...and why it is almost certainly not the right answer to the problem of having a logic level signal and a non-logic level FET.
Mark VandeWettering posted an entry concerning the problem of driving an IRF510 MOSFET from a microcontroller running at 3 volts. The basic problem is that the threshold voltage of the IRF510 is specified as something between 2 and 4 volts. With only 3 volts to drive the gate it might not even be possible to turn on the device at all, much less use it to control any significant current.
The best answer to the specific problem that Mark presented was provided by Lee Felsenstein(1) – use a 2N2222 transistor instead of the IRF510.
I proposed, as a hack, a different approach. Since the MOSFET was to be driven by a PWM signal, use a diode and capacitor to boost the voltage to the gate of the IRF510. This is not a good solution to the problem. However, in some contrived cases it can actually work. Instead of being limited to driving the gate with Vcc, it is possible to drive the gate with Vcc + a little less than the min threshold voltage. Assuming the threshold voltage is less than Vcc, a couple of resistors are needed to keep the gate voltage under the min threshold voltage when the IRF510 should be turned off.
In the case of a 3 volt supply and an IRF510 this means that it is possible to drive the gate with something close to 5 volts, which means that you can at least be assured of driving the device with more than the max threshold voltage of 4 volts. Of course, the right answer is to use a different device. But, if the right contrived circumstances ever arises, it is possible to adequately drive the gate of a IRF510 such that the min and max threshold voltages are meet using only a 3 volt PWM signal, a capacitor, diode, and a couple of resistors. Note that I have not actually built the circuit, but LTspice indicates that it works as desired… Edit: I tried it on a breadboard and it does work as expected. Of course, it is still a lousy way to drive a MOSFET.
(1) I’ve never meet Lee Felsenstein, but it turns out that he designed the Sol-20 computer. A science and math teacher at my high school won a Sol-20 kit as a door prize at a computer show (that dates me). I believe I actually soldered a couple of the parts to the boards, but at this point I cannot be certain that I did not just watch the instructor do all of the soldering. It was the first computer I ever used. I learned Basic and Assembly programming on it, staying late after school to use it on as many days as I was allowed. I have “fond” memories of hand assembling code for the processor as the school did not have an assembler for the machine. Not something I’d want to do again, however. So, a very belated thank you, Mr. Felsenstein.