Theoretical question: What are those MOSFETs doing in this schematic?

[KarenColumbo]

I tried to gather some clues about this schematic - I've heard the youtube demos and am trying to understand what does what in the circuit. Especially the MOSFETs – Q2 and Q5 – that are precariously positioned in the "fringe region" there make me wonder. Intuition tells me they have something to do with switching power für those FETs below them.

Can someone shed a light on this mystery?

[amz-fx]

They are active loads for the transistors that are below them (Q6 and Q3).

Q5 increases the gain available from the jfet. Not much is gained by using Q2 instead of a resistor since the Q3/Q4 circuit already has more than enough gain - C17 has been included to try to keep the fuzzface-like stage from oscillating.

regards, Jack

[R.G.]

Yeah - it looks like someone finally picked up on the lower output impedance version of the mu-amp with the resistor spacing between Q3 collector and Q2 source. Without that, the high gain of Q3/Q2 is lost with loading and a much lower gain results from Q3/Q4.

But there is probably plenty there anyway. It's kind of gilding a lily.

[amptramp]

They are part of a true µ-amp or SRPP (shunt regulated push-pull) stage.  The drive for the gate of the upper FET comes from the lower collector so when the collector goes low as the transistor turns on, this is coupled to the upper FET gate in a direction that tends to turn it off.  When the collector goes high, this is coupled to the upper gate to turn the FET on.  But both transistors are in series, so whatever current difference there is goes to the load.  The load always has to be a designed impedance - if you try to drive an open circuit or high impedance output, there will not be enough current going through the load and the output will limit at the top or bottom voltage.

To equalize the current change on the top and bottom (in the opposite direction), the source resistor for the upper FET should be the inverse of the transconductance of the FET.

[Kipper4]

Wow well that was thorough. I think I know what you meant too.
I was about to suggest an SRPP.
But it looks like you guys have this wrapped up.
Thanks Jack, RG and Ron

I've never seen this before with a mosfet in an SRPP.

I know that you have seen SRPP before though Andreas right?

[R.G.]

SRPP is quite old. It was developed first on tubes. The modern use in effects comes courtesy of National Semiconductor's inclusion of a variant of the SRPP wiithout the resistor separating the collector/drain of the bottom circuit. It was called a mu-amp in tube days because the gain is approximately equal to the mu of the bottom tube.

Here are a couple of articles on the topic from geofex.com in 2000 and 2001 respectively.

http://www.geofex.com/article_folders/foolwfets/foolwfets.htm


http://www.geofex.com/Article_Folders/modmuamp/modmuamp.htm

There are some other gems buried in geofex.com that turn up here fairly frequently.

[KarenColumbo]

Yeah, I've been reading everything about SRPPs I could get my mitts on for half a year now (geofex and muzique always being my first stops). From time to time I put something on my breadboard, too, and watch it with a wary eye. Since I really don't know what's happening in a circuit it's the repetitious reading that grows some synapses in the underdeveloped electronics region in my brain.

Thank you all for the detailed explanation - yet another piece of the veil is falling

[amptramp]

The best explanation is here:

http://www.tubecad.com/may2000/

A little bit of light reading and you will know as much about it as I do.

[KarenColumbo]

Tube Cad Journal rocks!