First - Just in case anyone thinks I dug this ancient thread up - I DIDNT! - It wasnt me, gov! ;-) But as its here, and as my will is too weak to stay away from TW...

Let us, for sake of getting a "starting point", assume that the iron spring has SOME effect on performance - it will have, even if for no other reason than that by changing its tension one will change its capacitive relationship in much the same way as changing the length of a telescopic antenna changes its capacitance.

Now if we do even the most exagerated evaluation of the spring antenna, and try to find a mechanism by which it could operate, we (or at least I) do not find any - I certainly do not find any based on the capacitive mechanism by which I believe theremins work (and which is how Lev and Bob believed it works, and is an explanation which is backed up and entirely explained by known physics and predictable and verifiable with mathematics, and which, other than for strange unverified apparatus, is entirely duplicable).

Theremins such as the Theremax and EPE which are NOT designed for inductive antenna equalization (The EPE-2008 did incorporate an antenna equalizing coil, but this was simply tacked onto the SC front end and DOESNT WORK - simply because the tank is designed for direct capacitance sensing and was not altered to make it suitable for being driven by an antenna resonant circuit - see my addendum below) require a MUCH larger antenna inductance than could ever be provided by a spring antenna, and modified tank circuit.. On theremins which are designed for antenna equalization, the extra 50uH or so provided by the spring antenna is less than 1/100th of the required equalization inductance, even running at 1MHz.

So the only way that I can imagine the spring antenna achieving the claimed linearization, would be through some unknown mechanism -

But heres the problem..

The "The Lev Antenna" has nothing to do with Lev - Lev didnt use spring antennas, he used resonant antenna circuit with equalising inductance working into antenna capacitance, and he used a series resonant tank with the "virtual inductance" created by variation in capacitance seen by the antenna imposed across one half of the coupled (balanced) tank inductance - This topology has been misunderstood or ignored for decades, with people proclaiming that the Lev theremins used a parallel tank circuit in an Armstrong configuration. Almost every theremin not designed by Lev has copied this misconception, and been based on parallel tank topology.

Lev did not design high frequency theremins - He did not use spring antennas - And he got good linearity.

The statement "Until someone validates the Lev Antenna within the parameters I list above it would probably be frustrating to venture outside of them." Is the hub of the problem.. We would all love to have a simple method of getting perfect linearity.. But any theremin needs to have reasonably consistant performance. With (as I stated in my initial posting here) a given set of environmental factors (including the "right" placement of grounded surfaces etc) it is possible to get linearity from even the most basic theremin - One can even arrange an array of conductive objects and strategically place these to alter the shape of the field, to improve playing field linearity..

But the above is no good for a practical instrument - One needs an instrument which works with reasonable linearity at least, even in non-ideal situations.

As for "Until someone validates the Lev Antenna within the parameters I list" this begs the question as to why, with so many people having attempted to get linearity from your antenna, within your parameters and following your instructions, over so many years, no one has yet "validated" it.. I am not saying that you are being untruthful - I believe that you believe you have seen perfect linearity - but it seems that no-one has verified this yet.

Fred.

Addendum: Below is part of an email reply to a friend:

The question of *“**coils and how they interact with the antenna and their playability* *“* is a deep one – As you know, I am not into “dark arts” LOL ;-) – But the whole nature of the front-end interactions are about as close to “dark arts” as one gets in electronics..

The reason for this is (I think) that the primary mechanism is extremely complex, even if one has ideal components – the math’s is on the edge of my ability to deal with (in fact, I can only compute linearity by using a basic macro within excel, which iteratively seeks the solution – this is effectively running a minor simulation, and takes calculated capacitance values from the distance table and works out the inductive loading effect on the tank inductor caused by the position of the frequency with reference to the antenna resonant frequency – I am not at all confident that it is accurate) – The effects of coil SRF / Capacitance / couplings etc have a huge effect on real performance – sometimes beneficial, sometimes detrimental.. With crudely hand-wound coils, even though I have taken care to make them as similar as possible, I have never managed to wind two equalizing coils which behaved identically! - This is why I have now standardized on the Bourns / Miller 6300 series – They at least have repeatable results!

Tacking an equalization coil into a theremin which is not specifically designed for equalizing inductance, from my experience, never gives good results – and sometimes has disastrous results.. The most classic case of this was the EPE-2008 theremin – A SC theremin front-end was used virtually without modification, and the results were much like you describe – the “ *the weirdness that I felt between zero feild setting and the antenna might be translated as moving my hand through water rather than through air**”* is almost exactly what xxx said to me when I tried to linearize his xxx (based on the SC) xxx and I had a hell of a job linearizing my theremins which were at the Royal Festival Hall – They had S/C oscillators – I ended up modifying them during the show (I had 3 I had modified before the show at the last minute, but ran out of time to complete the others – As I see it, the primary difference between “Equalizer friendly” and “Equalizer unfriendly” designs is this:

An “Equalizer unfriendly” design (as in, direct-to-antenna friendly) tends to have a larger tank inductance with a smaller tank capacitance – the antenna capacitance is in parallel with the tank capacitance, and the oscillator frequency is simply determined by the sum of tank and antenna capacitances in parallel with the tank inductance… In the case of the EPE / SC theremins using a 455kHz IF Transformer, the tank inductance is 680uH and the tank capacitance (inside the IFT) is 180pF. It can be seen that a small change in the capacitance (due to antenna capacitance) has a significant effect on the oscillator frequency.

An “Equalizer friendly” design operates in a completely different way – One is not looking for change in capacitance across the tank capacitor (in fact, one wants to eliminate this) one is looking for a change in inductance across the tank inductor as the means of changing the oscillator frequency.

A correctly tuned antenna circuit represents an inductor connected to ground, the inductance (or “virtual inductance”) is controlled by the capacitance “seen” on the antenna.. As this inductance is in parallel with the tank inductance, when it decreases, the oscillator frequency increases – and visa-versa ..

The “Law” or “conversion factor” between the change in antenna capacitance and the change in total tank inductance is what imparts improved linearity – one can get an idea of this if one looks at the relationship between XL and XC on a standard frequency plot – one is linear, the other exponential – doing this strange equalizing trick one is effectively taking the linear curve and transmuting it to an exponential curve, and this exponential opposes the natural square law curve, resulting in an approximately exponential result.. And an exponential curve gives a musically linear response! … ;-) … Oh, I am sure some mathematician would have a field day tearing my explanation apart – but it works for me! ;-)

The “Equalizer friendly” design requires that the “virtual inductance” has a significant effect on the tank inductance, so the tank inductance is reduced (100uH in the case of the EW) and the tank capacitance is increased (3300pF in the case of the EW) – it can be seen that the tank capacitance is huge by comparison to the SC type designs – to modify a SC for operation with a linearizing coil, one needs (among other things) need to decrease the inductance an increase the capacitance..

The role of the tank capacitance in an “Equalizer friendly” design is to set the oscillator frequency and not to vary as a consequence of changes to antenna capacitance – changes to antenna capacitance should only alter the tank inductance by merit of the “virtual inductance” in parallel with it.

There is also a huge “multiplication factor” involved in the conversion from capacitance to inductance – this multiplication factor (of sensitivity) is a complex interaction of the antenna equalization components and tank component values.

Which brings us back to the “mish-mash” behavior where Equalizer and tank component values are not carefully selected – A number of strange (and sometimes destructive) things can happen –

1.) The performance is critically dependent on the operating point of the oscillator / Equalizer -Antenna tuning .. Most theremins tune the null point by changing the reference frequency, so the real Equalizer -Antenna tuning varies and is not constant..

2.) Close to resonance (hand distant from the antenna) high currents can flow, and the virtual inductance can reduce to extremely low values – this can shunt the tank inductance, cause the oscillator to stall, and even blow up transistors.

3.) Tuning so that antenna resonance is never close defeats the whole object, linearity returns to equal or worse than an un-equalized configuration, and sensitivity is reduced.

4.) Both antenna capacitance and virtual inductance can be “seen” by the tank, causing cranky operation.