Maker Pro
Maker Pro

MOSFET's Gate Driver Circuit

BlackMelon

Aug 7, 2012
188
Joined
Aug 7, 2012
Messages
188
Hello,

I am curious in the circuit in the picture. Can we use only the part in the red box to drive the MOSFET? (I mean the collector of Qa connected directly to the MOSFET's gate). If so, must we make a darlington pair of two QAs to make it affordable for high gate driving current?

Thanks,
BlackMelon
 

Attachments

  • 2005FEB14_AMD_NT03.jpg
    2005FEB14_AMD_NT03.jpg
    14 KB · Views: 141

Harald Kapp

Moderator
Moderator
Nov 17, 2011
12,613
Joined
Nov 17, 2011
Messages
12,613
Can we use only the part in the red box to drive the MOSFET?
You can, but it has drawbacks: The turn on time depends on RA charging the gate of the MOSFET. This will be slow (assuming a reasonable RA for not overloading QA in the on-state). A slow gate drive imposes the risk of operating the MOSFET in the linear region for too long, thus bringing it outside its safe operating area (SOA) and in consequence destroying the MOSFET by overheating.

It is always a good idea to drive the gate of a power MOSFET from a low impedance source as Q1 and Q2 are in your diagram.
 

BlackMelon

Aug 7, 2012
188
Joined
Aug 7, 2012
Messages
188
A slow gate drive imposes the risk of operating the MOSFET in the linear region for too long, thus bringing it outside its safe operating area (SOA) and in consequence destroying the MOSFET by overheating.

Do you mean the region to the right of the transition line (dash line) in this picture? In my textbook, the regions of MOSFETs are defined in the opposite way of those of bipolar transistors. But in some application notes about switching devices, they defined both devices in the same manner.
 

Attachments

  • BB.PNG
    BB.PNG
    34.1 KB · Views: 151

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
Jan 21, 2010
25,505
Joined
Jan 21, 2010
Messages
25,505
SOA for bipolar transistors is a slightly different issue as it is limited by secondary breakdown in the high voltage/ high current corner.

Mosfets don't have this problem, and after generally limited by thermal conditions in the corresponding region.

When switching, the mosfet briefly goes through a linear region where the peak dissipation is typically half the power consumed by the load. The temperature rise depends on both how long the mosfet stays in this state, and how often it is in this state. The faster you can switch the mosfet, the lower the energy transferred to the mosfet and thus the lower the temperature rises. This can be really important if you have a device rated at (say) 2.5W switching a load of maybe 50W many times per second.

Because the gate has a capacitance, the amount of current required to change the voltage on it is inversely related to time taken to do it. High current (and especially low Rds(on)) mosfets have a higher gate charge, and thus a higher gate current is required to switch them. Gate drivers which can supply 4A of gate current are not unusual.

Knowing the gate charge, the Vgs(th), the load voltage/current, the switching speed, and the thermal characteristics of the device, you can quite easily determine if you need just a pull up resistor, a fancy gate driver, or something inbetween.
 
Top