Ya, Mooly shorted the input to the bias decoupling, and MJ, those bipolars are a bipolar VBE multiplier, not a common base amplifier. They have no ground connection, so the impedance is set by the CCSs and the MOSFET gates, and mostly the 10K load resistor, which could be 100K or nonexistent, another pointless feature.
Attachments
-
power-follower-output-stage 01.jpg120.1 KB · Views: 138
-
power-follower-output-stage 02.jpg95 KB · Views: 144
-
power-follower-output-stage 05.jpg80.1 KB · Views: 124
-
power-follower-output-stage 06.GIF16 KB · Views: 132
-
power-follower-output-stage 07.jpg86.1 KB · Views: 130
-
power-follower-output-stage 08.JPG356.1 KB · Views: 133
You might think that a gainless buffer would outperform a full amplifier but simulations I have done indicate otherwise. Maybe the output is responsible for most of the distortion and an IPS and VAS are needed to correct some of that distortion. If so, then a gainless buffer is, again, pointless.