In message , gareth
writes
"Ian Jackson" wrote in message
...

Ian, with your greater experience than mine, it is the concept of
single-signal reception in which I am interested.

Have you any clues about that, please?

The only clue I can offer is that 'single-signal reception' is vague - but
presumably self-explanatory, ie the filtering is very narrow, enabling you
to receive only one signal (unless you have more than one on or very close
to the same frequency). It's a term that I recall being around when I were
a lad, but I can't say I've heard it much since. I guess it's been
replaced by more scientific descriptions of how good the filtering is.

Thank you, Ian. I have a vague memory of something in BadCon from about 40
years ago which
related to setting up for single signal reception, which involved no further
adjustments to phasing or BFO once it had been set. I am fairly sure, hence
my
enquiry that it involved phasing out the audio image as well as involving
the peaking
that comes from a single series resonant crystal.

Hence my assumption that the BFO frequency must lie half way between the
peak and the notch.

I've no experience of such things. The HRO I had in my possession 20 years
ago did not posses
the crystal filter, but in an effort to speed up development of my RX
project, it seemed to me
that a single crystal filter would be an easier starting point than a ladder
filter.

You're not confusing SSB generation (and reception) by the 'phasing
method', are you? That requires something quite different from the
action of the elementary single crystal filter we're talking about.

While a single crystal filter can provide a fair amount of selectivity
(combined, if you choose to use it, useful suck-blow or blow-suck
frequency response), it is not really suitable for 'serious' SSB
filtering. Its frequency selectivity characteristics don't really use
clever phasing out of the audio image. The passband peak is really too
sharp for either the generation or reception of good quality SSB, and
you usually need a 'proper' flat-topped filter, a 'proper' SSB phasing
TX or RX (which also does exist in a direct-conversion form) - or if
you're really clever, a 'third method' phasing TX (or, I suppose, RX).
That said, I'm sure that 'KISS' transmitters and receivers have been
made using a single-crystal filter - albeit having a somewhat limited
performance.
--
Ian

"Ian Jackson" wrote in message
...

You're not confusing SSB generation (and reception) by the 'phasing
method', are you?

Neither that nor Weaver's Third Method.


While a single crystal filter can provide a fair amount of selectivity
(combined, if you choose to use it, useful suck-blow or blow-suck
frequency response), it is not really suitable for 'serious' SSB
filtering. Its frequency selectivity characteristics don't really use
clever phasing out of the audio image. The passband peak is really too
sharp for either the generation or reception of good quality SSB, and you
usually need a 'proper' flat-topped filter, a 'proper' SSB phasing TX or
RX (which also does exist in a direct-conversion form) - or if you're
really clever, a 'third method' phasing TX (or, I suppose, RX).

Actually easier to set up for the Third Method, because all the phasing is
done
at a single audio frequency. But that's not what this thread is about.


That said, I'm sure that 'KISS' transmitters and receivers have been made
using a single-crystal filter - albeit having a somewhat limited
performance.

It's not for SSB. CW forever!

However, ISTR G3VA (RIP) in his TT column discussing Stenode
correction when trying to resolve voice through a single-Xtal filter.

On Sun, 25 Aug 2013, Scott Dorsey wrote:


"Single signal reception" to me would imply a narrow-sloped bandpass
filter but it sounds more like marketing than engineering.

"Single signal reception" is specific, about no audio image. Before Lamb
came up with the filter (and some argue it was someone else before him),
receivers were generally 'broad", there was no way to get rid of the
image, though I suppose at the time there were some lab receivers that
used really low IFs for some high selectivity (or that famous experiment
that used sharp low frequency antennas to prove the existence of a carrier
and two sidebands on an AM signal).

So the term applies to CW, siince at the time, the late thirties, SSB
wasn't really used by hams, and AM has no image in this context.

But in terms of SSB, it still means no audio image. So if you use a
direct conversion receiver, you get an audio image, and there's nothing at
audio that you can do to get rid of the image (same with CW and a DC
receiver, the audio filter will get rid of adjacent signals, but not the
audio image). You can't knock out the interfering signal that's on the
other side of zero-beat.

If you use a DC receiver that has phasing networks to get
rid of the audio image, you have "single signal reception" since you've
wiped out the image.

Same with a crystal filter, it allows only one sideband to pass so
anything on the other side of zero-beat is knocked down in strength
significantly.

MIchael VE2BVW

On Sun, 25 Aug 2013, Scott Dorsey wrote:

=?ISO-8859-1?Q?philo=A0?= wrote:
Well, even though the guy is a troll, I always try to make to best of
everything. Though I had known about the "homebrew" group I had never
before known of the existence of the "boatanchor" group.

Welcome! It is a good place! Traffic is much lower than it used to be,
but there are still plenty of interesting people hanging out here.

I can now relive the good old days.

Through the years I have gotten rid of most of my "boatanchors"
but happily still have my HQ-140-X

I had one of those when I was a novice and eventually did a horse-trade
for an R-388 that made me a lot happier, but you can't really complain
about any of those old rigs. You turn on the receiver and there are
people talking and after a few decades that's still pretty cool.

Did the R-388 have a phasing type crystal filter, or did Collins put a
mechanical filter in it? I seem to recall the former.

Michael VE2BVW

"Michael Black" wrote in message
news:alpine.LNX.2.02.1308251248110.28924@darkstar. example.org...

Same with a crystal filter, it allows only one sideband to pass so
anything on the other side of zero-beat is knocked down in strength
significantly.

Except in the case of CW through a single-Xtal filter, when the carrier
and BOTH sidebands***** pass through, but an interfering signal
on the other sie of the BFO is phased out.


***** very close in, eg 12WPM is 10baud.

gareth wrote:
"Scott Dorsey" wrote in message

Perhaps you, as indeed do others seem, are trying to interpret a technique
from
the 1930s and 1940s in terms of the multi-pole Xtal filters that are the
norm today?

Well, yes. That is the point of this thread, isn't it?

No, it isn't.

I have a junk box going back 50 years from which I intend to make the sort
of RX that I dreamed of as a teenager in the 1960s, on the basis that if I
do
not make use of all those museum bits and pieces, the executor of my will
will be likely to bin the lot.

I am inspired by the ham-bands only Eddystone EA12 and am making slow
progress
in a DIY effort to manufacture the gears for the dial drive and am now
considering
the manufacture of a Catacomb along the lines of the National NC100X.

One technique from those pre-mechanical, and multi-pole or monolithic
Xtal, filters was to use a _SINGLE_ crystal early on in the IF chain, and it
is that
single crystal together with its phasing control that interests me at the
moment.

Yes, and you would like to understand how that device works in terms of
modern nyquist filter theory, correct?
--scott

--
"C'est un Nagra. C'est suisse, et tres, tres precis."

Michael Black wrote:
On Sun, 25 Aug 2013, Scott Dorsey wrote:

Did the R-388 have a phasing type crystal filter, or did Collins put a
mechanical filter in it? I seem to recall the former.

Yes, it had the crystal filter, which was useful in a CW pileup, but not
really all that great for AM or SSB.

I traded _that_ up for an R-390A which has mechanical filters with very
sharp skirts combined with a narrowband audio filter for picking one CW
signal out. Audio quality for phone is not very good, but you can hear
stuff way down in the grass.
--scott

--
"C'est un Nagra. C'est suisse, et tres, tres precis."

On Sun, 25 Aug 2013, gareth wrote:

"Michael Black" wrote in message
news:alpine.LNX.2.02.1308251248110.28924@darkstar. example.org...

Same with a crystal filter, it allows only one sideband to pass so
anything on the other side of zero-beat is knocked down in strength
significantly.

Except in the case of CW through a single-Xtal filter, when the carrier
and BOTH sidebands***** pass through, but an interfering signal
on the other sie of the BFO is phased out.

No, you put the BFO on the slope of the filter, no different from some
fancier filter, and so the image is attenuated because it's outside of the
filter bandwidth.

Michael VE2BVW

"Scott Dorsey" wrote in message
...
gareth wrote:
"Scott Dorsey" wrote in message
Perhaps you, as indeed do others seem, are trying to interpret a
technique
from
the 1930s and 1940s in terms of the multi-pole Xtal filters that are the
norm today?
Well, yes. That is the point of this thread, isn't it?
No, it isn't.
I have a junk box going back 50 years from which I intend to make the sort
of RX that I dreamed of as a teenager in the 1960s, on the basis that if I
do
not make use of all those museum bits and pieces, the executor of my will
will be likely to bin the lot.
I am inspired by the ham-bands only Eddystone EA12 and am making slow
progress
in a DIY effort to manufacture the gears for the dial drive and am now
considering
the manufacture of a Catacomb along the lines of the National NC100X.
One technique from those pre-mechanical, and multi-pole or monolithic
Xtal, filters was to use a _SINGLE_ crystal early on in the IF chain, and
it
is that
single crystal together with its phasing control that interests me at the
moment.
Yes, and you would like to understand how that device works in terms of
modern nyquist filter theory, correct?

Harry Nyquist is far from modern, I have somewhere an essay of his from
1924,
something along the lines of, "Certain topics in telegraph theory"


What I was after was the standard way of setting up the phasing together
with
the BFO for eliminating an interfering carrier that was equally spaced from
the BFO
frequency on the other side.

"Michael Black" wrote in message
news:alpine.LNX.2.02.1308251307180.28924@darkstar. example.org...
On Sun, 25 Aug 2013, gareth wrote:

"Michael Black" wrote in message
news:alpine.LNX.2.02.1308251248110.28924@darkstar. example.org...

Same with a crystal filter, it allows only one sideband to pass so
anything on the other side of zero-beat is knocked down in strength
significantly.

Except in the case of CW through a single-Xtal filter, when the carrier
and BOTH sidebands***** pass through, but an interfering signal
on the other sie of the BFO is phased out.

No, you put the BFO on the slope of the filter, no different from some
fancier filter, and so the image is attenuated because it's outside of the
filter bandwidth.


That's not the case with a single-Xtal filter because of the notch caused by
the parallel resonant frequency.