3 channel momentary preset selection

[R.G.]

What I've got sketched out is a 14 or 16 pin chip that takes in up to four momentary switches, and turns on one and only one of four outputs.  So far, so trivial.

But I think I can make the chips listen/drive two lines and work in concert with N other chips so that you can expand it from 1 of up to 4 to  1 of up to 8, 12, 16, 20, 24...

I'm pondering whether this needs a PCB which contains drivers for up to four relays (as either a chip or four transistors) or also contains the relays themselves. There's advantages and disadvantages to all of these. The disadvantage of putting all that on a PCB and not using it is that the PCB costs more. PCBs would be stackable to N as well by coupling the signaling wires across.

(To the programmers out there: two-wire and one-wire signaling schemes are the natural, but I think that's overkill. I think I can get the same result more simply and robustly other ways. That's why two wires plus power and ground.)

[defaced]

This is one of those applications that I really wish someone would open source in a language other than assembly.  I've thought about doing it as I have done a project like this (two chips, serial communication, 8 inputs/outputs), but alas, great ideas that never get published, partly because I have _zero_ desire to claim to be a programmer, or want to manage the project, or want to learn MIDI (which on the surface to a layman such as myself seems to be a suitable communication protocol).

[R.G.]

Yeah, I've done several designs like this. The desired result is straightforward. The programming in terms of what needs done is trivial.

However, I'm becoming more convinced that any time the words "menu" or "MIDI" come up with respect to pedals, it's self limiting. Only a few musicians will use it.

By far the most complicated part of any multi-chip communicating setup is going to be the communications protocol.  One-wire, two-wire, MIDI, RS-232, whatever, all carry a lot of overhead in terms of getting what needs sent encoded, sent, error checked, response checked, and decoded. It takes a fair amount of work to write down *exactly* what needs to be done at the bit level, then to translate that into a transmission logical protocol, then the physical protocol, and back again.

[jpwilksch]

R.G., perhaps split out the relay & driver into a separate small board. That way, you only buy the number of output boards needed (be it 2, 4, 7,  27, etc.)

[alexradium]

I could never get RG's circuit to work properly- dunno why. Anyway, I designed an alternative (four switches shown, but you don't have to use them all, obviously). All diodes 1N4148 or whatever. The transistors are just examples of circuits that could be controlled by it- they have nothing to do with the actual switching controller, so you could just connect a resistor+LED to each output. Also, the IC has both Q and /Q outputs (not shown), which can be handy.

thank you so much for this,Merlin!
i just did a 4 channel switching board for my preamp with no midi,using a 4066 to close the appropriate contacts.
Anyway,there is one minor point to this circuit,when i give power all the outputs are on,so i have to touch one time any button to "reset" to normal one at a time channel selection.
Is there any way to accomplish this reset?
i know that Soldano x88 uses a cap and diode on one input,but those buttons go to ground instead of V+.
Thanks for any advice,just an idea  could be of help.

[miknshla]

resurrecting this old thread...

I'm looking at options for 3 channel amp switching.
I have seen the Aiken design and the Valve Wizard design and I made a MIDI footcontroller about a year ago with a WEMOS MINI but I don't have experience with PIC controllers.

I think the ultimate design would:
1. Allow selecting channels from the amp front panel or the footswitch when the footswitch is connected
2. Have LED's on the footswitch to indicate the current channel
3. Only have one cable from the amp to the footswitch that is readily available and has enough pins.

If I go with the Aiken or Valve Wizard design I could use a 5pin din cable but I don't see how you could select channels from the amp whenever the footswitch is connected.

If I use the Midi controller it would allow more options than just 3 channels such as boost, reverb, etc. but to get footswitch controller to read the current state
I think you would need a 2nd midi cable so you would have MIDI IN and MIDI OUT. 

Any ideas have ways to achieve the three features listed above?

[njkmonty]

[njkmonty]

https://oshpark.com/shared_projects/ZAjPHO3a

[MetalGuy]

CD4028 also can do 3 channel switching.

http://www.seekic.com/uploadfile/ic-circuit/201172353854920.gif

[miknshla]

Sure, but read my post.  Once the footswitch is connected these designs won't allow channels to be selected from the amp, only the footswitch.  So you basically have to disable the channel select switches on the amp whenever the footswitch is connected.  No big deal, I guess.  I was just wondering if someone has solved this issue.

Basically the footswitch would have to update it's channel selection if a channel was selected  from the amp.

[miknshla]

I think one option is to forget the MIDI standard and just use a cheap micro-controller on each end utilizing it's serial port to communicate back and forth.
I have used the WEMOS D1 Mini Serial port before and can get them for about $2.67ea.

[PRR]

The clever, pre-cheap-logic, way was to send raw AC through a resistor to the pedal. Switches and diodes would short-off the +, -, or neither side of the AC wave. This can directly run LEDs in the pedal. Minimal logic in the head would detect the deformation in the wave as three states, to work LEDs and relays. Minimal more switching allows the head to work without the pedal, if forgotten lost or busted. 2-wire interface, spare guitar cord (but DON'T mix it with signal inputs).

[merlinb]

Sure, but read my post.  Once the footswitch is connected these designs won't allow channels to be selected from the amp, only the footswitch.  So you basically have to disable the channel select switches on the amp whenever the footswitch is connected.  No big deal, I guess.  I was just wondering if someone has solved this issue.
Basically the footswitch would have to update it's channel selection if a channel was selected  from the amp.

Interesting problem. Based on MetalGuy's clever suggestion, I *think* this would do what you want. (I haven't shown any LEDs but you would just attach them from the 4028 outputs to ground, with current-limiting resistors obviously)

[merlinb]

Thinking about it some more last night, the diodes render the feedback resistors redundant, meaning we can use two identical logic blocks on a continuous bus. I find the symmetry of this intensely pleasing...

[PRR]

> Interesting problem.

I was expecting you to show three different color LEDs and window-logic to get it all on 2 conductors, not five.

Fer bad example: my car gives three wiper-speeds on one wire and ground, by sensing 220-470-1K resistors. A similar gimmick gives three cruise-modes on one wire. (And when the switch gets a hundred Ohms of dirt in it, it starts mixing modes...)

[merlinb]

I'm such a dumdum. You don't even need the second IC, as long as you don't try to consume too much current for the LEDs the footswitch can be passive:

The clever, pre-cheap-logic, way was to send raw AC through a resistor to the pedal. Switches and diodes would short-off the +, -, or neither side of the AC wave.

Can you remember an actual amp that did this? It would be nice to see an orignal circuit.

I was expecting you to show three different color LEDs and window-logic to get it all on 2 conductors, not five.

Heh, two wires and still have LED indicators in the footswitch? Another interesting problem...

[merlinb]

Don't know why I bother 

[PRR]

> Don't know why I bother

Memory is bad and I have been distracted by electrical and medical problems.

Fender Frontman 212r

[Ripthorn]

First thing I thought when reading the OP was to use a microcontroller. I have recently gotten into them quite a bit. I prefer the ATTiny series and program it with Arduino. This strips away most of the assembly and allows most everything to be written in C/C++. Super simple if you have any programming experience (mine is predominantly Matlab, and even that works!). In fact, I just recently wrote up a guide on how to program ATTiny's with an Arduino and have a board layout for a programming board. I also love that the Arduino community is super engaged. But then again, I love that because I struggled a bit when first getting into MCU's. But they are so cheap and plentiful that I like to do it that way. I did a 2 button version of this when I first started and it works great. Extension to N buttons is not that big of a deal.

Anyway, if you want to see what I posted, you can find the board here: https://scientificguitarist.wixsite.com/home/attiny-programmer
and the guide here: https://scientificguitarist.wixsite.com/home/programming-attiny-mcu-s

I never even knew about ways to do it with solid state logic, and I work for a semiconductor company! 

[slacker]

The clever, pre-cheap-logic, way was to send raw AC through a resistor to the pedal. Switches and diodes would short-off the +, -, or neither side of the AC wave.

Can you remember an actual amp that did this? It would be nice to see an orignal circuit.

Fender does it, I had a late 90's Twin with a four button footswitch with LEDs using only a two core cable. Here's an example schematic http://ampwares.com/schematics/Twin_Pro_Tube_Amp_SchE45.pdf