Moog Theremini!

So why heterodyne? Is it expensive to derive numbers directly from an RF signal?

(Sorry if this has been covered in "Let's Design and Build a (mostly) Digital Theremin!" - that discussion went way beyond me several hundred posts ago. Just give the brief, non-techy answer here if you can.)

"So why heterodyne? Is it expensive to derive numbers directly from an RF signal?"  - GordonC

Not expensive, but difficult to do directly without fast dedicated digital logic.  Heterodyning brings things down to audio or near audio, where you can measure frequency with a fairly slow/anemic processor. 

Heterodyning also seems to increase resolution when fast dedicated digital logic is available, so I'm starting to look into it for my own stuff (but the heterodyning will likely take place digitally).

Kudos to the Open.Theremin project for developing and revealing this method (offset rather than baseband heterodyning)!

"So why heterodyne? Is it expensive to derive numbers directly from an RF signal?" GordonC

Yeah, it has been covered ;-) ..

With RF one has a high frequency varying by a tiny percentage, perhaps 5% or less, and one has extremely small period, so if one wants to capture this 5% from the period variation by determining its 'length' and deducting the baseline "ref Osc" value, you could need a clock up in the 100's of Ghz probably.

You can devide the RF down, say to 1kHz mean - but even doing this one doesnt get the required pitch resolution (14 bits or more) without an extremely high frequency counting clock..

However - If you heterodyne the front-end oscillators, you get rid of all the unwanted data (end up converting the 5% to 100%)  - you end up with a frequency from which you can get the numbers without excessive clocks etc.

The most recent posts on Dewsters "digital theremin" thread are dealing with some of this at quite a basic level right now! - And perhaps almost every aspect of the "how do we get the numbers" issue has been discussed there - a lot of it you will probably understand better than I do, because I think your maths comprehension is better than mine! .. I also think its been discussed in the "open theremin" thread.

I would love it if someone would collate important technical matters like this which are spread all through TWs database, into documents or folders - I think this could be one of the most useful technical repositories on the web, and not just for theremins! 

Fred.

For me, there are just 2 ways to get the numbers - One is using an analogue scheme like a filter (or PLL as in my "upside down" topology) to generate a voltage from the circuit tuned to resolve the RF frequency variation, the other is to use heterodyning or variants thereof.. I know there are other ways, but they are nearly too complex for me to get my head 'round, and certainly too complex for me to implement! 

"I would love it if someone would collate important technical matters like this which are spread all through TWs database, into documents or folders - I think this could be one of the most useful technical repositories on the web, and not just for theremins!"  - FredM

FWIW, once it jells a bit more (and if it doesn't kill me first :-) I plan to document my Theremin development efforts (similarly to my Hive design PDF).

"FWIW, once it jells a bit more (and if it doesn't kill me first :-) I plan to document my Theremin development efforts (similarly to my Hive design PDF)." - Dewster

I think its worth more than you could ever imagine or predict (and BTW, your Hive document is superb!)

OT rant ->

Without the belief (whether valid or not) that these efforts contribute to human understanding and development - that perhaps in some way the time devoted is more than just an interesting  distraction for us in the present, and actually has some "meaning" or "usefullness" , I would certainly be more depressed than I am ;-)

I have pondered why we do it - For me, Its perhaps some stupid notion (and it is just absolutely stupid!) of "imortality" - Something to "leave behind" which  wouldnt "have been" if I hadnt "been" - and a way of continuing what those who came before started.. even though I often despise the results of what they "started" ;-) .. Then there is also the "present" involved - Not sure how much is about wanting to leave something of value behind, and how much is about wanting respect and glory or whatever in the "now".. I think both components are present.

 Whatever, we do it - And I dont think we really have any choice in the matter! ;-)

If I devoted the time I spend on-line to spamming, I could probably be quite wealthy ;-) - Perhaps its all down to wanting to be loved and not hated... Or at least not hated for any 'valid' (in my view) reasons..  LOL.

I had the occasion to meet Urs Gaudenz, the Swiss inventor of the Open.Theremin last weekend. We talked a lot about the Open.Theremin project and I could give it a try. 

Basically, the principle of operation of both, the Open.Theremin and the Theremini are identical: There are heterodyning LC oscillators, a micro controller which takes the raw mixer products and drives then an external DAC after looking up the values in a wavetable and multiplying it with the volume value. Since the Open.Theremin project is already out in the public domain since more than a year, I even wonder if Moog did not "pull" some inspiration from it...

There are naturally differences: The actual software version of the Open.Theremin allows to load only one wavetable at a time, thus changing the timbre means still generating a new .c file with the discrete waveform definition, recompiling the software and uploading it to the Arduino again. There is actually also no pitch correction. But on the other side, there are still 27.5kB of the Arduino's 32kB memory free. That means that the ability to select between different wavetables and many other functions might still be realized.

The most important difference for me is the price. An Open.Theremin can be built for less than 100$. And it offers a (IMHO) better long term perspective since the community can help to improve the software.

I got an Open.Theremin.UNO shield from Urs and I ordered the needed Arduino.UNO R3 today. Until it will arrive, I'll have a second and a third look onto the software (had the first look last night). Already now I've plenty of ideas how to tweak, improve, and professionalize that thing. I just don't have enough experience with micro controllers (almost none) to estimate if my theoretical ideas will be realizable in practice. Fortunately the Arduino can be re-flashed several hundreds of times...

"Louie, I think this is the beginning of a beautiful friendship."
(Humphrey Bogart as Rick Blaine in "Casablanca")

The Open.Theremin UNO instructions say that the total antenna length should be 50cm.
The board is quite small and all the builds I've seen yet are almost as compact as a B3 Standard.

If I use wires to extend the antennas, they would have to become rather short to stay within the 50cm limit.
Is it possible to get more horizontal separation between the antennas somehow?

The O.T could be a ticket to my first Theremin. I already happen to have an Arduino UNO which is currently unused. Incidentally a fresh charge of Open.Theremin UNO boards has been made available to buy just a few days ago!

I will post a full writeup of my direct experience with this instrument and my interactions with the Moog Music people.  I just need to compile all the details.  I was at the Moog booth on all four days of the NAMM show, I gathered all essential details of the instrument from the company's point of view, and I had enough time (and space) on the fourth day to understand the Theremini from a thereminist's point of view.  

I'll make a new thread here on TW to clear the air a little bit. One interesting point to mention about my personal experience is that my current impressions of the instrument are vastly different than my initial impressions (in the positive direction).  It will be comprehensive so it will take me a while to write, but fortunately I have an excellent memory...

"I got an Open.Theremin.UNO shield from Urs and I ordered the needed Arduino.UNO R3 today."  - Thierry

OMG!  Thierry's going digital!  Anyone got a thermocouple down in Hades?  ;-)

"The Open.Theremin UNO instructions say that the total antenna length should be 50cm.
The board is quite small and all the builds I've seen yet are almost as compact as a B3 Standard.

If I use wires to extend the antennas, they would have to become rather short to stay within the 50cm limit.
Is it possible to get more horizontal separation between the antennas somehow?"  - Gibarian

Not sure, but I think if you stay within the adjustment limits of the heterodyning oscillators you might be OK.

I'd be interested in user's experience with thermal drift regarding the Open.Theremin.  It seems like it could possibly drift a lot (ferrous choke w/ crystal beat source).

"OMG!  Thierry's going digital!  Anyone got a thermocouple down in Hades?  ;-)" - Dewster

Strange days indeed!