Pro Level Moog Theremins

Posted: 9/23/2014 2:00:24 PM

From: Richmond Hill, Georgia

Joined: 9/18/2005

Jason and I had an interesting chat via social media. We both lamented the presence of a true pro level Moog theremin. The Theremini is, of course, a toy. At Etherfest 2008, the assembled board of all of us, including Lydia and Carolina, pretty much told Moog what we wanted in the next pro level theremin. What came out later was the Moog Etherwave Plus and then, of course, the Theremini.

I always hear that it's "too expensive" to do a new production run of already designed theremins. Here is my idea: why not run off a new batch of circuit boards / components and sell them as kits???

Once could buy a bare bones Etherwave Pro or whatever and go forth from there.

This is all wishful thinking as it appears that Moog isn't listening to the theremin community anyway.

Just my 2.3434 cents worth.



Posted: 9/23/2014 2:55:19 PM

From: züriCH

Joined: 3/15/2014

hey feeleep,

that kit idea is just great. with a lot of if's. i always wondered why they did not continue/get further/improve the pro. simple ignorance?  price? compared to a guitar: a simple fender strat costs about 1500$ here, and a ew+ about 560$.  so i would be willing to pay something extra for a good standalone theremin. maybe moog doesn't get it, that it should not be  another synthesizer, but a instrument on its own. with controlling functions.  i think it will take moog another six years for a 100th theremin annyversary model.  a usable one for the real concertistas among thereminworld and beyond.


Posted: 9/23/2014 4:05:25 PM

From: Eastleigh, Hampshire, U.K. ................................... Fred Mundell. ................................... Electronics Engineer. (Primarily Analogue) .. CV Synths 1974-1980 .. Theremin developer 2007 to present .. soon to be Developing / Trading as . ...................................

Joined: 12/7/2007

Moog are unlikely to release Pro kits -

When you supply a kit, you need to supply a schematic - And for whatever reason, Moog keeps the E-Vox and E-Pro schematics under wraps -

I suspect that a reason for this probably comes back to Bob Moog.. Bob developed a clever scheme for register switching on the E-Pro, and a different clever scheme for MIDI on the E-Vox, but both had a "uncool" aspect to them - something that people like me, back then, would have ignorantly scoffed at and derided - Bob also had, I suspect, some discomfort about revealing his "crude" approach, so chose to keep it under wraps.

Times have moved on - "digital" and "logic gates" used in analogue instruments is probably far more acceptable now than it was then.. Back then those in "the know" probably would have regarded an XOR mixer as well below "Pro" standard - I certainly would have counted myself in that group - and the purist in me still does - but the pragmatist in me acknowledges that logic mixing confers massive advantages.

So here's a question:

If someone was to sell Pro-level boards - boards which gave register switching, waveforms the builder could set to taste, response (linearity, span etc) and all sorts of adjustment the builder could either take to the control panel or set up on presets or perhaps via a USB lead, would there be a market?

Because this is what I am working on at the moment - I am hoping to have an instrument for a renowned thereminist to play by early next year.. But I will not be in any position to put these instruments into production if it was liked - So I have been thinking of selling boards for DIY constructors or hopefully OEM'S which they could put in the cases they choose or manufacture and sell.

The heart of this instrument is analogue, but it operates with a PSoC4 (or two if one wants a editor with LCD etc so one can edit sounds and configuration of probably 64 presets) -the topology is based on the "upside-down" theremin I proposed a while back..

I have a limited time in which I must fully complete the above - If I dont reach this deadline, then I will call it a day, publish everything I have done (including sound samples etc) and hope people get it up and into production without me.

I have absolutely no idea though about what the kit-builder market is like, how large it is, how competent it is, what sort of support builders need, what sort of price they would pay..

So would be interested..

None of the above will make any difference to my direction though - even if only one person was interested, I am driven by something else - probably insanity!



Below is a preliminary table of adjustable functions - there are a stack of heterodyning derived waveforms, each of these feeds a pair of analogue filters with adjustable highpass and lowpass filtesr (forming high, low or bandpass) - these waveforms are mixed, so one can morph from say a pulse to ramp to triangle to sine as pitch changes, and have any such configuration you choose - the level and frequency response of each waveform being independent.

Then there are FM functions which feed selected audio back to modulate the pitch oscillator - this gives tonal richness, particularly if taken via some formant filters.

Then there are four formant filters (2 pairs) which also have separate HP and LP filters so the formants can morph with frequency (this is an unnatural  sound, but interesting). The audio section has its own controls for global tone control and drives multiple DCAs to allow mixing of external input audio to any route, including to modulate the pitch oscillator or to feed to the preview output, so preview can play a mix of the 'preview' and theremin and external audio.

The system is being designed as two boards - one being a basic register switching theremin which just dumps out raw waveforms and has potentiometers or presets wired for user controls. The second board adds all the other features and user interface - one fits digital potentiometers in the first board, and the second board takes control with these.

Much of what is in the diagram is inside the PSoCs - The volume CV for example is converted to digital and drives Log DCA's - PWM's are generated by the PSoC which control other functions - The bit which is not affected by digital stuff is the audio generation, this is near instant (the latency comes down to any latency in the Phase-Locked oscillators, and should be down at microseconds).

The table below is just the skeleton - one needs to have some structure to start dealing with the digital aspects of a project like this and get some idea about what is needed - Names for waveforms / voices etc are just preliminart - particularly on the analogue waveforms (rather than the mixed-signal ones) as I plan to have several analogue mixers which can be switched, including a conventional diode mixer and a true multiplier and probably a Fet mixer.. I may choose to lose some mixed signal voicing if I run out of space for analogue voices.

Also, I have included everything essential I can think of in the protocol - but things like pitch and volume tuning will probably not be implemented through the UI, but will be potentiometers hard-wired.. I think there are some functions probably too important to have in a DUI, so these are likely to be free for other commands.

I intend to convert WICP to MIDI controller messages for transmission, and convert incoming MIDI CC messages to WICP so that external MIDI controllers can be used - I actually wanted to use MIDI internally, but the MIDI limitation of 7 bits is something I cannot live with - And although it would probably be sensible to use MIDI CC internally, at this time wasting an entire byte to transfer a single bit ..... :-( ..... I may change my mind though..



WaveCrafter Internal Communication Protocol  (WICP)                  
1. ) One command byte followed by one or two data bytes.                
2.) Like MIDI, command bytes always have their most significant bit set, data bytes always have this bit at 0.        
3.) Bit 0 of the command byte also carries the most significant data bit, allowing command and 8 bits data to be sent in two bytes.     
4.) The command defines the length of the data, and commands can initiate transfer of 2 data bytes, giving a total of 1+7+7 = 15 data bits.  
5.) Commands with one following data byte can be running status if only the least significant 7 bits change.        
7 6 5 4 3 2 1 0            
1               COMMAND MESSAGE     MSB            
Creating the Command Byte:                      
The command message is composed of 6 bits as detailed below, this value is shifted left by 1 bit, 0x80 is then orred with this as is the MSB of the
data byte into bit 0.                        
Commands - Group 0: This group is primarily for system level communications requiring 8 data bits.      
0 0   {TBD} Pitch tuning if used / not using multibyte.            
1 1   Pitch linearity 1 (far field)                  
2 2   Pitch linearity 2 (near field)                  
3 3   Pitch Span  {TBD}                
4 4   {TBD} Volume tuning if used / not using multibyte.            
5 5   Volume Linearity 1  (far field)                
6 6   Volume Linearity 2  (near field)                
7 7   Volume Linearity 3 (audio side)                
8 8   Volume Span  {TBD}                
9 9   Register Selection                  
A 10   Audio Output Level (to main)                
B 11   Audio Tilt                    
C 12   Audio Balance                  
D 13   Preview Level Out [Monitor]  {TBD}              
E 14   Main / Preview Mix [Monitor]  {TBD}              
F 15   External input -> Monitor  {TBD}              
Commands - Group 1: This group is primarily for general and waveform adjustments with 8 data bits.      
10 16   Voice 0 (analogue / Sine) level                
11 17   Voice 0 (analogue / Sine) High Pass              
12 18   Voice 0 (analogue / Sine) Low Pass              
13 19   Voice 0 (analogue / Sine) Send                
14 20   Voice 1 (Ramp) level                  
15 21   Voice 1 (Ramp) High Pass                
16 22   Voice 1 (Ramp) Low Pass                
17 23   Voice 1 (Ramp) Send                
18 24   Voice 2 (Triangle) level                
19 25   Voice 2 (Triangle) High Pass                
1A 26   Voice 2 (Triangle) Low Pass                
1B 27   Voice 2 (Triangle) Send                
1C 28   Voice 3 (PulseRamp) level                
1D 29   Voice 3 (PulseRamp) High Pass                
1E 30   Voice 3 (PulseRamp) Low Pass                
1F 31   Voice 3 (PulseRamp) Send                
Commands - Group 2: This group is primarily for general and waveform adjustments with 8 data bits.      
20 32   Voice 4 (Square / Pulse / Special)  level              
21 33   Voice 4 (Square / Pulse / Special) High Pass            
22 34   Voice 4 (Square / Pulse / Special)  Low Pass            
23 35   Voice 4 (Square / Pulse / Special)  Send              
24 36   Voice [Special] PWM                
25 37   FM from Mix                  
26 38   Formant A  F1 Frequency                
27 39   Formant A  F2 Frequency                
28 40   Formant A  Lowpass Frequency                
29 41   Formant A  Highpass Frequency              
2A 42   Formant A  Level                  
2B 43   Formant B  F1 Frequency                
2C 44   Formant B  F2 Frequency                
2D 45   Formant B Lowpass Frequency                
2E 46   Formant B  Highpass Frequency              
2F 47   Formant B  Level                  
Commands - Group 3: This group is primarily transfer of more than 8 data bits.          
30 48   {TBD} Pitch tuning if used / not using single byte            
31 49   {TBD} Volume tuning if used / not using single byte.            
32 50   FM Functions -                   
33 51                        
34 52                        
35 53                        
36 54                        
37 55                        
38 56                        
39 57                        
3A 58                        
3B 59                        
3C 60                        
3D 61                        
3E 62                        
3F 63                        
Posted: 9/23/2014 5:21:22 PM

From: Northern NJ, USA

Joined: 2/17/2012

"I always hear that it's "too expensive" to do a new production run of already designed theremins. Here is my idea: why not run off a new batch of circuit boards / components and sell them as kits???

"Once could buy a bare bones Etherwave Pro or whatever and go forth from there."  - hypergolic

That's an awesome idea!  The EWP case with that curved wood is likely a huge cost driver.  And one could maybe angle the controls so that the user can actually see them?  Alas, in this modern world most companies paradoxically aren't in business to give us what we want.  And kits reveal too much, which is also verboten lately.


"...I am driven by something else - probably insanity!"  - FredM

Oh get real!  You're doing it for the same reasons we all are - money, chicks, and fame.  The insanity thing is just icing on the cake.  ;-)

Posted: 9/23/2014 8:48:46 PM

From: Colmar, France

Joined: 12/31/2007

I'm not sure if "copying" the Etherwave Pro is a good idea. In the eyes of the professionals who play one and in my eyes who has to fix, to maintain and to modify them, it has ways too much design flaws:

- Rather synthetic/annoying and not organic timbres (thanks to the digital heterodyning)

- Limited pitch range of only 5 octaves without register switching (that's the price you pay for the good linearity)

- Too heavy and too bulky to travel with it (that's the price you pay for the extraordinary design)

- The analogue front-end needs often to be recalibrated to maintain the excellent linearity when the instrument travels a lot (and that is a long and complex procedure)

- The volume response is snappy without modification

- The headphones output is of bad quality

- The tuner output gives always the top 5 octave signal independent of the register switch (can be modified)

- There is no true pitch preview output (can be added if you sacrifice either the headphones or the tuner output)

- The pitch and volume cv outputs are often useless because there is no gate signal

- The whole construction is absolutely service-unfriendly, replacing one of the front panel pots or switches requires complete disassembly of the instrument and needs at least three hours (if you are experienced). After re-assembling everything, you have to recalibrate the analogue front-end (see above)

- The pitch and volume boards are multilayer pcbs, replacing components is a real pain

- Neither schematics nor a service manual have ever been published

- Moog music inc is obviously not longer willing to do repairs
(Roger Ballenger called them after his EPro fell on its front panel and almost everything was crashed and broken. They told him that they were not sure if they could do it since the people who could have done this repair were not longer working there, so Roger was very disappointed if not furious and decided to ship it from California to France to get it fixed)

- Moog music inc does not longer carry spare parts for the EPro
(I had one contact there, a guy called Perry, who found me almost everything, if not in the stock, then in the drawers of the techs. Unfortunately, Perry is not longer working there, too, and since then, I wasn't even able to get a simple mute switch for the EPro from them - "sorry, we do not longer carry these parts")

Thus, put no hope into support from Moog music inc and try to avoid the list of flaws above! :-)

Posted: 9/23/2014 8:57:10 PM

From: Eastleigh, Hampshire, U.K. ................................... Fred Mundell. ................................... Electronics Engineer. (Primarily Analogue) .. CV Synths 1974-1980 .. Theremin developer 2007 to present .. soon to be Developing / Trading as . ...................................

Joined: 12/7/2007


You may be right about money and fame - But not about chicks. I have had my last close encounter with these creatures a while ago - NEVER again!

As for EWP kits - I have heard that these boards are almost unserviceable when they fail, Nobody at Moog can do it, and they have in the passed just shipped out replacement boards .. Seems to me that they are not in any position to support anyone buying kits even if they chose to do so.

The digital (I should say logic) aspect of the EWP is quite extensive AFAIK, and this lot could probably be mopped up by an FPGA or PSoC, the whole EWP could probably go on a single board - the only real mystery / unknown that I can see is the oscillators, as these seem from all accounts to be really good.

But alas, those with access to these dont want to share -

A lot of whats in the EWP seems badly engineered - from the choice of the PSU which by all accounts is a standard SMPSU taken from one of their synths and not well suited, to other issues that people have had problems with and some of which have been corrected by non-Moog sources.

When Bob was there, there was some hope of things going right eventually (It was Bob who created the first fix for the volume problem, from which more advanced solutions were engineered) - But without Bob, unless schematics were made available, I think buying a EWP kit would be real risky.



Posted: 9/23/2014 9:02:50 PM

From: Colmar, France

Joined: 12/31/2007

@Fred: Have seen your fantastic block diagram and protocol above. You really try to cover everything. But when, on one side I look at this and on my list of EPro design flaws, and I see on the other side that other professional theremins like the RCA and the tVox tour are (from outside) much more simple, the RCA having just a pitch tuning and a volume tuning control, and a mute switch, and the tVox the same plus one knob to vary a little the brilliance of the basic timbre, I wonder if professional thereminists wouldn't prefer an instrument with less functions but with better music inside. A violin player for example knows that he has just four tuning knobs for the strings, all other parameters are "by design" of the instrument or a little by choice of the quality of the strings, but the biggest part comes from the player and his playing technique. Shouldn't it be the same for the theremin? 

I fear that if you don't take these thoughts into consideration, you risk to run into the Theremini effect : Lots of functions which will make untalented players think that these will allow him to be a virtuoso, but the true musician will perhaps not appreciate this engineering genius.

Just my 3.14159 cents...

N.B.: Please don't explode, these are only my thoughts, neither critics nor provocations!

Posted: 9/23/2014 9:58:30 PM

From: Northern NJ, USA

Joined: 2/17/2012

"I'm not sure if "copying" the Etherwave Pro is a good idea. In the eyes of the professionals who play one and in my eyes who has to fix, to maintain and to modify them, it has ways too much design flaws:"  - Thierry

Eek!  That's a rather nasty list!

"I fear that if you don't take these thoughts into consideration, you risk to run into the Theremini effect : Lots of functions which will make untalented players think that these will allow him to be a virtuoso, but the true musician will perhaps not appreciate this engineering genius."

To me the Theremini effect (great name, but how about Theremini Syndrome like The China Syndrome?) is not too much effort on the processing / synth side, but too little thought / engineering effort on the capacitive sensing side.  I was kind of shocked that there was nothing new nor innovative to learn in that regard, and in fact it seems to be a large step backward.

Posted: 9/23/2014 11:13:28 PM

From: Eastleigh, Hampshire, U.K. ................................... Fred Mundell. ................................... Electronics Engineer. (Primarily Analogue) .. CV Synths 1974-1980 .. Theremin developer 2007 to present .. soon to be Developing / Trading as . ...................................

Joined: 12/7/2007


No explosion from me on this matter! ;-)

I think your points are entirely valid - From all the commercially sound reasons - Really, there is so little interest in the theremin that any extra effort is probably wasted effort... certainly is in terms of commercial return.

The few who actually play the theremin are stuck with what they have - they dont have a choice,there is no instrument that is TRULY "professional" - there is no instrument that allows them to set up linearity or span the way they want, no instrument which allows them to fine tune the sounds to exactly what they want to hear.

And there is no incentive for any developer to give them this.

The directions I have gone in have been steered by my love for synthesis - But the thereminist isn't able to twiddle knobs or devote their attention to anything other than playing, and I have realized that simplicity is essential. For this reason I am looking at a digital interface and presets - so the thereminist can create the sound at leisure, and then just call it up with the turn of an encoder or perhaps even an inc/dec pedal.

I am perhaps unusually placed - I have wasted 7 years on numerous theremin ideas, learned a lot in the process, but this indulgence has brought me to total financial and personal ruin.. I had given up and completely lost heart, realizing that had I just put out some bog standard "ok" instrument I would be far better off - And seeing that it was too late to change direction even if I could bring myself to do this - my best hope lay in a high-end instrument I have mostly prototyped, but was too complex and expensive to get into production without laying out a lot of money that I dont have, and it would take too long, and be priced so high that it would have few buyers.

Then, thanks to TW and particularly Dewster's support, I started looking at using FPGA to reduce the complexity - but it was still way out of reach from a production perspective for me - just the damn SMD aspect prevents the kind of realization of something I evaluate and could get made at bearable cost.

Then a £4 PSoC 4 board appeared, and I started playing with it - putting the phase comparators in, changing these to get them to give operation I could not obtain from a 4046 type 2 comparator I had used on my prototypes, ran the PSoC with a LC oscillator I developed that has a wide range CV input and 'antenna' input in a closed loop, implemented my mixed signal wave-shaping in the PSoC, used the configurable IO to allow direct control of the PWM amplitudes, mixed these and put a simple 5th order LC LPF on the sum, and got fully adjustable waveforms... I then found I could actually do more with this PWM topology - I could insert tunable filters on each waveform allowing the output  harmonics to vary as a controllable function of pitch...

Then I managed to implement register switching on the analogue signals, so I could also drive analogue mixers..

And at this time it looks like I can get everything in my list into 2 PSoC4's and a load of not-too-expensive digitally controlled potentiometers and external analogue hardware - but all the really awkward board-area consuming stuff, the stuff that made my original high-end theremin a production/testing nightmare, into a couple of tiny boards costing £8.

However - there is a hell of a lot of work to be done - modules have only been tested in an ad-hoc way, and I have not actually got a system even started.. I have a PSoC 4 board plugged into a couple of solderless breadboards and a couple of oscillator boards,and theres a lot of stuff in PSoC creator that I havent got my head 'round yet - And rather than doing this I should have spent the last week on accounts and other legal matters for a big meeting with the suits on Thursday... So that will be tomorrows panic.

The reality is that this is an absurd distraction for me - I am fiddling while everything burns down.. Its not even taking risk anymore, the odds on me remaining solvent and being able to realize this absurd fantasy are almost zero... But at least theyr not absolutely zero, and I cant say that about anything else in my life.



Posted: 9/24/2014 12:07:44 AM

From: Richmond Hill, Georgia

Joined: 9/18/2005

Thanks Fred. I am afraid that my background makes any technical comment I may make worthless, other than to say what you put on here looks nice.

You've obviously put a lot of soul and treasure into your project and for that I wish you well.


You must be logged in to post a reply. Please log in or register for a new account.