cc06d51c   
   From: apogosso@tpg.com.au   
      
   "Patrick Turner"  wrote in message   
   news:73714727-d7cb-47f1-9854-cbf4e0b99a92@k13g2000prg.googlegroups.com...   
      
   >   
   > The 47R resistors in the collectors do not do anything because the   
   > transistors are not going into saturation. I do not know why these 47R   
   > resistirs are there.   
      
   Transistors need to be protected from themselves and unless collector   
   resistors are added to my circuit,   
   along with series base resistors, they can run away with enthusiasm   
   and short circuit themselves to death in less time you can say   
   beejaytee. Too be SURE the collector current is never more than Ek /   
   47 ohms, the 47 ohms are used. Its not a critical value.   
      
   >   
   > But I overlooked 1K resistors in series with the bases. These indeed   
   > soften   
   > the transistor opening.   
      
   Yes, indeed.....The angel or the devil is always in the detail.....   
      
   > Assuming typical beta=100, these 1Ks translate into   
   > virtual 10R in series with the emitters.   
      
   What makes you think beta = 100? most power transistors might be less.   
   But it ain't critical, because the local NFB around the bjt circuit   
   means that while ever Ik peak remains high enough to turn on the bjts,   
   they conduct, and they cease conduction when they don't have to.....,   
   or more if they do have to, until the job is done.   
      
   > That means that from 10R   
   > degeneration under 65mA it goes to about 5R degeneration above 65mA. Only   
   > 4%   
   > net transconduction variation.   
   >   
   > Yet it is crazy to mix distortion current, created by the exponential   
   > curve   
   > of the transistor, into the cathodes of the "smooth" tubes.-   
      
   Nah, you are always seeing craziness in other blokes' attitudes when   
   in fact they might be a whole lot more sensible than you are. The tube   
   **current** in AB amps has HUGE THD, but the bjts hardly make any   
   difference whatsoever. I suggest you spend 30 minutes converting an   
   amp you have laying around so that it works like mine, you'll be very   
   happy, and wonder why you were so negative.   
      
   Alex:   
   I just see illogical to add transistor distortion to valve amps. If you used   
   say 1R current sensing resistors instead of 10R and  x10 op-amp gain stage   
   before the base of the shunting transistor, I would consider it a "better"   
   solution.   
      
   Have you done the following experiment. While your amp is idling or playing,   
   intermittently short one of your 10R sensing resistors to ground. Will you   
   hear some crackling, even soft, not loud? If yes, then it is the amount of   
   "transistor-like" distortion which your circuit mixes in. (You can argue   
   that inherent THD is much greater, and/or masked by the loud musical   
   material, etc., but then why build tube amps at all?) Or perhaps you will   
   not hear any crackling from that experiment -- maybe your NFB absorbs it.   
   Then -- congratulations -- you "won" the argument.   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)