Uh-oh, trouble at t'mill...

What's this then?

This morning's operations started out well. 

Then I started to rerun the previous toolpaths with "stock to leave" unselected. This would remove the remaining 0.5mm axial and radial stock left by the previous operations. I'd considered using a solid carbide, helical cutter for this with a proper finishing toolpath, to get a nice surface finish and remove any trace of the adaptive toolpaths but TBH, there's nothing critical that requires such precision.

Sent the updated toolpaths to the controller and hit the tit. Off it went. 

Then......the Y axis moves started to misbehave. Instead of making contact at each side of the cavity, the tool only moved perhaps half of the distance. WTF?? Something's not right here. Finally, the Y axis packed up entirely. 

The Y axis servo seems to have gone on strike. Here's the red status LED. In fact, the red and yellow LEDS are both flashing (alternately), which seems to indicate either overcurrent or loss of position, according to the manual for the DG4S drive. Not sure which of those (it's not explained in any detail) but I can't imagine why we'd be getting overcurrent, unless there's some sort of mechanical issue on the machine I'm not aware of yet. Didn't sound like a mechanical problem. And nothing's hot in the area of the driver itself.

Time to dig out the USB interface thingy and flash up the "servoconfigurator3" software to see what's happening. Possibly the brushes are failing. Either that or some sort of problem with the encoder feedback or the power electronics. I favour the brushes myself - without jumping the gun of course.

One thing this reminds me of is the lack of feedback from the error output to the Centroid controller. Ideally the machine would have stopped when this error was flagged up. I suspect the shortage (lack) of spare inputs on the Acorn may have been partly to blame. However, I could still figure something out using a relay (SSR) or similar to interface with the limit switch or e-stop circuit. I think this is how the VFD achieves it. 

Obviously the standard length USB cable is too short to reach to the PC, so I'll need to install the s/w onto a laptop and use that to talk to the drive.

Being a brushed machine, I should be able to drive it directly from a bench PSU to see if there's a basic issue with the machine itself. That might require me to remove the belt first but I get the feeling I may be removing the motor shortly anyway....

Servoconfigurator diagnostics:

Well that didn't end well. The USB adaptor must be fucked, as it won't talk to anything. 

Meanwhile, the Y axis seems to be behaving again. That's about as worrying as it gets - intermittent faults have a habit of repeating themselves when you least want them to. I've thrashed the machine with an air cutting program containing lots of Y moves and so far it's behaved perfectly. The only difference is that I've had the cabinet fan enabled. I wonder if overtemp could be the cause?

I should now look at how to tie the error output back to Acorn. Whether I do or not remains to be seen. Depends how impatient I am.

In the meantime, I've ordered another USB programming "stick".....