Bad mesh compensation

[Ducati87]

@jay_s_uk

Y114 is the right probe position in front of the lead screw.

[Phaedrux]

your homeall just calls the individual homing files, can you post those?

0,0 is expected to be the front left corner of the build plate.
-x moves to the left, +x moves to the right
-y moved to the front, +y moves to the back

Does this match what you have?

[Ducati87]

@phaedrux said in Bad mesh compensation:

your homeall just calls the individual homing files, can you post those?

0,0 is expected to be the front left corner of the build plate.
-x moves to the left, +x moves to the right
-y moved to the front, +y moves to the back

Does this match what you have?

yes they are matched.

homex:

; homex.g
; called to home the X axis
;
; generated by RepRapFirmware Configuration Tool v3.1.4 on Mon Jul 27 2020 14:28:18 GMT+0200 (Ora legale dell’Europa centrale)
M400
M913 X50
M400
M201 X200
M915 X F0 H400 S1 R0
G91               ; relative positioning
;G1 H2 Z5 F7200    ; lift Z relative to current position
G1 H2 Z10 F7000
G1 H1 X-305 F5000 ; move quickly to X axis endstop and stop there (first pass)
G1 H2 X10 F7200       ; go back a few mm
G1 H1 X-305 F3000  ; move slowly to X axis endstop once more (second pass)
G1 H2 X15 F7200   ; lower Z again
G90               ; absolute positioning
M400
M201 X4000
M913 X100
M400

homey:

; homey.g
; called to home the Y axis
;
; generated by RepRapFirmware Configuration Tool v3.1.4 on Mon Jul 27 2020 14:28:19 GMT+0200 (Ora legale dell’Europa centrale)
G91               ; relative positioning
M913 Y50
M201 Y100
M915 Y F0 H400 S1 R0
;G1 H2 Z5 F7200          ; lift Z relative to current position
G1 H1 Y-305 F5000 		; move quickly to X or Y endstop and stop there (first pass)
G1 H1 Y10	
;G1 H1 Y-100 F5000        ; home Y axis
M913 Y100
M201 Y4000
G90               ; absolute positioning

homez:

; homez.g
; called to home the Z axis
;
; generated by RepRapFirmware Configuration Tool v3.1.4 on Mon Jul 27 2020 14:28:19 GMT+0200 (Ora legale dell’Europa centrale)
G91                ; relative positioning
;G1 H2 Z5 F7200     ; lift Z relative to current position
;G90                ; absolute positioning
G1 H1 Z10 F800
G1 X140 Y174 F7000           ; go to first bed probe point and home Z
G4 P750             ; pause
G91

;Probe Z
G30
G90
;G1 Z10 F500

[Phaedrux]

@ducati87 said in Bad mesh compensation:

G1 H1 Y-305 F5000 ; move quickly to X or Y endstop and stop there (first pass) G1 H1 Y10

This is not correct. The second move G1 H1 Y10 should not have H1 if it's a back off move it should just be G1 Y10

@ducati87 said in Bad mesh compensation:

G1 H2 Z10 F7000 G1 H1 X-305 F5000 ; move quickly to X axis endstop and stop there (first pass) G1 H2 X10 F7200 ; go back a few mm G1 H1 X-305 F3000 ; move slowly to X axis endstop once more (second pass) G1 H2 X15 F7200 ; lower Z again

H2 should only be used on Z moves to allow the Z axis to move before being homed. It should not be used on XY moves. There is also an error in the block above where the Z axis is lifted, but then the X axis is moved when the comment says to lower the Z axis.

M671 X-15:290 Y114:114 S6

Does driver 2 match the motor at the first XY position?
Does driver 4 match the motor at the second XY position?

G31 P500 X37 Y72 Z-1.044
You have a rather large offset on Y axis. If there is some tilt it can produce different trigger heights depending on XY position.

Your Z trigger height is negative. Is that correct? That would mean the nozzle is pushing a full 1mm into the surface of the bed when the probe triggers? Depending on the probe type this may be the case. What type of probe is it?

[Ducati87]

@phaedrux said in Bad mesh compensation:

@ducati87 said in Bad mesh compensation:

G1 H1 Y-305 F5000 ; move quickly to X or Y endstop and stop there (first pass) G1 H1 Y10

This is not correct. The second move G1 H1 Y10 should not have H1 if it's a back off move it should just be G1 Y10

@ducati87 said in Bad mesh compensation:

G1 H2 Z10 F7000 G1 H1 X-305 F5000 ; move quickly to X axis endstop and stop there (first pass) G1 H2 X10 F7200 ; go back a few mm G1 H1 X-305 F3000 ; move slowly to X axis endstop once more (second pass) G1 H2 X15 F7200 ; lower Z again

H2 should only be used on Z moves to allow the Z axis to move before being homed. It should not be used on XY moves. There is also an error in the block above where the Z axis is lifted, but then the X axis is moved when the comment says to lower the Z axis.

M671 X-15:290 Y114:114 S6

Does driver 2 match the motor at the first XY position?
Does driver 4 match the motor at the second XY position?

G31 P500 X37 Y72 Z-1.044
You have a rather large offset on Y axis. If there is some tilt it can produce different trigger heights depending on XY position.

Your Z trigger height is negative. Is that correct? That would mean the nozzle is pushing a full 1mm into the surface of the bed when the probe triggers? Depending on the probe type this may be the case. What type of probe is it?

thank you for your reply.

M671 X-15:290 Y114:114 S6

Does driver 2 match the motor at the first XY position?
Does driver 4 match the motor at the second XY position?

yes both of them.

G31 P500 X37 Y72 Z-1.044
You have a rather large offset on Y axis. If there is some tilt it can produce different trigger heights depending on XY position.

I reduced the gap between the probe and the bed and now the right value is 0.750.

The problem between the left and right side is a bit better than before but the compensation is not perfect.
The right side is still a bit upper than the left side but the compensation doesn't work as well.
Maybe I should to change the probe or install a bltouch in order to improve the accuracy.

[Phaedrux]

What is the probe? Reducing the distance between nozzle and probe may help.

Can you post a photo of your heightmap image?

[Ducati87]

@phaedrux said in Bad mesh compensation:

What is the probe? Reducing the distance between nozzle and probe may help.

it's a normal inductive probe NPN.

[Phaedrux]

Have you leveled the bed manually to get it more or less even to the nozzle tip first?

[Ducati87]

@phaedrux said in Bad mesh compensation:

Have you leveled the bed manually to get it more or less even to the nozzle tip first?

do you mean the z offset?

[Phaedrux]

No i mean manually leveling the bed to try and eliminate as much tilt as possible.

[Ducati87]

@phaedrux said in Bad mesh compensation:

No i mean manually leveling the bed to try and eliminate as much tilt as possible.

No I'm not. Yes but why there is this problem with automatic leveling?

[engikeneer]

@ducati87 you only have two z lead screws, at the max and min of the x axis (left/right), so the G32 levelling can only level along the x axis. From you heightmap image, it looks like your heightmap is very level in X as expected

You should probably try mechanically levelling your bed a bit better from front to back. From your first image, it seems that's the direction the issue is in?

[Phaedrux]

Yes it seems a large Y tilt, possibly related to the large Y offset of the probe.

[Ducati87]

@phaedrux
is there any chance to fix it?

maybe I should to move the probe in a better position.

[Phaedrux]

It's always best to have the probe as close to the nozzle as possible.

[Ducati87]

@phaedrux said in Bad mesh compensation:

It's always best to have the probe as close to the nozzle as possible.

ok I will try to move the probe in a better position. thank you so much.