Z offset stopped working
-
Hi,
I updated my Duet wifi to 3.3. It seemed my Z offset calibration routine stopped working.
In config.g the basic endstop and limit settings I have are,
; Axis Limits
M208 X-2 Y-4 Z0 S1 ; set axis minima
M208 X330 Y330 Z380 S0 ; set axis maxima; Endstops
M574 X1 S1 P"!xstop" ; configure active-high endstop for low end on X via pin xstop
M574 Y1 S1 P"!ystop" ; configure active-high endstop for low end on Y via pin ystop
M574 Z1 S2 ; configure active-high endstop for low end on Z via pin !e1stop; Filament sensor
M591 D0 P2 C"e0stop" S1 ; Filament runout Sensor for Extruder E0 on E1 Endstop input P1 = signal HIGH.; Z-Probe
M558 P1 C"!zprobe.in" H5 F500 T10000 ; set Z probe type to switch and the dive height + speeds
G31 P500 X-35 Y-4 Z4.85 ; this Z offset is just above the bed, I fine tune with my Z offset routine.My Z offset routine,
M291 P"Press ""OK"" if you would you like to calibrate the Z-offset for this Tool Cartridge." R"Calibrate Z-Offset" S3
M291 P"Homing, please wait..." R"Calibrate Z-Offset" S1
M208 Z-2 S1 ; allow movement below Z0
G1 X165 Y165 Z8 F4000 ; move to center of bed
M558 P0
G30 S-2
M500 ; save results
M501 ; load new data
M208 Z0 S1 ; restore minima
G1 Z10 ; drop build plate
G1 X165 Y0 F4000
M558 P1 C"!zprobe.in" H5 F500 T10000 ; restore probe type
M291 P"Z-offset calibration complete! " R"Calibrate Z-Offset" S1 T3What this leaves in config-override.g is,
G10 P0 Z0.26
After all this I used to be able to move to Z0 and and the nozzle height is where I want it.
Now, the Z height stops at .26mm and I can't go to Z=0. If I could, I would be where I want.
I'm not sure what changed.
Joel
-
I believe your redefinition with M558 resets the probe offset. Move the M501 until after your second M558.
-
From the docs:
M558 with P parameter deletes the existing probe with that K number (if any) and creates a new Z probe. This resets the G31 values for that probe to default values.
Frederick
-
Makes sense. I will move some commands and see what happens.
Thanks
Joel -
Moving the M501 did not help.
It does seem trying to restore the probe type is where things break.
What I'm not sure of is why Z won't go to Z0.
Right now my zero routine is still broken. Not sure how to fix it.
Any thoughts?
Thanks
Joel -
I did some more playing. I found a workaround that I don't really like but it will get me going again.
After the G30 S-2 command, I reset the Z min to 0, the Z axis would always stop at the probe offset amount. the extra offset returned is usually < .5mm. So I reset the Z min to -.5mm and now the bed can go to Z0 and my nozzle height is correct.
There seems to be something weird with resetting the Z min with M208 that it resets to what the G30 command returns instead of 0, M208 Z0 S1. In any case using M208 Z-.5 S1 makes it work for now. What I don't like is that this will allow to drive the nozzle into the bed if not careful.
Hope all that made sense
-
-
I tried the M564 idea M564 S0 to disable limits let me adjust the z-offset, M564 S1 to re enable limits gives me the same issue, Z will not reach Z0, stops at what ever the Z-offset is.
Joel
-
Can you test in 3.4 beta 3?
https://github.com/Duet3D/RepRapFirmware/releases/tag/3.4.0beta3
-
I did not see the zip package to install so I tried the duet2combinedfirmware.bin file and get an error trying to install it,
Error: M997: In-application programming binary "0:/firmware/Duet2_SDiap32_WiFiEth.bin" not found
I dont see the missing file anywhere.
-
It sounds like you might be missing some files from 3.3 in the firmware folder. I suggest you upload the 3.3 zip file again, and then try updating to 3.4 beta 3 again.
https://github.com/Duet3D/RepRapFirmware/releases/download/3.3/Duet2and3Firmware-3.3.zip
-
With 3.4 beta 3, it now seems to work using M564 to disable limits. Using M208 has the same behavior, M208 Z-2 S1 to allow below 0 and then M208 Z0 S1 to restore Z0 limit still stops as Z=offset.
Using M564 works for me.
Edit: I spoke too soon - restoring limits with M564 S1 allowed Z movement below 0. Still seems to stop at where ever the offset is. In this case my offset was actually negative, allowing it to reach Z = 0 ( negative meaning I had to increase nozzle to to table to get the clearance I wanted, it was a little tight ). I increased the default gap between nozzle and table and see the same behavior where Z stops at the offset. So in the end, 3.4 beta 3 did not change anything. I am back to using M208 allow a slight travel below Z0 to work.
Now the only M208 commands are in config.g
M208 X-2 Y-4 Z0 S1 ; set axis min
M208 X330 Y330 Z380 S0 ; set axis maxThanks
Joel -
I thought I came up with a cleaver workaround, but.....
I added a G92 Z0 after the calibration is done. Functionally that works, but then the Z axis display is not sensible. Lets say my offset was .42mm, going to Z = 0 goes to the right position nozzle to table, but the Z height displays as .42mm even though it is Z0 logically. I was expecting G92 to also reset the Z indication to what is specified. Is that not supposed to be the case?There also seems to be an execution issue. I initially had the G92 Z0 line right after the G30 S-2 line. In that case, the nozzle moved to the XY coordinate but stayed against the table. The Z display showed 10mm.
new temporary z offset routine,
T0
M291 P"Press ""OK"" if you would you like to calibrate the Z-offset for this Tool Cartridge." R"Calibrate Z-Offset" S3
M291 P"Homing, please wait..." R"Calibrate Z-Offset" S1
M564 S0; M208 Z-2 S1 ; allow movement below Z0
G1 X165 Y165 Z8 F4000 ; move to center of bed
M558 P0
G30 S-2
M564 S1; M208 Z0 S1 ; allow slight - Z so zoffset can work
G92 Z0
G1 Z10 ; drop build plate
G1 X165 Y0 F4000
M558 P1 C"!zprobe.in" H5 F500 T10000 ; restore probe type
G31 P500 X-35 Y-4 Z4.85
M291 P"Z-offset calibration complete! " R"Calibrate Z-Offset" S1 T3 -
@joel said in Z offset stopped working:
I was expecting G92 to also reset the Z indication to what is specified. Is that not supposed to be the case?
Can you post a screenshot of what you mean?
Can you also send M114 and see what positions are reported? -
Not sure why this got moved to beta, it is 3.3 release. 3.4 beta 3was just a test I did for you.
Here are some pics, to walk through them,
First I homed all three axis. This leaves my nozzle a little above the bed ( per my settings )
I then did the Z offset, in this case it was .56mm. Near the end of the z offset routine I drop the build surface 10mm.At this point I raise the build surface 10mm, the nozzle and surface have the clearance I offset to, a piece of paper. At this point the Z axis display shows .56mm. Note: this differs form the original issue in that the Z axis would stop at the offset, the nozzle would be .56mm above the surface. Using G92 now puts the nozzle at the right height but the Z display shows .56mm. See Z56.jpg
Next I did the M114 command and it shows Z .56mm
I then did G92 Z0 again, and M114, Z still shows .56mm. See M114.jpg
In the past I thought what ever an axis was at, doing G92 would reset it ( and the readout ) to what ever you specify. So if the Z axis was at .56mm and I do a G92 Z0, that would reset the axis and display to 0. See the steps in G92.jpg
Never minding this ( G92 not doing what I expect ), There still seems to be an issue with G30 S-2 changing something in my axis limits it should not be doing. It should simply be adding or subtracting from the initial offset in config.g, Not mess with endpoints at all.
Also, from here, G0 Z0 does not move the table, it just stays at .56, G90 G0 Z5 moved to 4.44mm, G91 G0 Z5 moved to 5mm ( after going back to Z0 ( .56mm displayed )). some error here in measuring, but the take away is something is affecting absolute ( Z offset )
Z56.jpg
M114.jpg
G92.jpg
-
@joel said in Z offset stopped working:
Not sure why this got moved to beta, it is 3.3 release. 3.4 beta 3was just a test I did for you.
The problem was still present in 3.4 beta3, correct?
-
Can you upload us your full config.g and homing files?
-
This post is deleted! -
