NewSoundWaves Synth Meet Feb 10

Thanks to everyone who attended the New Sound Waves meeting today.

View this post on Instagram

#newsoundwaves #tinrs #synthesizer #synthmeeting #modularsynthesizers

A post shared by jono (@dj_jondent) on Feb 10, 2019 at 4:28am PST

Lots of fun.


 

Thats a "wobbler" on the left and a "Rectangular Thing" in the centre.

Edgecutter & Tuesday

wE HAD THE PLEASURE of meeting Stijn Haring-Kuipers
Principal inventor, CEO of Void B.V. and Priscilla Haring-Kuipers, Resident media psychologist.of TINRS

Terry's video synth.

A 3-trins video synthesizer, Roland V4 mixer & Edirol P10 sample player.

Double Jomox Eurorack

Furrrr Generator    complex oscillator , Stillson Hammer sequencer

Love this DIY synth

Adding anchor links within the same webpage

This is not related to synths at all but may help some who are trying to add links within
a single page when using google blogger.
esp if the page is very long.and needs a index for example.

Inside the Html mode ,write the code like:

<a href="#1">Index</a>


The link :

<h1 id="1">index</h1>


<a > contains href element to refer the link
<h1> contains the ID of href element.

An evening with This Is Not Rocket Science

A reminder ... This Sunday TINRS will be demoing and work-shopping  this Sunday.

You can bring your synths too.

FaceBook

https://www.facebook.com/events/544919402680384/ 

 Location: 107 Project Gallery.

Address: 107 Redfern St, Redfern NSW 2016

http://107.org.au/

Buchla - CV vs Audio

Have you ever wondered why Don Buchla separated control voltages from audio signals.

In virtually every other modular synthesizer system there is no division between the two.
Eurorack use 3.5mm cables.
Serge use Bananas
Moog and the rest of the 5U community use 1/4 inch jacks.
This democracy of connectors seems very practical. All voltages are treated the same and this gives increased flexibility and scope for experimenting.

So why did Don decide to separate the two

I just read an old interview of Don Buchla which may shed some light on the subject.
http://www.vasulka.org/Kitchen/PDF_Eigenwelt/pdf/096-099.pdf

"I would say that philosophically the prime difference in my (Don Buchla) approach from that of Robert Moog was that I separated sound and structure, and he didn’t. 

Control voltages were interchangeable with audio (in the Moog world).  
The advantage of that is that he required only one kind of connector and that modules could serve more than one purpose. 

There were several draw-backs to that kind of general approach, one of them being that a module designed to work in the structural domain at the same time as the audio domain has to make compromises. DC offset doesn’t make any difference in the sound domain but it makes a big difference in the structural domain, whereas harmonic distortion makes very little difference in the control area but it can be very significant in the audio areas. 

You also have a matter of just being able to discern what’s happening in a system by looking at it. If you have a very complex patch, it’s nice to be able to tell what aspect of the patch is the structural part of the music versus what is the signal path and so on.  

There’s a big difference in whether you deal with linear versus exponential functions at the control level..."
D.B. 1964.

The last point raises questions re linear vs. audio taper, bipolar vs. unipolar, and shielded vs. unshielded cables.

Another article from SOS shines more light on the subject:

"One significant difference between Buchla's approach and that of Bob Moog was his separation of the signals used in synthesis into three distinct classes. 
First, there were the audio signals, which could be generated by oscillators, or injected into the system from devices such as microphones or tape machines. 
Secondly, there were the control voltages. 
Finally, there were timing pulses, which we nowadays call clocks, gates, and triggers. By today's standards, the audio levels were quite low (about 1V peak-to-peak) and the CVs and pulses were rather hot, with a maximum voltage of around 15V."
 http://www.soundonsound.com/sos/dec05/articles/buchla200e.htm


The precise  Buchla audio & CV levels fluctuated over the decades:
100 series Buchla had audio at 0dBV .... roughly 1 volt RMS.
Pulses are 15V and CVs range between 0V & +15V.
Early 200 series Buchla used audio levels of +4dBV... Professional line level  of +1.736 RMS
Pulses are 15V and CVs range between 0V & +15V. for early 70's 200.
Late 200 Buchla  dropped the CV range to 0-10V. Audio levels stayed the same at +4dBV
Buchla 200e dropped audio back to 0dBV. Pulses are 10V. CVs are 0 to +10V.

Inspite of these changes over time, one thing remained constant : the huge gap between CV and  audio levels. It's probably a good  enough reason for not mixing  CV and audio (or at least taking extra care) within a Buchla system .

This is making me wonder what the compatibility of the new RED Panel Buchla 100 will be like with my standard Buchla. I'd love to mix the two if that was possible. It's interesting that this separation of CV & audio is gone within that system. I guess this makes it more compatible with Eurorack.

According to the Doepfer manual, the Eurorack universe runs by these guidelines:
Audio voltages produced by modules such as a VCO or NOISE are typically in the 10Vpp range (from -5 V to +5 V)
Control voltages, as produced by modulation sources like a LFO and ADSR, are typically from -2.5 V to +2.5 V (5 Vpp) for LFOs, and from 0 V to +8 V for ADSRs.
These days, there are lots of companies that have +/- 5V CV ins and outs. (10Vpp)
Trigger, Gate or Clock Signals, are typically rectangle shaped signals with voltage levels of 0/+5 V

Synovatron make an excellent series of modules that do Buchla- euro & Euro - Buchla conversions easy.
https://www.muffwiggler.com/forum/viewtopic.php?t=136727

http://synovatron.blogspot.com/

Microcontrollers & Synths

This page is an index relating to microcontrollers (MCU - microcontroller unit) and how they relate to synthesizers.
It will grow over time so keep coming back

I'm seeing  these chips time and again when building synths.

They usually take the form of a single IC that contains these basic parts:
1. CPU - a microprocessor (4-bit or more)
2. non-volatile flash memory --- this holds the program to be run.
3. RAM - random access memory (volatile memory for data storage)
4. EEPROM - Electrically erasable programmable read-only memory 
    used for non volatile storage of program data
5. GPIO pins - General purpose in/out pins - used to interface with switches, LEDs etc etc.
    These can be configured to be either inputs or outputs.

Microcontrollers are basically tiny computers.
They perform things one step at a time and use a clock to trigger each step.
The clock is an oscillator (usually 1MHz to 20Mhz).

The GPIOs are very flexible ... can be ins or outs because they are controlled with software not hardware
unlike CMOS for example where the pins have a fixed function.
They are limited only by the speed and memory of the chip ( and the ability of the programmer)

Microcontrollers are also able to bridge the divide between digital and analog. So we can use them in the
world of analog synths.

There are plenty of examples of the use of microcontrollers in synthesizers.

Intel produced the 4004 in 1971 Arguably, this was the world's first microprocessor.
This led to the famous 8080 CPU and then the IBM PC's 8088, 80286, 486 etc.

The Buchla Series 300 system was introduced around 1973. It was a marriage of Series 200 modules and a computer system comprising an 8080 eight-bit CPU, a floppy disk drive, video monitor, interfaces to the synth modules, and a music language called Patch IV (developed specifically) for the system.

Buchla 360 - octal signal source

Motorola  produced the 6800 in 1974. It's a 8-bit microprocessor
    Some of the synthesizers using the 6800 & its variants were:
    The 6800 was used in the Fairlight CMI series II. The IIx used the later 6809.
    The Fairlight series III used the 6809 & 68000
    Ensoniq EPS-16 (68000),
    Oberheim Xpander, Oberheim Matrix, ( 6809)
    PPG Wave 2.x , PPG Waverterm A (6809)
    Ensoniq SDP-1  , Ensoniq ESQ1, Ensoniq SQ80 (6809)
    Quasar M8 (6800)

Fairlight IIx

 Texas Instruments produced the TMC0281 Speech Synthesizer in 1978.
 The TMC0281 was the world's first single-chip speech synthesizer.

Zilog  made the Z80 microprocessor in 1976.
This was used in many home computers of the early 1980s  like the Dick Smith System 80 which was my first computer.  

Synthesizers that used the Z80 include: Roland Jupiter 8, Sequential Circuits Prophet 5 & 10,
      Sequential Circuits Prophet 600, Roland MC4, E-mu 4060, Roland MSQ700, Oberheim OB-8,
      MemoryMoog, Emulator I and II, Akai 2700, E-mu SP-1200, E-mu Drumulator,
      Sequential Circuits Drumtraks.

Microchip Technology  introduced the PIC 16C84 Microcontroller in 1993.
      The PIC 16C84 used a new type of memory called EEPROM
      (electrically erasable programmable read-only memory).
      You will find this type of memory used in lots of drum machines from the 90's

These are some of the most common microprocessors used in modern synths. (It's by no means a full list).

Arduino

+ Snazzy Fx Ardcore

   I believe its powered by an Arduino Uno

+ Circuit Abbey Euroduino module.

 

Atmel (AVR, ARM, Arduino, ATtiny, ATmega, ATxmega, AT89, AT90, AT91)
+  Mutable Instrument
      Branches (ATMEGA88PA), Grids (ATMEGA328P), Edges (ATXMEGA32A4U)
      Shruthi - ATMega644p , CVpal
+  ADDAC -  VCC (Voltage Controlled Computer) Atmega2560
+ Elby AVR synth (AT90S8535)
+ Modular Synthesis AVR synth ATMEGA32
+ Bastl Grandpa - Atmega328-PU
+ Buchla Format - Sputnik 244 (ATtiny88 & 84)
+ Buchla Format - 204r ATtiny84A
+ Buchla Format -227r (rev2) ATTINY84A

+ Monome Teletype AVR 32 (32-bit)

ATMEGA88PA

Seeeduino.
+  meng qi voltage memory

Seeeduino V4.2

STM32 
+ Mutable Instruments
     Tides, Peaks, Streams, Yarns, Braids, Frames (STM32F103CBT6)
     Clouds, Elements STM32F405RGT6
+ Owl modular - STM32F4
+ PER|FORMER sequencer STM32F405RGT6
+ Buchla Format - 248r MaRF & 218r
+ DAFM synth - STM32
+ Axoloti -  168MHz STM32F427 microcontroller

STM32

 Teensy
 + Ornament and Crime (Teesny 3.2)
 + 16n Faderbank (Teensy 3.2)
 + Radio Music / chord organ (Teensy 3.2 with internal DAC as the audio output)
 + Temps Utile (Teesny 3.2)
 + Malekko Varigate8+ ???
 + Orgone Accumulator
 + TELEX Teletype
 + Euroshield from 1010music

Teensy is a brand of microcontroller development boards created by PJRC and designed by the co-owner, Paul Stoffregen.  Arduino +  32 bit ARM-based microcontrollers = maximum I/O capabilities

Teensy 3.2

PIC (MIcrochip)

+ BeasteK - Pixel Drum

+ Midi2CV - Hexinverter 

Microchip has created its own family of MCUs and baptized them as dsPICs.

The name PIC initially referred to Peripheral Interface Controller

They are also known as digital signal PI controllers.

Early models of PIC had read-only memory (ROM) or field-programmable EPROM for program storage

All current models use flash memory  

The device families: 

+ PIC10 and PIC12 - 12 bit

+ PIC 16 - 14 bit

+ PIC 17 & 18 (21bit)

+ PIC 24

+ dsPIC - released in 2001

    (PIC24 devices are designed as general purpose microcontrollers. 

     dsPIC devices include digital signal processing - DSP - capabilities  in addition.)

 + PIC 32

Silabs (Silicon Labs)

+ Buchla 259e, 260e, and 261e use a C8051Fxxx (Model 120 ??)

https://www.silabs.com/mcu/8-bit/c8051f12x-f13x/device.c8051f120

---------------------------------------------------------

Useful Links

+ List of common microcontrollers

+ Wikipedia

+ Atmel

+ AVR synth video

+ Popular IC's

+ Mutable instruments - flashing firmware

+ Snazzy Effects ArdCore -Installation Demo

+ ATMEGA16 microcontroller AVR Synth - Elby Designs

+ Jondent SynthLab

+ Microcontroller... What you need to program a MCU 

+ Microcontrollers... Programmers

+ AVR Programming -- AVRDude etc

Please let me know if there are any mistakes or omissions.

To be continued .........................

 ---------------------------------

+ Arduino Index 

-------------------------------------

Statues - NLC - Build notes

These are my build notes for the Nonlinearcircuits Statues Module.
Its in eurorack format.

The name comes from the children's game Musical Statues also known as Red Light Green Light.
This is a 1 to 8 multiplexor/multiplexer….with a thru and hold circuit attached to each output.
The active output is chosen by whichever combination of the x00, 0x0 & 00x inputs is high.

The multiplexer IC can be either a DG408 or DG508 (DIP package)

You can put any signal into these, crossing 1V will send it high. In the demo video, I used the /2, /4 & /8 outputs from a Divide & Conquer which gives it a regular 1-8 count.

Using any old signals will give a more random effect but patterns will still emerge.
IN is the signal getting processed. This module can be used to process CV and/or audio signals.
-----------------------------------

A movie quote? ... Inn't he scrummy? ..

I think this is from . from Ideal ..a British dark comedy television series originally broadcast on TV channel BBC Three, ..... He is known for the line 'Inn't he scrummy?' He often talks to Moz ... He isn't particularly clever, though he is somewhat crafty.

Back to the build..........

-------------------------
S1JL Power diodes are for Reverse voltage protection -Mouser Part No: 821-S1JL. Any similar rectifier with at least 1A rating should be okay, such as 1N4001, 1N4002, 1N4004. Get these sizes; SOD-123 or sub-SMA (DO-214 is a bit too big, so avoid).

Which multiplexer to use? the DG408 or DG508?....
The DG stands for Double-Gate .... its a CMOS chip..... ie there are 2 gates in the one device.

"Multigate transistors are one of the several strategies being developed by CMOS semiconductor manufacturers to create ever-smaller microprocessors and memory cells, colloquially referred to as extending Moore's law.^"

The DG408 is an 8 channel single-ended analog multiplexer designed to connect one of eight inputs to a common output as determined by a 3-bit binary address.
https://au.mouser.com/ProductDetail/Renesas-Intersil/DG408DJZ?qs=sGAEpiMZZMutXGli8Ay4kA%2fBPd2gF8jLzm9LT6sd4BM%3d

The DG 508 is a Multiplexer Switch ICs 8:1 CMOS Mid Voltage MUX
 https://au.mouser.com/ProductDetail/Maxim-Integrated/DG509ADJ%2b?qs=%2fha2pyFaduielkf44kdqC6FI6vY%2fkDwFpufrTM%252bGuAk%3d

I chose this one & it works.

---------------------------------------

Specs

HP: 8

Inputs

xOO: first bit of a 0-7 number
OxO: second bit of a 0-7 number
OOX: third bit of a 0-7 number
IN: cv/audio signal

If xOO = 1, OxO = 0 and OOx = 1, you have the binary number 101. Converted in decimal : 5, so the output 6 is selected (you have 1 to 8 channels).

Outputs

The output selected by the x00, 0x0 and 00x inputs is fed the input signal (thru). When the selected output is changed the signal is held at whatever voltage it happens to be (ranges from approx -11V to 11V) (hold)

Links:
+ https://www.nonlinearcircuits.com/modules/p/statues
+ BOM

 + DIY index
+ All about CMOS

-----------------------------------------------------------------------------------

You can find more NLC builds here.

http://djjondent.blogspot.com.au/2015/03/non-linear-circuits-ncl-index.html

---------------------------------------------------------------------------------------

hARMONIC experiments

Playing around with the Verbos 262v

I'm surprised how many vactrols it uses.

View this post on Instagram

#buchla262v #harmonicoscillator #buchlaformat Self generating patch ..Sorry about the sound quality.. just my camera phone.

A post shared by jono (@dj_jondent) on Jan 27, 2019 at 7:56pm PST

Apologies for the poor sound on this instagram video, but it does sound really amazing.
I'll do a proper recording one of these days.

It started me wondering how many other "harmonic" related VCO modules there are out there.
Not may I think.
In Euro the The Verbos Harmonic Oscillator (HO), the Mutable instruments Plaits & Braids and the Make noise Telharmonic were some modules I could find. The HO looks related to the Buchla 148, 262v.
Do let me know if you know of any other modules that you think fall into this category.
The old Hammond organ with its drawbars is an example of this type of synthesis (additive synthesis).
In the software world Native Instruments Razor is a good example.
This brings up the topic of additive vs subtractive synthesis.
and begs the question why are there so few Synth modules that do additive synthesis?

The other euro module that I think can be loosley added to this list is the Make Noise Echophon.

The Echophon's depth control of the pitch shifter can be used to roughly approximate the harmonic series of the imputed  audio .. these are the frequencies that are multiples of the fundamental. To do this the large black pitch control knob must be set to max.

Every musical sound has these harmonics. A sine wave has only the fundamental or first harmonic, a saw tooth contains all harmonics in inverse proportion to their number. A square wave has all the odd harmonics in the same ratio, etc etc.
So with the Echophon, we can for example, just use a sine wave which has only the fundamental harmonic, and then add more harmonics by turning the depth control knob (the blue one).

Maybe this is use of pitch shifting modules to obtain spectral shifting is worth further investigation.
I understand that shifting the pitch of signal components retains their harmonic relationships.
----------------------------------------------------------------

Other modules worth exploring are clock dividers & multipliers .
Eurorack modules in this category include the SSSR Labs VC Divider, the 4MS Atoner, the quad clock distributor, the RCD (rotating clock divider) and SCM (shuffling clock multiplier). Use audio-rate input sources instead of LFOs with the clock dividers & multipliers to get harmonically related outputs.

The Doepfer A-137 is worth a look at.

The basic idea of a wave multiplier is to multiply the waveform of an incoming signal (e.g. triangle/saw/sine from a VCO) within one period of the waveform. This leads to additional harmonics of the incoming signal. The period and consequently the pitch of the signals remains unchanged.
The A-137 works as a kind of "inverse low pass filter", i.e. it adds a lot of harmonics to a signal that contains none or only a few harmonics (e.g. sine or triangle waveform). In contrast to that a low pass filter (e.g. A-120) removes harmonics from a signal that contains a lot of harmonics (e.g. saw or rectangle waveform).
 -------------------------
Ladik make a Harmonic Generator in 8 HP
The H-020
 ------------------------

A little bit  more off this topic, but still  loosley related are filter bank style modules.
I think the Verbos Bark Filter, and the Frap Tools Fumana are worth investigating.
The Bark/Fumana seem to be inspired by the Buchla 296/295/294.
Also worth a google are the 4MS Spectral Multiband Resonator (SMR) and the Serge Resonant Equalizer.

Re the Buchla 294/295/296, the Verbos Bark, the Fumana, the 4MS SMR and the Serge Res Eq. ... maybe they can be better described as "spectral processors"..... they are much much more than fixed-filter banks.

 

 The 294 comb filter is on the extreme left.

 The 295 filter

The first two, the Buchla 294 & 295, Don called comb filters, but their evolution into the amazing 296 seems plain to see.

The 296, the Fumana and the Bark are 16-band filter/spectral analyzers. Amazing tilt-like controls, you can address odd and even bands. They also have a row of envelope followers -- one per band.

Mark Verbos has chosen to use the "Bark scale".
It is named after Heinrich Barkhausen who proposed the first subjective measurements of loudness.
The scale ranges from 1 to 24 and corresponds to the first 24 critical bands of hearing.

 The Frap Tools Fumana - Dual16 Band Spectral Editor

The outward style of each of these modules is very similar ... the upper section provides access to each filter's main output at the chosen fixed frequencies.

The Buchla 294 uses the scale:
<250/ 1K / 4K / >4K

The Buchla 295 uses the scale:
<100/ 200/ 350/ 500/ 700/ 1K/ 1.4K/ 2K/ 3.5K/ >7K

The Buchla 296 uses the scale:
<100/ 150/ 250/ 350/ 500/ 630/ 800/ 1K/ 1.3K/ 1.6K/ 2K/ 2.6K/ 3.5K/ 5K/ 8k/ >10K

The Frap Tools Fumana uses the scale:
 15/ 65/ 90/ 122/ 170/ 230/ 320/ 435/ 600/ 825/ 1.1K/ 1.5K/  2K/ 3K/ 4K/ 5.5K/ 22K

 The Verbos Bark filter uses the scale :
<100/ 300/ 510/ 770/ 1.08K/ 1.48K/ 2K/ 2.7K/ 3.7K/ 5.3/ 7.7K/ >10.5K

The published Bark band edges are given in Hertz as [0, 100, 200, 300, 400, 510, 630, 770, 920, 1080, 1270, 1480, 1720, 2000, 2320, 2700, 3150, 3700, 4400, 5300, 6400, 7700, 9500, 12000, 15500].

So the Verbos Bark uses every second Bark band.
Looks like the Frap Tools has more evenly spaced bands following something like a graphic EQ, whereas the Verbos focuses on frequencies that are more recognised by the ears.
The Fumana might be better for shaping bass sounds as it has 3 bands below 100Hz, but the Bark may be better for singling out harmonics that the human ear can distinguish ???

Quote: Fumana manual.
"Each of the two filter array is based on 16 parallel analog bandpass filters. The main filter bands from 2 to 15 are mainly based on Bessel calculation, while bands 1 and 16 are respectively a lowpass and highpass with a custom method to obtain better musical results. All bands on the main filter array use an 8th order slope (48dB/oct)." 

Links:
+ Top Octave Generators
+ Muffs - DIY Harmonic Osc