rjent wrote:BuckeyeDennis wrote:rjent wrote:Update!
PP running again! Took me an hour to change out the power unit. What a great design! I am more impressed with this machine than I was before I got to see her innards .....
Anyway got the PU yesterday afternoon FedEx and put it in this AM. Everything is running great so far.
Dick, are you running the PP on 110 or 220V?
110V. I honestly don't know why you would want to (or need to) run it on 220. The motor has so much power you slip the belts (I have done it on purpose just to see) before you stall the motor. I have never felt I needed more power ....
I was just curious about the root cause of your failure. If running on 220, that would be different than most, and might be a clue.
Several years ago, I designed a 10 hp servomotor drive that could be jumpered to run on either 240V or 480V 3-phase. All went well in testing. But once fielded, we had a couple of failures, with symptoms similar to what you reported. After the initial panic, we got the failed boards back for post-mortem. The problem turned out to be that the PCB layout tech was thinking in only two dimensions, when he needed to be thinking in three. The track clearances on the board were as they should be. What he forgot was that the electrolytic bus capacitors mounted on the board had some exposed conductor extending radially outward from the pins. So he inadvertently ran some ground-potential track right underneath that exposed high-voltage conductor on the cap, with only a small air gap separating it from a rectified 650VDC. Or even up to almost 1000VDC, under regenerative braking conditions.
So what happened was that under some combination of peak voltage and perhaps high humidity, that small air gap broke down and arced. Which ionized the air, greatly increasing it's conductivity. And then the entire DC bus (with about the same stored energy as a defibrillator) discharged through that air gap, releasing a huge amount of energy in just milliseconds. The thermal shock from the discharge blew the offending capacitor right off the circuit board, and vaporized anything in its path. IIRC, I had reviewed the PCB layout, and missed the problem as well.
Once the problem was identified, the solution was simple. In the short term, we had some FR4 circuit-board material cut to shape as insulators, and mounted them between the caps and the PCB. And then the next time we revised the PCB, we simply moved the track out from underneath the overhanging conductor.
So you see, it's all just physics. No voodoo involved.
