New theremin? Nano from Poland...

Hi Fred, I'll respond to your comment on the staccato pedal thread.

Hey everyone,

Sorry for my long absence here on the forum (a lot of stuff happened, also family vacation).

I've read all the posts and came up with a diagram, that should explain all technical questions.



Rick Reid, thank you very much for the comments on Nano.
--------------------------------------------------------

I've added to the pitch set:

1. 3/8"-5/8" mic stand thread adapter

2. Euro-US AC adapter

3. Pitch spy cable/adapter - 3pin plug to 3.5mm jack socket with level adjust, for any 32ohms phones.
   ( I've changed pitch spy plug - the new one have a mark, to connect it correctly )

I've removed wired pitch spy earphone set from product list
and replaced it with wireless pitch spy earphone set (more comfortable).

I've attached a "manual" with service drawings - http://assec.pl/sleeve.pdf

Again, thank you Rick.
----------------------

Nano is completely handmade. After it's assembled, it's tuned, then aged for 24h and tuned again.

Plastic box was only meant for shipping (i might add hardcases to products list).

There's jitter in mp3 samples, because there's high voltage power line near my lab causing a lot of interference.

This theremin's design is really miniaturized and it's not possible to add extra components
e.g. another socket with harsh or pure wave, VCO, VCA outputs, etc.

Turning off (mute) volume antenna doesn't turn off volume pedal/pad of pitch module.

Touching the volume antenna with your hand increases capacity of your body (body grounding via antenna housing)
so the pitch antenna reacts with increased/boosted tone frequency.

Pitch antenna - hand movement range - ca 60cm (24"). Volume antenna - hand movement range - ca 20cm (8").

Best regards,

Vladimir

Hello Vladimir,

Well, that diagram has thrown me completely ;-) ..

Looks like you have opted for an entirely CV route for implementing your theremin - not a logic gate in sight! ;-) .. Also, looks like you are using the oscillator frequency change directly (no heterodyning) so I presume you have a fixed BPF following each oscillator.

This does give you an easy option for adjusting the pitch span of the instrument - a simple attenuator following Urf..

I played with this topology years ago, and gave up on it - Mark Keppinger developed a theremin based on this topology, and he too seems to have given up on it. AFAIK the only theremin using a similar topology (but using fixed locked oscillator frequencies and deriving the CV from phase comparison, as I understand it) is the Moog 91.

Real interesting! Many thanks for sharing this with us.. To me, the CV topology has some big advantages you dont seem to be exploiting (because its a nano, and there's no way you could fit more in) these are the ability to adjust the span, and the ability to use the CV for control of a VCF that would expand the tone palette hugely.

Lots of clever ideas in your instrument! - I wish you well.

Fred.

Wow, I was utterly wrong as to how this one was designed!

Please correct me if I am utterly wrong yet again: it seems an adjustable (but during play fixed) pitch oscillator drives an LC tank with the C exposed to the playing environment, which forms a variable BPF (or peaky / high Q LPF).  The amplitude of this LC is rectified and low pass filtered (detected) and used to drive the VFO / VCA.

A possible downside to this approach is the AM detection, which likely lets a lot of environmental noise in.  A phase detector approach (as FredM describes the likely operation of the Moog 91 above) might work better in this regard but I haven't investigated it beyond seeing the effect often in my bench experiments.  Nor have I looked into the linearity of either approach, but the non-linearity of LC detuned phase / amplitude likely offer opportunities for improving the perceived hand/antenna pitch linearity.

Thanks for the very detailed info Vladimir!  (Do you know what the voltage swing at the pitch antenna is?)

Vladimir,

Below I am pasting a diagram of the instrument I am working on at present - there are some similarities.. The main one being that we both derive CV's from the antenna oscillators and use these for control of pitch and volume.

The differences are quite major though - I obtain the CV's by keeping the oscillators in a frequency/phase locked loop (the frequencies on the antennas dont change, the CV's are derived from the PLL error voltages) - The other major difference is that I use a HF VCO (in fact, all the oscillators are nearly identical series  LC oscillators)and derive the audio in the conventional (heterodyning) way.

The main reason for posting this is to show you how I get linearity and span control.. I am taking the raw error voltage (from 6) and distorting this (the output is adjustable with 2 controls, one to bend the near field, one to bend the far field) - My scheme is complex, but a single transistor be curve (if you compensate for temperature) can improve linearity greatly..

The above is probably too much for a Nano - But putting an adjustable attenuator between the CV output and the VCO input would give a span control- allowing one to adjust the field for say 3 to 7 octaves.. This may require little more than a potentiometer, and is something I think a lot of people would like!

Should just say that the diagram below hides a lot - much of what is shown is implemented in a PSoC 4 (blocks 5,13,14,10 and some of 12, and the CV from the volume goes via an ADC to give user adjustment of field size and position and drives a log DCA.. At this time I am not sure, but even the oscillators may use PSoC drivers)

Fred.

<see http://www.thereminworld.com/files/photos/15383/AE%20BD%201.jpg for the diagram - I dont want to paste it and hog your show ;-) >

" A phase detector approach (as FredM describes the likely operation of the Moog 91 above) might work better in this regard but I haven't investigated it beyond seeing the effect often in my bench experiments." - Dewster

My experience on this is quite bitter! ... Even for volume control (which is exactly what this seems to be - a "double volume sensor" system) interferers cause trouble.. One needs filtering to reduce these, and filters add latency,  acceptable on volume (as demonstrated by the fact that almost all theremins work this way) but I have not been happy with any experiments done when using this method for pitch.

But the Nano is out there, being used, and has at least one happy customer, so Vladimir must have managed better than I did - because if I had put my attempt at this method out, I dont think customers would have been happy - or at least not unless they were in an extremely quiet RF environment.

If this was being sold as a high-end high-priced instrument with bogus claims about it, I would be tearing it to shreds! ... But its not - its being honestly represented, information clearly disclosed.. Its not a topology I would adopt (particularly as the single biggest advantage of this topology - the ability to mess with the CV for adjusting span or shaping the control curve isn't being used) but that's just me probably - Let me re-engineer the Nano and you may end up with a panel full of knobs and twice the cost..)

But for the same price you could have heterodyned - this wouldn't have risked the problems above, but it would have been like and sounded like every other simple theremin... Unless you used my mixed-signal heterodyning which can give exactly the same waveforms that VCO does...

Fred