here is a great quote from Robin Whittle, creator of the Devil Fish modification for the TB-303. Don't worry if you can't understand the electronics technobabble, there are some real gems in here.

The Devil Fish does not have an external oscillator, but it does have an external input. The Devil Fish principle is for a single chain of processes (not two oscillators in parallel) to be doing things which range from "electronic naturalesque" to "mechanical overdrive" - through the use of unusual circuitry, high signal levels, and in one case, a filter FM feedback system which can elegantly approach chaos. Those who recognise that complex things like a love life can benefit from wise or reckless overloading of certain elements (for instance a hard spanking) are well on their way to understanding the Devil Fish. In this case the filter is driven well beyond normal limits. Like an ultimately appreciative spankee, its noises and gyrations provide depth, complexity and emotional heat which is exciting and highly satisfying for all concerned. (As when any gentleman heats up a derriere, there is no permanent damage.) Boosting the oscillator level to 20 times normal was good, but there was a promise of even greater excitement at even higher levels. I decided this called for the clothesbrush treatment and whilst examining various combinations of standard resistor values, decided that the pair which gave 66.6 times the normal level was the most auspicious. (Gentlemen, if you think that all that women want is what they say they want, then you need to read some romantic spanking fiction written by women, for women. An excellent source is "Sassy Ladies Magazine" - PO Box 4516 Lexington, KY 40544 USA US$12 for a sample issue. From their advert - "Independent, feisty women find they still need strong, caring men.")

Right then.

Time to start modding.

I'm actually in the process of looking at the Devil Fish modifications, and seeing how I could implement my own, and even looking into other mod ideas. All this is on my x0xd0x wiki, specifically the mods section.

This is Part III of a multi part series on building your own effects. See Part II and Part I Here.

I Finished the prototype of 'Bunny Cruncher', my cute little distortion unit. Thanks to 'D A F', Harry Bissell and Aaron Bader from the SDIY mailing list, I was able to get the output to a much more acceptable level. I changed the collector resistor to a much lower value, (10K) and added a biasing resistor. The Biasing resistor is super high in value, but it works. I also changed the caps to 10uF. Now it distorts just fine. A few more tweaks and revisions, and I'll have something hella cool.

Next up on my plate is to get a pure sine-wave sent to it, and start to really tweak with the diodes and see what kind of interesting effects I can make out of it.

I also did the schematic in eagle this time. MUCH BETTER. Eagle is much easier to use when laying out a schematic, but there is some VERY deep mojo going on there. If I were to try and make a PCB with the schematic as stands, it would fail. Badly. Mostly because there is a disconnect between the components that I have, and what is actually displayed. For instance, the 2222A transistor I have is in the typical black plastic crescent moon shape, but the one displayed in Eagle is one of the weird cylindrical metal units. Here is the schematic in Eagle Format

Ah well. Cross that bridge when I get there.

This is Part II of a multi part series on building your own effects. See Part I Here.

So, I cooked my own distortion. To the left you can see the schematic that I ended up with. This is modified from a schematic Here. The two bits on the left and the right are the input and output respectively. Please excuse the lameness of the schematic. Jonathan hooked me up with Dia, and while its relatively simple to use, I didn't really want to fart around with it that much, I just needed to get a quick and dirty schematic up. So it is quick, and dirty.

And It works! Its not perfect, but it works! It makes a really nice crunchy grindey distortion, but it is also somewhat 'clean' in its distortion. Its not at all like overdriving a Mackie. I do like the sound. The next step is to try and figure out why the output is so low. I tried playing around with the value of the 400K resistor, but it didn't seem to make much of a difference. After I have this thing in fully working order, I am going to play around with adding some more in the "feedback loop" between the 2222A and the 2 diodes. I am probably going to had a sweepable low pass and sweepable high pass filter.

Once I have all these basics down, I am going to look at making the thing pluggable. I want to eventually have it so you can change out the feedback generator (i.e. switch to an op-amp), Add things to the feedback loop, pre-distortion and post-distortion.

All this work has also given me an idea on what a good electronics prototyping breadboard needs. Right now I am using Shells "Vulcan Logic Trainer", which has all kinds of interesting bells and whistles, for... well... logic programming. Not so good for audio. Hell, when I was testing it, I was forced into holding 1/4 inch plugs up to the bare wire. Bah. Suckage.

This certainly isn't a new idea, but I've totally been bitten with the diy bug. My plan is a distortion unit that utilizes a 'pluggable' architecture. The idea is that you could plug in small modules at the various stages of the distortion so that you can change the sound. In any given distortion unit, you generally have 3 stages: Input, Feedback loop (where the distortion actually happens), and Output. The kind of modules that I am thinking about are filters (to color the sound) and specialized circuits on the feedback loop to change the shape. There is no reason why other circuits couldn't be added to this as well. Things like a ring modulator, resonant filter, or something.

Each plugin unit would have its own bypass switch and controls. Also, when a unit is not plugged in, it should just bypass the signal, rather then break the circuit. I'm not entirely sure what kind of plug I should be using for this part however. I do have the circuit block diagrammed, the next step is to block diagram the plugin elements, and then to do the physical layout. After that, I can think about the actual components of each piece.

As I keep going with this project, I'll post it.

Drew Rules. His set was absolutely stellar. His set is what my set needs to be. His sound is much more polished, and he has the crowd connection that I just don't have.

I think that overall my set went well. First of all, I think I should have done a better sound check. I need to get more and better experience at listening to my stuff on The Big Stereo, and be able to extrapolate how the rest of my set will sound.

I also need to get more interaction happening. This set was much more interactive (see the sidebar on the right for one of my favourite examples).

Hardware considerations for next time:

• Fix the fucking knobs on the A3k
• Get a rack system - For the longest time I have had some interesting PVC pipe ideas. It is about time that I sit down and design something that will work. A self contained table would be ideal, something to make sure all my devices are accessible, but also where I am visible would be optimal. If PVC works, then I could permanently snake midi and 1/4 inch cables through it, as some kind of combined rack/patchbay. I'll need to flesh this out more.
• Set Details. Write some kind of Scheme program to deal with, and display set details. BPMs, pattern numbers, that sort of thing. It was really handy having a backlit notepad, but the display was kinda wonky and hard to read using the goofy little database program. Huge letters is where it is at.
• Get important small gear. Here is my shopping/DIY list in order of importance:
  1. Proper Mackie 1604 Balanced 1/4 inch -> XLR cables - I'll have to solder these myself, but it could be pretty fun. And this is something I could do. I should probably be smart and go stupid long, like 30-40 feet.
  2. Proper lighting - Using flashlights/leds kinda suck.
  3. Cable Tester Not 100% necessary but would be nice.
• Get important large gear. Again here is a shopping/DIY list, in order of importance:
  1. Compressor - Actually, probably more of a limiter. Or maybe both. Really, something to make sure I don't completely overload the main output, and something else to compress the x0xb0x.
  2. Distortion Unit - This can be made easily with an opamp, and some basic components. I already have good ideas for making it modular too.
  3. Line conditioner - Something to take the sketchy AC power, and make it nicer. Much nicer.
  4. Sonic Maximiser - Or some kind of sweetener. You know, a black boxes of sound that just make it all better.
  5. Equalizer - A multi-band graphic equalizer would work in the short term. A parametric would rule.

'Soft' Considerations

1. Start Programming Channel Change messages into my sequences. The more I can give my knobs on the A3k A Break, the better. It would also help with any song-change-over-badness that happens.
2. Bassdrums on channel 13, Basslines on channel 14. Time to start moving to purely sampled bassdrums. I already made some 808ish and 909ish sounding ones. (with realtime control to boot.) Throwing the bassdrum and the bassline on their own channels on the mixer is just a smart thing to do. And because bass is omni-directional, I only have to really use one channel.
3. Reformat and Rebuild. A stupendous amount of my music is spread all about my samplers HD. Time to backup, and rebuild from scratch. Better and more organization.
4. Smooth Out the Tempo I need more options in the 145-160 BPM range. Time for more music. This is never bad.
5. Shells Song
6. Chu Chu Chu Chu Chu Chu Chu.... NA NAAAAAAAAAAAAAA NANANANANANA NA Na Na Na Na NAAAAAAAA! Either you get it, or you don't.

Note: This was actually posted on Monday morning at 1:14 AM, not Sunday morning at 1:14 AM, but I couldn't resist the pun.

So after making Hella Cute Hats, A business out of them, and of course, my own TB303 and even some twisted BDSM gear... (some of it public, some of it not.) I have come to the following conclusion.

It is fun, and cool. It is generally cheaper to make your own stuff, rather then buy your own stuff. This is especially true in the realms of BDSM sexXx0r toys and TB-303s (where the original will net you at least a grand on E-Bay, versus a $300 almost-perfect-clone kit). It is also a hellava lot more rewarding making your own stuff, rather than going down to the 'local' Wal-Mart, to further their construction of a evil-empire-like-pave-the-planet-fully-operational-battle-sales-station.

I also find that I spend more time learning about the world, and learning about my relationship to the world. Now, design, programming, and music creation are also DIY-styled-pursuits, but there is a difference between hacking code, (or designing a site, or building a song) and soldering a synth or making a hat. The latter has a certain physicalness to it. A sense of real-world tangibility to it that is very appealing. When you're done, you have something physical, and hopefully useful.

My next project is going to be learning how to sew PVC, and will probably involve some caulking as well. I have mad plans... MAD PLANS!

Wow, I've been Blogged at Make Zine. Kick ass!

As promised here are: Musics, Photos, Damages and Reflections.

Music

Here is an x0xb0x remix of Overload:
www.jonnay.net/music/AcidTechno/Overloadx0x.mp3

Pictures

In the beginning...	Filing down pots after cutting	Number 28 = 23 + 5!	Cataloging EVERYTHING
First time soldering something to the board	IO board when we got it. (USB pre-soldered)	First Milestone of IO/Power	My beautiful wife, the solder queen!
The VCO section is finished!	Power attached for testing.	Transistorize teh world!	Waveform Switch! First Knob!
VCF (ohhh!) VCA (ohhh!) and Mixer	VCO and Envelope	KNOBS! SWITCHES!	Almost done...
IO Board finished!	Testing the whole unit	Hmm, more 'testing' required.	Funky zoom on topside of board.
Finished! Mmmmmm. X Ohhhhh Sex Box!

Damage Done To Humans By x0xb0x

• Poked a LED through my thumb mounting it on the standoff
• Very minor soldering iron burn on my middle finger
• Lung Damage due to burning flux and solder
• Sleep Deprivation
• Sanity damage due to anticipation
• Anticipated: Ear damage

Damage Done to x0xb0x By Humans

• One 6V Voltage Regulator. Totally fuct. No hope for patient. Currently sitting on the floor behind our computers, sans its three legs.
• One transistor slightly wounded from soldering, cutting, and re soldering. Patient made full recovery.
• Amputated Headphone Jack

Testing the x0xb0x.

After the filter section, we opted to do no further testing until we had the thing completed. For better or worse. In retrospect, I would have liked to have done more frequent testing. Not having a desktop power supply really put a stumbling block in our testing. Using 2 double-a batteries connected together did work enough to do some very VERY basic testing, but certainly no calibration. Not having a scope meant that I had to use a pair of headphones for the testing. This sucks, because you can hardly hear the notes, and it doesn't sound ANYTHING like a TB-303 at this stage in testing, so deep inside the back of my head, there was a little voice that said "That sounds nothing like it. You've paid a bunch of money for something that makes thin, weedy noises, and they aren't even loud, and it might not even work when you have it all put together!".

I found the envelope testing to be hard to follow, so we decided to skip that until we could properly generate envelopes in keyboard mode. By that time, we were so close anyway, that we got the rest of it all put together.

Except... it didn't work. Until we put the microcontroller in. You know.. the brains. Duh. I coulda used some. I blame lack of sleep (it was almost 2:30 at that point), and flux inhalation. So we did get it working, and I twisted knobs on random mode.

Calibration of the x0xb0x.

Calibration of the box went surprisingly smooth. I did have some troubles with the software oscilloscope crashing, but it wasn't that big a deal. I used this one Here

I have the resonant frequency tuned (roughly) to the official roland specification of 500 Hz. Because I did the tuning after the envelope and Accent sections, I am probably off some, but if I am, it just means I have a higher resonance, which will hopefully account for the slight difference in square waves between the x0xb0x and tb303.