Z Skew when printing but not when manually moving

Basic Information:

Printer Model: PrintrBot Metal Plus
MCU / Printerboard: SKR 2
klippy.log (18.8 KB)

Describe your issue:

Hi All,

For some reason, all my prints are coming out 31% shorter than they should.

To test this, I’ve been printing a 20mm cube. It prints 20.15 on the X axis, 20.35 on the Y axis, and 13.7 on the Z axis.

The weirdest issue is that when I manually tell Klipper to move the z axis up 25mm, it moves exactly that distance. Meaning this issue only happens when printing.

I’ve tested this with other models (3D Benchy and some random models) and have gotten the same result.

I originally suspected there was an issue with the Rotation Distance so I recalculated and put a value of 1.6 which (I believe) is the correct rotation distance given that I am using 1/4" -16 single thread lead screws. That resulted in the z axis moving 2x the distance it should have during manual movements. I was unable to properly calibrate the z_offset with that 2x multiplier so I increased the rotation distance to 3.2 which is how I am now able to have accurate manual movements.

Any advice would be super helpful!

Thanks in advance.

Here’s the 20mm cube Gcode incase you suspect my slicer is at fault. I read through it to the best of my ability and it seems like it it’s moving 20mm.
20mm Cube - z step disabled.gcode (559.4 KB)

Hello @AlieGG !

How do you get to that clue?

One rotation is about 1.6 mm, so the the movement is 1.6 mm.

When you say, the cube is about 31 mm short, the corrected value for rotation distance would be about 4.7.
Are you sure about the lead screw pitch?

Hi @EddyMI3D ,

I got to that conclusion by measuring the upward movement of the hotend carriage using a caliper. When set to 1.6, the carriage moved 50mm up when I told it to move 25mm. But when set to 3.2 the carriage correctly moved 25mm up.

I see your point about modifying the rotation distance based off of the percent of height that I’m missing (about 31%). I can try that. It just seems like there is some other issue at play since it moves accurately during manual movements.

As for the lead screw pitch. I haven’t measured it myself, however, I’ve referenced the BOM for the printer which states that it’s using 1/4" -16 lead screws. Digging deeper, I looked for the spec sheet for that lead screw which is where I found the rotation distance of 1.6. Here is the link to it, it’s not the same material but it’s the same screw type: McMaster-Carr

Right now I’m re-flashing my mainsail SD card and updating everything else to see if the issue was related to a bug in the beta version of mainsail I was running. I’ll keep you updated with the result of that.

Is this lack of height (31%) always the same and does it only appear during printing?

Just finished updating everything and it didn’t make a difference.

No, it is not consistent. I’ve printed some models that are 20-19% off.

Yes, this only appears during printing.

Have you checked the driver currents of the drivers?

How fast does the Z-axes acually move?

In the gcode file the Z-axis moves with a speed of 7800 mm/minute (130 mm/sec).

G1 Z0.3 F7800

The Z-axes may lose some steps on the way up. Its quite fast.

The printer.cfg says max_z_velocity = 5 and that should prevent this.

@EddyMI3D you’re a genius!

I set the max acceleration on the z axis to 25 instead of 100 and now it’s printing perfectly!

Thank you so much for your help. :heart: :heart: :heart:

1 Like