Re: Effect of lengh of wires in matrix


roxlu wrote:
Luhan wrote:
roxlu wrote:
Hi,

I've created a small prototype where I connected 64 tiny dip-switches
in a matrix of 8x8. Using a
serial in shift register and a serial out shift register I'm scanning
through these buttons to check which one is pressed. This works
perfectly.

Now I'm gonna buy a lot more and bigger buttons (33mm, 256 pieces). The
matrix will be approximately 100cm x 100cm. As I'm not very experienced
in electronics I'm curious how this will afffect the workings of my
circuit.

I'm using a PIC16f628 for the scanning and I'm communication with a PC
using RS-232.


Sounds like a good design. My money is on it working just fine. There
is a built-in lag in reading via the shift registers.

Luhan


Luhan
But are those long wires to all those butttons (256) in a 100cm x 100cm
wooden plate gonna interfere my circuit in any way?

Thankls

When I built my pipe organ, the keyboards were 8x8 keying across 36"
(91cm) but only 3" wide. The scan rate was 3kHz. The matrix board was
connected to the microcontroller board with 18" of ribbon cable. Pullup
or pulldown (whatever is appropriate) resistors are needed. The stop
knob matrix was across the width of the console (60") and also had no
issues for the 13 years I had it. The exposed wiring did have a
tendency to make some RFI as I didn't shield anything. It mainly messed
up a nearby AM radio. The controllers were 5 8031s.

I was concerned about pulse rise/fall times and all that capacitance
from the ribbon cables so the 'scanning' of the stop knobs was not
raise 1 bit high and look for what comes back. Instead, the switches
all came into tri-state buffers and the buffers were enabled in
sequence. The software was identical but the long lines in the ribbons
were not being pulsed.

http://haskey.com/johnh/organ/organ.html

GG

.

Loading