Bob Okas suggested adding a 470 ohm series resistor to the NE602
supply line to reduce the 7808 regulator noise, and add a 10uf
electrolytic in parallel with the 0.1uf supply bypass capacitor....
Just looked at the noise spec from Motorola, on the 7808.
They say wideband noise is (10 Hz to 100 Khz) 80uv, typical. I've
measured the input noise (over the narrow bandwidth of the audio
filter) of a few microvolts. That works out about right.
That IS quite a bit, probably masking atmospheric noise @ U3's output.
I'd guess that the 7808-generated noise @ 12Mhz would be much less,
and its output impedance much higher, so that the 0.1uf supply bypass
capacitor across each NE602 would be very effective (for R.F.). But
not at audio frequencies, as you say. So U1 is likely safe, U3 is not.
A differential amplifier at U3, pins 4, 5 would certainly help reduce
the impact of audio noise on the supply line, but this is a major mod.
Some time ago, a surface-mount choke was added to the
offending +8v line feeding U3. The noise went down, but this mod
added a terrific thump to an already thumping 38s. So out it went,
replaced with a wire jumper. Still thumps, but more quietly. Perhaps
it was added too far back, so that filtered Vcc was feeding other chips
as well. Vcc changed quite a bit when switching from TX to RX, enhancing
the thump.
Sure hope you don't have a thumper. (some do, other's seem immune).
At 6.5 volts, you may not get enough TX signal when this chip
is used as the TX mixer. Don't know enough about the NE602 internal
biasing to tell if running at a lower voltage will reduce gain, or
increase spurs.
[
QRP-L Archive |
]
[
1993 |
1994 |
1995 |
1996 |
1997 |
1998 |
1999 |
2000
]