need a boost converter for gate driver

Thread Starter

pager48

Joined Nov 25, 2018
161
Going to use BJT totem pole for driving a MOSFET=<1Khz ~4.5A peak but have only 12v supply. What parameters does a 20v boost converter need to supply the totem pole which would be enough for peak current at 20v? Id rather use totem pole over discrete driver for this particular circuit.
 

Alec_t

Joined Sep 17, 2013
13,219
If the MOSFET is a standard or logic-level N-channel type and is used for low-side switching of a load you don't need a boost converter; the 12V supply will be fine for the gate drive.
 

Alec_t

Joined Sep 17, 2013
13,219
how much current does it need to output to handle the peak current draw from the gate?
Since you haven't specified the device type, how would we know? What does the datasheet say about the gate drive requirements? Can you post a link to the datasheet?
 

shortbus

Joined Sep 30, 2009
9,748
This seems to me kind of strange. Most non-logic mosfets have a maximum Vgs of 20V, why would you run some electronic at its maximum? Any spike at all in the circuit would be dangerous to the mosfet.

Why the resistance to dedicated drivers? Can't be due to board space because a boost convertor and the totem pole would take more space than a driver. And drivers are made to do the work.
 

ronv

Joined Nov 12, 2008
3,770
Its not a logic level MOSFET its a high voltage trenchfet. Simulation showed it will dissipate too much power with =<12v it needs a boost converter but how much current does it need to output to handle the peak current draw from the gate?

With higher Vgs it needs only a small heatsink.
The only current it needs to supply is enough to discharge the gate capacitance.
I picked a big ugly fet and simulated it at 1KHz. and the average current was 10.5 ma.
Just make the filter cap on the boost supply ~ 10 times the FET gate capacitance.
But I bet we could get a lot closer guess if you posted the proposed circuit and the FET part number.:)
 

Alec_t

Joined Sep 17, 2013
13,219
So from the datasheet we now know the FET needs a minimum Vgs of 13V to avoid thermal runaway, and has an absolute Vgs max of 22V. A Vgs of 20V would be uncomfortably close to the limit.
The typical gate charge is quoted as 42nC, so from that you can work out the mean gate current if you know the switching frequency and duty cycle. Peak current would be limited by gate resistor value and parasitic inductance.
 

Thread Starter

pager48

Joined Nov 25, 2018
161
So from the datasheet we now know the FET needs a minimum Vgs of 13V to avoid thermal runaway, and has an absolute Vgs max of 22V. A Vgs of 20V would be uncomfortably close to the limit.
The typical gate charge is quoted as 42nC, so from that you can work out the mean gate current if you know the switching frequency and duty cycle. Peak current would be limited by gate resistor value and parasitic inductance.
Are there specialty boost converters made for gate / totem pole drivers, or will any boost converter such as the mentioned XL6007 be adequate?
 

ronv

Joined Nov 12, 2008
3,770
Are there specialty boost converters made for gate / totem pole drivers, or will any boost converter such as the mentioned XL6007 be adequate?
A few observations when I simulate your FET.
You don't need to design your driver for 4.5 amps. With a 10 ns. rise time the most current you should see at 18 volts drive is about 1.8 amps.
upload_2019-6-30_21-14-11.png
The second is that because of the low frequency the switching losses really aren't a significant part of the power
dissipation.
The gate voltage is a big variable. About a watt per volt at 10 amps.
These are cute.
https://www.mouser.com/datasheet/2/468/RKZE-1369815.pdf
 
Top