I mentioned in Accelstepper that I had been working on a 5 Axis platform for a laser cutting project. Well I’ve sent the drawings in and parts are slowly starting to come! We’ve already ordered and have many of the smaller parts…motors, drivers, (not yet controller….that’ll be another post) and the rotary parts for our rotational axis Namely the C-axis…Remember the ABC are the ones the rotate around our XYZ.
Buuuut of course. Nothing ever works the way a person wants it to and here is a little tale…or maybe review of our purchase.
We decided to buy Igus slew rings¹ instead of designing a complicated set of axel+axialbearings to rotate a table that barely weighs 300 grams. They’re pretty expensive per part, and have all the fancy Igus PTFE² or whatever plastics they use (it’s teflon). Usually When one decides to go high precision, you can guess by the price of something an expectation of quality, one can also expect (wrongly) a certain kind of tolerance. If you want to rotate a thing…it should not move eccentric around any of the XYZs…Our expectation was exactly (and falsely) that.
Once the parts actually showed up, a person could literally shift the outer HTD ring around it’s sliding ‘elements’ Infact, as I just see now in the picture. The black core was ø37mm and the ptfe elements were ø37.25mm! That may not seem like a lot. But for a high precision machine like the one I’m building. It’s nearly 250 times what I can even tolerate.
Fortunately we arn’t working in the micrometres anymore, just in the 10s of micrometres! So it’s only 25 times larger than what I can deal with. Thats something I can work with!
How you ask?
This is a exactly defined (also with a tolerance but whatevs) strip of metal. The ones I have are 0.1mm 0.02 and 0.05. When a person measures things using something like a venier calipre you can expect that your measurements might be off…especially if it’s a low quality (again the quality here) something called Abbe-error³ shows up or you moved the things after trying to remove it from the thing you were measuring!
So it’s kinda trial and error. And thats why I had different thickness to play around with. I immediately tried to use the o.1mm band and found it was way too thick to use. Once reassembled the ring would barely move. Waaaay too much friction. So I next tried the 0.05 and the 0.02. (spoiler, they worked)
From my old work I learned that when you’re going to use something like these you should cut them at an angle instead of straight. The moving parts inside are less likely to catch the ends and maybe start ripping things up. Of course we have a part that is a little jagged and sharp, eventually the parts will wear. You can reduce this wearing by doing this simple trick! Not to mention it looks nicer.
So in the end we closed up at slop of 0.14mm which means (assuming the 250 micrometre was accurate) we’re down to 0.055mm eccentricity. Thats one I can deal with. So whats my lesson learned? Maybe just use the axel+bearings instead of going with Igus slew rings…