I have some questions about the diode selection and the feed to the AF amp so a couple of quick sims.
BAT46 diving a 50 ohm load. (45uV)
BAT54 driving a 1K load (100uV)
1N34A driving a 50 ohm load (80uV)
Now for a surprise???
1N34A driving a 1K load at plum new 1 mV. All I changed was the diodes and the load resistor and we go from 45uV to 1 mV. The 1N34 was developed to be used in radar some maybe it should not be a surprise that it would function well but what just happened? That thing about Z matching from a couple post back. The 1N34A is 40 - 50k Z while the BAT46 is 200K.Just to test/prove the point I put a couple 100k resistors across two diodes and .
Notice the negative is unchanged. So armed with this info I'll try some different diodes and output loads.
The BAT54 with a higher Zo is more respectable.
D18 diode at 5K. looking good
The 1N34A at 40k
As a final check, the 1N34A at 400K. Notice once the load goes above 40K we see little change. So now the question is do we have some 1N34As or D18s if not how to match the output?