Does anybody know this theremin ?

It would be interesting to know more details of the technology inside.

When reading the german description, I felt that it has not been written by a German but by a russian or polish native speaker. These guys aren't always up to date, but often have good ideas.

I have talked to them on the phone.....
They are german allright.

...hmmm...I could help you talking to them, if you still need more information...

Btw. regarding the information behind Gordon's links labeled

-'English Text' is an overall description of the features (like the german one on eBay)
- 'German Text' is the complete manual

I'm also tempted to order a set of pitch-and-volume-theremins to try them out... ;o)

It seems that these FS Theremins are based on the "Glasgow Theremin" circuit - I may be wrong about this, so please dont take what I say here as verbatim.. My analysis is of this circuit, I am assuming the "Glasgow Theremin" is the basis of the FS Theremin, as "stated" in http://www.myspace.com/thereministry "[i]Fritz Soll who built it after the block diagram of the so-called Glasgow Theremin developed by students at California University[/i] " [This site also has the schematic for the Glasgow Theremin, and it is this schematic on which I base my following observations] - I DO NOT KNOW FOR SURE THAT THIS SCHEMATIC IS WHAT IS BEING USED IN THE FS THEREMINS, SO PLEASE TAKE WHAT IS SAID BELOW WITH DUE CAUTION - I DO NOT WISH TO PUT ANYONE OFF WHAT MIGHT BE A GOOD PRODUCT USING DIFFERENT CIRCUITRY!


1.) The pitch and volume antenna circuits are the crudest implementation possible - and consist of an RC oscillator.. Linearity of this type of circuit is notoriously bad when used directly with a fixed frequency reference oscillator to produce the beat frequency.

2.) ... However, this circuit does not directly produce the audio as a result of mixing two oscillators.. What it does is uses the crude beat waveforms, and (via monostable multivibrator to give a defined width pulse) converts the beat frequency to a voltage..

3.).. Converting beat frequency to voltage introduces a frequency dependant latency, and at low beat frequencies I would expect this to be quite long - in the order of perhaps hundreds of milliseconds, at higher frequencies this may reduce to about 40ms.

4.) As both volume and pitch circuits employ the above technique, the above latency will apply to both.

5.) Voltages generated by the above technique are applied to a reasonable quality Voltage Controlled Oscillator (the output of which is the pitch signal) on the Pitch unit, and to an extremely good quality voltage controlled amplifier on the volume unit.

I have not played with the above, and have not even done a full circuit simulation, but I expect the following:

A.) A pure sine output which is boring and tiresome in comparison to the sound from a conventional hetrodyning Theremin - In a 'real' Theremin, the harmonics change, albeit sometimes marginally, as capacitance seen by the antennas (and interaction between oscillators) change - The nature of this Theremin has no mechanism for any dynamic change in harmonics, intentional or unintentional.

B.) I see no mechanism which would make the pitch to position relationship musically linear - The relationship (I think) will be inverse square law - to get musical linearity, one requires an exponential law.

c.) I think both pitch and volume response will be extremely sluggish.

Update: looking over the translated text, the following can be confirmed.. A) Output harmonics are constant (sine wave) so a sine VCO (probably 8038) is being driven, as per the Glasgow Theremin. The sound will be boring and tiresome unless one really likes pure sine waves without ANY variation. B) description of Volume antenna response is compatable with crude RC oscillator / antenna system of th Glasgow Theremin..
However - the claimed linearity for Pitch response does not fit the Glasgow theremin, so if these claims are true, then the pitch->voltage circuit must be different from the Glasgow Theremin.

I would not recomend this Theremin - The Etherwave [b]will[/b] sound much more interesting and probably be better in every other respect as well..
The concept of a 'hybrid' driving a VCO is, in my opinion, a good one IF IMPLEMENTED COMPREHENSIVELY.. but to make an improvement on the standard Theremin design requires looking at the issues which can be (and may benifit from being) improved (linearity,sensitivity etc).. WITHOUT losing any essential features such as tone quality. Driving a VCO producing a harmonically rich waveform, and feeding this waveform into a VCF (Standard subtractive synthesis) is a way to achieve this.. Without a VCF, and without any harmonics for the VCF to remove, the sound is what one would get from a sine signal generator - Ok for a few minutes, but not for much longer!

Do any of the the modular synth guys make a heterodyning VCO module for "the genuine theremin sound".

As far as I know, I am the only one who has gone down this path.. I thought it was a brilliant idea at first - but actually now think it is probably folly..

A VCO produces an output frequency proportional to the control voltage (in fact, VCO's are usually current controlled oscillators, where the current is defined by the voltage through a feeing resistor).. It is easy enough to make a HF VCO, mix its waveform with a fixed frequency reference oscillator to produce beat frequency output... BUT .. This output will NOT change (harmonically) as the voltage (derived from the pitch antenna) changes.. so one may as well stick with a (much simpler / cheaper) standard VCO.

A 'real' hetrodyning Theremin mixes the ACTUAL waveform at the pitch antenna with the reference oscillator waveform .. This waveform changes (albeit sometimes only slightly) as the sensed object (and therefore capacitace / frequency) changes position.. The changes in this waveform directly affect the quality (shape) of the output waveform.

There are also (sometimes excessive and therefore undesirable) interactions between the reference oscillator and the pitch oscillator - as the frequencies of both get close (at the low frequency or bass end) which can cause quite dramatic changes in output harmonics through oscillator synchronisation.. This interaction can be implemented with a hetrodyning VCO circuit - but (as with a conventional Theremin) can cause erratic behaviour, as the oscillators can lock into a synced state too early, resulting in limited performance at the bass end.

To me, it looks like there are two ways to implement a good sounding playable Theremin -

The first is Lev's concept.. I confess to having been sceptical about the persistance of this design.. I thought that, with advances made since the 30's, clinging to this "outdated" design was more sentimental than practical - I was wrong! The more I experiment, the more I realise the elegance and brilliance of the original Theremin concept.. It is still probably the easiest to implement, and its sound varies dynamically and is unusual in being both quite low in harmonics and yet interesting and musically pleasing.

Then there is the voltage control method..[i] When I started, I now realise that I was not looking to make a new Theremin, I was looking to make a subtractive Analogue synthesiser controlled by a linear capacitive 'Theremin' interface.. I wanted to play the great Mini-Moog lead lines (Emerson Lake + Palmer, Rick Wakeman / Yes) on a Theremin Controller instead of a keyboard.[/i]

A VCO producing a ramp waveform, fed into a VCF, both of which are controlled from a voltage derived from the Pitch Antenna,(followed by a VCA controlled by the Volume antenna) can produce a dynamically changing output which can accurately emulate the classic Theremin sound .. AND can also produce other classic synthesiser sounds (particularly if the unit is configured as a patchable synthesiser, so that volume and pitch voltages can be routed to VCOs, VCFs and VCAs).

What does not work, is simply taking a VCO and tacking this on to a Voltage generated by an antenna.. All one has if you do this, is a signal generator - Dull and musically uninteresting.

Gordon -
Your question has just inspired a new idea.. Thanks!

Taking a High Frequency VCO producing (say) a Ramp waveform, and feeding this into a High Frequency VCF (a simple varicap diode will do this) and THEN feeding this into a mixer, mixing HF sine reference to produce the audio output - This could be configured to give a true, dynamically changing "Theremin" output waveform..

Yet another module to work on!

You're welcome.

[i]Lev's concept.. I confess to having been sceptical about the persistance of this design[/i]

I know what you mean. I've just been browsing Gabby's Photos (http://www.thereminworld.com/forum.asp?cmd=p&T=3300) and I am once again astonished by the number of components in a tube theremin - so few - a really neat, tight, clever solution. But at the same time the structured programmer in me - knowing that software is only hardware by other means, so similar design criteria apply - points out that it combines two discreet functions - pitch determination and timbre creation (frequency and waveform) into a single, indecomposable unit, making it impossible to adjust one functionality without impacting on the other. A program written using such techniques is a house of cards - hard to get just right and difficult to maintain.

(But arguably worth the effort. video (http://www.youtube.com/watch?v=YvLkjuYunRw))

Great Video!

[i]A program written using such techniques is a house of cards - hard to get just right and difficult to maintain.[/i]

The analogy to structured programming is a good one - The dependency of the elements in a conventional Theremin does cause all sorts of 'problems' which only really show themselves when one wants to change something or add something.

Back to the "Glasgow" Theremin (which has seperated the circuit into functional independant blocks.. Ie a 'structured' approach) - This is the method I started with.. And it works well IF one is not primarily interested in a "Theremin" but more in a "Capacitance controlled Synthesiser". A good Synthesiser with VCO, VCF, VCA and sufficient signal routing options does allow a reasonable simulation of a 'classic' Theremin sound.. But to get a really good emulation is not as easy (or cheap) as it first seems.

And there are surprises.. I have been playing with a lash-up of the Glasgow Theremin since I first posted on this thread - Most of what I said has proved to be right.. But [b]Pitch Linearity[/b] is [b]Far Better[/b] than I had expected. I had overlooked the fact that the Pitch -> Voltage conversion was linear, and therefore did not put out "Volts per octave" scale, it puts out a "Volts per Hz" scale.. This, driving the 8038, with its linear control law, gives quite a good linear playing area for about 3 octaves. (the capacitive square law dictates that linearity cannot be extended without some changes to the HF oscillator / antenna)

[b]If a better HF oscillator / antenna tuning circuit is employed in this Theremin than is shown on the Glasgow Theremin circuit, it is quite possible that the claimed linearity is being achieved[/b]