figuring out more complex switching

[caress]

i'm curious about using eproms or digital logic in switching arrangements requiring the switching of many i/o at once.

i've been reading the tone god's articles on switching, as well as gathering info on eproms.

for one example, i'm thinking about things such as:

a----aa
b----bb
c----cc
d----dd

using a dpdt and switching d-dd off with one throw, c-cc and d-dd off with the other throw.
another dpdt to switch b-bb, c-cc and d-dd off with one throw, all off with the other throw.
hopefully working up to larger switching arrangements with  more i/o or more complex routings...

if i use digital logic, it seems that i will need to use a LOT of chips.  way more than i'd like to use, at least... i don't know anything about programming eproms yet...

any ideas here?

[Artie]

What do you anticipate your max number of channels being? If its less than approx half-a-dozen, then there's an easy way to do this with inexpensive relays, (like this), for example. Once the number of channels exceed 6 or so, its becomes less practical to use the relays. (Still possible, however.)

If you end up going with logic, there's still a couple of tricks for keeping the parts count down. Like using diodes to "program" gates.
I can post a diagram of what I mean, if you need.

[caress]

If you end up going with logic, there's still a couple of tricks for keeping the parts count down. Like using diodes to "program" gates.
I can post a diagram of what I mean, if you need.

that would be cool.  i don't mind using relays, but i'd like to be able to go up to probably 8-10 i/o using either a rotary switch or 4-5 dpdt/spdts as hardware...

[Artie]

I'll post a diagram showing both ways. One thing to keep in mind, when using relays or logic, to handle the switching chores, your "human-interface" switch can be much simpler. In other words, you only need a simple contact closing to accomplish complex switching. So, you can use a simple SPST switch for each function, or a less complex rotary switch.

[teletroy]

awesome- I've been wanting to do something along those lines too

[caress]

in the end i think one application i'm looking for is a simple way to use a 1P8T rotary switch to switch between:
a-a
a-a/b-b
a-a/b-b/c-c
a-a/b-b/c-c/d-d
a-a/b-b/c-c/d-d/e-e
a-a/b-b/c-c/d-d/e-e/f-f
a-a/b-b/c-c/d-d/e-e/f-f/g-g
a-a/b-b/c-c/d-d/e-e/f-f/g-g/h-h

would this be right -- 2 4066 chips just for the 8 individual switches plus the control logic.  then if i understand correctly, i would need to duplicate this for each switching section?  or is there a different way to interface it so that i would only end up with 3 or 4 chips for the switching?

[Artie]

You may be making this more complicated than need be. Is the above configuration hard set, or do you actually need and/or want, programmability, versatility, and adaptability?

I'll be posting the pic of what I was talking about shortly. I'm just tweaking it to match exactly what you want.

[Artie]

Ok, here's what I was thinking along the "relay" lines. And don't freak out when you first look at this. It isn't nearly as complex as it first looks. Here's the "master" switching matrix: (Completely programmable, expandable, and configurable.)

Switch Matrix 01

K1 - K8 are the relay coils. That "extra" diode on the bottom of each relay is to suppress back EMF when the relay turns off.

I only show the input/output section on the first two relays. Those two rows of little circles are like the jumper-headers that you see on motherboards. They're available at most Radio Shacks, and they're inexpensive. You just put those little jumpers on whichever channels you want to activate with each rotary switch position. Any combination of I/O channels can be "on" at any rotary switch position.

As an alternative, you can skip the little jumper headers, and simply install a diode where you want an active channel. For example, to configure this for the modes you specified, you'ld just do this:

Switch Matrix 02

The way I'd do it, (and I am building one of these for myself), would be to build each "channel" on one of those little Radio Shack project boards, (like this), and then orient them like this:

Switch Matrix 03

That way, you can just run your channel buss wires, (colored lines), straight through the holes on one side. Then connect the ends to your rotary switch. As you can see, it only takes one simple DPDT switch for a master bypass switch. When its "off", all the relays relax to their bypass position. Also, any number of channels and any size rotary switch may be used. Additional channels can be added at any time if you use the first version.

Make sense?

Digital logic version forthcoming. Stay tuna-ed.   

Artie

P.S. One of the things I like about using relays is that they aren't sensitive to power supply aberations, and you can make almost any voltage relay work with any voltage supply, (as long as the relay voltage is less), by simply including an appropriate resistor in the ground lead. Also, relays keep the audio signal out of the "electronic" domain. It only travels through relay contacts.

[caress]

cool.  this is actually more or less how i was thinking about it... just didn't know about using the diodes like that.

the digital logic configuration is pretty similar, i imagine?

[Artie]

Not necessarily. There's so many different options with logic. What might be a simple configuration for a manufacturer, with a complex circuit board, could be daunting for the DIY'er. I'm still doodling up a couple different options. I'll try to post something tomorrow.