DC42 - Improved Z probe, fan control, illumination board

So I've been running DC42's improved IR sensor board for around 2 weeks now. I'm extremely happy with it - the auto z height sensing has been working perfectly for finding the z-height in the centre of my bed, and the added illumination and better fan control is fantastic. I'm really pleased with the addition to my Ormerod, and highly recommend it to everyone.

I am having a minor problem when using it for auto bed compensation however - I'm not sure whether it's something physical or one of my printer settings. Usually my bed is pretty level within 0.2mm in all corners) and I print without any bed compensation. I worked on the printer today however, so it is slightly skew at the moment, and I figured this would be a good test for the auto bed compensation.

Essentially, when I zero the Z height in the middle of the bed, it zeroes perfectly. However, for some reason, when I measure the height in the corners of the bed, something seems to go awry.
For example:

My G31 command in my config.g is G31 P500 Z0.70. I use the standard A4 sheet of paper test to determine "true" 0.

In the top left corner of my bed, if I issue a G30, the sensor stops at 0.7mm, but the piece of paper gets touched at -0.4mm. So the difference is 0.7 - -0.4, so 1.1mm difference.
In the top right corner, G30 stops the sensor is at 0.7, again the paper is at -0.4, so 1.1 difference
On the bottom right corner the paper touches at -0.3 (1.0 difference).
On the bottom left corner the paper touches at -0.3 (1.0 difference).

However, if I go to the centre of my bed, and I issue the G30 and then do the paper test, the paper sticks at 0.7! So in my 4 corners, the sensor triggers at around 1mm, but in the centre of my bed, it triggers at 0.7mm.

I've switched off the light in the room, no difference.
I've issued an M561 before homing all axis, and repeating the test, and the result is the same.
I've even compensated for the offset of the sensor vs the print nozzle by doing the G30, and then moving the printer nozzle via the web interface, so that the print nozzle is exactly over the place where the sensor was.
My bed is plain aluminium spreader with glass on top, no kaptan or white paper/tape in the corners.
I've tested with both bed and print head at printing temperatures, and with bed at room temp.

I'm at a bit of a loss as to why the sensor value is slightly different in the centre of the bed.

My config.g is as follows:

; Configuration file for RepRap Ormerod
; RepRapPro Ltd
;
;M111 S0 ; Debug off
M550 PDieter RepRapPro Ormerod ; Machine name (can be anything you like)
M551 Preprap ; Machine password (currently not used)
M540 P0xBE:0xEF:0xDE:0xAD:0xFE:0xED ; MAC Address
M552 P192.168.0.14 ; IP address
M553 P255.255.255.0 ; Netmask
M554 P192.168.0.1 ; Gateway
M555 P2 ; Set output to look like Marlin
G21 ; Work in millimetres
G90 ; Send absolute coordinates...
M83 ; ...but relative extruder moves
M906 X800 Y1000 Z1000 E800 ; Set motor currents (mA)
M563 P1 D0 H1 ; Define tool 1
G10 P1 S0 R0 ; Set tool 1 operating and standby temperatures
;M92 E364 ; Set extruder steps per mm - 364 is for Black CADHouse PLA
M92 E424 ; Set extruder steps per mm - 424 is for Funky Blue CADHouse PLA
M574 X0 Y2 Z0 S1 ; Y End Stop is at the far end - DC Firmware specific
M558 P1 ; Use an unmodulated Z probe - DC firmware
G31 P500 Z0.70 ; Set Z probe height and threshold - DC firmware
M557 P0 X55 Y25 ; Five...
M557 P1 X55 Y175 ; ...probe points...
M557 P2 X215 Y175 ; ...for bed...
M557 P3 X215 Y25 ; ...levelling...
M557 P4 X135 Y100 ; ...DC Firmware style - DC recommends using the average of the X and Y's
M556 S78 X0 Y0 Z0 ; Put your axis compensation here
M201 X800 Y800 Z15 E1000 ; Accelerations (mm/s^2)
M203 X15000 Y15000 Z100 E3600 ; Maximum speeds (mm/min)
M566 X600 Y600 Z30 E20 ; Minimum speeds mm/minute

Any ideas? I'll level the bed by hand before I print something again anyway, but it would be nice if the auto bed compensation works too.

Hi pandionx,

Please can you check the following:

1. When you probe in the corners, is the sensor head is at least 10mm from both edges of the bed (preferably more), and well away from the bed clips? You could try doing the G30 a little further in from the corner.

2. Looking along the X axis, is the bottom edge of the sensor board parallel to the bed?

3. Try probing near the mid point of each edge of the bed: at those points, how does the trigger height differ from the trigger height at the centre and at the corners? The X and Y edges may behave differently. Again, make sure the sensor head is at least 10mm in from the edge, preferably more.

4. Is the aluminium heat spreader uniformly shiny/dull at the centre and at the corners? I have recently updated the instructions to recommend putting a piece of black paper or thin black plastic between the heat spreader and the glass, to remove the effect of reflections from the aluminium.

Edited 2 time(s). Last edit at 04/01/2015 05:12PM 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].