Possible ground loops? Feed the power to the device from an independent floating power supply and see if the gunk in the center of the band goes down. Should not be too hard to try out, I would do it myself but I have not received my order yet. Better yet supply the power from a set of batteries.
toggle quoted message
Show quoted text
Ahh the small board -- good question as to why is it so dang small -- I
think one of Tony's original goals for the gizmo was more of a
demonstration vehicle than an experimenter vehicle. I for one wish it had
more working room, but I will say the experience has vastly improved my
soldering and rework in small places.
The central hump you're seeing around 7.056 is normal. The gunk you're
seeing around this is a consequence of the software architecture. What is
happening is that the hardware takes the swath of RF centered and 7.056 and
downconverts it such that 7.056 is essentially at DC. Also down there
near DC is noise and 60 hz (and its harmonics) hum. When one tries to tune
down near 7.056, the local software oscillator will be down in the 100's of
hz's. The output of this will be Software LO +/- 60 Hz (and
harmonics). Things will always be a bit messy near the center frequency
-- quieter less hummy sound cards can help, but I don't believe there's any
practical way to get rid of all the gunk in the middle.
If you want to check out the hardware sans computer, and have a signal
generator and a scope you can put in a -30 dbm signal at 7.050 or so into
the antenna. Look at the audio output on the I and Q channels -- it should
be about 6 khz (7.056 - 7.060) and on the order of about 0.8 Vpp or so
(don't take my nums as precision numbers, they are all from memory). The
thing to look at is if both the I and Q channels are comparable in
magnitude -- if one is half the other something is not quite right in the
hardware. You can also take a look at the phase relationship -- it should
be about 90 degrees between teh two channels. While you've got the scope
out check the clocking signals to the FST3126. If you've got a generator
and no scope, you might be able to compare the two channels with a good RMS
voltmeter.
One last thing to check -- make sure you're using a sound card with a
stereo input. Using a mono mic in input would give the symptom of not
being able to null the image.
Hope this helps.
Regards,
Bill (kd5tfd)
Yahoo! Groups Links
--
Cecil
KD5NWA
<www.qrpradio.com>
I fail to see why doing the same thing over and over and getting the same results every time is insanity: I've almost proved it isn't; only a few more tests now and I'm sure results will differ this time ...
