RCA - Signals / Waveforms etc - "Real" .

I'd assume that the clara-min arrangement is somehow superior- stability, sensitivity etc- (at least on a #27 tube) .  It is lev's masterpiece and I doubt that he would change his design arbitrarily.  

"It is lev's masterpiece and I doubt that he would change his design arbitrarily." - Chobbs

I agree!

My guesses are that the tank components and antenna components (ratio of turns, capacitances etc) increase the sensitivity, and simultaniously cause greater loading on the tank (this increased the number of playable octaves as requested by Clara)..

Lev probably switched the topology to account for the greater loading - Primary loading was now on the upper (driven) coil - If this loading had been on the lower coil, I think Lev saw the possibility that, close to antenna resonance, the oscillator could stall (because the feedback to the grid would drop in amplitude)..

Not knowing tube technology well, this is only a guess - But as there is no 'down side' to connecting the antenna circuit to the anode rather than the grid, I suspect that Lev (in looking for ways to improve the grid-antenna topology) realised this and went for it.

The funny thing is that the grid connected topology is the one which jumps into the brain first, regardless of whether one is designing with transistors or fets or whatever - It is the most obvious, and the one described in Lev's patents. First look at the anode connected topology when I saw the anode-antenna connection, I thought Bob must have drawn it wrong! .. But a few moments later I realised it was obviously just as effective as grid connection..

I wonder if Lev had the same kind of "Eurika" moment when he thought of it ?

Fred.

"First look at the anode connected topology when I saw the anode-antenna connection, I thought Bob must have drawn it wrong! .. But a few moments later I realised it was obviously just as effective as grid connection.."  - FredM

If I'm following you correctly: the EQ coil connection point can be either the tank drive or tank sense point? 

If so, I noticed something similar a while back when playing around with (what I now call) mode 2 oscillators.  The Pi tank was about equally sensitive to delta capacitance on the drive and sense sides, which kind of surprised me at the time.  Using the drive side might be better because external world perturbers perhaps get filtered before being used as oscillator feedback?

Just stumbled again over this "old" thread... Has it gone to eternal sleep or is there hope to see more waveforms of Lev's designs (be it RCA or others)?

Best would be "calibrated" pics of the oscilloscope which indicate also DC levels because there are still some things which I'd like to understand. None of the audio-side tubes are biased, thus these positive pulses at the grid of V7 will most probably lead to grid current which risks to drive V7 into saturation. I am really interested in seeing the V7 plate voltage when the 106 speaker is connected because of the eventual interaction of saturation effects with the complex load.

Another point would be the V7 grid at different audio frequencies. From my mental simulations and conclusions these peaks should have a "flat roof" at lower frequencies and I'd like to see that confirmed...

Thank you in advance!

I would love to see actual oscilloscope traces of the cyclic waveforms (the pictures posted previously look like frequency spectrum displays, not waveforms.) I'd like to try to reproduce the waveform with an analog synthesizer (and then play the synthesizer with a theremin.) I suspect it's a triangle wave with some extra overtones, but I'd like to see how it changes with the pitch.

"I would love to see actual oscilloscope traces of the cyclic waveforms " - Joe

So would I  !!! ;-) A full set at didfferent frequencies and different volume levels, from every possible point in the theremin... But even a set of good  snapshots from the output would be great!

"I'd like to try to reproduce the waveform with an analog synthesizer" -

I do not believe you can get a good replication using subtractive synthesis.. I think you need additive synthesis where one is able to control each harmonic as a function of the audio frequency.

"I suspect it's a triangle wave with some extra overtones, but I'd like to see how it changes with the pitch." -

It changes a lot with pitch, and not in a simple manner (as in, not just simple roll-off functions). The waveforms contain both odd and even harmonics, and the best way I have found to get an idea of their complexity is by analysing the waveforms from some of Howard Mossman's recordings - I will find the links and post them here ASAP.

some here: http://www.thereminvox.com/filemanager/list/19/index.html

And this: http://www.thereminvox.com/filemanager/download/78/Jungle%20Lullaby.mp3

and some waveforms: http://rcatheremin.com/tone.php

Not a triangle wave! ;-)

Fred