Cannot get Duet 3 MB6HC and BLTouch working

Phat32

I have been trying to get a BLTouch setup on the Duet 3 for a little while now but have so far blown 2 BLTouchs trying to get it setup. I have a 3rd but am a tad terrified connect it cause if it blows I really can't justify to my wife getting a 4th (she doesn't know about the 3rd... I will pay for that later)

I know it is blown because when I issue the M280 command to test I hear a loud pop from the BLTouch and the PCB becomes extremely hot to the touch when connected to power after that.

There is lots of conflicting info out there for the Duet 3 from what I can tell. In the documentation it shows the wiring layout and what to connect where, but then in the configurator I see it says to connect power to 3.3v instead of 5v. Then I use the config tool to create the config.g file, but it's config is different for the Z than I have seen in others using the same pin layout I am. I am especially confused as RRF Config Tool adds the pull up resistor to io7.in but in other configs posts I don't see that included and the documentation it seems it isn't needed on io7.

Just wondering is someone can help guide me in the right direction.

config.g Z-Probe

; Z-Probe
M950 S0 C"io7.out"                           ; create servo pin 0 for BLTouch
M558 P9 C"^io7.in" H5 F120 T6000             ; set Z probe type to bltouch and the dive height + speeds
G31 P500 X10 Y-35 Z2.5                       ; set Z probe trigger value, offset and trigger height
M557 X15:215 Y15:195 S20                     ; define mesh grid

deployprobe.g

M280 P0 S10 ; deploy BLTouch

retractprobe.g

M280 P0 S90 ; retract BLTouch

Wiring:

Empty-> 3.3v
White -> io7.in
Black + Brown -> GND
Orange -> io7.out
Red -> 5v

Phat32

A small update.

I found this post: https://forum.duet3d.com/topic/18261/bltouch-v3-1-toolboard-resolution

I followed their steps and when M280 P0 S60 is issued the LED will light up. However when the M280 P0 S120 is issued the LED become starts flashing rapidly and very dimly and just doesn't seem to do anything.

When I issue M280 P0 S10 the LED turns off entirely and it doesn't seem to do anything. The Z in the interface also stays at 0 and doesn't change to 1000.

I am getting super annoyed at this. It should just work but yet just doesn't want to. Ugh haha

Phat32

I found the problem and OMG I am livid.

The colours for the wires from my BLTouch are not the ones from the documentation at all.

The pattern on mine is as follows (From left to right)

White, Black, Yellow, Blue, Red.

So this means the pin layout is

Red = io7.in (Z Probe Pin)
White + Blue = GND
Yellow = io7.out (Servo Control)
Black = 5V

The only colour that matched up is Yellow. I am extremely surprised the BLTouch is still working with just about everything connected incorrectly.

The probe is working now, at least in testing. I need to fix the wiring going to the actual print carriage and put it through its paces

Phat32

Just got it connected and while it deploys and retracts properly now, it doesn't send back that the Z has been triggered. It detects it in the probe itself, the light lights up and everything, but nothing in the DWC. The Z stays at 0. Willing to bet putting 5V through the Z probe fired something. Now I am down 3 BLTouches. Yay.

droftarts

@phat32 are these clone BLTouch, or genuine? The trigger registers very quickly, often too quickly to show on DWC. See https://duet3d.dozuki.com/Wiki/Test_and_calibrate_the_Z_probe

If your Z probe produces a short pulse when it triggers (e.g. Smart Effector, Precision Piezo, FSRs with John SL board, BLTouch) then the pulse will probably be too short for you to see. Proceed to the dynamic test.

Ian

Nuramori

@phat32 half the time I don’t see a trigger value in dwc, but it is registering/triggering properly (since my platform stops

The pulse is so short that the web interface won’t always catch it.

dc42

@phat32, I'm glad you found the problem. Was it a genuine BLTouch, and if not, what brand was it? If we can identify the brand then I'll add a warning to your documentation about it. The genuine BLTouch manual is at https://5020dafe-17d8-4c4c-bf3b-914a8fdd5140.filesusr.com/ugd/f5a1c8_d40d077cf5c24918bd25b6524f649f11.pdf and clearly states the wire colours on the first page.

On Duet 3 and 3 Mini it's best not to enable pullup resistors on ioN.in pins. They all have permanent pullup resistors anyway, and if you enable the MCU ones as well then the noise margin is reduced.

droftarts

@phat32 Thanks for the wiki edit. I lightly amended it, also adding the colours. https://duet3d.dozuki.com/Wiki/Connecting_a_Z_probe#Section_BLTouch

Ian

Phat32

@droftarts Genuine as far as I can tell. Sadly when I tried to use the sensor to level it just drove the bed into the nozzle and would give the alarm flashing on the BLTouch. Pretty sure pumping 5V through the Z Probe killed something that is preventing it from sending a signal back.

And no problem This was an expensive lesson and I want to make sure others have the info needed to prevent it from happening to them

@Nuramori sadly my platform doesn't stop moving and drives into the nozzle and the BLTouch alarm activates. Oh well, lessons learned right? haha

@dc42 Thank you From all apperances and what not it appears to be genuine but in a Creality CR-10 V2 kit. I saw posts elsewhere of 2 others who had this happen, it could be the sellers and including alternate cabling or something.

Thank you! I thought that was the case but was unsure. I will make that change in the config It would be good if the RRF Config Tool can be updated when configuring the BLTouch on the Duet 3 to remove the pull up, if that is possible

Nuramori

@phat32 Do you have the correct offset from the nozzle? I've found the ideal distance to be 6.1mm from the nozzle tip to the flat base where the pin opening is. If it's too far away, then you can get the situation you're describing.

Nuramori

@dc42 Is the pullup info you mentioned also apply to toolboards? I had a devil of a time with my BLtouch until I enabled the pullup on the duet3 before I used a toolboard, and continued to use the pull up on the toolboard as a habit.

Nurgelrot

@dc42 You can't even trust real BLTouch I have one I know is genuine and the wiring was reversed in the small connector at the BLTouch. Always trace the wires.

Phat32

@nuramori Unfortunately it happened during the initial probe process, at that point the z offset doesn't really take effect. It will stop when the probe trigger is detected, the offset comes into effect when doing the print operation itself. To be safe I also set a very low initial offset so I can adjust from there

Nuramori

@phat32 I’m not talking about the g-code setting - I’m talking about the physical distance of the bltouch itself vertically from the tip. If it’s too far then the probe tip doesn’t move enough to trigger before the nozzle crashes.

Phat32

@nuramori Ah my misunderstanding, but also yes haha. The BLTouch when retracted is barely above the height of the nozzle

droftarts

@phat32 are the probes working now, despite miswiring? I thought they were quite robust, and as they run on 5V, miswiring shouldn’t make the magic smoke come out.

You should be able to test without actually homing, by sending M401 to deploy the probe, then lightly pushing in the probe pin, or moving bed up/nozzle down in small Z increments until it just pushes in the pin, and makes the red LED come on in the BLTouch. DWC should report a probe reading of 1000. If you push too hard you trigger the alarm mode on the BLTouch (flashing red LED); clear it with M42 P0 S160.

Ian

Phat32

@droftarts I can confirm the miswiring causes the z probe to not report back to the main board. I tried many times and while the probe would deploy it would not show any value in the UI but 0 when manually touched.

My 4th was just delivered and after swapping it out it worked perfectly instantly. With the wrong wires going to ground and the Z probe acting as the 5V had fired something for sure.

With the 2nd one I did 100% hear a pop when I issued the M401 command.