On Saturday, January 25, 2014 7:39:25 AM UTC-5, haitic...@gmail.com wrote:   
   > Hi George,   
   >    
   > You must understand, I am bringing to this conversation about the   
   intelligence of a house fly.    
      
   OK I hope I wasn't too rude.     
   >    
   > I am merely looking at what Horowitz says in Art of Electronics around page   
   1027 et seq.    
   > And I see, gee whiz, the signal emerges after just adding many measurements.   
   And I feel like the native confronted by an explorer who flicks a lighter to   
   make a flame.    
   > Despite this, I'm sure it has its limits. I pushed it at Phil, since he has   
   a nose for noise, to see if this was widely useful.    
   >    
   > One the problems is that signal conditioning has many tricks and methods,   
   sometimes operating simultaneously. "Lock-in" gains you say 40 dB due to band   
   width narrowing, and another X dB due to phase alignment.    
      
   Hmm the X dB (X=3) 3dB gain from the phase is interesting.  Though the 3dB is   
   certainly real, a colleague has me convinced that there's no real gain in   
   noise when compared to a band pass filter (with the same Q).  In a nut shell   
   he agrues that though the    
   LI gets rid of 1/2 the noise in one phase, it also mixes in noise from both   
   positive and negative frequencies near the switching frequency... so no net   
   gain... of course it's (a lockin) a big win in other areas!     
      
   George H.   
   >    
   > It's telling how Horowitz deals with this issue. He doesn't. He just   
   presents examples.    
   >    
   > I was involved with the Astronomy Department at Harvard at one time, but I   
   never met Mr. Horowitz. Has anyone ever met him? Just curious.   
   >    
   > JB   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)