MOSFETs with gate-drive specs below 1.8 volts

Depletion mode has a negative Vt. JFETs are commonly depletion mode. No MOSFETs known for discrete products, only NMOS ICs had depletion mode or zero threshold at Intel in 1980.

I ran google on "logic power mosfets for 1.8V systems" and got a bunch of hits.

At DigiKey you can even select the voltage of the logic level FET

Random search:

Defined down to 900mV

Only in low Vds types.

On a related note, there are 3.0V logic level protected-switch FETs from Infineon (and probably others). Very handy, and you don't need it to be fast...

Tim

-- Seven Transistor Labs, LLC Electrical Engineering Consultation and Design Website:

I tried some of the self-protecting fets and wasn't impressed. One oscillated; another one didn't actually protect itself.

What, did you zap it?

Cheers

Klaus

No, I just applied a reasonable drain voltage and cranked up the current.

I don't remember all the part numbers now. One was NCV8406. There was an NXP part too.

We wound up using regular mosfets and applying our own protections by measuring voltage drop and current.

torsdag den 14. marts 2019 kl. 16.29.20 UTC+1 skrev John Larkin:

hitting the thermal limit with a (too) large inductive load?

No, just power.

Yep, you zapped it.

It was supposed to survive thermal overload. I verified that it didn't.

Self-protecting FETs sound like a good idea, but I haven't found one that I'd design into anything. They seem to have fairly complex shutdown electronics that is powered by the gate drive. Some have specific requirements for that gate drive.

I'd like a fet that just draws more gate current as the chip gets hot. I can take over from there. An actual temp sense output would be better, but would need another pin or two.

Maybe they are only protected if zero or full (several volts) gate drive is applied. Some lower voltage may turn it on but not power the protection circuit.

PMV40UN NXP

- Michael Wieser

Thermal as well as over-current is why I prefer to use De-Saturation protection. As the FET heats up, its resistance goes up and the Desat kicks in earlier.

I remember playing with self protected FETs many years ago (Philips maybe ?) and it didn't work very well. They weren't very good FETs to begin with.

And low Vds threshold FETs are scarry in that you must make sure you keep them turned OFF in a half-bridge configuration which means either adding Miller-Clamp circuitry OR having to use higher voltage to drive the gate negative.

This might work:

you'd need a to differentiate between on and off

Sure, you can only detect the desat when the fet is supposed to be on. So no desat protection is going to be practical for high frequency switching.

Wow! Very cool, even if not good for very much current (tested at 0.1A?).

Seems to speak more about their process control (or sorting...) to have such a narrow Vgs(th) range and still be willing to spec that.

Another jellybean is BSS138, which is usable nearly that low, but not spec'd for it. And, obviously, much lower current. :)

If teeny jellybeans are acceptable, I'm fond of RUM001L02, which is spec'd max 18 ohms at 1.2V, among other Vgs(on)'s.

Tim