Archive for the ‘The VA Synth’ Category

Back in business

November 12, 2009

So after having a lot of other things that had to be done, I got some time over to continue the project. First I had to cut the 29 potaxels down to 12mm (1/2 inch) to fit the knobs.

I then started to mount them one by one on the frontpanel.


As I tightened the nuts I managed to make some small scratches in the panel but I can live with them as they are not that obvious.


Whohaa! A whole forest of pots. This will require some amount of wiring I guess.


And finaly the pots and knobs mounted.


This time the pictures are not a simulation but the real thing.




The next update will be the LCD-display and a rotary knob for patch/parameter change.

Until then…



Cold in the north

October 17, 2009

Sorry for not making any updates lately but I have to work on the house heating and it’s getting quite cold here in Sweden. I promise I’ll be back as soon as I have a warm home.


Some progress and a binary tutorial

September 28, 2009

Ok, while waiting for the hardware to arrive, i’ve made some progress.


I had a comment regarding my schematics and how come both the left MUX, the right MUX and the LCD are all connected to the same wires in the LPT port. Is that gonna work?

Yes it will, it’s all down to binary numbers and how computers “talk” to different chips on the same wires (in parallel). It’s called a databus.

But if they are all on the same wires, won’t there be collisions? No, because we have something called a chip select (CS or E signal).

All the chips listen to the same bussignals in parallel but only the chip that receives the chip select signal will use the data on the bus and the A/D converter and the LCD have their chip select on different lines.

There was also a question about how an Inverter can be used to select between the left and the right MUX.

A value of 0-15 are represented with 4 binary digits (A0-A3 in the schematics)  so we can use that to select one of 16 inputs in both multiplexers in parallel.

What happens when we output the number 16? Well the 5th line goes to a logic 1. This means that the left MUX will be disabled because the E input of the MUX needs to be a logic 0 (or off) for the MUX to let anything through. So by putting out the number 16-31, the left MUX will disconnect but at the same time the Inverter will “Invert” that logic 1 to a logic 0 so that the right MUX will put it’s voltages through instead.

This means that from 0-15 the left MUX delivers and from 16-31 the right MUX takes over. Easy or what 🙂

Then for you that understands exactly how it works also know that when putting out values to the LCD the MUXing will jump around like crazy but that doesn’t really matter as we are not reading any values from the A/D converter because it’s chip select line is off.

I know this is technical but if you build it like this there will be a minimum of wires to connect and it will just be a matter of putting out the right values to the LPT port.


Analog scanning and display hardware

September 22, 2009

These are the hardware chips that will be responsible for scanning the potentiometers.

2 x  CD4067 16 to 1 Analog Multiplexer

1 x TLC549 8-bit Serial A/D Converter.



The multiplexers will enable the PC to choose 1 of 32 analog inputs for A/D conversation by the TLC549 A/D converter. The endpoints of the potentiometers are tied to +5v and Ground and the center of each are routed to an analog input of the CD4067. The frontpanel program scans through each of the pots and registers any changes, transmitting these as MIDI-CC.


The display part is a standard 20×4 LCD with a yellow backlight connected to the PC with a 4-bit interface enabling the frontpanel program to display any characters or graphic on it.


The display is very easy to interface with thousands of examples on the Internet.

The frontpanel buttons are connected to the standard PC-keyboard scanner resulting in normal keypresses to the frontpanel program.

Here is the complete schematics with the 4 chips and display connected to the parallelport.


The 4th chip is just an 74LS04 Inverter used to extend the adress range for the multiplexers to 5bit linear adress  instead of 4bits per chip.

There is actually a bonus using an A/D converter, it converts analog voltage to a MIDI-CC so if we would like to have a CV input, no problem, just connect a jack to one of the multiplexer inputs. It’s that easy.


Measuring the Panel

September 21, 2009

I have been busy printing and cutting out the panel vinyl. Actually it’s mostly CAD work, making sure that the finished vinyl keeps the original measures.


The blue spots represents a 10mm potentiometer and the orange spots a 6mm button. I also have to make a large hole for the 20×4 LCD display.

I still don’t like the compact-flash sticking out so I’ll just ignore it. When I’m done I’ll mount it inside and patch up the hole.


The final stages on the VA

September 16, 2009

So we have reached the final 2 stages on the synth, building a panel and making it tick. This is actually the most difficult part.

Boxing a computer and a keyboard is easy but doesn’t give any personality or soul to the keyboard. What needs to be done now is give it life (as Dr Frankenstein said).


This is just a bit of the parts needed to make it work. What is even more painfull is that it has to be connected on the inside to the computer to make it come to life.

I’m still working on the graphical design of the panel (as it didn’t turn out the way I planned) before I can put all the hardware in the picture in it. If you thought the cabling looked messy before… It will get worse.


The panel

September 12, 2009

Tomorrow I’m off to the shop to print and cut my panel so you will see some results in the next few days.

I also have a confession to make, I actually overclocked the board to 250MHz. It ran from the beginning at 200MHz and worked like a charm but I need the extra 50MHz because I have to run extra software for handling the frontpanel and I want it to have a real response. If you twist a knob on the panel and it doesn’t respond instantly then the feel is lost.

I also want you tell you that my next project online will be: A copy of the EMU Emulator II, one of the best sounding sampling keyboards in history. I’ll copy this baby in perfection. I have already written the code for the 8-bit uLaw playback generators and the code for the SSM 4-pole filters.

This will not be a VSTi but a dedicated CPU/Software/Hardware emulation. And I will show you the code aswell (except for the SSM filters) so you may build it yourself.

If anyone is thinking that I’m a fan of Depeche Mode… well, you are right, I am.


Just a thought

September 9, 2009

This is a post about old and new gear. There are people saying “why should you buy old stuff when there is fresh gear to buy?”.  If you buy old stuff you know what sound you are buying.


There are endless discussions saying new sounds are better than old ones.

Are they?

2 days ago I bought a Yamaha CS1x synth second hand. So, all of you are saying  “It is outdated and old!”, why?


The CS1x is the youngest board in my collection but in my ears it sounded nice. I love my keyboards and I hate to depart from each one of them but to call one useless from the other… NO WAY! Every each one of them have their special sound and I love all of them.

In my ears, play your boards and lov’em or ditch’em.

But never say that one is better than another!



September 6, 2009

Now, this looks nice 🙂



Panel design

September 6, 2009

I have recived all the pots, knobs and buttons for the frontpanel so the designpart can begin.


The design is made using Corel Draw and Photoshop. As my wife runs her own T-shirt and print shop I have access to a wide format printer and a Roland GX-24 vinylcutter I can just print and cut the design in full scale and attach it to the panel.


This is what I had in mind. A total of 29 knobs and 37 buttons.