Shorted Mosfet due to measurement

[caspian]

Dear technicians,

When I measure SMD Mosfets on Laptop mainboards, I face shorted Mosfets sometimes. But I think it might be possible that I made some of them shorted by my measurements. Specifically, I measure resistance between pins of Mosfet while power is being applied to the mosfet and the mainboard. Is that possible this measurement make mosfet shorted?

For example: I measured 91 ohms between gate and source of a Mosfet while the mainboard was powered on. The next time I powered on that mainboard, I measured 0 ohm between gate and source of the same Mosfet. So, I think I made it shorted.

note: I measure Mosfets using my multimeter either in Diode mode or in Resistance mode.

thanks

[petehall347]

Re: Shorted Mosfet due to measurement

resistance tests should be done with power off .

[caspian]

Re: Shorted Mosfet due to measurement

at the time of power off, the mosfet showed high resistance between source and gate. But at the time of power on, the mosfet showed about 91 ohms between source and gate. So I needed resistance at the time of power on to check the mosfet.
Is this method wrong?
Is 91 ohms between source and gate a sign of being shorted when powered on? or is it normal?

[petehall347]

Re: Shorted Mosfet due to measurement

measure volts with power on .

[caspian]

Re: Shorted Mosfet due to measurement

It was a P-channel Mosfet PQ25 model: AO4433
on laptop mainboard: Sony MBX204
described in Page 54 of the attached schematics.
===================================
When powered on:
on source: 19.58v
on gate: 19.31v
on drain: first 19v then goes to 16v after 1 second.
between gate and source: 91 Ohms , after a number of shutdowns/restarts went to 0 ohm
between gate and drain: I forgot to measure
between source and drain: 0 Ohms
====================================
Resistance when powered off:
between gate and source: 110 kOhms
between gate and drain: 55 kOhms
between source and drain: 50 kOhms

Diode mode measurement when powered off:
between gate and source: 2.8 v
between gate and drain: 2.9 v
between source and drain: 0L one way and 0.5v the other way
====================================
no other voltage rail exists on the mainboard except the 19v rail.

what do you think about this strange Mosfet?
has my measurement made this mosfet shorted between gate and source?

[diif]

Re: Shorted Mosfet due to measurement

Yes checking resistance with power on will short components. It's just the same as bridging the component with piece of wire.

[petehall347]

Re: Shorted Mosfet due to measurement

only real test you can do with a meter is see if shorted... or if turns on and off there is a good chance its good if it does both things .
saying that i have always found bad ones shorted or blown to pieces ..

[eccerr0r]

Re: Shorted Mosfet due to measurement

You can't measure resistance when there's another circuit powering it. This is not to mention that when measuring in-circuit there are parallel paths, and that parallel path will cause your measurements to go wonky. Just think of the second power source when powered up, that's another short that power can go through, *really* throwing off your measurement.

BTW 19V from gate to source is nearing or exceeds the voltage limit of most MOSFETs. MOSFETs are typically clamped at 15V or so, some much less, to prevent permanent gate damage. Then again I'm not exactly sure how you're measuring voltage there. Perhaps an oscilloscope trace of the pins would tell you more information than a multimeter - if you have one.

[momaka]

Re: Shorted Mosfet due to measurement

...
Specifically, I measure resistance between pins of Mosfet while power is being applied to the mosfet and the mainboard.
...
Is this method wrong?

Yes, this method wrong. NEVER measure resistance with the power turned ON. (And sorry to say it directly like that. I don't mean to offend you or anyone else.)

Measuring resistance with the power turned On can not only damage your board but also can be very dangerous if you are working with line voltages (i.e. relatively high voltages and currents), especially if you are using a cheaper multimeter (they can blow up and easily cause fire).

Please be careful.

To check MOSFETs correctly, use resistance or diode check ONLY with the power OFF *and* all batteries removed from the device. In addition to that, when checking resistance in devices with high-voltage caps, first always make sure those capacitors are discharged. Again, this is to prevent you from blowing more parts up and for your own safety, too.

[caspian]

Re: Shorted Mosfet due to measurement

Thanks for the replies. I need help of electronics technicians sometimes. I hope I help you back in some laptop repair.
there is about 0.3V between gate and source.
there is 19.31v on gate and 19.58v on source. Thus, this p-channel Mosfet should become off and shows infinite resistance from source to drain. But It does not and passes voltage from source to drain. Do you think it is faulty although it shows good Diode mode measurement when powered off?

[caspian]

Re: Shorted Mosfet due to measurement

When I power a laptop off, I have to start measurement as soon as possible. How long do laptop capacitors hold the power after powering off?

note: Laptop mainboards contain mainly SMD ceramic capacitors and also a few SMD tantalum capacitors too.

[eccerr0r]

Re: Shorted Mosfet due to measurement

You may have other circuits causing a voltage to be sent. At 0.3V it should be subthreshold and should be mostly 'off', maybe some leakage depending on the transistor. The drain should be mostly floating and follow whatever voltage other devices are pulling it.

Sine it's an apparently high side p-channel, what is it powering, if it's shorted, it should be constantly powering whatever it's controlling -- is this the case?

If it's not shorted out of circuit, the mosfet isn't shorted. You should be able to tell leakage through it out of circuit too, S-D should be virtually infinite resistance when G is connected to S. G should also be infinite resistance to the other pins too when out of circuit - you can't measure this in circuit as the parallel circuit elements tend to dominate the measurement.

Why would you need to start measurement soon after poweroff? Heat dependency? Most capacitors discharge fairly quickly, just takes seconds or so as the LEDs turn off fairly quickly ... ???

[caspian]

Re: Shorted Mosfet due to measurement

During laptop testing, I have to repeat the process of "restarting and measurement" in a fast way, because lots of points may be required to measure on the mainboard. Thus, I can not wait after a restart for more than a few seconds.
I finally figured out PQ25 is probably not faulty and some other component of the mainboard pushes the 19v on the drain of PQ25 down to 16v.
This was the most strange laptop I ever faced. I want to replace the entire mainboard because it took too long to repair.
A very cute discovery on this mainboard: when I connect the battery to the laptop while the AC adapter is connected, I measure a variable voltage on the drain of PQ25 that periodically changes between 12v and 16v similar to a sine wave.