P Mosfet Gate Threshold for OFF state

Hi all, newbie here, trying to decode datasheets.

**broken link removed**
Datasheet: http://datasheet.octopart.com/FQA36P15-Fairchild-datasheet-125386.pdf

Am I right in thinking I can get away with a gate-source voltage of -2.5V ish and the mosfet still be fully off? I'm looking at the transfer characteristics graph.
Cheers
Nick

Sorry to DP but there doesn't seem to be an edit button to modify my original post any more so I can add more info.

In the specs at the top I see the gate-source threshold voltage given to be Min: -2V and Max: -4V. Am I right in assuming this is a tolerance and represents the range at which the fet will begin to conduct? So from unit to unit and various temperatures this could be anywhere between -2V and -4V? In which case I should be looking at keeping the gate at -2V at the very most to keep the mosfet off?

Hi,

Two informations you can find in the datasheet:
* gate threshold voltage. It is -2V ...-4V. Below that voltage (in direction to 0V) the FET is considered to be OFF.
With a higher negative voltage (in direction to -10V ) the MOSFET is ON = conductive
* Zero gate voltage drain current.: here you see that the max current with 0V gate voltage is 10uA (100uA)

For switching applications the gate voltage should be near 0V / -10V.

Klaus

Thanks, in the test project I'm going to build which has a 12V supply the gate voltages will be supplied by a driver, but the driver's VIN will be via a diode (to prevent stress on its bypass caps) so I'm going to get a diode drop of about 0.6-0.7V on the driver output. Meaning the ON stage should be fine, as the driver will be acting as a 7 ohm resistor to ground giving me -12V at the gate. In the OFF phase, the driver will be outputting about 11.3V giving me a gate voltage of -0.7V which I'm hoping will be enough to switch the mosfet off significantly enough that any drain current is negligible. I guess I could drop the margin further by using a schokkty as well. Do you think that's an acceptable margin of operation for this mosfet?

Hi,

(to prevent stress on its bypass caps)

Click to expand...

Usually not necessary.
If you want to reduce ripple current, then a (10 Ohms?) resistor is most likely more effective than a diode, without the 0.7V voltage drop.

If you need more assistance, then show us your circuit.

Klaus

The graphs in a datasheet are for a "typical" device that you cannot buy. You gamble that you might get one with spec's at "typical" or better but you might lose and get one that does not work in your circuit that is designed for a typical one. If you design for the printed range of spec's then all of your circuits will work.

I do not think that the range of spec's is random. I think that a production run could produce mostly worse than "typical" ones but they still pass and another production run could produce mostly better ones. Then you get whatever they have.

KlausST said:

If you need more assistance, then show us your circuit.

Click to expand...

I haven't designed a circuit yet. I'm trying to get more information about things before I go and do that. This includes asking elementary questions, watching videos and attempting to verify if what I am assuming is correct or incorrect. Are you saying here that assistance won't be offered on elementary questions unless the asker has actually designed a circuit?

Hi,

Are you saying here that assistance won't be offered on elementary questions unless the asker has actually designed a circuit?

Click to expand...

No.
But you tak about a circuit with diode voltage drop and so on... you have a circuit in mind..
Such things are hard to describe just with words. Here a simple sketch says more than hundred words.

Klaus

VT is a specific attribute measured in a specific way.
The conditions may or may not be described, or well.
MOSFETs do conduct below VT ("subthreshold") and
your circuit might or might not care about enhanced
leakage when you're between VT and Vgs=0. There
is a gap between where IDSS0 is spec'd (Vgs=0) and
where VT sits. You're unlikely to get a subthreshold
slope value from a power device datasheet, to know
how much leakage you'll get at what distance below
the VT you got, and of course you can't know that
either before you take delivery.

If you want the IDSS0 specified you need to meet
the conditions under which it is specified.

- - - Updated - - -

NicNewb said:

Are you saying here that assistance won't be offered on elementary questions unless the asker has actually designed a circuit?

Click to expand...

You want specific advice without specific context. Many
people will not waste time in unbounded speculation if
they don't think they know what you are really asking
for.