In another discord moment, someone was asking about the difference between electrolytic capacitors and multi-layer ceramic capacitors. That, plus some desire to re-rate the moteus, inspired me to do another sweep and measure the DC bus voltage ripple for various power levels. I captured this plot with a 24V power supply, with a 5008 motor with 0.061 ohm of winding resistance or so, and each current being applied for 300ms. The voltage ripple is peak to peak measured at the power connector.
At the peak power I tested of around 740W, the phase current was 114A. At that level, the motor coil was getting hot much faster than the FETs on the moteus, which implies I need even more capacitance to take advantage of the full current capability of the FETs on the board. Also, the voltage ripple at the peak power I tested was higher than some applications could support.
A Modicum of Fun 
Voltage ripple will be determined by input impedance (inductance and resistance of the battery/PSU and wires going to the drive) as well as local capacitance, so it might be helpful to give some of those details as well, if you can measure/estimate them.
LikeLiked by 1 person
I could estimate it, but I don’t expect such measurements to be all that useful. I’ve measured similar ripple results with a range of lab supplies and a range of wiring. Clearly, the results could be a lot worse if the supply impedance was really bad, but it isn’t going to be much better without more local decoupling.
LikeLiked by 1 person