Newbie Testing P Channel Mosfet

[orientalsniper]

Hello all, I've read across different sites (badcaps included)

To test a P-Channel mosfet, I have concluded the following:

1) - plug to gate, + plug to source = open circuit reading
2) + plug to drain, - plug to source = open circuit reading
3) + plug to gate, - plug to source = open circuit reading
4) + plug to gate, - plug to source = short circuit reading

Now, the mosfet I'm testing is this

I've performed the test to 3 same mosfet in a laptop motherboard and I've got different measurements, they are still soldered to the board, would that make any difference, or did I misunderstand the basic tests?

[capkid]

Re: Newbie Testing P Channel Mosfet

I'd de-solder them if your results are inconclusive.

[eccerr0r]

Re: Newbie Testing P Channel Mosfet

Can't test these in circuit, there tends to be enough circuit stuff that will discharge the gate before you can make a good reading.

To test a P-channel device basically you can do the exact same measurements as an N-channel, except swap the probes.

In circuit there should still be no "shorts" unless the transistor is somehow on (which it shouldn't if it's not powerred up). If you see any shorts when testing in a dead circuit, likely that transistor is fried and something else killed it.

[momaka]

Re: Newbie Testing P Channel Mosfet

In circuit there should still be no "shorts"

Not true.


The lower drive MOSFETs on CPU and GPU VRM may (and usually do) appear shorted. It all depends on the GPU and/or CPU, though. Some GPUs/CPUs have low internal resistance, others not so much. So just beware when checking those. You'll be troubleshooting a problem that is not there.

[dj_ricoh]

Re: Newbie Testing P Channel Mosfet

In circuit there should still be no "shorts".

90% of all mosfet near cpu show short untill you take out the CPU

[orientalsniper]

Re: Newbie Testing P Channel Mosfet

Thanks guys, so the general consensus is to desolder them and test out of circuit.

Also can someone confirm the 4 steps test I'm doing is correct?

[capkid]

Re: Newbie Testing P Channel Mosfet

Here's how I test N-channel mosfets with my Fluke 87. P-Channels will be the opposite:

(1) + Drain, - Source = Open
(2) - Drain, + Source = Reading around 0.5V
(3) + Drain, - Source then quickly touch + to Gate and back to Drain = Voltage climbs steadily as mosfet is charging
(4) Touch Gate and Source simultaneously with your finger(s) while maintaining probes in position to discharge = Open
(5) Repeat step 3: + Drain, - Source then quickly touch + to Gate and back to Drain = Voltage climbs steadily as mosfet is charging

You can also build your own cheap mosfet testers (see attachments).

[redwire]

Re: Newbie Testing P Channel Mosfet

orientalsniper, I think your 4 steps are wrong.
First, what do you consider a "short circuit"? For a proper test, I would use diode test on the multimeter and also ohms. Both give different and good information. Write down your readings.
Second, the gate (on a good mosfet) will always read open circuit (your step 4 shows a problem) if the mosfet is out-of-circuit. Only N-ch mosfets used in car ignition modules have a zener diode connected D-G.
Third, I have seen when testing a mosfet out of circuit, stray static electricity or multimeter can charge the gate, the mosfet turns on and testing says D-S leaky or shorted, when really the mosfet is just turned on. Shorting G-S ensures the mosfet is off for the D-S test.

I do a multimeter test in-circuit and if I get suspicious readings, I pull it out and then test out-of-circuit.

Most mosfets fail with a shorted D-S or G-S so I look for that first.

[eccerr0r]

Re: Newbie Testing P Channel Mosfet

Ah that's true, it does depend on the circuit, the resistance is pretty low for the D-S junction for low voltage high current loads (I'm too used to the main switcher in on-line SMPS). In any case G-S/G-D should not be shorted even in circuit, and yes removing from circuit is needed for accurate measurement.

[orientalsniper]

Re: Newbie Testing P Channel Mosfet

Haha, let me see if I get that.

Basic test for P channel mosfet (assuming N channel would be opposite):
-G +S = Open
-G +D = Open
-D +S = Open

If they are shorted, mosfet is gone gone.

[dj_ricoh]

Re: Newbie Testing P Channel Mosfet

Haha, let me see if I get that.

Basic test for P channel mosfet (assuming N channel would be opposite):
-G +S = Open
-G +D = Open
-D +S = Open

If they are shorted, mosfet is gone gone.

but you have to test it in diode mode to and see the value.

[redwire]

Re: Newbie Testing P Channel Mosfet

I made this table, I think it might help, or add confusion...

[trebo]

Re: Newbie Testing P Channel Mosfet

@redwire
thanks redwire
your post and table really help us to understand the testing method of the mosfet
can you just clarify for us, as to wether we short G-S (as in #8 post) or G-D (as in #12 post) to turn off the mosfet, or is it okay to use either method to turn of mosfet.???
and thanks to all who have contributed to this helpful thread