Duet 3 6HC Rev. 1.01 Faulty

NeueKlasse

When the communication gets lost the Heater is still Heating, although M303 is working with all that Communication Loss Errors (no Header)

The PSU is the Original One so i think there must be plenty of Room for the Power...
(1000W 24V)
i was also surprised that the fuse was blown... i dont measured the Bed Amps but
i found that the Bed must have around 300-400W (i hope that's correct)
so around 12-16A

The Thing is it doesn't matter if all is connected to the board or when i disconnect everything from the Board.... i have Communication loss Errors every second!

So the Bed Heater doesn't have to Play with it to get hundreds of Errors...
(yesterday i thought so... but today i'm not sure what to think...)

Veti

@NeueKlasse said in Duet 3 6HC Rev. 1.01 Faulty:

i found that the Bed must have around 300-400W (i hope that's correct)
so around 12-16A

the max current on the bed is 15A. So you ware most likely exceeding that.

measure the resistance of the bed and subtract the resistance of the multimeter with no connection

NeueKlasse

@Veti

Yep the Fuse is only 15A (Mosfet on the Duet is rated for 18A i think) but even with
that (i had just tested the Option without extra Mosfets...) i had Communication Errors..

Measured the Bed... 1.5 to 1.6 Ohm, so at 1.6 V/A ... the Bed drawn Current must be around 15A

NeueKlasse

This post is deleted!

NeueKlasse

What does the Point SPI Overruns/Underruns say? (M122)

NeueKlasse

Although, is it possible to exchange the board? I need the Printer to work..

NeueKlasse

I remembered that i had connected the Duet and Pi4 in that way that the Duet supplies the Pi 5V before i put everything in the printer...
Then i wanted a extra Powersupply to not overload the Duet...

Today i disconnected the Pi Powersupply and added the needed Jumper... et voila... problems seem gone.

I don’t understand why? I tried the original Pi power supply and even with this i had communication losses ....

I hope the board is tough enough...

jay_s_uk

the board was originally designed with the pi 3 in mind and not the pi 4. The pi 4 requires more power compared to the pi 3 so its really advised to use a separate psu to power the pi

NeueKlasse

@jay_s_uk

The problem is it doesn’t work with my board ..
and i don’t understand why ..

Maybe it’s possible to cut the int5V and use the external 15W Powersupply together mit the 5Vext connector??

NeueKlasse

So.. how much Current is the Int 5V capable of?
and how much current does the Board in Ext 5V Mode work with? The documentation lacks of information there

The Communication Errors are completely gone with the Int 5V Jumper! 10Hour Print without Problems... the Toolchange is slow (directly switch from t1 to t0 without waiting for temps (pre and past toolchange script empty) gcode file without additional commands... the change can take from 3s to 20s

dc42

@NeueKlasse said in Duet 3 6HC Rev. 1.01 Faulty:

So.. how much Current is the Int 5V capable of?
and how much current does the Board in Ext 5V Mode work with? The documentation lacks of information there

https://duet3d.dozuki.com/Wiki/Duet_3_Mainboard_6HC_Hardware_Overview#Section_Operating_limits

@NeueKlasse said in Duet 3 6HC Rev. 1.01 Faulty:

the Toolchange is slow (directly switch from t1 to t0 without waiting for temps (pre and past toolchange script empty) gcode file without additional commands... the change can take from 3s to 20s

Have you anything in the tfree files?

NeueKlasse

@dc42
Oh Thanks, 5V3A mentioned... and what about external 5V power, is it the same thing there?

dc42

If you are applying external power, then the current taken is the quoted 200-300mA from the Duet itself, plus anything you have connected to the IOx, LED strip or Laser/VFD connectors that draws power from them.

NeueKlasse

@dc42

But what Amperage can the Duet handle via Ext 5V?

My target is to get the int5V overload risks
out of the way while powering the SBC from the Duet otherwise it doesn’t work for me (communication problems)

dc42

@NeueKlasse said in Duet 3 6HC Rev. 1.01 Faulty:

I remembered that i had connected the Duet and Pi4 in that way that the Duet supplies the Pi 5V before i put everything in the printer...
Then i wanted a extra Powersupply to not overload the Duet...

Today i disconnected the Pi Powersupply and added the needed Jumper... et voila... problems seem gone.

I don’t understand why? I tried the original Pi power supply and even with this i had communication losses ....

I hope the board is tough enough...

Somehow there was differential noise between the ground of the Pi and the Duet ground when the bed heater was operating. However, from your wiring photos, I can't see the source of that, because you have connected the power wires to the bed heater mosfet directly to the PSU.

I have a few other suggestions to try, with the 5V_INT jumper removed and the Pi powered from the buck converter:

Temporarily disconnect both power input wires to mosfet switch. Then check with a multimeter that there is no connection between the bed heater output wires and the negative output terminals of the PSU. Assuming there is none, reconnect those wires and proceed to #2.
Currently you have the red and black input wires to the buck regulator connected directly to the middle pair of output terminals of the PSU. Try removing those two spade terminals from those PSU outputs and connect them instead to the same outputs as the Duet, leaving the middle output terminals free. It's possible to fit two spade terminals under one screw, but you may have to bend one of the spade terminals a little.
If #2 doesn't help, connect those two buck regulator input wires directly to the VIN terminals of the Duet instead, along with the two power wires.

NeueKlasse

@dc42

i'll test these Variations asap! Thank you!

NeueKlasse

@dc42
I tested everything, the 3 Output terminals V+ are bridged and the 3 Terminal V- are Bridged on the PSU. it had no Impact when i Wire the 24V lines from the Vin Terminals to the Powersupply or from the PSU Terminals.
also no change when the Mosfets are disconnected from the PSU
(5V->SBC not jumpered)

While the Printer was Upsidedown the Opportunity was there to Change the noisy
PSU Fan to an Noctua 60mm PWM fan (Wired to out4 of the Board and jumper changed to 12V)

after Reassembly and Disconnection from the PI Powersupply i reapplied the
5V->SBC Jumper. First all seems ok, last time the Print ran 10hours!!!! without any problem....
but now after Start of an Print i get connection losses every... 10-20seconds! again.

Interesting thing... after 30min Printing and several Connection losses the Hotend cooled down due to the Connection losses but the Printer was still on its path! Gcode was Driven without an problem!...
Why does the printer don't stop in that case?

is the Pi4 tested while Board development? or only with the Pi3?
i have no Pi3 here to verify..

i am about to setup a TP-Link Nano and get rid of the Pi4...

A Former User

This post is deleted!

NeueKlasse

Short Update, maybe helpful for somebody.
I had still Communication Problems, doesn't matter what i tried,
now my Configuration looks a bit different and i'm running very happy

Duet 3 MB6HC in Standalone Mode to TP Link Nano, and a
Raspberry Pi 4 with Octoprint to Provide a Camera URL, Octoprint doesn't need to Control anything... so the Stream isn't laggy by any means

All Together in the Original Chiron Sheet Metal Cabinet/Box, + 1 PWM 60mm Noctua Fan, no flimsy extra space needed....
since that i've run 3-4 over 22Hour Prints without any Issue.