Pedal Switcher?

[shawnee]

How hard would it be do build a pedal switcher like the Cral Martin Octaswitch?

[chip46]

The octaswitch is very involved. I would say without a background in Electrical Engineering it might prove rather difficult. That's not to say it couldn't be done with the proper research or if you have a good background in DIY projects and electronics, but from what I remember it wasn't as simple as it looks on the outside. I remember someone on a forum built one, but the majority of the posts were in spanish I believe, so I couldn't follow it too much, but I guess you could copy and past the posts into a translator.

Something like a standard TB looper is much easier.

[shawnee]

You are probably right Chip. I did find this on R.G.'s website:
http://www.geofex.com/article_folders/fxswitchr/fxswitchr.htm

The new Octaswitch 2 can do amp channel switching too so I may have to bite the bullet on that one of these days.

[jkokura]

I think the Octaswitch type of programmable switcher is actually one of the simpler ones. It involves a bunch of DIP8 switches to make your presets, and then a series of relay based switches that may need to have a PIC system that turns off the last preset when you turn the next on. Check out the Cusack Pedal Board Tamer for a really cool switcher, but it's much harder to build something like that.

Anyway, the point is that for DIY that route isn't really cost effective. It would be cool to build yes, but unless you were going into production on a similar product and hope it sells really well, it's doubtful that you're going to really save money over buying one.

Jacob

[chip46]

That's another good point that I forgot about Jacob. Something like the OctaSwitch or PBT aren't very cost/time effective to build DIY style, especially for the OctaSwitch and all the holes that will go into fitting the dip switches to the enclosure and what not. Definitely a pain to do and buying the real thing would save some money and sanity I think.

[JakeFuzz]

I totally agree that this wouldn't be cost effective and I also don't really have a need for this sort of thing but I think it is sort of interesting to think about how to make it in the simplest way possible. Here is how I would do it...

Get a microcontroller with at least 8 digital IO pins and 1 analog in pin, something like an ATmel 8 would work. Hook up 8 NO momentary switches as a big voltage divider with a common switchable output. This will divide 5 volts into 8 voltage ranges and go into the analog input pin of the MC. Hook up digital IO pins 1-8 to the bases of 8 different transistor gates. All of the collectors are connected to a 5 volt supply and the emitters each go to one side of 8 separate DIP8 switches. To clarify one whole side of each DIP8 switch will be at 5 volts when their respective control pins goes high. The other sides of the DIP8 switches are all common to each other, pin 1 connected to pin 1 for each switch and so forth. Now these 8 MUXed outputs would go to the control coils of 8 individual 5 volt DPDT relays, one pole to activate or bypass the respective effect loop and the other for the loop indicator LED.

The code would be super simple and you could probably write it in a few lines minus all the necessary declarations. How it works: When you step on one switch it sends a specific divided voltage to the analog input pin of the MC, lets say the 3rd switch so that means 3.125 volts. The MC will detect this and be able to discern it from a non active state (ground maybe?), of course some voltage range needs to be built in due to resistor tolerance. So now the MC knows which footswitch/loop has been selected by the user and will send one of the digital output pins high (and the remaining 7 low) based on which loop the user wants. This will switch the transistor into saturation mode (conduction between C and E) and send one of the DIP8 switches high. Depending on which loops the user has set for that particular DIP switch, the switch will allow 5 volts to the appropriate relays and activate those loops and LED's.

[shawnee]

I had to read that twice but I think I get it now.    Probably too involved considering all of the other things I am working on. I was hoping that it would be simpler to put together. Anyway, thanks for the input guys.

[cjkbug]

I used to have the rocktron patchmate floor switcher. it's very similar to the octaswitch and can even send and receive midi cc messages, and I think it's a little cheaper than the octaswitch. It was really well built and easy to use. it also takes up about the same footprint.

[jubal81]

JakeFuzz, you are really on top of things!
I looked up the AVR chips and controllers to program them and this looks like something I really want to learn about.
Got any educational links?

[JakeFuzz]

JakeFuzz, you are really on top of things!
I looked up the AVR chips and controllers to program them and this looks like something I really want to learn about.
Got any educational links?

Haha thanks. I love trying to figure out weird little puzzles like this.

I would start with an Arduino module (I actually need to buy me one of those as well). You can pick them up for around 30 bucks. That comes with the Atmel MC and a programming board. I like the looks of those because they have header pins and built in LED's which makes it nice for prototyping.

I would say after that start learning to program in C whether by an introductory book or online tutorials. It isn't too bad to pick up and you really don't need a lot of the advanced stuff for embedded programming. I would say the numerical processing and the basic program flow is the most important. I've had to take several classes in C and Fortan (ancient I know  )  so I don't really know where to point you as far as online resources go.

After that you would need to learn the internal syntax of the MC and compiler. These are just the specific commands you will use in your C code to interface with the controller like reading data from digital/analog pins and outputting a digital signal to specified pins. Here is the Arduino reference website.

http://arduino.cc/en/Reference/HomePage

Then I would start going through as many sample programs as possible starting with the most simple ones. Like making an LED flash at a specific frequency. I would go through every line and understand exactly what it does. Then I would start modifying parameters to see how program changes affect the physical operation (like sampling rate). Don't be afraid of doing something wrong; the compiler wont let you build a program with errors and will usually point you straight to your mistakes. Here is the Arduino sample program site.

http://arduino.cc/en/Tutorial/HomePage

From this point I would say get creative. Start with something simple (like the program suggested on this thread) or something where the passive components are pretty simple/foolproof and start writing programs that change/control some of the circuit parameters. I would think an LFO function generator would be a cool program to try. I think the most important things to focus on interfacing with (as an analog pedal person) are control voltage analog inputs, opto-isloators and digital potentiometers.

[jubal81]

Great stuff. Thanks, Jake!
I never really realized how much you could do with digital components to control analog circuits while keeping all that digital nonsense out of the signal path. Now my mind is going nuts thinking of all the crazy possibilities.
I spent about half a day watching arduino and other digital hocus pocus on Youtube with my mouth hanging open.
In college all I wanted to do was play guitar and drink beer. It's only now that I have a career in the way that I have all this stuff I want to learn ....
... so it goes. (sigh)

[jkokura]

As always Paul, thanks for the in depth info. I'm going to have to come back to this...

Jacob