Z motor catch and dropping only during startup sequence.
-
- Posts: 59
- Joined: Mon Mar 21, 2016 2:18 am
Z motor catch and dropping only during startup sequence.
Hello!
I have a M2 Rev. D, upgraded to Rev. E , and upgraded to Dual extruders for this particular issue. It's been intermittent for a few months now, but has recently become unavoidable.
During the start up sequence, after homing, when the build plate is brought up to the nozzle to purge it catches about a cm above the z endstop switch. After the it stop trying to raise it to the platform for that specific command, it returns to functioning normally.
What's particularly confusing, is I cannot make it catch by any other means. If I raise and lower the Z using the console, it never catches. If I repeatedly home the axis with the console, it never catches. Start up a new print? Catches every time.
Previously when this issue would arise, I could use my hand to help the axis past this point, and it would then print normally. It now catches even with my support.
Things I've tried:
-Starting the print without the build plate to see if weight is the issue.
-Manually helping the build plate rise past the point with my hand it different locations.
-Adjusting the starting scrips to move the Z axis in smaller increments, at separate times, and slower via F.
Here's a video of the issue: https://www.youtube.com/watch?v=-yvl6xYH1iM
The video starts with me homing the Z axis with S3D control panel repeatedly without issue. I then raise and lower the z axis by 100 mm repeatedly without issue.
Starting at around :45 I start a print and show how it catches after homing.
Any ideas? The Z stepper was purchased about two years ago, I don't believe it should be at the end of its lifespan yet.
Thanks!
Cameron
I have a M2 Rev. D, upgraded to Rev. E , and upgraded to Dual extruders for this particular issue. It's been intermittent for a few months now, but has recently become unavoidable.
During the start up sequence, after homing, when the build plate is brought up to the nozzle to purge it catches about a cm above the z endstop switch. After the it stop trying to raise it to the platform for that specific command, it returns to functioning normally.
What's particularly confusing, is I cannot make it catch by any other means. If I raise and lower the Z using the console, it never catches. If I repeatedly home the axis with the console, it never catches. Start up a new print? Catches every time.
Previously when this issue would arise, I could use my hand to help the axis past this point, and it would then print normally. It now catches even with my support.
Things I've tried:
-Starting the print without the build plate to see if weight is the issue.
-Manually helping the build plate rise past the point with my hand it different locations.
-Adjusting the starting scrips to move the Z axis in smaller increments, at separate times, and slower via F.
Here's a video of the issue: https://www.youtube.com/watch?v=-yvl6xYH1iM
The video starts with me homing the Z axis with S3D control panel repeatedly without issue. I then raise and lower the z axis by 100 mm repeatedly without issue.
Starting at around :45 I start a print and show how it catches after homing.
Any ideas? The Z stepper was purchased about two years ago, I don't believe it should be at the end of its lifespan yet.
Thanks!
Cameron
Re: Z motor catch and dropping only during startup sequence.
broken wire that only looses connection when the Y axis is moving??
Re: Z motor catch and dropping only during startup sequence.
What's different: the different speeds & accelerations used during homing vs. manual jogging vs. normal operation.
The Z-axis motor must produce plenty of torque to lift that massive aluminum plate. If there's any mechanical binding / crud on the guides / whatever, that'll require enough additional torque to stall the motor when it's running at whatever the critical speed might be.
HOWEVER, the basic problem is mechanical, rather than electrical or configuration, because it used to work just fine.
Remove the screws holding the follower nut to the Z-axis stage, spin it all the way to the top (I think it's attached to the top surface; the one on my M2 is on the bottom), and tape it to the frame so it doesn't spin down on its own.
Lift the Z-axis stage using two fingers on either side of the leadscrew, so you can feel how much oomph the motor must produce. Lift the stage to the top, lower it to the bottom, repeat slowly as you feel where it catches / snags / hangs up / requires more effort, then examine the linear bearings & rods in those regions.
Most likely, slobbering oil on the rods & working it into the bearings will cure what ails it, but if there's a distinct hitch in the git-along near the bottom, perhaps you should dismantle the rods + bearings for a deep cleaning.
-
- Posts: 59
- Joined: Mon Mar 21, 2016 2:18 am
Re: Z motor catch and dropping only during startup sequence.
Thank you as always Ed!
I followed your cleaning instructions and reassembled, no luck. I felt little to no resistance on the linear rails, drowned them in oil anyway.
Is there anyway to chance the accelerations during homing to match that of jogging?
Cameron
-
- Posts: 59
- Joined: Mon Mar 21, 2016 2:18 am
Re: Z motor catch and dropping only during startup sequence.
Another bit of info. I had a thought for another test; manually pressing the z stop with the z axis only half way down the z screw. Same exact issue, immediately catches upon raising the platform post homing. So that proves there is no specific spot that causing it, to motor simply cannot lift the build plate anymore during the startup lift, but can do so at all other times just fine.
Re: Z motor catch and dropping only during startup sequence.
Now that I look at the source again, at least for (obsolete) Marlin 1.1.0, there's only one acceleration setting. Seeing as how homing works, it's probably OK.KaiborgStudios wrote: ↑Sat May 16, 2020 3:51 amaccelerations during homing to match that of jogging
That first move is in the startup G-Code prepended to the model's code by the slicer, right after the G28 Z that homes the Z axis. My startup code is heavily customized, so look for the next G0 with a Z axis value, along the lines of:during the startup lift
Code: Select all
G0 Z2.0 F1000
The real solution will be to limit the maximum Z speed to that value by tweaking Configuration.h and recompiling the firmware:
Code: Select all
#define DEFAULT_MAX_FEEDRATE {450, 450, 100, 95} // (mm/sec)
With the Z-axis limited to what you (now) know will work, no G-Code can exceed a sane speed and it shouldn't stall. Layer changes and long motions won't be quite as peppy as they used to be, but I think "working" always beats "fast". [grin]
-
- Posts: 59
- Joined: Mon Mar 21, 2016 2:18 am
Re: Z motor catch and dropping only during startup sequence.
I appear to have made the mistake of not consulting Occam's Razor, but the problem is solved! I'll just leave a picture of the illusive issue. Thank you for all your help Ed!ednisley wrote: ↑Sun May 17, 2020 1:09 pmNow that I look at the source again, at least for (obsolete) Marlin 1.1.0, there's only one acceleration setting. Seeing as how homing works, it's probably OK.KaiborgStudios wrote: ↑Sat May 16, 2020 3:51 amaccelerations during homing to match that of jogging
Re: Z motor catch and dropping only during startup sequence.
what are we looking at????
Re: Z motor catch and dropping only during startup sequence.
Well, that's obvious! In retrospect, anyway. [grin]
The third item in the
Code: Select all
Debugging Rules poster
You did manage to "Make it fail" consistently, which got you most of the way to the goal.
Next time it'll be something entirely different, though …
-
- Posts: 59
- Joined: Mon Mar 21, 2016 2:18 am
Re: Z motor catch and dropping only during startup sequence.
I need myself one of those posters!ednisley wrote: ↑Mon May 18, 2020 1:59 pmWell, that's obvious! In retrospect, anyway. [grin]
The third item in theover my workbench applies: "Quit thinking and look". It seems I can always come up with a complex fault that doesn't actually match the obvious-in-retrospect evidence, alas.Code: Select all
Debugging Rules poster
You did manage to "Make it fail" consistently, which got you most of the way to the goal.
Next time it'll be something entirely different, though …