In my last blog post I reported on the arrival of my new laser cutter.... Today some videos of the first day and a half of playing with it. First let me say, this machine is AWESOME. I was expecting at least a week of frustration before it did anything sensible, but it basically just started up and ran right out of the box. I have not even had to touch the laser alignment despite its having been shipped across the country, tipped on its side, etc, since it was first put together.
It was a good call putting this thing in our side-building (a former drive-through bank, basically a bunker with bullet proof glass and a completely separate ventilation system from the main office building). It came with a powerful blower to extract fumes, and this turns out to be really quite important. Burning plastic smells terrible! I'm paranoid about setting things on fire, so I replaced the supplied plastic ducts with metal ones. The box in the middle is a water cooling unit that circulates cold water through the laser tube. (The shelf above it is the former money drawer, for passing things to customers outside in their cars. Sadly it doesn't work anymore.)
There is a lot of stupidity with file formats trying to get patterns into the machine, but that's not the fault of the machine, it would be the same with any model. Remarkably, in some cases I have had to resort to using the DST file format to transfer a design from Mathematica into the software that operates the machine. Why? Because bugs in Mathematica's DXF export make that not work, and I happen to have an elaborate body of code for exporting DST files, developed for use with our quilting robot. Why is this silly? Because DST is a decades-old format originally implemented on punched tape. It's limited to 0.1mm resolution, which is not good enough, so I have to expand the pattern by a factor of 10, then shrink it back down again. Oh well, I'll get some workarounds in place eventually.
Anyway, here is a video of the first real thing I tried to cut, a model of a spring-dial scale:
The version being cut here didn't work, but after several iterations I got a working device.
Making springs out of acrylic is sketchy, because the stuff is very brittle. But with enough zig-zags, I got springs that are able to stretch as far as they need to without danger of breaking. The scale is not "calibrated" in any sense of the word, and is super inaccurate, but that's not the point. The purpose it to let you understand how the mechanism works with a simplified, cross-sectional sort of mechanism. (When placed side-by-side with photos/videos of a real spring-dial scale, everything will make sense, I promise.)
After making the model scale, I started cutting and engraving transparent replacement dials to go on a pair of real scales (one spring scale, and one huge no-springs Toledo scale). My idea is to replace as many as possible of the cover plates and the dial face with clear acrylic, so that we can photograph the scales in operation, but you can see how they work inside. Creating the pattern for this involved spending a couple of hours carefully measuring the original dial, and writing Mathematica code to recreate the tick marks and letters as closely as possible. (Toledo uses a very strange font, and they are currently looking to see if they can get me a version of their old corporate font to make the numbers closer to the original.)
This scale dial is huge! 60cm, or nearly 24 inches, in diameter. Interesting, I think, that all the dimensions of the dial are clearly in metric units, even though it was made in Toledo probably in the 1950s. How did that happen? I have a book on the history of the Toledo scale company that will hopefully explain things to me.
Here's a picture of the smaller spring scale with its transparent replacement dial temporarily in place. Sorry for the mess, it's my garage. We will clean up and repaint (the scale) before taking the real pictures. Notice how the mechanism inside mirrors the acrylic model from above, except it's harder to see how it works. In both cases there are a pair of springs, and a rack-and-pinion gear system to turn linear movement into rotation of the dial pointer.
Next I'm working on a lock mechanism, which isn't working very well yet.
Meanwhile, Connor is having fun not cutting all the way through a piece of 1/2" acrylic. This is a time lapse of his depth test, varying the laser power level. I'm surprised how consistent the depth is. I didn't think you could do this with a laser cutter: