From: apogosso@tpg.com.au   
      
   "John L Stewart"  wrote in message   
   news:John.L.Stewart.b9551a8@audiobanter.com...   
   >   
   > Hi John,   
   >   
   > What is the purpose of the cathode followers in front of 6V6s? At 320Vdc   
   >   
   > they can deliver as much as they are capable safely in AB1. Overdiriving   
   > the   
   > amp with Vg2=320V and going into control grid positive zone will   
   > either:   
   > a) burn the screen grids if nominal Ra-a load is present;   
   > b) burn the plates if the load happens to short out.   
   >   
   > You are correct Alex, the loadline needs to be adjusted from Class AB1   
   > conditions. Otherwise the screen grids draw excessive current & are in   
   > danger of failing. For the example I gave in the previous post I set the   
   > loadline at 5.6K. The attached PP 6V6 plate family illustrates this but   
   > these are not the actual loadlines.   
   >   
   > From the trace we see the B+ changes by about 13 volts in 9 ms. The   
   > current is coming from a 100 microF cap. If CE = IT, then the current   
   > pulse on the PS is 144 mA. So each 6V6 sees a total plate & screen current   
   > increase from nominal of 72 mA. The standing current in each is about 38   
   > mA. So peak current during this test was 110 mA. And Ik adds a bit more to   
   > account for the G1 current while +ve. Running as a vertical deflection   
   > amplifier the 6V6 cathode in peak rated at 115 mA, so looks OK. But in a   
   > real audio situation it is unlikely the amplifier would be run at this   
   > level. The program material would see to that.   
      
   With Vg2=250V (as per your load line plot) yeah, AB2 might be justified to   
   suck everything from the poor 6V6. But for Vg2=+320V in pentode (as per your   
   schematic) I guess AB2 is too much for a poor 6V6..   
      
   >   
   > I've attached here another trace, this time of music by Peter Gabreil, the   
   > beginning of his performance of Salsbury Hill. This one drives into a   
   > loudspeaker load. All done while my wife was out, being very loud, far   
   > more than one needs while inside! For this one the loadline was at 6.8K.   
   >   
   > This trial in Class AB2 is similar to what has been published in the older   
   > tube manuals showing 6F6s running PP Class AB2 in both triode & pentode   
   > connexion.   
   >   
   > My practice is to set the loadline in a PP pentode stage to be somewhat   
   > lower impedance than that which results in the maximum output. It is   
   > claimed that this helps to lower the 3H. It is also useful where the   
   > loudspeaker impedance rises somewhat above nominal at the resonances in   
   > some speaker systems.   
      
   This makes sense. Since in PP you do not have 2H anyway (in theory), you   
   have nothing to lose.   
   Then, the control grid characteristic emphasizes the peaks because of the   
   "law 3/2" curvature, but plate-to-screen current distribution tends to   
   compress the peaks. These two effects somewhat cancel each other at a   
   certain load, thus minimising 3H.   
      
   >   
   > Cheers, John   
   >   
   >   
   > +-------------------------------------------------------------------+   
   > |Filename: 6V6GTA 14W 4 PushPull C 5W.jpg                           |   
   > |Download: http://www.audiobanter.com/attachment.php?attachmentid=345|   
   > |Filename: 2013_03_21_001 Peter Gabriel Salsbury Hill 8W.jpg        |   
   > |Download: http://www.audiobanter.com/attachment.php?attachmentid=346|   
   > +-------------------------------------------------------------------+   
   >   
   >   
   >   
   > --   
   > John L Stewart   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)