OT: Bowerbird mating dance

And on similar topic... Has anyone ever heard of building a magnetron for 10MHz? Magnetrons are apparently based on a ring of little GHz tank circu its, no actual 1/4-lambda cavities in there. So, very weird magnetron freq uencies might be feasable. (I'll use this to attract the female.)

The "original" non-cavity magnetron tube used an external resonator, which was often an LC tank.

I have some training textbooks for WWII Navy techs. They mention the non-cavity magnetron as a low-power microwave generator, as an impractical curiosity. I think that was to throw off the Germans and Japanese that might nab some books.

--

John Larkin         Highland Technology, Inc 
picosecond timing   precision measurement  

jlarkin att highlandtechnology dott com 
http://www.highlandtechnology.com

No no, I mean microwave oven magnetrons. Their cavities are not "cavities. "

Take one apart. They're not based on quarter-wave resonators (which is wha t, 4cm at 2GHz?) Each "cavity" in oven magnetrons is about 1cm diameter. I was surprised by this when I finally took one apart. Too small to be res onant cavities, so they're nearfield components. Might they be behaving as LC resonators for GHz: a conductive ring with a series capacitor; a narrow gap in a 1-turn inductance?

Of course if I try to build a 10MHz version with weak magnets, external tan k coils, and 100nS electron transit rather than 0.5nS, it probably won't pu t out any kilowatt for 2KV plate voltage. Snail corp "Food saver" polycarb vacuum chamber, 1C epoxy feed throughs, with a turbo and an Edwards pump. On the other hand, a couple of 812 tubes pushpull power oscillator works well, but without the fancy e-beam switching magnets stuff.

Trivia: I just saw a very old article saying that the Magnetron inventors h ad been trying to make a negative-resistance oscillator for VHF. They actu ally seem to have done so. The Magnetron's filament-choke creates a consta nt-current source as far as RF is concerned. I think we get a negative-R e ffect as two segments interact, one stealing the constant current from the other, as with those weird, dual-anode "Negatron" tubes of the 1920s.

Doh, the bird mating-dance joke should have been started at 2:36, like so:

I wonder why the power would be low. Maybe it gets terrifically hot, while a compact "cavity" device could avoid the huge losses.

((((((((((((((((((((((( ( ( (o) ) ) ))))))))))))))))))))))) William J. Beaty Research Engineer beaty a chem washington edu UW Chem Dept, Bagley Hall RM74 billb a eskimo com Box 351700, Seattle, WA 98195-1700 ext. 3-6195

I'm not sure what available tube you could harness to produce a swirling electron flow past resonators.

You could perhaps harness a heptode (a real one, without the g2-g4 and k-g5 ties) by putting resonators on each grid, thus making a lumped element, beam-coupled* "TWT".

*With respect to the grid voltages and currents: tubes are multiply non-reciprocal devices. Of course s21 >> s12 because of forward gain and backwards shielding (assuming screen grid or better), but there's also s13 != s31 because the electron beam intensity influences grids as it passes by, without allowing the same influence in the backward direction. For a 6BE6 heptode, the effect is on the order of 100umho g1 to g3, so this can have real impact on LO leakage from an AM radio set, for example.

The effect is used to advantage in FM quadrature detectors, where g3 is wired to a tuned circuit only; it gets excited by space charge, and generates a phase shift which in turn controls the current to the plate. Thus, detecting FM, quite handily.

This was further refined in the 6BN6 and similar tubes, which are "gated sheet beam" types. The transfer curve (plate current vs. g3 voltage) is unusually sharp, going from ~off to ~on over about 3V of grid swing (where "on" is proportional to g1's current, thus allowing one to construct a single-quadrant PWM analog multiplier), with little additional change (no significant change in plate current for higher Vg3, or additional cutoff or other funny business for lower Vg3) up to 30V of swing. In FM IF/det. lingo, it makes a fine limiter, to the tune of 20dB input amplitude range.

Tim

--
Seven Transistor Labs, LLC 
Electrical Engineering Consultation and Contract Design 
Website: http://seventransistorlabs.com

You do not need so many grids. The old TPTG (tuned plate tuned grid) oscillator runs with a triode. It was also accidentally re-invented when building an IF amplifier.

--

-TV

Are those birds good eatin'?