I had some old receivers with similar issues, and, as you point out, the
problem sounds a lot simpler than it really is.

In one instance I increased the value of the "gimmick" capacitor until I
could receive SSB. Unfortunately, the BFO then was so strong that it
controlled the AVC, which was, of course, designed for AM or CW reception.
Also, on a strong CW signal, the BFO would pull, giving a weird chirp
effect. Of course, it was then great for autodyne reception of AM, and gave
listenable results even with serious selective fading, but that was not the
intent!

It's hard to fake a good design.

"Michael Black" wrote in message
...
"TerryJ" ) writes:
I have restored a Hallicrafters S77A . I could not receive SSB signals
no
matter how I twiddled the CW tone control and even CW tones were weak. I
guess it was not designed to receive SSB. Any way, I had this idea to
change
the injection point for the BFO signal. Originally the BFO was coupled
to
the plate of the second detector by a 2 pf 'gimmick'. I moved the BFO
injection to the screen on the second IF tube by adding a 680 ohm
resistor
between the screen and screen supply and connecting a 100 pf cap from
screen
to BFO output. After recentering the BFO in the IF bandpass I tried it
on
the 40 meter band. It worked! Sound is great and not too difficult to
tune.
It drifts and has to be readjusted periodically but never the less it is
usable for SW listening. I have looked around for schematics and none
that
I have found use this circuit. Could some of you pros explain why my
change
works so well and also why it wasn't used way back when.

It's because you don't need a strong local signal for hearing CW, just
a signal that beats against the incoming signal. Up until SSB came into
common useage, no receiver was designed for SSB. They all had an AM
diode detector, and the "CW" switch merely turned on the local beat
oscillator, ie the BFO. If you didn't need a strong BFO, there was no
sense in expending the effort for it, and besides, a strong BFO is
more likely to get into parts of the receiver where it isn't wanted.

But with SSB and a simple diode detector, the BFO has to be considerably
stronger than the incoming signal. With a weak BFO, it comes across
as "overmodulation", which isn't really a surprise since at the receiver
you are in effect creating an AM signal (the incoming sideband, plus
the locally generated "carrier"). When I bought a Hallicrafter's S120A
("A" for transistorized) receiver back in 1971, it was all I could afford
and ever dollar I spent on it was wasted, I too couldn't receive SSB,
much to my disappointment. I'm sure I interpreted the problem as
"distortion". It was only as I learned more that I realized the problem
was a too weak BFO.

So when SSB started coming in, there were plenty of articles in the
magazines about how to receive it with an old receiver. You can find
some, complete with better descriptions of what is happening, in
the first 2 or 3 editions of the ARRL SSB manual.

Since the BFO is too weak, what you need to do is attenuate the
incoming signal in some way. This changes the ratio of the incoming
signal to the BFO, so the latter is strong enough, and then the SSB
signals become receivable. As someone else said, the articles would
also mention turning up the audio control, which makes sense since
the audio stages are used for gain, to compensate for the loss of
signal when you attenuated the incoming signal.

When you moved the point of injection in your receiver, you have
found a point where BFO will get some gain (or perhaps the IF stage
is now acting as the mixer), so it's now stronger in reference to
the incoming signal.

Another scheme from when SSB was first coming in was to inject
at the antenna, on the signal frequency. That way, the whole
receiver is amplifying the beat oscillator as well as the incoming
signal. And when injecting on the signal frequency, the stability
of the receiver was not a concern. If one assumes the external
beat oscillator is more stable than the local oscillator in the
receiver, not too much a stretch given the time, then it will
stay on the same frequency as the incoming signal. As the receiver
drifts, the beat note between the incoming signal and the beat oscillator
remains the same, so all you have to worry about is retuning
the receiver when it comepletely drifts the incoming signal
out of the passband.

Of course, front end injection, or injecting the BFO before
the detector has the disadvantage that it will be a constant
signal, and thus the amplifier stages become more prone to overload
if something else comes along. It will also likely make AGC ineffective.

So when receivers intended for SSB came along, they not only had
SSB bandwidth filters but product detectors so one could go back
to a relatively weak BFO.

Michael VE2BVW