MOSFET gate discharge circuitry.

[drillz]

Hi,

I am driving an N-Type MOSFET in an offline flyback SMPS. At the moment , the drive circuit is as follows...

However, I believe that the gate capacitance must have a discharge path to ground to make the MOSFET switch quicker...therefore, i have dreamed up the following circuit....

I would be most grateful if readers could possibly offer any comments on this, and whether you may know of websites with better gate discharge circuits?

(Incidentally, Vout = 24V, 2A
Lp = 866uH,
Np = 80
Ns = 5.5
DC Bus = 320V
Frequency = 224KHz
Max duty cycle = 50%

This is for orders that i am getting for 24V smps for LED displays for a marquee-hire company.)

 

[ScottyUK]

Why not use one of the many gate driver ICs from (for example) Interational Rectifier. They are designed to make interfacing logic signals to rectified mains safe and simple. The IR21** series would be a good starting point. Have a look on their website.


----------------------------------

Sometimes I wake up Grumpy.
Other times I just let her sleep!

 

[IRstuff]

Did you read the datasheet:

http://datasheets.maxim-ic.com/en/ds/MAX4426-MAX4428.pdf

This device has symmetrical rise and fall performance.

Are your diodes going to beat 20 ohms? Why not use 5 ohms as recommended as the low end in the datasheet? For that matter, do you really need the series resistor at all?

TTFN

FAQ731-376

 

[ScottyUK]

Oops, my bad... didn't recognise the Maxim p/n and should have looked for the data before posting! Just do what the datasheet says unless you know enough about the IC to make changes.


----------------------------------

Sometimes I wake up Grumpy.
Other times I just let her sleep!

 

[LionelHutz]

You have to charge and discharge the capacitance to turn the device on and off.

The right diode and 100ohm resistor are basically always in circuit so you might as well replace the 10k with a 100ohm and save 2 components. But, as already mentioned, special considerations to charge are discharge the device are not necessary when using a properly chosen driver IC.

Do you have a reason for wanting to change the design?

Did you ever figure out the proper way to do the feedback voltage in the question you asked before or are you just forging ahead with the poorly done thrown together design anyways?

 

[Comcokid]

Although there is no circuit in the data sheet, reading the general description for the MAX4427 there is the statement: High-current output drivers rapidly charge and discharge the gate capacitance of even the largest MOSFETs to within millivolts of the supply rails.

If you think you have a switching speed off-time issue, you could try another MOSFET driver or a different MOSFET. Also look at you layout - you could have some parasitic inductances around your MOSFET that could cause fast switching problems. This is especially true if your your circuit is an initial crude breadboard bench circuit.