3.4RC1 1XD board BUG?

[Marcossf]

I think we found a bug in latest RC1.

With version 3.4beta7 when making a Z home whose axis is controlled by a 1XD plate, after triggering the probe, the axis retracts and rises at the speed set in the homeall.g, after setting Datum Z=0 and everything works correctly.

On the other hand, with version 3.4 RC1, after triggering the Z probe, it tries to up to the set position after this operation, at an very very very exaggerated speed (it tried do an instant movement). It's so much speed that the external driver (closed loop absolute encoder) to which the Z axis is connected, goes into alarm and stops/block for safety.

If revert to 3.4beta7, it backs to normal sucessful homeall behaviour and all it works as expected.

@dc42 Any hint? Related with the canbus delay correction, maybe?

None on the external driver config was changed or duet config files.

Homeall.g:

; homeall.g
; called to home all axes
;

M564 S0 ; disable axis limits
G29 S2
;G29 S1 P"heightmap.csv" ; load heightmap
G4 S1 ; wait until loading the heightmap

T-1 ; unselect tool

M584 P5 ; 5 axis visibles 

M584 Z43.0 U44.0 V45.0
M574 Z2 S1 P"!43.io0.in"
M574 U2 S1 P"!44.io0.in"
M574 V2 S1 P"!45.io0.in"

M208 Z100 U100 V100 S0 ; change max limits

G91 ; relative positioning
M208 S0 Z100 U100 V100
G1 H1 Z100 U100 V100 F2000 ; lift ZUV relative to current position
M400
G92 Z100 U100 V100
G1 H1 X-405 Y-255 F3000 ; move quickly to X and Y axis endstops and stop there (first pass)

M584 P3 ; 3 axis visibles

G90 ; absolute positioning

G1 X51 Y0 F3000 ; move to physical X51, Y0 (logical X0,Y0)

M400 ; wait for moves to finish

G92 X0 Y0 ; establish current position as X0,Y0

G30 K0; z probbing
M114
M208 Z0 S1 ; Z axis min limit = 0

G1 H3 Z100 F2000 ; lift Z relative to max position and update M208 Z max limit

M400 ; wait for moves to finish

G91 ; relative positioning
G1 Z-5 F2000 ; move down 5mm for allow compensation
G90 ; absolute positioning

M400 ; wait for moves to finish
G29 S1 P"heightmap.csv" ; load heightmap
G4 S1 ; wait until loading the heightmap

set global.noToolMax = {move.axes[2].machinePosition}
M208 Z{move.axes[2].machinePosition} S0
M564 S1 ; enable axis limits

M114 ; report positions

Config.g related:

;################################# Z PROBE 0 CONFIG 
M950 S0 C"out4"                      	    ; create servo pin 0 for solenoid on Out4
M558 P8 C"!io3.in" A1 H12 R0.5 F800 T4000      ; set Z probe type and the dive height + speeds
G31 X0 Y0 Z0                 	    	    ; set Z probe trigger value, offset and trigger height

; ############################## DRIVER PARAMETERS SETTINGS 
M584 X40.0 Y41.0:42.0 Z43.0 U44.0 V45.0 ; axis drive mapping
M350 X1 Y1 V1 Z1 U1 V1 I1 ; configure axis microstepping with interpolation
M92 X150.53 Y150.53 Z100.12 U100.12 V100.12 ; axis steps per mm
M566 X900.00 Y900.00 Z500.00 U500.0 V500.0 ; set maximum instantaneous speed changes (mm/min)
M203 X18000.00 Y18000.00 Z12000.00 U12000.00 V12000.00 ; set maximum speeds (mm/min)
M201 X300.00 Y300.00 Z500.00 U500.00 V500.00 ; set accelerations (mm/s^2)

[dc42]

@marcossf thanks for your example. This sounds to me the same fault as https://forum.duet3d.com/topic/27248/3-4-rc1-1xd-probe-problem-after-update but your example may be easier for me to replicate.

[Marcossf]

@dc42 Hi David

We've tested the 3.4RC+2 files from your Dropbox link in our printer and it worked. This problem is solved.

Currently we have main 6HC with 3.4RC1 fw, DWC 3.4RC1, all boards 1XD and LC1 with 3.4RC1+2. Tested standalone, not in SBC due to global RPI shortage.

For the moment we have not found any other problems with this version.