[Elecraft] Help needed with SSB problem

[Donald Wilhelm]

Fran,

What you are experiencing has happened to me early in my trials with the
filters and BFOs.  The problem seems to be that you are using a single
signal to align the filters - and that is a hard way to do it.  You really
need to use a wideband signal input first rather than a single signal
because using a single signal adds the VFO tuning as a variable where the
wideband source does not depend on the VFO setting.

For those without a wideband noise source use a 'dead' band - some noise
will be present.  Spectrogram or some other waveform indicating device
allows  you to see the filter bandpass correctly unless something in the
path is overloaded (turn off the AGC, use the external speaker jack, and
keep the computer audio gain at mid-setting or below).

After you have set up the filters and BFO frequencies with the wideband
noise input, then you can use a single signal source to doublecheck your
work - some small change to the BFO frequencies may be more pleasing to your
ear, but the change should be quite small - just one or two 'clicks' from
that obtained with the wideband noise source.

Hope this helps.
73,
Don W3FPR

----- Original Message ----- >
>     3.  I had done a rather precise job of adjusting the CW filters using
a
> frequency counter and one of the birdies.  Yet the center frequencies on
the
> plots are at 800 Hz and 500 Hz.  I seldom use CW-Rev but I do remember
that
> I had set things up such that the note did not change when I was spotted
> right on a signal.  My sidetone pitch is set for 600 Hz.  I used
spectrogram
> to center the 40M birdie at 600Hz.  The frequency readout was 7000.01 at
> that point and I could detect no difference between CW and CW-Rev.  I
added
> the noise generator and turned the RF gain back down to 12 O'Clock.  I
looks
> like there are 3 peaks across the top of the CW filter and I am centered
on
> the left peak.  The CW-Rev filter only had one off-center peak that the
> birdie is centered within.  This is curious but I don't know if there is a
> problem indicated here.
>