I would like to build an old-style receiver using a BC-453 stuck at 500-kHz as
IF chain. A possibility would be to put a pentode in front of the BC-453 mixing
the incoming RF signal with the signal generated by a variable local oscillator
(VFO). To improve selectivity, one could adopt a regenerative arrangement
whereby part of the pentode plate signal is fed back into the grid (by inductive
coupling), and the cathode resistor is then adjusted just before the tube starts
oscillating. In other words, a kind of Q-multiplying converter (I think this is
called "Q-dyne" receiver). What I am not fully sure about is if the increased
selectivity I so obtain turns into a higher rejection of the image frequency, or
just into a narrowing of the received bandwidth (which is already narrowed down
by the tight BC-453 85-kHz IF transformers). My feeling is that said
regenerative scheme would offer no advantage in terms of image rejection, but I
would value very much your opinion on that subject.

73

Tony I0JX

Antonio Vernucci wrote:
I would like to build an old-style receiver using a BC-453 stuck at 500-kHz as
IF chain. A possibility would be to put a pentode in front of the BC-453 mixing
the incoming RF signal with the signal generated by a variable local oscillator
(VFO). To improve selectivity, one could adopt a regenerative arrangement
whereby part of the pentode plate signal is fed back into the grid (by inductive
coupling), and the cathode resistor is then adjusted just before the tube starts
oscillating. In other words, a kind of Q-multiplying converter (I think this is
called "Q-dyne" receiver). What I am not fully sure about is if the increased
selectivity I so obtain turns into a higher rejection of the image frequency, or
just into a narrowing of the received bandwidth (which is already narrowed down
by the tight BC-453 85-kHz IF transformers). My feeling is that said
regenerative scheme would offer no advantage in terms of image rejection, but I
would value very much your opinion on that subject.

For CW reception, the Q-multiplier can give you tighter selectivity without
needed crystal or mechanical filters in the signal path. It's a pain in the
neck to operate, though.

Build yourrself a conventional super-regen receiver, then add a q-multiplier
stage after you have the rest of it working.
--scott


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

On Mon, 05 Nov 2007 21:01:09 +0100, Antonio Vernucci wrote:

I would like to build an old-style receiver using a BC-453 stuck at 500-kHz as
IF chain. A possibility would be to put a pentode in front of the BC-453 mixing
the incoming RF signal with the signal generated by a variable local oscillator
(VFO). To improve selectivity, one could adopt a regenerative arrangement
whereby part of the pentode plate signal is fed back into the grid (by inductive
coupling), and the cathode resistor is then adjusted just before the tube starts
oscillating. In other words, a kind of Q-multiplying converter (I think this is
called "Q-dyne" receiver). What I am not fully sure about is if the increased
selectivity I so obtain turns into a higher rejection of the image frequency, or
just into a narrowing of the received bandwidth (which is already narrowed down
by the tight BC-453 85-kHz IF transformers). My feeling is that said
regenerative scheme would offer no advantage in terms of image rejection, but I
would value very much your opinion on that subject.

Your feeling is correct -- regeneration in the IF will narrow the IF
response, but won't correct for images before the IF. If the following
receiver suffered from images you'd correct them, of course.

I've seen designs in old handbooks that used regeneration in the RF stage
to sharpen up the response there -- that would, indeed, help the image
rejection at the expense of intermodulation performance.

--
Tim Wescott
Control systems and communications consulting
http://www.wescottdesign.com

Need to learn how to apply control theory in your embedded system?
"Applied Control Theory for Embedded Systems" by Tim Wescott
Elsevier/Newnes, http://www.wescottdesign.com/actfes/actfes.html

Your feeling is correct -- regeneration in the IF will narrow the IF
response, but won't correct for images before the IF.

I agree, but in my case regeneration occurs in the converter stage, not in the
IF amplifier! That is reason why I am wondering about.

With a 500 kHz IF, I would have images only 1 MHz away. So I am worrying about
the image frequency.

73

Tony I0JX

"Antonio Vernucci" ) writes:
I would like to build an old-style receiver using a BC-453
stuck at 500-kHz as IF chain. A possibility would be to put
a pentode in front of the BC-453 mixing the incoming RF
signal with the signal generated by a variable local oscillator
(VFO). To improve selectivity, one could adopt a regenerative
arrangement whereby part of the pentode plate signal is fed back
into the grid (by inductive coupling), and the cathode resistor is
then adjusted just before the tube starts oscillating. In other
words, a kind of Q-multiplying converter (I think this is
called "Q-dyne" receiver). What I am not fully sure about is if
the increased selectivity I so obtain turns into a higher rejection
of the image frequency, or just into a narrowing of the received
bandwidth (which is already narrowed down by the tight BC-453 85-kHz
IF transformers). My feeling is that said regenerative scheme would
offer no advantage in terms of image rejection, but I would value
very much your opinion on that subject.

A) There's really no reason to "boost" the selectivity of the BC-453.
They were once seen as the ultimate in selectivy, at a time when
there were few options. If the IF is too wide, then you need to
adjust it.

B) The Q-Multiplier works by adding regeneration, but the prime
point is just before it oscillates. Which means that when you need
it to oscillate to provide a beat note, you lose control of selectivity.
And the proper point is so sharp that external factors will kick
the circuit into oscillation. Which is where the superregen came in,
keeping it at that crucial point of highest gain without the need for
crucial tuning (but incidentally bringing other problems into play).

C) If you have a self-oscillating converter, which seems like you
are talking about, note the regen to the point of oscillation is
at the oscillator's frequency. That's not going to do a thing on
the signal frequency, and it's not going to do a thing on the
IF frequency. A mixer by definition has a different frequency on
the input from the output, which does not make for proper regeneration.

Unless they were modified, most BC-453's were used in multiple conversion
schemes. Either tapped into the 455KHz IF of an existing receiver (and
if the existing receiver was single conversion, leaving its image problem
intact), or as a tuneable receiver with some sort of converter ahead of
it. The better ones used two stages of conversion, to get around the
image problem that arose with a high signal frequency dropping immediately
down to 455KHz.

Michael VE2BVW

Tim Wescott ) writes:

I've seen designs in old handbooks that used regeneration in the RF stage
to sharpen up the response there -- that would, indeed, help the image
rejection at the expense of intermodulation performance.

That was a weird period, because it seemed to only exist after there
were single conversion receivers with IFs in the HF range, and then
disappeared somewhat later.

Indeed, it seemed to all be tracked to a receiver using a 7360 as
the mixer, and I can't remember whether the Q-multiplier was in
that famous article in QST by Squires circa 1963 (and if so, it
would have been in that famous Squires-Sanders receiver of the same
period), or if the construction article in QST that seemed based
on the Squires article introduced it.

So for almost the next decade, others would put that front end
q-multiplier in their single conversion receivers with the 7360
mixer (and there was even a mosfet based version).

The secondary articles were clearly doing it because the first
article did.

But I've always wondered about the first article. Since yes,
the receiver used a tuned circuit into a mixer, so the receiver
did lack front end selectivity. But, the mixer was about the
best for the time, and the IF was in the HF range. So it was
far less likely to overload than many mixers of the period, and
image rejection was easy with the high IF.

Later, the notion of high IFs became more common, and nobody
particularly worried about image rejection, even though they
too were using single tuned circuits into the mixer.

So did Goodman toss in the front end Q-multiplier because
of a perceived or real need for extra front end selectivity?
Or did he simply want to play with the concept?

What never made sense was that he went for the front end
q-multiplier with it's added tube and added ability to overload,
when any other receiver would have simply added better front
end selectivity.

Michael VE2BVW

Antonio Vernucci wrote:
Your feeling is correct -- regeneration in the IF will narrow the IF
response, but won't correct for images before the IF.

I agree, but in my case regeneration occurs in the converter stage, not
in the IF amplifier! That is reason why I am wondering about.

With a 500 kHz IF, I would have images only 1 MHz away. So I am worrying
about the image frequency.

Well, is the regeneration feeding back energy at IF or at RF?

--

Tim Wescott
Wescott Design Services
http://www.wescottdesign.com

Do you need to implement control loops in software?
"Applied Control Theory for Embedded Systems" gives you just what it says.
See details at http://www.wescottdesign.com/actfes/actfes.html

Well, is the regeneration feeding back energy at IF or at RF?

It is feeding back IF energy into the grid where also the RF signal is applied

73

Tony I0JX

A) There's really no reason to "boost" the selectivity of the BC-453.
They were once seen as the ultimate in selectivy, at a time when
there were few options. If the IF is too wide, then you need to
adjust it.

Agreed. In fact I am looking for extra selectivity in terms of image rejection,
not in terms of smaller channel bandwidth. The BC-453 selectivity is enough for
me!

B) The Q-Multiplier works by adding regeneration, but the prime
point is just before it oscillates. Which means that when you need
it to oscillate to provide a beat note, you lose control of selectivity.
And the proper point is so sharp that external factors will kick
the circuit into oscillation. Which is where the superregen came in,
keeping it at that crucial point of highest gain without the need for
crucial tuning (but incidentally bringing other problems into play).

There is no need to push the regenerative converter into oscillation for the
purpose of generating the beat note. I can easily get the beat note from the
BC-453 built-in BFO.

C) If you have a self-oscillating converter, which seems like you
are talking about, note the regen to the point of oscillation is
at the oscillator's frequency. That's not going to do a thing on
the signal frequency, and it's not going to do a thing on the
IF frequency. A mixer by definition has a different frequency on
the input from the output, which does not make for proper regeneration.

As I wrote in my initial post, I am supposing to have a separate local
oscillator that feeds the conversion frequency into the converter stage.
Therefore the converter is not a self-oscillating one. Nevertheless the
converter is regenerative because its plate coil is linked to its grid coil, so
that it would tend to self-oscillate if one does not properly control the
cathode resistance. Such circuit is called Q-dyne, see QST 1938

.. Unless they were modified, most BC-453's were used in multiple conversion
schemes. Either tapped into the 455KHz IF of an existing receiver (and
if the existing receiver was single conversion, leaving its image problem
intact), or as a tuneable receiver with some sort of converter ahead of
it. The better ones used two stages of conversion, to get around the
image problem that arose with a high signal frequency dropping immediately
down to 455KHz.

I know, but my purpose was to get good image rejection with just a single stage
(too easy with two!) and I was wondering whether the Q-dyne approach would do
the job.

73

Tony I0JX

The BC-453 was used to sharpen IF selectivity on superheterodyne
receivers by tuning around the passband of the ~455 "kc" IF frequencies.
It was known as a "Q-Fiver," and added a second conversion stage down
to the 85 kc IF of the command set. Why not play with regeneration at
the BC-453 IF level, if you think you need it? (The regeneration scheme
works--I used it on several old receivers, and a WW II tank transceiver.)

If you are worried about Images with your mixer scheme, you may want to
look into some RF selection/amplification before the mixer. Back in the
Fifties and Sixties, one-tube PreSelector projects were popular,
(several good receivers, such as the Drake 2B, had them built in). The
extra tuned circuit should reduce your image.

--W9NPI

Antonio Vernucci wrote:
I would like to build an old-style receiver using a BC-453 stuck at
500-kHz as IF chain. A possibility would be to put a pentode in front of
the BC-453 mixing the incoming RF signal with the signal generated by a
variable local oscillator (VFO). To improve selectivity, one could adopt
a regenerative arrangement whereby part of the pentode plate signal is
fed back into the grid (by inductive coupling), and the cathode resistor
is then adjusted just before the tube starts oscillating. In other
words, a kind of Q-multiplying converter (I think this is called
"Q-dyne" receiver). What I am not fully sure about is if the increased
selectivity I so obtain turns into a higher rejection of the image
frequency, or just into a narrowing of the received bandwidth (which is
already narrowed down by the tight BC-453 85-kHz IF transformers). My
feeling is that said regenerative scheme would offer no advantage in
terms of image rejection, but I would value very much your opinion on
that subject.

73

Tony I0JX