So um.. I guess a sine wave is pretty much the holy grail?

[spinoza]

I've searched google and the forum and it seems like a sine wave is pretty hard to do using simple analog circuitry. I was thinking of making another synthstick, but using a sine this time, for a "good vibrations" feel.

So is this a "forget it man, go build a fuzz face clone" or a "well it's not THAT hard" situation? And is this ( http://www.interq.or.jp/japan/se-inoue/e_ckt18.htm ) any good?

[gez]

It's not really a holy grail, just requires a little more circuitry than most LFOs and it can be a bit problematic getting a distortion free wave form over a wide range of frequencies.

Don't know anything about the 'synthstick', but by the sound of it you'd need a large range?  In that case, it's probably 'easier' to think in terms of shaping a triangle into a sine.  Then again, what you linked to might work fine...would help if you explained what you were doing/linked to a schematic.

[R.G.]

As gez points out, it depends on what you're doing.

If you mean the theremin-like thing in the song "Good Vibrations", that can be done.

The circuit you show from Japan is a classic Wien Bridge oscillator; it's a good source for low distortion sines, and is easily tuned.

The easiest-to-tune sine wave is digital. If you set up a digital shift register to recirculate a lump of all-1`s and all-0's correctly, you can use resistors on the outputs of the shift register to sum the outputs to ... a sine wave. The output looks like a stepped approximation, but with even minimal smoothing, the sine is a very good waveshape and lower harmonics than many of the analog synths. You tune the clock of the shift register, which is trivially easy. The output is big and constant amplitude with frequency.

You might like the results you get with only two ICs, a ring counter and a clock, maybe a hex inverter, under $2.00 for the chips. And it will do frequencies from LFO through ultrasonic.

[Paul Perry (Frostwave)]

It's very easy to convert a triangle to a sine, by overdriving an OTA (eg half a LM13700) by just the right amount. Only problem: you want the triangle and the supply rails to be fixed, not something to do with a battery that goe from 9 to 6 volts.
In my optical theremin, I have a waveshape control that works by progressively overdriving a CA3080 (but now I'd keep those 3080's for repairs!). As the level of the triangle increases, the output goes from triangle to sine to rounded square.

[amz-fx]

I've searched google and the forum and it seems like a sine wave is pretty hard to do using simple analog circuitry.

A sine-wave is easy to do...  a very low distortion sine is somewhat harder to accomplish...  for audio fx projects it won't make any difference if you are using an ultra-pure sine or one with 1% distortion...  in fact,  the one with the distortion might be just a bit more realistic and pleasing.  Very low distortion sine waves are sought for testing purposes...

regards, Jack

[Mark Hammer]

As has been summarized very nicely, there is an abundance of simple and easy designs for things that get really close to a "pure" sine-wave, but assuring that the waveform is ONLY composed of a single solitary fundamental with no harmonics or other content (e.g., little glitches from crossover distortion at the junction of the positive and negative halfwaves) requires something more complex.

That being said, Spinoza notes in his post that he simply wants something for a "good vibrations" feel.  I assume he wants something that gives an oscillator-type tone.  To the best of my knowledge, the theremin used on the original (and the oscillator used on the Tom Polk-built Tannerin which the Beach Boys used on the road - and which I had the good fortune to look at closely over 35 years ago) did not HAVE to use a sine wave to produce the target sound.  Indeed, I imagine a triangle or sawtooth with or without a bit of filtering would nail it to his satisfaction more than adequately.  As such, I wouldn't worry at all about the perfection of the waveform produced by a simple oscillator.  In the application described (the synthstick), the pitch instability resulting from the rather crude controller will produce a quality that quite simply overrides any distractions the waveform might pose.  That's not a criticism, merely an observation that when one quality stands way out, small variations in other qualities are glossed over.

[Paul Perry (Frostwave)]

On the subject of tannerins & derivatives..... how does THIS position sensor work???
http://www.therevox.com/et-1.php?showImage=6

[goodrevdoc]

What about an XR2206? using one of the app notes on the datasheet, i made a handy little square/triangle/sine wave function generator. Freqency (pitch) is controlled via a 1M pot which is just about the resistance of thr VCR tape on my synthstick. I don't have a scope to look at the waveforms, but they sound pretty clean and the sine can be quite "theremin-like". I was actually thinking about applying this to a synthstick type design, with a switch selectable wave output, if that makes any sense...
Only thing to get around is that i run it off of 12V, so batteries and portability would be a problem...
Hope it helps,
-justin

[gez]

On the subject of tannerins & derivatives..... how does THIS position sensor work???
http://www.therevox.com/et-1.php?showImage=6

One rests ones finger on the spike and applies pressure by varying degrees.  The more pressure, the higher the pitch of ones scream.  This is mere supposition on my part...

[Mark Hammer]

When I saw the one Mike Love was using 35 years ago, it was essentially a fancy long-scale ribbon-controller strip.  To make it easy for him to use in concert (after all, it doesn't have frets, keys, strings, holes or any of those other things layers normally have available to dientify notes), the notes were indicated in white-out stripes against the dark background.  Given that he was only using it for the one tune (Good Vibrations), there were only 5 notes indicated.

The controller Paul links to is presumably intended for people who don't know in advance what notes they have to play so it shows them all the notes.  So, like the Tannerin, it is a ribbon-style controller but with a physical interface intended for the user to treat it like a note-demarcated instrument.  The dual-tube slider simply provides smooth lateral movement.

[ErikMiller]

I'll second what justin says about the XR2206.

If you go to http://www.euthymia.org/ and click on the link for DIY resources, you'll find a schematic for a minimum parts count audio range sine-only oscillator using an XR2206. I built mine 2 years ago and am still on the first 9V battery.

The datasheet says that the chip needs more than 9V, but since the chip can't read, it will happily create a 4V sine wave using a 9V power supply.

It makes a great Tannerin, especially if you further limit the range of the frequency knob.

[RDV]

Is that Tannerin thing what was used on "Runnaway" by Del Shannon?

OT, but I'm sorry.

RDV

[RDV]

I'll answer my own question.

RDV

[Mark Hammer]

Is that Tannerin thing what was used on "Runnaway" by Del Shannon?

OT, but I'm sorry.

RDV

I was always under the impression that was an Allen/Allan Explorer organ.  But I may well be wrong.

[Eric H]

I'll answer my own question.

RDV

Great link. There's a million stories out there --I love that stuff.

-Eric

[spinoza]

Hey thanks for all the replies!

Erik's osc looks like a nice project. This is surely going inside my next synthstick!