[amptramp]

I think there is a difference between the radio schematic and your drawing of an equivalent circuit.  The two diodes to ground are separated by a capacitor, not a resistor and the AVC side has a resistor which is returned to the negative supply.  A lot of sets use a single diode and get the AVC off a lowpass filter from the top of the volume control.  The problem with this is that it is possible for the AVC to back-bias the detector.  Suppose for example, the AVC voltage was returned to ground rather than the bias supply and was at -2.7 volts.  If this voltage was developed at the top of the 500 K volume control, the top of the volume control would be at -0.5 volts due to the divider effect of the AVC resistor and the volume control.  But if it is back-biased by this amount, the diode would not begin to conduct as soon as the transformer output goes above zero because it has to overcome the reverse bias, in this example, 0.5 volts.  This becomes important when you get near full modulation.  Keeping the AVC diode and the audio detector diode separate means you get the same audio regardless of the value of the AVC voltage.


The connection between the diodes cannot be a resistor, it has to be a capacitor to get this effect.  The separation of the VAC and audio diodes allows the audio to be returned to ground whereas the AVC is returned to the negative bias supply.