Print skewed to the right as Z goes up - Issue Found

Thought I would try here first as it is Delta specific I think.

Went to print some light bulb covers today and the print looked like the leaning tower of Piza.

Approximately a 3 degree lean to the +X AND +Y direction as Z goes up. (Edited after printing a cube.)

This was fine yesterday and I have printed multiple 200mm tall items that were vertical.
The last part printed last night was the green fan duct in the bottom photo

Steps taken so far,
Checked all towers square to build plate, worst is the right tower at 89.90, the other two are 90.00

Completely redone calibration.

Checked x and y movement is correct, 100mm is exactly 100mm in both directions.

The only concern is the Z tower belt required tightening and visually looks not right however X/Y movement is correct

Now out of ideas, anyone point me in the right direction.

This is a 24mm cylinder, 30mm high, it is actually round and correctly dimensioned but has a approx 3 degree tilt.




Edited 2 time(s). Last edit at 06/30/2016 11:23PM by aussiephil.

Might be loose pulleys on the steppershaft?
Is it leaning exactly in X+ direction? Then it would be a combined error of two towers.
If it was along the Y axis, the z-tower stepper would be my best guess.

It leans to the right in the x+ direction. I would have expected loose pulleys to reflect in the independent x and y travel measurements.

OK A little bit more information....

I printed a square and the near three degrees slope leans towards the +X and +Y corner.

What electronics and firmware are you using? If it's RepRapFirmware:

1. Make sure you haven't accidentally enabled orthogonal axis compensation;

2. After or during a print that shows these symptoms, run M122 and check the 'Step errors' field. It should be zero.

Do your stepper motors have flats on the shafts?

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

two possible causes

1) lost motion due to slippage
2) orthogonal error

you can test for 1 by starting your printer and doing a normal calibration and then mark where the nozzle touches the bed and the carraige positions when at XYZ=0, if you can measure the carriage positions accurately.

Then, run a program, doesnt have to use filament, just run the program for a couple of hours

After the program has run go back to XYZ=0, is everything in exactly the same place? if NO then you have a lost motion problem.

---

RepRapPro Mendel 3 Tricolour
RepRapPro Fisher
-Carbon Arms
-Easy adjust Carriage+effector
-axis stiffness mods
HE3D -600 delta
-Duet 0.8.5
-PanelDue
-DC42 Height probe
-RobotDigg metal components
Simplyfy3D
RS Design Spark CAD

Quote
dc42
What electronics and firmware are you using? If it's RepRapFirmware:

1. Make sure you haven't accidentally enabled orthogonal axis compensation;

2. After or during a print that shows these symptoms, run M122 and check the 'Step errors' field. It should be zero.

Do your stepper motors have flats on the shafts?

Duet 085 with RRF1.13, this is the machine that's getting the DuetWiFi

1. already checked, nope
2. yep was zero.

Yes shafts have flats

See next post for "feel like an idiot for not checking first" details

Quote
bgkdavis
two possible causes

1) lost motion due to slippage
2) orthogonal error

you can test for 1 by starting your printer and doing a normal calibration and then mark where the nozzle touches the bed and the carraige positions when at XYZ=0, if you can measure the carriage positions accurately.

Then, run a program, doesnt have to use filament, just run the program for a couple of hours

After the program has run go back to XYZ=0, is everything in exactly the same place? if NO then you have a lost motion problem.

No issues with 1 or 2 and it was the completely accurate sizing for x y and z that was throwing me off.

For the record here's what i found once i decided to start pulling the machine apart and hopefully this helps others though i feel dumb for not finding it first.

This is a 1.6M high delta that has been fully enclosed and was fully square and vertical till a few nights ago after a nice -4c overnight low.

Nothing looked different and all the x y z calibration measured correctly included measured moves.

So started to strip the machine down and plumb bob all the vertical elements and that's when I got the shock of the day......
Z tower was essential vertical left/right but was leaning in towards the middle.
X and Y towers however where completely out of whack with the same lean to the right looking from the front.

Seems the entire frame twisted around potentially from differential thermal expansion/contraction values of different materials though why it skewed/twisted in that fashion seems rather bizaar.

The upshot is that a skewed print that is sized correctly directly points to leaning towers on a delta.

Now to loosen all bolts and square things back up.