Uneven Z-travel in XY plane

[lumpy_potato]

Hi all! I broke things again in the pursuit of higher quality prints.

I tried tightening down my belts further - so they had a nice 'snap' when pulled instead of a loose flump. All three feel about the same.

However, since then, all accuracy as far as travel has gone out the window. Z0 will change each time I do a g1 z0, and using the script in the SeeMeCNC delta config video (going from each tower to center in the same z.2 plane) I can clearly see the Z bobbing up and down.

So far, my thoughts are this:

* In tightening the belts, something has gone terribly wrong

And my solution is this:

* Undo the belts and take the Tricklaser arms off the cheapskates
* Re-do the cheapskate calibration so they all fall at the same rate again
* Re-do the belts aiming for a nice twang and about 1" slack when pulled to a nice twang.

So far I've removed the belts and am checking the cheapskates and noticed that they are 'bumpy' - like they don't ride evenly on the carriage. Some of the wheels look like the wear was a little uneven - any way to fix this? I've re-done the tighty/loosey bits and they are all falling at the same rate now - is there any advantage in using loctite or something similar to stop the eccentric cams from slipping over time?

Is there anything else I might be missing? Towers should be properly square as far as I can tell.

UPDATE:

Everything is back together, and I think I see what part of the issue is, but I don't know why.

The X axis, at the top, seems to pause and then keep going. This shifts the entire thing. When trying to move the carriage by hand, it seems to sort of pause in the area (roughly between Z370 and z375).

I DID just notice that the X toothygearthing has frlipping slipped and is no longer flush with the shaft. I have a sinking feeling this might be contributing in a major way, as the belt will be somewhat offset, and possibly binding at a certain point on the tower. It's the only thing that seems different as far as I can tell.

I just got it all nice and tuned up too oh well, thats what I get for not using loctite the first time around. I'll post a video of some sort of the oddity in action shortly.

POTATO VIDEO UPDATED POTATOTIME

UPDATE: Damn. At some point the grub screws came loose, but not loose enough to dislodge. So the bloody grub screws bit into the shaft a bit and threaded the shaft! So the toothgearthing is kind of hard to work with now. Maybe polish it with some sandpaper to it slides on like normal again, but man. Was not expecting that. Not sure if this is actually the problem, but it couldn't have been helping. TL;DR the grub screws rotated loose but the tension from the belts held it in place, such that it cut into the shaft a bit and threaded itself on.

[geneb]

...and this children, is why I say to use thread locker on the grub screws.

Class dissssssmissed!

I'm glad you got it figured out.

g.

[lumpy_potato]

I'm leaving it off FOR NOW, only because I'm planning on taking the whole motor assembly out for dampers and other fun stuff. But when I'm done with thati t will be THREAD LOCKER ALL THE WAY BOI


Unfortunately that wasn't the core of the problem. I'm still seeing uneven travel in all directions -- the X skate for some reason does not seem to be travelling at a uniform pace compared to the other towers.

At this point I've undone and re-done the belts several times - they should all be at similar tension levels, but I don't think that could be the only thing going on here. I'm going to keep plugging at it, but right now there is simply no direction in which the head travels in a straight line. There is always a variation in the X, Y, or Z direction.

FOUND THE ISSUE BUT NEED A FIX

On Z movements, the Y motor now has slop. Its very, very odd:

On Z DOWN the first time, the X and Z motors engage normally, but the Y does not.

THEN after that all three move at the same rate.

Do a Z UP, and the Y motor freezes again, and then subsequent movement they move in sync.

This is very odd and doesn't feel mechanical to me - why would

Maybe a firmware wipe is in need as well, but I can't imagine _why_

Could use some ideas here :/

Edit: Yeah the toothygear has come loose off the X motor again. So either the shaft is wrecked, or the toothygear is. Hopefully its the gear...does introduce some slop, but that isn't my main issue now.

NEVERMIND THE STUPID GRUB SCREWS ARE LOOSE ON EVERY SINGLE MOTOR YAYYYYYY

I am a little concerned that so soon after tightening by hand the screws are coming loose again, however. But that might be because I'm not tightening properly as well.

Looks like you're 100% on the money geneb. You win a llama. If I end up with a llama I will send it to you.

[geneb]

Vibration is not your friend.

Do you know the difference between a one "l" lama, a two "l" lama and a 3 "l" lama?



One is a Tibetan holy person, one is a South American beast of burden and the last is one hell of a fire!

g.

[lumpy_potato]

Thanks Gene,

One question while I've got you - its been fine for nearly a year now, but the belts were not as tight as they could (should?) be. My thought is that tightening the belts transmitted more vibration than previous, vibrating the grubscrews loose. While the belts were relatively loose (enough for everything to move nicely, but not enough to create any sort of twang when plucked), there was still vibration transmitted, but not as much or not enough to loosen the grub screws.

At least thats all I can think of. Regardless hopefully once this is all sorted I'll be back to printing, followed up by tweaking, followed up by breaking bloody everything all over again

[geneb]

Belt tension is kind of a weird thing. There's no real tool that people have easy access to in order to adjust them. You want the belt tight, but not too tight. The tighter it is, the more force the stepper motor has to apply in order to move it. You certainly don't want it tight enough that plucking it gives you a good note. A fairly reliable gauge is to touch your middle finger to the tip of your thumb and then press on the "meaty" bit of your hand below your thumb joint. You want your belts to have a similar feel to them when you press in the middle of the belt on the inside track with the carriage moved all the way to the top or the bottom.

g.

[lumpy_potato]

Thanks - actually that metric is the first I've had of anything concrete as far as belt tension. There are a few threads were people were aiming to have a level of pluck where you could use a guitar tuner to pick up the tone!

May I suggest that, if its not already in there, to add that particular instruction to the build guide?

[geneb]

I did at one point, but it turns out that the hand position shown is rather rude in some countries and I got yelled at.

g.

[lumpy_potato]

Pity. I'll be using it moving forward at least.

Appreciate your help in all of this.