Z 0

[darkdvd]

I'll try to ask my question in English (Google-English)

I do the levelling of the bed with a steel gauge from 0.10 mm thick.
Question: What is the value used by the printer for the placement of the bed in Z at 0 (zero)?
In other words: after a bed leveling at 0.10 mm to the nozzle, what is the height used by the printer or the slicer (CURA) between the bed and the nozzle?

En français maintenant
Je fais le niveau du plateau de mes UM2 en utilisant une jauge d'épaisseur en acier de 0.10 mm.
J'ai donc 0.10 mm entre la buse et le plateau.
Quelle est la distance utilisée par l'imprimante ou CURA entre le plateau et la buse ?

David

[Amedee]

Si tu jauges a 1/10ème, tu auras 1/10ème, si ce n'est que la calibration se fait à froid et quand ça chauffe ça se dilate
Cura te met à Z=0 pour la calibration et si ton premier layer est de 1/10ème, Cura positionne à Z=1/10ème donc au total 2/10èmes a froid...

Cura does not adjust, it positions the head at 0 during calibration and consider this as reference when it prints. There is however a difference because calibration is typically done with a cold head, so you have a difference when it is warm.

[darkdvd]

Merci pour la réponse Amedee
Généralement je fais la calibration à chaud, soit juste après une impression soit en faisant chauffer la buse et le plateau.
Donc il n'y a pas moyen de spécifier que l'on a calibré à 1/10 et soustraire cette valeur quelque part ?
Je dis ça parce que j'ai besoin de sortir une pièce dont un élément en contact avec le plateau doit avoir 4/10 d'épaisseur sachant que j'imprime l'objet en couche de 1/10 (0.10 mm)

Usually I do calibration hot or just after printing either by heating the nozzle and the bed.
So there is no way to specify that it was calibrated to 1/10 and subtract this value somewhere?
I need to get out a piece including an element in contact with the bed must have 4/10 in thickness knowing that I print the object in layer of 1/10 (0.10 mm)

[LePaul]

Cura does not adjust, it positions the head at 0 during calibration and consider this as reference when it prints. There is however a difference because calibration is typically done with a cold head, so you have a difference when it is warm.

This threw me off a lot when I first installed the heated bed on my UMO. Based on all the videos I had seen on the UM2, I just leveled the bed cold. Then, to my dismay, on my first print, nothing was coming out (because now, with heat, the bed was too close)

I was curious if the firmware took into account the difference a heated bed would have over a cold bed. Apparently not. Since bed leveling with the bed at 60, prints have gone well, with minus adjustments while the skirt is printing.

[Amedee]

Donc il n'y a pas moyen de spécifier que l'on a calibré à 1/10 et soustraire cette valeur quelque part ?

Ecrit toi même un bout de GCode qui positionne la tête à 1/10ème et calibre à cette hauteur

If you calibrate with a warm hotend, then write a piece of GCode wich moves the head to the desired height and calibrate for that height.

[darkdvd]

GCode n'est pas mon ami ! Il devrait mais non
Gcode is not my friend ! It should but not

[Amedee]

Out of my head...

Code: Select all

G21                        ;metric values
G90                        ;absolute positioning
M82                        ;set extruder to absolute mode
M107                       ;start with the fan off
G28 X0 Y0                  ;move X/Y to min endstops
G28 Z0                     ;move Z to min endstops
G1 Z0.1 F12000             ;move the platform down 0.1mm (or whatever height you want to calibrate)
; Next are calibration loops where you bring the nozzle to the point you want to calibrate
; Adjust the coordinates to get the head exactly where you want it...
G0 F12000 X10 Y10          ; front left
M0                         ; pause
G0 F12000 X200 Y10         ; front right
M0                         ; pause
G0 F12000 X100 Y200        ; back middle
M0                         ; pause
; copy the above for multiple passes:
G0 F12000 X10 Y10          ; front left
M0                         ; pause
G0 F12000 X200 Y10         ; front right
M0                         ; pause
G0 F12000 X100 Y200        ; back middle
M0                         ; pause
; ...
; Finished...
M84                         ;steppers off

In case somebody would copy this blindly, it assumes that your bed is already calibrated, otherwise you could crash the hotend into the plate

[Neotko]

I do level at 0.08 metal gauge feller when cold and with a 0.04-0.05 gauge feller.

Um2 autocalibration sets the head at + 0.1 for you to calibrate. Umo home moves the head to 0. Cura just moves z to 0.

[Neotko]

Nice gcode amedee! I would add a bed temp and at least 80-90C for the hotend (so it could be cleaned also). There's nothing more disturbing that getting ready to calibrate and seeing a bit of cold pla on the nozzle.

[Amedee]

Um2 autocalibration sets the head at + 0.1 for you to calibrate.

Oh, thanks for this info, I didn't knew that.

I would add a bed temp and at least 80-90C for the hotend (so it could be cleaned also).

Yes, so we get french sausages well cooked at the end of the calibration

[darkdvd]

...french sausages with Belgian fries...

[Anders Olsson]

One thing worth to mention is that the UM2 firmware had numerous bugs in the early firmwares which affected the bed leveling,
So if anyone is using an old UM2 which has not had the firmware updated lately, update it before doing any further testing/troubleshooting!

Some people at the old forum, including me, eventually managed to track down all these bugs which were causing major headache for Ultimaker.
If I remember correctly there were at least three independent issues:
1. The routine that finds the home position of the Z-micro switch did not move the bed far enough up to properly deactivate the micro switch before starting looking for the zero position. So if the bed had dropped by gravity or was pushed down, the Z position could suddenly be really far off.
2. The calibration routine used 0.2 mm as the thickness of the calibration paper, while Ultimaker recommended to use a regular A4 sheet type paper which is 0.1 mm thick.
3. These 0.2 mm were not properly subtracted until power cycling, so before power cycling the bed was 0.1 mm too close and after power cycling it was 0.1 mm to far away (or if it was the other way around?).

[darkdvd]

Thanks for this information, I understand better now and I'm going to validate all these points.

David.

[darkdvd]

Little test this morning.
I draw a little staircase : size X 100 mm x Y 10 mm x Z 1 mm + 0.1 mm every 10 mm

Here is the result of the print :
nozzle : 0.4
layer : 0.1
initial layer : 0


the average difference between the staircase and the "reality" is 0.11 mm

I think the 0.11 mm are lost in the first layer

[Amedee]

So your bed is 0.1 to high...
I you use the same calibration procedure and add 1/10th to your gauge, you should be good