Hi everyone,

I am using this MOSFET IRF5305 and I'm afraid I don't quite understand how this device works. I am following an application note by Microchip called AN857:
ww1.microchip.com/downloads/en/appnotes/00857a.pdf

The other MOSFET in the circuit IRLI3705 I understand. I give the gate a voltage (12V) and the drain to source conducts current. When I don't give it a voltage then it does not.

The IRF5305 MOSFET however, despite the fact that I give the gate voltage or not, it seems to conduct. Now I notice that the voltage drops down to 11.62 volts when I do the switch between applying 12V to the gate, and applying ground to the gate. Can you clarify this to me. I do not understand this MOSFET at all.

 

Its a P-fet it requires a negative voltage between Gate and Source to conduct, its the opposite to the N-fet.

 

You are using P-channel and N-channel MOSFETs, they operate differently from each other, notably, creating a channel in the P-MOSFET.

look at your threshold voltages...

operation of MOSFETs
http://en.wikipedia.org/wiki/MOSFET#Modes_of_operation

 

Those two are complementary transistors.

An N-MOSFET uses positive drain-to-source and gate-to-source voltages for normal operation.

A P-MOSFET use negative drain-to-source and gate-to-source voltages for normal operation.

 

Thank-you for the info everyone. I'm still very new to all of this so I apologize for any trivial questions I have. Now looking at this circuit, and given your information about the p-channel mosfet and the need for a negative voltage between Gate and Source to conduct, at what point will there ever be a negative voltage between the gate and source? Looking at the circuit below:

http://i.imgur.com/PQak3lJ.jpg

The TC4469 is going to be outputting 12V or 0V based on the input which will then be inputted into the MOSFET. Now this will give a positive potential between gate and source, when the input to the gate is 12V. When will there ever be a negative potential between gate and source seeing as the source is always going to be low on the IRF5305, regardless of whether its pair the IRLI3705 is on or off? The TC4469 is never going to output a negative voltage (-12V) into the gate of the IRF5305.

Once again I apologize if I'm not seeing something obvious, thanks again for the help.

 

The source of the p-channel MOSFETs is the upper pin that is connected to +12V.

The voltage at the gate of the P-channel MOSFET is either 0V or 12V with respect to ground, therefore the voltage from gate to source is either 0V or -12V.

+12V is the reference in that case. Vgs is the voltage from gate to source.

 

Remember that all voltage polarities are relative, not absolute. So even if you only have a plus 12V supply (with respect to the terminal called "common") you can "stand on your head" and say that the plus voltage is common and the previous common terminal is then called -12V. Thus the relative voltage between two terminals of a device can be negative or positive, depending upon how they are connected.

 

Hi everyone,

Thanks for all the replies. I finally realized what my problem was thanks to praondevou's reply. I had read the schematic wrong, and accidentally connected my p-channel mosfet (IRF5305) incorrectly and had my drain and source confused. I had accidentally connected my drain to 12V and had the source connecting to the other mosfet. Thanks for all your replies. Appreciate the help.

I'm now having a problem with the output on the entire circuit. I am getting a waveform on the output (it is supposed to be a sinusoid). Check out the attached images, as you can see it's a weird shape and not quite a sinusoid (look at the top most waveform, that's the difference between the two phases). The two waveforms below were the waveforms from the respective phases. It also seems to be pulling quite a bit of current off the power supply (I only had 1k resistors across the phases a-b, a-c, b-c). I am thinking that my output sequences from the PIC may be wrong. Please let me know if you can see anything wrong. Thanks for all the replies.

 

From the circuit you posted you cannot get a sine wave because there is no load. You would need a LC at the output. Then you get a sine wvae with the right modulation.

Where exactly did you measure this wave form?

 

I connected resistors 1k from A-B, A-C and B-C. I then connected one probe to output A and another probe to output B. I then used the oscilloscope's math function and measured the difference between A-B, to get the resulting waveform.

 

I connected resistors 1k from A-B, A-C and B-C. I then connected one probe to output A and another probe to output B. I then used the oscilloscope's math function and measured the difference between A-B, to get the resulting waveform.

As I said, the signal over a resistor cannot be a sine wave in your circuit.

Connect a resistor from A to ground, B to ground, and C to ground. Start by checking if each of them is switched correctly and in the correct sequence. I would start with a 50% duty cycle on one leg and let the others off.
Verify each leg step by step.