Theory of operation info?

[Consul]

I was just wondering if anyone had ever discussed or written about the theory behind the hows and whys of the concepts behind the Lunettas. I just discovered these (and re-discovered electro-music.com) thanks to the Make Blog. I am, however, a big-time physics and math nerd, going for a degree in physics, and am most certainly not afraid of the occasional differential equation or discussion of Kirchoff's Law or other such things.

Eventually, though, I do want to get some parts and start playing around. Until my workspace and money issues are worked out, though, I thought I'd try to learn as much as I can.

Thanks for the help, and for the great forum, and especially for these simple synth ideas.

[RF]

Hi Darren -
Nice to have you here!
Interesting - I just read a post about Lunetta's where the author was saying essentially "I need to think less and build more" Now here you are bucking that sentiment.

I don't know of anyone who has written much on the math/physics of these things. I've been searching out everything I can find on them and have only seen a few references from Stan Lunetta interviews on the math - and those were pretty sketchy.


I look forward to seeing what you come up with - Have at it!

bruce

[Consul]

Well, really, I was just hoping for an explanation on how these ideas work the way they do. I've never dove (dived? diven?) into digital logic electronics theory in any great depth, other than what I need to know for programming computers (which is really different knowledge).

But like it's said a lot around here, 4000-series chips are cheap as, well, chips, and it would probably easy enough for me to start playing around and figure a few things out. Thanks for the reply.

[droffset]

If someone wants to write articles here are some starting points.

My understanding is spotty at best, but I'll share what I can. If I type things you already know then skip ahead. Also correct me if I'm wrong.

Digital logic is basically:
- High and Low
- 1 and 0
- Full Voltage and No Voltage
? Rail to Rail

A Squarewave is an inherently digital function then?
A squarewave can be a clock, providing a timing reference.
A squarewave can be a musical tone, with the pitch determined by it's frequency.

A 40106 IC can make 6 squarewave oscillators with some added components. Others can too, but 40106 is very popular because it's easy to use.

So the different CMOS ICs each have a function. You input something and something else gets spit out the other end. There might be more that one input and more than one output. The outputs can either change/modulate) the pitch(frequency) of an oscillator or the same output can be used to clock another IC.

A Lunetta machine, then, is a way to study how these different functions can interact.

Some ICs just count up to a certain number. The 4017 IC counts up to ten over and over again. The input is a clock, just a squarewave of a tempo that you choose, and there are ten outputs (0-9). So every time a clock ticks it'll step through each of the outputs and make it go High. People build sequencers with this.

Some of them just add two signals together, some divide a signal by a certain amount(4040 is my favorite), some act as switches depending on whether the input is high or low. It can get really complex.

In a Stanley Lunetta Interview posted on this forum he described using the counter as the frequency base for a note's pitch.

People have also converted the high/low outputs into a variable stepped voltage stream that can interact with a voltage controlled oscillator.

It's a node-based approach to building a complex system. It's a fantastic educational tool because it's tactile, it's visual, you can get tangible results and a great deal of understanding in a short amount of time. It's fun because it's both simple and challenging and can make super cool sounds.

It's like Lego...Really awesome Lego...

[RF]

I misunderstood, Consul ;0
Drofset - nice job on the basics.
Yea, I really encourage you to build a few of these things. It's kind of hard to wrap your head around parts of it (at least it was for me) until you start using these chips in this way.
Ask questions and we'll all do the best we can answering them, though.

bruce

[Uncle Krunkus]

It aint rocket science, so I don't know if you'll find any complicated maths or physics lurking in these circuits. Ironically, rocket science is not as complicated as a lot of other stuff which has been developed since. Rocket science hasn't really changed very much over the last 40 years, (basically because we stopped funding anything interesting which used rockets! )
Anyway,
Certain input states lead to certain output states with respect to time.
I think the real trick is to know the functions of as many chips as possible. This is where the "just get building and see where it leads you" philosophy pays off. There are some chips which can do things which you may never have thought of until you start playing around with them. Like many systems which are made up of simple building blocks, complex results are just a matter of time.
BTW, just because these are "digital" circuits, don't think they can't produce some very complex organic outputs. I've recently discovered this while playing with my ReAnimator circuit. The main stumbling point is how many chips are you willing to employ!
http://electro-music.com/forum/topic-33381.html

[Consul]

Thank you all very much for the information. I'm beginning to see that Kirchoff and differential equations, both essentially analog domain ideas, might be the wrong approach here. Putting together Lunetta Systems (I'm sure he'd love that term ) is definitely much more an exercise in logic and puzzle solving. And the first important puzzle is having an understanding of what each piece can do, as Uncle Krunkus has stated.

It starts with square waves, but it's easy enough to see how to get to pulse waves from there. PWM could be next, then.

I like thinking in terms of nodes in a directed graph, since that's something my mind can already grasp. Linear algebra was a favorite subject of mine. That's probably the approach to take with this.

[blue hell]

[amplex]

droffset: nice post!! really great explanations of the lunetta basics.

Blue hell: That wouldnt be the same Moore as Gordon Moore cofounder of intel and Moores law of exponentially accelerating IC transistor count would it? (kurzweil reader here =D)

mathmatically describing whats happening in a lunetta (or any other piece of complex electronic equipment) seems like more than a headache to me, even broken into stages. i applaud your attempt to explore the base mathematical/physical functions of these lunettas, if you figure out any mathmatically relationships that're easily explainable plz post it =]

if you are only interested in the workings of these lunettas, im sure they are simulatable in some sort of computer software, not really sure about circuit designing software so far, but there is probably something perfect for your needs out there already, and possibly free.. anyone? Would ChucK labs work? I know that you can build basic synths and it is basically an audio programming language (Csound comes to mind), and there is even a GUI for it for Mac. I've never really used either of them though, so i'm basically worthless in this thread =]]

[droffset]

I hope you found that useful, hehe.

There's so much info on this forum that a really good introductory tutorial document could be put together, sort of a text, starting with general concepts and working through the first 3-4 blocks to build for your own Lunetta. After that you have enough knowledge to find your own path.

If you wanted to get into really deep theory you should look at some of this stuff through an oscilliscope and analyze the relationship dynamics between frequency and pitch, and how at really high speeds you can make sounds that seem like they're going through some crazy filter when it's not.

Or you can just build it and have fun.

[blue hell]

[Inventor]

[blue hell]

[Inventor]

[Consul]

I guess it's time to find some bulk sources for banana cables and jacks, 4000-series chips, and the like.

Actually, what I should really do this weekend is get the basement finished so I can have my lab/workshop ready.

[RF]

Glad to hear you are going to do this! Place the orders...then work on the basement.

Maybe you already have a good source, but FWIW I use
Digi-Key for the chips & resistors
Jameco for Pots and LEDS
Mouser for EF Johnson Banana Jacks
Radio Shack for all my cell phone needs.

Buy in bulk- or at least in larger numbers and you'll save a bunch.

bruce

[Consul]

Radio Shack truly has fallen from grace for home electronics hobbyists. Still, they're a convenient source of piezo discs, which I play around with a lot, and not just for musical uses.

Inspired by the idea of nodes in a graph, I think I'm going to build mine as small modules in Altoid tins. Discrete nodes which can then be connected in every which way. Powering them is the only issue I see. I'll probably build a common power bus which can feed power via minijacks or something like that. Just plug in one cable, and voila! Power.

[RF]

"You've got questions...We've got Batteries" - Radio Shack.
It's almost painful for me to make fun of them. Many hours of my formative years were spent in the parts section of Radio Shack. Not a month went by that I didn't whip out my "Free Battery a Month" card to get a 9V battery to power whatever my current project was.

I love those little Altoids boxes..They are just begging to be used for something like this. Good idea.
The Johnson banana jacks won't fit in those, though - except maybe if you mount them on the sides. I recall someone mentioning mini-banana's somewhere in the forum - maybe they would fit? You could also use 1/8 jacks - those would work.

Keep us posted on your progress

bruce

[Inventor]

You know, the notion of putting separate Lunettas into Altoids boxes sounds an awful lot like an eChucK project (OK, I'm getting to sound like a broken record but that's what's on my mind this morning)...

[Consul]

What would be really cool is if little Lunetta modules could be induction powered. Just plop them down on the induction pad, and they "magically" get the juice they need. Then the only wires will be the banana cables running signals everywhere.

[Consul]

And then integrate accelerometers so that moving the boxes around affects the patch.

Okay, I'll stop now.

[droffset]

Nifty idea, i wonder if the famous BEAMbot photovore circuit would work, some big capacitors charge up on sunlight and periodically blurt out some bleeps and bloops.

Now I really want to try it, heh.

[Rykhaard]

[Inventor]

Very resourceful scrounging, Rykhaard. Me, I try to usually buy from DigiKey, but then I pay too much, don't I? Still, a great supplier for lots of goodies.

[Rykhaard]