Print quality - VFA's?  

RE: Print quality - VFA's?

Oh mate, you wouldn't believe this...

Dragon: Prusament PLA silk, stable fw, 20mm/s external perimeters.
Planter: 3DJake PLA silk, stable fw, 25mm/s external perimeters.
Black thingy: Prusament ASA, IS fw, 45mm/s external perimeters
Rectangle silk thingy: printed at an angle, the same as the angle required for VFA to appear on those other prints on the photos, IS fw, quality - 25mm/s, struct - 45mm/s, speed - 100mm/s.

Those motors are faaar from "no VFA". The screenshot shows the angle at which they occur, also when mirrored along X axis.

RE: Print quality - VFA's?

To add to that - the VFA existing at all is not the issue.

The issues are:

1. Marketing the printer as "no VFA " when they are still there even on 20mm/s, can be seen with harsh light on shiny filaments and easily with no effort on silk filaments. This is false advertising, especially when they can be so easily noticed on Prusament PLA blend (silk) prints, you know, their own filament.
2. The VFA are at their peak on the structural profile, which is the IS successor to quality profile. You know, the supposedly best looking profile meant to be used when you need utmost print quality. The VFA are the strongest there, so strong they can be felt with a physical touch. It's that bad. Not even my MK3S+ had its VFA so prominent I could feel it under fingers. This is a big big problem and they should absolutely adjust the profile to avoid the VFA zone AND look into the printer's hardware again and try to minimise this artefact further. Especially if they market the printer as "no VFA". If their best looking profiles show Bambu levels of VFA with this marketing, there's something seriously wrong there and I do feel cheated a bit. At least make the default profiles VFA free.

Another issue is that increasing speed to avoid the VFA zone only makes the orange peel artefact more prominent on Y axis. Some filaments show it more, some show it less, but it's there and it's affecting everything the Y axis touches - meaning only pure X movements are smooth. I can understand that on +100mm/s speeds, I absolutely cannot understand that on 45mm/s. This should also be looked at, there is something wrong with the motor or the hardware itself. I couldn't find this artefact on other bedslingers ran at those speeds, MK3 included, so something went really wrong with the MK4's Y axis. This absolutely needs to be fixed if they want to maintain their printer being marketed as producing high quality prints.

I do sincerely hope they will look into it. I am in contact with CS about both issues and I really want to help them resolve this. I really like everything else about the printer and I just want to get the quality I was promised in the marketing material. I want this printer to get better.

RE: Print quality - VFA's?

Thanks for sharing your findings Shushuda, much appreciated.

This is seriously disappointing to hear. I'm in the middle of assembling my MK4, and one of the main reasons I bought it to replace my MK3S+ was for print quality improvements, especially in regards to VFA. It's strange that I've seen somewhat conflicted reports on the VFA issue. I've seen comparisons to the MK3 where it appears pretty much gone, and then I've also seen prints and complaints on them still being there.

I'll be testing mine thoroughly to see how bad it is when I finish assembling it. I value print quality above all else, speed means nothing to me if the print is ugly and has more pronounced VFA.

RE: Print quality - VFA's?

I do have to admit the VFA are a lot less prominent on quality profiles. If you're not printing silk nor super shiny filaments, you won't notice. The ones that were present on pure X and pure Y movements on a MK3S+ are absolutely gone on a MK4, period. But they are still there on a specific angle, smaller, but still enough to see them on silks and super reflective filaments.

They also seem to have their peak around 35-40mm/s, which is reached on the structural profile during accels and deaccels. This is why they're so prominent there - the ASA print I did is a great example. This angle is also the one best for cooling, which is how I've noticed it - I oriented my models for best cooling, which made the VFA appear on almost the entire side wall of the print.

While this is disappointing, it's almost invisible on normal filaments on 25mm/s so it can be worked around - I currently run IS structural profiles with 25mm/s external and small perimeters for this.

Another issue is that orange peel. It seems to start appearing at over 40mm/s and only gets worse with speed. So escaping the VFA by going faster (exactly what Bambu is doing, although they need to reach +160 for that, MK4 is WAYYYYYY better in this regard) is easy. But then the issue is bad quality of walls that aren't pure X. This has to be worked on as priority, in my opinion, because this will allow to just increase the speed a bit more to escape VFA, satisfy people that want high speeds and make the motors really "no VFA" as promised.

As for the overall quality compared to MK3S+, there's no comparison. MK4 prints are stellar compared to MK3S+, no questions asked, the extrusion is very consistent, the cooling is great, it just prints good. At least on 25mm/s walls - the VFA are minimal, but it's not a "no VFA" printer yet. They're still there which annoys me on silks, but on normal filaments you need to put some harsh light to see very high resolution lines, so in this case I am satisfied. It's just a shame silks show them so clearly, they shouldn't do so with this marketing.

On +40mm/s tho? The orange peel thing makes some filaments look bad. Some you don't notice that much. This is the true issue that needs to be fixed, especially with the stable IS fw on the horizon. This will allow to completely escape the VFA zone with no consequences.

If you need quality, I suggest lowering external and small perimeters to 25mm/s and keeping all IS accels and other speeds intact. I am getting reasonably fast and clean prints this way as long as I don't use silks. For silks I need to pay attention to the print orientation to keep VFA to a minimum.

RE: Print quality - VFA's?

Thanks for elaborating, I'm relieved to hear that VFA can be mostly eliminated at the right speeds.

I hope I can still get some use of the input shaper by adjusting perimeter speeds. I'll try it at 25mm/s and then adjusting up or down. Are accelerations not a factor at all?

I'm looking forward to seeing what kind of quality I can get, as I've been quite unhappy with print quality on my MK3S+ for the past two years I've had it (my first and only 3D printer), even though I print very slowly, and I'm really eager to get rid of it.

RE: Print quality - VFA's?

Input Shaper is basically ANC for vibrations, it reduces ringing. While it allows for faster speeds and accelerations because it offsets the increased ringing caused by those, it's not necessarily equal to fast printing. I'm using the Structural profile in the slicer, which includes IS activating gcode, and I still print those external walls at 25mm/s and see the ringing just completely gone. It will absolutely benefit you at lower speeds.

I leave other speeds at IS values, accelerations too. IS takes care of ringing caused by these increased accelerations, especially at such a low wall speed as 25mm/s. Internal perimeters and everything else - it doesn't need to be printed slowly, so I leave it fast. It works great like this. I don't really use Speed profiles, so I don't have an opinion on those.

For silks, I use IS accelerations etc, but copied Quality profile speeds onto the Structural profile, all of them but travel. I've noticed the infill doesn't lay down properly at these speeds with silk filaments I have. The increased accelerations are fine, it's just the speed I lower - silks like to be printed slowly.

The biggest improvements you will notice is cooling and extrusion consistency. Parts printed directly on supports are looking magnitudes better than on my MK3S+. In this regard I think you will be impressed.

RE: Print quality - VFA's?

Scratch that, I was wrong. I just didn't print at the right angle. I did a few low poly prints recently and they're all showing ridiculous VFA. At 25mm/s.

This printer is falsely advertised as no VFA when it has more VFA than a MK3S+ even at 25mm/s. Buy something else if you want high quality prints.

https://forum.prusa3d.com/forum/english-forum-original-prusa-i3-mk4-general-discussion-announcements-and-releases/heavy-vfa-on-a-no-vfa-printer/

RE: Print quality - VFA's?

I think the artefacts are not caused by the "anti VFA" Stepper. Because the stepping angle is much smaller. Those ripples are caused by the belt and mechanical resonances. I did some investigation and you can either slow down the speed on the outer perimeter and/or you increase the belt tension - higher than suggested from the frequency app.

RE: Print quality - VFA's?

No. Those are way too thin to be caused by the belt. Belt would cause multiple times wider artefacts. These are the exact stepping angle. They are exactly twice the resolution of the MK3S+ VFA, which had 1.8 motors. MK4 has 0.9. Those are quite literally VFA. They are affected by speed and speed alone. They happen at a specific range of angles that involves both motors.

What you're talking about is a completely different issue that I also had and it was fixed by a new pulley on the Y axis. I've already replaced all plastic parts on the printer with ASA, replaced Y belt, replaced Y bearings, loosened up the belt, overtightened the belt - no difference whatsoever. Belt tension doesn't change a thing. This is VFA only.

Your issue is most likely a pulley. After replacing mine, I no longer get those ripples at all on neither axis at any speed. But I still get the exact same VFA at the exact same speeds and angles.

I've even reprinted this wing with changing speed every 1cm to check. They are 100% VFA.

Posted by: @volker

I think the artefacts are not caused by the "anti VFA" Stepper. Because the stepping angle is much smaller. Those ripples are caused by the belt and mechanical resonances. I did some investigation and you can either slow down the speed on the outer perimeter and/or you increase the belt tension - higher than suggested from the frequency app.

RE: Print quality - VFA's?

Ah, unless you mean the artefacts from the OP. These I think are ringing. IS introduces this weird low resolution ringing on edges, corners and seams, at least in my experience. These don't look even enough to be belt ripples, I think, they look like they happen for a while after an edge and then disappear. Belt ripples would be visible on the entire wall, I think.

Posted by: @volker

I think the artefacts are not caused by the "anti VFA" Stepper. Because the stepping angle is much smaller. Those ripples are caused by the belt and mechanical resonances. I did some investigation and you can either slow down the speed on the outer perimeter and/or you increase the belt tension - higher than suggested from the frequency app.

RE: Print quality - VFA's?

At least my investigation: it is not ringing (sorry used the word resonance) but caused by the belt. I checked the distance between the artifacts and compared it with the raster of the stepper-motor. If you move the print-plate with the hand you feel the ripple of the motor. And this ripple has a smaller distance to the plate than you see on the objects. If you compare it to the raster of the belt it is compatible. And also: I found an improvement by tensioning the belt which is also a sign that it is not related to and VLC of the stepper motor.

Also: those artifacts are on all parts from Prusa. And those parts are printed with high speed (I’m sure about this). And I can find it on my MK3 (with much less effect). MK3 has other motor and is driving much slower.

I think it is caused by a higher regulating current of the motor to enable higher acceleration.

Just my thoughts.

RE: Print quality - VFA's?

Which artifact are you talking about? The one from OP or my VFA?

Because in my case it is VFA. It's super fine, exactly half of the MK3's resolution, they're not affected by belt tension at all, increasing speed in 5mm/s increments changes the placement of the artefact (slightly different angle) and it almost disappears at 55mm/s and completely disappears at 60mm/s. Happens at a specific range of angles. It's like 0,5mm between ripples, it's super thin. It's VFA, or rather MRR.

Another person just did a print for me, same exact settings, even used a marble PLA filament to be as close as possible. 25mm/s. His print has literally zero VFA whatsoever, perfectly beautifully smooth everywhere. Mine has severe VFA. There's something wrong here.

Mine:

His:

Posted by: @volker

At least my investigation: it is not ringing (sorry used the word resonance) but caused by the belt. I checked the distance between the artifacts and compared it with the raster of the stepper-motor. If you move the print-plate with the hand you feel the ripple of the motor. And this ripple has a smaller distance to the plate than you see on the objects. If you compare it to the raster of the belt it is compatible. And also: I found an improvement by tensioning the belt which is also a sign that it is not related to and VLC of the stepper motor.

Also: those artifacts are on all parts from Prusa. And those parts are printed with high speed (I’m sure about this). And I can find it on my MK3 (with much less effect). MK3 has other motor and is driving much slower.

I think it is caused by a higher regulating current of the motor to enable higher acceleration.

Just my thoughts.

RE: Print quality - VFA's?

Take a look at following discussion:

Vibration artefacts on Y-Axis – Assembly and first prints troubleshooting – Prusa3D Forum

RE: Print quality - VFA's?

I made posts in this discussion. This is not my issue. Your artefacts have like 2mm of space between each ripple. Mine have half a millimeter if not less.

Posted by: @volker

Take a look at following discussion:

Vibration artefacts on Y-Axis – Assembly and first prints troubleshooting – Prusa3D Forum