Hi I want to bypass an Inrush Limiter Thermistor. It is between the rectifying bridge and the smoothing capacitors, I was considering using a solid state device.
After the rectifer current only flows in one direction, so I considered using a mosfet because I have good quality ones lying around. However, would the charged capacitors maintain a reverse voltage on the mosfet when the AC wave goes down? In such case would a mosfet be unsuitable?
Then my next choice would be an SCR/Triac right?
p.s.: voltage is 12V to 15V AC, 4A max, I have 6600uF of capacitance.
Best Answer
The rectifier doesn't enforce a decrease in voltage, only that the output is greater than or equal to the input minus a diode drop or two. Otherwise the caps wouldn't do much except possibly blow up from excessive ripple.
If you leave the FET on, then it's basically a low-value resistor that can pass current in both directions. Might as well call it a wire unless you're really concerned about it. To keep it on, you simply need a gate-source voltage that satisfies the threshold.
When the FET is off, it theoretically works like an open switch to both polarities...except for the parasitic diode in parallel that's impossible to get rid of. So you'll orient the FET so that the diode will block the normal flow of current and choose the channel type to control it conveniently given that orientation and which rail it's on.
Seems to me like it ought to work. I assume you're going to start with it off, to let the thermistor do its job of limiting inrush, then turn it on and leave it on?
Thinking through several configurations in my head, I think you'll be stuck with the gate referenced to the line side of the thermistor instead of the load side, unless you can boost it beyond the normal supply rails. But if you put a resistor between gate and source to hold it off and then pull it to the opposite supply rail to turn it on, you should be good. As long as the controlling circuit can handle the full, unfiltered rectifier while the main caps charge.