G32 keeps exceeding bounds of M557

dc42

@vlex G32 is not deprecated. The G30 commands in bed.g are allowed to exceed the normal bed limits, because on delta machines it is advantageous to probe between the towers outside the usual printable bed radius, where the mechanics supports that.

vlex

The G30 commands in bed.g are allowed to exceed the normal bed limits, because on delta machines it is advantageous to probe between the towers outside the usual printable bed radius, where the mechanics supports that.

@dc42 Well, this confuses me - I've got a mini kossel and executing G32 is exactly what broke my effector, because a couple of times I wasn't fast enough to hit the emergency break before the effector twisted the rods on one of the axis. Is this meant for only effectors using magnetic joints, or can you give me an example where this probing outside of the printable area won't kill the suspension of the effector like it did mine? If this is not limited to only magnetic suspension it means that either my printer is somehow defective (although it very well might be Layer8, due to the numerous improvements I've done), OR I'm severely overestimating the build radius of my printer...

P.S. Can the orthogonal axis correction in M557 be useful if the pillars geometry is off or askew? Or is it only available for cartesian machines?

P.P.S. and slightly offtopic: Whilst fiddling with the settings, my sd card gave up so I had to work out the fw to get the right DWC and had to get a vanilla configuration from the RRFCT and good that I posted here my config.g, to effectively back it up, lol. This saved me from reconfiguring everything from scratch

dc42

@vlex whether or not the machine can probe between the towers when outside the usual printing radius, and how far outside it can go, depends on your mechanics and the shape of the bed.

M557 is unlikely to be useful on a delta because if your machine is exhibiting XY skew then it is almost certainly exhibiting distortion too. You can print a pattern of lines in both X and Y directions over the whole bed spaced at e.g. 25mm intervals to check for this.

vlex

@dc42 That makes sense.

I've got a weird new problem though - after init of the duet, the prints are not great, dare i say even bad - usually there's gaps between the hotend and the bed or sometimes it drags across the bed. After running G29 it seems to work slightly better, BUT as soon as I run G32 I think the geometry goes right out of the window. I'm trying to print a spool holder and one of the parts is a big triangle with a circle inside it:

All is good and dandy until the G32, after it all of the sudden the circle becomes an ellipse and a quite elongated one. I'm assuming something ain't right in the setup above (I'm reusing it except I added springs under the bolts of the bed so it's now about 10-12mm higher, so the Z-height is less now). The elongation is across practically the Y-axis (imagine a line between the furthermost pillar on the pic above and the centre of the section).

Any ideas what I'm doing wrong OR how to troubleshoot this? Thanks in advance!

jay_s_uk

@vlex can you post your bed.g?

vlex

@jay_s_uk sure thing:

; called to perform automatic delta calibration via G32
;
; generated by RepRapFirmware Configuration Tool v3.3.15 on Fri Feb 24 2023 13:57:20 GMT+0000 (Greenwich Mean Time)
M561 ; clear any bed transform
G28  ; home all towers
; Probe the bed at 3 peripheral and 3 halfway points, and perform 6-factor auto compensation
; Before running this, you should have set up your Z-probe trigger height to suit your build, in the G31 command in config.g.
G30 P0 X0 Y40.43 H0 Z-99999
G30 P1 X38.22 Y-22.07 H0 Z-99999
G30 P2 X-47.63 Y-27.5 H0 Z-99999
G30 P3 X0 Y15.47 H0 Z-99999
G30 P4 X14.87 Y-8.59 H0 Z-99999
G30 P5 X-23.82 Y-13.75 H0 Z-99999
G30 P6 X0 Y0 H0 Z-99999 S6
; Use S-1 for measurements only, without calculations. Use S4 for endstop heights and Z-height only. Use S6 for full 6 factors
; If your Z probe has significantly different trigger heights depending on XY position, adjust the H parameters in the G30 commands accordingly. The value of each H parameter should be (trigger height at that XY position) - (trigger height at centre of bed)

The diff between the posted above config.g and the current one is here:

;Current                                    ; Previously posted
; Set delta radius, diagonal rod length, printable radius and homed height
M665 R105.6 L218 B85 H333.5                 ; M665 R105.6 L218 B85 H360

; define mesh grid
M557 X-50:50 Y-50:75 R75 S20                ;M557 X-50:50 Y-50:75 R55 S10

Atm there's no config-override.g so not posting one and the heightmap.csv atm looks like this (not sure whether it helps for me to post it, but can't hurt, can it?):

RepRapFirmware height map file v2, min error -1.601, max error 1.398, mean -0.163, deviation 0.747
xmin,xmax,ymin,ymax,radius,xspacing,yspacing,xnum,ynum
-50.00,50.00,-50.00,75.00,75.00,20.00,20.00,6,7
      0, -0.896, -1.552, -1.479, -1.331, -0.823
 -0.137, -0.397, -0.472, -0.389, -0.280, -0.023
  0.056, -0.385, -0.198, -0.004,  0.360,  0.852
  0.284,  0.220,  0.220,  0.329,  0.729,  1.180
 -0.159, -0.699, -0.389,  0.082,  0.683,  1.398
 -0.782, -0.665, -0.476, -0.190,  0.501,  1.342
      0,      0, -1.601, -0.950,      0,      0

jay_s_uk

@vlex nothing jumps out at me as looking wrong.
Maybe it's time to take the plunge and update to 3.4.5?

dc42

@vlex what type of Z probe do you have?

vlex

@dc42 it's an inductive LJ12A3-4-Z/BX. From what I've tested it triggers consistently at 0.69 height for the nozzle (the sensor is higher than the nozzle by another 0.4mm or so).

vlex

@jay_s_uk if i did that, would i be able to reuse my config? Also I tried upgrading when my sd card gave out, but nothing worked, so i had to revert to whatever fw i had and a compatible dwc.

jay_s_uk

@vlex there would be some changes required but it wouldn't be insurmountable.

T3P3Tony

@vlex said in G32 keeps exceeding bounds of M557:

All is good and dandy until the G32, after it all of the sudden the circle becomes an ellipse and a quite elongated one

are you 100% sure you have the probe offsets correct? if you have the probe XY offsets wrong that could explain some of the behaviour you are seeing.

See:
https://docs.duet3d.com/en/User_manual/Connecting_hardware/Z_probe_testing#measuring-probe-x-y-offset