EMRFD Message Archive 3219

Message Date From Subject
3219 2009-06-19 08:02:11 vicker40 Re; Tunnel Diodes
Hello Fellas,

Indeed, I would enjoy sharing notes about tunnel diode projects with
N7FKI, Bob!

Art, KY1K, I believe you are correct in saying nobody has made a
tw0-way, all-tunnel diode QSO to date; at least a "skywave" QSO.
However, Seab, AA1MY and I have completed two QSOs on 80m at a distance
of 160km using an all-tunnel diode station on this end.


Mikums is an 80m transceiver using a tunnel diode as the sole active


There was tremendous enthusiasm for tunnel diodes in the early 60's;
mostly on account of their high-speed capability. However, people soon
realized this virtue was accompanied by a number of faults which I won't
bother to enumerate here. On the bright side, tunnel diodes seemed to
encourage folks to "think outside the box." As a result, some very
interesting circuits appeared in the early to mid 1960's.

By coincidence, I spent yesterday afternoon in the shack working on a
new, all-tunnel diode transmitter. Monarch (named after the "DX"
butterfly) uses a pair of 1N3720 (20mA peak-current) diodes in a
push-pull, synchronous oscillator having an output power of 1.5mW. I
realize that 1.5mW doesn't sound like much, but it's actually a huge
amount of power for Germanium tunnel diodes working at HF. A crystal
oscillator built from a 1N3712 (1mA peak-current) device synchronizes
the push-pull "power oscillator" frequency.

I hope to first use Monarch in an all-tunnel diode, 30m, QRSS beacon.
My hope is that its tiny signal might flutter across the Atlantic Ocean.

I ought to mention the push-pull pair works as a semi-relaxation
oscillator. As Van der Pol pointed out quite a long time ago, the I/V
signal excursion limits are determined, in part, by the L/C ratio. A
relaxation oscillator, for example, uses a large L/C ratio in order to
work well past the peak and valley current points. For maximum power
production it's best to build relaxation oscillators; even though they
produce more harmonics. Even after filtering, you'll end up with more
energy at your working frequency.

The oscillation limit-cycle decreases along with the L/C ratio,
producing a progressively smaller, but "cleaner" output signal. The
pretty sinusoid disappears altogether at a sufficiently low L/C ratio.

In another recent "just for grins" project, I used a cascade of two
synchronous tunnel diode oscillators (x9 -> x12) to multiply a 32,768Hz
wristwatch piezo-tuning fork up to 3538.7kHz. This project was inspired
by a tunnel diode quartz-crystal chronometer that appeared in the early
1960's. It was described in a number of places at that time, one of
which was Carl Strong's wonderful "Amateur Scientist" column in
Scientific American Magazine.

As for where to find tunnel diodes; I purchased my pair of 1N3720's from
Brad, AA1IP. Brad seems to have a nose for finding interesting
components. What's more, he's not a "scalper." He knew full-well that
his 1N3720's were incredibly rare, but he still charged me less than
four bucks a piece for them. Brad went so far as to characterize my two
diodes on his curve tracer before he shipped them. If you drop him an
email I''m sure he'd be happy to send you his current list of

Otherwise, check the usual places. A year, or so, ago I found ten N.O.S.
tunnel diodes stuffed in an old cardboard box at the Deerfield, NH
hamfest. Most of the tunnel diodes made by GE were produced in a
distinctive, bullet-style package that was typically gold, and less
often silver-plated. Bear in mind, there were tunnel diodes made with
peak-currents of several amperes once-upon-a-time. Granted, their
junction capacitance was enormous, but if I ever come across one at a
hamfest I'll likely need to be defibrillated.

In any case, I'm happy to see you guys talking about the likes of tunnel
diodes and parametric amplifiers in an age when one can buy a radio on a
chip. Obviously, I feel there's still plenty of ore left that might be
mined from these old devices and plenty of flags left to plant with
them. They're also just plain fun to tinker with.

Happy tunnel diode hunting/homebrewing,
Mike, AA1TJ

[Non-text portions of this message have been removed]
3221 2009-06-19 13:04:01 ai9e_qrp Re: Re; Tunnel Diodes
If you haven't already, you should check out Nyle Steiner's website

He is making zinc-oxide point-contact tunnel diodes from galvanized steel! He also has made alcohol-flame triodes, TEA lasers, copper oxide semiconductors, electrolytic rectifiers, and so on. Be forewarned, you can easily spend an afternoon reading his stuff.

73 - Dave NM0S