Modifying the tone of the Etherwave Standard

UPDATE:

I have just run some simulations - ok, these are never fully reliable (particularly when Theremins are involved, LOL)..

The simulations do exactly as I expected so far.. Increasing C23 reduces harmonics progressively AND reduces signal level - at 470nF the volume should be almost inaudible with a 1kHz audio output frequency.

I am simulating the diode mixer and full VCA - The big difference between simulation and reality is that I am driving the Reference and Pitch oscillator signals from signal generators which do not allow interaction. I have some EW oscillator models I will try later to test interactions.

If only real Theremin circuits behaved like simulations do, I think life would be a lot easier - waveforms are nice, brightness and waveform controls do what they should and are not vulgar..

I will play some more - but the fact that the simulation does what I expected at least gives me comfort - Theoretically, my suggestions were sound.

I will keep editing here rather than posting new messages..

Some strange things are starting to happen as I change the input waveforms - The brightness control does seem to exhibit a 'volume control' effect when signal input is extremely low.. If I mix square waves, and increace C23 to 220nF, the signal level input to the VCA is tiny (and quite clean - sort of gentle ramp waveform at 1kHz.. but amplitude on input is about 16db below the 'normal' level so the attenuation caused should be heard!)- and brightness affects the output level but does not effect the waveform.

I find this all quite fasinating. A little over my head to be able to contribute but if an Etherwave improvement mod is born... bravo!

- I only wish I REALLY had the time to do this..
Once again, I divert and therebye delay my primary task.. However..

I have just played with R31 and C24 as suggested by Eitherspiel..

Some quite interesting results - but the bottom line is that the volume level affects the tone as I thought it would.. One is, effectively (by changing R31 and/or C24) changing the capacitance / resistance ratio.. Think of the LM13700 as a current controlled resistor - the variable resistor sitting between pins 3 and pin 5.. (this is a crude 'visualization' but it will do) .. A capacitor between pin 5 and ground will form a simple current controlled filter. Put a rresistor between pin 5 and ground, and you have a current controlled attenuator (in fact an amplifier, because it is not as simple as the 'visualization') .. put both a resistor and capacitor to ground, and you have a mix of VCF and VCA determined by the ratio of resistance to capacitance. It is somewhat more complex than this - but this is the basic idea...

In order to get significant filtering from this point, one ends up with a VCF.

Watch this space.. I will update here until someone else posts.

Eitherspiel.. Sorry, only just noticed your last posting.. page changed! [i]"Another possibility would be to change the values of C26 and R34 - these might be the best choice for post VCA processing. Again a rheostat type control to replace R34 would have a very nice effect of controlling the harmonics, no?"[/i]

Alas, No.. R33 + R34 form an attenuator - Change to either of these will affect the amplitude (volume) more than the filter TC.

Last update for tonight..

Found some interesting things about diode mixer I never thought of / realised before.. I never use them..

A distortion is added due to change in capacitance of the diode.. when both input signal are negetive, the diode is forced into reverse bias.. This decreases capacitance which accentuates the harmonics. The capacitance of the diode is directly proportional to the voltage across it (when reverse biased).. As this diode is connected to both oscillators, this capacitance modulation will feed back as a frequency modulation signal to the oscillators... This will, I think, produce sidebands ---

Eeeee- Yuk!

I have always hated the sounds I got when I tried to use diode mixers - I was never 100% sure if I was doing something wrong.. The maths of heterodyning through a non-linear element eludes my full understanding - If I dont have full understanding of something, I dont trust my conclusions about it... All I knew for sure was that, to me, a LM633 or a MC1496 always sounded better than a diode mixer.

Now I have at least one reason which makes perfect sense - at least for unbuffered oscillator inputs as seen on the EW. Two little FET's and 8 resistors is all it would have taken to buffer the oscillators..

Give up.
To have a chance of really fixing the problem one needs to buffer the oscillators - of that I am now sure.
To do the job even better would probably require replacing the diode mixer with a MC1496 at least.
The best sounds I have heard from a heterodyning mixer were obtained with a LM633 - So if doing the mod, I would pay the extra £6 and use this.

Someone want to take this on as a project? A modification board to fit in the EW... I will design the schematic and provide tech support - You do a layout which fits tidily on the EW board, and build it - and build another for me.. We test / debug together communicating on-line + phone if required.. Then, if it is truly wonderful, we sell these boards for as low as possible and donate any profit to www.streetkidsrescue.org (or you keep 1/2 the profits and donate my share to streetkidsrescue).. You handle building / selling / shipping etc, taking fair pay for the time + effort.

Any takers?

I'd buy one if it was truly wonderful. Especially if someone who already had a reputation for producing an excellent mod board for a moog theremin was involved. I can't think of anyone like that off-hand, but I'm sure they must exist. :-)

well this would be a sort of crazy expensive solution, but the moogerfooger freqbox lets you reach further down before the pitch breaks up. i tried to do a demonstration of this but my camcorder has a really crummy mic, sigh.

(youtube video here: pls excuse lack of practice) (http://www.youtube.com/watch?v=cKCgSKx8jWY)

Continued Here (http://www.thereminworld.com/forum.asp?cmd=p&T=4132&F=3&p=1)

Hewson,

Yes - despite the audio quality, the advantage of using CV as a medium to overcome the intrinsic problems are clearly demonstrated.. In order, however, to achieve this you need a Theremin with CV outs, a VCO, and a VCF ..

Certainly for new Theremin designs, the CV route has big advantages - The traditional heterodyning design has problems but also has advantages..

In my new Theremins, I hope to exploit the best of both worlds .. But for those with standard EW's and small budgets, I think a simple board which fixes the the sound generation problems at source, will be their best option.