The schematic of the amplifier in question is posted on Bob Carver, Corp website:
https://9d0691c9-2bb0-4a51-acf2-a4e121c ... 28d241.pdfThere was some speculation about whether there is anything from a circuit perspective to compensate for the low power iron used, but this circuit looks to be a modified Mullard front end circuit with a pentode PP output stage, and with more complex feedback/compensating networks than usually seen.
1. Zobel across the secondary
2. Zobels from the B+ CT on the primary across each primary winding.
3. Fairly complex feedback network from transformer secondary + output to cathode of input tube.
4. A Zobel at the output of the input tube (R35+C18)
5. Compensating network (R20||C22 into R42) at each output tube grid. This is similar to what is done with the Hafler/Keroes 6550 UL Williamson
6. Interesting way of lowering the grid leak resistance when drawing grid current, with (R45+D3)||R42 at the output tube grid. I guess this helps maintain bias stability. This is interesting.
It appears there is alot of work to try to make the small output transformers work. BTW, the biasing instructions indicate that the 4 output tubes are biased to a combined 100mA current, or 25mA per tube (for KT120s). This means the tubes are significantly underbiased by conventional standards.
David