Author |
Message |
frijitz
Joined: May 04, 2007 Posts: 1734 Location: NM USA
Audio files: 54
|
Posted: Sun May 27, 2007 1:27 pm Post subject:
10-Phase Chaos |
|
|
I've finally gotten around to putting up a schematic and some sound clips for my 5-stage cyclic hyperchaos generator on my website.
This rather large system comprises 5 identical damped integrators and cubic-nonlinearity circuits. The inputs and outputs are patchable, so loops from 2 to 5 cycles can be set up. Each stage's nonlinearity may be used or not. Additionally, some extra, variable cross-coupling can be switched in.
The frequency and damping of the system are under full voltage control, although all the stages' controls are all ganged together.
When the full system is set up for periodic oscillations, the outputs are 36 deg apart in phase. With inverters on each output the system is a decature (10-phase) oscillator.
I've includede clips showing the effect of using different phases in synth patches, including a cool patch that sounds like random bleeps until you change the phasing, at which point you hear that the the parts are correlated differently -- a pretty dramatic illustration of the difference between chaotic and random control signals.
I've included patches using the continuous chaotic outputs as well as patches including chaotically-timed events using my TGTSH module.
And I'll bet nobody has seen Lissajou figures like these!
Ian
http://home.comcast.net/~ijfritz/ch_cir4.htm |
|
Back to top
|
|
|
blue hell
Site Admin
Joined: Apr 03, 2004 Posts: 24079 Location: The Netherlands, Enschede
Audio files: 278
G2 patch files: 320
|
Posted: Sun May 27, 2007 1:47 pm Post subject:
Re: 10-Phase Chaos |
|
|
frijitz wrote: | And I'll bet nobody has seen Lissajou figures like these!
|
Dunno, but I didn't - pity you don't have a color scope
Some nice samples as well ! _________________ Jan
also .. could someone please turn down the thermostat a bit.
|
|
Back to top
|
|
|
andrewF
Joined: Dec 29, 2006 Posts: 1176 Location: australia
Audio files: 4
|
Posted: Sun May 27, 2007 3:59 pm Post subject:
|
|
|
(pre-morning coffee = vocab is minimal)
really amazing stuff, the poor old 'scope is working overtime
do the Int and NL outputs of each stage patch into the Ins +/- of the next?
or am i missing something? |
|
Back to top
|
|
|
frijitz
Joined: May 04, 2007 Posts: 1734 Location: NM USA
Audio files: 54
|
Posted: Sun May 27, 2007 4:41 pm Post subject:
|
|
|
andrewF wrote: | do the Int and NL outputs of each stage patch into the Ins +/- of the next?
or am i missing something? |
For the basic cyclic patch each NL out connects to the (+) In of the next stage, with the last stage coupling back to the first (loopwise).
But you don't have to use all the NL circuits, ie you can couple some of the (+) Integrator Outs directly to the following (+) In.
Or you can put an inversion in the loop by connecting one of the NL Outs to the next (-) In -- this usually gives you periodic signals rather than chaotic ones.
Or you can switch in extra coupling paths with a pair of switches (not shown) one for selecting the stage to couple to and one for the polarity of the coupling and a pot to set the coupling level.
I should have mentioned that the inverters for obtaining all 10 phases (not shown) are taken from the Integrator outs, so there are 2 inputs and 3 outputs for each stage.
On my actual box, the 5 I/O points for each stage come to panel jacks, so the configuration of the loop is set with patch cords. This means that you can add external driving, or break the loop and insert other circuits/modules, etc. |
|
Back to top
|
|
|
toppobrillo
Joined: Dec 10, 2005 Posts: 766 Location: oakland, ca
G2 patch files: 1
|
Posted: Sun May 27, 2007 11:36 pm Post subject:
|
|
|
super stuff
very nice sound samples too |
|
Back to top
|
|
|
StephenGiles
Joined: Apr 17, 2006 Posts: 507 Location: England
|
Posted: Mon May 28, 2007 2:08 am Post subject:
|
|
|
Just so a humble guitarist can share this - can you please explain what is going on? |
|
Back to top
|
|
|
frijitz
Joined: May 04, 2007 Posts: 1734 Location: NM USA
Audio files: 54
|
Posted: Mon May 28, 2007 7:00 am Post subject:
|
|
|
StephenGiles wrote: | Just so a humble guitarist can share this - can you please explain what is going on? |
Yes, this is a topic that gets super geeky very quickly. I have a short introductory page on my site, which I tried to make as simple as I could. You might try starting there:
http://home.comcast.net/~ijfritz/ch_over.htm
Or just google around for "chaos" "chaos music" etc. |
|
Back to top
|
|
|
StephenGiles
Joined: Apr 17, 2006 Posts: 507 Location: England
|
Posted: Mon May 28, 2007 7:44 am Post subject:
|
|
|
Many thanks, I've regretably ignored this branch of music electronics over the years, but time to change that! |
|
Back to top
|
|
|
frijitz
Joined: May 04, 2007 Posts: 1734 Location: NM USA
Audio files: 54
|
Posted: Mon May 28, 2007 8:13 am Post subject:
|
|
|
StephenGiles wrote: | Many thanks, I've regretably ignored this branch of music electronics over the years, but time to change that! |
Most, if not all, of the musical applications of chaos theory are in the area of computer music. A few folks have made electronic chaos generators in the past, but this hasn't really caught on with any composers that I know of. I think there is a lot of potential in this area because of the mixed regular/irregular nature of the signals that you can get. I'm sure people can go musically well beyond what my simple demo clips have illustrated. |
|
Back to top
|
|
|
StephenGiles
Joined: Apr 17, 2006 Posts: 507 Location: England
|
Posted: Mon May 28, 2007 9:33 am Post subject:
|
|
|
Of course a great many guitarists play nothing but chaos |
|
Back to top
|
|
|
Fernando
Joined: Dec 30, 2006 Posts: 286 Location: Barcelona & Emporda, Spain
|
Posted: Wed May 30, 2007 4:53 am Post subject:
|
|
|
Ian, it is wonderful! And very inspiring, thank you!
I think 2 systems per pcb would be great =:)
I'd like to ask... could this be replicated or at least approximated using a vc slew limiter + VCA for the vc integrator and something for the non-linearity like a tube or diode soft clipper? (I'm thinking in Ken's tube vca, wave multiplier, etc: CGS27/29/52/65) |
|
Back to top
|
|
|
frijitz
Joined: May 04, 2007 Posts: 1734 Location: NM USA
Audio files: 54
|
Posted: Sat Jun 02, 2007 8:36 am Post subject:
|
|
|
Fernando wrote: |
I think 2 systems per pcb would be great =
I'd like to ask... could this be replicated or at least approximated using a vc slew limiter + VCA for the vc integrator and something for the non-linearity like a tube or diode soft clipper? (I'm thinking in Ken's tube vca, wave multiplier, etc: CGS27/29/52/65) |
Hi Fernando!
Thanks for your interest.
PCBs, hmmm....
I think I'll at least wait to see how people do with the other two boards. Quite a few folks have bought them, but as far as I know only a handful have actually built them out and used them.
One issue that would have to be resolved is that the different stages need to be carefully trimmed to make them all the same. This includes careful offset compensation as well as selection of OTAs for low offset variation. Also, the zeners should be matched and some adjustment made for differing outputs. I spent a looong time doing all this!
Your idea of patching up something similar is interesting. A soft clipping nonlinearity probably wouldn't work, though. But a wavefolder might -- after all, that's what the x-x^3 function I've been using is.
Ian |
|
Back to top
|
|
|
Fernando
Joined: Dec 30, 2006 Posts: 286 Location: Barcelona & Emporda, Spain
|
Posted: Sat Jun 02, 2007 9:26 am Post subject:
|
|
|
frijitz wrote: | Thanks for your interest. :D |
thanks for the inspiration!
frijitz wrote: | PCBs, hmmm....(...) |
I'm shure there's many people interested.
The offset compensation means to adjust a trimmer or something more tedious?
The precision of the circuit is critical to get it running at all or it could run but just less adjusted to the equations?
frijitz wrote: | (...)But a wavefolder might -- after all, that's what the x-x^3 function I've been using is. |
that comment triggered my idea. So it might work... great! |
|
Back to top
|
|
|
frijitz
Joined: May 04, 2007 Posts: 1734 Location: NM USA
Audio files: 54
|
Posted: Wed Jun 06, 2007 8:17 am Post subject:
|
|
|
Fernando wrote: |
I'm shure there's many people interested.
The offset compensation means to adjust a trimmer or something more tedious?
The precision of the circuit is critical to get it running at all or it could run but just less adjusted to the equations?
|
The offset adjustment was pretty tedious and I'm not even sure I can remember how I got it to work. The problem is that the individual stages are integrators, so by themselves they run off to the rails. I think I opened up the damping feedbacks, added a resistor across each integrating cap and zeroed the offsets, one stage at a time. Then I got a stable periodic signal with all five stages and adjusted all ten offsets until the lissajou figures were the same for each pair of stages.
You don't get symmetric patterns without careful offset adjustment, unfortunately. Maybe if I used CA3280's for the OTAs it would be less of a hassle.
I think I will still wait to see how folks do with the simpler circuits before getting into a board for this one. (Not much response so far.) |
|
Back to top
|
|
|
|