Artefacts on walls
-
Anybody got a suggestion where these could come from? you can not realy feel them but can see it in light reflections.
the top one is printed at 80mm/s the bottom one at 40mm/s
my printer is a corexy (hevo fusion) with linear rails.
-
There's been a few good threads about troubleshooting issues like this. Search for vertical ribbing, or banding, or waves.
Usually it's a mechanical issue. Something loose. Belts not meshing with pulleys smoothly. Toothed side of belt running on a smooth idler. Oblong pulley, or not running on center. Bad bearing. those are the things I would check first.
-
Here's my recent thread:
https://forum.duet3d.com/topic/9222/part-wobble-formerly-need-gates-gt2-belt/23
Mostly mechanical, but in my case motor choice/swap helped. I've not finished yet. Dampers, pulleys, and tl smoothers (note: I'm running v0.6/0.8.5, v2) are due over the next few days.
-
Looks like the wobble on your parts are about stepper major step size?
Presume a Duet Ethernet/WIFI?
-
Its a duet maestro.
it could be the stepper major size.
i will try swapping my x and y stepper motors with the z ones.
-
Alongside the already popular polymer dampers there is also mass inertial dampers to consider. That said I've not heard of these being used on printers yet and would limit the max accelerations the steppers could achieve without stalling. Whether that effects the accelerations and instant speeds used in printing is another question.
-
Are you already running on the current limit of the Maestro or 80% of stepper max current whichever is smaller?
-
I am currently running my steppers at 67%
I can give 80% a try.
-
after a long time of testing and adjusting
chaining to different motors -> Still there.
switching to 16 tooth pulley -> Still there but small ripple distance decrease
using Spreadcycle instead of Stealthchop -> Still there , just more motor noise
chaning all toothed idler to smooth ones and twisting the belt -> Still there.
modifying some parts for better belt alignment -> Still there.
upping the current -> Still there
relubricating all the rails -> Still there.i then rotated the part by 45 degrees. this causes the part to be printed by only one motor at a time.
-> The ripples decreased in distance.So i guess i will some 0.9 degrees steppers and combine them with a 16 tooth pulley and see if that helps
-
What are the motors?
-
Model 17HS8401
Step Angle 1.8 (deg)
Motor Length (mm) 48
Rated Current (A) 1.8
Phase Resistance (Ohm) 1.8
Phase Inductance (mH) 3.2
Holding Torque (N.cm) 52
Detent Torque (N.cm) 2.6
Rotor Inertia (g.cm2) 68
Lead Wire (No) 4
Motor Weight (g) 400
-
What type of printer is it, which firmware version are you running, and what is your extruder steps/mm ?
-
its a Hypercube Fusion CoreXY with linear rails.
Duet Maestro with Firmware Version: 2.02(RTOS) (2018-12-24b1)
Extruder Steps are 415 for Bondtech BMG
-
I'm not and expert on CoreXY machines. I've seen a number of threads about similar artefacts on CoreXY machines. AFAIR the pitch of the artefacts generally matches the belt pitch, and the cause is generally friction. Using non-toothed idlers may also contribute.
Until you resolve this issue, I suggest you use spreadCycle mode for the X and Y drivers, also run them at higher rather than lower current.
-
the steppers havent arrived yet.
but this seems to be a common problem
see these benchys from a prusa.
https://youtu.be/FAXUjZZER5E?t=563
-
@veti said in Artefacts on walls:
the steppers havent arrived yet.
but this seems to be a common problem
see these benchys from a prusa.
https://youtu.be/FAXUjZZER5E?t=563That looks like plain old ringing to me though.
-
ok my 0.9 stepper motors arrived.
Looks ok from this angle
Same Cube, different angle, but still a lot better than before
However now i am getting layer shifts (as seen at the bottom layer), so i have to do some more tuning.
-
the layer shifts were due to the lowered stealthchop -> spreadcycle switchover.
its not that important but i think it would be more consistent to specify the V Parameter in M569 as mm/s rather than clock cycles
-
@veti said in Artefacts on walls:
its not that important but i think it would be more consistent to specify the V Parameter in M569 as mm/s rather than clock cycles
Then we would need to reprogram the stepper drivers when you set the steps/mm. So I compromised with setting the TPWMTHRS value but reporting the corresponding speed too. Also bear in mind that these are belt speeds, which for CoreXY, delta. SCARA| etc. printers are not the same as head speeds.
