[sdiy] Modular patching rules

[Tim Parkhurst]

On 2/13/07, Steffen <stffn at dibidut.dk> wrote:
>
> On 10/02/2007, at 22.53, Tim Parkhurst wrote:
>
> > the "standards" for modular synth gear are:
> > 100K input impedance
> > 1K output impedance
> > +/-5V Audio levels
> > 0 to +5V control signals (0 to +10V in some cases)
>
> Does 100K input impedance mean that there need to be a 100K Ohm
> resistor at every CV in? And likewise, a 1K Ohm at every output? I'm
> asking since
That's pretty much it, a 1K R in line with the output. You can also
get a 1K output impedance with a 2K in series and a 2K to ground (and
in this case, you've just reduced the signal by 1/2, which sometimes
is needed). The 1K at the output is there primarily to prevent a big
rush of current if the output is accidentally shorted to ground (this
can happen even when plugging and unplugging patch cords). It doesn't
have to be exactly 1K, but that's pretty easy to achieve most of the
time.

> 1) the word "impedance" make my nervous; I'd don't really understand
> it's meaning in the electronics context.
"Impedance" is resistance to AC. There are some cases where a circuit
will present different loads to AC or DC, but in the case of a CV
summer with resistors on the inputs, the input resistor sets the input
impedance.

> 2) i'm working on a module that have 49.9K Ohm resistors to every CV
> in. Will simply chancing them to 100K Ohm likely make the design tilt?
>
The input resistors are usually part of a circuit that sets the gain
of the CV summer. You will see a lot of summers with 100K input
resistors, and a 100K feedback resistor, which sets the gain to 1. You
might want to try changing the feedback resistor to match the input Rs
if you want to change the 49.9K parts you have. OTOH, 49.9K is still a
relatively high impedance and really shouldn't be a problem. What
you're trying to do is set the input impedance high enough so that it
doesn't present a big load to whatever is plugged into the inputs. In
a properly designed system, you can use something like an LFO output
to drive several VCOs (using a multiple) without loading down the LFO
circuit. What happens if you load down (i.e., draw a lot of current
from) a tuned circuit like an LFO? You'll notice the frequency of the
LFO change as you use it to drive more VCOs. Hopefully, your VCOs have
a nice high input impedance, and so they draw very little current from
the LFO and don't change its frequency. The same thing (frequency
change) can happen with VCOs if you plug them into a module that draws
too much current from them.

>
> I maybe should post this in a fresh thread. But while I'm at it: My
> scope is quite slow, more then 4kHz signals is out of it bunch. But i
> could let my computer act as an aprox. freq. counter. To do that -
> and in general to feed the synth sound to a mixer outside the system
> - I need to make the +/-5V signal into a line signal, which is 1Vpp,
> right? Any pointers to such "converter"?
>
Sounds like all you need is a couple of resistors to make a voltage
divider. Or use a pot and an inline R if you want a variable divider.
I cooked up something similar to record my synths on my PC. And a 4kHz
scope is still okay for a lot of audio frequency stuff (it's better
than no scope at all), but you might want to look at something with a
little bit more bandwidth in the future.


Tim (better than no servo at all, but only slightly) Servo
-- 
"Imagination is more important than knowledge." - Albert Einstein