Assorted questions

[i]" My understanding of a non-linear theremin is slightly different; yes, notes are closer together the higher you get (in the highest octaves), but also can get closer together the lower you get (in the lowest octaves) ... at least that's how I've understood it, but I'd be happy for someone with more expertise to either confirm or negate this!"[/i]

The natural non-linearity (the result if inverse square capacitance / distance relationship) is for notes to be compressed close to the antenna (high frequency) and stretch as distance increases..

Methods are employed to correct this non-linearity (most often by having an equalising coil in series with the antenna). Non-linearity at the low frequency end (particularly when distance between intervals get smaller as the audio frequency gets lower) is the result of balancing or tuning the equalization circuit incorrectly. Getting the balance between good linearity at the mid -> high frequencies, and maintaining good linearity in the low -> mid frequencies is one of the biggest challenges when designing or setting up a Theremin.

I believe (but may be wrong here) that the E-Pro linearity is achieved by careful positioning of the equalizing coils above deliberate capacitive pads on the circuit board.. the wire ended equalizing inductors can be moved manually to an optimum position (both vertically and horizontally) for fine tuning with the antenna.. I suspect that Moog have a jig of some kind which they use to optimally align these inductors during the assembly / testing phase..

But the above is only a guess based on some photos I have seen of the boards.

Fred Mundell
Fundamental Designs Ltd.
Electronics Consultant.
<- See Profile Image for Email.
Designer of Theremins and other alternative electronic music controllers and instruments.

Thank you, Fred - that's clear to me now. I didn't think I'd imagined that the very lowest notes on the E-Standard got gradually closer together again ... but I certainly didn't understand why they did that and didn't just keep getting wider apart.

Although (on each of the 3 range settings) the Epro's middle and upper octave seem to have perfectly spaced notes, the lower octave seems a little less reliable to me ... although this may well be down to my playing arc, and that I'm the less reliable part! However, I'm pretty sure I'm not imagining that the field needs to be retuned slightly after switching to a different range setting, especially the lowest one.

Have you seen these 4 photos of an E-Pro being assembled? Perhaps they might give you some clue ... (although only the last one shows a glimpse of the circuitboard).
-1- (http://www.flickr.com/photos/wobblymusic/32568176/sizes/l/) * -2- (http://www.flickr.com/photos/wobblymusic/32568177/sizes/l/in/photostream/) * -3- (http://www.flickr.com/photos/wobblymusic/32568180/sizes/l/in/photostream/) * -4- (http://www.flickr.com/photos/wobblymusic/32568182/sizes/o/in/photostream/)

[i]"Have you seen these 4 photos of an E-Pro being assembled?"[/i]

No, I hadn't ! Ye gods! Not at all what I expected to see - its no surprise to me that the E-Pro was discontinued looking at those photos - all the effort went into the housing! Beautifull, but ooohh so costly!

Not so sure about those inductors now.. You can see the track 'fingers' under them - my thought was that these were acting as capacitance between the wqindings and earth - capacitance which could be trimmed by slight re-positioning of the inductor.. But I now think I was probably wrong about this.

One other thing regarding low-end linearity.. The zone furthest from the antenna has incredibly tiny capacitance change for a given movement.. up near the antenna one can get say 1pf change for a given distance, but at 60cm the change can be less than 1/1000th of this (can get swamped by background capacitances) - The body also acts as a large 'background' capacitance - so as the hand gets close to the body some other strange non-linear stuff can happen. My maths is poor, but it seems to me that, particularly when dealing with the equalization resonance combined with the variable oscillator centre frequency, getting the tuning right to compensate for the (comparatively) major changes in capacitance close to the antenna, and cater for the incredibly tiny changes furthest from the antenna, and compensate for background capacitance, and then compensate the resulting curve to provide an exponential function ideally suited to give an audible output which sounds linear to our non-linear auditory physiology / psychology - well, quite honestly, it is astounding that we get such good results from such simple circuits!

Don't compare the EPro with other theremins! I've got some insight during my recent activities. One day if I'll have a few time I'll write an article about this.

Just so far: It is easier to create a EPro theremin circuit with "only" 5 linear octaves and a register switch later for virtually extending this to 7 octaves than building an EStandard, WFClassic or tVCox tour circuit which is expected to have 6 to 7 linear octaves without octave shifting... ;-)

Another article will soon come up which deals with low range linearization. There's a lot the player can do himself in order to improve this, but please be patient. Sometimes I happen to have to work for money and I have to prioritize this, not so much for me, but for my 4 daughters.

[i]"It is easier to create a EPro theremin circuit with "only" 5 linear octaves and a register switch later for virtually extending this to 7 octaves than building an EStandard, WFClassic or tVCox tour circuit which is expected to have 6 to 7 linear octaves without octave shifting" - Thierry [/i]

Thierry - I await your illumination of this subject with great excitement! .. As you know (from prior private correspondence) I abandoned the traditional approach to linearization more than a year ago.. I opt instead for a voltage corresponding to the variable oscillator frequency being fed back to the fixed oscillator (tuning voltage) in a proportion which linearizes the resultant difference frequency.. (This frequency either being used as audio output, or being converted to a CV - in which case the laws of preceding stages are somewhat different) Since then this basic idea has been expanded / honed, so that the feedback voltage undergoes transformation which can be adjusted to allow adjustable linearity (effectively an analogue exponential converter with adjustable law)..

The above is complex and quite expensive.. so if there is a simple / cheap way of obtaining linear response, this information would be hugely welcome - particularly if placed in the public domain so that good cheap Theremins could be made by everyone. I fully understand the commercial reasons that caused Moog not to publish the design / ideas.. But it is sad that something as important as this has been restricted to one, now discontinued Theremin.

However - while on this subject.. I did not personally like the 'feel' of the E-Pro when I played it for the first time on Monday.. It was undoubtably the most linear Theremin in the workshop, but I found it more difficult to play than the Tvox - I LOVED the Tvox, despite it having quite a bit of non-linearity, this non-linearity was evenly distributed over the playing zone.. I loved the full 7 playable octaves - It just felt absolutely 'right'.. The Tvox is, in my view, the instrument I now regard as the "Gold Standard".

I absolutely hated the EW Standard - could not get the notes I wanted, the field seemed squashed and irregular - just horrible! .. It may just have been a badly tuned EW, or it may be because I had just played the E-Pro.. But I was extremely glad to have had my lesson later on the Tvox.. If it had been on the EW Standard, I think I would have been embarassed by not being able to play anything!



[b]**** This thread is now continued [/b] Here (http://www.thereminworld.com/forum.asp?F=557&T=4110&cmd=p) [b] **** [/b]

Fred Mundell
Fundamental Designs Ltd.
Electronics Consultant.
<- See Profile Image for Email.
Designer of Theremins and other alternative electronic music controllers and instruments.

Thanks for your help Thereminstrel, and I'm glad I've helped trigger a bit of techy conversation! The Wavefront is definitively on my wish list now. Maybe when I win the lottery...



Going back to the very first reply, GordonC said Clara's theremin was one of the most linear theremins ever - is this really correct? To be a bit of a snob, I'd assumed that strong linearity (equal spacing) was in more expensive and modern development, easier to master than tradition weak linear theremins? I guess that theory goes out of the window!