shifted layers

[kervert 3]

hi. I just put together an UMO+ and my first two prints were fine, but then the layers began shifting, always on the y axis and always beginning around 15-20mm. I've tightened down the set screws on all the pulleys, including the one on the y motor, checked that the axis are all even and square, pulled down on the y motor to create more tension in the belt, lubed all my rods with machine oil and most perplexing of all I used a black marker to make a hash mark where all pulleys meet their axis and after each print that shifts, which is all of them now, I check to see if anything has shifted and the marks still line up. im completely confused. any help or advice would be greatly appreciated. thanks

[gr5 2,170]

Can you supply a photo please?

Are you saying it did or didn't shift pulley versus shaft? It seems to say that it did shift which implies they can still be tighter.

if the shift is sudden it's usually a pulley or an end stop switch. The end stop switches should pop up a message on the ulticontroller I think. Maybe.

If the shift is gradual it's usually that the belt is touching (just barely) the wood wall and the motor needs to be moved away slightly using some washers as spacers.

[kervert 3]

Can you supply a photo please?

Are you saying it did or didn't shift pulley versus shaft?  It seems to say that it did shift which implies they can still be tighter.

if the shift is sudden it's usually a pulley or an end stop switch.  The end stop switches should pop up a message on the ulticontroller I think.  Maybe.

If the shift is gradual it's usually that the belt is touching (just barely) the wood wall and the motor needs to be moved away slightly using some washers as spacers.

 

Snapshot@2015_0623_180634.bmp

it did not shift, according to the marks i made. they all still line up. how would the limit switches create this shifting? the belt is not touching anything

Snapshot@2015_0623_180634.bmp

[kervert 3]

[kervert 3]

 

i broke off the shifted layers and ensuing rat's nest

[gr5 2,170]

i broke off the shifted layers and ensuing rat's nest

 

Well then those pictures aren't helpful.  I'm confused - why would there be a rats nest just because it shifted. Doesn't it just keep printing upwards but over to the side?

endstops - if the endstop is triggered then that servo is told not to move any farther in that direction but it's okay to go the other direction.  So you get a shift.  It could be triggered by an electrical short or a bad switch, or something mechanical bumping the switch.  You could just unplug the switches for that axis.  The homing procedure will be horrible and noisy.  You could hit the switch mid print just to see what happens - see if you get a message on the ulticontroller.

Edited June 24, 2015 by Guest

[gr5 2,170]

These problems all seem to happen around the same height. So I am going to go with overheating. If the stepper driver overheats this will happen also - this would make complete sense. Check your bottom fan! Consider printing with the printer on it's side and add an extra fan or something. Is your house hotter this summer than it was when the printer was last working? Maybe a heat sink fell off the Y stepper driver.

[gr5 2,170]

Oh wait - UMO+. No bottom fan. Add a fan temporarily to see if that helps - the printer will print on it's side (or even upside down) just as well (no better or worse really) than rightside up.

[SandervG 1,521]

These problems all seem to happen around the same height.  So I am going to go with overheating.  If the stepper driver overheats this will happen also - this would make complete sense.  Check your bottom fan!  Consider printing with the printer on it's side and add an extra fan or something.  Is your house hotter this summer than it was when the printer was last working?  Maybe a heat sink fell off the Y stepper driver.

 

Except for the bottom fan ( ), I am also inclined to think it is overheating. Is it very hot over there?

Edited June 24, 2015 by Guest

[kervert 3]

[/media]

this last print made me realize that it's not a height issue, but rather the amount of time it has been printing. this contact lens case shifted after about 35 minutes, despite only being a few mill tall

[kervert 3]

[foehnsturm 969]

Occasionally, I see the same with my UMO. My best guess so far: wire cross talking. I just wiggle the flat ribbon cables and reinsert the connectors and the issue usually disappears .

Edited June 25, 2015 by Guest

[gr5 2,170]

Did you check on the cooling issue? Put your printer on it's side and have a desk fan blow on the board. It prints sideways just fine.

Those pictures are helpful. It is a large skip and always the same direction and most layers are fine.

Look the issue is most likely:

1) Pulley slipping

2) Friction

3) hot stepper driver

4) end stop switches

5) crosstalk?

did you put marks on all 6 of the pulleys for the bad axis? Or only 4? It's almost always one of the 2 pulleys on the short belt. 90% of the time. Are you absolutely sure this isn't the problem?

You need to eliminate all of the above until you figure it out. Hot stepper driver is easy to test by using fan.

end stop switch -- I would test this by hitting switch ON PURPOSE and see if a message comes up on the display and see if the behavior is the same or different. If the behavior is identical (no message on screen in both cases) then we haven't proved it's limit switch but haven't eliminated it either - also you can figure out *which* switch it is by the direction it moves because each switch would affect a different direction and one of the switches is TOTALLY USELESS so just disconnect that one please. The one not used for homing.

crosstalk - yikes - I never head of this - I guess reseat the cables? Put some tape between the tables maybe to keep signals farther apart?

Friction - is one axis much much higher friction than the other? Push head around with power off. Sometimes if the long belts are too tight or much more likely if the end caps are pushing too hard on the rods you get very high friction and the stepper motor can skip steps. This seems unlikely in your case but the quick check is toosen the end caps a bit for the 2 thick rods on the bad axis. Only one endcap for each axis.

Think of this as a fun puzzle! Don't let it depress you!

[foehnsturm 969]

Sounds weird I know, but when I slightly touch the wiring at that time I often get the crashed Ulticontroller screen with funny symbols as well. So on my UMO it's definitely something electrical.

[kervert 3]

i'm embarrassed to admit that i think i overlooked tightening one of the gears. the one attached to the small belt on the y axis. it wasn't loose, but i gave a little extra tightening and now i think the sliding is gone. fingers crossed. thanks all

[CB1 0]

Hi kervert,

Did that tightening resolve this issue?  

We're having the same problem on a UM2E+, one month old.

[SandervG 1,521]

There is a fair chance it did, and if not I would definitely recommend to give it a try.