One user’s lack of results does not mean nobody would benefit from it. Other users have had benefits, and I had a significant benefit from wrapping my pulleys with teflon tape to dampen the tooth impact. Since this is a reversible (though tedious) process, I would recommend testing that and judge from the result whether it is worth pursuing the process of installing toothed pulleys.
For testing purposes, a single wrap of electrical tape is easier to install/remove, but the overlap is pretty much unavoidable and will introduce a new artifact, but should demonstrate whether the 2mm artifacts are reduced.
It is important to test the result identically, in terms of the model, and sliced file. The 2mm VFA is most intense on travels parallel to the X-axis near the rear of the machine. The artifacts diminish near the front, so if you were to run a VFA test tower on the front and back you would find differing results.
The best test IMO is to run a square with rounded corners as a vase mode with a speed change every 2-3mm.
Except that, unlike all the other Orca Slicer tests, the Orca Slicer “tutorial” for VFA is non-existent. I can run the test, but then what? Seriously, where do I find how to interpret it?
You look at the model and see which directions and speeds have better / worse surface finish.
If you make changes to your machine, you can reprint it and see if anything changed. Pretty simple. I don’t know why you would need a tutorial for that.
I do not like the Orca Slicer VFA test in this context - we do not really need to see the angles at all as the VFA is happening on X-axis, and that specific aspect of that test does make it a little hard to tell just what is going on. It works, but results in printing more stuff that is not necessary for the efforts.
My preferred way is to make either a square with rounded corners and run layer height modifiers all the way from bottom to top with an unlimited volumetric flow to ensure it’s accurate to the speeds. That can either be one huge one that spans the entire bed or 4 small ones, biased toward the 4 corners.
I also favored a C-shaped version of same which gives bidirectional test option so you can see left to right as well as right to left, not that it seemed to have any consequence in my observation.
Based on the layer height modifiers or the values you’ve chosen in the Orca Slicer it is pretty simple matter to see at what speed the VFA manifests the worse and either avoid it or use that as a benchmark to test troubleshooting efforts.
It genuinely is pretty simple, so maybe @NeverDie is overthinking it?
I disagree about the VFA test. You may be observing the worst VFAs along the X axis (0* in the OrcaSlicer test), but that doesn’t hold true for everyone. At some speeds my worst VFA is at 30*.
With respect to 2mm VFA, it only occurs on the X-axis. It manifests on the diagonals as well, but only as they have an X-motion component. It is otherwise quite negligible and the rectilinear VFA test is more concrete to assess whether conditions have improved or worsened for 2mm VFA.
I disagree. It is visible on pure Y axis moves for me as well (only at certain speeds, and less prominent).
So, it seems worthwhile to test different angles (like the O.S. test provides), so you can make sure that different speed combinations of each axis are tested, accounting for constructive/destructive vibration.
Again, my worst 2mm VFA are at 30*, not 0*, which proves my point.
I disagree - it proves nothing. A 30 degree travel still has a large X component so it is being affected by the X VFA problem. Whether there is little or no ripple on Y travels is irrelevant here as it is a thread specifically dealing with 2mm VFA on X-axis, not Y. The treatment for the problem will not be resolved by methods discussed here.
I remember having seen those 2mm VFA on the railcore discord some years ago. There was an entire channel dedicated to that topic. It was observed, that using fishing lines instead of GT2 belts removed the pattern. The fishing lines were too flexible though to be used to permanently.
It’s not about “proving” anything. It’s about having a more comprehensive before/after test. Even if X-axis specific VFAs are the subject of this thread, what people really care about is VFAs in any direction. Why would you evaluate only one direction out of many?
Because VFAs on the X-axis are the ones of main consequence on the Bambu printers and attempting to resolve all of them muddies the water than the fundamental understanding of WHY the X-axis ones are occurring had not previously existed. You were also the one who claimed you had proved something, not me, and I am objecting to your statement of proof.
Just wanted to know if anyone had any update or new information for this issue? I saw a new post on Facebook of a new member reporting this exact same issue we are experiencing…
Personally it really ruins my experience so far with a Bambulab printer.
I cant sell any prints as they all show VFA in PETG and PC. And for my personal use, PLA is of no use (engineering). Printing at over 200 mm/s is also really not a solution since it reduces tolerances accuracy.
Bambulab really needs to give us a better answer than printing fast. Since the time this post was started, they had far enough time to do some investigation on the issue.
I had to replace my X axis after the belt retaining posts snapped during removal and that didn’t improve it at all. At the same time I was going from a genuine Gates belt to a chinese continuous belt that I split. Collectively there was little improvement in the print, but the generic belt had significantly less VFA than the Gates. After replacing the X axis and belt, I ran tests and once again went through things such as changing tension, belt tracking, etc without improvement. At last, I’ve gone back to wrapping the final pulleys with Teflon tape and this is the only thing that has improved my print quality. More than ever before, I believe replacing the smooth idlers with toothed idlers would probably fix the problem to a major degree, but it is an involved process.
Still dealing with this and regretting not swapping pulleys before installing the axis.
Has anyone else done the swap in addition to @Alkerion@APEX86 and @Tayas?
No need to worry about offset, the belt will be at the top of motors pulleys, no issue at all.
The belts on A1 and Qidi are interesting.
Seems Bambu (who will never officially accept they made a mistake with non toothed pulleys on toothed belt) finally changed their mind.
Interesting belts. Wonder if it’s possible to get a version of these made in the 2mm belt for the X1. Seems like Bambu could easily have these made for that platform. But yes I agree, it will be interesting to see what they do for the XL.
I wonder if we can overmold the last bit of toothed section with silicone?
That’s a pretty fantastic idea. Or since I’ve found a printed pulley wrap with a thin 0.1mm backing to be pretty compliant, it could be adhered to the belt with minimal dimension alteration and be reversible.
That said, when I wrapped the belt with vinyl and with gaffer’s tape, I didn’t see much change except that the thicker gaffer’s tape pulled the head to one side, but still showed the VFA. That kind of also has me wondering if it’s an extruder vs linear motion thing so I’m doing a couple tests printing at 0.3 and 0.6 wall thickness to see if the spacing on the lines changes at all.
