regenerative sets and selectivity
John Marvin Zelle
A recent Popular Electronics article described a regenerative
shortwave receiver using a dual-gate MOSFET for the RF oscillator. It
is well-known that the regenerative approach significantly increases
the Q of the resonant circuit. I was wondering, however, if the Q is
sufficient to provide the selectivity needed on the international
broadcast bands. Can such a receiver really separate relatively
powerful signals spaced every 5 Khz? I would appreciate hearing from
anyone who has tried such a circuit for shortwave reception. Is the
selectivity as good or better than superhets using a simple ceramic
filter?
As a side note, I am interested in building a small, portable, SW
receiver, and the simplicity of this approach (filtering via a single
tuned circuit) is quite appealing. Any other ideas regarding the
building of a small, battery-efficient receiver would be appreciated.
--John
--
*** John Zelle e-mail: ze...@cs.utexas.edu ***
*** Taylor Hall 2.124 motto : I'd rather write programs which ***
*** University of Texas write programs than waste my ***
*** Austin, TX 78712-1188 time writing programs ***
John Lundgren
as the superheterodyne receivers. That's why the superhet became so
popular. And now most good receivers are double conversion, so they're
even better. No matter how high the Q, a single tuned circuit can't
have as sharp skirts as several stages of IFs. And then we haven't even
talked about ceramic IF filters.
--
=====================================================================
| John Lundgren - Elec Tech - Info Tech Svcs | Standard |
| Rancho Santiago Community College District | disclaim- |
| 17th St. at Bristol \ Santa Ana, CA 92706 | ers apply.|
| jlun...@pop.rancho.cc.ca.us\jlu...@eis.calstate.edu | |
=====================================================================
Cecil_...@ccm.ch.intel.com
John Lundgren <jlun...@news.kn.PacBell.COM> wrote:
>No matter how high the Q, a single tuned circuit can't
>have as sharp skirts as several stages of IFs.
Hi John, I use a direct conversion receiver whose local oscillator is
a DDS circuit controlled by my PC. The audio output goes to my sound-
board for DSP-type processing. The biggest problem is images but the PC
lock-steps the DDS circuit to minimize the problem. I (blush) use this
system for CW only and it works pretty well.
In a super-regen circuit, the skirt selectivity problem can be overcome
by using DSP techniques on the audio. However, I don't know any easy
way to avoid the image problem. Does anybody else?
73, Cecil, KG7BK, OOTC (Not speaking for Intel)
--
Intel, Corp.
5000 W. Chandler Blvd.
Chandler, AZ 85226
James Speer
"Q-Multiplier" (a regenerative IF stage) to my not-so-hot single conversion
superhet and VASTLY improved selectivity.
My problems with regen rx's were always that they were insensitive, unstable,
and a damned nuisence, not that there were too broad.
--
If RST>519 Jim Speer, K5YUT
Then cut_power; f_sp...@ccsvax.sfasu.edu
Drew Diamond
>From: ze...@cs.utexas.edu (John Marvin Zelle)
>Subject: regenerative sets and selectivity
>Date: 10 Aug 1994 17:25:22 -0500
>
>A recent Popular Electronics article described a regenerative
>shortwave receiver using a dual-gate MOSFET for the RF oscillator. It
>is well-known that the regenerative approach significantly increases
>the Q of the resonant circuit. I was wondering, however, if the Q is
>sufficient to provide the selectivity needed on the international
>broadcast bands. Can such a receiver really separate relatively
>powerful signals spaced every 5 Khz? I would appreciate hearing from
>anyone who has tried such a circuit for shortwave reception. Is the
>selectivity as good or better than superhets using a simple ceramic
>filter?
>
>--John
>--
>*** John Zelle e-mail: ze...@cs.utexas.edu ***
>*** Taylor Hall 2.124 motto : I'd rather write programs which ***
>*** University of Texas write programs than waste my ***
>*** Austin, TX 78712-1188 time writing programs ***
John, from my experience of building regenerative sets, you can obtain quite
acceptable AM selectivity for SW BC work. The trick is to adjust the
regeneration to the point where the set "takes off", (you will hear a
gentle "plop") then back it off again slightly so that the detector just
stops oscillating. The better sets have an RF amplifier to isolate the
regenerative detector from the antenna (which can cause "dead spots, and
maybe radiate an interfering signal).
For SSB and CW reception, the detector is deliberatly put into the
oscillating condition, where once again, selectivity is quite adequate
for most conditions. SSB and CW signals sound remarkably clean by the way (a
bit like seeing the sky on a clear night away from the pollution of the
city-- signals stand out with a remarkable clarity because there's no
mushiness caused by multiple tuned circuits, AGC loops etc).
73, Drew, VK3XU. Telecom Australia Research Laboratories
Robert Wieland
>Subject: regenerative sets and selectivity
>Date: 10 Aug 1994 17:25:22 -0500
>
>A recent Popular Electronics article described a regenerative
>shortwave receiver using a dual-gate MOSFET for the RF oscillator. It
>is well-known that the regenerative approach significantly increases
>the Q of the resonant circuit. I was wondering, however, if the Q is
>sufficient to provide the selectivity needed on the international
>broadcast bands. Can such a receiver really separate relatively
>powerful signals spaced every 5 Khz? I would appreciate hearing from
>anyone who has tried such a circuit for shortwave reception. Is the
>selectivity as good or better than superhets using a simple ceramic
>filter?
>
>--John
fifties, when my uncle coached me through the assembly of a regen
using a single type 76 tube. I spent my high-school years
listening to a Knight Kit Ocean Hopper, and I currently have a
3-tube Heathkit I picked up at the GoodWill. So I'll say I'm
experienced with "jennies".
First, with jennies, everything gets worse as you go up in
frequency. With your RF tuned circuit giving you all your
selectivity, the "bandwidth" is a constant *percentage* of your
center frequency, and the absolute bandwidth (in Hertz, say)
increases in direct proportion with your operating frequency.
Additionally, your problems with hand capacitance, antenna
loading, drift, and "touchiness" of tuning all go up with
frequency. My experience agrees with the ham magazines of the
1930's: jennies are great at 4 MHz, good at 7 mHz, only fair at
14 MHz, and poor on 29 MHz.
But this also means things get better with decreasing
frequency. On the broadcast band, a jennie can be tuned to "cut
the sidebands", and give mushy audio; you actually have to back
off the regen control to let the sidebands through. At 455 KHz,
regenerative Q-multipliers gave IF strips responses like narrow
band (CW) crystal filters.
But that is mostly for CW work. The weakness of a jennie that
always got me on bands crowded with AM signals was that the
machine would produce perfect (interfering) audio from strong
stations on nearby frequencies. A jennie, necessarily, is an
envelope detector, so any AM that gets past the tuning circuit is
detected perfectly into AF, no matter what its frequency. So any
signal that was really strong would show up in the audio even
after the "Q-multiplied" tuning circuit was finished attenuating
it. And some of these signals can be *really* strong. In the
early sixties, I used to get Radio Moscow on 41 meters on a
crystal set... but crystal sets are another story. Needless to
say, on the jennie, Radio Moscow blanketed most of the 40 & 41
meter bands!
Jennies are still fun. They still the hold the record for
the most performance from the fewest components, and they need
no alignment or calibration; that's why they persisted as the
dominant homebuilt through the hard-time 1930's. But today
components are cheaper, and performance expectations are higher.
For AM work on SW, I believe a direct-conversion (DC) receiver
with some AF filtering would be the simplest receiver that might
satisfy you. Look around in the magazines, now that a balanced
mixer comes on a chip, very usable DC receiver projects have been
published, you can even buy a ready-made circuit board. No,
these are not SOTA, but they're not toys, either. A superhet
would be better still, and for AM, a ceramic filter would be a
big plus.
But me, I have been wanting to build another jennie for the
150-190 KHz "experimenter's" band. It will have all the
sensitivity I need, and atmospheric noise down there is so bad
that the few milliwatts of QRM radiated by a jennie will be
unnoticable. Jennies are at their best when relatively tightly
coupled to antennas that are electrically short (a small fraction
of a wavelength); other receiver input stages need to be rather
loosely coupled, preferably to resonant (reactanceless) antennas.
Well, on this band, all practical antennas are electrically
short... a potential problem for any receiver but a jennie.
Michael Silva
>In article <32c5hi$m...@ohlone.kn.PacBell.COM>, jlun...@news.kn.PacBell.COM (John Lundgren) writes:
>> Regenerative receivers never had and never will have selectivity as good
>> as the superheterodyne receivers. That's why the superhet became so
>> popular. And now most good receivers are double conversion, so they're
>> even better. No matter how high the Q, a single tuned circuit can't
>> have as sharp skirts as several stages of IFs. And then we haven't even
>> talked about ceramic IF filters.
>>
>>
>"Q-Multiplier" (a regenerative IF stage) to my not-so-hot single conversion
>superhet and VASTLY improved selectivity.
>
>My problems with regen rx's were always that they were insensitive, unstable,
>and a damned nuisence, not that there were too broad.
Yes, at 455kHz, but what happens at 15MHz?
>--
>If RST>519 Jim Speer, K5YUT
> Then cut_power; f_sp...@ccsvax.sfasu.edu
>
I actually heard somebody give a 119 last week. Maybe he had code
dyslexia and meant 449? (or even worse, 669?)
Mike, KK6GM
Michael J Dower
jlun...@news.kn.PacBell.COM "John Lundgren" writes:
> Regenerative receivers never had and never will have selectivity as good
> as the superheterodyne receivers. That's why the superhet became so
> popular. And now most good receivers are double conversion, so they're
> even better. No matter how high the Q, a single tuned circuit can't
> have as sharp skirts as several stages of IFs. And then we haven't even
> talked about ceramic IF filters.
Actually, that's something else I've been wondering about. Just how good are
ceramic IF filters these days, and what for centre frequencies are they
available (apart from 455 kHz)?
73's de VK2ENG (in G-land)
--
Michael J Dower
'Quoth the raven, "Never more".' ... Poe
John Lundgren
Michael J Dower (Mic...@arkas.demon.co.uk) wrote:
: In article <32c5hi$m...@ohlone.kn.PacBell.COM>
: jlun...@news.kn.PacBell.COM "John Lundgren" writes:
: Michael J Dower
10.7 MHz and 4.5 MHz (TV IF) should be readily available. I bought some
from Rat Shack several years ago, but they don't handle them any more. I
would guess just about any freq could be custom ordered, in quantity.
I've had some replies from regen receiver enthusiasts, seems like they
like 'em. But with the performance so good from relatively inexpensive
receivers nowadays, it would be hard to match them with a simple circuit,
no matter what it did. I ran across something recently in one of the
mags. It was a simple SW recvr using a NE602. I remember they were a hot
item when they first came out. Maybe a receiver using one would be
adequate and have acceptable performance.
Fred McKenzie
> In article <32c5hi$m...@ohlone.kn.PacBell.COM>
> jlun...@news.kn.PacBell.COM "John Lundgren" writes:
> > Regenerative receivers never had and never will have selectivity as good
> > as the superheterodyne receivers. That's why the superhet became so
>
> Actually, that's something else I've been wondering about. Just how good are
Michael & John-
My first Ham receiver was a Boy Scout Short Wave Set, which was a
regenerative receiver using plug-in coils, battery power and one or two
tubes. I made a number of contacts on 40 Meter CW using that receiver and
a one-tube (6L6) transmitter. Of course you're right about the
selectivity. However, what made the regenerative receiver so special, was
the amount of gain available without extra stages of amplification. The
trade-off was simplicity versus tuned stages.
In actuality, the term, regenerative, refers to the detector. I see no
reason why a regenerative detector couldn't be used in a superheterodyne
configuration, to take advantage of the selectivity. However, doing so
would offer no practical advantage over use of a BFO or product detector,
since other stations within the bandpass tend to "pull" the regeneration
frequency.
This discussion started with "regenerative sets", which were used for CW.
The "super-regenerative" receiver (detector) should also be considered in
the discussion. I never understood how they worked, other than
regeneration was increased far above the amount used in ordinary
regenerative receivers. I suspect they might be useable in a
superheterodyne configuration.
I've seen ceramic IF filters and discriminators used in less expensive
NBFM communications receivers, including much of the Two Meter Ham gear on
the market. Their advantage is that they require no tuning. The only
tuning problem might be incorrect impedance matching at either the input
or output of the filters, which could cause "ripple" or skew the shoulders
of the bandpass curve. They are usually considerably smaller than
equivalent LC filters, and cheaper than crystal filters.
One drawback of ceramic IF filters, is that the ultimate out-of-band
rejection may not be as good as can be achieved with other filters. It
may also have more insertion loss. However, insertion loss isn't usually
a problem with an FM system, where noise figure is determined in the front
end, and the detector is preceded by limiter amplifiers.
73, Fred, K4DII
Jeroen Belleman
Fred McKenzie <fred-m...@ksc.nasa.gov> wrote:
>The "super-regenerative" receiver (detector) should also be considered in
These things have always puzzled me too.
As my current understanding goes, an oscillator tuned to the
approximate frequency of reception is periodically stopped and then
allowed to restart at a rate of, say, 40KHz. If some received
energy at the right frequency is coupled into the resonant
circuit, the restart tends to be quicker, with a noticable effect
on the current drawn by the oscillator.
The current drawn by the oscillator contains the output signal.
Is this how it works?
Best regards,
Jeroen Belleman
David J Bell
>>the discussion. I never understood how they worked,...
>These things have always puzzled me too.
>As my current understanding goes, an oscillator tuned to the
>approximate frequency of reception is periodically stopped and then
>allowed to restart at a rate of, say, 40KHz. If some received
>energy at the right frequency is coupled into the resonant
>circuit, the restart tends to be quicker, with a noticable effect
>on the current drawn by the oscillator.
>The current drawn by the oscillator contains the output signal.
>Is this how it works?
>Best regards,
>Jeroen Belleman
Not to my recollection... I missed earlier portions of this
thread, but as I understood it, the regenerative circuit was
an attempt to boost gain of a receiver stage by adding positive
feedback, up to *almost* the point of oscillation. If this
point was exceeded, the receiver would become a transmitter,
and all received signal would be swamped by the oscillator's
current. The super-regen was a design that eliminated the
critical hand-tuning of the regen (which had been by tuning
the feedback control for each received signal) by allowing
the stage to go into oscillation, which (mumble, mumble)
increased grid current, which built up a DC voltage by
charging a cap across a multi-meg grid-leak resistor, which
reduced the gain, driving it out of oscillation... The effect
was similar to a sawtooth relaxation oscillator controlling
the gain (via grid bias) of the amplifier stage. The operating
points were chosen so that the stage 'hovered' about the critical
gain level, getting a voltage gain of thousands(?) from a single
amplifier stage.
Or, something to that effect... :{)
Wayne Covington
> approximate frequency of reception is periodically stopped and then
> allowed to restart at a rate of, say, 40KHz. If some received
> energy at the right frequency is coupled into the resonant
> circuit, the restart tends to be quicker, with a noticable effect
> on the current drawn by the oscillator.
> The current drawn by the oscillator contains the output signal.
> Is this how it works?
This is what I remember. These were used mostly on VHF? Maybe ancient
versions of the ARRL Handbook have good explanations of what is going
on.
Wayne
Kevin White
That is pretty accurate. Usually the circuit can be made to be self quenching
by utilizing the base emitter junction to charge up a suitable time constant
RC network so that the transistor is biased off and oscillation stops.
They are commonly used for garage door opener receivers, burglar alarms
X-10 RF remote control in the 300 MHz range, they can be very simple and
low power but fairly sensitive (1 microvolt).
My first radio control receiver I built 30 years ago used this technology
which was originally developed in the 30's.
They are very broadly tuned, and have a very high noise output on no-signal.
They are also very tricky to design and get going. I find it helps if the
transistor is fairly close to its maximum frequency, otherwise it oscillates
too readily and is not very sensitive.
kevin white
Dave Horsfall
wa...@fc.hp.com (Wayne Covington) writes:
| > As my current understanding goes, an oscillator tuned to the
| > approximate frequency of reception is periodically stopped and then
| > allowed to restart at a rate of, say, 40KHz. If some received
| > energy at the right frequency is coupled into the resonant
| > circuit, the restart tends to be quicker, with a noticable effect
| > on the current drawn by the oscillator.
| > Is this how it works?
|
| This is what I remember. These were used mostly on VHF? Maybe ancient
| versions of the ARRL Handbook have good explanations of what is going
| on.
Sure that's not a SUPER-regenerative receiver? "Super" meaning
super-sonic i.e. the chopping frequency.
--
Dave Horsfall (VK2KFU) | da...@esi.com.au | VK2KFU @ VK2AAB.NSW.AUS.OC | PGP 2.6
Opinions expressed are mine. | E7 FE 97 88 E5 02 3C AE 9C 8C 54 5B 9A D4 A0 CD
Wayne Covington
: In article <33gi2t$q...@tadpole.fc.hp.com>,
: wa...@fc.hp.com (Wayne Covington) writes:
: | > As my current understanding goes, an oscillator tuned to the
: | > approximate frequency of reception is periodically stopped and then
: | > allowed to restart at a rate of, say, 40KHz. If some received
: | > energy at the right frequency is coupled into the resonant
: | > circuit, the restart tends to be quicker, with a noticable effect
: | > on the current drawn by the oscillator.
: | > Is this how it works?
: |
: | This is what I remember. These were used mostly on VHF? Maybe ancient
: | versions of the ARRL Handbook have good explanations of what is going
: | on.
: Sure that's not a SUPER-regenerative receiver? "Super" meaning
: super-sonic i.e. the chopping frequency.
: --
Yes, the poster who was speculating about their mode of operation was
talking about SUPER.
Another poster has mentioned that superregenerative detectors are still
used in some garage door opener receivers. Apparently they don't radiate
enough energy to exceed FCC or CISPR limits.
Wayne
Michael Silva
come upon issues dating to the "dawn" of the superregen, so maybe I can
add a little background. The concept, invented by Armstrong, was
introduced to the world in 1922, and it was immediately heralded as the
receiver of the future, rendering current technology (regeneration)
obsolete. Most of the original circuits actually used a separate
oscillator to provide the switching signal to the received-frequency
oscillator (terminology my own!). Within a few months, the euphoria
was wearing off, and QST was publishing articles asking "how come our
superregens aren't any better?" Within a year they seem to have faded
from amateur view, although they did resurface with the move into 10,
5 and 2.5 meters in the late '20s / early '30s.
For anyone interested in a great story on the early days of radio, let
me recommend "Empire of the Air" by Tom Lewis (I think, not in front of
me), published about two years ago.
Mike, KK6GM
Peter Simpson
receiver in this week's EDN (8/18/94). It's in the Design Ideas
section, page 98. Charles Kitchin of Analog Devices is the author.
---
Peter Simpson, KA1AXY Peter_...@3com.com
3Com Corporation (508) 836-1719 voice
Northborough, MA 01532 (508) 393-6934 fax
I speak only for myself, 3Com doesn't pay me to speak for them, so I don't.
A.R. Bicknell
Build an amplifier with a gain of mu. build a network (attenuator)
to connect the o/p to the i/p with a gain (loss) of beta. The amplifier
multiplies the signal mu times and the output is fed back to modify the
input beta times the output. The output is input*mu*beta. If beta is
negative (out of phase) then the bandwidth is extended, the noise is
increased, the distortion is decreased. (audio). If beta is positive the
bandwidth is decreased, the noise is decreased, and the distortion is
increased (rf). I highly recommend 'high Fidelity curcuit design'by Norman H.
Crowhurst and George Fletcher Cooper, Copyright Gernsback Library, Inc. 1956.
--
He who laughs last lives longer.
Bert Bicknell
Dave Newkirk (WJ1Z)
: By chance, I had access to some old QST's a while back and happened to
: come upon issues dating to the "dawn" of the superregen, so maybe I can
: add a little background. The concept, invented by Armstrong, was
: introduced to the world in 1922, and it was immediately heralded as the
: receiver of the future, rendering current technology (regeneration)
: obsolete. Most of the original circuits actually used a separate
: oscillator to provide the switching signal to the received-frequency
: oscillator (terminology my own!). Within a few months, the euphoria
: was wearing off, and QST was publishing articles asking "how come our
: superregens aren't any better?" Within a year they seem to have faded
: from amateur view, although they did resurface with the move into 10,
: 5 and 2.5 meters in the late '20s / early '30s.
[*Empire of the Air* cite deleted]
The "switching" oscillator provides the *quenching* frequency. We have a
quench oscillator breadboard attributed to Armstrong in the ARRL Museum.
The main basic "can't do" with the superregen is that it can do heterodyne
reception. That lets out CW and SSB demodulation--the reason (well, CW at
least :-)) it didn't last in the pre-VHF era. Its broad selectivity also
makes it largely useless for narrowband FM, so a superregen is pretty much
only an AM and wideband-FM detector.
It's good to see that relatively little traffic on this topic has dwelt long
on the issue of how much a regen or a superregen radiates. (Any text that
says something like "Don't use those detectors; they radiate" is another
case of the specific case, or implementation, being taken as the general
case. An RF-amplifierless "direct-conversion" receiver, and likewise an
RF-amplifierless superhet with a poorly balanced mixer and insufficient
input selectivity, will also radiate, sometimes quite strongly.) A
reasonably unliteral RF amplifier stage, along with proper shielding, can
take care of detector or mixer radiation in any of these cases. Input
filtering (in superhets) also helps.
Data point: The 40-meter regen I published in Sep 1992 *QST* radiates--at a
level of roughly 12 picowatts into a 50-ohm antenna load.
Regards,
David Newkirk, WJ1Z
Senior Assistant Technical Editor, *QST*
Brent G. DeWitt
: Michael Silva (mjs...@ted.win.net) wrote:
: Regards,
In general, regens DO radiate! The FCC requires that they be CERTIFIED
with the FCC before commercial sale, while super-hets only require
notification. None of this means a thing as an experimenter building one
item for his/her use, but it does imply that history has shown regens to
be pretty good transmitters......
--
Brent G DeWitt (bde...@tuvps.com)
TUV Product Service
Boulder, CO USA
phone (303) 449-4165 fax (303) 449-3004
T J LeMense (Tom)
: Dave Newkirk (WJ1Z) (dnew...@arrl.org) wrote:
: : It's good to see that relatively little traffic on this topic has dwelt long
: : on the issue of how much a regen or a superregen radiates. (Any text that
: : says something like "Don't use those detectors; they radiate" is another
: : case of the specific case, or implementation, being taken as the general
Hear Hear! I cannot agree more. Sure, there are lots of examples of crude
superregens that radiate like banshees, but they are just that--crude
receivers. Superregens are superior to 'hets in terms of sensitivity vs.
cost and simplicity, IMHO. Same sort of baseless statements as: MOSFET's
are better than bipolar transistors, RISC is better than CISC, and Coke is
better than Pepsi. Of course, the last is a TRUE statement, but I digress..:-)
: : case. An RF-amplifierless "direct-conversion" receiver, and likewise an
: : RF-amplifierless superhet with a poorly balanced mixer and insufficient
: : input selectivity, will also radiate, sometimes quite strongly.) A
: : reasonably unliteral RF amplifier stage, along with proper shielding, can
: : take care of detector or mixer radiation in any of these cases. Input
: : filtering (in superhets) also helps.
Very good points indeed. Having direct experience with designing both
types of receivers, I completely agree. A superregenerative transistor
oscillator usually runs with about 200 uA of collector current, while the
LO in most wireless-type superhet designs consumes at least a couple mA.
When both circuits are preceeded by a well designed amplifier with a
decent amount of reverse isolation, say 40 dB or so, the radiated emissions
will meet US FCC, Canadian DOC, and the various european regs.
: : Data point: The 40-meter regen I published in Sep 1992 *QST* radiates--at a
: : level of roughly 12 picowatts into a 50-ohm antenna load.
Another data point, the 315 MHz superregenerative recevier inside the upcoming
Explorer's Remote Keyless Entry system is about 3-4 dB off the noise floor
at 3 meters. I would consider both my design and David's to be "state of
the art" in terms of superregen receivers.
: In general, regens DO radiate! The FCC requires that they be CERTIFIED
: with the FCC before commercial sale, while super-hets only require
: notification. None of this means a thing as an experimenter building one
I don't consider that the FCC only requires notification for 'hets to mean
that they are superior from a re-radiation standpoint! Remember, the FCC
had to mold and shape the broadcast world that we now live in, and their
plan was to make the receivers cheap, simple, and easy to market. The
rule for Notification for the superhet's simply makes it easy to market,
and takes them off of the hook of having to tightly regulate all the mfgr's
of broadcast sets. The economics and business reasons for this rule are,
IMHO, more significant than any technical reasons.
: item for his/her use, but it does imply that history has shown regens to
: be pretty good transmitters......
Seeing that history has spent much more effort improving the superhet, as it
supports both FM and AM broadcast, that is not that hard to believe. I have
a very thorough collection of superregen articles and papers, and they fill
about 6" of file cabinet space. I wouldn't even attempt to collect all the
papers and patents re: superhets.
The reason for the regen's bad rap is that in thermionic times (that's
toobs to you and me) an RF stage was an extra capacitor gang, an extra tube,
and more W of filament power--real product cost/benefit considerations.
Now, mass-produced superregens are relegated to wireless control receviers,
a job which they do very well, but ultra cost sensitive -- an RF stage adds
components, board space, current draw, etc. -- again, real cost/benefit
tradeoffs. As wireless products become more complex, you will see them move
away from superregens to other topologies that support more complex & compact
modulation methods, but only when costs allow and functionality demands it.
But of course, I am biased as I design superregens for a living... I
would be happy to correspond with anyone wanting to discuss the merits and
pitfalls of this simple, sensitive, and cost effective receiver...
--
Tom LeMense
RF Product Design Engineer
Ford Motor Co - Electronics Division
tlem...@rchp33.eld.ford.com
'82 XV750 Virago DoD #1074
Alan Bloom
: : It's good to see that relatively little traffic on this topic has dwelt long
: : on the issue of how much a regen or a superregen radiates. (Any text that
: : says something like "Don't use those detectors; they radiate" is another
: : case of the specific case, or implementation, being taken as the general
: : case. An RF-amplifierless "direct-conversion" receiver, and likewise an
: : RF-amplifierless superhet with a poorly balanced mixer and insufficient
: : input selectivity, will also radiate, sometimes quite strongly.) A
: : reasonably unliteral RF amplifier stage, along with proper shielding, can
: : take care of detector or mixer radiation in any of these cases. Input
: : filtering (in superhets) also helps.
: : Data point: The 40-meter regen I published in Sep 1992 *QST* radiates--at a
: : level of roughly 12 picowatts into a 50-ohm antenna load.
: In general, regens DO radiate! The FCC requires that they be CERTIFIED
: with the FCC before commercial sale, while super-hets only require
: notification. None of this means a thing as an experimenter building one
: item for his/her use, but it does imply that history has shown regens to
: be pretty good transmitters......
People generally use regenerative detectors to reduce circuit complexity,
which implies that the rest of the circuit should be simple too. For that
reason, most regen's lack an RF amplifier in front of the detector, so they
radiate like crazy.
However, as Dave mentions, it is quite possible to eliminate the problem
by including an RF amplifier with good reverse isolation. Another way
is to use the regenerative detector at an IF frequency, as part of a
superhet receiver: Most mixers have enough isolation to eliminate any
radiation problem.
AL N1AL
Pete Rossi
>Brent G. DeWitt (bde...@csn.org) wrote:
>: Dave Newkirk (WJ1Z) (dnew...@arrl.org) wrote:
>
>: In general, regens DO radiate! The FCC requires that they be CERTIFIED
>: with the FCC before commercial sale, while super-hets only require
>: notification. None of this means a thing as an experimenter building one
>
>I don't consider that the FCC only requires notification for 'hets to mean
>that they are superior from a re-radiation standpoint! Remember, the FCC
>had to mold and shape the broadcast world that we now live in, and their
>plan was to make the receivers cheap, simple, and easy to market. The
>rule for Notification for the superhet's simply makes it easy to market,
>and takes them off of the hook of having to tightly regulate all the mfgr's
>of broadcast sets. The economics and business reasons for this rule are,
>IMHO, more significant than any technical reasons.
>
>: item for his/her use, but it does imply that history has shown regens to
>: be pretty good transmitters......
Superhets radiate too!! I discovered recently while trying to listen to
a repeater on 444.0 MHz I kept hearing various carriers and noises come
and go as I drove around. They sounded very "local" since often they were
full quieting but they would have lots of very deep fades then disappear a
half mile down the road... then come back... then disappear.
Turns out what I was hearing was the local oscillators in other car radios
tuned to a popular local FM station on 100.3 MHz
100.3 + 10.7 = 111.0 MHz LO frequency x 4th harmonic = 444.0 MHz
With some car radios (mine) you can even hear some modulation from the
FM station way way way down in the noise.
If you have 440 mobile, pick the 440 frequency that matches a local FM
station and see how many car radios you can hear while you are driving
around ;-) You will be amazed! Some of those suckers really put out
some signal!
FM STATION LO LO 4th harmonic
99.50 110.20 440.80
99.70 110.40 441.60
99.90 110.60 442.40
100.10 110.80 443.20
100.30 111.00 444.00
100.50 111.20 444.80
100.70 111.40 445.60
100.90 111.60 446.40
101.10 111.80 447.20
101.30 112.00 448.00
101.50 112.20 448.80
101.70 112.40 449.60
=================================================================
Pete Rossi - WA3NNA ro...@vfl.paramax.COM
Unisys Corporation - Government Systems Group
Valley Forge Engineering Center - Paoli, Pennsylvania
=================================================================
Bill Newcomb
Pete Rossi <ro...@VFL.Paramax.COM> wrote:
>Superhets radiate too!! I discovered recently while trying to listen to
>a repeater on 444.0 MHz I kept hearing various carriers and noises come
>and go as I drove around. They sounded very "local" since often they were
>full quieting but they would have lots of very deep fades then disappear a
>half mile down the road... then come back... then disappear.
>
>Turns out what I was hearing was the local oscillators in other car radios
>tuned to a popular local FM station on 100.3 MHz
I found out last weekend that the mysterious audio noise marker
generator that I had been hearing on 2 m was actually my CD-boom box!
Every 90 kHz a nice bit of noise. Anyone know what it might be?
servo-positioner for the laser diode, maybe?
Bill "TVI, nuthin!"
--
Bill Newcomb "Name them."
nu...@netcom.com -Dilbert
Dave Newkirk (WJ1Z)
: In general, regens DO radiate! The FCC requires that they be CERTIFIED
: with the FCC before commercial sale, while super-hets only require
: notification. None of this means a thing as an experimenter building one
: item for his/her use, but it does imply that history has shown regens to
: be pretty good transmitters......
I just took at look at Part 15 in the most recent CFR 47 I could get my
hands on rapidly -- 1988 -- and although I again see that although FCC
requires superregnerative receivers to be certified -- superregenerative
receivers for the CB band, for 30-901 MHz, and for 935-940 MHz, that is
[don't have the book right in front of me, so I may have one or two of these
ranges slightly bollixed] -- I can't find anything about *regenerative*
receivers. I'd appreciate receiving information to the contrary.
John Wells
: I found out last weekend that the mysterious audio noise marker
: generator that I had been hearing on 2 m was actually my CD-boom box!
: Every 90 kHz a nice bit of noise. Anyone know what it might be?
: servo-positioner for the laser diode, maybe?
: Bill "TVI, nuthin!"
Most likely it is a bit of the clock that is used for the serial input
D/A converter. The base frequency is 44.1 (I think) so that would a 2x
oversample clock of 88.2 kihz. Most likely a bit of coperclad carefully
placed around the D/A converter area would fix it up fine.... No guarentees
by the way ;-)
John WA0LHB in the Fort