Auto bed leveling G29 data ignored during print

[Zuru1]

Hi

I decided to add auto bed leveling/trimming on my printer using a PNP inductive proximity sensor switch (LJ12A3-4-Z/BY).
At the moment I have managed to get the initial set up, were to probe, how far and replaced my normal z endstop switch.

To verify my set up, I decided to use this as a test print in order to verify bed level feature.
So far I have been unable to successfully make this work, get a levelled bed, part of the bed is ok but not the other part.
I have tried to use two different points in order to get the z probe offset, the bed centre and the start of the bed (0,0) which differ with 0.1mm.
Both yield the same result.

Here are G29 results run after each other with a -1mm z prob offset from the nozzle
SENDING:G29
Bed x: 0.00 y: 0.00 z: -0.96
Bed x: 155.00 y: 0.00 z: -0.83
Bed x: 155.00 y: 180.00 z: -0.60
Bed x: 0.00 y: 180.00 z: -0.89
Eqn coefficients: a: 0.00 b: 0.00 d: -1.00
planeNormal x: -0.00 y: -0.00 z: 1.00
echo:endstops hit: X:0.01 Y:179.99 Z:-0.89
SENDING:G29
Bed x: 0.00 y: 0.00 z: 1.00
Bed x: 155.00 y: 0.00 z: 1.16
Bed x: 155.00 y: 180.00 z: 1.38
Bed x: 0.00 y: 180.00 z: 1.09
Eqn coefficients: a: 0.00 b: 0.00 d: 0.97
planeNormal x: -0.00 y: -0.00 z: 1.00
echo:endstops hit: X:0.01 Y:179.96 Z:1.09
SENDING:G29
Bed x: 0.00 y: 0.00 z: 1.08
Bed x: 155.00 y: 0.00 z: 1.22
Bed x: 155.00 y: 180.00 z: 1.45
Bed x: 0.00 y: 180.00 z: 1.16
Eqn coefficients: a: 0.00 b: 0.00 d: 1.05
planeNormal x: -0.00 y: -0.00 z: 1.00
echo:endstops hit: Y:179.99 Z:1.16

Any ideas on how to solve this?
Seems like the data obtained by G29 isn't used at all even if I can clearly see the Z axis move while printing.

Thank you in advance

[Roxy]

Those numbers do not look correct. What version of the code do you have? It might make sense to being up RCBugFix which is the most current and fixed version of Marlin. You can get it here if you decide to do that: https://github.com/MarlinFirmware/Marlin/tree/RCBugFix

It would also be good to see a M48 V4 on your current setup so we can see how the Z-Probe is behaving.

[Zuru1]

Those numbers do not look correct. What version of the code do you have? It might make sense to being up RCBugFix which is the most current and fixed version of Marlin. You can get it here if you decide to do that: https://github.com/MarlinFirmware/Marlin/tree/RCBugFix

It would also be good to see a M48 V4 on your current setup so we can see how the Z-Probe is behaving.

Hi,

I'm on Marlin version 1.0.2-2.

Front-left point
>>> m48 v4
SENDING:M48 V4
M48 Z-Probe Repeatability test.   Version 2.00
Full support at: http://3dprintboard.com/forum.php
Positioning probe for the test.
1 of 10   z: 0.906500 mean: 0.906500   sigma: 0.000000
2 of 10   z: 0.908125 mean: 0.907312   sigma: 0.000813
3 of 10   z: 0.909250 mean: 0.907958   sigma: 0.001129
4 of 10   z: 0.908875 mean: 0.908187   sigma: 0.001055
5 of 10   z: 0.909500 mean: 0.908450   sigma: 0.001080
6 of 10   z: 0.910125 mean: 0.908729   sigma: 0.001167
7 of 10   z: 0.911000 mean: 0.909054   sigma: 0.001341
8 of 10   z: 0.911875 mean: 0.909406   sigma: 0.001563
9 of 10   z: 0.912875 mean: 0.909792   sigma: 0.001833
10 of 10   z: 0.913250 mean: 0.910138   sigma: 0.002025
Mean: 0.910138
Standard Deviation: 0.002025
echo:endstops hit:  Z:0.91

Front-right point
>>> m48 v4
SENDING:M48 V4
M48 Z-Probe Repeatability test.   Version 2.00
Full support at: http://3dprintboard.com/forum.php
Positioning probe for the test.
1 of 10   z: 0.853875 mean: 0.853875   sigma: 0.000000
2 of 10   z: 0.857125 mean: 0.855500   sigma: 0.001625
3 of 10   z: 0.859375 mean: 0.856792   sigma: 0.002258
4 of 10   z: 0.860250 mean: 0.857656   sigma: 0.002463
5 of 10   z: 0.861750 mean: 0.858475   sigma: 0.002745
6 of 10   z: 0.862250 mean: 0.859104   sigma: 0.002874
7 of 10   z: 0.863500 mean: 0.859732   sigma: 0.003073
8 of 10   z: 0.864375 mean: 0.860313   sigma: 0.003259
9 of 10   z: 0.865500 mean: 0.860889   sigma: 0.003478
10 of 10   z: 0.866750 mean: 0.861475   sigma: 0.003739
Mean: 0.861475
Standard Deviation: 0.003739
echo:endstops hit:  Z:0.87

Back-right point
>>> m48 v4
SENDING:M48 V4
M48 Z-Probe Repeatability test.   Version 2.00
Full support at: http://3dprintboard.com/forum.php
Positioning probe for the test.
1 of 10   z: 0.156000 mean: 0.156000   sigma: 0.000000
2 of 10   z: 0.156875 mean: 0.156438   sigma: 0.000437
3 of 10   z: 0.158125 mean: 0.157000   sigma: 0.000872
4 of 10   z: 0.158500 mean: 0.157375   sigma: 0.000996
5 of 10   z: 0.159875 mean: 0.157875   sigma: 0.001339
6 of 10   z: 0.168000 mean: 0.159563   sigma: 0.003966
7 of 10   z: 0.160875 mean: 0.159750   sigma: 0.003701
8 of 10   z: 0.160750 mean: 0.159875   sigma: 0.003478
9 of 10   z: 0.160250 mean: 0.159917   sigma: 0.003281
10 of 10   z: 0.160625 mean: 0.159987   sigma: 0.003120
Mean: 0.159987
Standard Deviation: 0.003120
echo:endstops hit:  Z:0.16

Back-left point
>>> m48 v4
SENDING:M48 V4
M48 Z-Probe Repeatability test.   Version 2.00
Full support at: http://3dprintboard.com/forum.php
Positioning probe for the test.
1 of 10   z: -0.233375 mean: -0.233375   sigma: 0.000000
2 of 10   z: -0.232250 mean: -0.232813   sigma: 0.000562
3 of 10   z: -0.231875 mean: -0.232500   sigma: 0.000637
4 of 10   z: -0.232125 mean: -0.232406   sigma: 0.000575
5 of 10   z: -0.232125 mean: -0.232350   sigma: 0.000527
6 of 10   z: -0.232125 mean: -0.232313   sigma: 0.000488
7 of 10   z: -0.232250 mean: -0.232304   sigma: 0.000452
8 of 10   z: -0.232125 mean: -0.232281   sigma: 0.000427
9 of 10   z: -0.232125 mean: -0.232264   sigma: 0.000406
10 of 10   z: -0.232125 mean: -0.232250   sigma: 0.000387
Mean: -0.232250
Standard Deviation: 0.000387
echo:endstops hit:  Z:-0.23

And G29 results
SENDING:G29
Bed x: 1.00 y: 1.00 z: 0.94
Bed x: 155.00 y: 1.00 z: 0.87
Bed x: 155.00 y: 179.00 z: 1.11
Bed x: 1.00 y: 179.00 z: 0.75
Eqn coefficients: a: 0.00 b: 0.00 d: 0.83
planeNormal x: -0.00 y: -0.00 z: 1.00
echo:endstops hit:  Z:0.75

I could try the the RC version but I have had problem with that version since I couldn't obtain z-probe offset without disabling the min and max software endstops

[Zuru1]

Update:

Tried the Marlin RCBugFix version and the same problem that I had before still prohibits me from lowering the nozzle so that I can calculate the z-probe offset.
Even if I change the Z min to -2 I couldn't lower the nozzle, so I'm basically back to square one.
I think I will get a new glass since the one I have ain't totaly flat, a few 0. mm apart from corner to corner and try that out and see if it helps.

Q: Does it matter if G90 is placed after or before G29 in the gcode?

[Roxy]

What are these values in your Configuration.h file?



#define min_software_endstops false //true // If true, axis won't move to coordinates less than HOME_POS.
#define max_software_endstops false //true // If true, axis won't move to coordinates greater than the defined lengths below.

If you set them to false... It is very possible your Z probe comes alive.

[Zuru1]

They are both set to true in both versions.

If I disable them (set them to false), won't that be totally removing any safety measures for travel over the a given axis length?

Does it matter if G90 is placed after or before G29 in the gcode?

[Zuru1]

So I changed to the RCBugFix version and disabled min_software_endstops and was able to calculate z probe offset.
Same behavior as before when printing the calibration print no matter how close the nozzle is without being to close to the bed(without scratching the bed), no adjustments where made during printing as can be seen in the picture.

Attachment 9866Attachment 9867

I will try to get a new glass tomorrow and see how that goes before i revert back to not using ABL since I have been at it for while now and haven't come any were since I configured this feature.

Edit: Here is the output of G29 V4 and print out put on RCBugFix firmware
>>> G29 V4
SENDING:G29 V4
G29 Auto Bed Leveling
echo:busy: processing
echo:busy: processing
Bed X: 155.000 Y: 0.000 Z: 0.409
echo:busy: processing
echo:busy: processing
Bed X: 0.000 Y: 0.000 Z: 0.962
echo:busy: processing
echo:busy: processing
Bed X: 0.000 Y: 185.000 Z: 0.900
echo:busy: processing
echo:busy: processing
Bed X: 155.000 Y: 185.000 Z: 0.769
Eqn coefficients: a: -0.00220726 b: 0.00080608 d: 0.85656232
Mean of sampled points: 0.76006250
Bed Height Topography:
+--- BACK --+
| |
L | (+) | R
E | | I
F | (-) N (+) | G
T | | H
| (-) | T
| |
O-- FRONT --+
(0,0)
+0.14019 +0.00894
+0.20194 -0.35106
Corrected Bed Height vs. Bed Topology:
+0.00000 +0.21088
+0.21088 +0.00000
Bed Level Correction Matrix:
+0.999998 +0.000000 -0.002207
+0.000002 +1.000000 +0.000806
+0.002207 -0.000806 +0.999997

Print started at: 23:10:22
G29 Auto Bed Leveling
Eqn coefficients: a: -0.00225887 b: 0.00070338 d: 0.84381237
Bed Level Correction Matrix:
+0.999997 +0.000000 -0.002259
+0.000002 +1.000000 +0.000703
+0.002259 -0.000703 +0.999997

Also my initial beginning of my gocde:
; generated by Slic3r 1.2.9 on 2016-11-21 at 20:37:21

; external perimeters extrusion width = 0.40mm
; perimeters extrusion width = 0.67mm
; infill extrusion width = 0.67mm
; solid infill extrusion width = 0.67mm
; top infill extrusion width = 0.67mm
; support material extrusion width = 0.40mm

M107
M190 S60 ; set bed temperature
M104 S215 ; set temperature
G28 ;Home all axises
M117 Homing all axises
G29 ;Level bed
M117 Auto bed leveling
G1 Z5 F5000 ;Lift nozzle
G92 E0 ;zero the extruded length
G1 F200 E3 ;extrude 3mm of feed stock
G92 E0 ;zero the extruded length again
G1 F9000
M117 Printing...
M109 S215 ; wait for temperature to be reached
G21 ; set units to millimeters
G90 ; use absolute coordinates
M82 ; use absolute distances for extrusion
G92 E0
G1 E-1.00000 F2400.00000
G92 E0
G1 Z0.250 F7800.000

[Zuru1]

Well I replaced the glass and got a flat bed and recalibrated the hole printer twice just to make sure and still no change.
I also tried RC7 firmware version to see if that would solve anything, still same end results.

So my only solution now is to go back to the original endstop since I already have invested a lot of time and still have nothing to show for it.