Author Topic: High Side Switching N Channel Mosfet  (Read 1840 times)

0 Members and 1 Guest are viewing this topic.

Offline EteslaTopic starter

  • Regular Contributor
  • *
  • Posts: 149
  • Country: us
High Side Switching N Channel Mosfet
« on: July 22, 2018, 02:20:08 am »
I am trying to better understand high side switching with N channel mosfets using discrete components. The problem I am running into is as follows: In what I understand to be the standard bootstrap configuration like in the picture below, the voltage from the gate to the source of the power mosfet is always about equal to the supply voltage. If I am trying to use a supply voltage higher than what the mosfets gate to source rating can handle, like above 20V ish, I blow up the fet... My initial thoughts are to use a zener to drop the voltage after R1, or to add a smaller resistor from R1 to ground to act as a voltage divider. I feel like both of these solutions are pretty crap simply because of the power losses associated with them and the large components they would require to work out... Im not willing to accept using a higher value resistor for R1 because I want to still be able to switch at high speeds, preferably above 40khz or so eventually. I am also trying to avoid the addition of another isolated power supply, as I would like everything to have a common ground... Gate drive transformers are also out of the question. Anybody know how this is done without IC's in real life? Thanks!
 

Offline duak

  • Super Contributor
  • ***
  • Posts: 1038
  • Country: ca
Re: High Side Switching N Channel Mosfet
« Reply #1 on: July 22, 2018, 03:34:12 am »
Where to start. 

As you've found, Vgs max for M1 is about +/- 20 V max.  A 15 V Zener diode with the cathode on M1's gate and anode on M1's source will protect M1 from excessive Vgs. 

Look at the reverse breakdown voltage for D1.  I think it's like 35 V or so.

This circuit has a few other problems and is probably not practical as it sits.  The FETs can be easily damaged with an improper load.    Without a floating gate power supply, the output voltage will be a few volts below the power supply voltage.

ICs bootstrap the high side drive with diodes and capacitors and will have a maximum cycle time requirement. ie., they can't maintain the high on state indefinitely.  I think some of the early app notes for the IR 21xx FET drivers show a simplified diagram.

Cheers,
« Last Edit: July 22, 2018, 03:37:48 am by duak »
 

Offline EmmanuelFaure

  • Regular Contributor
  • *
  • Posts: 154
  • Country: 00
Re: High Side Switching N Channel Mosfet
« Reply #2 on: July 22, 2018, 10:42:09 am »
For high frequency/non-DC signal the simplest and cheapest way is this one. It's not suited for DC or low frequency operation because of the capacitor discharge and gate/diode leakage currents. With 1nA leakage and a 1µF capacitor, drift = 1mV/second. The input PWM signal provides the energy to charge/discharge the gate. At startup, it takes some time for C1 to charge (Here < 1 milisecond), but it's perfectly fine in most cases.

If the supply voltage increases significantly it takes some time for C1 to charge again. If the supply voltage decreases significantly D2 is here to protect the gate against breakdown.

If you want to keep your schematic, the capacitor voltage limitation with a zener is the way to go.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf