Ok, sorry to come back to this post so late but I had other issues and I could not test further...
Now I managed to free other 3 heater ports on the Duex, so I got rid of the buck converter as I could connect all 4 servos to 4 pwm ports on the Duet.
Nothing works anymore.
I tried also this configuration but nothing... The only visible thing that happens when I invoke a M42 or M280 command is that the corresponding heater led lights up (and remains up). So, it could be that the ports do not deliver the 5v? Before, using only the signal pin they worked with an external 5v supply...
I solved it, didnt read enough... The Duex5 inverts the signal depending on how you name the pin in firmware...
The spam bot wont let me change the first post anymore I guess
"Sometimes a pin has multiple names.
These four all refer to the heater3 pin, either on the expansion connector (where it is pin #8 of 50) or on the heater output terminal block of a DueX2 or DueX5. However, there is a difference. The DueX boards buffer and invert the signal. Therefore, when you use one of the forms prefixed with duex the firmware knows it has to invert the signal, but it doesn't invert it if you use one of the other forms. In contrast, when you referred to logical pin 3 in previous versions of RRF, the firmware always inverted the signal - so if you were driving a servo or BLTouch from this pin you would have needed to use the I1 parameter in your M280 command."
Thanks for the quick response. I made a couple of test leads and measured a good 3.3V between the 3.3V, TWD, and gnd. However when I measured TWC, twice, it registered only 136mV. The Duex5 has been swapped and although much longer than the provided ribbon cable, I had an old 50 pin SCSI cable which I tried with the same negative result. So I assume that would point to a problem with the Duet-wifi card.