MATRIXARCHATE: User's guide. page 7 of 17
PASSIVE CONNECTIONS CONSIDERATIONS
Here we should understand two things. As soon as a row is connected with the column, they form a
NODE, and any additional connection with the same column or row will join this common node. And a
node shall not be confused with a mixing stage, logical 'AND' or a dedicated isolated connection. It's just
a straight electric node tying all the endpoints together. So, you will get a predictable result only
connecting ONE source to one or multiple destinations. When you connect multiple sources together,
the signals will not be added. Instead their currents will be added and you will get a mean value of their
voltages and only if the source output impedances are close to each other. The Doepfer standard
suggests 1 kΩ output impedance but some modules may have significantly lower impedance such as 220
Ohms or even 100 Ohms, so their signal will dominate over others.
Also, you may think: OK, I have two trigger outputs, and a sensitive logical module, so even at the half of
the logical one's level it will be triggered, so I can make a logical OR, interconnecting multiple sources in
the same node. Most of the time it is so but if some logical module have an open collector going directly
to the output jack, it will literally tie this line to ground on the logical zero's state, so everything
connected to the line will be grounded as well. However we will further assume that a full standard
compliant modules with 1 kΩ output impedance are used.
Example / Nerd Info:
Input 1: Vout1=5 V, Rout1=1 kΩ
Input 2: Vout2=8 V, Rout2=1 kΩ
Average signal = (5+8)/2 = 6.5 V
Input 1: Vout1=4 V, Rout1 = 100 Ω
Input 2: Vout2=8 V, Rout2 = 1 kΩ
Average signal = 4+((8-4)*100/(1000+100)) = 4
4
11
V
Please keep these considerations in mind when planning your patch.
Let's assume that in our example we want to also control the cut-off frequency with the pitch CV. Make
an additional connection at 1-F:
N91092093093092094109090
N9090909090909090
The correct note tracking is gone and envelope now modulates the pitch. Why is it so? Because we just
joined Row 1 with Row 4 via Column F. And also, we joined Column A and Column F via Row 1, so we
have a common node joining together pitch CV, envelope's output, V/Oct input of the Kotelnikov module,
and also the VCF's input. If we untie envelope from the pitch CV, the pitch tracking will get back to
normal operation. Of course, you may experiment with tying many lines together and sometimes get
interesting results, but first you need to learn how to make predictable patches.
