I had the chance to inherit a retired Geeetech A10 3D printer, minus many of the 3D printing specific parts. I gave it some power and devised a replacement for the missing Z-axis end stop. While this was not enough to restore it to 3D printing ability, it is now a functioning 3-axis motion control system. What can I do with it?
The problem is not trying to come up with ideas… the problem is trying to decide which idea to try first. Motion control systems used to be strictly in the realm of industrial machinery, but 3D printing has brought such capability within reach of hobbyists and now we are here: systems getting old and available for re-purposing.
I’ve decided the first idea I wanted to investigate was a camera based measurement system. Using a camera mounted to the carriage, measure and calculate dimensions of things placed on the bed. I’ve wanted this kind of capability many times in past projects, designing enclosures or brackets or something else for 3D printing to support an existing item.
Most typically, I’ve wanted to quickly measure the dimensions of a circuit board. Sometimes that’s because I have a salvaged piece of equipment and I wanted to repurpose it into something else. Other times it’s because I bought some electronic component on Amazon and I wanted to build an enclosure for it. It’s easy to use a caliper when they are rectangular, but they’re not always so cooperative.
People have asked if they could get dimensions from a photo. This is possible if the camera has been calibrated and its optical characteristics known. Lacking that information, a photograph is a 2D projection of 3D data and this transformation loses data along the way that we can’t reliably extract afterwards.
But there’s another way: if the camera’s movement is precisely controlled, we can make our calculations based on camera’s motion without a lot of math on optical projection. Is it easier or harder? Is it more or less accurate? It’s time to build a prototype of something that can be thought of as a crude optical comparator.