Glen and the Gang,
I have experienced the very same problem with the very
same components, but with the NorCal Cascade!. The 7808 is a
noisy part, with typically 40-50 uV of wideband noise appearing
on the output. In my experience, some parts are lower noise than
others. My solution for the Cascade was to install socket pins
at the 78L08 regulator site and hand select the quietest part.
One trick analog designers use is to filter the power
supply inputs to each device. For opamps, which already have
decent power supply rejection characteristics, the rule of
thumb is a 100 Ohm resistor in series with the supply lead and
a 0.01-0.1uf cap to ground at the device. Such a technique
might be applied here.
The NE602's are spec'ed for a maximum suppply voltage of
8.0V, so they're being run at the limit. I've found some 602's
that get pretty noisy with this supply voltage. A noise reduction
approach would be to cut the supply traces at pin 8 of the NE602's
and install a 470 Ohm resistor in series with the 8v supply. Make
sure there are 0.1 uF caps at pin 8 of each IC. This does two
things: 1.) It lowpass filters the power supply (3400 Hz corner
frequency) 2.) It reduces the supply voltage at pin 8 to about
6.5V. Tacking a 10 uF electrolytic at each 602 would take care of
any low frequency noise.
Since my 38s is just about ready for initial checkout, I haven't
tried these mods yet (any volunteers?). If mine turns out to be
noisy, I'll try the mods and report my findings.
Bob - W3CD
[
QRP-L Archive |
]
[
1993 |
1994 |
1995 |
1996 |
1997 |
1998 |
1999 |
2000
]