for aberrations with the corrector removed
  2. A linear stage would be a very helpful and perhaps my imprecise setup is preventing me from seeing the shape well. https://gr5.org/bath/ sells linear stages along with the material for a bath interferometer which would also be a lot of fun to try for characterizing the optics of the telescope.
  3. This setup up should also be capable of Schlieren imaging which would be a lot of fun!
3D Printing, optics

Telescope Eyepiece Holder Replacement

Ok This first project is less about optics and more about solving a problem. Ever since I have had my telescope when I bought it at 15 on Craigslist, it has been missing the eyepiece holder on the focuser tube.

For the most part this wasn’t a problem as the focuser tube is very close to 1.25 inches and fits the eyepieces pretty snuggly. There is a little wobble which my impact image quality a bit and would just move the telescope to have the best image quality on the object of interest which was fine for me. The biggest risk was just the eyepiece falling out.

With the 3D Printer up and running I went about design a replacement. The final design can be found here on OnShape https://cad.onshape.com/documents/6c2b93c301d81fd13156a65c/w/dbd2b71c3528a5cccf87b7b3/e/df8993ec8ee1d7e4efd36834.

Model of the eyepiece holder.

It is a simple design with threads at the bottom to match the focuser and 3 holes to take threaded inserts for thumbscrews. On cheaper models there is only one screw, which is probably fine give my use case but three is supposed to allow you to hold the eyepiece more evenly and was easy enough to do. The best eyepiece holders use a compression ring which is a thin piece of metal with a gap around inside the ring, and the screws press in to shrink ring and press more evenly and prevent marring of the eyepiece. Maybe if I have free time someday I will try to make one like that.

The toughest part of this build was getting the thread pitch and tolerance correct. For the pitch I used calipers to measure the pitch to be about 0.75 mm and guessed it was likely a round number so when with that which turned out to be correct. The thread is an equilateral triangle of 0.7mm and the inner diameter was tuned to be a diameter of 35.6 mm by doing some quick test prints adjusting until it screwed snuggly onto the focuser. The thickness was just eyeballed to seem sturdy.

The final result with the thread inserts already placed and some placeholder screws until I can buy some thumbscrews:

3D Printing, Projects

Mechanical Clock Mechanism

(Conducted ~Nov 2024)

There is a popular Escapement Desk Toy on Thingiverse (https://www.thingiverse.com/thing:3364860) I have been wanting to print. With the printer now functioning well, I got about printing the parts.

Improvements With Remixes

https://www.thingiverse.com/thing:3408585 has a pallet fork holder that is not as wide, attempting to place the pallet fork more in line with the escape wheel. And the original impact pin house has square edges and the pallet fork gets caught on it. The remix adds chamfers which reduces the amount it gets stuck.

This remix https://www.thingiverse.com/thing:3753245 tweaks the angles of the pallet fork to help it release from the escape wheel. In the original Desing they are prone to bending.

These remixes helped but the biggest issue I found in my print was the pallet fork’s axis was too low causing it to bind. The lower fork was too far in and would impact pin wouldn’t move it far enough out to let the escape wheel move and the upper pallet fork was too far out and would often not stop the escape wheel, instead letting it skip. Lifting up the balance wheel and pallet fork sub-assembly with a couple of washers between the frame and the base got the mechanism acting more symmetrically on the escape wheel, and letting the mechanism run more reliably.

Remaining Issues

The mechanism is still pretty finicky. It can run for 5-10 seconds, but eventually something catches. There are other larger issues with the design. The biggest of which is the impact pin (and thus the balance wheel/spring) are in contact with the impact fork as the impact swings up. This is not desirable in a watch mechanism because the free oscillation of the balance wheel is what allows for accuracy (un-impeded oscillators have a constant period well without friction) and efficiency (i. Its operation is more like an early clock mechanism called a verge escapement which still has a spring, but each oscillation it jams into the escapement instead of swinging freely and leaving the pallet fork behind to hold the escapement.

I am not sure how much further I will take this build. For the moment, I have fulfilled my goals of having a toy mechanism I enjoy and is fun to show off, and to learn more about how escapements work. It might be fun to continue to tune this build to test my knowledge or start from scratch with my own design, or pick one that is better already published. For now, I will probably turn my eye towards other projects I have been wanting to try.

3D Printing, Projects

First Prints of a Benchy

(Conducted Oct 2024, Posted March 3, 20205)

Left: First print. 2nd from Left: Reduce print and bed temperature, 2nd From Right: After fixing the fan, Right: More success with another filament.

Explanation

Just got a printer 2nd hand from a craigslist . It is an Ender 3 pro with a Micro-swiss all metal hotend with bltouch bed sensor upgrades.

The first print came out pretty ugly, the bumps on the hull indicate that the print isn’t cooling sufficiently.

Given this, I turned down the print temperature and be temperature so it is easier for the filament to cool. This significantly improved results but the hull still showed slumping. The print temperature was now significantly lower than expected and much more sensitive to temperature than it should be. So I started looking for other cause of problem.

At this point, I realized the printer was suspiciously quite which led me to checking the filament cooling fan which was not running. After some poking around with settings which didn’t have any effect, I found that one of the wires for the fan had broken off. I got out the soldering and reattached it and…

Voila print #3 came out great! I threw in another filament (eSun black PLA+) I had just bought to try it out and make sure it was repeatable and also great.