Failing FETs

[Miracletech]

For what reason would a MOSFET drain and source not be a Short circuit, but the MOSFET is blowing up?
Detail: I was testing some of my MOSFET in my stock (FQP50N06). I applied +12v voltage at the drain and -12v voltage at the source. I left the Gate unconnected. To my greatest surprise the MOSFET that I tested( even for RDS(on) and every other thing including body diode ) sparked and blew up!
Is it that I did something wrong? Because I tried this with 3 FETS out of 10 and same thing happened. Or are they FAKES? In my understanding VGS turns on the MOSFET. Or is there something I am missing? Is there some kind of protection that can prevent this from happening? Help would be greatly appreciated.
Maybe these will help.

https://sc01.alicdn.com/kf/HTB1gxqwfRDH8KJjy1zeq6xjepXaO.jpg

https://www.mouser.com/datasheet/2/149/FQP50N06-1009503.pdf

 

[be80be]

What did you power them with a car battery

 

[crutschow]

You can't leave the gate unconnected, as that puts the gate voltage at an undefined and arbitrary value.
Either connect it to the source to turn the MOSFET off, or to a positive 12V to fully turn it on.

 

[be80be]

I was just going to say that. The gate floating
Probably killed them or a power supply with over 50 amps of output current.
But even so the poster still needed to tie the gate to ground or it will come on floating.

 

[Miracletech]

What did you power them with a car battery

Yes I did.

 

[Miracletech]

You can't leave the gate unconnected, as that puts the gate voltage at an undefined and arbitrary value.
Either connect it to the source to turn the MOSFET off, or to a positive 12V to fully turn it on.

Is it that the gate was partially on, or some other problem?

 

[Miracletech]

I was just going to say that. The gate floating
Probably killed them or a power supply with over 50 amps of output current.
But even so the poster still needed to tie the gate to ground or it will come on floating.

But I didn't even give it a gate voltage at all. I used a 12V 6AH li-ion cell. But when I tested gate-source resistance(without gate voltage), it was infinite, or are they ESD damaged?

 

[Miracletech]

According to https://electronics.stackexchange.com/questions/45494/what-happens-when-a-mosfets-gate-is-left-open

"A high-impedance floating N-channel gate can pick
up charge from a variety of sources, (via the
parasitic gate-to-drain capacitance, via handling -
touching the gate with your finger, etc.) and
spontaneously turn itself on. Any gate that could
have a high-impedance control (a digital I/O from a
micro, for instance) really needs a gate-to-source
resistor to bleed off these spurious charges and
keep the device state defined (=off) when the
control is high-Z."
Looks like why they Got damaged.
Should I also add a pulldown resistor when switching at frequencies higher than 100HZ? And where? Before or after gate resistor?

 

[Diver300]

You need to pull the gate voltage up or down at all times that the circuit is powered. Dedicated MOSFET driver ICs will often pull up and down, so they don't need any other devices, but the gate voltage is controlled by a switch or a transistor, a resistor is almost always needed.

 

[be80be]

12v 6ah li-ionm
You need a load and you need the gate pull to ground or VDD
You have limit the inrush to a safe level.
Your battery look like a dead short and Lithium Battery can dump a lot of amps down the drain. LOL

 

[Miracletech]

I'll give it another try. Maybe it,ll work

 

[Miracletech]

12v 6ah li-ionm
You need a load and you need the gate pull to ground or VDD
You have limit the inrush to a safe level.
Your battery look like a dead short and Lithium Battery can dump a lot of amps down the drain. LOL

I didn't get " your battery look like a dead short". Could you explain more?

 

[be80be]

if you hook a big battery positive to drain negative to source.
its going to be a dead short.
You need a load to keep the MOSfet in spec.

The battery you say your using can output hundreds of amps.

You need load to limit it under 35 amps
Then it the gate is not pulled high or low to turn on or off the MOSfet will get hot
but if there no limit to inrush it will burn up fast.

 

[crutschow]

But I didn't even give it a gate voltage at all.

That's the problem,
If you don't give it a gate voltage, then it can assume any arbitrary voltage.
A open-circuit is not the same as zero volts for a MOSFET gate.

if there no limit to inrush it will burn up fast.

Also called a fast blow fuse.

 

[OBW0549]

I was testing some of my MOSFET in my stock (FQP50N06). I applied +12v voltage at the drain and -12v voltage at the source. I left the Gate unconnected. To my greatest surprise the MOSFET that I tested( even for RDS(on) and every other thing including body diode ) sparked and blew up!

You were surprised by that? Good grief...

Is it that I did something wrong?

Yeah, you sure did. First, you left the MOSFET gate unconnected, which allows the gate to assume whatever electrostatic potential is floating around in the surrounding air-- positive, negative, or anything in between. That potential may be enough to turn the MOSFET completely off, enough to turn it completely on, or make it partially conducting. Never leave a MOSFET's gate unconnected, because you have no idea what it will do.

Second, and worse, you connected the source and drain directly across a heavy-duty voltage source having more than enough current delivering capacity to utterly fry the MOSFET by greatly exceeding its maximum drain current rating. Always use a resistor or some other load in series with the drain, to limit the maximum drain current to a safe value.

Or is there something I am missing?

Yes. You're missing even the most basic knowledge about how to use MOSFETs. Stop "testing" these parts until you know how to do it properly.

Because I tried this with 3 FETS out of 10 and same thing happened.

The very first one to fail should have been a bright, red "STOP!" sign screaming at you and telling you to go no further until you figure out what went wrong. And "I screwed something up" should ALWAYS be your very first, default assumption when this happens-- NOT "the parts must be fakes."

 

[Miracletech]

Ok. I now understand my mistakes. Thank you very much.

 

[Miracletech]

A mosfet has a body diode across drain and source. Looks like I exceeded the diodes maximum ratings, making drain and source tho short, damaging the fet. But, what if I am switching about 40A? How do I prevent this? Or I go for another fet?

 

[Diver300]

A mosfet has a body diode across drain and source. Looks like I exceeded the diodes maximum ratings, making drain and source tho short, damaging the fet. But, what if I am switching about 40A? How do I prevent this? Or I go for another fet?

The body diode is in the direction where it won't conduct when the MOSFET is just doing simple switching. If you have an n-channel MOSFET, for example in this circuit:- https://en.wikipedia.org/wiki/MOSFET#/media/File:Mosfet_n-ch_circuit.svg

then the drain voltage is always larger than the source. When the MOSFET is off, the drain-source voltage is the supply voltage, and when the MOSFET is on, the drain-source voltage is just about zero.

The body diode will start to conduct when the drain voltage is less than the source voltage, when the power is applied the other way round.

The rating of the body diode doesn't matter unless it is going to be turned on.

If you take a MOSFET, and short the gate to the source, and then use a multimeter to measure the drain - source resistance, it will be open circuit one way and will conduct in the other direction. The MOSFET itself is off because there is no gate to source voltage, so what is being measured is the body diode.

 

[Nigel Goodwin]

A mosfet has a body diode across drain and source. Looks like I exceeded the diodes maximum ratings, making drain and source tho short, damaging the fet. But, what if I am switching about 40A? How do I prevent this? Or I go for another fet?

The current doesn't flow through the diode, so it doesn't apply.

I suggest you tell us what you're trying to do - as you seem to doing random stupid things.

So far, it's as if you've bought a load of new light bulbs, and decided to test them with a hammer - and then suspect they might be fakes when they don't work any more.

If you don't understand what you're doing, or why, as is obvious - tell is what you want and ask for advice before doing such things!.

 

[Miracletech]

This is what happens when you try to test something out of curiousity . I thought is was from mr mosfet, but it was from me.