How is X Axis Position Determined or Sensed? (And a BUG in the Prusa firmware?)
Using an MK3S+ here I assembled it, but it worked fine for many months.
Short question: How does the Prusa i3 MK3S+ know where the printhead is on the X axis? Does it jus assume it's where it has put it or does it have some way of knowing where it is?
I'm having repeated issues that seem to boil down to X Axis issues. I think I've fixed it, the printer works fine, I do my prints, then a couple days later, finish a design, send it to my printer (in another building), and think, "Nice. I'll pick it up when it's done." And then I get an error reported through OctoPod. This is the start of an infuriating pattern. I'll go to the printer and, in the meanwhile, get another error from OctoPod that it's paused. This will happen every 2 minutes until I get to the printer and can reset it. Then it's a few hours of recalibration and things act like they're okay - until a few days later.
When I send a print down, the printer does the bed level and, in the middle of it, instead of the print head going all the way back to the left on the X axis, it goes partway back. The printer doesn't know this and keeps moving the head to the right, so far it is no longer over the print bed. It goes down and can't detect the bed and gives me the warning that it needs to recalibrate the Z Axis and starts doing that. OctoPod gets an error and sends me a notification it's paused. So the printer tries to recalibrate the Z axis, but it's moved so far to the right the print head is no longer over the print bed and can't find it, so it finds an error, I get send a notification, and it starts again - it tries to do another calibration. It becomes an endless loop. If I've left the premises, I have to get back otherwise the printer will do this forever. (Can't stop it through OctoPrint!) Also, while this is happening, when the printer is adjusting to try to calibrate on the Z- Axis, it will come down and push the right edge of the print bed down.
Again, this is an endless loop - I see this as a significant bug, since it can damage the print bed while it keeps doing this over and over while I (or someone else with the same issue) may not be able to get back to the printer immediately.
I have had times I try to stop the print and start over. No joy. I try everything - resetting the printer. Nothing works with this EXCEPT to push the print head all the way back to the left and recalibrate XYZ, then readjust the Z offset (which takes time to do properly!).
Today I saw something weird. After this issue, I used the AutoHome function in the LCD menu. It put the print head far on the right side of the print bed. Tried it again. It placed it closer in the middle. Tried again. It put it a few inches from the left. It repeated that last one several times, until I manually moved the print head all the way to the right and it finally did what it should in AutoHome.
All this comes from the X Axis. It thinks it's a Z Axis issue, but that's because my printer has this problem with the X axis control - and ONLY when it hasn't been used for a few days. Never had it until about a month ago, and now it's all the time.
I've asked some questions about this and adjusted the belt and the tension on the X Axis. I've found the report of belt tension on the X axis, as reported by the printer on the LCD screen useless. When I tried to put it in the range it should be in (I think 270-290), it made the belt super-loose and non-functional. Someone has posted a link to a belt tension tester, but the issue with that is it requires Prusament filament because the specs for it are so precise. Yeah, I can get that on my next order of filament, but it'll be another few weeks before I can include that in the budget.
At this point I've tried the belts loose and tight and at adjustments in between. I'm genuinely surprised that, on a Prusa, the printer seems to have no way to realize the print head is not returning to the start of the X Axis - or does it? And what can I check on for this, other than the belt tension and using the adjustment screw that helps position the stepper motor for the X Axis?
RE: How is X Axis Position Determined or Sensed? (And a BUG in the Prusa firmware?)
Have you updated to the latest software version?
Also check your belt for a defect.
RE: How is X Axis Position Determined or Sensed? (And a BUG in the Prusa firmware?)
Short question: How does the Prusa i3 MK3S+ know where the printhead is on the X axis? Does it jus assume it's where it has put it or does it have some way of knowing where it is?
It's counting steps since the homing of the axis. That's all, no sensing of any kind after homing, just dead reckoning, or in other words: it assumes it is where it put it.
If at first you don't succeed, skydiving is not for you.
Find out why this is pinned in the general section!
RE: How is X Axis Position Determined or Sensed? (And a BUG in the Prusa firmware?)
PS: to check the belt tension without having to buy Prusament PETG: There is an app for that: https://help.prusa3d.com/article/adjusting-belt-tension-mk3-mk3s-mk3s-mk3-5-mk4_112380
If at first you don't succeed, skydiving is not for you.
Find out why this is pinned in the general section!
RE: How is X Axis Position Determined or Sensed? (And a BUG in the Prusa firmware?)
Have you updated to the latest software version?
Yes, on the latest. Back in December I had an issue because I had upgraded and the version I went up to had issues with the SuperPINDA not sensing the print bed when it was right over the screws in the bed. I upgraded to the RC version at the time. I'm actually thinking of downgrading to the version I had before that issue developed, since this didn't start until after that.
Also check your belt for a defect.
Thanks! I had not thought of that as an option! I'll look it over. Once you said that, I was thinking that the noise it makes when I'm present and hear it mess up could be explained by there being an issue with the belt and it just not being able to grip it and pull it through. I'm eager to check that out later today. (Having a barn we could renovate is wonderful - gave us a guest "suite" for friends, a rec room, a studio for my wife and a workshop for me. BUT - the problem is my printers and CNC are in a different building, so I have to do things there as I can get down there.)
It's counting steps since the homing of the axis. That's all, no sensing of any kind after homing, just dead reckoning, or in other words: it assumes it is where it put it.
That's actually a bit disappointing, since it leads to issues like this. I was hoping it that, at the least, it had a way to verify whenever it went to one end or the other of the axis.
PS: to check the belt tension without having to buy Prusament PETG: There is an app for that: https://help.prusa3d.com/article/adjusting-belt-tension-mk3-mk3s-mk3s-mk3-5-mk4_112380
I'll be working with that later today (see my comment above about the issues with my printer in our barn, not the house!) Nice that it's on the web and I don't have to worry about if it's out for my phone OS.
I didn't realize there is Prusament on Amazon - yes, not but so big a selection, but I went on and ordered a spool of Prusament PETG because I really just need to get this fixed. (I can print sometimes, but it's inconsistent, but a smaller item like the gauge in question should be printable.) I think between the app and the meter (the PETG arrives tomorrow), that should do it - and give me methods to adjust it in the future.
------------
As for the Z Adjust - that is a loop. Once it starts the process, it keeps repeating it, even after resetting the printer. It won't stop unless it's physically turned off, which can be a problem. It goes into a pause state that times out, then it starts on the adjustment again, can't do it, and pauses and repeats. There really needs to be a stop to that.
RE: How is X Axis Position Determined or Sensed? (And a BUG in the Prusa firmware?)
I had this working -for a freaking day and now - ka-put!
I adjusted the X Axis belt and it was reading right in the middle. I've had NO trouble with the Y axis at all, so I didn't check it. Then my 3.5 upgrade kit arrived and it layer-shifted along the Y Axis. (Remember Saturday Night Live in the 70s - "Jane, if it's not one thing, it's another...") So I go to the web page to check the belts and start on the X. Remember, less than 36 hours ago, the X was testing right in the middle - 99 to 100 Hz when I strummed it. At that time I had loosened the adjustment on the stepper all the way back, then carefully tightened the bolts holding the X stepper in place. I had to be REALLY careful because just a bit of movement and I'd get a too-tight reading. But it worked!
(Looking back at it, I think it's odd that it could go from okay to too tight just by fastening down the X stepper where it is. That's NOT much of an adjustment at all.)
Now it's a nightmare. If I loosen up the adjustment screw for the X stepper, I can't strum the belt - not enough tension. But tighenting it the least little bit and it's between 160-190 Hz, which is too tight. There is NO middle ground.
I have a spare belt. I'm thinking of cutting a length off the spare and replacing the current belt. I'm wondering if it's just too short so if I put on any tension, it's too much.
RE: How is X Axis Position Determined or Sensed? (And a BUG in the Prusa firmware?)
1. Got it working fine and since that was fixed, I went on and upgraded to 3.5, so I'll see if things are fixed or continue to haunt me.
2. I still think there's a bug in the firmware that sets up a continual cycle of calibrating the Z axis over and over unless you're right there to stop it. However, I won't be able to test or evaluate that anymore, though, with the upgrade. I'll see if it shows up for me in the 3.5 firmware.