Zeke Yewdall <[email protected]> wrote:
> Yeah, something seems wrong.
> Perhaps it's that the current isn't actually fluctuating all the way
> from 1000A to 0A at 10kHz... because of the inductive effects of the
> motor. It's actually a fairly large DC current, with a superimposed
> 10kHz AC current... and perhaps the AC portion is relatively small
> compared to the DC portion.
No, the motor inductance won't do anything to keep battery current
constant. The inductance of the motor will keep the current through
the motor constant, but that is different from the current through the
battery. When the controller is in its off-time, the FET is turned
off, so the motor current passes through the diode, skipping the
battery altogether. Treating the controller as a simple Buck converter
and treating the batteries as an ideal voltage source, the batteries
would experience alternating 0A and 1000A if the motor was using
1000A. Of course, that's not what happens in real life.
The battery current will be mainly DC with a smaller AC component; it
won't have the full motor current as its peak to peak. This isn't due
to the inductance of the motor; it's because the inductance of the
battery cable and the controller's input capacitance form a low-pass
LC filter. Depending on the cable inductance and controller
capacitance, this could produce almost smooth current from the battery
or extremely choppy current from the battery. Even with a tiny cable
inductance, though, it'll be nowhere near a square wave.
However, with 2/0 wire at 15kHz, the skin depth is 23 mils. That means
that the AC resistance at 15kHz is 8 times the DC resistance of the
wire! So any ripple current will have 8 times more resistive losses
than the DC current. As long as the inductance of the battery cable
and the controller input capacitance form a good enough LC filter,
this won't be a problem.
For subscription options, see