XY movement issue

Hello,

what i am seeing is related to the x,y directions on the print bed and best described as: print a square, when the print head travels in one direction, it goes twice as fast as the other direction.

the faster direction is when travel is from -x to +x and +x to -x. example, travel from x10, y-10 to x-10, y-10. is twice as fast as from x10, y-10 to x10, y10
.
what this is causing is on a test square, two opposite sides are smooth, two opposite sides have holes or thin spots because of the fast movement.

marlin 2.1.2, no issues with the z axis movement, here are all the settings related to this in my firmware.

#define XYZ_FULL_STEPS_PER_ROTATION 200
#define XYZ_MICROSTEPS 16
#define XYZ_BELT_PITCH 2
#define XYZ_PULLEY_TEETH 20
#define DEFAULT_XYZ_STEPS_PER_UNIT ((XYZ_FULL_STEPS_PER_ROTATION) * (XYZ_MICROSTEPS) / double(XYZ_BELT_PITCH) / double(XYZ_PULLEY_TEETH))
#define DEFAULT_AXIS_STEPS_PER_UNIT { 80, 80, 80, 137.38 }
#define DEFAULT_MAX_FEEDRATE { 2500, 2500, 2500, 300 }
#define DEFAULT_MAX_ACCELERATION { 1000,1000,4,750 }

delta radius and diag rod length have both been measured.

i am not sure what to check or modify to try and adjust the speed of movement in the x direction on the print bed.

Frankly, I don't see how what you describe is possible. Are you really running a delta printer rather than a cartesian? Post the model of your printer please.

Assuming you do have a delta, the max acceleration is wrong the X,Y, and Z values should all be the same.

The max acceleration was a typo, didn't realize i was touching the touchpad and modified it when i pasted it. should have been 1000, 1000, 1000, 750

it is a delta. an old rebuilt Folgertech Kossel 2020 from 2016 that i decided to resurrect. updated the firmware and configuration settings. did a few prints and not been happy with the results so i have been doing any calibrations i can think of. after doing an extruder calibration and starting flow calibrations i noticed the way the print result was and started watching exactly what the printer was doing. after watching a couple 30mm test cube vase prints, i started to see the speed difference.

i have previously measured and adjusted all three towers so they are all within 0.5 mm height (tightening the screws makes them shift a hair)

i have done multiple G33 auto cals and tweaked everything accordingly.

i am going to try changing my slicer and print application as well.

currently controlled with octoprint server on Rpi and Cura to slice.

ill try using Cura through usb connection to print directly, then ill try prusa to slice and pronterface to print. and see if there is any change.

I decided to recheck the configuration and then flash the firmware again and the speed difference seems to have gone away. i would suspect i had a typo somewhere in my prior config.

Good to hear. Should you ever happen to discover what the problem was. I'd very much like to learn what to cause a different speed in one direction only. I must misunderstand when the delta kinematics are applied.

it looks like it may have been related to your first comment.

i have been going through a couple revisions of my firmware configs and found one that was default_max_accelleration 1000, 1000, 100, 750

quick scanning through the config i missed it the first time. since that's the z value, and i think the z tower is directly in the line of the y axis if I draw x/y axis lines on the print bed, any time the print head was traveling along the y axis direction the issue occurred since the max accel of the z motor was slower than x and y.

Edited 2 time(s). Last edit at 11/02/2023 06:42PM by alzyrian.