RE: VFA Artifacts on X+Y Straight Edges
@sgtcaffran If it was a "Polygon Effect" as they state it would occur at most speeds, in particular it should be approximately linearly better with speed. This effect comes and goes and will come back with different speeds strongly imply some resonance issue. Since the same motors, as per previous posts, will repeatably exhibit the effect if good and bad pulleys are swapped then it's not non-linearities in in the motor angular velocity, it's problems with the pulley geometry.
@jondoe
I think that all these printers may have hit what I think of as a 'valley of instability' that is I think their belt length (from being corexy), motor mass/rotational inertial and extruder mass are all in a region where they will resonate with each other. These use much stiffer metal frames and metal/fibre reinforced joints for their mechanisms allowing resonances that were previously damped on vorons due to their much lower stiffness.
RE: VFA Artifacts on X+Y Straight Edges
This thread went from very interesting to a whole different level, awesome!
My constant thought: why didn't someone try V-belts yet? Would the required tension too much for the Core One or would the maximum acceleration before slipping happens awfully low?
My Fanatec CSW 2.5 simracing ffb Wheelbase is using V-belts and the forces and acceleration seem to be more than enough for a printer.
But it's a very different system with thicker axles and shorter+wider belts, so I don't really have useful knowledge about it.
RE: VFA Artifacts on X+Y Straight Edges
That's exactly what Prorifi3D did, though: https://prorifi3d.com/pages/technical-details
No. Prorifi fixed motor resonance. NOT belt ripple we have in this thread. Completely different artefacts, completely different pitch.
RE: VFA Artifacts on X+Y Straight Edges
What about using some PVC, rubber or other slightly elastic plastic tube on the screw going through the idler in the tensioner? That with blue arrow pointing at it here:
Should be quite easy to test.
Now that's the sort of academic rigour that's needed here! Thank you for this post.
Would an additional vibration damping idler do the trick, or is it better to convert one of the existing idlers? What would a vibration damping idler look like?
I was thinking of redesigning one of the existing idlers to include a vibration damping plastic spring in it that has an adjustable tension and adjustable constraint on the range of motion for the spring so that it can only be active for a few micron of motion. I know the composite filaments (pccf,ppa-gf or cf etc) are both good at absorbing a wide range of vibrations and don't creep anywhere near as much so they would hopefully work.
RE: VFA Artifacts on X+Y Straight Edges
Read the page 😉
RE: VFA Artifacts on X+Y Straight Edges
Read the page 😉
Maybe you should lol. I've read it 5 years ago, it's old news. They solved motor resonance on i3. And their separate solution for the belt ripple applies only to i3 printers. You know, bedslingers. They didn't test double belts on a coreXY. It's also different kinematics.
RE: VFA Artifacts on X+Y Straight Edges
The topic was about using two belts with a phase shift. That's what's also on the page. It was suggested that that wouldn't work but I think it does. Any effects that occur due to teeth meshing might be counteracted by the meshed teeth of the out of phase pulley.
It is, however, not a very practical solutions since there are no easily sourced parts. Same as herringbone belts.
RE: VFA Artifacts on X+Y Straight Edges
@raaz-2 I think that the primary reason is probably due to the open loop nature of the toolhead positioning, trapezoidal pulleys can have progressive loss in positioning errors due to many annoying reasons, toothed belts keep the positioning more predictable That said however the loss of positioning error may be negligible and it was just assumed to be otherwise.
@ersvo Give it a go if you find something and let us know if it helps!
RE:
Since I barely print at high speeds due to either using my 0.25 or 0.6 mm non-HF nozzles for small or thick-walled objects, I'd be willing to test V-belts.
Does someone have the time or knowledge to find fitting V-pulleys and belts?
And ofc, attaching them to the print head could be a bit.. Interesting... Maybe I could cut some teeth into both ends.
RE: VFA Artifacts on X+Y Straight Edges
I once printed on my KE bedslinger with a belt reversed and highly tensioned as a test and it worked fine at low accels and moderate speeds. Fixed the belt ripple for sure! Another option could be a capstan and kevlar lines. It's hard to beat a toothed pulley though for this application, and I think with the right design toothed belt ripple can become a small enough issue that it's considered fixed. The lowest hanging fruit is just getting rid of the belt teeth where they run over idlers where possible. Bambu do this on the A series and I've done it on my ad5m.
Since I barely print at high speeds due to either using my 0.25 or 0.6 mm non-HF nozzles for small or thick-walled objects, I'd be willing to test V-belts.
Does someone have the time or knowledge to find fitting V-pulleys and belts?
And ofc, attaching them to the print head could be a bit.. Interesting... Maybe I could cut some teeth into both ends.
RE: VFA Artifacts on X+Y Straight Edges
I am sure y'all might have seen this already:
RE:
"I finally have a VFA free Core One"? Seems like a bit of a clickbait title. In the video he sounds different: "Did we fix it? I would say: partially" and "This looks like a good temporary fix; let's see what Prusa comes up with."
(His approach is to swap in GT1.5 belts, by the way.)
RE: VFA Artifacts on X+Y Straight Edges
that is not the point, the point is that a solution is possible.
Prusa will be hesitant to go this route as it will result in a huge financial loss. Replacing all belts in all the already sold machines, will be a huge hit in their bottom line.
How they resolve this will dictate how people, myself included, will conduct business with them in the future. This product is not finished and they released a Beta at best...
On that note, my 2nd Core One ALSO has VFA, so this issue is I would say, occurring in ALL core ones. I dont have time to change pullys and do experiments, I paid for a finished product, not to be a beta tester. I will wait for Prusas fix on this but I will be much more careful with their products going forward.
This is my first prusa, and, I have no intentions to becoming a fanboy. I do have to say this printer produces one of the most accurate prints I have ever seen. I am yet to experience an H2d, but I was never able to produce this level of print in all my others bambus.
Now they need to solve this, and also make a mmu that is caught with the times... hopefully we see some news soon.
RE: VFA Artifacts on X+Y Straight Edges
that is not the point, the point is that a solution is possible.
The point (my point at least) is that an improvement is possible. Nowhere, except for the clickbait title, does the video claim to have a full solution. And the guy is still hoping for something better to come from Prusa.
RE:
Despite whatever official solution they offer, I really hope they are willing to include a permanent scaling option in the firmware for those who choose to go the extra mile to swap belts/pulleys. It's a simple and cheap enough fix for the users who are willing, and I'm a bit worried that they'll resist offering a FW accommodation because they don't want to send out replacement parts, and to do one without the other might garner bad sentiment.
The reality is that many of us would be happy to spend our own time/money to do the optional belt swap but would be annoyed by having to deal with the scaling Gcode after each reboot.
-J
RE: VFA Artifacts on X+Y Straight Edges
His final print of the lighthouse model (from his original g-code, because he couldn't find the model I think) shows an extremely clean print in shiny filament. I think he's justified in saying VFA-free for that print - "all the VFA is gone, it is just amazing how clean this Prusa CORE One prints now with the belt conversion. So, did we fix the Prusa VFA issue...hell yeah brother!".
I took his use of the term 'temporary fix' to mean that the 1.5mm belts are a complete fix, but he's waiting for an alternative official fix.
RE: VFA Artifacts on X+Y Straight Edges
that is not the point, the point is that a solution is possible.
The point (my point at least) is that an improvement is possible. Nowhere, except for the clickbait title, does the video claim to have a full solution. And the guy is still hoping for something better to come from Prusa.
The "typ" is also here in the thread, maybe he will comment on it.
I just did phase stepping with the new RC version and am printing a VFA test.
Let's see if anything changes.
If the 1.5 belt proves itself for others, I would also convert.
I don't mind the effort.
But first, let's wait and see what solution Prusa comes up with.
RE:
Despite whatever official solution they offer, I really hope they are willing to include a permanent scaling option in the firmware for those who choose to go the extra mile to swap belts/pulleys. It's a simple and cheap enough fix for the users who are willing, and I'm a bit worried that they'll resist offering a FW accommodation because they don't want to send out replacement parts, and to do one without the other might garner bad sentiment.
The reality is that many of us would be happy to spend our own time/money to do the optional belt swap but would be annoyed by having to deal with the scaling Gcode after each reboot.
-J
Agreed. Maybe they can somehow roll the 1.5mm conversion into an upgrade kit to make a 'CORE One S'. I won't be surprised, or even too critical, if the revised belt tension becomes the official 'solution' with GT2, because that video does show a clear improvement with the higher tension. An upgrade kit with 1.5mm belts+pulleys should not be too expensive, and together with Prusa's customary crystal clear instructions would fit their ethos quite nicely. Some other improvements at the same time might also sweeten the deal for any existing owners feeling hurt by the situation.
This would also mean that they would have to include the correct scaling into the firmware. I'm cautiously optimistic 🙂
RE: VFA Artifacts on X+Y Straight Edges
I really hope it just wont be just a software fix. And if it is just a software fix, I hope they allow us to make permanent modifications to the start g-code or make a scaling option. I would love to have their blessing to fix it myself with whatever belt and pulley we wish.
One thing I really find annoying is the lack of specific info to the parts built into the core one and to stuff like phase stepping.
There needs to be more document ion thats accessible to us.
For example, where would you find out What is the inner diameter of the pulley, how many teeth does it have, ect, how to enable Phase stepping or specifics about the stepper motor.
Honestly I thought everything would be more open, with all documentation in one place.
Blog posts are nice but I feel a proper wiki like Bambulab has would be better.
If that exists and i am too blind to find that info, please send me a link.
RE: VFA Artifacts on X+Y Straight Edges
I did the phase stepping calibration five times and got the following X/Y reduction percentages: 69/77, 35/34, 39/45, 41/46, 64/44. With the last calibration I printed the VFA test (top without, bottom with calibration). The printer is definitely a bit quieter and the print seems to have improved slightly. The calibration still seems to produce lots of variance in results, so I'm not sure how effective it is for my motors.
