Idea for Theremin Staccato Pedal

Attack, attack, attack, attack, attack, attack, attack is what we lack.

-- Phil Lynott, Yellow Pearl

A Brief History of Staccato

The pitch circuit of the theremin was developed first. Before Theremin devised a circuit for the volume loop he made instruments that had a pitch rod, an expression pedal and a hand held mute button, an "articulation regulator". These also featured in similar instruments that were developed elsewhere. 

On this web page, Peter Pringle describes this in a bit more detail, and provides a short sample of music made with a modern theremin and a simple hand held articulation regulator. http://www.peterpringle.com/ondes2.html

In this video, Martin Taubman plays his theremin variant.

Development of this theremin variant was continued in Russia by Theremin's assistant, Kovalsky. The Kovalsky theremin has a pitch antenna and an volume pedal, and the articulation regulator has developed into a panel of switches to control various aspects of the sound, including staccato playing. In this video, Olga Milanich plays a Kovalsky theremin.

Mention should also be made of Pamelia Kurstin's staccato loop technique (as she calls it. It is more commonly referred to as "walking bass".)

My Experimental Experiments

I have explored variations in my recordings on the various approaches that have been used, with varying results. 

This was the first, using an etherwave theremin and a home made articulation regulator held in the pitch hand, and using both the etherwave antennas. I also tried holding it in the volume hand it was confusing - it definitely seems better not to have these two different volume functions - expression and articulation - with two control devices - in the one hand. Having the articulation regulator in the pitch hand was an encumbrance, and would conflict with aerial fingering techniques. For my purposes I found it usable, but limiting.

I have also experimented with a Kovalsky style arrangement, fabricated from a conventional theremin and a soft modular synthesiser running on an iPad mini which acted as the control panel with a USB MIDI expression pedal connected to it. I do not have any recordings of this, as I was very unsatisfied with it. I found relinquishing the volume loop in favour of a volume pedal too limiting. In my opinion the volume loop is as important to the theremin as the pitch rod. (Perhaps even more so - a volume only theremin is a very limited device indeed, whereas my experiments with controlling the volume of a fixed volume instrument with the volume loop of a theremin leads me to feel that the volume loop is a useful tool in itself. (Specifically I ran the output from a stylophone - a very inexpressive instrument with a crude sound through the volume loop of my etherwave (thanks to Thierry, who modified it in a way that made this possible) and it was a vast improvement on the unadorned stylophone. No recordings - I have no stylophone skills worth sharing.) 

I have also tried staccato volume loop technique, using my Kees Enkelaar theremin, which has a particularly snappy volume loop. In this video it is the "bleep" sounds, not the noises, which are not from a theremin. 

 I have also experimented with an effects pedal called a Chopper. This modulates an audio signal with either a sine wave or a pulse width modulation wave (a square wave with variable mark space ratio) or a combination of both, the speed of which is controlled via a treadle. With some settings it "chops" the sound into regular pieces with brief silences in between. 

https://soundcloud.com/beat-frequency/snippetymen

Finally, I have a thin wire attached to the near side (player side) of my volume loop that extends onto the body of the theremin, where it is held down with gaffer tape. This provides a very small portion of capacitive field where I can tap out a staccato rhythm. Iy can only be used for playing quietly, because of its proximity to the volume loop. It can be heard occasionally in this recording.

https://soundcloud.com/beat-frequency/ouroboros-drive-for-barbara

Theremin Staccato Pedal

My idea comes from the simple observation that many musicians, and people in general, are very comfortable tapping out a rhythm with their foot. Indeed you can even get literal stomp boxes - little foot drums, nice wooden boxes fitted with contact mics, preamps and audio out sockets. http://www.log-jam.co.uk/logarhythm-mk3.html

So it seems reasonable to keep the expression functionality in the expression pedal, and assign articulation to the foot, reversing the arrangement of the theremin variants.

I want articulation to be achieved with a capacitive sensing system rather than a mechanical foot switch for several reasons - a switch by itself is very "clicky" - it requires additional circuitry to provide some envelope shaping rather than just blunt on and off. This would have to be preset. With a capacitive sensing plate, just as with the rod and the loop, there would be subtle variations introduced by the player. If there is magic in the theremin, it lies in the interaction between the player and the capacitive fields. And it would allow the player to move his or her foot away from the plate and continue playing.

It also seems reasonable to me to have the capacitive field operate in the same way as the volume loop, so that the closer the foot gets to it, the quieter the sound, so that playing it means punctuating a continuous sound with brief silences, rather than being able to tap out very short notes. And also so that the player can move his or her foot away from it and continue playing. Other than that, it should have a topology suited to articulation - i.e. a very small, very snappy field - the opposite of what is required for expressive loop playing.

Mention should be made of the Z-Vex Probe range of effects pedals, which have a capacitive sensor in place of a treadle controlled potentiometer. The tremolo in particular (http://www.zvex.com/trem.html) is quite close to what I envisage. (Except that it introduces distortion, which I do not want, the volume logic is reversed and tuning the antenna comes with a warning not to do it!) The Probe range leads me to think a device such as I describe is feasible. I would very much like to hear from the electrical engineers amongst us on this point. 

Click image for larger version.

 Gordon,

{ADDED} One thing I was forgetting with what I wrote below, was the effect of the floor in relation to the sensing plate.. I dont think that simple theremin volume circuits will be quite as easy to implement as I had thought.. But I do have a technology using shielding which will work.. But again, the floor = soft is the problem, because thermal drift moving the "mute" position beyond reach near the floor would be a real problem, whereas if maximum volume was to vary marginally at the plate, this wouldnt be a problem.. Adding thermal compensation to a shielded sensor is essential if one wants to have tight control and low drift - it can be added if essential, but would be less hassle / cost if I can get away without it.

A capacitive foot sensor should be extremely easily implemented.. In fact, there are several possible routes to this, from standard theremin circuits to CapSense .. For example this little PSoC kit is a single antenna cap sensor that could be configured easily into an auto-tuning foot sensor and is fine for short range detection (it can sense several feet, but is musically useful over about max 6")

I have a couple of these DK's somewhere, the DK's are pricey (£87) but the chips are cheap (whole circuit < £10).. I gave up on these due to their limited range, but they may well be ideal for this application - let me know if you are interested - I could post a DK to you (if I can find them ;-) for you to play with (the code for basic evaluation is all pre done and there is a GUI to allow tweaking - you can of course write your own code, and use the sensed value to drive a PWM or whatever - The PSoC even has MDACs so one could probably implement a DCA to do everything in the chip..

<< The above kit would be configurable for shielded operation >>

The other approach would be a standard theremin volume circuit - You could use the SC theremin kit to do this - simply leave out the pitch components and build the volume section - These kits drive the mixer IC directly so dont have a "real" VCA, so the mixer side would need to be re-configured to accept audio (could also double as a ring modulator / frequency doubler) but I would need to look into that... Care would need to be taken regarding the HF frequency as this will be radiating and could interfere with other theremins..

One could use a fixed frequency volume scheme like that in the EW - In fact, if you buffered the EW volume oscillator and had this output to a foot "pedal" containing a duplicate antenna circuit / detector / VCA both could be run simultaneously without any risk of interaction..  A little board in the EW containing a beefy buffer, and piping the oscillator and power supplies to a socket on the back.. The "pedal" having the inductors / detector / tuning / VCA (duplicates of the EW circuit).

The really simplest would be a dedicated circuit containing a fixed frequency oscillator and a simple tuned antenna circuit without linearization driving a detector and VCA - I will have a scrounge in my decimated lab next time im allowed into my house, and see if I can find one - or the circuit for one.

<< The above could have problems operating with a plate on the floor >>

What makes capacitive volume circuits "complex" is the requirement for range and "linearity".. The pedal wont need much range I would think - {Have just seen specification of 2cm so this is mostly redundant} I guess 6" for maximum volume?   The "law" might be a problem - It would be much easier if "foot down" was maximum volume - I think foot physiology probably makes for more extreme movement with the foot close to the floor than when its raised, which is really the opposite of what is needed if minimum volume is near the floor.

To get a "linear" response with minimum volume at floor level, one probably needs to compensate both for the capacitance effect increasing inverse square as the foot gets closer (making the change more rapid) and for the foots mechanical inclination to being more rapid as it gets to floor level - what one needs (if floor = soft) is change to be slowest as the foot is lowest - Probably requiring quite a lot of curve 'bending'.

<< the 2cm specified in the diagram removes any issues about foot control, but could cause difficulty in other areas - It might be difficult to keep the field so small - may need more like 4 or 5 cm >>

If floor = loudest, then things might be almost right without any compensation.. It could still be "snappy" but would also allow for control of dynamics - I think that an uncompensated floor=soft scheme would almost be an "on / off" control with extremely little control over dynamics.

Fred.

PS - I would be happy to build this for you.. Even happier if it was interesting enough to put into production and some competent seller was to market it - or better still, take the whole process of purchasing / manufacture / marketing off my hands and give me a commission.. ;-)

Just been looking at the ZVEX pedals - £200 up ..

-----------------------------

Information on the Wah pedal:

----------------------------

How it works

The probe circuit generates a small (one or two inches high) "bubble" of RF energy at about a million cycles per second above the copper plate.

Lol, someone else who thinks "bubbles" ;-) .. Ok, 1MHz, 2" sensing ... Thats REAL simple!

As your foot or hand (or any wet or metallic object, for that matter) approaches the copper plate, the RF field is disturbed and the circuit reacts by increasing the brightness of an LED, which drives a photoresistive cell and controls the Z. Vex wah circuit.

The wah circuit is actually the Seek-Wah sound circuit without the sequencer control. It's also the same wah circuit as the one in The Drip guitar that I make. The only difference is the boost circuit in this Wah Probe model.

Adjustments: The simplest one of these is the volume level of the "drive" knob... just set it so that when you kick on the wah it's doing its stuff at the right intensity. It goes from slightly lower than the normal guitar volume to completely bonkers when turned up all the way. I personally don't think it's necessary to turn it up all the way, in fact, it seems to pretty much obliterate the wah sound, but, to each his own. Abuse those transistors, please!

The next adjustment is the RF field screwdriver adjustment on the bottom of the pedal. If you turn it in the direction of the arrow (usually to the left for that one) it will make the probe circuit more sensitive and set the "bottom" setting of the wah higher and higher.

It's best to hold the pedal completely upside down so that the cords are dangling away from the antenna, with your hand and arm completely away from the antenna, and strum your guitar's open strings (you probably have to have it on a stand or strapped on) thru the Wah Probe. Gently adjust the RF pot with a tiny screwdriver in the direction of the arrow until you hear the wah get to the level you'd like it to be when you are not putting your foot near it. There's a region where the sound of the guitar is kind of "muffled" and the wah kicks in after your foot gets near the pedal, which is how i ship the unit. If you like the wah to be partway cocked into the lower wah region just park the adjustment right where you like it. As you get closer to the unit, it will wah higher.

The other adjustment isn't too intense... it's the LED current control. You can usually leave this one at exactly halfway. That's how I usually ship the unit. If you think the wah is too bright when your foot is all the way on it, just turn this control to the left to make it a little duller at the top end. I ship it so that with a pair of heavy Doc Martins (the rock and roll shoe of choice for many musicians) my foot will make it go from muffled to very bright.

And it all looks really crude and low tech..

So question: What are is the maximum tolerable sensing range, assuming 2cm is the minimum.

No other questions really - The big one is about floor = soft or floor = loud, and how important this is.. But with <5cm span, either is possible.. floor = loud would be easiest technically, and be less difficult to tune / less subject to drift.

Thank you for your thoughts Fred. 

I should make it clear that I have no intention of rushing into this. I would like to garner as much feedback as possible before a prototype is constructed - both regarding the technicalities of the circuit and from a variety of players. This idea is the synthesis of a lot of my personal experience, as described, and I am very aware that my experience differs from that of conventional players. (By conventional I mean those who play music that observes such conventions as 12TET and A0=440Hz.) On a practical note I also have a limited budget, and a theremini is at the top of my shopping list. So I am happy to proceed very slowly. 

I also seriously doubt doubt there is anything but a tiny market for this, possibly only one will ever be required, by me, especially given that similar ideas have existed since the theremin was first invented, and they have not caught on.

I would say that around 2cm is the maximum, not the minimum range. I want to be able to tap it very quickly, both in terms of shortness of attack and in the period between taps. The former because a rapid change in volume changes the timbre of the note via amplitude modulation, and this I want to emphasise, and the latter because not being able to play rapid successions of notes is a major limitation musically.

The pedal would have to be near=mute, far=loud for the reasons I noted. Not least because I want to be able to step away from the pedal while playing to use other effects pedals, but also because my experiments have suggested that this is preferable from a playing point of view. (Particularly, varying the mark space ratio on the chopper has led me to the conclusion that punctuating a continuous sound with very brief periods silence is musically preferable, at least for me, and this is best achieved with a near=mute configuration.)

i note in the description of the z-vex tremolo he describes an aluminium shielding on the underside of the instrument. I guess the effect of having it, particularly if it is earthed, would be to make any variation in the capacitance of the floor beneath it negligible.

"Under the whole mess is a 3 5/8" by 7" sheet of aluminum that provides shielding from annoying ground effects and wires under the stage."

i suspect you are right that it is more complex than might appear at first sight (isn't that always the way with theremins) - I mentioned this idea in passing to Thierry a while ago, and he has been scratching his head over it since then. (BTW I believe Thierry is rather busy with more important things at the moment - it would definitely be courteous to wait for his input here as he has had some time to mull the matter over.) 

" it would definitely be courteous to wait for his input here as he has had some time to mull the matter over." - GordonC

Oh yes! - Theres a lot of frustration (rightfully IMO) when one spends time and computing power mulling over a problem, only to find it thrown public and someone else "stealing" any "glory".

I agree with your assesment that "there is anything but a tiny market for this, possibly only one will ever be required, by me" - I think that a unit with <2cm control span, and "silent when floored" will have extremely limited  appeal! ;-)

I do, however, think that a generic capacitive "pedal" with perhaps 5cm "travel" could have a moderate market.. If the whole thing can be produced for say £30 and sold for say £180 then it may not result in a Ferengi death sentance! ;-) .. Producing your version would be equivalent to suicide!

Fred.

Oh, metal grounded screen under box.. Yeh, thats the "shield" - Trouble is though, its not a good shield - there is capacitive coupling to the sensor which will load this, variation in this coupling due to temperature etc will cause huge drift problems - the closer the shield and sensor are, the worse it will be.. so unless the pedal is thick as a brick, its not going to be good...

My thinking now is on the lines of a "Driven" shield that doesnt load the sensor and has minimal thermal effect - for the 2cm or 5cm versions this could work well, because the sensor voltage doesnt need to be high, and driving the shield should be easy.. The small sensing distance makes this feasable even for your absurdly small distance! - My original thinking was on the lines of a TX shield RX sensor, but the driven shield is better because its coupling to the sensor is less critical.

But you wont get away with a simple sensor - If you want silence with your foot down, and want 2cm to be max volume, the circuit is going to need to be really stable - Any drift on the "null / mute" point could make it impossible to silence the instrument - so I recon you probably need to tune this point to about 5mm above the sensor plate, and limit the control span (to maximum volume) to 15mm..

Its almost worth looking at optical sensors - a 4cm to 15cm Sharp analogue optical sensor (£12) has a "back field" detection range from about 2cm to 4cm - I have used this field in the past for near-field detection, and built logic that determines when it switches to "far field" operation and back to "near field" - This way you could sense 2cm of movement by adjusting the sensors position in the enclosure... Only real problem perhaps is latency - cant remember but I think it was a bit longer than 10ms, might even have been 20ms.. I imagine that you want extremely low latency for this application, probably something <= 6ms.

Even ultrasonics might work ....

I can see the headline now - Theremin Genius Commits Kill Switch Suicide.

But seriously, yes, this is why I want the thoughts of other thereminists. 

"But seriously, yes, this is why I want the thoughts of other thereminists. "

Thats not what you said originally!  You said " I would very much like to hear from the electrical engineers amongst us on this point" - not even a mention of "thereminists" LOL ;-)

Ok, ill just go skulk of into some corner - I can see when im not wanted! 8-(~

As Dewster says:

I keed.

I am delighted to receive any relevant feedback, positive or negative.

(Yes, I know you keed. :-)

"2 cm" -- GordonC

 
Why are you hooked on the distance sensing device? You'll get the same problem as that of theremin in general -- the lack of the tactile feedback.

IMHO the pressure sensor is a more preferred way. 

"A Brief History of Staccato" - GordonC

Excellent!  You present strong evidence that all things benefit from the addition of spicy lingerie videos!  ;-)

The Taubman video shows him standing on two separate things, one is the volume rocker pedal, the other perhaps a ground plate?

The portion of the Z-Vex Probe video where he operates it with his fingertips is quite interesting.

So you want something that will give you quick articulation, but will also do gradual articulation?  These goals seem to be somewhat at odds with each other?  Maybe a dual pedal, one for quick and one for gradual?  Or are you interested in quick and somewhat less quick?  Sensing the velocity and having it influence response might be an option.  Though limiting the range as you have to 2cm or less might be sufficient with a non-linear response.

A flat, thin sensor on the ground (like your drawing, and opposed to the Z-Vex high angled plate) makes particular sense if you are planning on using it while standing at the Theremin, though anything that threatens to destabilize the required stock still statuesque stance could be problematic. 

IMO, the rub with doing any of this is doing volume control itself.  You can do quick variable resistance with a FET or transconductance IC, but the signal must be kept small to avoid significant distortion.  You can brute force it with A/D D/A and software but this might give delay if you aren't careful, and aliasing could be an issue if you drive it with strong high frequency content, though doing this sky's the limit, and we have reached the point of cheap processors with built-in A/D D/A that this should perhaps be the default approach.  Maybe something in-between, like PWM to ground?

Aside: The Kovalsky would kill if it could do chords!

Thierry is actually ways to much eaten up by the fight for the daily bread...

But I have nevertheless been scratching my head over this staccato pedal idea. I rejected the capacitive principle because of the proximity to the floor/ground.

I wonder if a simple foil key could do the job, they are available with a mechanical action way of less than 1mm and acting forces below 0.5N which corresponds to 50 gramms. Such a foil key, if mounted/glued on a solid ground plate, could be used to simply short (mute) the audio signal and be acted with one toe or the heel but without the risk of destabilizing the player's posture.