Duet 0.8.5 and G32 not adjusting? - FIXED!

So I've started to print some larger items on the bed and noticed one corner of the print appears to be squashed compared to the other corners of the bed. I've checked the bed level with a spirit level and it looks really good, but I have not checked it with various Z probe points. However, I'm using G32 after a G28, so I would have thought it would have adjusted accordingly.

I do see this message in the G-Code Console:

Bed equation fits points [30.0, 10.0, -0.107] [30.0, 130.0, -0.257] [170.0, 130.0, 0.393] [170.0, 10.0, 0.355]

But is there any way to know if its actually adjusting the print based on the Z-plane? The part that appears to be squashed is near the 30,130 area. Did I miss something else in my start code to use the Z plane from G32? This is a CoreXY setup if that matters.

My current start code in slic3r appears as the following:

T0
G21; set units to millimeters
M190 S[first_layer_bed_temperature]; set bed temperature and wait for temperature
G10 P0 S[first_layer_temperature_0] R160; Set tool 0 operating and standby temperatures
M109 S[first_layer_temperature_0]; Set tool 0 operating temperature and wait for temperature
G28 ; home X and Y and Z
G32 ; Probe Z and calculate Z plane
M83 ; Set extruder to relative mode

I know the M109 and G10 are almost redundance, but it wasn't waiting for temp with G10 and it does with M109.

And these are the relevant sections of my config.g file

; The following M557 commands are not needed if you are using a bed.g file to perform bed compensation
;*** Adjust the XY coordinates in the following M557 commands to suit your build and the position of the Z probe
M557 P0 X30 Y10						; Four... 
M557 P1 X30 Y130						; ...probe points...
M557 P2 X170 Y130					; ...for bed...
M557 P3 X170 Y10						; ...levelling
;M557 P4 X100 Y100					; 5th probe point for levelling

; Tool definition
M563 P0 D0 H1						; Define tool 0
G10 P0 S0 R0 X0 Y0					; Set tool 0 operating and standby temperatures
;*** If you have a dual-nozzle build, un-comment the following 3 lines
;M563 P1 D1 H2						; Define tool 1
;G10 P1 S0 R0	X0 Y0					; Set tool 1 operating and standby temperatures
M143 S275 							; Define Maximum hot-end temperature

; Z probe
;M558 P1 X0 Y0 Z1						; Analog Z probe, also used for homing the Z axis
;G31 Z1.20 P500						; Set the probe height and threshold (put your own values here)
M558 P1 X0 Y0 Z1 H3 F200 T5000			; smart IR Z probe, used for homing Z axis only, dive height 3mm, probe speed 200mm/min, travel speed 5000mm/min
G31 P500 X0.0 Y0.0 Z0.52 				; set threshold and offsets

;*** If you are using axis compensation, put the figures in the following command
M556 S78 X0 Y0 Z0						; Axis compensation here

Edited 2 time(s). Last edit at 06/27/2016 09:31PM by PDBeal.

You can tell when bed compensation is being applied because the Z motor will be turning a little as the head moves in the XY plane.

I suggest you check that the Z probe trigger height is the same at each of the 4 probe points. If it isn't then you won't get accurate compensation.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

Checking Z Probe trigger height would be the same as the calibration when I send it a G30 S-1? skipping the G92 Z0 part?

And as I watch it, its not moving Z at all until layer change or retractions happen.

Edited 1 time(s). Last edit at 06/26/2016 06:28PM by PDBeal.

So after moving the head around to my 4 points and adjusting the 3 leveling spring screws, it triggers the Z probe around 0.60-0.63 for all 4 corners. I kept moving Z to 10, then doing G30 S-1 to get a distance on each corner. Sometimes they report 0.61 other times it reports 0.60 depending on the corner and other corners with no screw seem to report 0.62-0.63. At this point I think this is the accuracy of the differential IR probe reading my MK2b heatbed through a piece of glass? With this setup, G32 should offset the Z plane? Or do I need to get them even closer still?

After this setup, I then re-calibrated the Z offset for the differential IR sensor with the hot end at 240C. Loaded up the same print and the same rear corner (30,130) was squashed. I got the following message in the G-Code Console

Bed equation fits points [30.0, 10.0, -0.005] [30.0, 130.0, -0.180] [170.0, 130.0, 0.420] [170.0, 10.0, 0.420]

But I do not see the Z axis moving whatsoever during the print (outside of layer changes and retraction). Oddly enough the corner that is squashed is the left side of the single screw/spring side with the screw being in the middle. If it matters, I'm running 1.13b firmware on the 0.85 Duet. Am I missing something with my calibration, or is something not quite right somewhere?

Edited 2 time(s). Last edit at 06/27/2016 12:12AM by PDBeal.

Quote
PDBeal
So after moving the head around to my 4 points and adjusting the 3 leveling spring screws, it triggers the Z probe around 0.60-0.63 for all 4 corners. I kept moving Z to 10, then doing G30 S-1 to get a distance on each corner. Sometimes they report 0.61 other times it reports 0.60 depending on the corner and other corners with no screw seem to report 0.62-0.63. At this point I think this is the accuracy of the differential IR probe reading my MK2b heatbed through a piece of glass? With this setup, G32 should offset the Z plane? Or do I need to get them even closer still?

After this setup, I then re-calibrated the Z offset for the differential IR sensor with the hot end at 240C. Loaded up the same print and the same rear corner (30,130) was squashed. I got the following message in the G-Code Console

Bed equation fits points [30.0, 10.0, -0.005] [30.0, 130.0, -0.180] [170.0, 130.0, 0.420] [170.0, 10.0, 0.420]

But I do not see the Z axis moving whatsoever during the print (outside of layer changes and retraction). Oddly enough the corner that is squashed is the left side of the single screw/spring side with the screw being in the middle. If it matters, I'm running 1.13b firmware on the 0.85 Duet. Am I missing something with my calibration, or is something not quite right somewhere?

With that bed equation, the Z axis should definitely be moving a little. Are you certain that it isn't? If your machine is an Ormerod, you can watch the printed Z gears; but on other machines with the Z motor(s) driving threaded rods or leadscrews directly, it may be difficult to see.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

No, my machine was a Smartfrendz ALU kit that I've since put the Duet 0.8.5 controller on. The Z axis is a GT2 Belt with 20 teeth and a 608 bearing. I would have thought it was moving too, but I don't see it move at all besides layers and retraction. Is there any way to check and see if G32 has modified the Z plane, or modified the Z points?

Is it worth trying an older firmware to double check something didn't get accidentally broken in an update? I don't think the G32 was working correctly with 1.12a that I started with either, but I never leveled the bed at that point with the IR trigger distances (G30 S-1).

Edited 1 time(s). Last edit at 06/27/2016 11:16AM by PDBeal.

I'll do a test print with bed compensation enabled on my Ormerod running 1.13b later this evening.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

Is there anything I needed to put in my config.g file to enable bed compensation beyond the M557 points? At this point I'm only using M557 and then the G32 command in my start script, but now you have me wondering if I'm missing something else to enable the z plane offset.

The firmware will use a bed.g file if you have one, but if not then it will use the M557 points.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

I have just done a test print using firmware 1.13b with bed compensation enabled, and the Z axis is indeed moving to make height corrections during the print. The amount of movement is very small. I can only see it because the Ormerod has visible Z gears and I can see them rotating a little. If I grip the Z threaded rod then I can also feel it.

Edited 1 time(s). Last edit at 06/27/2016 03:41PM by dc42.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

Is that using bed.g or the M557 points?

I know there's not supposed to be any difference, but it has me wondering if I create a bed.g and see if there's a change. If it's a bed.g file, any chance you can attach a copy of it? I've seen on the wiki about a few samples, but it seems to be more complex than the 4 M557 lines in config.g.

Edited 2 time(s). Last edit at 06/27/2016 03:48PM by PDBeal.

Ah, I found my issue. There was a small piece of orange kapton tape under the glass that was under the area that was being squashed which I presume would interfere with the IR probe. The kapton tape was covering the electrical connections of the MK2b PCB. Once I peeled this piece of tape off, it wasn't squashing that corner anymore and I could see the Z axis moving ever so slightly. This was after I went through different iterations of bed.g vs M557 vs 4 points vs 5 points. It looks to be working now. I could see the Z axis moving when I enabled the 5th point, it was quite abrupt due to the kapton tape point. Now it all seems to be working.

Edited 1 time(s). Last edit at 06/28/2016 02:43PM by PDBeal.