Some days ago I finally finished my Further Fabrication’s Y-1200. I’ve been fine-tuning the parameters and testing some materials when I noticed some wavy lines when cutting or marking at 20mm/s or more.
It only happens when accelerating decelerating. Everything is fine while engraving.
I’ve checked this and other forums in search for a solution but nothing has solved it yet. Do you have any idea what might be causing this? All my current parameters and vendor settings are attached in the pictures.
The machine uses a Ruida 6445S with DM542 Stepper controllers, dual shaft Nema 23 for the Y axis and a Nema 17 for the X axis.
Things I’ve checked so far:
Coupler winding: Changed the standard flexible couplers for better ones. It improved a bit but there are still wavy lines.
Lowering acc values: Don’t seem to make much difference.
Decreasing acceleration and corner acceleration: Improved quite a bit but still noticeable.
Head and nozzle play: Checked and both are tight and fixed in place.
Loose lens and mirrors: Checked, they’re secured in place.
Belt and pulley rise: Tried to compensate with backlash values but it made no difference.
Linear rails damage or bumps: Nothing, Both axis slides smoothly.
Tightening and loosing the belts: Same, no difference.
All pulleys and gears are firmly in place.
Motors are set to their peak amperage and running at 3200 pulse/rev but already tried up to 12800 with no results.
Any help is really welcomed as I’m at my wits’ end.
Many thanks!
edit: Added a better detail picture or the waves or vibrations.
Yes, I’ve already checked multiple times the lens, mirrors, belts, pulleys, couplers, both X and Y carriages and linear rails etc. I really don’t know what else to check that I haven’t checked 4 or 5 times.
The waves are pretty consistant and are not random on its pattern so I’m pretty sure is not the lens or mirrors as the waves form should be random then.
As per the accelerations, the factory settings were pretty conservative and I managed to increase the X acc to 12k mm/s only for engraving as X is the scan axis. It engraves perfectly without skipping steps or overscanning in excess. Nevertheless I also reduced X acc to 2k just to see if something changed and no luck.
For line cutting X is set to a much much lower acc. Y acc is lower than factory settings in order to mitigate the issue but with limited sucess. Factory settings for X acc was 5000 mm/s and 3000mm/s for Y.
I think you’re seeing the inherent rigidity of a large box-style machine: the frame distorts by that much.
A rectangular frame built from extrusions depends on the corner joints to hold everything in proper alignment, but there’s a huge big lever arm on those screws and angle brackets. As a result, the whole frame can distort very slightly as a result of the reaction forces from the motors.
Addional diagonal bracing, perhaps with tension cables, might stiffen it a bit, but the “wiggles” look to be on the order of 0.2 mm, which is already pretty good for a machine that big.
A somewhat different problem producing a similar amount of error:
I was beginning to think that as well but this is happening at only 15-20 mm/s. Isn’t that too slow to distort the frame?
Also, I’ve noticed this only happens when both axis move at the same time. Circles, text, diagonal lines etc. Changing acceleration doesn’t seem to do much and changing speed only shortens or spaces the sine waves. They’re still there, just closer or farther.
I think this is some kind of vibration or resonance that only happens when the 2 axis are moving. Do you have any idea of how could I dampen or reduce this?
Thanks all for your help, at least I’m learning a lot from this situation!
edit: added pictures for reference. Done at 20mm/s
One of the first rules of debugging is to stop thinking and start looking: you can easily rule out the cause by deciding that can’t possibly be it.
Seems reasonable, but that points right back to rigidity. Whether it’s the entire frame or just one section is hard to say.
Which seems remarkably slow for a line engraving. If the laser tube has a lot more power and you’re throttling it down for a low-speed test, that’s fine.
However, you may have just identified a corner case to avoid: "If it hurts when you do that, don’t do that.
As you observe, speeding up or slowing down changes the spacing, which suggests a mechanical resonance: the frame (or whatever) shakes at the same frequency, so varying the speed spreads the same pattern over different distances.
Despite mentioning cross-bracing earlier, the least-awful way to add rigidity & damping will be fastening sheets of plywood / MDF (not acrylic / aluminum) across all those nice open spaces between the extrusions, with plenty of screws locking the sheet to the rails. Identifying which spaces need filling comes down to trial and error, although I’d start with the biggest ones.
If you come across a well-hidden / hard-to-reach joint with loose screws, well, maybe the sheets will be superfluous.
High truth: you learn more from failure than success!
