Go Down

Topic: Mosfet gate resitor (Read 2348 times) previous topic - next topic

pito

#15
Aug 01, 2017, 10:57 pm Last Edit: Aug 01, 2017, 11:11 pm by pito
The mosfet is NOT current driven device. It is voltage driven device. The input resistance of the mosfet is hundreds or thousands of megaohms.
In static operation (no switching), NO current flows into the gate.
So no resistor is needed.
If the output voltage of 555 is 10V there will be 10V at the mosfet's gate for any reasonable resistor wired in between (ie. 1k or 10k, or 100k, or 1000k, or 10mega) in static operation.

Mind the threshold voltage of power mosfets could be 0.5 - 10V. No input current in static operation.

In case you are switching it (ie rising/falling edge of the gate voltage will be 10ns) the gate current could be 100A in peak.. That current comes not from gate, but from the "gate capacitance" - imagine there is a capacitor wired between G and S (and D).
You have to read about RC constant in the link above.



polymorph

The gate acts a lot like a capacitor. A resistor between the Arduino pin or 555 timer will charge it just as if it were an RC circuit, it will reach about 63% of the Vcc of the Arduino or 555 after one time constant, which is called Tau and is equal to the product of the resistor and the gate capacitance.



So within 5 time constants, the Gate voltage will reach Vcc. As you can see, the current drops off as the Gate charges, so less voltage dropped across the resistor as the Gate charges.

Note: It is a simplification to call the Gate a capacitor, but it isn't too important in this case.
Steve Greenfield AE7HD
Drawing Schematics: tinyurl.com/23mo9pf - tinyurl.com/o97ysyx - https://tinyurl.com/Technote8
Multitasking: forum.arduino.cc/index.php?topic=223286.0
gammon.com.au/blink - gammon.com.au/serial - gammon.com.au/interrupts

MarkT

[ Cringes at the obviously inaccurate shape of the exponential curve in those diagrams ]

Here's a better one: http://expeyes.in/experiments/rc-transient.html
[ I will NOT respond to personal messages, I WILL delete them, use the forum please ]

pito

#18
Aug 02, 2017, 12:07 am Last Edit: Aug 02, 2017, 12:08 am by pito
And, when the gate is at 10V, and the 555 goes to 0V, the gate capacitor discharges via the Resistor with the same RC constant into the Ground (because the 555's output is at "GND").
The current via Resistor will have an opposite direction, however..

pito

#19
Aug 02, 2017, 12:23 am Last Edit: Aug 02, 2017, 12:26 am by pito
This is how it looks like in reality. The charging and discharging the gate is more complex because of other effects involved, but it illustrates what we are talking about.


pito

#20
Aug 02, 2017, 12:35 am Last Edit: Aug 02, 2017, 12:53 am by pito
This is with R=100ohm. You may see below the gate current is much higher (+/- 75mA in peaks) but it charges/discharges much faster. The gate current is most of the time zero..

Mind it is an simulation with a Mosfet model only, the generator is 25kHz 1:1, 10V, 100ns r/f edges.



MarkT

Indeed, MOSFET gates aren't ideal capacitors by any means, the plateau is very evident in those traces.
In real life there's always ringing and other artifacts as well.
[ I will NOT respond to personal messages, I WILL delete them, use the forum please ]

pito

#22
Aug 02, 2017, 12:44 am Last Edit: Aug 02, 2017, 12:51 am by pito
And, the talented students may calculate the total input gate capacitance from the above waveforms and their timings :)

johnfg

Thank you guys! Ill take a look at these!

polymorph

That plateau is due to something called the Miller Capacitance or Miller Effect.

As I said, calling it Gate "capacitance" is a simplification.
Steve Greenfield AE7HD
Drawing Schematics: tinyurl.com/23mo9pf - tinyurl.com/o97ysyx - https://tinyurl.com/Technote8
Multitasking: forum.arduino.cc/index.php?topic=223286.0
gammon.com.au/blink - gammon.com.au/serial - gammon.com.au/interrupts

Go Up