Stacking MOSFETs

Is there a canonical topology to stack them in series such that one gets a switch with a higher Vds rating but maintain approximately the same switching speed?

Reply to
bitrex
Loading thread data ...

Yes, each one needs its own {isolated} high-speed high- current gate driver. Usually some effort is put into matching the gate-driver delays, but MOSFETs don't mind avalanche breakdown and are tolerant of using this to protect themselves for a short time. But since you're talking about speed, one danger involves dealing with the voltages developed from high dI/dt.

--
 Thanks, 
    - Win
Reply to
Winfield Hill

Win is right, but you might also consider silicon carbide fets. How much total voltage do you need? Can the switch sink to ground? How fast? How often?

Sometimes one can arrange to drive only the lower, grounded-source fet in a string.

--
John Larkin         Highland Technology, Inc 
picosecond timing   precision measurement  
 Click to see the full signature
Reply to
John Larkin

I vaguely remember a circuit in AoE3 like that... with resistors and /or zeners to turn on the other gates? But not as fast I'd guess.

George H.

Reply to
George Herold

Figure 9.111, pg.697 GH

Reply to
George Herold

I was mostly hoping for an expedient, temporary solution as I need to test a circuit that requires 100 volt-rated FETs but I only got 50-volts in stock and I'm waiting on a shipment, and I'm impatient. the drive frequency is only about 1 MHz switching under an amp.

In this case that might be an option to get on with business for a while.

Reply to
bitrex

Isolated high-voltage drivers require special circuits, usually involving transformer-powered circuit fragments. But at voltages below 600V, you can use high-side driver ICs, like Fig 3x.108, Table 3x.5. Drive sets of stacked MOSFETs. Probably none of those in stock either, huh?

--
 Thanks, 
    - Win
Reply to
Winfield Hill

Sigh....

Reply to
bitrex

Try two series FETs, with the high-side FET shorted gate-source. Then drive the low-side FET. The high-side will avalanche when it sees the full 100V. Will this work? I doubt it. But at a glance I can't see why not...

Reply to
sea moss

If the duty cycle is reasonably far from 1 and bipolar drive is not a problem: Ze Transformah...

Best regards, Piotr

Reply to
Piotr Wyderski

If the duty cycle is reasonably far from 1 and bipolar drive is not a problem: Ze Transformah...

Best regards, Piotr

Reply to
Piotr Wyderski

Low voltage NMOS and high voltage SiC JFET in cascode configuration?

formatting link

--
Cheers 
Clive
Reply to
Clive Arthur

formatting link

Reply to
plastcontrol.ru

The OFF condition should be OK. But for the ON condition the high-side FET will remain off, except it may avalanche at some voltage. At any rate, the output won't be pulled down to GND.

--
 Thanks, 
    - Win
Reply to
Winfield Hill

That circuit looks good. Nice find!

The secret is the gate capacitors to GND, with zener protection diodes for the FETs Vgs. If there's no load for pullup, a high-side switch will also be required. But it's not clear why the high-side switch can't use the same trick as the low side, employing only one isolated gate drive on the bottom FET of the stack.

--
 Thanks, 
    - Win
Reply to
Winfield Hill

???????, 6 ????? ? 2020 ?., 22:24:34 UTC+3 ??????? ????? bitrex ?????? ?:

formatting link

formatting link

Reply to
plastcontrol.ru

Look at AoE3 figure 9.111, Maybe Paul put this one in there?

GH.

Reply to
George Herold

No, I put that one in there, but I'm not sure they'll work the same. Our caps were meant to help equalize Vds drops, whereas the paper's circuit, with a separate cap to ground for each MOSFET's gate, insures that it'll be aggressively turned on if its source is pulled down. Anyway, it's a quick easy circuit, nicely satisfying the O.P.'s request.

--
 Thanks, 
    - Win
Reply to
Winfield Hill

A brute-force way to do this is to use some commercial (or home-made) gate driver circuits and float them on each of the fet sources, with dc/dc converters to power each one. DC/DC bricks with high isolation and low capacitance are cheap nowadays. Get the signals up there through an IC isolator or a small transmission-line transformer. Works for high and low sides.

--
John Larkin         Highland Technology, Inc 

The cork popped merrily, and Lord Peter rose to his feet.  
 Click to see the full signature
Reply to
jlarkin

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.