what the *&%^ is wrong with this machine?
This is a machine at my school. I have worked on it before, took it home and got it working. But now after a couple of months back at school it is acting very strangely.
It seems not to understand itself. It can't auto home, puts the extruder in the middle of the X axis. Is the X stepper to blame here? I can't tell if it's locking up. As I said, it turns without power so it's not physically locked but is there some logic or switch that's making it think the X axis is a fraction of its real size?
It fails the belt check, as it can't move the extruder more than a third of the axis from the right. Without power, it can be moved just like mine at home. I did run the wizard/first run and it failed with the first error message.
The Z axis is also messed up. It will press the extruder/nozzle below the level of the heat bed.
I have no idea how to help it find its way. It seems to have lost any idea of what it is. And what could be making the Z axis act this way? The PINDA seems to work, it lights up and the calibration process seemed to almost complete the bed leveling, just failing at the very end.
Firmware is probably one version behind, it was current when I returned it.
What can I check or look into here? What kind of reset might be warranted here?
RE: what the *&%^ is wrong with this machine?
Check if the sheet is placed really correct and flat.
Do z calibration (not xyz). Seems this reset the presettings for mesh bed leveling.
We will do what we have always done. We will find hope in the impossible.
RE: what the *&%^ is wrong with this machine?
Started with Z calibration, it failed. It looks like factory reset is my next step. I had forgotten about that option. I hope that resolves it.
RE: what the *&%^ is wrong with this machine?
I don't think a factory reset will help. So better not (saves time in the end).
Let's work back.
If Z-calibration does not work, we have to go back to XYZ-calibration. It will probably fail too. I have a MK2.5S (the closest to MK3). And had the same problem. Somewhere in the firmware the measurements for height differ from reality. So we have to trick the machine a bit. This what works at my machine:
1. Heat up, and take nozzle out (so you are sure it will not hit into the plate/heatbed.
2. Do XYZ procedure (no sheet). It will probably fail. If not you are good 🙂
3 If it fails: loosen the PINDA and move it up about 2 mm.
4. Run XYZ-again. By moving the PINDA it it looks like your extruder is smaller and there is more room.
Let's hope it is successful like this. If not you might even move the PINDA more upwards (I had to go to about 4 mm).
5. As soon the XYZ was OK (and not earlier):
6. Place the nozzle back. let it cool down (do not give commands now, because when homing now, it will crash into the sheet!!!)
7. Place a creditcard below the nozzle. Whith settings - Z-axis lower the nozzle till it touches the creditcard and you hardly move the card.
8. Place a second creditcard on top of the first and below the PINDA. Lower PINDA until it rests on the second card. Tighten the PINDA (not too much!!!). Now your PINDA is about 0.8 mm higher then your nozzle.
9 Do the Z-calibration now (with sheet on it). It will check 9 points and set presets for the meshbed leveling.
10. DONE.
We will do what we have always done. We will find hope in the impossible.
RE: what the *&%^ is wrong with this machine?
I am more concerned with it failing the belt test than just Z-calibration. It has lost its mind completely from what I saw yesterday. Resetting it costs nothing and means that I am starting from a baseline. If it fails from there, I can tackle the more involved process.
RE: what the *&%^ is wrong with this machine?
This is concerning.
I did reset it — twice — and still no luck. The belt tension seems OK but the motor sounds like it's skipping teeth. The extruder still moves without power or with the steppers off but will not move properly with power. I don't know how it determines the end stop for the axis. It seems like it doesn't know where 0 is. I'll look into that as I will be bringing the machine home to sort it out.
RE: what the *&%^ is wrong with this machine?
Well, the solution seems to have been quite simple. The rods were bone dry, and the bearings were unable to move the extruder head. I wiped some white grease on the rods, ran the head back and forth a couple of times, and the machine passed all the configuration/self-tests. It's printing a first layer test right now. I can't remember where I got this idea but the fact that the head would move without power to the steppers but was unable to move when they were powered was a clue. Still looking for a reliable way to verify belt tension that doesn't interpolate some magic number where the higher the number, the lower the tension.
Anyway, this seems to have gotten resolved. I'll do some more tests and see if this fix holds up.
RE: what the *&%^ is wrong with this machine?
Still looking for a reliable way to verify belt tension that doesn't interpolate some magic number where the higher the number, the lower the tension.
Well that is good news, and interesting how you found the error, coming from Z-axis problems to greasing the X-rods.
As to belt tension. There is a model on printables (to be printed in PETG) what could help.
Prusa has this app (tuner) for this and that works well. And in doubt: more tension is better then too less. But is some work to change these belts (if no tensioners are on the printer). You could just leave it as it is, untill you get shifted layer issues.
We will do what we have always done. We will find hope in the impossible.
R
You are looking for: https://www.prusaprinters.org/prints/46639-tension-meter-for-the-gt2-belts-of-i3-mk3s-or-prus
It MUST be printed in Prusament PETG or the spring tension will be wrong.
School, university and other 'public' use printers (and other equipment) are constantly abused and rarely properly maintained, for a classic example of how NOT to do it see the first 15 seconds of the video in this upload: https://www.printables.com/model/902366-jumping_robot
I assume this is some class exercise as there were several similar designs uploaded about the same time. It's not a robot btw, just a shock-release latch. If I recognised the College I might email the tech staff there as a heads up...
Cheerio.
RE: what the *&%^ is wrong with this machine?
Yeah, I have an unopened spool of PETG just for this. I have a hard time accepting that machines as well-designed as these are lack a belt tensioner that doesn't require you to guess or flip the machine over and work around the belts to adjust. Also a puzzle how no one has designed one, given the open nature of these machines. So much conflicting info about how to tension them, from Prusa's own audio tester to an often-cited YouTube that sets both belts at the same tension. I don't have any investment in Creality but their cheapest machine has two belt tension adjustment knobs right out front. They almost make up for how frequently you have to mess around with the four leveling wheels.
As for the rod-greasing, I have no idea how the Z-axis issues were resolved by that but it could have been the resets I performed. In any event, it is in the middle of a job and working fine.