https://en.wikipedia.org/wiki/Common_gate
The P-channel mosfet sourcefollower/buffer provides current gain and the common gate n-channel (Jfet in this schematic) provides voltage gain. The feedback loop to the Jfet gate is a normal Schade feedback. If the Schade-feedback resistor voltage devider is connected before the output capacitor it can be used to bias the N-channel device (enhanced mode).
Cheers,
Johannes
The P-channel mosfet sourcefollower/buffer provides current gain and the common gate n-channel (Jfet in this schematic) provides voltage gain. The feedback loop to the Jfet gate is a normal Schade feedback. If the Schade-feedback resistor voltage devider is connected before the output capacitor it can be used to bias the N-channel device (enhanced mode).
Cheers,
Johannes
From Nelson Pass article http://www.firstwatt.com/pdf/art_zen_iv.pdf
I think he describes the potential benefits with the common gate much better then I could, so I just quote him here.
Cheers,
Johannes
There are many ways to do this.
I am using a p channel buffer at input, but I prefer using the source of the n channel device as input node since it is usually much more linear and has much higher bandwidth then the gate.
In my experience the common source circuit is the worst in pretty much every way, except pure simplicity since you only need one single active device to build an amp. A common drain coupled to a common gate is much more linear and has much higher bandwidth with the same type of devices. With a common gate you evade much of the gate to source capacitance, with is a large contributor to the total amount of distortion in the amp, and especially so at high frequencies.
Nelson Pass often writes that he does not consider a cascode device to be a another gain-stage, since all it does is shield the gain-device from the voltage variations drain to source. It does not contribute much of its own character to the amp. I guess it is much more of a voltage-regulator to fix some voltage at a more useful value. (Sweetspot article)
Since a common drain (sourcefollower) is the most linear way to use a gain-device, it is quite logical to use a common gate cascode device to create voltage gain and shield the common drain from large voltage variations. The p channel buffer is perfectly suited to amplify the amount of current needed, so with a very low impedance high current, low voltage output, it seems very strange and mismatched to drive a many hundred mega Ohm non-linear gate instead of the very linear and low impedance source of the main gain device.
This is my reasoning behind this circuit, "but each to his own" as you say...
Cheers,
Johannes
I like picture books.
An externally hosted image should be here but it was not working when we last tested it.
I like those Cree devices. The best part is the "Ease of paralleling" in the spec sheet.
I can imagine a 2 kW cacoded version of the BAF2015. Say 700 volt powersupply and 100 ampere of constant current... The local steel plant would complain of reduced voltage to their electric arc blast furnace available from the powerplant whenever you fire up our surround system...
Better to live near a large lake so you can watercool the amps. No more Ice-skating wintertime though....
View attachment 550812
I like those Cree devices. The best part is the "Ease of paralleling" in the spec sheet.
I can imagine a 2 kW cacoded version of the BAF2015. Say 700 volt powersupply and 100 ampere of constant current... The local steel plant would complain of reduced voltage to their electric arc blast furnace available from the powerplant whenever you fire up our surround system...
Better to live near a large lake so you can watercool the amps. No more Ice-skating wintertime though....
You might need to build yourself a nuclear reactor first to power it.
Because of you, I just came up with a brilliant idea for a different project.
These ideas always seem to come when on the toilet. Don't ask why.
Hahahaha
