4
$\begingroup$

0.200 mm layer height, single wall

0.075 mm layer height, 2walls, no infill

  • These lines exist on all prints, PLA, ABS.
  • They're 0.8 mm apart [20T 2GT pulley, 1.8°/step motor = 1 line/4 full-steps]
  • Start to disappear above 75 mm/s but will still appear on slower axis when printing diagonal lines or curves
  • Custom built frame, cross bar (Ultimaker style) using linear rails
  • dual-motor (4 total on X-Y) + separate driver (1 motor/driver)
  • Running Smoothieware on Smoothieboard 5X [A5984 drivers, 32 microstep]

Tried all these with no improvement:

TL;DR problem is somewhere between drivers and motors

  • Switched to 6.625:1 geared extruder
  • Tried parallel, series, single coil on the 8 wires motor and 2 other different motors
  • Enable fast decay mode on A5984
  • Motor current from 0.5 A to 1.8 A
  • PLA temp from 170~200 °C
  • Acceleration as low as 100 mm/s2
  • **Changed 20T to 16T pulleys. The pattern scaled down proportional to the change in tooth count. Ruled out mechanical issues.
$\endgroup$
6
  • $\begingroup$ Have you tried 1/16 micro steps? $\endgroup$
    – 0scar
    Jan 20, 2019 at 13:26
  • $\begingroup$ No, the drivers are soldered on-board and require tiny SMD resistor soldering to change microstepping, no jumper. THB6128 drivers and expansion board on the way. $\endgroup$ Jan 20, 2019 at 15:34
  • $\begingroup$ that looks like ghosting... $\endgroup$
    – Trish
    Jan 20, 2019 at 15:52
  • $\begingroup$ I've tried 100mm/s2 acceleration and it still appeared. Also they always appear perfectly vertical near curved edges. Definitely not ghosting. $\endgroup$ Jan 21, 2019 at 6:16
  • $\begingroup$ @0scar tried 1/16. Did not help. $\endgroup$ Jan 21, 2019 at 6:42

3 Answers 3

1
$\begingroup$

The fact that these are all perfectly spaced, and don't mirror the edges of irregular prints, makes me think it's definitely not ghosting. That said, I can't see the Y direction on either print, just the X direction, so this all assumes it's only happening in one direction. One thing to think about: Your motors have typically 2 opposing coils, and they get activated by taking 4 steps: (North, off, South): N/o o/N S/o o/S If these are spaced out exactly 4 steps apart, that would imply that one of your coils is either underpowered or overpowered on the motor controlling that direction's movement. That would lead to your motor torque dipping and increasing, leading to slightly uneven print speed. This is 100% speculative and might be a goose chase since you've got 4 X/Y motors and it seems to happen in both the X and Y axis. The chances of having that many motors exhibit the same deficiency is astronomical.

That said, I've got little experience with multiple motors per axis. Another thing you might look into is whether the motors are fighting each other at all. If, for example, the motor-side pulley isn't aligned exactly the same way on both X motors, or the motors get out of sync with each other, because of the way the motor's holding torque falls as you get away from a full step position, you might find that one motor is holding the other back slightly, or pulling it forward towards the nearest full step. Again, this is all speculative, but it might be worth looking into. You can typically figure out the full step location by killing power to the machine and letting the motors settle into a full step on their own without the belts or other drive mechanisms attached. I'd unhook the belts, kill the power, get the motors settled (with a bit of a nudge if necessary), and then see if your belt perfectly settles into both pulleys in that location. You might find that the belt teeth don't quite line up on both pulleys, and the only way I can think of that would fix that specific problem is spinning the motor until it matches, or even physically relocating the motor closer or farther relative to the other on the same axis.

YMMV, best of luck.

$\endgroup$
0
$\begingroup$

I know this is a pretty old discussion, but I had the same trouble recently. In my case, it think it could be a pattern caused by the extruder gears biting into the filament. I found that the problem seemed to be worse with increased retractions. That makes some sense to me, the more retractions you let the slicer make, the more the extruder gears are going to go back and forth through the same length of filament.

To try and confirm this, I turned off retraction completely and cut off and removed the filament that was already past the extruder (this length of filament was presumably marred by many retractions, the filament outside of the extruder would've been clean). My resulting print still had the pattern but not nearly as contrasted and obvious.

I did try adjusting accel, jerk and print speed and the pattern didn't seem to change in frequency with any of that, so that led me to the extruder gears which are constant. I was also going to test if my hotend fan was contributing vibration artifacts but haven't gotten around to it.

Not sure there's much else that can be done with this. Maybe some filaments would have less of an issue than others. It's possible the issue is more obvious on a dual gear extruder? A single gear extruder will only bite into one side, which might be a bit more hidden.

$\endgroup$
2
  • $\begingroup$ Welcome to 3D Printing SE and thank you for your contribution. When you get a chance, please take the tour to understand how the site works and how it is different than others. $\endgroup$
    – agarza
    Oct 25, 2021 at 21:09
  • $\begingroup$ Do you have any photos showing the difference in the pattern with retraction on and off? It would be a useful illustration of your point, if you could edit and add them to your answer.. $\endgroup$
    – Greenonline
    Oct 26, 2021 at 4:45
0
$\begingroup$

This is a great blog post, Taxonomy of Z axis artifacts in extrusion-based 3d printing, explaining what I think I am seeing, although I'm not entirely sure what I'm seeing in your photos.

The short version is if your z-rods are threaded in an imperial unit, inches/fractions of inches, instead of metric, you get a repeating decimal value (think about converting 1/3 to decimal), which can lead to a rounding error depending on the chosen layer height.

Without knowing how exactly you built your machine it's hard to say more but the post goes over a number of things I almost did wrong (like anti-wobble caps on top of the z rods for Prusa i3s...)

$\endgroup$

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .