What seems like the final scaling problem in the receive chain was
discovered and repaired today. Eric did a nice job of setting up a test
scenario and giving it to me and it took about five minutes to figure
out the bug. If we take in a level 1.0 power tone, in the passband,
we would like to get a level 1.0 output for all inputs inside the AGC
range. We do now. Before this scaling problem was discovered, we got
amplitude 0.5 or 6 dB less power out from the DSP than we put in. The
issue is that we assured power 1.0 except in the very last step, when
we threw away the imaginary part to send the reals to the speakers. DUMB.
Eric tested with 0.1, 0.01, 0.001 inputs and we got that as outputs with
0 dB gain in the AGC and got that out.. With 20 dB gain in the AGC,
0.01 produce 1.0 to within a few hundredths.
Sorry for the low audio output.
Bob
N4HY
--
Laziness is the number one inspiration for ingenuity. Guilty as charged!