I am not an expert of motor control and I am hoping that someone with some experience in this area can give me a hand. I have a custom motor controller that drives a dc brushless motor for an underwater vehicle. I am using a 2 quadrant pwm switching scheme. The battery is connected to the motor driver through a diode (in another subsystem) that protects the circuit from reverse current flow.
The motor works fine except that as the RPM rises above I start to see a voltage waveform across the diode. This voltage causes the bus voltage to other subsystem to rise up to 10 Volts, which is a problem. If there were no series diode in the line, the battery charges during these cycles and keeps the bus voltage nice and steady. Unfortunately, this series diode is in another subsystem and has to be there for other reasons.
I believe that a lot of motor controllers put a very large capacitor across the supply to smooth this. I can't do this because this system has to operate in an underwater environment under high pressure. Electrolytics and other large capacitors cannot tolerate pressure.
I don't want to put a zener in to clamp the voltage because it would have to dissipate a lot of power. If I put a series diode near the motor, this would block the voltage from appearing on the bus of the other systems. However, I would be dissipating a lot of power in the diode since the motor has peaks of 10 amps. The system is battery powered and power consumption is critical.
Is there an easy solution to this problem? Can someone point in the right direction? Would a different modulation scheme help?
Thanks, Bob