@DieterSw For a PNP device, see the example here for how to connect: https://docs.duet3d.com/User_manual/Connecting_hardware/Z_probe_connecting#pnp-output-normally-open-inductive-or-capacitive-sensor

I've added a wiring diagram to make it clearer how to wire a PNP sensor. See above text for values for R1 and R2.
alt text

Ian

Well I am dumb the 5V jumper was missing. I have absolutely no idea how the jumper could have vanished off the board. When I looked to see if the jumper was there the first time I accidently looked at the wrong jumper...

I if found the github for the schematic, there is no regulator. If your using the 5v off the wifi it is directy feed off the jumper.

@Miasmictruth 23fb4d0c-e660-4158-9890-54c4f39639f4-image.png

@moth4017
So I tried on the bed heater OUT0 line - and I got the same response with buzzing!

As I was poking around thinking the buzzing was a weird wiring short, I was able to change the pitch when I touched the power inlet ferrite inductor beads. I removed them and the problem was solved.

But WHY????

Ferrite inductors should smooth the voltage at the power inlet, although also probably changing the impedance. Anyone have an idea why that would cause buzzing from the board (mosfets?)?

@Phaedrux I did not change any config when I swapped inputs. I purchased it from Filastruder in August 2023,

@chrishamm it works well on my machines, thank you again for your time on my issue

@droftarts ok, so a 5v voltage regulator. Makes total sense that the board doesn't work if it is broken. I will try to get a replacement and hope for the best

thanks!

@Nazar it shares the SPI connection with the SBC interface. So in principle you might be able to connect it to the SBC connector; but without spending more time looking at the differences I can't promise that it will work. In particular, I think the firmware may detect the board version and only make WiFi available on the 1.02 and later boards.

USB error -32 seems to be related to unstable or low power situation, other sites recommend adding a powered hub. If one is built into your printer, check that it still is powered.

USB error -71 is likely an initialisation error as a result of the power fluctuation.

Another suggestion I found elsewhere was to completely depower the system, remove the USB devices, reboot and try plugging them in one by one.

I hope that helps.

Unfortunately the motor wire colors and orders are not standardized. the four wires build two phases, and you will need to have one phase on one side of the connector and one phase on the other. Properly measuring this is straight forward if you have a multimeter: the phase pairs will have resistance. If you do not have a multimeter you can use a DuPont wire to "short" each pair (while not connected to the board of course) and trying to turn the motor by hand. Should be harder when a pair is shorted than when it's not.

@DocTrucker said in Powering 12V_EXT on Duet3 1.01:

@droftarts cheers. Missunderstood your first message as 2A limit when using 12V derived from VIN, rather than 2A limit per output when the jumpers are set to feed V_IN to the outs.

That said I'm surprised the selector-jumpers itself doesn't limit the current more than it aparrently does.

Hmm... actually, I think you are correct, as the selector pins are Molex KK too. Molex KK has a current limit of 2.5A (see the 'product specification' links on this page), and that applies to the voltage selector as well as the actual fan pins. Jumpers are rated at around the same https://www.farnell.com/datasheets/2706775.pdf. So the limit per bank is 2A (to be on the safe side), though if you are supplying power separately, and just using the MOSFET on the GND side, 2A is fine for the header. Also, the MOSFET DMT6018LDR appears to be rated for 8.8A: https://www.diodes.com/part/view/DMT6018LDR

I'll check with the others, and will likely update the documentation.

Edit: we use Wurth pin headers, which are rated at 3A, as are the jumpers, so maximum total current for each group of fans fed from a voltage selector is 3A.

Ian

@claustro Thank you for the link. I couldn’t imagine the impact on layer height of the effect. Please keep us informed if PID helps.

@dc42 Finally got around to re-routing the cable, can confirm that fixed this problem. Just thought I'd post in case someone finds this thread relevant. Thanks again!

@droftarts
Thank you for answering this so clearly, I appreciate it ! I'm back running again.
I may have questions in the future, but this will do for the moment.

Cheers!

@dc42
Thank you, that's the Info I needed.

Thank you so much for all your help!
How do I switch to driver 4? Which port on the board is driver 4? What needs to change in my config file?

@o_lampe

Thanks

Will update the same

@janjoh Yeah, After pondering this I think stealing a heater from the expansion connector makes more sense.

@droftarts you wont belive what was behind my problem 🙈 🙈 🙈

So, everything is set as it should be but still the X carriage hits the rod. I start digging and eliminating probable causes. I get to the probe itself. I click home x while holding my finger near the probe - it stops upon reaching my finger and now I start to wonder - whats the difference???

As it turns out, the difference was that my finger was closer to the probe than that piece of paper.

When the probe would get to the paper, it would detect the value of about 500, when It got to my finger, the value was arround 900 and that triggered the probe well and good.

It was the bloody distance from the Z probe to the paper. For some reason, 500 was not enough for it to send the signal back home that it reached the end stop.

This is how "close" the paper was to the probe before.
Senzor slika2.jpg

Now I turned the orange x stop bar so that the longer part is up over that black thingie and now the probe stops when it detects the x stop bar.

But also makes me wonder - 500 value wasnt enough, 900 was. Does this have anything to do with the probe itself, or maybe with the firmware? or a certain config.g setting? Maybe something is telling the probe that 500 value does not mean there is an endstop, but 900 does.

@nikscha The Duet 2 WiFi is set to spreadCycle continuously. You could try running it in spreadCycle on the Mini 5+, and to avoid coil whine have it switch to stealthChop when stationary (set changeover speed at below the jerk value). The spreadCycle on the Mini 5+ is different from the Duet 2; they use different drivers, and different versions of spreadCycle.

Are the other axes noisy? What happens if you swap motors over?

Ian

Hello Ian, that's right, that's how it works without the M950 command.
Thank you very much and problem solved now, thank you.