Photovlaltaic Isolator (PVI) as FET Gate Drive: exploring the concept

Hello!

Excuse my ignorance but I was reading about Photovoltaic Generators used as FET drivers, and the implication appears to be that operation only requires an initial pulse signal to drive a FET Vth on, with the PVI driver having it's own floating voltage source to manage the FET Vgs threshold. Is this correct?

It's hard for me to wrap my thought around this because I am used to thinking of a FET as a switch having a constant voltage at its gate in order to turn it on, which seems to occur in the PVI, correct? See one of the app notes I read on this: IR's AN-1017. I would like to create a Solid State Relay and the PVI seems to be a great choice as it only requires the current source signal to initiate the PVI to drive the gate, but its hard for me to accept that the FET can be "latched" on by a single pulse signal from the PVI and "stay on" like a mechanical relay. I seem to want to think that the independent voltage source of the PVI on the Vgs would eventually dissipate.

Just trying to get to a better understanding of how this device works.

Thanks!

I tried this myself a very long time ago, and it is not without its problems.

The first issue is getting enough voltage to turn on a mosfet really hard. While there are now a wide range of logic level mosfets to choose from, it will still probably take several individual PV cells in series to do the job.

At around 0.6 v per PV cell the required number of cells take up space, which can be inconvenient.
The other problem is speed of operation. PV cells are current sources, and small cells don't produce very much current. Charging up and discharging the gate capacitance is slow.
And for a switch, if its not going to burn up, switching on and off needs to be both fast and clean.

All things considered its not really a practical solution.

Many isolators have limited common-mode dV/dt before
they lose the control signal in the slewing "noise".

Capacitor types are especially challenged. There exist
isolated gate driver ICs (Avago, and I believe others).
A signal isolator and a >1A, driver isolator will have very
different outputs (strength, range).

I used in the past a part from Vishay. The VOM1271.

The application wasn't critical...fully switching on and off a DC (1/2HP) motor.
Coupled to the proper Mosfet, it worked fine.
But I kept the LED on all the time.