In article ,
(Bruce Kizerian) writes:

A regen operates at the hairy transition of self-oscillation where
positive feedback yields a tremendous increase in gain. Doesn't
matter whether it is vacuum tube or transistor. The only difference
between remote-cutoff and sharp cutoff characteristics (transistors
of the bipolar junction type are very sharp cutoff equivalents to
tubes) would be on the amount of spurious garbage created when
the regen jumps into full oscillation.

But it does matter. In the heirarchy of regen devices tubes provide
the "smoothest" regeneration, followed by FETs, with bipolar
transistors generally taking a distant third. I am not asking because
I have never built a regen. I have built DOZENS of them, and I sell a
simple version on my website...but I'm always looking for a new
approach.

I know of no solid state equivalent (without some sort of AGC feedback
loop) for the remote cutoff pentode. Does anyone know if there is?

"AGC in a regen?"

Here, I was speaking in more general terms and not referring to
regenerative circuits.

If you are speaking in general terms then there are plenty of gain-
control elements out there. For a single-IC type of device, the
old Motorola MC1350 (8-pin DIP) is a sort-of Gilbert Cell arrangement
inside, differential-in, differntial-out, constant parallel Z-in of 5 KOhm
in parallel with a couple pFd each input. The gain control portion is
done by "starving" or actually redirecting the DC emitter current in the
input differential pair. Jameco (www.jameco.com) still sells this IC
at around $1.20 (?) in singles and has a copy of the Motorola data
for download on their website. That datasheet has a schematic of the
insides. That's good up into mid-VHF.

Some of the older JFETs had non-linear source-drain curves v. gate
voltage curves...see the difference in biasing for "depletion" and
"enhancement" mode operation. I don't know if they will work to
the top of HF range, though.


For experimentation purposes, a high gain-bandwidth product op-amp
IC might produce some interesting results. The gain-bandwidth (or
0 db open-loop gain frequency) of some op-amp ICs is up at 30 to
70 MHz now and the DC open-loop gain is enormous in comparison
to vacuum tubes. I sense possibilities of an op-am regen or even a
superregen on up through HF. Just a thought... :-)

The voltage gain of an effective regenerative stage is can be as high
as 100,000 as reported by Charles Kitchin. That's 100dB...not many op
amps have that kind of gain at say 10MHz.

I was suggesting using op-amp ICs WITH positive feedback. :-)

I'm just not going to buy "100 db voltage gain" in a regen based on
my own experiences...unless I see the bench layout and check the
calibration stickers on the test equipment determining this. :-)

Frankly, to get a simple receiver for HF, the Tayloe Mixer and its
separate LO followed by a low-noise AF range op-amp has the
most sensitivity for the fewest parts...and with little possibility of
re-radiating the oscillations of a regenerative due to a twitch of the
regeneration control. As a direct-conversion receiver, it can handle
CW or SSB and, with a stable LO, can take in conventional AM
with a lot less tweaking than a regen with a touchy regen control.

The most stable regen I ever built (three in all, the first using that
1T4) was with a 117N7 beam-power pentode and rectifier diode, AF
out driving an old, old high-impedance dynamic speaker. Plate curves
of that tube were of the sharp-cutoff variety. AM BC band, not much
else to listen to for entertainment in 1948. I suspect it was acting
more like a "plate detector" but it had the sensitivity of a common
"all-American-five" AM table top radio receiver.

Some of the minimal-tube-complement receiver designs of older times
used regenerative detectors at the IF for sensitivity improvement. I
know of one old Hallicrafters S-38 (?) reworked that way. That receiver
had essentially an "all-American-five" tube lineup and no power
transformer. Somewhat unsafe, but useable.

Nostalgia isn't what it used to be...

Len Anderson
retired (from regular hours) electronic engineer person