Standalone integrator and non-linearity modules for exploring chaotic systems

Buchla, Serge, Studio.h, Northern Light Modular, Keen Assoc., 1979, Vedic Scapes, etc. Banana systems
Be sure to look into MANUFACTURER SUB-FORA as well..
Post Reply
User avatar
rumblesan
Learning to Wiggle
Posts: 29
Joined: Sun Aug 30, 2020 8:37 am
Location: Bristol, UK
Contact:

Standalone integrator and non-linearity modules for exploring chaotic systems

Post by rumblesan »

Hey all,

a couple months ago I got interested in exploring some of the chaotic circuits on Ian Fritz website, and instead of just building them as their own thing, I decided I'd make some modules for my system with the basic integrators and non-linearity... thing... that way I figured I could just patch up different configurations and play around a bit more easily.
It sat on the back burner for a bit but I've just finished up the second revision (fixing a few stupid mistakes) and I think they're at a point where they're worth sharing.

At the moment there's two modules.
One is just a dual, simple integrator.
The other is a dual integrator, with VC control over the rate, and a separate inverter and a x(1-x^2) non-linearity circuit that ian details on his site.
IMG_2129.jpg
Here's an example of the non-linearity working on a triangle wave that may well
https://photos.app.goo.gl/LySFrqD1sUiCC8Wh6

If people are interested then I'll try to get some examples of the XY oscilloscope output for some of the circuits ian outlines on his site and stick them here as well.


I'm not really sure how I should go about making these available, as I'm using bits of Ian's circuit designs, and he's not gotten back to the email I pinged over, so for the time being I figure I'd just make the gerbers available on the github repo.
If anybody has any thoughts/issues then let me know I guess.
https://github.com/rumblesan/chaos-suite
There's gerbers for front panels as well.

Hope that's of interest to people.
User avatar
KSS
Super Deluxe Wiggler
Posts: 10430
Joined: Mon Jan 25, 2016 7:28 am

Re: Standalone integrator and non-linearity modules for exploring chaotic systems

Post by KSS »

:yay: Well done! :tu:.
User avatar
the bad producer
Super Deluxe Wiggler
Posts: 2669
Joined: Tue Mar 17, 2009 8:43 am
Location: Clapton Pond

Re: Standalone integrator and non-linearity modules for exploring chaotic systems

Post by the bad producer »

Nice!
atomd
Wiggling with Experience
Posts: 254
Joined: Tue Nov 02, 2021 5:47 am

Re: Standalone integrator and non-linearity modules for exploring chaotic systems

Post by atomd »

I'd be glad if you'd link source as I'm curious how you did x(1-x^2)
Scrog1er
Learning to Wiggle
Posts: 37
Joined: Mon Jul 19, 2021 5:43 pm

Re: Standalone integrator and non-linearity modules for exploring chaotic systems

Post by Scrog1er »

Keen on this. Thank you!
User avatar
rumblesan
Learning to Wiggle
Posts: 29
Joined: Sun Aug 30, 2020 8:37 am
Location: Bristol, UK
Contact:

Re: Standalone integrator and non-linearity modules for exploring chaotic systems

Post by rumblesan »

atomd wrote: Mon Oct 10, 2022 4:28 am I'd be glad if you'd link source as I'm curious how you did x(1-x^2)
x(1-x^2) is pulled pretty much verbatim from Ian's site so full credit to him.
https://ijfritz.byethost4.com/Chaos/ch_close.htm
User avatar
rumblesan
Learning to Wiggle
Posts: 29
Joined: Sun Aug 30, 2020 8:37 am
Location: Bristol, UK
Contact:

Re: Standalone integrator and non-linearity modules for exploring chaotic systems

Post by rumblesan »

Right, bit more information, an example patch and some sounds. sorry for the delay, have been having a reshuffle at home and the serge was out of action whilst it was relocated.

So firstly, Ian has gotten back to me, and is happy for me to distribute these as long as it's small numbers for people DIYing stuff. My plan is to sell them direct to people who DM me on here, and not putting them on pusherman/synthcube. In the unlikely event there's hundreds of requests then I guess I'll have a chat with him to sort something else out, but I don't really see that being the case because these feel pretty niche. Possibly I'd just create a version without the X(1-X^2) circuit to distribute but I think that's probably the most interesting thing on here.

Which brings me to the next thing. I've spent a bit of time this weekend getting to grips with these, using them in the driven double well example on Ian's site.
https://ijfritz.byethost4.com/Chaos/ch_cir1.htm

They're definitely a lot of fun, and there's interesting stuff that can be done with them, but it requires a very fine touch a lot of tweaking, and probably a scope. I find it difficult to not have one or both of the outputs slamming into the voltage rails whilst oscillating for example, and getting the nice double attractor images has proved pretty elusive so far. It's not clear to me how much of this is my patching, or requires more work, or if there's something fundamentally wrong with the design of these (I've not trimmed for or even thought about any DC offsets in the lm13700 outputs for example)
Of course none of this is necessarily a bad thing depending on your point of view, just that I think a lot of people interested in chaotic oscillators and signals may be better served by getting a complete ChaQuO/Hypster/etc module.


Anyway, with that in mind, have a patch and some sounds.
IMG_2143.jpg
This is essentially the double well example.

Integrator A -> Integrator B -> Non-Linearity -> Mixer + Driving Oscillator -> Integrator A

The scaling buffer is being used for negative feedback/damping around Integrator B, as well as controlling the rate on both integrators simultaneously and letting it be modulated slightly by a triangle LFO from a cycling slope. The driving oscillator here was the variable sin/saw output from the CGS compact oscillator, with the volume control for that on the mixer corresponding to the drive control in Ian's circuit.
IMG_2141.jpg
I should try to get some better images of the scope, but I was recording the sound at the same time and so was busy tweaking as well hah.


Recording as well, X output to the left channel and Y output to the right.
https://soundcloud.com/rumblesan/integrator-chaos

I'll periodically update here as well with other interesting things I find and patches I make using them. Might try patching up a state variable filter later and see how that goes.
User avatar
Joe.
Wired for sound
Posts: 5840
Joined: Sun Jan 30, 2011 10:58 pm

Re: Standalone integrator and non-linearity modules for exploring chaotic systems

Post by Joe. »

rumblesan wrote: Mon Oct 17, 2022 10:12 amSo firstly, Ian has gotten back to me, and is happy for me to distribute these as long as it's small numbers for people DIYing stuff. My plan is to sell them direct to people who DM me on here, and not putting them on pusherman/synthcube.
Thank you for following the Forum's guideline's on Copyright & Intellectual Property, contacting the designer like you have. It's great when people's designs are respected and acknowledged.

When you are getting a little closer to a final board don't forget to review the updated Forum Rules, there's been some recent changes with DIY group buys as well as 'expressions of interest'.

Looks like a really well thought out project so far :tu:
User avatar
rumblesan
Learning to Wiggle
Posts: 29
Joined: Sun Aug 30, 2020 8:37 am
Location: Bristol, UK
Contact:

Re: Standalone integrator and non-linearity modules for exploring chaotic systems

Post by rumblesan »

Joe. wrote: Tue Oct 18, 2022 2:40 am When you are getting a little closer to a final board don't forget to review the updated Forum Rules, there's been some recent changes with DIY group buys as well as 'expressions of interest'.

Looks like a really well thought out project so far :tu:
thankyou for the heads up, I hadn't seen these updates but all makes sense. I'll keep this thread to being just my investigations about what I get up to with the module and the ideas Ian put out, and once it's at a point where I think any kinks have been knocked out I'll create a separate thread anouncing and linking to my website where interested parties can get in touch.



On that note, have been digging in some more this morning and I think I'm starting to find some sweet spots. Having a voltage processor around each integrator to provide some negative feedback and a little bit of offset compensation, as well as greatly increasing the negative feedback after the non-linearity gives some really stable results.
DS1Z_QuickPrint4.png
DS1Z_QuickPrint1.png
DS1Z_QuickPrint3.png
Still requires tweaking to balance it up but it's all less twitchy.
Above plots are all the X-Y output straight out the integrators, below is part of the waveform with the output of the non-linearity as the red trace.
DS1Z_QuickPrint5.png
Post Reply

Return to “4U Format Modules + EMS”