Can a high current switching load damage a bench power supply?

Question

I am powering an electronic speed controller (ESC) driving a BLDC from a bench supply capable of supplying 40V at 128A. The maximum load the ESC presents is 42A.

The wires from the supply to my setup are about 3 feet long (long!), and thus I added a capacitor bank at the ESC end to protect the electronics on the ESC, since the inductance of the wires will create large voltage spike at the ESC. I use the supply's sense lines to ensure I regulate the voltage at the ESC end.

The supply started to misbehave recently, after months of use testing my ESC design. Initially, the ammeter on the supply stopped working. Now, I believe the supply is having difficulty regulating when the switching current load is high.

Is it possible that voltage spikes (maybe capacitor bank is undersized - need to confirm this) at the ESC supply input damaged the sense lines and control circuitry in the bench supply over time? Is there typically protection circuity to limit voltage excursions on these lines?

I found the the manual for this old Agilent 6684A supply, but at first glance, I didn't find these types of specs --I'll look more carefully. Are there typically specs provided on what sense lines can take, or how fast the supply responds to step changes to the sense line voltages, etc?

I have another supply to use, but I don't want to kill it as well overtime if I could be potentially damaging it testing the ESCs.

Answer 1

Is it possible that voltage spikes (maybe capacitor bank is undersized

• need to confirm this) at the ESC supply input damaged the sense lines and control circuitry in the bench supply over time?

Yes, it is possible. It would be safe to assume that the supply does not have input protection for overvoltage. It would be best to use diodes or some other form of input protection like this circuit:

Source: http://kaktuscircuits.blogspot.com/2014/07/reverse-polarity-and-overvoltage.html

A better way would be to monitor the voltage spikes with an oscillocope at the input of the supply to see what your up against. Very short spikes could be taken care of with extra capacitance, or minimizing cable inductance and a diode. Very long spikes might need something like the circuit above. At any rate, it's probably not good to present voltage conditions beyond the rating of the supply to it's inputs.

Answer 2

My suspicion is similar to the answer by @schnedan. When the motor speed is reduced, the ESC will attempt to transfer energy from the rotating mass to the DC supply. In addition, the inductance of the motor will cause the AC current to the BLDC motor to lag the voltage. The resulting reactive current is supplied as ripple current added to the DC source current. That must be supplied by the DC capacitor. The capacitor must be large enough to supply that without too much DC ripple voltage.

Note that the DC supply may have difficulty with the initial charge current of the capacitor. Some means of dealing with that must be provided.

In general, it would be better to supply the ESC from an unregulated DC supply. The ESC should be able to regulate its own output voltage with input voltage that fluctuates to the same extent that mains voltage fluctuates. However it would be useful to have electronic overcorrect protection supplied in the DC supply.