Logic for noobs who can't do arithmetic

[Gold]

I've done a little googling and searching around these parts but haven't been able to find anything that gives any practical examples for switch matrices. An interlocking mute, solo or source select application is what I have in mind.

I went looking for application notes but didn't find much.

 

[NewYorkDave]

Can you be more specific? Mechanically interlocking pushbuttons? A rotary? Direct or electronic switching?

 

[Gold]

Electronic switching. Momentary action pushbuttons that control electronic switches or relays via TTL. It would mimic an interlocking switchgroup. Pushing one button closes a relay or electronic switch while simultaniously opening all other switches in the group along with LED indicators of the switch status.

 

[NewYorkDave]

Did you look at this yet?

http://en.wikipedia.org/wiki/Logic_gate

 

[mikep]

well, first youve got to debounce the switch. this is way more important than you think! it can be done a least a dozen different ways, I wont get into that. then there is the logic, which for your one-at-a-time matrix would be pretty straightforeward, as these things go. then you need the actual switch element, either an open collector driver + relay or the SSM's that were mentioned in the crossfader thread.

you could eliminate 90% of the parts by using a microcontroller. this is really the best way to do it, IMO. everything but the momentary switch and the relay/SSM can be done in software. I know that sounds scary but really it isnt. you wont believe how simple the resulting PCB layout will be compared to a 4000 series CMOS approach. maybe you can get someone here to help? a PIC development board is relatively cheap and IMO pretty self expanatory...
http://www.ccsinfo.com/product_info.php?cPath=Store_Development-Kits&products_id=18F452kit

mike p

 

[stickjam]

It so happens that I'm working on the same thing.

Here is a drawing: http://home.comcast.net/~stickjam/InterlockSwitchIdea.gif

My design objective (and I'm guessing yours) is to make this work just like the good-ol Schadow interlocking pushbutton banks--including the ability to push in more than one simultaneously and have more than one remain activated.

I haven't breadboarded it yet so I'm not 100% sure it'll work as expected, but it really seems like it will. It shows only four switches but could be expanded to any number.

It uses TTL 74xx74 D with Preset/Clear flip-flops as the latches. The button switches are debounced by the RC network and squared up with the 74xx14 inverting Schmitt Triggers. The diodes form a n-Input AND gate into another inverter to reset all the latches to zero when a button is first depressed. The idea is that the rising edge of the clock inputs reset the switches by loading the D value (strapped to 0) on all flip-flops except those whose ^Preset inputs are being held low--those will be turned on.

I haven't come up with hard values for components, but I'd start with 5.1k for R1, R3, R5, R7 and R11; 100K for R2, R4, R6; 10K for R12-15; For the caps, I've got a ton of 4.7uFs--thought I'd start there. Transistors are whatever you need to handle the current to drive LEDs and relays tied to the +5v rail; with appropriate series resistors on the LEDs and suppression diodes on the relay coils of course. You'll also want to sprinkle around some .1uF caps between +5V and ground to decouple the power rails.

You can add a "Cancel all" button by adding a single 5.1K resistor between +5V (Vcc) and the buss connecting all the ^Clear inputs of the Then add a button between the ^Clear buss and ground.

You can also designate a button as the power-on-default by putting a cap (maybe 10nF) and a big resistor (maybe 470K) in parallel across the desired button.

Hope this helps. I'm open to suggestions, but this seemed to be the simplest digital "hardware" solution.

--Bob

 

[JohnRoberts]

+1 on using latching FF but I'd be tempted to simplify. My favorite FF for such stuff was CMOS 4013.

A simple diode OR from outputs could be cap coupled in to reset everybody. You could probably debounce with an RC so the schmidt trigger is perhaps overkill.

If you're doing more than just switching a cheap PIC could do everything inside without any glue parts needed but that's a bit much of a learning curve and excess horsepower for such a simple task.

Debouncing CMOS can be a PIA but it's doable. I recall years ago doing a variant on this for a tape recorder remote foot switch.

JR

 

[Gold]

Thanks all. These are great replies. I have enough to go on for a while.

 

[stickjam]

[quote author="JohnRoberts"]+1 on using latching FF but I'd be tempted to simplify. My favorite FF for such stuff was CMOS 4013.

A simple diode OR from outputs could be cap coupled in to reset everybody. You could probably debounce with an RC so the schmidt trigger is perhaps overkill.

If you're doing more than just switching a cheap PIC could do everything inside without any glue parts needed but that's a bit much of a learning curve and excess horsepower for such a simple task.

Debouncing CMOS can be a PIA but it's doable. I recall years ago doing a variant on this for a tape recorder remote foot switch.

JR[/quote]

Like this? http://home.comcast.net/~stickjam/InterlockSwitchCMOS.gif

It is a bit simpler in CMOS with the active-high set/resets. In this case I used diodes to implement a rising-edge reset input via the clock inputs.

 

[sonicmook56]

You can also get the HEF4013 with a schmitt trigger built into the input, so you can loose a chip or two.

Get the book "cmos cookbook" A very good book that I learned 90% of what I know about logic out of. Should be cheap, I think mine was $17 from a used book store.

 

[Ptownkid]

uuugh, I had to take two digital classes and i still don't get this *****...

 

[Gold]

Thanks Sonicmook, I ordered the book last night. I found a little PIC kit I'll order and see if it makes any sense.

 

[ChrioN]

I'm working on a mixer, and I pretty quickly found out I couldn't just use dpdt switches for solo facilities...
Anyone got any simple circuits for this?

 

[SSLtech]

Can you be specific about "solo facilities?"

You can do either PFL or AFL solo with a dpdt, dozens of manufacturers do it.

You can do in-place solo with an SPDT, so long as you already have logic controlled muting (which is going to be a prerequisite for ANY in-place solo design).

I don't understand why a DPDT switch should be a limit...

Keith

 

[ChrioN]

What I meant was, to use just a dpdt switch, without any logic circuit. Is that possible?

 

[JohnRoberts]

I don't know about logic but there is often some electronic switching in the master section to listen to the AFL/PFL bus instead of normal CR feed.

I did a trick solo/mute grouping circuit for a friend's custom console decades ago. It allowed him to set up several groups and either mute the group or solo the group. This Solo was destructive, as in it just muted everything not solo'd in the two-mix but he found it useful for mixing (Reggae I think).

I forget the details of the logic but it was probably discrete to interface with the JFET channel mutes. He used a 10 position thumb wheel on each strip to assign the channel to a group which could then be either solo'd or muted as a group by on-off-on switches in master section. Up for solo, down for mute. There were also bi-color status LEDs on strip so you could see what was going on.

JR

EDIT: BTW the standard channel SOLO requires 3 poles. 2 for the stereo signal and one for control, PFL could be done with 2 pole switch and unpanned mono signal. I guess one could be clever and superimpose DC on the stereo SOLO signal to control downstream switching with only 2 poles, but I haven't seen that done.

 

[Fredrik]

[quote author="ChrioN"]What I meant was, to use just a dpdt switch, without any logic circuit. Is that possible?[/quote]

I´we been thinking about the same thing. What about splitting the signal just after the input and feeding the signal to two summing networks. One main bus and one solo bus.

On the feed to the solo bus you put a DPDT switch where one pole adds the audio signal to the solo bus and the other pole closes a 5 volt circuit which activates a relay that switches from the main summing net to the solo net just before the summing amp. Then connect pole 2 on all the switches on all channels in paralell, that would switch in the solo bus nomatter witch switch you closed or if you closed one or many.

 

[dshay]

I'm partial to C++, if you are too you may want to look into microcontrollers of the AVR variety by Atmega, quite cheap and reliable.

Hook straight up to your computer to load your code and do other serial communications via your USB port.

There is an open source prototyping platform/community like Arduino for it. (or Wiring, the competition to PIC)

http://www.arduino.cc/

http://www.wiring.org.co/