RE:
Any shiny PLA (silver/bronze) seems to show it well
https://drive.google.com/file/d/1nb2EncvBSVV_NSiCUyAXijazUL4r6CNP/view?usp=drive_link
RE: VFA Artifacts on X+Y Straight Edges
Looks exactly the same on my P1S, have to print below 50mm/s or above 160mm/s.
With my Voron Trident i have absolutely no problems like this, smooth as butter.
I have a Core One on order, but reading forums carefully.
Somehow i have more confidence in Prus to solve this than Bambu (they'll never solve this).
RE: VFA Artifacts on X+Y Straight Edges
Any shiny PLA (silver/bronze) seems to show it well
https://drive.google.com/file/d/1nb2EncvBSVV_NSiCUyAXijazUL4r6CNP/view?usp=drive_link
I didn't have a silver filament but I used some shiny black PLA.Here is the test print:
The quality isn't that great because 230C was way too high for that filament and print speed. As you can see there are some light VFA's at X2 and a little less on X1. Interestingly I didn't see any on my other prints.
I did some more testing and I found out there are some noticeable VFA's at around 80mm/s. Coincidentally that is the exact speed at which the machine makes the louder "vibration" noise, like the XL did before it got the phase stepping update.
I think every belt driven 3D printer has some VFA's (more or less) at a specific speed, only the the exact print speed differs between the machines.
Have you tried to print with different print speeds? I found this VFA test gcode someone made for the Core One: https://www.printables.com/model/1189256-vfa-test-prusa-core-one
For me the stock structural and speed profiles with 45 and 200mm/s wallspeed are actually both good to mostly avoid VFA's.
RE: VFA Artifacts on X+Y Straight Edges
After swapping the smooth pulleys for toothed pulleys on the x axis of my XL, I can confirm that there is no noticable difference in the print quality.
As a disclaimer my VFA's were not bad to begin with. I mainly did this upgrade because I wanted to replace the motor mounts with PCCF, so I went ahead and upgraded the o-rings and the pulleys at the same time as it didn't make sense to me to run the toothed portion of the belt over the smooth pulley.
Based on my XL, not the Core One, I don't think having the smooth pulleys on the X axis are the problem. I suspect motor tuning as some others have suggested.
RE: VFA Artifacts on X+Y Straight Edges
I suspect motor tuning as some others have suggested.
But on the XL, don't you have phase stepping (and the accelerometer-based calibration for it) already in place? And it still leaves you with some VFAs, I understand, though not terribly strong ones?
RE: VFA Artifacts on X+Y Straight Edges
I suspect motor tuning as some others have suggested.
But on the XL, don't you have phase stepping (and the accelerometer-based calibration for it) already in place? And it still leaves you with some VFAs, I understand, though not terribly strong ones?
That's correct, however an additional difference is the use of 1.8° motors on the XL vs. .9° motors on the Core One, so the final print quality might not be the same. Prusa promoted that switching to the .9° motors on the MK4 and now on the Core One, improved the appearance of VFA's.
I think my main conclusion here was that the smooth pulleys on the X Axis don't seem to be a major contributing factor to the VFA's.
RE: VFA Artifacts on X+Y Straight Edges
I think my main conclusion here was that the smooth pulleys on the X Axis don't seem to be a major contributing factor to the VFA's.
Fair enough. As I recall, the problem with smooth pulleys is not that they are the Great Satan of VFAs, but that they introduce another harmonic into the system resulting in certain speeds (which dovetails with your point) exhibiting artifacts where good quality toothed idlers do not. And they also potentially increase wear over properly fitted, quality toothed idlers/pulleys. It's a fine point, perhaps, but IMO stacking tolerance mitigation is absolutely a factor unless you're only printing penis keychains to sell to people on reddit.
RE: VFA Artifacts on X+Y Straight Edges
Any shiny PLA (silver/bronze) seems to show it well
https://drive.google.com/file/d/1nb2EncvBSVV_NSiCUyAXijazUL4r6CNP/view?usp=drive_link
I didn't have a silver filament but I used some shiny black PLA.Here is the test print:
The quality isn't that great because 230C was way too high for that filament and print speed. As you can see there are some light VFA's at X2 and a little less on X1. Interestingly I didn't see any on my other prints.
I did some more testing and I found out there are some noticeable VFA's at around 80mm/s. Coincidentally that is the exact speed at which the machine makes the louder "vibration" noise, like the XL did before it got the phase stepping update.
I think every belt driven 3D printer has some VFA's (more or less) at a specific speed, only the the exact print speed differs between the machines.
Have you tried to print with different print speeds? I found this VFA test gcode someone made for the Core One: https://www.printables.com/model/1189256-vfa-test-prusa-core-one
For me the stock structural and speed profiles with 45 and 200mm/s wallspeed are actually both good to mostly avoid VFA's.
Very interesting Boris! Thank you for doing that.
Surprisingly your artifacts don't look a million miles away on the X axis. My Y axis is definitely more pronounced.
I have printed a VFA tower with my results in this thread. Certain speeds definitely VFAd more than others but wasn't expecting this on a default print profile.
I'm actually now more worried that they won't fix it!
RE: VFA Artifacts on X+Y Straight Edges
I really hope they do. Hopefully people return their printers so Prusa can have a look.
If you read the Bambu forum about this problem you don't get your hopes up for that printer, the forum dates back more than 2 years and all Bambu can say is: retighten your belts....
RE: VFA Artifacts on X+Y Straight Edges
I am now on the wagon of having an accelerometer can make improvements in print quality, assuming you can calibrate the phase stepping with it, and not just the input shaper values.
As I experimented more today my Y axis on my XL looked similar to the Y axis on the pictures that @boris-3 posted. When comparing to my test print before replacing my parts with PCCF, the print before the change did not have these lines.
So I experimented with different belt tensions repeating the test part each time. I saw a slight improvement with lower bent tension, (around 80hz on the belts) but it was still worse than before making changes. So I decided to rerun phase stepping again since I had changed belt tension, and the result was much improved. I plan to test tighter belts again and running phase stepping again because before the change I had been running about 90hz on the belts.
What this tells me is that belt tension can affect the phase stepping values enough to improve VFA's. Now remember, I'm not talking input shaping, which is related to machine vibrations, but phase stepping which is specifically aimed at reducing motor vibrations.
So I'd recommend buying the accelerometer, as its pretty cheap.
Can anyone verify if you can currently calibrate the phase stepping on the Core One?
RE: VFA Artifacts on X+Y Straight Edges
I bought the accelerometer but on the Core One you can currently only calibrate input shaping, not phase stepping. No difference changing the input shaping.
RE: VFA Artifacts on X+Y Straight Edges
I have VFA's too on my core one (as well as the ugly sounding resonance sounds). I found that turning my model 45 degrees completely eliminates the sound (as well as printing at a faster speed) and also helps with the VFA's, but they're still there. Nothing as smooth as my mk4s+.
After learning that the gantry actually requires both motors to turn to keep straight movement (and if one motor moves it moves in a 45 deg angle) I'm starting to wonder if this gantry design is fundamentally flawed. The mk4's just needs to spin one motor to make straight movements.
I've tried belt tightening, and I have the accelerometer kit as well, which made no difference.
Kind of a bummer, the core one hasn't been impressive to me so far. This is in addition to the stupid bug where if you pause a print and then resume, it won't retract the filament (which causes oozing), and then finally it resumes at an incorrect z level which then results in all hell breaking lose. You essentially have to end the print after that.
RE: VFA Artifacts on X+Y Straight Edges
I had high hopes for the Core One, but i already have a printer with lots of VFA, which i wanted to replace with this one...
Think about cancelling my order and wait for a solution before purchasing.
RE: VFA Artifacts on X+Y Straight Edges
I would keep your order open or you'll end up back of a long queue.
I don't doubt prusa are working on a solution for the problems, phase stepping, QC etc. They must be absolutely overwhelmed with shipping a brand new product but the blanket silence response to these problems isn't instilling me with confidence
RE: VFA Artifacts on X+Y Straight Edges
They must be absolutely overwhelmed with shipping a brand new product but the blanket silence response to these problems isn't instilling me with confidence
Yes, probably they are overwhelmed. But I think Prusa is missing a major opportunity by totally ignoring this forum, which seems to be a general policy of theirs. They should not only follow the threads, but also post updates here.
Potential buyers of the Core One will be reading here, and get to read only about problems, not about a single solution from Prusa (or even a path towards a solution) yet. I bet Bambu Labs are reading here and are having a blast...
RE: VFA Artifacts on X+Y Straight Edges
They must be absolutely overwhelmed with shipping a brand new product but the blanket silence response to these problems isn't instilling me with confidence
Yes, probably they are overwhelmed. But I think Prusa is missing a major opportunity by totally ignoring this forum, which seems to be a general policy of theirs. They should not only follow the threads, but also post updates here.
Potential buyers of the Core One will be reading here, and get to read only about problems, not about a single solution from Prusa (or even a path towards a solution) yet. I bet Bambu Labs are reading here and are having a blast...
I too really wish Prusa would chime in with at least an acknowledgement of the problem and the fact they are working on it. I have several MK4S to Core One upgrades on order, but am starting to feel a bit worried after reading all these posts. I know these forums tend to amplify the bad issues though, and am also hoping that some of these issues get resolved within the next month or so.
RE: VFA Artifacts on X+Y Straight Edges
I've been fighting this for a while. It's called belt ripple. VFA is much finer and is quite reduced with 0.9 motors. Tension makes a difference, so does speed. Tooth vs smooth idlers didn't make a difference for me. It has to do with the resonance of the motors. Some users have found that their pulleys are not concentric and has helped when changing to a concentric one. I've changed all my pulleys 3x over with genuine and Aliexpress specials and it made little difference. Try a speed tower, I think higher speeds will make the largest improvement.
RE: VFA Artifacts on X+Y Straight Edges
Not that it's of direct help to the topic, but interestingly the Peopoly Magneto X has the VFA issue as well, and that has no belts at all... Is it possible the belts aren't the primary factor here but just influence how the issue presents itself?.
RE: VFA Artifacts on X+Y Straight Edges
I'm thinking it can be a combination of things, but primarily some type of motor resonance. That's why the MK4 with it's .9° stepper motors, Prusa said drastically improved VFA's (and the print quality on the MK4 proves it imo). Now why those same motors in a core XY setup don't seem to work as well I suspect comes down toa combination of motor tuning and the fact that the belts are much longer, which change the dynamics of the system. We'll have to wait and see, but I suspect there will be some kind of software improvement in the future.
RE: VFA Artifacts on X+Y Straight Edges
Not about the Core One, but my XL.
After straightening the CoreXY gantry (I used the margin of error on all screws—try this at your own risk), I finally achieved a perfectly aligned X-axis with both belts tensioned to 85Hz.
After dialing in the optimal speed, flow, and temperature settings for PETG, I’m now getting acceptable VFA on perimeters parallel to the X-axis and a really nice finish on one wall parallel to the Y-axis (printing from front to back). However, the second wall still shows VFA on my X- and Y-symmetrical vase. Tomorrow, I’ll contact Prusa support because I suspect one of the steppers might be faulty.
5 kg of PETG and around 100 hours of tinkering and slicer adjustments later, I’m really disappointed with the so-called legendary Prusa quality. I do a lot of tinkering to achieve the best quality—even with Bambu printers—but with Bambu, I’m refining the print to look better than injection-molded parts. With Prusa, I’m just struggling to get anything decent at all. 😅