So I’ve been struggling with my Z axis to some extent since before I even got the GO up and running. While putting the kit together I decided I wanted to switch to metric lead screws for the Z axis to save myself from having to do some math every time I wanted to print at a different layer height, then a short while later changed to larger diameter (but still metric) rods and now have gone and changed them yet again.
For the 5/16″ zinc plated all-thread lead screws (18 threads per inch) that came with the GO to be used to maximum positive effect by all of the metric software driving them, you have to settle for some very specific layer heights for reasons that are intriguing and/or painful, depending on your particular persuasions, hence my desire to do away with them without even installing them in the bot (“standard” American/English measurements are bad, metric are good).
The 6 mm brass threaded rods I started out with however arrived slightly bent and when I started to notice some wobble when the nozzle was moved up and down AND observed some waviness in the sides of the prints, I decided to upgrade to 8 mm lead screws. As with the 6 mm lead screws, I went with brass rods and steel nuts since the rods are slightly easier to replace than the nuts as far as disassembly of the bot goes. The bad news? The wobble was still there, and maybe even worse.
I decided to just suck it up and purchase some proper lead screws. The threaded rods I’d been using as lead screws up to that point were not designed for precisely linear motion, and so the threads could not be expected to be as consistent as ACME or trapezoidal lead screws that were purpose built for the task. I shied away from ACME because of the old standard vs. metric problem and when I searched for trapezoidal lead screws (both ACME and trapezoidal lead screws have very similar thread profiles, but trapezoidal threads are metric) I came across igus.
As far as I can tell, if you’re building something that moves, igus has a solution for you. Every kind of linear guide, bushing, bearing, rod or table you could want, built around their proprietary, self-lubricating plastics. After a little more searching I discovered a number of RepRap folks using their iglide linear bushings in place of ball bearings for movement in the X and Y axis and I decided to give them a call and talk to them about lead screws. Although it was intended as a reconnaissance mission only, I had just sold a watch and had a few dollars in my pocket and made a purchase over the phone. I ordered two 8 mm trapezoidal lead screws in 290 mm lengths and two of their fancy DryLin™ flanged nuts and a day or two later they arrived.
The bad news was that the nice lady on the phone had given me the wrong part number for the trapezoidal lead screws and so rather than receiving 8 mm rods with a 1.5 mm pitch, they were 8 mm rods with 1.25 mm pitch and I immediately cross-threaded one of the fancy plastic nuts when I tried to test fit it. Ouch. Luckily, igus realized the problem even before I did and sent me out replacement rods which also arrived lickety-split. The other bad news being that the nut was damaged by my test fitting AND was on back order (from Germany?) and is still yet to arrive.
So rather than just sitting on my hands and waiting for the replacement nut to arrive, I decided to cut three flutes into the very end of one of the rods and use it like a tap to clean up the threads inside the iglide nut and the procedure allows it to move much more freely, if not quite as slippery-smooth as the undamaged nut still.
Wow, this is a super boring blog post so far. Not even one picture yet. Look, here’s my rabbit.
OK, so where were we? Oh yeah – so I cut down one end of each of the 8 mm rods from igus to 5 mm to fit inside the PTFE bushings I’d been using at the top of the bot and decided to keep using the printrbot couplers for now. And it was around about this time also that I realized that I am a total idiot. My wife has a Lytro camera. If you don’t know what that is and are too lazy to click on the link, let me summarize by saying that it takes pictures that are kinda 3d. The fact that I have access to one of these and have not yet used it to document any of my 3d printing explorations is inexcusable at best, so let me remedy that right now.
These are the rods after cutting them down to 5 mm. Next to a 6 mm brass rod.
Next puzzle: how the heck am I gonna attach these big flanged nuts to the X axis carriage (I’ve taken to calling this a gantry by the way, it makes me feel sophisticated)? I thought about just screwing the flange of the nut to the underside of the gantry, but I’d have to cut out some of one of the panels on each side to make it fit, and doing that and getting everything nice and level and looking presentable sounded like a pain. Moreover, I had a moment of extreme envy while at Maker Faire when I saw Brook lift the printing head of a printrbot JR up to clean or fiddle with something and then just drop it back down on the nut for the Z axis. Until that moment I hadn’t really realized how cumbersome it is at times to have the nuts for the Z axis on the GO completely trapped inside the gantry (what a great word). Arguably, the completely trapped nuts offer some serious advantages for a kit that folds up into a suitcase, but I’ll cross that bridge when I come to it.
So I decided to print out some mounts for the iglide nuts and screw the mounts on the underside of the gantry and just let gravity and precisely fabbed alignment posts do the rest.
These first attempts required a little bit of filing to fit properly, but they fit well enough to allow me to start using them and of course print replacements in ABS.
This is the “final” version of the nut mount, in ABS. I say “final” because I designed this version to be very easy to remove and replace if I get a hankering to do so, and may print them again in nylon at some point.
The funny thing is that the Z wobble is still there, albeit dramatically reduced. My new found ability to easily lift the gantry to test its freedom allowed me to identify some binding around the lead screws (they’re slightly larger than the original threaded rods) and after grinding away some of the wood to expand the holes that the lead screws pass through, the wobble has been reduced considerably, but yes, it is still there. I think part of the problem may be the fact that the lead screws are not directly in line with the linear rods, so the gantry is cantilevered ever so slightly, but honestly, I can’t say if that’s the real problem or not. I’m not sweating it at the moment.