58b0f3e6   
   From: apogosso@tpg.com.au   
      
   "Ian Iveson"  wrote in message   
   news:9bba5458-ecee-4d69-9a7d-afe7a22f4ffc@cc2g2000vbb.googlegroups.com...   
   > Flipper's posts disappear so the 10c thread looks like David and   
   > Goliath, with Alex facing up to the ranting Butcher all alone. Comical   
   > but hard to follow. I guess a clue lies in the term "trough detector",   
   > but I can't imagine what that means.   
   >   
   > So, Alex and Flipper. Can you tell me if I've got this right?   
   >   
   > A problem with self-regulating bias is that asymmetric current in the   
   > ouput valves, particularly at high signal levels, contains a DC   
   > component that the regulator interprets as an increase in bias   
   > current. The regulator will tend to respond by reducing the bias   
   > current, thus making the asymmetry worse.   
   >   
   > Triode operation in particular produces predominantly 2H distortion,   
   > so a symmetrical input signal will always result in a DC component in   
   > the output, increasing with signal amplitude. In addition, in PP AB,   
   > asymmetry increases rapidly when the output stage is driven into class   
   > B. The result, as with ordinary cathode bias, is a bias current that   
   > always wanders and sometimes strays considerably.   
      
   Interestingly, the concept of Brook was (and is) so smart that even today   
   certain people (including enlightened Patrick) can not grasp it.   
      
   While the total average current from both cathodes changes and wanders   
   greatly, the whole trick is that the minimae (troughs) of this current   
   remain the same. The minimum remains the same at idle (at idle there are no   
   troughs -- the current is stable) and at the loudest passages.   
      
   Thus the whole idea is to control the bias voltage measuring the troughs,   
   not the average. To measure the troughs you need the trough detector, not   
   just an averaging DC amplifier. Trough detector servo system can always   
   maintain a required target bias current regardless of the music content,   
   load and +B voltage.   
      
   In the days of Brook the implementation of the trough detector on a triode   
   was not perfect as the cutt-off knee point in a triode 6SN7 is not perfectly   
   sharp, but today, using Si diodes or transistors the trough detector system   
   can be brought to perfection (if anyone cares, I do not).   
      
   Regards,   
   Alex   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)