Weird Z axis issue

Hi all,

I assembled a Mendel a few days ago and noticed a very strange behavior of the Z axis.

While printing a layer, sometimes the left Z axis stepper goes one step forward while the right stepper goes one step backwards. This results in the X axis not being level after some printing and screwing up the prints. This glitch happens mid-layer, not when any Z move is supposed to happen.

It's particularly visible while printing the simple .5mm thick square wall. It draws the first two segments fine, but when it starts the third segment, the Z steppers have this glitch. I thought this could be hardware, like too low current for the motors, but after some experimentation I noticed this never happens if there is no subsequent height change in the gcode (I modified the square wall gcode to remove all the Z changes).

If I manually send the gcode layer by layer, waiting for a layer to complete before sending the rest, the glitch doesn't happen either, so I'm thinking this could be a firmware issue maybe? I'm running Marlin 1.0 on Melzi electronics.

Any ideas?

Thanks,
Theo

Did you ever solve your problem?

I have the same issue right now, except I'm in the calibration phase and haven't done any printing yet. I'm set up with RAMP 1.4, Sprinter firmware, and using Pronterface to carry out the movement, and I've noticed that clicking the +1 button repeatedly on the Z axis produced fairly random results. Sometimes the motors move in the same direction, other times they are opposite, and again sometimes they reverse. I don't know if anyone else has had this issue, but it's driving me mad!

I think I've ruled out the soldering on the RAMPS as I have swapped the z-axis plug onto the x & y and it still produced the same results. X & Y axis are both working fine, extruder works great, heated bed heats up wonderfully! Any suggestions are greatly appreciated! I'd love to get printing!

Travis

I have a similar issue on my machine, but I think I know what it is. After I calibrate my machine, it works fine for a while, but eventually the Z axis height will change. I adjust my endstop to compensate, but this only works for a while, Eventually I hear the sound of missed steps from one of the motors. The axis is no longer level and I need to go through the calibration steps again.

In my case, I think one of the Z motors is silently losing a step periodically. Since the Z axis only makes large moves at the start of a print when it is homing, this is not a major issue initially. Over time, the cumulative effect of losing steps results in what I see.

My Z motors are hardly warm at all, so I think I'm going to try turning the pololu trimpot up a bit to see if it helps. Otherwise it's probably a result of the Z axis components being slightly out of alignment (although I DO have Z constraints top and bottom on both threaded rods).

I don't think this will help 'thedude', but for Travis and jbernardis, make sure your Z constraints (nuts) are all the way in their collars. For example, if you can manually grab your x-axis chassis and lift it up and down along the Z-axis, even a small amount (like 3mm - 5mm) then that may be the cause of your unpredictable z-axis behavior. At least that's what got me.

My fix involved taking off the z-axis motors, and repositioning the top and bottom restraining nuts (on the z-axis threaded rods) so that they were snug against the separating plastic within the collar (this can be tricky, you have to 'unthread' the x-axis chassis all the way to the bottom of the threaded rod, make sure the top nut is flush against the bottom of the collar, then put the bottom nut flush against the top of it's collar, and thread the x-axis chassis back on the threaded rod. Confusing?). Since I've done that my z-axis has been dead-on with every print. Before this fix, I felt like I would spend 15 - 20 minutes before each print just trying to get the z-axis in a decent starting position, then it may or may not have worked for the print - very frustrating. So, I hope this helps you guys.

Hey guys,

Ya, I found out what was my problem. The firmware on my electronics was configured to release the Z steppers during a layer and only activate them when used to change height. This caused the glitches when the steppers were powered back on.

Edited 1 time(s). Last edit at 11/26/2012 06:12PM by thedude.

Thanks for the responses guys! Cheeseweasel, I think I will have to explore your z-axis fix as I certainly have lots of play in between the top and bottom nuts (likely inches of play now that I attempted the removal of the top nut to see if gravity was enough to force down the axis). However, I have the motors off right now and have some tape on them to see the rotation and I see the randomized direction changes at this point. I'm not sure why the motors would turn in opposite directions...they are connected in serial and should receive the same signal provided enough current is sent to get to motor 2. I'll take apart the z-axis this week and rethread them...as I obviously missed the requirements for precision at this point.

Is everyone else using the serial connection of the z motors? I know the old Mendels likely were configured for parallel connectivity and I wonder if I should attempt to go that route if nothing else works.

Travis

hello theDude.


Ya, I found out what was my problem. The firmware on my electronics was configured to release the Z steppers during a layer and only activate them when used to change height. This caused the glitches when the steppers were powered back on.

where I can find this in the firmware?
I use ARDUINO, marlin / sprinter.

thanks,Gert

Gert,

You should see the settings in the Configuration.h file for Marlin. They should look something like this:

// Disables axis when it's not being used.
#define DISABLE_X false
#define DISABLE_Y false
#define DISABLE_Z true
#define DISABLE_E false // For all extruders

Bobby

Bobby,
I've found it, thank you very much
solves a problem.

Gert