A "step" theremin? Also a few q's

Hi everyone,

I want to make a theremin! And I had an idea that I'm positive must have been thought of before. I've heard (and experienced, the couple times I've actually used them) that theremins are very hard to make sound good.

So I was thinking, what if I made one that "divided" up space so that if your hand was anywhere in a certain range, it would still only be a single specified frequency (rather than the continuum of frequencies that that range would normally contain). This would make playing most songs pretty simple, though I guess it would almost defeat the purpose of it being a theremin...

Anyway, has that been done? If I attempted it, I'd most likely use a programmable microcontroller because I have some experience with them and I think it could do the job.

But I shouldn't get ahead of myself, I've never built a theremin at all. Could someone suggest a very simple model, one I could maybe do mostly on a breadboard? I've found some plans online but they seem to be pretty intensive and I wondered if there was a standard "beginner" circuit I could assemble, to learn about it.

Thank you!!

Welp, I'm an idiot -- a quick search shows that this simple one does what I was talking about, "discrete" mode.

Hello TG, Welcome to TW!

Yeah, the apogee kit is cheap and simple - but its certainly not a "theremin" from any purist perspective.. its range (sensing distance) is tiny, and its a toy.. In terms of having a "step" function, I think you probably found the only "theremin" (other than MIDI theremins at the high end) with this function. IMO, the ONLY thing the theremin really has going for it is the fact that the player has complete control of the pitch - that this can be slid between notes, and that it is possible to add player controlled modulations both on pitch and volume..

Take the above away, step in tone increments (not even semitones, LOL ;-) and IMO you are far better off building a toy electronic "piano" using a 555 timer some switches and some variable resistors!

The Apogee kit will teach you nothing about theremins or electronics - all functions are embedded in a small PIC - its a bit like one of those musical greeting cards, and probably less musical than most of these! - But as a simple cheap project for amusement, go for it! .... But DONT think that it will give you ANY idea about what a "REAL" theremin is... IMO, its enough to put someone who doesnt understand this completely off theremins!

There is a long-standing hobbyist theremin web site where peaple are warned against me (although un-named - I am referred to as the "TW Devil") and one of the "identifying features" declared about me is that I recommend the SC/Jaycar theremin!

Well - I am going to do so again!  ;-)  ... I DO NOT regard these as "good" theremins (and have said so many times, and given full exposes of the technical reasons why I think they are not well designed) - but I do think they represent excellent value for money, are simple to construct and get running, and incorporate both pitch and volume at half the price of a little 3 transistor pitch-only board (which the webmaster of above site praises) .. IMO, the Jaycar theremin is the best value kit available, and the theremin one gets is usable IF the electronics is put into a bigger box so that pitch and volume antennas are further apart.. (and if the horrible speaker is removed and a line-out provided, and the audio amplifier is removed or the circuit modified as per the Enkelaar)

Also, there is a lot of builder / user generated data on modifications, and the kit can easily be converted to a Enkelaar theremin by more advanced builders.. In fact, some highly regarded (at least one ;-) theremins on sale for more than $500 started life as Jaycars, and got modified ..

But - It does not have a "step" mode.. A way to a step mode is to feed the audio (difference) frequency into a small processor like a PIC, and get this to produce the stepped audio output.. Even better, feed both high frequency oscillators into the MCU, and from this data derive the difference frequency and produce the semitone-quantised audio.

One other note - the simple Jaycar kit does not have a way to get the difference frequency before the VCA.. You would need to add a buffer and D-Flip flop (4013) to get a clean constant difference frequency..

My Element-14 Theremin group has loads of articles related to the Jaycar (including my H1 thereremin schematics that show how to interface it to digital, and schematics for the Enkelaar conversion) and other stuff ..

ps - (there is a Jaycar MIDI theremin kit - this will probably do what you want - if you have a MIDI instrument to 'drive' -  but I have no expierience with it so cannot say.. It is real easy to produce a simple MIDI controlled tone generator using the simplest MCU)

Fred.

This link:

http://www.thereminworld.com/Forums/T/29010/alternate-ways-to-generate-tone?Page=0

Is a good place to (I think) find some of the "extra" data required to modify a Jaycar .. It has links to my H1 theremin, and shows where one can get the required signals.. Ok, the H1 front end was based on the more elaborate EPE-2008 "Jaycar" theremin, but its similar enough that you can get a fair idea about what needs to be done.

It is possible to add a discrete mode to an existing theremin by using either a pitch to MIDI convertor or pitch quantisation (the dreaded auto-tune) in either software or hardware. I have been experimenting with this recently.

As Fred observed, doing so will reduce the expressive capacity of the instrument. It will not make it totally expressionless, a large proportion of the expressive capacity of the theremin comes from the volume loop - it is significant that this is also called the expression loop. (Note: as pitch to MIDI conversion turns the volume loop into an on/off switch you would need to track the volume using some sort of envelope follower).

For melodic playing this is not the main problem. The main problem is that you do not know where the boundary between one note and the next is - you may be right on the edge of one note and a tiny hand movement will trigger a note you did not intend to play, or you may be a (relatively) long way from the next note and it will take a tadge longer than you intended for your hand to get to the next note, throwing your timing off. It may be possible to get around this by providing with player with an uncorrected "pitch preview" that they can listen to in an earpiece whilst playing. As I am not a melodic player I cannot estimate how mind-bogglingly distracting this might be to the player, but I suspect it may be a very bad thing.

Also, if the next note you want to play is several semitones away from the current note your hand will have to pass through the intervening notes, triggering them in rapid succession. With continuous pitch this sounds as a smooth change in pitch and, if done fast enough, is barely noticeable. But with discrete notes, the faster you jump from one note to the next the more obtrusive the intervening notes are.

All of this means that it may be easier to play "in tune" but it does not make it an easier instrument to play. At least melodically. As a non-melodic player making somewhat aleatoric music it just opens up a new range of possibilities for me and I am enjoying the exploration of these quite a lot. :-)

Here are some examples of what can be achieved...

This is the super-cool John Armstrong, feeding his theremin through a rack-mount autotune hardware device.

This is Tim Blake (of Hawkwind!!!!!) with a cheap and cheerful pitch to MIDI box.

This is the sound of me using autotune software on an iPad.

And finally here is the masterful Peter Pringle showing us what he can do with a high-end MIDI theremin.

I hope this helps. :-)

Thanks for posting those, Gordon..

What really surprises me watching / listening to each of those clips one after the other, is that, to me, the most competent thereminist (Peter) got IMO the least satisfying results - Peter without the IMO horrible MIDI backing always produces wonderful music with the theremin - but here, with the best MIDI theremin available, I think his performance and the music produced is dull and lifeless compared to all the other quantized performances.

I really like John Armstrong's use of auto-tune, and I like everything Tim Blake did / does, so no surprise I loved that clip.. But the piece I found most interesting was what you did - Really different to your other compositions - but wonderful!

In all the clips, IMO the theremin is lost - its become a controller for a chromatic quantised synthesiser - even if this synthesiser is in fact a "theremin" .. To me, the major defining charactaristic of the theremins sound is its freedom - the fact that it is not bounded to any scales, not even microtonal ones.. IMO the second most important feature is, as you noted, the freedom to control dynamics - quantised / auto-tuned "theremins" may retain this latter quality, and I think this imparts some degree of "theremin-ness" to the quantised renditions on some of the pieces - But I do not think that Peters rendition exerts direct control over dynamics of the MIDI triggered sounds - this may be a reason why, to me, it sounds the most distant from the theremin of them all.

What these clips have done is to change my mind about the usefullness of quantising / auto-tuning.. I can see (more importantly, hear) that it has real musical application - For SOME "types" of music.

But, as to whether the theremin is a good controller for quantized sound production, I personally doubt it - I think that something like a ribbon controller where one can actually see the position of notes and choose to 'skip' those you dont want to play, is probably far better suited both to "somewhat aleatoric music " and to "melodic" music.

Fred.

"(Note: as pitch to MIDI conversion turns the volume loop into an on/off switch you would need to track the volume using some sort of envelope follower)." - Gordon

Actually, I think the easiest is to break the connection between the theremins mixer and its VCA - Take the pre-VCA signal out to the quantiser / auto-tune, and return the quantised audio back to the VCA.

Alternatively, the volume CV is usually quite easy to find in a theremin, and with one opamp can be buffered and taken to external devices.. the results are FAR better than messing with extracting both pitch and envelope from a composite audio signal having changing dynamics.

But yes - I know.... This does mean messing with electronics, or finding someone to do the job for you :-|..

Theremin pitch quantization might best be done by having the quantization kick in at very low volume levels so that new notes start on pitch, then at higher volumes remove the quantization.  I think this would require up to +/- 1/2 semitone offset be applied to the continuous range after the quantization is removed (so there is no pitch step at this threshold) which might throw off experienced players.  This wouldn't be too difficult to implement in a digital Theremin, but sans actually playing one with this feature I'm not convinced it would be an improvement over the standard configuration.

[EDIT] OK, just watched Gordon's videos (thanks!).  Sparse quantization to various scales and such is quite interesting.  I wonder how Peter is changing chords, I don't see any foot switching going on?

[Geek EDIT] I believe Gordon's video is on how to make quartz crystals for oscillator use.  The angles at which it is cut in relation to the 3D crystalline pattern are critical for various properties in-circuit.

Dewster wrote: "I believe Gordon's video is on how to make quartz crystals for oscillator use."

You are correct. The source of the video is the film Crystals Go To War (1943) "A story in pictures of the preparation and manufacture of quartz crystals for radio communication."

http://archive.org/details/6101_Crystals_Go_to_War_01_20_16_21

" I wonder how Peter is changing chords, I don't see any foot switching going on? " - Dewster

I believe the chording was automated by the synth Peter was using - I think I remeber this correctly, but cannot be sure .. it was discussed in some thread here I havent been able to find..

""I believe Gordon's video is on how to make quartz crystals for oscillator use." ----- " You are correct. The source of the video is the film Crystals Go To War (1943)" "

A truly wonderful clip IMO!  - So it was actually real !?- it looked like some SF movie ! - kind of stunning that we now have tiny crystals in throw-away digital watches, and can get high frequency crystal oscillators encapsulated in tiny modules one can buy for a few £/$ ... I remember playing with old whisker contact crystals (different technology application) for crystal radios as a kid and what fun it was.. oh, a real pain to get it to work, but fun nonetheless.... Replaced with germanium diodes - but there was something about actually having a physical crystal to probe..

Dewster wrote: I wonder how Peter is changing chords, I don't see any foot switching going on?

Watch the index finger of my left (volume control) hand. It's very subtle so watch CAREFULLY.

"The source of the video is the film Crystals Go To War (1943)"  - GordonC

*Sigh*  At what point did every geek with a soldering iron turn into Dirty Harry?  The movie they'd make today would probably be "Flying Death Robots Are Winning The War On Terror" or something ++good.

"Watch the index finger of my left (volume control) hand."  - coalport

Thanks!  It always comes clear after the magician explains the slight-of-hand.  I think you put the elaborate footswitch on the floor there just to throw us!