Oscilloplasm - complex VCO pair

Earthside communication channel for 6U format devices

Moderators: Kent, slow_riot

Post Reply
slow_riot
Super Deluxe Wiggler
Posts: 4002
Joined: Tue Jun 30, 2009 8:58 pm
Location: uk

Post by slow_riot » Fri Jun 22, 2018 8:46 am

Next week I will have units ready for shipping. Contact address is on my website - ask me anything you like.

The faceplate will look like this, with deep red knobs.

Image

This module and future ones revolve around a form of synthesis that is more like additive, where specific function blocks allow generation of sinus partials through oscillation or resonant filtering, with similar ideas for low frequency control (most of the modules will have a wide enough range to be used for both). With analogue circuitry this technique means higher signal to noise mainly because subtractive generates extraneous harmonics which are ultimately removed from the output signal but nonetheless can be radiated between sub circuits by the different forms of coupling that exist. It is, however, more complex and depending on the user or the project this may be a good or a bad thing. I've found the oscillator less satisfying to use in some respects than other products, where a single voice and a simple control such as wavetable morph or pulse width can be immediately satisfying. But I don't think there is a good argument against novel synthesis techniques.

In general the production style is to simplify design where possible, but to realise it in the absolute best possible way. Most appropriate technology is selected without regard for aesthetic such as "modern" or "vintage", for example the PCB is a very modern 4 layer type with dedicated layers for power and signal 0V, but there are also NOS parts from 1980-90. Don Buchla said synths should be built from the outside in, I personally give the circuits themselves autonomy in the design process.

My favourite thing that the VCOs can do is mid to low frequency bass (drones or staccato shapes), the ability to detune partials and extremely high accuracy FM really opens up that range compared to drum voices or subtractive mono synths.

2 new modules are in the later stages of development. One is an all analogue physical modelling filter that extends the range of the classic string filter, as well as massively simplifying it compared to Haible/Moog etc. I'm also keen to explore generation of chaos, both for control and as a form of distortion.

slow_riot
Super Deluxe Wiggler
Posts: 4002
Joined: Tue Jun 30, 2009 8:58 pm
Location: uk

Post by slow_riot » Tue Jul 03, 2018 12:39 pm

OK go!

Image

slow_riot
Super Deluxe Wiggler
Posts: 4002
Joined: Tue Jun 30, 2009 8:58 pm
Location: uk

Post by slow_riot » Mon Jul 16, 2018 5:14 am

Published the schematics for this (under a Creative Commons license).

https://aether-machine.com/schematics/

User avatar
neil.johnson
Super Deluxe Wiggler
Posts: 1825
Joined: Thu Apr 08, 2010 3:51 pm
Location: Cambridge, UK
Contact:

Post by neil.johnson » Wed Jul 18, 2018 4:16 pm

Interesting design, thanks for sharing.

One quick question: on page 2 the Vee connection to the two THAT2180s - are you trying to make a constant current source with the J111 FETs?

Neil
Random ramblings :: http://www.njohnson.co.uk
SSI parts and more :: https://www.cesyg.co.uk

slow_riot
Super Deluxe Wiggler
Posts: 4002
Joined: Tue Jun 30, 2009 8:58 pm
Location: uk

Post by slow_riot » Wed Jul 18, 2018 5:09 pm

Thanks Neil!

Yes it's a constant current source, I added that part to the schematic in a hurry and got the wiring wrong on the diagram, shall correct it shortly.

Do love the guarded praise and criticism of the engineer mind though... :deadbanana:

There are a few things that I'm quite pleased with, the triangle wave is accurate in the Y domain to within around 5mV. I haven't seen the technique using a FET switch like that before. The Legendre equation calls for a term T Squared, but squaring an analogue saw directly would cause a brief 0-10V glitch once per cycle. No saw core oscillator would avoid this. I received a wonderful tip from Graham Hinton that you can generate a glitch free saw by multiplying a triangle by a square, but in fact a triangle wave squared is the same as a saw squared anyway, so that step can be omitted, and a triangle can be used for that part of the computation for perfect operation.

An important part of the design, other than simplifying where possible, was in setting the gain of signals feeding the analogue multipliers as hot as possible, +/-10V ideally. Offset, non linearity are problematic otherwise. There are quite a few places where a two quadrant multiplier could be used instead of the more expensive 4 quadrant type, the AD633 was used in those places because of it's compact footprint. It was important to squeeze everything onto one board for low crosstalk.

Probably the most important thing is that it all works, more or less as a user would expect, which probably reflects quite well on me given that several experienced engineers strongly advised against such a complex computer using analogue techniques. Plus it was the first thing I ever designed, some companies started with passive format adapters or multiples...

User avatar
thee ghost ov n_phay
Common Wiggler
Posts: 211
Joined: Sun Jan 25, 2015 11:59 am

Post by thee ghost ov n_phay » Wed Dec 05, 2018 7:19 pm

https://soundcloud.com/theda-electronic ... rst-tryout

First tryout, an active drone, double tracked and panned with a bit ov echo.

slow_riot
Super Deluxe Wiggler
Posts: 4002
Joined: Tue Jun 30, 2009 8:58 pm
Location: uk

Post by slow_riot » Thu Dec 06, 2018 8:31 am

Thanks Norman, hope you like the module. The embedded soundcloud player isn't appearing for me, direct link is here if it doesn't for others either:

https://soundcloud.com/theda-electronic ... rst-tryout

I wonder what your user experience is like? The modulation sine with first and second harmonic outputs is unique in the modular world, crossfading between the 2 when used for FM is a good workhorse patch.

The master VCO is also novel, there are some settings that are quite similar to digital wavetable VCOs (e.g. the E350 which provided the mathematical basis for it), but other chord like settings which are unique. Subtle detuning is also a unique effect and means beating or phasing effects can be perfectly controlled.

Hope to hear some good tonal patches too! Although I've personally not got bored of years of drones from the thing yet.

this beautiful mess
Learning to Wiggle
Posts: 7
Joined: Mon Oct 07, 2019 11:26 pm

Post by this beautiful mess » Thu Nov 28, 2019 8:13 am

Hi all, sorry about the weird name change, hope you can trust there was a good reason for it.

I've had a bit of a sabbatical whilst planning for the future, and figuring out the past. Not that any of that is relevant here. I'm selling 2 Oscilloplasms at a reduced price of £700 to raise funding for future designs. Contact me via email please if interested. Thanks!

Aether Machine
Learning to Wiggle
Posts: 35
Joined: Mon Jan 27, 2020 2:32 pm
Location: elsewhere
Contact:

Re: Oscilloplasm - complex VCO pair

Post by Aether Machine » Sun Jan 17, 2021 3:06 pm

New demo comprised of 2 Oscillplasm VCO modules plus digital reverb, and some drums added for compositional effect from a Reaktor Newscool ensemble.

Polite reminder that these modules are extremely capable of thick complex evolving drones.

https://aethermachine.files.wordpress.c ... ms2021.wav

User avatar
frequenzteiler
Learning to Wiggle
Posts: 47
Joined: Wed Nov 14, 2018 12:46 am
Location: Berlin

Re: Oscilloplasm - complex VCO pair

Post by frequenzteiler » Sun Apr 11, 2021 3:47 pm

im amazed by this .absolutely great to see with how much passion you make these .love the design very much ! im happy there is ppl like you out there :) im really curious how you get those faceplates done .

Aether Machine
Learning to Wiggle
Posts: 35
Joined: Mon Jan 27, 2020 2:32 pm
Location: elsewhere
Contact:

Re: Oscilloplasm - complex VCO pair

Post by Aether Machine » Mon Apr 12, 2021 6:36 am

frequenzteiler wrote:
Sun Apr 11, 2021 3:47 pm
im amazed by this .absolutely great to see with how much passion you make these .love the design very much ! im happy there is ppl like you out there :) im really curious how you get those faceplates done .
Hey thanks for the feedback!

The faceplates are acid etched, then hand painted with alcohol inks, then the most important part is to use a hardener/coating so that the finish is durable.

Oscilloplasm drones all over this album here:


User avatar
notmiserlouagain
Ultra Wiggler
Posts: 850
Joined: Thu Feb 23, 2012 11:04 am
Location: Hamburg
Contact:

Re: Oscilloplasm - complex VCO pair

Post by notmiserlouagain » Mon Apr 12, 2021 7:35 am

Very interesting sounds!
He who eats with most pleasure is he who least requires sauce.
Xenophon

Aether Machine
Learning to Wiggle
Posts: 35
Joined: Mon Jan 27, 2020 2:32 pm
Location: elsewhere
Contact:

Re: Oscilloplasm - complex VCO pair

Post by Aether Machine » Mon Apr 12, 2021 9:25 am

Thanks. Patch notes are: 2 Oscilloplasms, a Triadex Muse, some Bugbrand sequencing modules and effects processing (grain delay and reverb) via Reaktor.

The Triadex Muse does the single VCO lead, and the Oscilloplasms do the layered drones, the Bugbrand subdivides the shift register outputs from the Muse to get slower sequences that follow the "lead".

The focus on FM has still held interest for me, although it can be fiddly, more so due to the amplitude compensation circuit which causes compression artifacts with inharmonic settings.

Aside from FM it would also pair very well with a frequency shifter as the primary oscillator provides complimentary 90 degree phase shifted outputs.

Post Reply

Return to “Aether Machine”