RE: VFA Artifacts on X+Y Straight Edges
I'd show you more photos but, to be honest, phase stepping bricked my printer. I had a smooth print I showed earlier, right? I ran IS calibration today, then phase stepping calibration, then redid the print. I got new diagonal lines I've never had before. I've never even seen anything like that before. I've reset phase stepping calibration. The lines are still there. I've reset the printer to factory settings with the hard reset option. The lines are still there.
RE: VFA Artifacts on X+Y Straight Edges
phase stepping bricked my printer. I had a smooth print I showed earlier, right? I ran IS calibration today, then phase stepping calibration, then redid the print. I got new diagonal lines I've never had before. I've never even seen anything like that before. I've reset phase stepping calibration. The lines are still there. I've reset the printer to factory settings with the hard reset option. The lines are still there.
Just saw your posts on Github too. This seems really strange -- especially the fact that resetting the phase stepping calibration or the overall firmware does not revert the effect. Are you really sure that the diagonal artifacts are caused by phase stepping running wild? Could they be due to some other (hardware) change you made recently?
RE: VFA Artifacts on X+Y Straight Edges
Literally the only things I did between getting a clean smooth print and a print with diagonals are:
- ran IS calibration
- ran PS calibration
That's it. Nothing else.
phase stepping bricked my printer. I had a smooth print I showed earlier, right? I ran IS calibration today, then phase stepping calibration, then redid the print. I got new diagonal lines I've never had before. I've never even seen anything like that before. I've reset phase stepping calibration. The lines are still there. I've reset the printer to factory settings with the hard reset option. The lines are still there.
Just saw your posts on Github too. This seems really strange -- especially the fact that resetting the phase stepping calibration or the overall firmware does not revert the effect. Are you really sure that the diagonal artifacts are caused by phase stepping running wild? Could they be due to some other (hardware) change you made recently?
RE: VFA Artifacts on X+Y Straight Edges
where is phase stepping ? which point in the settings?
Follow me on : twitch-vitaprint3d --- youtube-vitaprint3d --- instagram-vitaprint3d
RE:
where is phase stepping ? which point in the settings?
It's not in the menu yet, but you can run it by putting 'M1977' into a plain text gcode file and 'printing' that file.
RE: VFA Artifacts on X+Y Straight Edges
ok
Follow me on : twitch-vitaprint3d --- youtube-vitaprint3d --- instagram-vitaprint3d
RE: VFA Artifacts on X+Y Straight Edges
I don't want to pollute the GitHub thread as per the dev's wishes, so I will write here. I've reflashed an older firmware, then flashed a new one again. Artefact is gone.
No idea what happened.
I will proceed with final test prints and belt tension for you guys and post them once I decide on what looks best on my printer.
RE: VFA Artifacts on X+Y Straight Edges
Hello all - I am a Mk4 owner and was planning to order a core one kit. I have been following this thread and to be honest, at this point I cannot. I have been very disappointed in Prusa’s absence here. It seems like while I’m sure they are aware of the problems, they seem willing to continue shipping the troubled printers and let the user community flounder around trying to solve it. They have always stood behind their buyers, but it is questionable if this is enough to get the market to continue to support supplying products that are not up to Prusa’s stated standards. I have always believed in Prusa but it is getting increasingly difficult.
RE: VFA Artifacts on X+Y Straight Edges
Just wanted to clarify something, is getting the gantry perfectly square more important than getting the belts tuned to exactly the same frequency? As on my machine it's either one or the other.
RE:
(I used to write firmware for a living a while back) it sounds like the current fw doesn't clean up/reset to original value the area where the phase stepping are stored so, once overwritten by the wizard, they survive intact. Installing another version of the fw and reinstalling the latest one probably triggers a reset (or they get overwritten). Probably one of the reasons why the wizard hasn't made it into the official menu yet...
A notice in GitHub, just in case this particular case has slipped the developer's attention, is probably a good thing.
I don't want to pollute the GitHub thread as per the dev's wishes, so I will write here. I've reflashed an older firmware, then flashed a new one again. Artefact is gone.
No idea what happened.
I will proceed with final test prints and belt tension for you guys and post them once I decide on what looks best on my printer.
RE: VFA Artifacts on X+Y Straight Edges
Hey, so as I promised - I did a speed test. Speeds from 25 to 70mm/s.
Tldr, 45mm/s looks best on my printer. So stock Structural profile.
Best looking speeds on my printer - 30, 35, 45
Sliiight motor resonance - 50
Visible motor resonance - 25, 40
I don't care much about higher speeds as they're no longer reliably shiny all around depending on filament. But you can take a look at them on the photos if you wish. Keep in mind I've been nitpicking them a lot and photos don't capture everything I've noticed in person.
Overall I consider the issue on my printer 100% fixed and can't wait for my MMU3 to arrive. I'm not happy I had to put my own cash into third party hardware to get this printer working well, but c'est la vie. It's perfect now. Well, a bit noisy and resonating at certain speeds, but 45mm/s seems quiet enough. I am happy with the printer now.
Still won't recommend getting it without also getting like 4-8 Mellow/Gates pulleys and trying them all out. If these pulleys solve your ripple, the printer gets actually really really darn good and worth the price. If.
RE: VFA Artifacts on X+Y Straight Edges
Also as promised, here's a before and after, you can tell which is which. The before is one of the first prints, after is from today after all my fixes. Same speed - 45mm/s, stock structural profile.
RE: VFA Artifacts on X+Y Straight Edges
What do you attribute to being the fix? Just changing the motor pulleys or the combination of the toothed idlers and the pulleys?
The fact you sampled multiple motor pulleys, did the replacements all provide better results than the ones delivered from Prusa or were some just better as they matched up to your motors?
RE: VFA Artifacts on X+Y Straight Edges
You mean that after is the print at the top and before at the bottom of each picture?
RE: VFA Artifacts on X+Y Straight Edges
What do you attribute to being the fix? Just changing the motor pulleys or the combination of the toothed idlers and the pulleys?
Tbh I don't know if the idler did anything. The pulleys certainly did at least 95% of the fixing. And overtightening the belt to 95Hz smoothed out the remaining ripples.
The fact you sampled multiple motor pulleys, did the replacements all provide better results than the ones delivered from Prusa or were some just better as they matched up to your motors?
No actually. I've tested 4 Mellow pulleys. One was slightly worse than Prusa. Two were a lot better (the ones I've used). And one was in the middle. And the Prusa pulleys weren't the same either, one was worse than the other.
You mean that after is the print at the top and before at the bottom of each picture?
Yes. Top print is after the fixes. Bottom print is before the fixes, actually one of the first prints I've done on this printer.
RE: VFA Artifacts on X+Y Straight Edges
It's pointing to a possible QA issue with the pulleys, were the 'good' pulleys an improvement whichever motor they were placed on?
Have you been able to determine why the motor pulleys were different? e.g. a concentricity issue, bad bearing, or where the teeth are machined?
RE:
No actually. I've tested 4 Mellow pulleys. One was slightly worse than Prusa. Two were a lot better (the ones I've used). And one was in the middle. And the Prusa pulleys weren't the same either, one was worse than the other.
How did you test them?
Disasembled the printer, changed 1 pulley, reassemble and then test print?
Each time?
How did you concluded it was better or worse? Just on print result?
RE: VFA Artifacts on X+Y Straight Edges
- Loosen the belts as much as you can on both ends. One turn on the left, one turn on the right, one turn on the left etc. Do it in small increments to keep the gantry squared.
- Remove the left side metal panel.
- Remove the left side plastic panel.
- Undo 4 screws that hold the left motor.
- Lift the motor and lay it on its side.
- Remove old pulley.
- Place new pulley.
- Put the motor back in, taking care to catch the belt loop inside the plastic part.
- Screw the motor in.
- Retension the belts the same way as you loosened them but in reverse.
- Print a wall - perfect 45 degrees diagonal that goes from the front left to the back right. The exact same motion as when the printer does solid infill.
- Loosen the belts the same way as in step 1.
- Unscrew the motor.
- Lift the motor.
- Remove pulley.
- Put it aside together with the print. DO NOT confuse them all, you will have multiple pulley-print pairs.
- Repeat steps 7-16 until you run out of pulleys.
- Compare the prints and pick the best looking 2. DO NOT confuse them or you will have to repeat the process.
- Put those best pulleys aside, marking the absolute best with a marker.
- Now you can put those pulleys onto the motors for good. Use the absolute best pulley on the RIGHT motor.
- Reassemble the panels, retension the belts.
The right side metal panel is a bit tricky to remove. Just undo the rivets and gently drop the panel so it pops out and hangs onto the connected PTFE. Leave it like that and you can then reach the motor screws just fine.
RE: VFA Artifacts on X+Y Straight Edges
In what way do the pulleys differ from each other? Is there any visible difference? If not, can you speculate on what might be different?
RE: VFA Artifacts on X+Y Straight Edges
It's pointing to a possible QA issue with the pulleys, were the 'good' pulleys an improvement whichever motor they were placed on?
Yes. The artefact follows the pulley no matter which motor.
Have you been able to determine why the motor pulleys were different? e.g. a concentricity issue, bad bearing, or where the teeth are machined?
No. They look identical to me.
In what way do the pulleys differ from each other? Is there any visible difference? If not, can you speculate on what might be different?
No idea. I can't see nor feel a difference.