Crazy (?) theoretical / technical ideas

I think this idea is great mainly from an educational perspective - it clearly demonstrates the dramatic things that can happen through capacitive coupling, whether intentional or not.

I dont tend to use RF connectors (50-239 or the like) - my antennas on recent prototypes have used standard 6.35mm stereo jack sockets / plugs - particularly useful for directional antennas as they allow the antenna to be easily rotated.

The issue of antenna capacitance is an interesting one - its something I am playing with and recently changed my perspective on..

Large antenna capacitance is a real bother when one uses a series tuned scheme for linearization - and probably also if one uses a "tankless" series scheme - but its no bother for a directly connected parallel tank - with these one has a camparatively large tank capacitance anyway (usually 100pF or greater) and the antenna capacitance is simply in parallel with this - so even if the antenna components add up to 50pF, one can simply reduce the fixed tank capacitance if needed - A 455kHz IFT has 680uH || 180pF (454.9kHz) so a 1pF change (+1pF) gives 453.7 kHz, 1.2kHz/pF.. But its easy to increase the sensitivity, simply by adding a series inductor to the 680uH IFT - for example, a 5mH in series gives an adjustable 5680uH inductor, strap a 50pF across this (combined tank and antenna capacitance) and one gets an operating frequency 'round 295kHz, and a sensitivity of about 3kHz/pf.

The above gives no linearization - and its unusable in its raw form - but add electronic controlled capacitance, and one can 'squash' this excessive sensitivity and use it to implement linearization.

IMO, the 'bulk' antenna capacitance is only a bother with conventional theremin topologies - One gets into the habit of doing everything one can to minimise this, and avoids using screened cables etc because one usually gets into theremins via the conventional route - But I am starting to think that its a non-issue if your topology implements pure capacitive sensing, and isnt trying to be a musically linear sensor.. If the focus is on getting capacitive data which one processes and corrects further down line.

Fred.

Actually, some of what I said above is nonsense - The above facts apply equally to series and parallel tanks - they only dont apply when one uses seperate oscillator tank and antenna resonators and capacitively loads the antenna after the EQ inductor.. IMO, its this combination (Seperate tank and antenna resonant circuits) - which was invented for good reasons (linearity and antenna amplitude) which is IMO limiting for new designs.

I am really annoyed with myself for not seeing the issues above far sooner - My Skywave H1's were needlessly awkward to construct simply because I wanted the board close to the antenna, so ended up having to pipe power supply and audio wiring from the lower active loudspeaker (which contained the power supply and amplifier) up to the top speaker (which contained the theremin circuitry)

I could easily have had the theremin in the lower speaker with the other electronics, and simply run an RF shielded cable up to the antenna with its EQ inductor placed there - even if this cable had 40pF I could simply have reduced the tank capacitor by this amount.

"I dont tend to use RF connectors (50-239 or the like) - my antennas on recent prototypes have used standard 6.35mm stereo jack sockets / plugs - particularly useful for directional antennas as they allow the antenna to be easily rotated."  - FredM

The things I like about UHF connectors are: their capacitance relatively is low and fairly controlled, they are physically rugged, they lock down nicely in a variety of orientations, 90 degree elbows are available, they readily accept 3/8" tubing, they are 2 conductor (though 3 would be better for some applications), and no one can confuse them with audio connections.  They are a bit pricey, but I suppose most connectors are (even clunky old plumbing stuff can cost a fair amount).

"I could easily have had the theremin in the lower speaker with the other electronics, and simply run an RF shielded cable up to the antenna with its EQ inductor placed there - even if this cable had 40pF I could simply have reduced the tank capacitor by this amount."

(Warning, nitpicking ahead!)  Yes, but wouldn't the cable capacitance be of the "uncontrolled" (e.g. non NP0) variety?  Something you maybe wouldn't want as an integral part of your precisely tuned circuit?

LC tanks with large C and small L can be quite odd when coupled with large L small C resonators, so as you note there is the interesting opportunity to load the tank down with all sorts of capacitance and not have it negatively influence sensitivity very much.  I guess I haven't thought about this a lot, having come at things from a somewhat different direction, and I'm fairly reluctant to explore directions that are likely more difficult to manufacture and maintain (like most double tuned - LCLC - Theremin circuits). 

All kinds of unexpected nuance in fairly simple circuitry!

"and no one can confuse them with audio connections" - Dewster

Yeah - this is a perpetual "problem" - All sorts of connectors have pre-defined or "standard" functions - Becomes something of a pain when one wants to use them for other functions - For example, I want an expander socket on my theremins, and the ideal parts are designed for video - HDMI, SVGA, that kind of thing - the big advantage is one can buy good quality ready made leads with screened cables for RGB, Audio an other signals - ideal for piping buffered HF signals from one unit to the next..

All I can do is study the specs of the "standard" signals, and ensure that what I connect to the pins isnt going to damage or be damaged if some idiot wires the theremin to a PC or whatever, and add any extra protection circuitry required to ensure this.

At the end of the day, if the sockets are clearly marked "Not VGA" or "Not Audio" or whatever, then that should be enough.. After all, one doesnt expect some radio operator to go plugging their transmitter output into your 50-239 socket ;-)

Your points about  cable capacitance being "uncontrolled" is really valid - and for long cable lengths will certainly cause drift if not thermally compensated - My H1's however drifted like hell anyway - the front-ends were basically (badly) modified S/C oscillators with a series resonant antenna - I learned a lot from them by the mistakes I made, but if I hadnt been rushed and contracted to have them on the show, I would never have released them - They sounded wonderful, but that was all they had going for them... The environment was the worst possible - huge temperature and capacitance fluctuations as spotlights and equipment was moved around, and people crowded or vacated the small area - There was never any hope of stability.

"I guess I haven't thought about this a lot, having come at things from a somewhat different direction, and I'm fairly reluctant to explore directions that are likely more difficult to manufacture and maintain (like most double tuned - LCLC - Theremin circuits). "

Oh, I certainly dont think LCLC can be used with large capacitive antenna loadings -

Fred

"Oh, I certainly dont think LCLC can be used with large capacitive antenna loadings"  - FredM

I guess I meant the kind of thing in the EW: small L | big C tank, and big L EQ | small C antenna.  It could also mean a "lopsided" series tank I suppose, but I've never explored this.

I think the reason for the small L + Large C on the tank, compared to the small C large L on the resonator, is intrinsic in the equalization concept.. The antenna LC operates at a frequency where it effectively becomes a L in parallel with the tanks L - for this "virtual" L to be significant, the tank L must be small.

For the antenna resonator to work, it must be in its inductive zone at the operating frequency and over the entire range of antenna capacitances. I have not explored series-series tank/antenna schemes except for the Lev oscillator which is quite peculiar - with this oscillator I think the coupling capacitor (mid-way between the tank windings) allows the antenna resonator to operate in its inductive zone.

But my days of messing with these dual resonant topologies is past - or I certainly intend them to be past! ;-)  ... One oscillator with one resonant circuit, Fu8K the response / linearity etc, that can be taken care of by electronics I am at home with! ;-)

I will leave the high mathematics, the poles and asymptotes and complex resonance curves, and all that stuff to others - I have probably spent 6 solid months in the last 6 years stretching my mind to grasp these in a meaningful / useful way, but have little greater mathematical comprehension for this effort.. and a simple analogue curve shaper employing multipliers gives me adjustable linearity and span without needing to go near the front-end other than by inserting an electronically adjustable inductance  or capacitance in the tank.. I have learned just enough knowledge of this mathematical stuff now to keep me out of trouble and give me clues when things dont work as expected, and to know that its not an area I will ever master .

So I shall speak of maths no more, and leave that to TW's mathematician (who is one of the only people I know who can often present a mathematical idea in a way I can understand!)

Fred.

Everything is said here (scroll the page downwards). A musically satisfying linearization is only obtained for Lser/Lpar ratios of >= ~150. At the same time, the maximum pitch which can be played (at about 1cm from the antenna) can be approximated by 3/(16*pi*sqrt(Lser*Cpar)). That shows that the bigger Cpar, the smaller Lpar on one side which gives a better linearization and the more the pitch range will be limited which gives more stable behavior.

Let's take the example of the ultra-linear Etherwave Pro:

Lpar = 100uH
Cpar = 3.325nF (The free-run frequency of the VPO must be 276kHz without the series tank connected)
Lser = 35mH

The Lser/Lpar ratio is thus 350, and the approximated maximum pitch with the above formula around 5.5 kHz which corresponds to what can be found in reality.

"Everything is said here " - Thierry

Yes, it is - and extremely well said!

But thats it - Everything is fixed by Lpar, Cpar, Lser, Cant, and VPO Free-run frequency - sensitivity is fixed, linearity is fixed, and Cant is constrained.

Which is fine...

But if one wants adjustable sensitivity - as in, you want the playing zone to be linear and to be able to set the number of octaves / note spacing within this playing field, well, you cant - not without radical changes to component values.

The designer of a theremin using the above formulae must decide on one build - must define their priorities with regard to range (sensitivity) and linearity, and go for (what they deem to be) the best available compromise within the constraints imposed by the topology.

I hate being constrained! (especially by a straight jacket ;-) - I want to have a theremin that can be adjusted for 3 octaves over 60cm, or for 5 octaves over 70cm or whatever I want between say 3 and 7 octaves. I want linearity to be adjustable, so one can have a almost completely linear field (regardless of how many octaves you span) if you are a keyboard player - or a curve which compresses in an evenly distributed manner over the play zone towards the high end, if you are used to this behavour from acoustic instruments.

And I now believe it is possible to achieve the above - I believe its been possible for many years, and could have easily been implemented on the 91 series theremins, or any theremin employing voltage controlled audio oscillators even more easily than my method of voltage controlled HF oscillator.

And I dont believe one needs to mess with the antenna stuff at all - one can do it with a simple single (tank) resonator, either series or parallel - there is no need to implement linearity or sensitivity correction at the antenna side.

For me, its pure liberation! The maths reduces to simple stuff - all I need to compute is the dc correction curves and the mixing of these functions in a user interface - and thats not "maths" its standard analogue design.

But I havent done it yet - so I may yet find some unpleasant surprises! ... But all I can say is that, at last, I think I have a topology which fits right in the centre of my comfort zone! ;-) So damn annoying that I didnt think of this 5 years ago..

Fred.

 if it ever gets made, im calling it the Liberation I think..

Fred, on one side I understand your wish for freedom. On the other side I see that almost every traditional music instrument has its constraints. I'd never ask a luthier to modify a violin in a way that I could play Schuman's cello concerto on it, but rather play it on a cello.

Many instruments exist as a family: Strings from the violin to the double bass, brass from the trumpet to the tuba, reed from the oboe to the bassoon and so on. Every instrument of a family has not only its characteristic pitch range but there are also differences in the timbre characteristics and in playability which made that with the time, many virtuous works have been written with respect to the individual characteristics. Why then should just the theremin be a highly flexible all-in-one device for every purpose? Isn't it already difficult enough to play?

Can you imagine a grand piano which allows to adjust the width of the keys??? No, people have to learn to deal with the given standard width as others have to learn to deal with the non-linearity of whatever string (which is much worse than most theremins since the tone spacing is divided by 2 for every octave). They will also have to learn to deal with the particularities of a theremin. Who said that everybody should be able to play it with ease? Let it rather remain difficult to play in order to separate the wheat from the chaff...

There are standard dimensions for most traditional music instruments which have evolved over many, sometimes hundreds of years, with respect to the human anatomy and thanks to the constructive interaction between professional musicians and committed luthiers. We will hopefully see a similar evolution for the theremin one day.

"Can you imagine a grand piano which allows to adjust the width of the keys???" - Thierry

No - but I can imagine an electronic instrument with a keyboard that can be configured for microtonal intervals.

This I think is the core of our "disagreement" - I do not see the theremin in terms of being an acoustic instrument "equivalent" - Acoustic instruments, by their nature, are constrained - Some electronic instruments, mainly due to their player interface (keyboard or whatever) are also constrained.. Also, you speak of evolution - The theremin has not evolved - its mostly almost identical to the first prototype! .. Its had 80 years to evolve, not hundreds of years like the acoustic instruments you refer to.. IMO, its time it did start evolving! - NOW, not "one day".

80 years is a long time to stick with redundant technology when every other electronic musical instrument has evolved hugely in the same period... And thats what the theremin is, its an electronic musical instrument - not a violin or a cello, not a trombone or an acoustic piano - it does NOT fit in the class of acoustic instruments - it is, in fact, a synthesiser with a capacitive user interface...

Its not luthiers who design and build theremins, its engineers who do this! - A luthier is as useless for purely electronic musical instruments as an electronics engineer is for purely acoustic musical instruments! - The only thing we hopefully have in common is understanding of the physics of sound and biological limitations - Perhaps on the latter the luthier is likely to have superior understanding. Lev Theremin was an engineer not a luthier, Bob Moog was an engineer not a luthier.. its not luthiers who will advance the evolution of theremins!!

Where would music be today if the crude early instruments had not evolved, if the difficulty playing these instruments had been deliberately retained and enshrined in order to "seperate the wheat from the chaff" ? - And What a cruel idea this is, in my mind anyway - to deliberately deny people the pleasure of being able to play a musical instrument just to protect the interests of an elite few.. I would love it if everybody could play any instrument brilliantly - This would, I think, do more for peoples happyness than anything else ever could! (oh, sorry - I forgot.. not everyone would be happier - some of the elite few would have their noses put a bit out of joint ;-)

Yeah - I kind of wonder about statements like "Let it rather remain difficult to play in order to separate the wheat from the chaff..." - Im sure you dont really believe this or apply it - Are you tuning peoples theremins for optimum linearity in order to make them more difficult to play ? ;-) .. I think you are just saying that to be argumentitive! - or at least I hope this is the case, LOL!

The only constraint on the theremin has been imposed IMO by not-so-clever technology.. Oh, it was clever 80 years ago, but its not clever today! - The actual instrument can cover any frequency span, play any intervals, be a "theremin" or be a controller for any synthesis possible..

Lev's 3-4 octave RCAs are sought after both for musical and historic reasons - but if one wants a modern 3 octave theremin, where can one get one? Designers have generally chosen to opt for greater span in order to appeal to what consumers believe they want. So you cant get a grand piano with stretchable keys - but likewise, with a piano you are not constrained to 60 cm or so.. If you want a 3 octave keyboard, you can get one, or a 4, 5, 6, 7 or 8 octave keyboard - You dont have this choice with theremins - and there is no reason for this other than that the technology hasnt moved forward much in 80 years!

My thinking is that a newcomer starting with a 3 octave span might find playing easier - if they are able to increase the span as they become more competent, they wont need to upgrade.. With register switching, a 4 octave span will probably be well suited to most players - but if they have the ability to expand this field to 5, 6 or 7 octaves and choose to do this, its in their hands.

From a commercial angle a single build catering for all requirements is far saner than tying oneself to one set of specifications which may be unatractive to a fair percentage of the tiny potential market.

And if someone wants to emulate the conventional span and linearity of say an EW, they are free to do this - IMO, its all about giving the musician complete control.. As a certain coalminer here often states - us engineers dont know what musicians want (to which I reply that bloody musicians dont know what they want, so how can we know!) - my answer is to effectively give them an instrument they can easily set up exactly the way they  choose! - And this includes everything - pitch span and linearity, volume linearity and span, and much wider range of tones / waveforms.

I have a dream, LOL ;-)

Fred.

Oh, I know I am probably being an idiot commercially publicly exposing my thinking - particularly as someone is developing a 91 series clone and could easily incorporate my ideas in this - but I may not get anything to market - at least theres a chance that exposing these ideas will give theremin evolution a (IMO) badly needed kick in the (IMO) right direction..

ps .. I must just add that I think it most likely that theremin evolution will not go the way I am headed - Everything I have talked about is implementable or at this time nearly implementable with digital technology, and I believe it almost inevitable that future (certainly 10 years from now) new theremins will be primarily digital.

Now (after about the 20th edit) I go to bed - My schematic hasnt evolved in the last two hours (or more) - I need to stop clicking on TW or this theremin has no hope of moving from the dream world to reality.. I think I have said all I am going to say on this matter - been in this same discussion more times than I care to count.. Time for me to leave it, permanently.

Crazy idea #6966

Wire Theremin circuit ground to the microphone stand flange mount.  Then the microphone stand itself will form something of a ground network while supporting the Theremin.  Crazy?