on jerk motions (high accel.) how do you plan on preventing step-loss
I removed the teeth from the belt only in the place where it runs over a smooth pulley in a place where they are not needed. I donāt know when Iāll get around to it, I donāt think the main problem is in this place where the toothed belt hits the smooth pulleys.
I would say that it is a combination of the toothed pulley, belt tension, and how square the printer is. We saw in earlier posts that properly squaring the printer when retightening the belts does seem to help.
This seems like a decent guide for people to follow, with regards to how square your printer is and some belt tensioning tips.
1 Like
Ok. I tried to square it properly and retighten the belts. It is marginally better (really hard to see, but my banding wasnāt severe to begin with), but it is still there and only on the X axis. So the geared pulleys might be onto something. But I am definitely not disassembling the axis just to replace the pulleys. But I donāt understand why they arenāt there by default. I guess my Voron has them there for a reason.
The simple fix (not perfect but doesnāt require a total teardown) is to wrap 6-7 random lengths of teflon tape onto the final idler pulleys. It is not as effective as a new pulley would be, but it is a HUGE improvement.
I verified my printer is square, attempted variations on belt tracking, adding/removing mass to places, high/low belt tension, etc. Nothing solved the problem at all except for altering the final pulleys.
1 Like
They are mechanical, for me at least. I know this because Iām printing with the same exact files and filament, and it started happening on one printer when I started messing with the belts. Then I messed with them on the other printer for w/e dumb reason and now Iām stuck with 2 printers that print like ā ā ā ā (a p1p upgraded to p1s, and a p1s). A guy on youtube showed how to adjust the tensioners in the back so that the belt doesnāt ride the top or bottom of pulley, but when I did it, I screwed my printers up somehow and canāt seem to get them back to normal. When I do the normal tensioning it doesnāt fix them. His video is to get rid of ringing, and I did it to fix a separate issue and ended up with the bad ringing lol.
this is also killing me on my prints - will try to see if changing speed helps, would love to put in better pulleys but looks like a total missionā¦ i have 2x P1S and 2x X1c and they ALL have this issueā¦ tired of doing full calibrations and cleaning rods it doesnt make any difference.
The tracking adjustment is a waste of time. The belt path is not planar and it seems that the OEM belts are not cut straight so no matter how much effort you put in, it will always track up and down.
I used those guides to verify squareness, tried every combination of high and low tension, tracked up and down, and nothing improved it.
For those who do not want to or donāt have the time yet (like myself) to do the pully replacement, you can try wrapping the final pulleys with electrical tape first - verify this improves your print quality except for the overlap. If it helps, itās easy to remove, and you can replace it with several layers of different lengths of Teflon. It is important to use different lengths so that they donāt all have the same start/stop point and create mounds on the pulley. This is still possible though tedious to remove, so it is a long-term temporary solution, assuming that the electrical tape showed a reduction in VFA in the first place.
ALL Bambu X1/P1 have this to some degree. Some are worse, some are better. It is likely that the result of the adjustments youāve tried is negligible and youāre just pickier about it now that you know it exists. For most of a year I did not realize it was there and it did not bother me. Now it bothers me immensely and I see it on every printer.
1 Like
I wouldnāt say belt tracking adjustment is a waste of time. Youāll almost never getting holding completely steady. But having it gradually tracking between the top and bottom flanges is much better than having it parked on one flange the entire time. Once you get the hang of it, itās a quick and simple adjustment.
1 Like
Specifically I mean pursuing tracking as a means to resolve 2mm VFA is a waste of time. It might be worse or better, but it will always be there no matter how perfectly the belts are tracking. If itās slammed way out so itās rubbing the flange AND striking the pulley it could make it worse, but it would be unusual for that to be the case.
You seem to be contradicting yourself here. You say that belt tracking may make VFAs better or worse. So, why wouldnāt you want to adjust them so that itās better? No one is claiming belt tracking alone will eliminate VFAs. You might as well eliminate as many sources as possible.
if your flow rate is low, you will need to open the extruder and clean the teeth of the extruder gear. it will get clogged like mud in your tire treads.
Yeah I mean I have had 3 x1cās in the past year all of them start printing perfectly and after a few hundred hours to a thousand they get pretty messy. better off replacing the belt. you can order one on their website.
I actually replaced my belt with a genuine Gates one - the VFA got massively worse. I printed a VFA benchmark before the belt swap and then afterward and it was incomparable. Wrapping the pulleys with teflon still improved it after around 10 times wrapping (which is probably about 30-40 layers of tape), but it never was close to as good as the last belt and wrapping. The Gates belt is thicker and stiffer and has a more textured and aggressive toothed surface than the OEM belts. I tried printing some extremely thin toothed pulley wraps with a 0.15mm back piece but it didnāt help and was hard to get to retain.
I ordered a couple 1524mm closed belts to split and install (open belts are spiral-cut from a wide belt which means that the fibers do not run continuously through it and the teeth are ever so slightly off perpendicular to the belt length. My hope was that a perpendicular tooth on a straight-cut belt would improve the situation but while trying to remove the belt, their belt-securing system failed and snapped the retaining post off the carriage so Iām now going to replace my X-axis which just arrived today.
I was debating going through the trouble of installing toothed idlers, and was rereading the thread for input, but itās not entirely convincing that it is the definitive culprit. Iām going to go ahead and install the new gantry with my 1524mm closed belts and see how that goes first. Having removed all the panels so many times now, the process is not so daunting and the option to replace the idlers with toothed ones remains open in the future, especially with a spare/broken axis to practice on.
Interesting topic, but my God, this must be the longest thread on the entire forum.
Donāt shoot me if I havenāt read all 576 of the preceding posts, but have you ruled out the effects of the double pinch rollers on the extruder? I seem to remember reading somewhere that they were found to be a major unexpected source of VFA, and the reason why the A series uses just one instead of two, and consequently why the A-series has comparatively little VFA.
2 Likes
Can you clarify what this means? Iāve never heard of this terminology.
Iām 100% certain the main VFA issues are related to the belt/pulleys as it almost exclusively affects X-axis travel and the effect is altered by doing anything to the final turning pulleys on that axis. You can also feel a very intense cogging effect that makes the toolhead step forward and backward in Y when moving slowly across X by hand and you can actually hear the buzz at higher speeds when it travels swiftly across X.
1 Like
Dual Drive: In a double pinch (or dual-drive) extruder, both gears or rollers actively engage the filament from opposite sides. This contrasts with a single-drive extruder, where only one side has an active gear, and the other side is passive (usually a bearing or idle roller).
Iām by no means an expert on this topic, but IIRC the theory was that in double pinch unless both rollers were perfectly synchronized, by special tuning or some other means, it would result in VFAās.
Youāve just given a good reason to think that in this instance itās not that: the difference in VFAās on the X vs Y axis motion. That in turn reminds me of a criticism of some of the recent creality printers, where the input shaping was shown to only be tuned on one axis and the results were blithely copied and just assumed to work the same on the other axis. Well, naturally, that assumption was false, with visible results. Maybe itās the same kind of issue with the X1C? Maybe X1 Plus can reveal whether that is so, or is it beyond reach given the closed firmware?
Iāve probably said too much for what little I know, but for a thread thatās gone on for so long with no resolution, I figure whatās the harm in a fresh set of eyes taking a shot in the dark? If enough fresh sets of eyes do that, maybe one of us will get lucky and touch upon an actual possible solution, to be vetted by people such as yourself, who are invested in finding one. That would be a win for everybody. 
By the way: out of all the printers on the market, which ones have the fewest VFAās? Anyone know? Iām in a mode right now where Iām buying a variety of purposefully different printers, where each is good at the particular things it does well.