Here we have the finalized design for the balanced PP amp:
Current values have been added.
Component values in blue signify changes from the previous schematic.
The treble performance of the amp was still bothering me. Although the highs were all there,
they seemed a bit down in level, compared to the rest of the music spectrum.
To cut a long story short, it was Morgan Jones, "Valve Amplifiers" (4th ed) that once again came up with the answer.
Because this is a balanced shunt pot volume control, effectively centre tapped to a virtual earth,
the series resistors actually needed to be
half the value you would normally fit to a 100K pot.
A carefully matched pair of 91K, 1% metal films were parallelled up to the existing 91Ks and we now had the correct upper frequency response.
I didn't stop there. A bit more reading of Lynn Olson's Amity write-up and a pair of 30uF PIO 450V motor-run caps were shoehorned into the chassis
and connected up WE style to the HT where it fed the centre taps of the OPTs.
Now with Ultralinear, with FB operation, the ultrapath caps suggested by Mark had not been a success, but with feedback-less
triode-strapped operation of the output stage, the difference was shocking; night and day better in favour of the WE connection.
HF transparency was very good, soundstaging and bass performance were quite a way ahead of the simple un-bypassed cathode resistor I had been using up to then.
Interesting.
Wonder why the WE connection was so awful with Ultralinear and plate-to-grid feedback, but is so obviously good with triode operation, sans feedback.
Sgt. Baker started talkin’ with a Bullhorn in his hand.
