Chopping AC at circa 1kHz (PWM) FET's?

Thread Starter


Joined Sep 4, 2010
Hi folks,
I have an idea that I would like to test ....

If it works on the bench at low voltage and current I may buy some commercial gear and build a panel but that would be out of the scope of this board due to the voltage

Back to safe voltages ... I know capacitors can be added to an induction motor, three phases in my case, to form a self exiting generator and I have an old 48v / phase motor I can experiment with.
The caps will be in parallel with each winding and the motor will need to rotate at synchroness speed + slip.

Now conventional wisdom suggests that the caps should be sized to give the correct voltage within a small RPM and power window.

But that is of course far to simple to be interesting so I wounded if I could add oversized caps and then modulate them with a PWM signal.
I suppose I could use phase angle and therefore triacs but that would involve 3 separate 0 crossing detectors that would either have to be trailing edge or employ current detection, because of the load charismatics ... I think

So my question is two-fold.

1. I would like opinions as to weather I am likely to be able to effectivly control the excitation of the machine and therefore the output voltage at various loads and speeds.
(The output will be rectified anyway so Frq isn't an issue and it load will be resistive)

2. Can I use a MosFet (N) to do the switching buy placing it between the DC legs of a bridge rectifier that has its AC legs in series with a Cap.
In that configuration it wouldn't it always be low side, provided that the gate drive circuit could float?

Fig1: Standard generator config.
Fig2: Modified config with modulated Caps
Fig3: Individual switch circuit
Values and actual gate drive circuit omitted for clarity :-
Will be BJT totem-pole and associated bits.

Any and all comments will be much apriciated