From: apogosso@tpg.com.au   
      
   "John Byrns"  wrote in message   
   news:byrnsj-F07199.20475916072011@news.giganews.com...   
      
   > How do you end up with "better sensitivity" if you reduce the IF gain?   
      
   Alex:   
   Obviosly, if your AM detector is more sensitive you can achive the same   
   sensitivity referred to the aerial input with less IF gain, or having the   
   same IF gain, you can achieve (possibly) a better sensitivity referred to   
   the aerial input. Of course the later becomes useless if the aerial   
   sensitivity is already limited by the internal noise. (This is true for   
   communications receivers. But in some boooooring radios it is the AM   
   detector which limits the ability of hearing the weakest station.)   
      
   >> There is another less known effect. Not only the heavy C distorts HF   
   >> modulation due to slew rate limitations, it also reduces original   
   >> modulation   
   >> index at HF, working like sort of high-cut tone control. To understand   
   >> the   
   >> phenomenon without a deep maths, consider that on a steeply rising RF   
   >> envelope the detector has to charge the load capacitance. This sucks   
   >> extra   
   >> energy from the hi-Z IFT on top what is to be dissipated in the resistive   
   >> component of the load. On the steeply falling RF envelope slopes, R is   
   >> being   
   >> fed from a discharging C instead of the diode. Thus the IFT gets unoaded   
   >> on   
   >> the falling slopes. It is easy to see that the peaks are thus "cut and   
   >> rounded" and the troughs are "filled". Modulation virtually reduces. This   
   >> reduces the slew rate distortion, replacing it with a HF cut. In the end   
   >> the   
   >> sound is still crappy.   
   >   
   > This sounds like a good "bean counter" scheme, the extra high frequency   
   > loading   
   > reduces the slewing distortion, and the HF roll off can be used as part of   
   > the   
   > deemphasis needed to overcome the preemphasis applied at the transmitter.   
      
   Alex:   
   Then you need to establish a universal standard  level of crappiness of  the   
   AM detector, so that this level of "unmodulation" is the same for all the   
   boooring radios from all the beancounting manufacturers, so that all   
   radiostations can apply the same level of pre-emphasis (on top of   
   ear-wearing compression they already use).   
      
   >> (Those who are familiar with the operation of a ratio   
   >> FM detector, as opposed to a Sheeley discriminator, will see many   
   >> parallels.)   
   >   
   > I am familiar with the ratio detector, what are the parallels I should   
   > see?   
      
   Alex:   
   You do not have to see, you even should not see, but you might see (in the   
   ways how AM modulation is eliminated).   
      
   >> (I use an active integrator based AGC amplifier for that purpose, so that   
   >> all the stations are levelled up to the optimum level, but Partick's AGC   
   >> into the mixer only is very primitive and inefficient. Besides it   
   >> introduces   
   >> more distortion by nonlinearly loading the IFT. However it may be a   
   >> separate   
   >> thread to discuss.)   
   >   
   > What is the integration constant of your active integrator AGC, what   
   > effect does   
   > it have on tuning feel and fading?   
      
   Alex:   
   In my radios AM detector works at 1V carrier level. Output of the AGC   
   integrator can swing from 0V to --30V (if the radio uses initial self bias   
   for the 6BA6s IF (or from --3V to --30V if it is a radio based on old tubes   
   6A8, 6K7, cathodes of which are hard grounded).   
      
   Time constant of the integrator itself is 0.1sec, but it is sort of   
   "multiplied" by the effectiveness of the AGC: by how many dBs gain changes   
   for 1V change in the AGC voltage. In comms receivers, where AGC is applied   
   to both IF stages and to the RF stage (not to the mixer!!!!, compare to what   
   Patrick does!) the AGC becomes more efficient thus reducing overall reaction   
   time. If this issue stirs up more interest I will elaborate more.   
      
   > Couldn't slewing distortion be eliminated by using a pure resistive load   
   > on the   
   > detector diode, eliminating the capacitive component of the load   
   > impedance?   
   > This could be accomplished in one of several ways.   
      
   Alex:   
   It certainly could. But a detector working on a resistive load has to have a   
   relatively small load (under 30...50K), it has a gain of  about 1 / pi even   
   with an ideal diode and has quite a nasty feature of generating IF harmonics   
   which can interfere with the radiostations. It must be followed by a cathode   
   follower, and then a RC filter, not the way around.   
      
   Actually the best is to use a full wave rectifier working of a paraphase   
   transformer into a resistive load. Gain is 2/pi.   
      
   Again if further interest is expressed, I can elaborate.   
      
   > Discuss "optimum bias of an unbiased detector"?  Isn't that an oxymoron?   
   Alex:   
   No.   
   Unbised detector -- is the one in which the average diode current is   
   approximately proportional to the carrier level.   
   Biased detector -- is the one where the average diode current is almost   
   independent of the carrier level.   
      
   Bias of a diode is simply initial voltage across it with no signal.   
      
   For example, if you use a silicon diode detector and apply a 0.45V BIAS to   
   the diode, still the whole detector will be UNBIASED since at 0.45V there   
   will be no current through the diode at zero signal.   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)