I think my hard disk is knackered, can anyone tell me what this means?

Prototype board

So, the prototype board finally arrived for the power supply. Unfortunately the spacers that I had intended to use wouldn't fit in the tiny holes already provided, and the holes in the board were so close to the edge that I couldn't just drill them out, so I have drilled out some little holes a bit further in, not ideal, but still was just about right. I'm only loosing a small amount of space, and in the project space isn't at a premium, so it doesn't matter. I have found the pillar drill at work so no more excuses for shoddy hole drilling! Although they seem not to have any kind of vice or clamp over there, so i will have to source something like that if I want to drill a huge hole in the wooden front panel for the pots. Here is a picture of the idea anyway, probably small components and op -amps and stuff on the bottom, and big power hungry components on the top, so that I can get maximum air flow around them, for cooling.

Bench Power supply circuit is nearly ready!

Hi,

   So we are getting very close. The first circuit creates our un-regulated supply, a 38V supply for the op-amps for the circuitry and a 12V supply for the internal cooling fan.

Then we have an op amp based power supply design with current limiter based on the LT3080 voltage regulator. In this circuit we are using 2 in series to double the amount of current we can deliver. These regulators rely on a minimum load of 1mA generated by I1. I2 is simply for testing (to represent the load on the output of the supply). This supply should give between 32V and 0V, and 0A to 4A. I believe that the transformer I have will only provide 0.5A at 48V, but I have designed the circuit to support more current if I choose to change the transformer in the future.

This has taken me over a month, and is only what I guess is a 'first draft'. I need to get some people who really know how these things work to give it a once over before I start designing the PCB!

Bench PSU Housing

Hey, So I didn't give up, but I built a few board up and decided I needed a bench power supply. I have been designing a bench supply for a while now. I have started putting the case together. I have cut up the front of an old draw and am going to make it into a box to house the design. Last night I marked up the wood with the components I want on it.

 Then I have drilled and sanded for my life until the holes are the right side for the components. Here is the back, and below is the inside of the back. I have made a little bit of a hash of the front so it needs more work. I have the circuit for the supply nearly complete, but i'm still grappling with a few calculations (although im sure it'll all change when I get the design out of the simulator anyway!) - I have purchased a transformer which very shortly I will put a dynamic load that I have at work over it and figure out exactly how many bolts and amps I am going to get out of it. I intent to try and make a printed circuit board, etch it myself (because I can't sure veroboard for this project, as I'd smoke the copper track due to the high current). I'd like to finalise the circuit in the simulator before I get to the task of laying out a single sided PCB (as I beleive double sided will be too tricky). Watch this space for final schematics. Hopefully I will finish it soon so I can get on with the task in hand, which is building the synth! (Project update: I have the first two boards completely built up, thats Midi to CV, and Waveform generator. I just want a current limiter to test them with to ensure I don't smoke any of my lovely components!)

Sorting Capacitors

I ordered some small value ceramic capacitors from ebay. for the princely sum of £3 I got 50 different values, 20 of each type including postage and packing! A bargain I though, until I came to the problem of how to organise them so I could use them. I really have run out of cash this month so didn't want to get ripped of buying another rubbish plastic monstrosity from Maplins. The solution I found was much more elegant ;)

Writing the resistor code number on left me with only the simple addition of a paper clip on each bag, then a bit of string strung up in my lab, hey presto, capacitor sorting method! Now when I come to picking a capacitor it is as easy as choosing a sirt from the wardrobe (and worryingly similar) :) 

Midi to CV - Arduino Code

So, I have written the code that handles the MIDI in and whacks a CV out. This very basically chucks out a number to the DAC that is based on a lookup table. At the moment the values in the lookup table are junk, and that's OK, because the ICL8038 is not looking like it is going to be as precise as I want it to be, so this will give me the ability to be flexible with it, when I find out what it needs.


#include  // this library has the "ShiftOut" command in it
#include  // Add Midi Library

#define gatePin 13
#define midiIn 0
#define clockPin 5
#define dataPin 6
#define latchPin 4
#define midiSwitchPin 2
#define cvPot A0
#define sweepSwitchPin 3
int oldNote;            // this is the last value that we did set
int newNote;            // this is the value we want to set
int noteCount;          // here we will count how many notes we have pressed at any one time
int midiLookup[144]     // here we define the MIDI lookup table, this is 12x12 (12 octaves)
                        // and will define C-2 to C10 (NOTE: This table WILL need adjusting)
= {   12,   13,   14,   15,   16,   17,   18,   19,   20,   21,   22,      // Octave -2
      23,   24,   25,   26,   27,   28,   29,   30,   31,   32,   33,      // Octave -1
      34,   35,   36,   37,   38,   39,   40,   41,   42,   43,   44,      // Octave 0
      45,   46,   47,   48,   49,   50,   51,   52,   53,   54,   55,      // Octave 1
      56,   57,   58,   59,   60,   61,   62,   63,   64,   65,   66,      // Octave 2
      67,   68,   69,   70,   71,   72,   73,   74,   75,   76,   77,      // Octave 3
      78,   79,   80,   81,   82,   83,   84,   85,   86,   87,   88,      // Octave 4
      89,   90,   91,   92,   93,   94,   95,   96,   97,   98,   99,      // Octave 5
     100,  101,  102,  103,  104,  105,  106,  107,  108,  109,  110,      // Octave 6
     111,  112,  113,  114,  115,  116,  117,  118,  119,  120,  121,      // Octave 7
     122,  123,  124,  125,  126,  127,  128,  129,  130,  131,  132,      // Octave 8
     133,  134,  135,  136,  137,  138,  139,  140,  141,  142,  143,      // Octave 9
     144,  145,  146,  147,  148,  149,  150,  151,  152,  153,  154       // Octave 10
  };
    

// Below is a function that will be called by the Midi Library
// when a MIDI NOTE ON message is received.
// It will be passed bytes for Channel, Pitch, and Velocity
void HandleNoteOn(byte channel, byte pitch, byte velocity) {
  if (velocity == 0) {//A NOTE ON message with a velocity = Zero is actualy a NOTE OFF
    noteCount = noteCount - 1;  // we remove 1 from the 'note on' count
  }
  else
  {
    newNote = midiLookup[pitch]; // set the newnote to be the value defined by the lookup table
    noteCount = noteCount + 1;   // we add one to the 'note on' count
  }
  if (noteCount >> 0)
    {digitalWrite (gatePin, HIGH);}
  else
    {digitalWrite (gatePin, LOW);}
}



void setup() {
  pinMode(clockPin, OUTPUT);        // configure the clock pin to be an output
  pinMode(midiIn, INPUT);           // configure the midi in pin to be an input
  pinMode(gatePin, OUTPUT);         // configure the gate pin to be an output
  pinMode(dataPin, OUTPUT);         // configure the data pin to be an output
  pinMode(latchPin, OUTPUT);        // configure the latch pin to be an output
  pinMode(cvPot, INPUT);            // configure the pot to be an input
  pinMode(midiSwitchPin, INPUT);    // configure the midi switch to be an input
  pinMode(sweepSwitchPin, INPUT);   // configure the sweepSwitch io be an input
  digitalWrite(clockPin, LOW);      // Here we write the clock pin to be low
  digitalWrite(latchPin, LOW);      // Here we set the latch high to tell the device we dont need it
  oldNote = 0;                      // the last value we set at this point is invalid, so we set this to 0
  newNote = 1024;                   // we want to start off somewhere sensible so ballpark? 1024
  noteCount = 0;                    // assume that all notes are off when the unit powers on
  
  MIDI.begin(MIDI_CHANNEL_OMNI); // Initialize the Midi Library.
  // OMNI sets it to listen to all channels.. MIDI.begin(2) would set it
  // to respond to channel 2 notes only.
  
  MIDI.setHandleNoteOn(HandleNoteOn); // This is important!! This command
  // tells the Midi Library which function I want called when a Note ON command
  // is received. in this case it's "HandleNoteOn".
}



void writeClockAndData (byte TopNumToSend, byte BottomNumToSend)
{
  digitalWrite(latchPin, HIGH);
  digitalWrite(gatePin, LOW);  // default position at the start is notes off
  digitalWrite(clockPin, LOW);  // we are interfacing with a device that is clocked by rising edges, so we must first set the clock pin low
  shiftOut(dataPin, clockPin, MSBFIRST, TopNumToSend);  // here we send the data top end of the data
  shiftOut(dataPin, clockPin, MSBFIRST, BottomNumToSend);  // here we send the data bottom end of the data
  digitalWrite(latchPin, LOW); // driving the latch high updates the DAC
}



void loop () { // Main loop
  oldNote = newNote; //transfere the old value into oldNote before updating newNote
  if (digitalRead(midiSwitchPin) == HIGH) {
     MIDI.read(); // Continually check what Midi Commands have been received.
  }
  else
  {
    noteCount = 0; // this is a soft reset incase we get MIDI notes stuck on, 
                   // we can flip in and out of MIDI mode to clear the problem
    digitalWrite(gatePin, LOW);  // this does a hard clear of the 'note on' count
    if (digitalRead(sweepSwitchPin) == HIGH) {
      delay(10);
      newNote = newNote + analogRead(cvPot);  // set the note to be incremented by the value of the analog pot
      if (newNote > 4096) {newNote = 0;}      // reset the sweep if it hits max.
    }
    else
    {
      delay(10);
      newNote = analogRead(cvPot);            // set the value of the note to be played to the value of the analog pot.
      newNote = newNote * 4;                  // scale the 0 to 1024 value from the pot to 0 to 4096 for the DAC
    }
  }



  if (oldNote != newNote) {                    // if the note value has changed from the last application scan...
    writeClockAndData(highByte(newNote),lowByte(newNote)); // here we set the data
  }
}

MIDI to CV circuit design

So, I am led to believe that this circuit should do the trick. I am going to lay it out on veroboard so there is enough room for tweaks in case I need to do something more to smooth the Control Voltage, but if I'm honest, I am not going to know until I start playing with the ICL8083 (waveform generator chip). At this stage I feel that its best that I get this end built up. The fine calibration is going to have to happen afterwards, when I have the 8083 to play with, because I really need to 'listen' to the results to see if they are good enough.

At the moment I'm finding progression hard because I do not have a stable platform. After I have the 3 core modules (CV, Waveform, Mixer/amp) sorted, the subsequent modules will be open to a lot more prototyping.

Description: U2 is the Arduino Chip that will need to be bootloaded and programmed up with my desired software (more discussion on this later). U7 is the opto-isolator that is required for the 'Midi In' stage. U1 is the DAC (that apparently doesn't like being de-coupled). U3 is the reference generator (the application notes in the data sheeet don't ever mention decoupling caps, but I will leave space for this on the board just in case). U8 is an op amp, I'm not sure how essential this is, but because I do not know exactly what we will be doing with the CV afterwards I thought this was a good idea to stabilise the system, hopefully I won't accidental put too much strain on the DAC. J6 J7 and R9 are the extras that will allow the right information into the arduino to create the static voltages and the sweep. J4 will we software controlled and provide a "Gate" output, 0V to +5V. This will be used later by the VCF.
This is the power and bus requirements of the circuit, R2 and R3 provide us with a split +12v and -12v power to play with, but because this circuit does not require -12v have have not included the -12v VREG. The Arduino however requires regulated +5v so I have provided one. The reason I have opted for the top end out the +24v to be used is because In the next module, the waveform generator, I would like to give the chip +/-12v, the common ground will be 0V. because all these modules will be linked I wanted the Common ground between the modules to be the same. Making 0v on the PSU -12v below the gorund for these circuits. (kinda making my +12 and +5 on my PSU bus redundant, but... Que Sera...)

MIDI to CV - Getting the DAC working

OK, so it's been a while since my last post. I found a bit of software, "NI - Multisim" for simulating my circuits, while simulating my first circuit i found that it really wasn't kicking the correct voltages out, the design was all crappy. I'd built it on veroboard and made a right mess of it. The module was only supposed to be a half way house anyway, so I have decided to scrap it and start all over again. Doing things properly and not rushing it.
Sooooo...
The first module in the system needs to generate my Control Voltage, but I want a keyboard to control it, so I thought I could use an Arduino to take MIDI in and turn it into a stable CV. I have a 12-bit DAC (LTC1257) that I found at the bottom of a cupboard gathering dust. So first things first, get the Arduino talking to the DAC.
   I wrote a little program to send the clock and data using an arduino external library calling the "ShiftOut" command. Because it's a 12-bit DAC and the shiftout command only works on bytes, I've had to split the integer in half and send both halves separately. Sorting out the endianism was a bit of a nightmare, getting my clock and data the right way round always floors me, and  controlling the 'Load' pin of the DAC HIGH or LOW at the right time so the DAC pays attention was also a challenge, but the end result is fairly positive.
   I have had to buy a scope at this stage, and I have opted (due to cheapness) for a Hantek 6022BE which is essentially just a data acquisition module that then displays its data on the computer. Quite nifty really, I'm a fan, dead cheap for what it does. Anyway, the end result you can see on the picture.

The application is driving the DAC up in steps of 90! which is massive for what seems like little change (the DAC is railed from 0v to +12v). In an effort to remove noise out of the DAC I have tried many different values of decoupling capacitors, but have found that each one of them make things worse not better. I have decided to invest in a Voltage Reference generator chip for the DAC (which is mentioned by the DAC datasheet) in the hope that it will produce a more stable voltage out.
   In addition to the MIDI in, I have decided to include a software feature that will allow a selectable voltage out or a selectable sweep speed of 0 to 12v (simply because the arduino chip has spare pins and it seems a shame not to use them.

Voltage source components on the board

So this weekend I have been inputting schematics into 'tinyCAD' and then exporting the resultant NetLists to a bit of software called VeeCAD. This is a veroboard layout package. It won't do the work for you but it'll tell you when you have got it wrong. A design package like this allows us to optimise for physical space on the veroboard. The first circuit has been laied down, all that is needed is that the controls on the front panel be wired in, and the first module will be complete.
(the second module is CV splitter and mixer, allowing frequencies based on the root note to be generated, to save space in my rack I have decided to combine these 2 functions. After that we'll get to actually getting the thing to make some noise!)

Building the first panel

Covering the panel in masking tape will stop marking while being held in the vice. This panel has been marked up to show the exact place for each component. A centre punch (made out of an old filed down screw driver) was used before some honking great big holes were drilled in the panel.

 The components were then added (see photos below) with some consideration taken to make sure they were clear of the edges so as not to impede in the way of getting in the way of the case.

Now we know how many components we have on the front panel, we can start designing exactly what circuit we need for the board.

Choosing the right panel size

So its about time I started picking the right sized panel for the first module. The first module I need to be a voltage source for the oscillator. I have decided that for a voltage source I will be wanting at least one main control voltage (+1v -> +10v), and also I thought it would be nice to have a couple of low voltages (0v -> +1v and 0v -> +2v) that can be mixed with a main control voltage and used as an offset. There also needs to be some kind of gate, so I have opted for the middle size panel showed in this image. The panels I have got are 8HP, 10HP and 14HP. To fit at least 4 pots and 4 jacks, the smaller panel would be too small. The next step will be to drill holes.

Sonic Wonderland: A Scientific Odyssey of Sound

I heard an interview on radio 4 yesterday with a guy called Trevor Cox. He is a sound engineer working at Salford Uni, and he has just written a book called Sonic Wonderland. It was pitched as a travel guide, but instead of the destinations being visual feasts they are auditory! It sounds wonderful, can't wait for it to come out!

Ice Music Festival

The scandinavian "Ice Music Festival" consists of improvisations on instruments made of ice. It's the kind of thing that deserves a very nice sound system to listen to it on :)

   Some of the marimba like instruments are fantastic, not only do they have a beautiful resonance but the sustain on them is phenomenal. One of the key individuals is called "Terje Isungset" whose music is available on spotify.

Radio 4 have done a wonderful documentary on it.

Power

There we go, easy peasy, one power bus, giving us 24v, Regulated 12v and regulated 5v. Now for something a little more interesting

Making things easier

OK, so it looks like I have taken some bad advice on buying a power supply. I think a +24v PSU is going to struggle to supply both +12 and -12v, so for my power bus I'm going to stick to +12 and +5, and if I need negative voltage, then I'll have to do it on a module by module basis. No big deal. What I do lose is compatibility with existing systems, but if Im honest, I am not building a re-salable, integratable project, I am building a cool sounding synth that sounds cool and looks cool! (just so everyone is sure about that... cool)

Negative Voltage

I think I have a problem... This circuit seems like it might give us the +12v and -12v supply we need, but I think the lack of relative GNDs scuppers me. Can anyone help with this?

Synth Project - Concept

Modular Synth

So, I watched I Dream of Wires and I thought I'd build a nice analogue modular synth. I'd like to practice my electronics, and I thought a modular synth might be a nice project that would take us step by step through analogue electronics. The nice thing about it being modular is that each module can be separated from the rest of the system so that the focus can be entirely on the functionality of each module and not a complicated juggling act of balancing circuits operations. I thought I'd blog this (due to several request).

To start this we need a standard to stick by, so that all the modules can operate happily in a common environment.

System Design

Looking at analogue synths, the top end of the market is eurorack, for which I think the Doepfer stuff looks pretty cool, so I thought Id model my system on their. (Why invent a new standard when one already exists, plus there is a lot to be said for compatibility these days!)

Here is an example of the kind of think i'm going for:

So first things first, I need a supplier for the case and the module face plates. I found that the Doepfer stuff all runs on a 3U 19" rack. Schroff are a company that supply this kinda stuff, they also have an ebay store.

Power

Very helpfully, Doepfer provide a lot of information about the basic spec of their system which allowed me to get the information required to build the system environment (or, power supply and chassis)

This diagram specifies +5, +12 and -12 power bar. I've ordered a PSU that'll take 230v AC and kick out +24v DC and give me 3 amps. I don't think I'll need more than 3 amps for this project, but If I find things are going screwy later because I don't have enough juice, it's a simple change to make.

To get the power I want I'm going to have to use a DC-DC converter and some Voltage Regulators. I found a converter on RS that'll take +24 and kick out +15 and -15. It's called a TEL32423 and is quite expensive, but I can't see another way of getting negative voltage onto my system. (I'm guessing the negative will be important for when I try and generate sine waves).

For each connector I will need a +12 VReg and a -12 Vreg. (most vregs need an overhead of a few volts to work, which is why I need a 24V PSU stepped down to 12V), and because I want each module to be correctly isolated, I think I need the vregs for each unit in the bus. Instead of building this power isolation on each board I want to keep it on the power bus, because it makes sense to keep functionality organised and together. So I've ordered a bunch of them. The connectors on this standard bus are 10 way "IDC" connectors, so I have ordered a bunch of plugs, sockets and cables also.

The CV and Gate on the bus are a lovely idea to potentially remove a little bit of the patching from the final front panel. Electrically there is nothing going on here, this is just a bus that allows modules to communicate with other modules without using patch leads. CV is designed for the control voltage (to control pitch) and Gate is designed for the 'On/Off' signal, although these can be used for anything. To isolate these there should be jumpers on the board so this can be configured depending on the system. I am not even sure how useful this will be, but as we are adhering to the standard it doesn't do any harm to do things properly, it might even come in handy later, who knows.

So with all the components on the way, next we will turn to the breadboard to check that we are getting the voltages that we expect before building the power bus board.