Apologies for taking so long to reply @Phaedrux, the end of 2020 was kind of a shit show for me. So with a few months left on my warranty. I have some higher resolution photos of everything that was or seemed off. Let me know if you want photos of anything specific, I couldn't confirm any issue with the diodes, because my board actually looks like the diodes are different... I think.

Whole Board Front:

Whole Board Back:

The header that came loose:

The dodgy looking soldering I noticed on the back of the board:

Hi @Veti Thanks for the endstop recommendation, and yes I intend to upgrade the extruder later on. Also I put the PCBs for mounting the Magballs to the carriage, on the front in order to avoid pushing the belt too far out of alignment, it seemed to be adding unnecessary friction, I'm planning to print better belt retainers once its running and then mount the PCBs better to the new tensioners.

I've finally had time to check all the wires so on to the results.... I've done resistance/continuity checks up and down the channels. Connectivity is good at all points and across all the joints... I even checked it across the stepper motors themselves, and did round trip checks with everything plugged into the stepper motor from the board side of the cables. All checked out. I've got a relatively cheap oscilloscope so I can probably check the output I knew how to wire it up to the Stepper motor connectors on the board. It would be good to rule out any issues with the board hardware. Since I had some kind of solder join quality issue with one of the other connectors, my IO_2, header detached clean from the board during wiring, all the other IO_* headers seem ok, but it does make me consider the board itself may be the cause.

Any recommended resources/links for how to check that the stepper motor output from the board is good?

So I've got a retrofitted AnyCubic Kossel Linear Plus, its just your run of the mill delta frame, steppers etc. I've pretty much got this thing in ideal state to troubleshoot the config... but I have no idea what to adjust based on the behaviour I see... This is my 3rd printer, and the first where I've really done anything close to building it myself, my previous two were run stock as bought.

So hardware wise it's a Duet 3 main board (the production revision), with a RPi3 host connected via the 20 pin cable, and also the 7 inch Panel Due for status information and such without having to fire up the web interface. I've also swapped the stock effector for a duet smart effector, and replaced the arms with Magballs & Rods from Hayden. As it currently stands, the only original parts are the frame, linear rails, bed heater, and the stepper motors.

It took a bit to sort out the wiring and theres still a few temporary breakout boards and such which I'm using to keep things reconfigurable until I get it all working. I'm pretty sure the wiring is ok, but I'm obviously not certain. Once the printer is working I plan to drill out the case and setup more permanent wiring, with connector sockets, shielded twisted pair wire for the stepper motors, and so on. If my problems turn out to be a wiring thing I'm not worried because I haven't set any of the wiring in stone yet.

So... Pics, video, and my current config.json follow... I think the config is in the ballpark of correct since I followed previous threads in this forum regarding the AnyCubic delta models, but again... I have no idea what to adjust based on this behaviour. I hit "Delta Calibration (G32)" and it starts doing what you see in the video.

config.json

https://imgur.com/a/4MKPzI5 - Videos

The belts do flow freely, with less force than the stepper motor is pushing back. I can get them started with a little more force than they used to need, but when the printer is making that shake/vibrate noise, I cant push the rail carriage, its like the stepper motor is "holding in place" with more force than I need to get the carriage moving... so I don't think its too tight/needs more lubrication. But I'm happy to be corrected ... thats why I'm posting this.