I am going to use a N channel JFET as a normally-ON switch. I am wondering if it would cause any problem in the long term reliability having the PN junction forward-biased. In other words, would it be considered a reliable design if I let the gate voltage be greater than the source voltage for some time (*) during operation as long as the current flowing into the gate pin is limited much less than the absolute maximum of the forward gate current, Igf? Igf < 50mA is specified for MMBFJ201, the one I am trying to use.
(*: Roughly 2 seconds of Vgs>0 for every 30 minutes of operation)
Schematic of my circuit is:
Best Answer
This won't cause a problem. You've addressed the Gate current issue with a current limit resistor (R1) and this will not be an issue. However, you do say that you want it as a normally on switch yet you also say Vgs = 0 for every 30 minutes of operation - I can't see in your circuit how the JFET will ever turn-off. It looks like it will be permanently on (and therefore activating M1 and therefore driving current into R4 (load?). Maybe you mean it the JFET will turn-off for 2 seconds every 30 minutes - this rings true because of the circuit driving it when U1's output goes low?
In summary, the JFET will survive but maybe you can clarify your intention for the circuit as it appears to contradict your words. I'm also assuming that Vs is not going to be greater than (say) 35V because the maximum voltage rating on the JFET is 40V. I guess not because the mosfet is only rated at 30V!!