X-axis suddenly forgets its place!

[dry 0]

 

Hello dear UM community,

 

Help needed! My colleague and I received our pre-assembled Ultimaker earlier this week for our design studio.

 

It printed very well the first couple of tests, then yesterday the X-axis suddenly jumped 9-10mm to the left. This was printed using the latest version of Cura for Mac (via the UltiController). Then my colleague tried with a different STL and with Cura for Windows and the same thing happend (only at a different height and different lengts - 3mm off track).

 

I'm including an image my third try (with the error happening at a different height as the others). All the belts are tight and I can move the XY axis easily. My settings in Cura (13.03) were:

 

Layer height: 0.2mm

Wall thickness: 2,8mm

Retraction: ON

Fill Density: 20%

Print speed: 50mm/s

Print temp: 210

Support: Everywhere

Raft: ON

Duplicate outlines: ON

(the rest are the standard settings)

 

https://dl.dropboxusercontent.com/u/3673145/photo.JPG

 

https://dl.dropboxusercontent.com/u/3673145/Skjermbilde%202013-04-19%20kl.%2010.25.36.png

With best regards,

Daniel

 

[snowygrouch 1]

Looking at your print, my best guess is that the print is warping upwards (this is always worse when you have

large angled beams like your model has).

This has happened to me several times, what occurs is that the print warps - the printhead nozzle then hits

the upturned edge on one of its moves with such force that a motor skips steps.

Im not saying a slicing error is impossible, but I have not heard of any such problems with

13.03 (and have used it myself recently). Which leads me to suspect that its very unlikely to be the problem.

Solving it is not easy, but you can try the following things.

1) rotate the STL model so that the parts where it warps UPWARDS are at a 45 degree angle to the

YX axis. This just slightly softens the blow when it does hit (because the force is then distributed

between both motors). However this only really works when printing roughly rectangular geometries.

2) Try printing the part in a different orintation, or separating then bonding together after printing.

3) Search out threads here on lowering warping and follow tips in there.

4) I dont know if the latest CURA has the feature, because I only just started using it a few days ago,

but the latest Beta version of Kisslicer has a "hop" feature. This lifts the printhead during moves so that

the nozzle does not hit the print. I have no tried this feature so cannot say if its implimented correctly.

5) Increase your layer height. This will give more clearance for moves during printing. I would suggest 0.3 to

be max sensible height.

6) Decrease the distance between support and model, such that the support "ties down" the angled beams from

moving upwards. Makes removal harder of course.

7) Edit the STL to have perhaps a small diameter post as part of the model that achieves the same things as #6

I mainly use Kisslicer, so perhaps one of the CURA experts can come in with more specific suggestions tha will be helpful to you.

C.

EDITS:

Couple of extra thougts.

8) Although fan cooling as a whole, does make the part "warp" more - LOCALLY

more cooling will lower the warping of thin edges of the kind you have on those

leading edges of the 3 angled beams. So perhaps try to increase cooling there.

9) Increasing infill % might give the beams more stiffness and be less liable to

warp. However how it will behave thermally is highly "form dependant" so

its not a rule that will automatically be guaranteed to make the leading edge of the 3

round beams stay flatter.

10) I dont know if CURA has it (I think it does), but you might want to increase the "minimum layer time". This gives things like thin edges a bit more of a chance to cool...

probably its set to 5 seconds or something right now. Try increasing that a bit...

 

 

[dry 0]

Hello C.

Thank you for your response! I did not expect that fast and such in-depth feedback.

Yes, there were some warping on this model. I will try to print a single vertical tube and see what happens. If there are no errors then the warping can be the cause of this problem.

All best,

D

 

[snowygrouch 1]

Just make sure you watch the printer working when you have problems like this. Otherwise diagnosis

is alot harder.

If you can see the printhead hitting the warped egde, and you hear a BOINK ! sound - probably the belt

skipped a tooth. If the motor is skipping you will also hear a very unhappy grumbling sound from the motor.

When its a phyical problem like this, sitting and watching is very beneficial. If you are too buisy to do that

then get a webcam and film your prints (with audio). Then if the print fails you can just quicky

skip through the video footage to find what happened.

C:

 

[gr5 2,230]

I think you may have a loose set-screw on one of the 5 pulleys on the X axis. You need to tighten all 5. Especially the screw that is on the motor pulley.

This loose set screw would explain the shift you got. While you are at it, tighten the X motor by sliding it further down the slots.

You want to tighten these pulley set screws very tight - almost to the point of breaking, but don't break them! There should be spare set screws in case one is lost.