Well, I think this is just a hypothetical solution until some bench measurements can be made. There are some unknowns to be determined, such as does this provide any pickoff point for monitoring odd harmoncs for a common mode loop. Another question I have is does this work with pentodes and triodes. And how well does it work compared to just a CCS or resistor. The SS versions are pentode like. And tubes have a 3/2 power characteristic, but Mosfets are closer to square law. Triodes are closer to linear rule with some 3/2 power mixed in.
The split tail scheme is still an approximate solution I think, so I wouldn't rule out some of the other earlier ideas. I especially like the idea of using the driver screen grids for feedback.
Need some way to get a very clean differential test signal to test this and some of the other approaches with a sound card analyzer. When we have fully characterized the circuits, then go for bench testing an output stage with cross coupled feedback or partial feedback (as in Baby Huey).
Don
The split tail scheme is still an approximate solution I think, so I wouldn't rule out some of the other earlier ideas. I especially like the idea of using the driver screen grids for feedback.
Need some way to get a very clean differential test signal to test this and some of the other approaches with a sound card analyzer. When we have fully characterized the circuits, then go for bench testing an output stage with cross coupled feedback or partial feedback (as in Baby Huey).
Don