People, I need some seroius voodoo magic help here: X axis is having absolutely incomprehensible step misses. Right after assembly all worked well, Sanguinololu with the assortment of drivers (radiator-less). Printing happily on test cubes and such.
Then all of a sudden - X starts missing steps. Aaaand never stops. Missing steps. See the 5-mm-step pyramid below.
Here is what I have done to it already:
changed the driver - that helped things a lot, motor stopped overheating so much and new driver is way cooler. Still missing steps though.
Lubricated rods, checked belt tension - tried looser, tighter, moderately tight. Same thing. Checked if the rods are parallel - adjusted them. 37.3 mm to 37.1 mm left to right. Removed belt, tried rolling extruder without it - rolls like breeze.
But - with belt on it rolls with some difficulties. Found belt it passing through a couple of tight places next to motor and idler where it has to pass between two plywood walls. Turned out these walls (that also hold the 8 mm rods higher and lower) are being bent to leave gap almost 6 mm wide - the width of the GT2 belt.
OK, disassemble the whole axis, add spacers everywhere, add washers to the idler bearing. all to bring the width of all places the belt passes on its way to 8 mm. Succeeded, idler bearing rolls smoothly, belt does not scratch walls anywhere on its way. Adjusted motor pulley at the same time, marked pulley with permanent marker to check if it ever turns - it does not. No luck, steps still missed…
Alright, swapped motors between X and Y (type KH42JM2B140E, I haven’t mentioned yet…)
Installed a ferrite ring on X cable next to motor.
Cranked up the voltage. 0.8 volts - still missing steps. 1.3 volts - motor is more loud but at least I have managed to print a little square cube. Test pyramid - not as bad as photos, but still missed one time. Driver radiator is at 44C, motor shaft 57C…
The question is - what did I miss?? Acceleration is at decent 1500 mm/s2, and the latest prints were done in the lowest possible speed settings, like perimeter 5, solid infill 15, sparse infill 30.
I even thought I have managed somehow to wear down the belt teeth and tried to print in various places all over the bed - no luck with that either.
So what is left? Replace motor with a beefier one?
@Jan_Wildeboer
well, that’s what the ferrite ring was added for. I must say that the whole wiring is done via a piece of UTP cable (which, IMHO, should be somewhat less induction-prone) - a pair for X endstop, another couple of pairs for the motor coils. Yeah, changing wiring is another thing I am thinking of - but why would it cause such trouble on X but in no other part? Y, Z and extruder work flawlessly, while being wired almost the same way…
Did you check if the motor does not slip in the gears? I had that problem on my extruder axis where the screw fixing the gear came lose.
Took some time before I found what the problem was.
@Igor_Larine with the pulses flowing through the wires to the steppers, the risk is quite high. I had similar problems and separating the end stop wires completely (wiring them far away from the stepper wires) solved it for me. Just try with separate wires for the x end stop and see if it helps.
I would definitely examine the wiring. UTP is “UNtwisted Pair”, so your assertion that it’s less prone to induction is fallacious. One quick test would be to run the motor wires through a separate cable, possibly shielded, but away from the rest of the wiring. That’s what I did on mine…
@Kristof_de_weyer
you mean whether the pulley moving the belt slips on the shaft of the motor? yes, checked that - marked a spot on the shaft and pulley and checkaed after all these unsuccessful prints - not a fraction of movement. But I have aluminum pulleys with 2 screws - it should take a lot to have one slip on a shaft. Belt slips faster.
I do however agree that network cables are not perfect candidates for 3D printers. They simply are not made for the relatively high currents that flow to a stepper.
@Alan_Weber
yeah, but that would mean steps missed on the Y axis, right? Here it’s the X - and believe me I have meditated a long time in front of that moving machine trying to understand where it slips. Pity the bloody thing does not work as it should - I had plans to print a holder for my phone to make movies of print jobs. Just for these purposes…
First thing when I saw this, I thought “solution: gotta be a pulley loose” – I’ve seen this on an Ultimaker where the print just slanted 1 direction. Turned out to be a loose pulley set screw. So check for a loose pulley on the stepper motor, tighten that up.
I’ve had similar issues, solved them lowering maximum acceleration of X axis: my controller came with 9000mm/sec2 default settings, had to set it at 1000mm/sec2
Hope it helps.
Is your X axis motor hot when it skips? Do the skips matchup with your belt tooth spacing, ie. does it always skip in multiples of your belt tooth spacing or is it sliding randomly? Do you see any scratching on your guide rods that would indicate a linear bearing catching?
@Jan_Wildeboer
Unshielded it is. And my eyes tell me it’s twisted It looked like a better candidate than the ribbon cable that’s commonly used - it’s a good quality cable with silicone outer sheath, the wires are 0.2 http://sq.mm, so should theoretically hold up to 1.57 Amps sustained - and since out currents are very short-termed, plus the length is not more than 50 cm… Well, in the end I might as well cut it up and resolder using 1 mm thick wires. But I will really leave this option as last
It looked like a better candidate than the ribbon cable that’s commonly used - it’s a good quality cable with silicone outer sheath, the wires are 0.2