[sdiy] Control - was Radfio Shack

[TIm Daugard]

>From: "Czech Martin" <Martin.Czech at Micronas.com>

>This was not the point.

>It is about CONTROL. Not audio processing.

*** Long, off the point, but definetitly synth DIY discussion follows ***

I deleted the orginal message so, I may be off point. That is not unusual, my
wife claims I'm way off balance anyway.

For control I use a pedal board (remember I play bass). The pedal board has
eight switches multiplexed on a three wire cable. I have figured out a way to
add another bank of eight switches, but figured that would be overkill. The
eight switchs are decoded by a controller, and sent to a dual four way latch.
That gives me a 4 X 4 control matrix. This means 16 patches available. In my
current patch set, one switch clears the synth to a straight through signal
path - bass to LNA to amp driver and the out of the synth. Six other switches
control two banks of three - three effects and three different path modifiers
such as; filter banks, envelope genesrator and ADSR, auto wah with an envelope
follower and a VCF. The last switch is waiting on the construction on a switch
on / switch off buffer. The switch controller debounces the switches so the
ON/OFF switch should be fairly easy. It's just there are more pressing things I
want to play with on the bench.

This controller plus patch cord flexibility has provided as many options as I
needed. Half the time, when I play with the group, I'm not allowed to use some
of the patches I've come up with.

This entire control system is analog. Not a digital chip or a microprocessor in
the batch. The trick came to mind while I was sitting with my wife in a small
concert that was supposed to be chamber music but actually was three Religious
singers backed by 3/4ths of a quartet. I had been thinking about the problem for
a few days and realized I was thinking digital. Three wires meant two logic
levels and a ground. I realized that if I used ON / OFF / GLASS HALF FULL then I
had three conditions to play with. This made eight switches easy.   (The logic
table actually has nine conditions 3X3.)

I'm all for digital - spent my US Air Force career repairing mostly digital
radar systems. I was the expert on the automatic test equipment run by a
Honeywell computer. 16K of memory and a 5M hard drive with 12 inch platers. I
actually learned to program using that equipment (and spending hours in the
factory learning the hardware and software while the contractor built it.) I
started really learning about analog signals while repairing the computer. It
had 24? (I don't remember exactly) circuit boards and a wire warpped back plane.
Most of the circuits were implemented with open collector - pull the line down
logic. If a part was bad, a signal line could be low, high or some where in
between. If a wire wrap was burnt open, a line could be low, high or in between.
The first trick I and a civilian AF technician figured out was that in between
meant, you were looking at the level the logic floated to when nothing was
attached. Next trick was discovering that if you pulled out a board unrelated to
the problem, that might just be where the pull up resistor was.

To make this shorter, I went from that to building computers including one on
radio shack 44 pin edge connector boards with wire wrap and solder and a back
plane! (okay, I brought it back to radio shack, maybe the guitarist will
recoginize that I'm back on the right chord changes and join back in. . .
please??? now to get the tune back to control.)

Computers are great espcially for controlling things. Many a time I have looked
at using a microprocessor. I went on websites as recently as last week (I think
it was ATMEL). The main reason I haven't used microprocessors are:

001) Commercial is as good for the price.

010) I can learn and write and debug with a new language or I can practice bass
and banjo (anyone know any good banjo books? My wife was going to get me a
twelve string bass for my birthday, but it was goine. So my wife bought me a
banjo and my son bought me a digereedoo (spelling?).

011) If you think analog parts disappear fast, look at the life span of
micrprocessors.

100) Microprocessors require a programmer. They cost money - that I would rather
spend on a good scope.)

101) Did i mention that the goal was each module under 5 USD. That was easy when
I picked up a bunch of NJM324s for under 20 cents. 8 inverters using a 324s is
cheaper than a hex CMOS inverter (and better power supply range to boot.)

110) This is a hobby for me? I like trying to figure out an analog solution to a
digital problem. It usually results in me thinking outside the box and building
in even further capability.

111) I'm ornery, you just can't make me do something I don't want. (Actually the
real answer is time and development equipment cost)

Tim Daugard
AG4GZ 30.4078N 86.6227W
Finished rebuilding the front sprinkler system from the huricane - maybe we can
save some flowers. I'm almost dried out so lunchtime and I think cutting logs is
next.