Es ist mal wieder soweit – Doktor Andy ihm selbst sein Vater dem sein Sohn baut einen weiteren Midi-Controller. Hintergrund ist, dass wir unsere diesjährige Festivaltour etwas mehr aufpeppen wollen, als es bislang der Fall gewesen ist. Und -frag’ mich nicht warum- im Kopf war das Flag gesetzt für ‘einen Midi-Controller bauen’. So ganz ist noch nicht klar, was damit eigentlich gemacht werden soll, der Gedanke kam mir aber erst ins Bewußtsein, als ich schon 4 Stunden dabei war. Die Zeit bis dahin hatte ich quasi aus dem Muskelgedächtnis abgespult.
This post is about building an arduino-based Midi Timecode generator. It incorporates tempo-tap, nudge and a granular tempo-control via an endless encoder. It aims to providing nudge-funtionality (and improved tempo-control) for Native Instruments Maschine MK2 but can easily be used with other devices and programs as well.
Oben links der teilweise selbstgebaute Controller, den ich tatsächlich schon seit über 5 Jahren nutze. Das Ding hat sich definitiv bewährt. Die Plattenspieler müssen eigentlich nicht sein, sind aber noch da, weil’s manchmal einfach tierisch Spaß macht, zumindest so zu tun, als hätte man es derb drauf. Verwendet wird das dann in einem komplett wirren Mix aus Navigation-in-den-Tracks-per-Controller, Beatmatchen-via-Plattenspieler-mit-Timecode-Platten und Zwischendurch-einfach-Play-und-Beatmatch-per-Controller. Abenteuerlich und schön anzuschauen, ganz bestimmt – in Summe bringt’s das aber nicht mehr. Die Plattenspieler hatte ich vor ~10 Jahren gebraucht gekauft. Der Zahn der Zeit ist nicht spurlos an ihnen vorüber gegangen. Man könnte das alles wieder hübschmachen, aber ich hab’ auch einfach Bock auf was Neues.
Ich hab es an anderer Stelle schon einmal erwähnt: Seit mehreren Jahren arbeite ich mit Leuten an einer Lösung zur Steuerung von Geräten. Der ganze Bimmbamm hat sehr wenig bis ganz viel mit Automatisierung zu tun, geht aber an manchen Stellen darüber hinaus und ist an an verschiedenen Stellen sowieso ganz etwas anderes. Das Ding nennt sich BetaTouch.
Einer der Wege, die wir mal verfolgt haben, waren die ‘Zauberkästchen’. Daraus ist kein fertiges Produkt geworden, aber es gibt ein paar Bilder und ein wenig was zu erzählen. Hervorragend also für die Webseite.
Im Auftrag eines Kunden habe ich ein Thekenmeldesystem entwickelt und gebaut.
Ziel ist es, in einer Diskothek von mehreren (bis zu 10) Theken aus an zentraler Stelle Meldungen absetzen zu können (Ruf nach Security, Geschäftsführung, etc). Meldungen sollen durch einen Text und farblich entsprechendes Blinken auf einem Bildschirm dargestellt werden. Zum Beispiel: Rotes Blinken + Name der Theke, wenn es darum geht, die Security zu informieren. Für den Fall, dass von keiner Theke eine Meldung kommt, sollen Animationen abgespielt werden (Veranstaltungshinweise etc). Die Texte, Farben und Videos sollen vom Kunden angepasst werden können. Das Ganze soll Luft nach oben haben: Erweiterung um den Anschluss an ein Bussystem, mehrere Monitore, Ethernet, etc .. .
An idea that came up during the 31C3. The guys from VisualPhi had some motion sensors lying around and wanted to use them to control their VJ-software. That’s why I built them a Motion-Sensor-to-MIDI-Converter.
As usual it all starts on a breadboard. Most of the times I draw the schematics parallel to building the circuit on a breadboard. Guess that’s the usual way.
The circuit itself is rather unspectacular. 8 inputs are polled from a 74HC165. Then there’s a little bit of logic implemented within an Arduino and then there’s 16 LEDs, a rotary encoder and MIDI out.
This project is the first one to benefit from my new 3D Printer. Due to the fact that I don’t have a dedicated toolshed anymore it’s kind of impossible to reliably manufacture the case anymore. Seems as if I don’t need one from now on.
I really think the fixation of the rotary encoder is one of the smartest pieces ever done by mankind. Ever =)
The LEDs are driven via Charlieplexing. It’s rather easy to implement but you really need to concentrate while soldering. By the way: If everything else fails I guess I’ll become a Soldering-Artist one day.
The function of the device is easy to explain. Every input is triggered when the state of a connected switch changes. This is indicated by the red LED below the channel. The green LEDs indicate the channel that’s influenced by the rotary enoder: The encoder gives the possibility to set the time that has to pass from the moment the input is triggered until it can be retriggered again. Something like a ‘Retrigger Threshold’. The value can be set to values between 0 and ~2 seconds. When the lower / upper limit of the value is reached the green LED flashes. Pressing the rotary encoder (it has a built-in switch) switches to the next input.
Or: Creating an audio-signal with an Arduino, feeding it into a mixing desk, altering the frequencies via the mixer’s eq and analyzing the processed audio with another Arduino which then turns it into a MIDI-signal. Yes, that is Digital-to-Analog-to-Digital-to-Analog-to-Digital-conversion. Phew!
Here’s a picture of the setup:
On the left side there is a circuit consisting of an Arduino and an Attiny85. The Arduino creates a low-frequency ( ‘bass’ ) sine-shaped tone which perfectly sits in the frequency range of the mixer’s bass-EQ. I did some testing here to find the perfect frequency. Testing means: Using Ableton Live to create a white noise, feed this into the mixer, feed the mixer’s output back into Ableton and use the Spectrum-tool to see which frequency gets most influenced by the bass-EQ (C2, that is).
Creating a somewhat true sine needs some effort since the Arduino’s analog outputs only do PWM which isn’t very useful when talking about low-frequency audio signals. PWM basically creates a square-wave signal with a certain pulse-pause relation. While this might be okay for dimming an LED, this becomes quite unusable when dealing with audio because you can simply hear that it’s no sine – the lower the frequency the more the signal turns into some sort of ‘click’-noise. No wonder, the bass-EQ doesn’t influence this to any convience.
That’s why I used this solution to make the Arduino spit out something that’s a little more sinewave-like. I ommitted the circuit as you may see on the picture below. I didn’t have the necessary parts lying around and it worked nevertheless.
The Attiny85 is used to create the second tone. It’s a simple PWM signal at 480 Hz. This time the PWM-nature of the signal can be used for our benefits: A square-wave signal has a recognizable amount of harmonics. You don’t hear one but (at least) two tones. Perfect for us because the mixer I used perfectly influences (well … “perfectly” ) the signals with its mid- and hi-EQs.
The code for the Attiny85 looks like this:
void setup(){
pinMode(3, OUTPUT);
}
void loop(){
buzz(3,480,100);
}
void buzz(int targetPin, long frequency, long length) {
long delayValue = 1000000/frequency/2; // calculate the delay value between transitions
long numCycles = frequency * length/ 1000; // calculate the number of cycles for proper timing
for (long i=0; i < numCycles; i++){ // for the calculated length of time…
digitalWrite(targetPin,HIGH); // write the buzzer pin high to push out the diaphram
delayMicroseconds(delayValue); // wait for the calculated delay value
digitalWrite(targetPin,LOW); // write the buzzer pin low to pull back the diaphram
delayMicroseconds(delayValue); // wait again or the calculated delay value
}
}
I guess I found it over here and adapted it to my needs.
the two microcontroller’s output signals are fed into a 7408 (Quad AND) and then sent out into the analog world by a circuit I found over at the MunichMakerLab. This is my first audio-circuit with an Arduino. it’s probably spine-crawling for those who do this on a more professional base but I was getting the best results with this circuit.
[Edit] As someone pointed out in the comments section for this post on Hackaday this might read like I didn’t know at all what I am doing here or that it’s all just a big coincidence. This is not correct. The AND gate protects the audio sources from interfering with each other for a certain amount. I tested that, it simply sounds cleaner. At least I had a certain intention when I added the gates to the circuit (…not that I completely remember….). Looking at the circuit I am still wandering about _why_ but that’s one of the things that I file as ‘Audio things’ for now. [/Edit]
The signal is fed into the mixer and from the mixer sent to another Arduino which does the processing. the circuit looks like this:
To be honest: I cannot really remember where I got this circuit from. Somehow all the solutions I found while trying to find the circuit again look a little different. Again this might be spine.crawling for some of you but… it’s a fun project and it works. The code for the realtime audio-analysis is based on the FHT library by openmusiclabs and expands an example I found over at dontquityourdayjob.
// Adjust the Treshold – what volume should make it light up?
#define THRESHOLD 40
// Attempt to ‘zero out’ noise when line in is ‘quiet’. You can change this to make some segments more sensitive.
int oct_bias[] = { 600, 600, 1, 100, 50, 50, 50, 50 };
// Divide Threshold by 2 for top octave? 1 – yes 2 – no. Makes highest frequency blink more.
#define TOP_OCTAVE_DIVIDE false
/////////////////////////////////////////////////////////////////////
// Hard Customizations – know what you are doing, please.
/////////////////////////////////////////////////////////////////////
// FHT defaults – don’t change without reading the Open Music Labs documentation at openmusiclabs.com
#define LOG_OUT 0 // use the log output function
#define FHT_N 256 // set to 256 point fht
#define OCTAVE 1
#define OCT_NORM 1
// Delay – defines how many cycles before the lights will update. OML’s algorithm at 256 samples (needed for our 8 octaves) takes
// 3.18 ms per cycle, so we essentially throw out 14 cycles (I used mechanical relays, you can lower this for solid state relays).
// 15 cycles = 47.7 ms update rate. Be careful here and don’t change it too quickly! I warned you!
#define DELAY 15
#include <FHT.h> // include the library
#include <MIDI.h>
void setup() {
Serial.begin(31250); // use the serial port
TIMSK0 = 0; // turn off timer0 for lower jitter
ADCSRA = 0xe5; // set the adc to free running mode
ADMUX = 0x40; // use adc0
DIDR0 = 0x01; // turn off the digital input for adc0
}
/**********************************************************************************
Loop – includes initialization function and the full loop
**********************************************************************************/
const int NUMREADINGS=10;
int readings[NUMREADINGS]; // the readings from the analog input
int index = 0; // the index of the current reading
int total = 0; // the running total
int average = 0;
int bassVal;
int midVal;
int hiVal;
int bassValOld;
int midValOld;
int hiValOld;
const int OUTTHRESHHOLD = 4;
void loop() {
// True full loop
int q = 0;
while(1) { // reduces jitter
cli(); // UDRE interrupt slows this way down on arduino1.0
for (int i = 0 ; i < FHT_N ; i++) { // save 256 samples
while(!(ADCSRA & 0x10)); // wait for adc to be ready
ADCSRA = 0xf5; // restart adc
byte m = ADCL; // fetch adc data
byte j = ADCH;
int k = (j << 8) | m; // form into an int
k -= 0x0200; // form into a signed int
k <<= 6; // form into a 16b signed int
fht_input[i] = k; // put real data into bins
}
fht_window(); // window the data for better frequency response
fht_reorder(); // reorder the data before doing the fht
fht_run(); // process the data in the fht
fht_mag_octave(); // take the output of the fht
sei();
if (q % DELAY == 0) {
//—-Smoothing
// subtract the last reading:
total= total – readings[index];
// read from the sensor:
readings[index] = (fht_oct_out[1] – oct_bias[1]);
// add the reading to the total:
total= total + readings[index];
// advance to the next position in the array:
index = index + 1;
// if we’re at the end of the array…
if (index >= NUMREADINGS)
// …wrap around to the beginning:
index = 0;
// calculate the average:
average = total / NUMREADINGS;
//—-
The whole mechanism is not THAT precise but it gets the job done and it’s a fun thing to watch. The bass-frequency has to be smoothed-out quite a bit in order to make it all work. After spending a little more than a day with this some might ask “what for?”. I tell you what for: for the sake of finally doing it. I had this idea for over a year now and it was well worth trying.
The system is quite slow in its reaction (mainly caused by the necessary smoothing) and results are still a bit unpredictable but turning an audio-mixer into a midi-controller just by using hardware of ~10€ ain’t too bad, isn’t it?
There are numerous hacks out in the wild adding some weird functionality to it. For quite some time I wanted to something similar. This is the documentation of how to make the Easy-Button a MIDI-USB device based on Atmega328 (Arduino).
Many hacks have in common that they are either relatively expensive (like…involving something with a dedicated teensy device) or rather ugly (due to holes just being sawed into the Easy Button’s case).
My first goal was to build a device that automatically identifies itself as an HID-compliant USB-MIDI device and gives simple MIDI functionality by using an Atmega328 and V-USB. In order to achieve a correct enumeration and to get a useful starting point I used the demo-versionof USBLyzer to get the necessary data from my KORG nanoKey and altered them in various places (Vendor ID etc..). (You will remember this one later.)
When the button is pressed a ‘MIDI Note ON’ signal is sent. Upon release it sends a ‘Note OFF’ message.
The second goal was to give it a clean overall look.
Let’s see…
This is the Staples Easy Button with the cap and the clicker already being removed (which makes it just a pretty unidentifiable bunch of electronics and some plastic)
we don’t need the speaker so it will be gone soon….
The black line shows how deep the cap goes when it’s pressed.
This is the spot where the USB connector will be placed
I might not be the best craftsman around (already mentioned that once before) but after some filing this one looks pretty decent:
The plastic on the inside has to be cut as well
This DOES look quite well
Now I need to add the circuit board. Due to the speaker being thrown out there is lots of space for that now.
The circuit is basically a 1:1 copy of the V-USB keyboard example. The Button is connected with data pin 6 of the Atmega. It involves some creative soldering of the diodes because I didn’t care too much about the circuit’s layout before I started soldering.
Everything’s coming to an end soon
It seems impossible for my camera to do any decent shots that contain the color red.
Anyway, you might get an idea of how the circuit fits into the structure.
Some detail of how the actual button-mechanism is connected to the Atmega (the blue wires, you guessed it)
To make future additions a little easier I added an ICSP connector which fits nicely into what has formerly been the battery case
Finally… in all its glory.
I had this in the back of my mind for ~2.5 years now. Shortly before starting with this built I started wandering whether this might make sense or be worth the money or time invested or…..
F*CK IT. Never let doubts get in the way of your creativity. If this wouldn’t make sense than I probably wouldn’t have built it. Word! =)
The whole Arduino project (I did that arduino IDE 1.0.3) can be downloaded here. Unzip and copy the folder ‘Nanokey’ to the libraries-folder of your arduino IDE. Feel free to contact me if anything is left unclear.
Ich gucke gerade das erste mal Rick und Morty und plötzlich ist die Fantasiewelt, die seit Jahren meinen inneren Frieden zusammenhält, bei weitem nicht mehr das Bekloppteste hier. #rickandmorty
