Ok, this is pretty weird. First of all, the basics. Prusa I3 variant, Marlin 2.0.x-bugfix (latest and version from about 3 weeks ago)., Ramps 1.4.
I've been printing happily along for quite some time now and then last night, something strange started happening. Mid print, the printer would act like the z axis stopped going up and the nozzle would end up dragging through gouging into whatever was printed so far. It happened on one print last night. The object is about 6mm thick and the problem started showing about 1/2 way into the print. Tonight, on a totally different print which is about 30mm tall, it seemed to happen after printing about 10mm. Then on a second attempt, it started noticeably failing at about 15mm (though that might not be the case based on what I found). On the first print (the 6mm object), I did some testing afterwards and found the printer was about 4mm off from where the printer thought the z-axis was. On the first print of the 30mm object, it was about 8mm off from where it thought it was.
Before the final attempt, I swapped out the stepper driver on the z-axis and made sure the voltage was adjusted to what the previous one was.
In all cases, whatever was happening was affecting both z motors as the carriage didn't end up higher on one side.
For the third attempt I watched it printing the whole time. What I saw was that on every layer change, I would see the z-axis spin to move it to the next level, but before starting to print, it would tick back a little bit. I wasn't sure if that's what it's always done or not, so I kept watching. The 30mm object prints almost just a plain rectangle for about the first 8mm or so. Then after that, the layers have some more complex features, so it does multiple retracts/unretracts during a single layer. That's when I was able to see what was happening. On every retract, the z-axis would move down some, and on unretract, it would also move the carriage down.
After I bailed out of the print, I then heated up the hotend and manually issued retracts/unretracts (G10, G11) from the RepetierServer command line input. I could easily duplicate the problem there. I updated my Marlin bugfix 2.0 code to the latest, but it still has the same issue.
On normal extrusion, it doesn't affect the z motors at all, only on retract/unretract.
It sounds like there might be electrical interference of some kind. But the wires for the z motors don't run anywhere near the extruder stepper wires. The z motors are mounted at the bottom, and the extruder is on the x carriage. the closest the wires are to each other is at the ramps board itself.
I'm not sure what else to try. Could the ramps board develop some kind of issue that will cause electrical problems like this? I'm pretty much stuck until I figure this out. I hope someone has some ideas.
