RCA Sounds / Recordings / Samples

Christopher,

There is undoubtably a strangely high level of noise "shimmer" on the waveform - and some (not too unpleasant, in my view) distortion.. But I am impressed nonetheless.. Two transistors and an op-amp? Not bad!

Can I ask for a few technical details:

1.) Was this directy recorded, or recorded via other processing..

2.) Is it perhaps possible that your circuit failed to completely remove all high frequency (Sum) components from the signal.. ?..

If there is noise modulating the frequency of either HF oscillator, the result of this modulation gets easily filtered from the difference frequency*.. However, if any sum components get through to your recorder these can beat against the sample (or bias, if doing analogue recording) frequencies and this can result in exactly the kind of noise 'shimmer' I hear on your samples.

I suggest trying it again, but placing a good filter on the output of your circuit so that no signal (more than 80db attenuation) with frequency above 20kHz gets through - Then either post or send me a sample.

Tip:
If you have a problem hearing noise or signal generated from HF leakage, turn off one of the oscillators (or turn off both sequentially - ie, only run one oscillator at a time) and view the output with a 'scope or spectrum analyser.. Any leakage will show up, and you can turn your 'scopes gain right up to view low level leakage - something you cannot do if both oscillators are running and producing a (comparatively) high level difference signal. Note - Any signal you see will be about 1/2 the frequency of what you would get if both oscillators were running and producing a sum frequency - so the (actual) leakage signal is likely to be about 3db lower than the level you observe using this method.

*Unless an oscillators frequency is extremely noisy - in which case this FM noise will be demodulated and appear in the difference signal.. This is quite possible when using simple oscillators.

Christopher's hissing sound .,. my guess is using the TL082 opamps rather than TL072. His goal was to build the thing strictly out of Radio Shack parts and they don't carry the low-noise opamps.

Sadly it's nearly impossible to accomplish this goal now, since Radio Shack is hell-bent on becoming just another cellphone store :(

Don

As for "tube sound" ... since most players are not wanting to "turn all the knobs to eleven" and blast their neighbors, there's not a whole lot of use in having a completely tube amp.

Things like Marshalls and All-tube Fender amps don't really drive the output stages hard until you are playing near full volume.

The preamp stages, however, do most of the tone shaping (i.e. 2nd harmonic distortion). So you can come pretty close to "tube sound" with just a tube preamp stage. Many on this thread are doing just that. The downside of this approach is that most theremins are line-level and must be attenuated to run through these preamps. This means you are adding more noise ("resistor noise") and not running either the preamp or the theremin at its "sweet spot".

So you could use a tube stage inside the theremin just past the mixer stage, or as I propose to do, use a cathode-follower stage as the front end of an amplifier. This will be followed by all solid-state circuitry.

Theoretically, all that "tube goodness" will still be there. You are just taking your output from the cathode rather than the plate.

Don

@don_pb: [i]"The downside of this approach is that most theremins are line-level and must be attenuated to run through these preamps."[/i]

Luckily, the ART "mini" preamp I use has both mic (XLR) and line (1/4 phone) level inputs. Indeed, going through the mic input heavily overdrives it. It also has mic and line outputs, so it makes a nice DI box for stage PAs or recording consoles.

[i]"my guess is using the TL082 opamps rather than TL072" - Don_pb[/i]

I really doubt that the noise was due to the op-amp used.. There is not a huge difference between the TL082 and the TL072 (25nV vs 18nV for ON parts, but a TL082 from National Semiconductor is specified at 16nV 0.01pA)

In fact, opamps often get wrongly blamed.. Even a LM358 can have noise levels which one cannot hear - PROVIDED the circuit is optimally designed. Areas such as active filters are the ones where close attention to detail is most required, and where selection of the 'right' opamp is most important.

I would be willing to bet a Theremin that the noise in the sample was not soley the result of a noisy opamp.

I am still of the opinion that the noise was due to noise modulation on the oscillator/s..

Back in August when I had my BASIC Theremin going, and was preparing for HO2009, I got (from Gordon) the details about the theme of the music to be created in the workshops (The elements).. I thought - If I modulated an oscillator with noise, I could generate "air" sounds (wind, perhaps rain and thunder etc) .. so messed about with this idea.. Well, nothing useful or pleasing resulted from feeding controlled levels of noise into the modulation input - but I did learn just how important it is to be sure that no noise modulation ever occurs - tiny quantities of noise cause the effect heard in Christophers sample. As for the "air" effects I was seeking, the only way I could get these was by putting HF noise through a BPF tuned to the fixed oscillator frequency, and using its output instead of the fixed oscillator... This works a treat - but to be tunable requires a HF VCF.. It is one of the "extra" features I am not sure whether to include in my Theremin or not.

I'll keep this going here (http://www.thereminworld.com/forum.asp?F=1&T=4243&cmd=p&p=1) for fun and new thought posting.