Is this how it should be tuned?

I recently bought a Moog Etherwave Plus. While tuning the L5 and L6 inductors inside, I messed up and completely lost the output (one of the inductors was way off). I turned both of them several turns until I found the signal again. Then I did the tuning as explained in the manual.

Is the "absolute" tuning of L5 relevant?

I mean, the output is based on the interference between two HF oscilators, so the absolute tuning frequency should not matter, but only the difference, right? Somewhere on the net I saw a text that said that you should tune L5 so that the oscillator produces 289-290 kHz at one point in the circuit (I don't remember where but I can look it up). In my case, the tuning frequency is probably something quite different, not 289-290 kHz, but I do get a reasonable signal output.

Does the "absolute" tuning of L5 have any effect on the linearity of the theremin? In my case, the highest two octaves are clearly "compressed", i.e. I need smaller movements there. Is this normal, or can I retune L5 and L6 to make all octaves approximately the same "size"?

Also, when I touch the antenna, I get 3 kHz output, but as soon as I remove the finger (even if it's still very close) the frequency jumps abruptly down by 1/2-1 octave. Is this also normal?

The absolute position of L5 affects range and linearity That's why I always tell people to not to touch at it!!!).The 1 or 2 top octaves remain still somewhat compressed by design, absolute linearity is not possible with the E-Standard.

Your idea that it's only the difference of both oscillators which counts is basically right, but not for theremins with linearization coils. Thanks to these coils, you have a third resonant (series) tank circuit, formed by the big coils and the static capacitance of the pitch antenna. So when you change the frequency of the variable pitch oscillator with L5, you change also the relation between the oscillator's frequency and the (fixed) resonant frequency of the antenna circuit which affects range and linearity. Fine tuning of the antenna circuit is possible by bending the wire more or less far away from the aluminum foil which changes the static capacitance by several hundreds of picoFarads. Bending it towards the foil should normally stretch the field somewhat in the higher octaves.

Absolute tuning rules cannot be given since component tolerances and environment factors tend to be more important compared to the sensitivity of the antennae towards your hands. That's why classical tuning instructions as for a radio set ("connect frequency counter here, adjust for 287.344 kHz there") cannot be given. Fine tuning a theremin so that it meets your personal requirements on linearity and range in an optimal manner needs years of experience. Having tuned my Etherwave for about 100 times (after several modifications on the oscillators, lin-coils, mixer) and having tuned about 30 Etherwaves for friends and clients, among them renowned thereminists like Carolina Eyck, Thorwald Jorgensen, Wilco Botermans, Gordon Charlton, I know what I'm speaking about.

Thanks a lot (and nice to get a comment from you who are obviously very experienced in tuning)!

Actually, today I did the tuning all over again. I made notes for myself to be able to repeat the procedure if needed. Please check if you agree :)

1. Remove wooden cover. (This will alter the apparent tuning, but I compensate for it as described below.)

2. Set the Pitch control to its center position.

3. Hold antenna with one hand, and adjust L6 with the other hand until zero beat is reached.

4. Rotate L6 clockwise until the audio frequency rises from zero to approximately 4.2 kHz.

5. Let go of antenna, and pull hand away slowly until it's about 9 cm from the antenna (as far as the rightmost edge of the front panel of the instrument is from the antenna). Does the audio frequency then go down continuously to zero or some very low frequency?

6a. If yes, tuning is finished.
6b. If zero beat is reached before 9 cm, and frequency starts rising again, rotate L5 counterclockwise and start again from step 3.
6c. If zero beat is not reached at all up to 9 cm from the antenna, rotate L5 clockwise and start again from step 3.

7. Replace wooden cover. This "stretches" the intervals so that zero beat will now be reached a lot further away from the antenna than 9 cm as used above.

With this tuning method I was able to get the result I wanted.

NOTE: Step 4 differs from what the manual is saying! This makes me wonder. If I do as in the manual, i.e. rotate L6 COUNTERCLOCKWISE after finding zero beat, my tuning ends up being reversed so that pulling my hand away from the antenna will RAISE the frequency (with no zero beat in between)! (Is this a typo in the manual?)

Also, I found today that the 3 kHz recommended in the manual isn't high enough. When I use 4.2 kHz instead, I get an usable range of up to C at approx. 2.1 kHz, which I can play with my fingers still a couple of cm from the antenna, and the upper octaves aren't so compressed anymore. If I tune for 3 kHz as described in the manual, I find that I must keep my fingers very (read: inconveniently) close to the antenna to reach C at 2.1 kHz.

Comments?

The tuning method described in the manual is very basic, but it has the advantage that it will however work.

With this basic method you may get a 90% result in 10% of the time compared with more sophisticated methods which is not too bad.

When you tune it for about 3kHz when holding the pitch antenna, you match the range of older Etherwave theremins which had other linearization coils and thus were designed to have a playable range of only 2.5 octaves above middle C ~1.620kHz.

Tuning it for 4 to even 5kHz (the latter gives still a little headroom above C ~2.160kHz up to Eb or E) is ok with newer production runs where the windings of the linearization coils are divided by three in order to reduce parasitic capacitance.

Do not go higher although it seems possible with this method, you risk that a too high resonance current through the linearization coils will virtually shorten the oscillator's tank circuit and overheat the transistors to death, especially when there is a low antenna capacitance, i.e. when you walk away some meters from the instrument.

There is no typo in the manual, it just hadn't been updated when the adjustable coils (L5, L6 and L11) where changed since 2003/2004:

Older Etherwaves have Japanese coils whose cores can be turned with a normal (plastic) screwdriver and which react in the classical way: Turning clockwise increased the inductance and the frequency got lower.

Newer Etherwaves have Chinese coils whose cores require a hexagonal tool and whose reaction is reversed: Turning clockwise decreases the inductance and thus raises the frequency.

Very interesting about the change of coils!

However, I'm more confused now... if the direction of L6 changed as a result of new kinds of coils, then why are the directions I need to turn L5 still according to the manual? (In my instrument both L5 and L6 are of the same type.)

Also, in both L5 and L6, the ferrite core sinks deeper into the coil as you turn clockwise, which means that the inductance should increase when turning clockwise. Isn't this the same as in the old coils...? Looks more like a typo to me.

Also, your text above indicates that pushing the pitch antenna wire further down would stretch the upper octaves somewhat. So to get an even more linear performance, I should push it further down (it's not straight, but already now it's bent towards the ground plane below it) and then again tune to 4.2-5 kHz when touching the antenna. Correct?

[i]However, I'm more confused now... if the direction of L6 changed as a result of new kinds of coils, then why are the directions I need to turn L5 still according to the manual? (In my instrument both L5 and L6 are of the same type.)[/i]

Nothing to be confused. The manual is not very precise concerning L5 - both directions may be correct, depending on which side of the resonance curve you start from.

[i]Also, in both L5 and L6, the ferrite core sinks deeper into the coil as you turn clockwise, which means that the inductance should increase when turning clockwise. Isn't this the same as in the old coils...? Looks more like a typo to me.[/i]

No, the windings are on top, so lowering the core's position will decrease inductance. The older coils had pot cores which surrounded the windings partially.

[i]Also, your text above indicates that pushing the pitch antenna wire further down would stretch the upper octaves somewhat. So to get an even more linear performance, I should push it further down (it's not straight, but already now it's bent towards the ground plane below it) and then again tune to 4.2-5 kHz when touching the antenna. Correct?[/i]

Correct, but don't expect too much... An Etherwave Plus is basically an Etherwave Standard with additional circuitry only on the AF side of the mixer while the oscillator circuits are unchanged, so you'll never get a perfect linearity. The linearization coils have tolerances of +/- 10% and are not adjustable which might be much more important that the correction of some fractions of a picoFarad which you can obtain with the wire. The Etherwave Standard/Plus is and remains a cheap entry level instrument.

Thanks again for great information!

OK, is there any point in trying to adjust L5 to the other side of the resonance and tune again? Would that perhaps even improve linearity? When you tune theremins of this particular kind, have you found any difference depending on from which direction you tune L5?

I also see the point about the Etherwave Plus not being particularly fancy as regards optimization. Would it be possible to get a completely linear pitch response from the Etherwave Pro? If only it were still in production... :(

Or is ANY theremin at all capable of "almost perfect" linearity?

[i]OK, is there any point in trying to adjust L5 to the other side of the resonance and tune again? Would that perhaps even improve linearity? When you tune theremins of this particular kind, have you found any difference depending on from which direction you tune L5?[/i]

No difference, the two points where you get the result fall finally together.

[i]I also see the point about the Etherwave Plus not being particularly fancy as regards optimization. Would it be possible to get a completely linear pitch response from the Etherwave Pro? If only it were still in production... :([/i]

The EPro can be still more linear when well tuned (which is much more complicated). But you pay a price for that:
- limited pitch range of "only" 5 octaves
- extending the range with a register switch which allows to transpose the 5 octaves by 1 or 2 octaves down.
- The latter is done by digital frequency division with simple CMOS circuits which requires that the beautiful oscillator waves are cut down to simple square signals. These are responsible for the rather synthetic and less organic timbres of the EPro. The advantage of this digital waveshaping is that you don't hear the distortion generated by the dynamic coupling between the oscillators at lower audio frequencies.

[i]Or is ANY theremin at all capable of "almost perfect" linearity?[/i]

No. Simple Theremins without linearization coils require a completely different design of the oscillator's tank circuits with another L/C ratio. That's why you can't linearize a simple theremin just by adding a coil. There was a EPE design where they tried it and the theremin could be tuned more linear with an octave distance of about 4cm - LOL!

There are physical limits which are the more obvious the more you want to extend the pitch range towards higher frequencies.

Now I feel enough pestered with questions and I suggest you that you learn to play your theremin first. After several years of practice you will find that slight unlinearities will not prevent you from playing in pitch. Take the example of Lydia Kavina: I've seen her in front of home-tinkered non-linear theremins which she never played before. She tried a few tones and jumps in order to feel the pitch field, turned somewhat the pitch knob and started to play in tune. Practice and experience is more worth than a mathematically adjusted antenna circuit. The tVox Tour theremin on which Lydia plays most time has no linearization coils at all...

[i]Thierry says: "The tVox Tour theremin on which Lydia plays most time has no linearization coils at all..."[/i]

The above is true - and I agree with almost everything you say, Thierry.. I have seen Lydia play the most abysmal theremins perfectly.

However, linearization coils are not the only way to linearize a theremin.. it is the oldest way, and technically perhaps the simplest way - but it is not the only way .. and in my opinion it is probably not the best way, to achieve good linearization.

I believe (from discussions with its designer - who does not give much information away, LOL ;-) that the Tvox linearization is mainly achieved by [b]tuned[/b]coupling between the
variable and reference oscillators - As you know, variable and reference oscillators are often deliberately coupled (in the case of the EW, this is done crudely via C2 and C6) causing the oscillators to 'pull' each other into lock as the frequencies get closer.. My understanding is that with the Tvox, the coupling is via some form of filter which causes the oscillators to interact over a wider span and therebye linearize the response over a wider area..

The above is, however, mostly speculation on my part - based on a few 'clues' slipped to me - I MAY be wrong! .. However it is done, the Tvox is, in my opinion, one of the best 'linear feeling' theremins I have ever played.. but I have not had a chance to measure how linear it actually is.

I did play about with tuned coupling, and it seems like a viable method to achieve linearization.. but I have my own way. One big advantage of the tuned coupling method would be that the waveshape would change continuously over the playing frequency range, due to oscillator interactions affecting waveshape.

Thanks for the extremely useful information you presented regarding the new EW coils.. It seems to me that, what with these changes and changes to the tuning circuit, it is time for an update to the EW "hotrodding" manual!

Fred

Thierry, I can understand your point about learning to play first... BUT before I start to practice and memorize hand positions, I want to be really sure that my instrument is as linear as it can be, otherwise I'm wasting time and making it harder than it needs to be :)

Also, my hands tend to tremble now and then (I'm not nearly as steady as the best ones out there), and therefore compressed high octaves lead to vibrato or inaccurate intonation where I don't want that to happen...