Am I missing something? It is my understanding that if Vgs is less than Vtn than the n-mos should be in cutoff with no current flowing from the source. I have drain hooked up to 12v in series with a 3k resistor and source to ground. When I send 0V to the gate I get 0mA flowing from the source. As soon as I increase the gate voltage to even .03V I start getting current from the source!?!? The data sheets for the mosfets I am using all have Vtn anywhere from 1-2V. I have tried this with several different logic level n-mos transistors and get the same result. Is this normal?
The threshold voltage is specified as a certain gate voltage for a certain drain current. The current used varies from datasheet to datasheet. For example, for the PMZB200UNE, the following is in the datasheet:
The datasheet says that at somewhere between 0.45V and 0.95V the drain current will reach 0.25mA (assuming room temperature). You can safely assume the drain current will be less than 0.25ma at (say) 0.4V. You are interested in the sub-threshold behavior of the MOSFET. This particular one happens to have curves describing the behavior:
From this you can see that if you need the drain current to be fairly negligible (~1uA - which is comparable to the maximum guaranteed leakage) you should limit the gate voltage to less than about 200mV.
This is a particularly fully-specified part, with maximum Rds(on) guaranteed at 5 different Vgs values- as low as 1.5V at 25°C.
If you are seeing relatively high current with a MOSFET with higher threshold voltage at only 30mV, then maybe your voltage source has some noise on it.