so i've been building this 40m direct conversion receiver from scratch, based loosely on the Neophyte design but i swapped out the audio stage with an LM386 because thats what i had in the junk box. overall the thing actually receives, which honestly surprised me after all the rework i had to do on the band pass filter section.

the problem now is the audio quality is just... bad. like there's this low frequency hum that i cant seem to track down and on strong signals i get this weird distortion that almost sounds like motorboating. i checked the power supply filtering and added a 100uF cap across the rails which helped a little but its still there. the LM386 is running at minimum gain right now with a 10uF between pins 1 and 8 removed, so its at like 20dB not 46.

anyone dealt with this before? im wondering if its RF getting into the audio stage somehow or if my ground plane is just garbage. i built it on perfboard which probably didnt help things. the hum is there even with the antenna disconnected so i lean toward a grounding issue but not totally sure.

the LM386 motorboating thing is super common, almost always a decoupling problem. put a 0.047uF cap directly across the supply pins of the chip itself, as close to the body as you can get on perfboard. also try a 10 ohm resistor in series with the supply to the chip with another 100uF on the chip side of that resistor. basically you're making a little RC filter right at the chip. that usually kills the motorboating dead.

hum with antenna disconnected does sound like either a ground loop or you're picking up mains from somewhere nearby. how are you powering it, wall wart or battery? if wall wart try running on a 9v battery temporarily just to rule that out. i built something similar a few years ago and spent like two evenings chasing a hum that turned out to be my bench power supply inducing into the circuit through a shared ground path. wasnt even connected electrically, just proximity.

yeah perfboard dc receivers are a pain, the layout really does matter more than people think. i did a 20m one a while back and had similar issues, ended up having to run a dedicated ground trace around the whole board just to keep the audio ground separated from the RF ground. might sound overkill but it worked.

also just something to check — whats your bypass cap situation on the VFO or BFO section? sometimes the oscillator noise couples right into the audio and looks like hum but its actually oscillator garbage riding on the audio. hard to tell without a scope honestly.