SE Class-A developing

[ilimzn]

Your easoning is fine, and Hugh i think, is saying that while charging current can be quite high (because the BJT can pull the gate down), dischargng is limited by the voltage across Rgs and because this is not very high, the resistor limits the slew rate in that direction (when output signal is going towards ground).
Your estimation of the required current is fine... for me, I tend to design this sort of thing at twice the calculated value, so I'd set the current at about 8mA, but this would be a sort of rule of thumb thing.

[bogdan_borko]

Thanks ILIZMN you`re always been nice.

I`ve seen picture of yor designs on the Audofil net forum.(Crvena i plava iverica and I`ll join the forum soon as I find some mail (that is not yahoo) to be able to register. All of your replys are realy great and it`s a pleasure to read them. It would be very nice of you to join:
http://www.hifi-forumi.com

There`s not many experts round there (except Bora) so I`m inviting you to join.

Cheers!

[AKSA]

Hi Bogdan,

Yes, Ilimzn is right, there is a big difference between charging and discharging the gate capacitance of the CFP.

Charging is via the CCS, so is limited to whatever current you select. At very high frequency, there will be so much current diverted into the gate capacitance that the driver may even turn off. When this happens, the amplifier clips, since the driver has lost control of the mosfet. Discharging is much quicker, since the transistor will turn on harder to remove the charge from its collector to the emitter, which is very low impedance since it is connected to the load.

Thus the limiting factor will always be the current source, and Ilimzn's suggestion of 8mA is a good one. But watch the dissipation in the driver; with 20V across it, this is 160mW, acceptable, but I would not go much over 200mW if you are using a small TO92 device.

Cheers,

Hugh