Theremin based on Induction Balance vs. BFO?

Hi, I understand that Theremins are built on the Heterodyne BFO principle between a fixed and 'influenced' RF oscillator.  One related application of this is metal detection - the other related detector scheme uses Induction Balance where two oscillators are electromagnetically couple and setup in a way that they null each other out - bring a ferrous object near them and they fall out of balance and create the error/detection signal.  

My question is if the Induction Balance principle could be applied to Theremins and if there would be any advantages?

Thanks, Elie

Hi Elie,

As you note, metal detectors operate by disturbing one or both inductors with metal, with the LC tank capacitor sized and situated to avoid capacitive pickup from the environment.  Theremins are the opposite, with the variable C exposed to the environment.  So I suppose for something like induction balancing to work in a Theremin, the C's of both tanks would be exposed via twin pitch antennas, with the player placing his/her hand between them.  It could work I suppose, but there would be strong interaction, and I'm not sure anyone is super interested in things that don't look and play like traditional Theremins.

Thanks Dewster, I just came across the IB system and thought it might have some related uses.  Otherwise, I've been wanting to build a proper theremin for sometime, last I did something was a very basic loopstick based one which I breadboarded back in high school. 

RF stuff is less familiar to me than say, Analog circuits so I'm hoping to find something I can DIY without the big bucks required for something like a Moog Etherwave with Midi etc.  I'm even thinking that Vacuum Tube based designs might have some advantages of sensitivity due to their inherent high impedance etc. but I'm not familiar enough with RF to know if that's a fact or not.

On the more modern side, the Moog Theremini is more like a consumerized version so I'm guessing it's a little more dumbed down and less sophisticated as far as sensitivity, linearity and range.  Otherwise, I can imagine the nice thing of a digital interface is that you can translate the native response to something more practical.

"I'm even thinking that Vacuum Tube based designs might have some advantages of sensitivity due to their inherent high impedance etc. but I'm not familiar enough with RF to know if that's a fact or not."  - keyman2

No, for high Q, you either drive a series tank with very low impedance, or a parallel tank with high impedance (current source or small capacitance).  Sensing can also be done either low or high impedance.  But the "sensitivity" you're probably thinking of, for an analog Theremin anyway, is mainly high voltage swing at the antenna and high Q (these things tend to be linked).  And possibly linearization. These things give you a larger useful pitch field (genuinely something to behold).  Analog Theremins often have quite a bit of capacitive padding, which actually drastically lowers absolute sensitivity, but perhaps stabilizes the oscillators (?) and sets the audio pitch range.  Real, absolute sensitivity has almost nothing to do with note spacing, which is pretty much fixed on simple analog Theremins.

"On the more modern side, the Moog Theremini is more like a consumerized version so I'm guessing it's a little more dumbed down and less sophisticated as far as sensitivity, linearity and range.  Otherwise, I can imagine the nice thing of a digital interface is that you can translate the native response to something more practical."

The Theremini has low voltage swing at the antenna, and makes up for this with very low bandwidth gestural response (a classic filtering tradeoff).  The one I bought (original firmware) was also quite non-linear, with no easy way to adjust this.  I thought it was kind of neat, and made a great conversation piece, but I ended up selling it as it lacked basic controls and the response was sluggish. Bob's dead and there's no one left at Moog who understands Theremins like he did.  End of an era.

Okay, so I think the high voltage seems to play a part in providing a more useable 'pitch field' (as you put it), which translates to better control (yes?).  Does the 'Q' translate to a sensitivity peak in the 'pitch field'? (assuming that makes sense), whereby low q spreads out the pitch and vice versa?  Just attempting to parlay my own related knowledge in case that works. ; )  I understand the 'padding' to mean a higher ratio of capacitance within the system vs. what your body would contribute, that would explain the stability, but perhaps it means less sensitivity?

That's the impression I get, the newer consumerized theremins are more 'gimmick' than professional instrument - okay if you want to make noises or limited note range, but not up to par with the etherwaves etc.

There was a Burns unit here a summer or two ago, but again, it seemed a compromise....

"Okay, so I think the high voltage seems to play a part in providing a more useable 'pitch field' (as you put it), which translates to better control (yes?).  Does the 'Q' translate to a sensitivity peak in the 'pitch field'? (assuming that makes sense), whereby low q spreads out the pitch and vice versa?  Just attempting to parlay my own related knowledge in case that works. ; )  I understand the 'padding' to mean a higher ratio of capacitance within the system vs. what your body would contribute, that would explain the stability, but perhaps it means less sensitivity?" - keyman2

High Q gives you higher voltage swing with less energy input, and gives you better selectivity (noise rejection).  A large voltage swamps external interferers, and the selectivity tunes more of them out.  Less energy in means less heating, so less drift.  It's all win.  High Q is also associated with oscillator stability and accuracy - quartz crystals have crazy sky high Q.  High Q gives you a massive phase gain around the LC resonance frequency, which makes the region more well defined.

C padding is often right at the antenna, which can be orders of magnitude larger than the delta hand capacitance of roughly 1pF, or the static capacitance of the antenna of roughly 10pF.  Low frequency analog Theremins can possibly employ the C to give a certain playing range I suppose, but it is a voltage killer too.  Beyond that I've never quite gotten why there's sometimes something like 220pF at the antenna of some Theremins.  Though this is a field many seem to dabble in.

Adding parallel C at the antenna will alter the playing range in terms of what note will play when your hand is say 12" away, but it won't alter the number of octaves.  This is fairly counter-intuitive, and has to do with the C response fit to the ear's log pitch response.

"That's the impression I get, the newer consumerized theremins are more 'gimmick' than professional instrument - okay if you want to make noises or limited note range, but not up to par with the etherwaves etc."

Pretty much.  Kinda sad, because at this point everyone could have a fantastic Theremin for like $150 or so.  There isn't that much to them.  The Theremini is an OK synth with fairly shoddy C sensors, too bad it isn't the other way around.

"There was a Burns unit here a summer or two ago, but again, it seemed a compromise...."

IIRC Thierry stated the Burns uses RC oscillators.  Ugh, no wonder they drift.  I've looked into a lot of approaches and LC is the only way to go.  Even the spread spectrum approach uses LC (with low-ish Q to work over a larger range of frequencies).