[sdiy] Re: distortion circ

[Tim Daugard]

Sorry Ian (and this message actual expands beyond our discussion so I'm
going to put it to the group (and not mention at all what we were
discussing))

Using the LFO on my web site schematic page -

LFO Buffer -> current limiting resistor -> output pin -> wire -> input
pin -> resistor R2 -> system ground -> power supply -> LFO (buffer) is
the first current loop.

this develops a voltage across R2 that provides a DC reference for C2
negative terminal.

. . . current limiting resistor -> . . .  -> input pin -> C2 -> D1 -> R4
& C4 -> ground . .  is current loop two

this develops a voltage across D1 and network paralleled with Rv1 and
R3. This provides the DC reference for the positive terminal of C3 (and
the positive terminal of C2).

. . . current limiting resistor -> . . . -> input pin -> C2 -> loading
by the clipping circuit -> C3 -> R6 -> ground . . . is current loop
three.

The resistors R5 and R6 provide a DC reference and current source that
the current from the opamp (signal) rides on.

. . . current limiting resistor -> . . . -> input pin -> C2 (and RF
filter C1) -> loading by clipping circuit -> C3 -> output network and RF
filter C5 -> wire -> input to the next module (100K DC load impedance
and at least 90K AC load impedance.

Is the final major current path through the module. No where in the
system is there a capacitor that can be tied to another capacitor by
patching or anything else without there being at least one DC path to
one or both of the supply rails between them. Every output and input is
bypassed for RF (the LFO in the schematic sample has been updated with
the RF filter caps.)

Because every input and output has DC paths, there is often a DC current
flowing through the signal cables. These DC currents are limited at the
source (generally outputs) and lightly loaded at the destination
(generally inputs.) This wastes some power (usually 50 uA or so) but
ensures that there is never any pure AC to AC (cap to cap) links in the
system.

Like the old style current loop data and telegraph systems, my system
relies on currents to transfer signals. This eliminates the problems of
noise riding voltages on patch cables and makes the system very quiet.
Any noise that a cable picks up through radiation, etc. does not have a
complete path through. Basically anything that is not output by the
system signal source is common mode and ignored.

The only place I have any problems is with the start of the system. My
signals come from, guitars, bass and or microphone. Microphones are low
enough impedance that I can usually eliminate the noise through proper
impedance matching and termination at the end of the cable. For my bass,
I've been tempted to open the bass and eliminate the coupling capacitors
so that the path is DC from the coil on.

As far as the write up on the page, that is one circuit for a module
that can hold three. The write up is mostly so that if I decide to build
another copy or make design changes, I can remember what my thoughts
were. I wrote a book with more detailed write ups for publication, but
havn't found a publisher or anyone to help proof read.

Tim Daugard
AG4GZ 30.4078N 86.6227W Alt: 3.7 M
http://home.sprintmail.com/~daugard/synth.htm

And to concurr with, I think it was Harry, Ohms second law and
Kirschoff's are essential. Filter design and transistor circuit design
are easier with them. Once I understood them, I realized that everything
that comes out of the output of an opamp has to go somewhere and if I
planned where the opamp currents were going to go, my circuits would
behave better.

{so what kind of currents come out of an LM308 and how does the
compensation capacitors change the current flows? ;-}