Building a Theremin

Hi FredM!
I´m working in a analogue Theremin.
My theremin hasn´t to be a good theremin. it has to be something that works well enough. I´ve built a wien bridge oscillator as the fixed one, and I´ve have tried also to build the variable one as the hotrodEtherwave does, but I´don´t know how to connect the variable coil(it´s TOKO and I couldn´t find the configuration of the pins...?)do you know if the output of this oscillator is a sine wave? after this i want to mix these two signal with the mixer AD633... after that the filter and then the amplifier.
what do you think? very complicated?
If you can explain me an easier way I will follow it...
i don´t want to follow the idea of the F-V converter and the function generator. I want something simple which can demostrate I really undestand how a theremin works.

Hi Esther,
ok.. You are wanting to build a conventional analogue Theremin, with no real concern for its performance in terms of playability etc.. You just need something to demonstrate that you understand the principles and can design?/build it.. Yes?

You have "built a wien bridge oscillator as the fixed one" and are now looking for a suitable "variable one"..

You intend to do the heterodyning "with the mixer AD633" (this is a good choice for a quality Theremin, but a bit expensive and certainly "overkill" for a "proof of concept" Theremin - but it is easy to use.)

You dont want/need a "F-V converter and the function generator"

You ask: "what do you think? very complicated?"

Answer: Not at all complicated! There are loads of designs on the web, and pages of documentation giving detailed information and schematics to do exactly what you are asking.. You have complicated things because you never really did the groundwork and research before you started building and asking questions - And you have made your (and everyone elses) life more complicated by not getting your thoughts together and working out what you wanted to do, earlier.. This is fine - But learn from it! An hour of sitting down and planning can save many frustrating hours later.

EW Theremin (http://www.cs.nmsu.edu/~rth/EMTheremin.pdf) is a link to EW construction article.. EW Coil (http://www.thereminworld.com/forum.asp?cmd=p&T=3790&F=3) is a link to a TW thread where the coil connections are detailed.

This should give you everything you need for the EW Oscillator - or you could use the Smirnov oscillator.. you now have links to all the resources you need.

I will be too busy over the next few weeks to spend as much time on TW (I need to spend more time to try to raise money for Street Kids Rescue (http://www.streetkidsrescue.org/),As the economy has caused a sharp drop in donations, and made the situation for the children more difficult).

Good luck,
Fred.

Hi fred!
As I told you, I´ve built a wien bridge oscillator as the fixed one. the problem now, i can reach the range of frequencies that i need. when i calculate the values for 200kHz, i only get 97khz, and when i try to change the components sometimes the signal disappears. I´m using an 741opamp and some diodes for the distorsion like appears in this link :
link (http://www.calvin.edu/~pribeiro/courses/engr332/Handouts/Wein%20Bridge%20Oscillators%20Presentation-Darren.ppt) can you give some idea about what it´s happening?
thank you very much



[i]fixed the big link TW-Staff[/i]

As Fred seems to be too occupied, I'll try to give you an answer in his place: The 741 opAmp is too slow. As you can see in the above cited PowerPoint file, it has already difficulties in maintaining stable oscillation at 15kHz...

Try a TL071 (rarely available, so take the half of the dual version TL072). It also costs only a few cents but has all properties to give you stable results at 200kHz.

Hi Esther,

Your basic wien-bridge circuit looks like it should work.. But I do not understand what you are trying to do in the modification to the circuit.. Z2 must feed back to the +ve input of the op-amp.. But you have this grounded! - I presume you are showing this for purpose of analysis - but it doesnt actually prove anything - U2 out will be a low Z source of 0V!

Following the link, I do not see the distortion you are talking about.

>>"when i calculate the values for 200kHz, i only get 97khz, and when i try to change the components sometimes the signal disappears"

The first thing I would try, is changing the opamp to something MUCH faster.. The 741 slew rate can be as slow as 300mV/us (GBP 0.43MHz) and I would not even try to to do anything above 50kHz (certainly nothing above 100kHz)with a 741!

To give you an idea - I have abandoned using opamps for HF oscillators because they are too expensive - a couple of transistors costing <10p each can make a stable oscillator (either using an inductor, or constructing a wien bridge) and an opamp doing the same function reliably could cost £2. The cheapest opamp you may get away with is the TL071 which has a slew rate of 8V/us (GBP 2.5MHz) - but there are many others.. I tend to use the LM61xx family as they have near R-R I/O and some are able to drive highly capacitive loads - But they are expensive if bought 1 off.

I have not checked over the values of components in your circuits, or confirmed the operation of your diode limiter.

If you do not have a good reason for adopting a wien bridge oscillator for this function, I would abandon it, and go for a duplication of your pitch oscillator.

Good Evening Thierry! ;)

We must both have been replying at the same time.. Good to see we are still like minded! - Ps - Looking forward to hearing about how the SC kit goes.

So Esther.. Both Thierry and I agree - Change the opamp! (I sort of wonder where anyone gets 741's these days - and more to the point, Why?)

By the way, is that a good idea to take two different oscillator designs for the fixed and the variable pitch osc?

I'd rather use the same design in order to have equal properties concerning temperature drift and so on...

@Fred: Although I had a lot of other work to do, I started with the SC Kit yesterday. It took me 1 hour and a few minutes to place and solder all components to the circuit board. Since all was clearly packed and described, there were no problems. And so I expect it working from the beginning. The only trouble I encountered was with my eyes and the 5 band color codes on blue ground of the resistors. I remember times when I was younger, needed no progressive glasses, and resistors had 4 broader bands on a beige ground...

Perhaps I'll wire the external hardware later but I have still another job to do before.

[img]http://www.tfrenkel.com/picserv/cb_top.png[/img]

[img]http://www.tfrenkel.com/picserv/cb_bot.png[/img]

>>[i]"By the way, is that a good idea to take two different oscillator designs for the fixed and the variable pitch osc?
I'd rather use the same design in order to have equal properties concerning temperature drift and so on..."[/i]

I agree Thierry - And also, having 2 identical oscillators makes setting signal levels, and debugging, far easier - There are good reasons why this is the common way of doing it..
But - building a good, stable theremin is not the object of this project (as I understand it) - the object is to pass some exam / assessment.. We have not been given the specification for the assignment, so I assume a wien bridge oscillator is required (?) somewhere..

>>[i]"I remember times when I was younger, needed no progressive glasses, and resistors had 4 broader bands on a beige ground..."[/i]
Yeah! me too... Many years ago! Now I take great care to keep my resistors in labelled containers, and use my little Peak Atlas LCR40 (passive component analyser) to 'read' resistors capacitors and inductors for me when I get them muddled

That SC Theremin looks real tidy - look forward to hearing more..

I imagine that heterodyning two waveforms that vary slightly in different ways from being perfect sine waves (i.e. produced by different sorts of oscillators) will tend to produce a more harmonically rich lower band than either alone.

It could be interesting to try, although I note on wikipedia that the Wein Bridge is described as a precision sine wave oscillator, so perhaps not the best choice for the experiment.

Gordon, you are absolutely right. Mixing (multiplying as some mathematicians say) whatever RF signal with an almost pure RF sine wave will give an almost pure sine wave in the audio band.

But as the Etherwave standard demonstrates, the wave shaping may be done later on the audio side by adding a kind of "controlled" distortion.