Project wishlist

[Dave_B]

I thought it might be a good idea to have a thread for ideas.  I don't think one has been started, has it?

Example:
A µC'd footswitch that engages/disingages an effect when you tap it, but acts as a momentary switch when you hold it for longer than 1 second.   

The reason I want it: It's really nice to have a momentary switch on an effect in addition to a standard 3DPT.  IMO, it would be nicer to have one switch that can handle both functions. 

[Peter Snowberg]

GREAT idea! 

[Mark Hammer]

The CMOS switch in the old Anderton EPFM book will do something like this quite easily.  It uses a 4016 or 4066, and is normally enabled by a latched SPST stompswitch.  But you can easily parallel a momentary with the latching switch, and stick them both in the same box or beside each other in the same pedalboard.  The momentary is moot when the latched switch is engaged, but if the latched one is open, you can step on the momentary for as long as you want the effect/loop on.

I used to have an arrangement with two sets of switches laid out: latched A -> momentary A -> momentary B -> latched B.  If I wanted I could position my foot to step on both momentaries at once, or just one at a time.  VERY handy, and well worth doing.

[Quackzed]

i had a cool idea for a rotating mic thingy... but it had been done, so i figured i'd put a new slant on it...
what about a box with maybee 4 rotating stereo unidirectional mic pairs in a row.... each mic pair could have volume and speed knobs...might yeild some interesting phaser univibe sounds, plus you could mic anything with it, not just "electric" instruments...
   ...or it could be made like a 4 layer cake with a wider orbit for the mics witch would add some doppler shift in a la "rotating speaker"...
I love mechanical effects... and weird gizmos that take sound manipulation "out of the box"
i'm sorry ... i didn't mean it!  blasphemy!

also a collapse switch for mono (1 side of each rotating stereo pair)...

[Dave_B]

But you can easily parallel a momentary with the latching switch, and stick them both in the same box or beside each other in the same pedalboard. 

True.  The thing is, my stage space is really limited where we rehearse.  If I can save a foot (as I pretty much would the way I'm calculating it) that's a big bonus for me.

[R.G.]

A µC'd footswitch that engages/disingages an effect when you tap it, but acts as a momentary switch when you hold it for longer than 1 second.

At the risk of being redundant, has anyone looked at
http://geofex.com/Article_Folders/footswitch_pancake/footswitch_pancake.htm#A%20computer%20controlled%20pancake which shows all of what you need to do that. The only embellishment is to make the uC look for instantaneous taps versus hold-down.

As for ideas, try searching the build-your-own-stompbox page with "R.G." and "PIC".

[Dave_B]

At the risk of being redundant, has anyone looked at http://geofex.com/Article_Folders/footswitch_pancake/footswitch_pancake.htm#A%20computer%20controlled%20pancake which shows all of what you need to do that. The only embellishment is to make the uC look for instantaneous taps versus hold-down.

I've looked at that page several times R.G., thanks.  I know you've provided a lot of the building blocks and in many cases some of us are playing catch up to things you've thought of five or six years ago.  Did you write source code for that switch?  It would be really nice place to start for people who haven't seen a practical application of PIC's and guitar effects. 

As for ideas, try searching the build-your-own-stompbox page with "R.G." and "PIC".

Well, the thing is, I'd like to get some brainstorming going.   Some really good ideas come from people who are too fresh to know the rules, ya know?  If no one else takes the bait, I'll write the dual-purpose switch code and post it when I can test it with hardware. 

[R.G.]

Did you write source code for that switch?  It would be really nice place to start for people who haven't seen a practical application of PIC's and guitar effects.

Let me see if I can find it.

As I remember, it was written in PIC Basic and did something like set a timer for about 10mS. The program looked at the timer and when the timer went to 0, the program went off to do switch processing.

Processing was something like:
Define variables for current switch state, previous switch state, current pin logic value, previous switch value, and a count of the number of times the pin has been read the same.

The main loop just checks to see if the timer has clicked over, and if it has not, loops back to itself.

The break out of the main loop when the timer clicks over is once every 10mS.
At that interval, the program turns off the timer (just to keep it from doing spurious things, if any) and then reads the current value of pin.
If the pin is the same as the previous value of the pin, no change has happened, so start the timer back up and go back to Main.
If the pin is different from the previous value of the pin, then a change MAY be happening.
If so, count up to four reads of the pin in the same new value before changing the state of the switch.
Any less than four consecutive reads of the new value resets the counter for four reads.
At four consecutive reads the same (40mS), the switch must have changed. The current and previous values of the switch state are updated, and I go to the I/O subroutine to do whatever happens when the switch changes.
Once the I/O is over, I clean up variables an jump back to Main.

Notice that the pin condition is separated from the "switch state". The switch state is a conceptual state, based on the history of what the pin that senses the switch has done. The pin is periodically read, and when it changes, it must stay changed for four consecutive readings to eliminate switch bounce. When that has happened, the "state" is updated.

The state may be updated by several different logical rules, and this is where you get the luxury of choosing different actions. If the state follows the pin's action, then the result is a momentary switch, where the switch state as seen in the I/O follows the pressure on the actual switch. If the state is set to toggle on each pin transition from one to zero (for instance), then the state is alternate action, or a latching switch. If you want other actions, you have only to look at the current state (not pin value!) and decide what to do based on the state, the previous state, and perhaps a timer if you're after the auto switching from momentary to alternate.

Notice the important points: timing is NOT dependent on the external devices; the condition of the external switch determined by reading a pin only tells you that something has happened; WHAT happens is independent of the timer, and the condition of the pin; this lets you substitute in any action to take when the pin changes, including ignoring one switch direction entirely. The connection between these modules is entirely through well defined state variables, and only through those variables.

It's the independence of the modules that lets this be flexible. You can easily write a tight, simple switch reader in line with other code - and you'll be tinkering with it every time you mess with the code. Like all very flexible schemes, this is bigger and slower - but infinitely more adaptable.

I'll see if I can find it.

[gaussmarkov]

to add to the wish list things that seem (to the uninformed and inexperienced) like good starting points:

-a signal generator, if this is easy then it would be a useful bench tool also

-a sampler that loops, seems like a project that could start simple and get more complicated 

[Dave_B]

A signal generator is a great one!  If we included sweep and burst generators, all the better. 

I'd add to the list a simple MIDI program changer.  I've got an old Midiverb II that I'd like to start using again.  If I could set one button for program 99 and another for one of the delays, I'd be happy.

[Dave_B]

A l-o-o-o-o-o-n-g delay.

Hopefully using the old 128k and 512k DRAM chips we're all accumulating.  I know there are articles on the internet for using dram with µ's, but I've not seen not one written specifically for making joyous noise.

[Primus]

I want a ready-to-go DSP board with a standard microcontroller that I can just build into the enclosure of my choice. I want to be able to add inputs and outputs to the microcontroller and program the DSP by USB and I want it to cost <$75. And I want a pony.

Failing all that I want just a DSP-in-a-box with a rotary preset selector that is programmable by RS-232... for <$75.

[idlefaction]

I think a good one would be a small board that mounts on the back of some standard, high-quality momentary-action stompswitch, containing:

Stomp switch
uC
latching-coil relay
LED, possibly bicolour or RGB
5v regulator
voltage inverter
Input and output buffering for the effect

The uC could debounce the switch, toggle the LED and relay, giving you true bypass with LED.
It could have a MILAN input for setting bypass state.
It could measure your battery voltage using an onboard ADC and warn you via the LED if your battery's getting low.  Maybe it blinks or something.
The onboard buffers make your projects easier to design.
Perhaps a standard header with -5V/0V/5V power, audio send/return, and a status line that lets you put your circuit into low-power mode when bypassed if applicable.

Dropping 9V to 5V and then inverting gives you +/-5V which is great for opamp effects.  If you use a low-drop regulator your circuit will work with a 9V battery that's quite flat, while still putting out 10V.  These often have reverse-voltage and over-voltage protection also. 

One problem with this is that even assuming 100% efficient voltage conversion, if you have a circuit that pulls 1mA at +/-5V, you pull about 2mA through the 5V regulator, which pulls 2mA from the battery.  So your 10mW circuit draws 18mW from the battery, 8mW of which is dissipated as heat in the 5V regulator.  :-(  Is there a better way?  Can we somehow use the uC to do voltage converting and just use the 5V reg to run the uC?

If you fit the DC apapter and input socket onto the board as well, and add a connector for the 9V battery clip, you can do all the clever input socket power switching on the wrapper board too.

Now all your projects consist of is your Wrapper board, output socket, and the main board.

I have the parts here, i just have to design it, but I'll post it when I do.  :-)

[Dave_B]

Man, I've been hung up on a "preset board" spurred by Transmogrifox's post on digital pots. 

I'm envisioning a general purpose µC board that does the following:
* stores eight presets
* has a simple binary 3-LED display for program number (only because it's easier than cutting squares in your box for 7-seg displays)
* has two momentary switches for "up" and "down"

You'd clip the wires to your pots and solder this board in between. 

Assumptions:
* you can get the correct digital pot values to replace yours
* the pot values you're using aren't somehow going to cause problems for the ADC or power supply (otherwise replace them with 10k's).

Obviously not thought through all the way, but here's where my head is at.  I'm sure more than a few people are thinking similar things.

[Quackzed]

i want a little digital board that i can mount in my guitar with 2 digital pots so i can control...
volume and tone,or.. switch to
gain and fx blend, or
pan and stereo spread
controll my delay mix and d-time
controll my monitor mix on stage
turn on my flanger and echo at the same time...
etc...alert the bar that I need beer!
start stop my digital recorder!:)
turn on/off ac powered devices( rotating speaker )(strobe)(smoke machine??!?!!)lights etc...
My intention here is to think of my "real situation" needs and wants and just brainstorm some ideas ,throw them out i guess... rather than burden you with a "chore list" of things to do...
nut it would be great to have a UNIVERSAL module consisting of 2 or 3 digital pots,1 to scroll through all the devices,1or 2 pots to controll them...
then you can add a a/c module... or adjust your pedals controls from your guitar or
scroll to your reverb and turn it up via a pot... scroll to the strobe and 3 4 STROBE!or scroll between cleen and distorted sounds for a DAAaaAAaaAAaaAAaa a  a  a
I THINK A GOOD SOLID STANDARD THAT EACH MODULE MUST CONFORM TO WOULD BE A MUST!
able to make it with readily availabe parts and clear code. make it expandable... with standardised connectors(easily available)and i think it will (must) be a huge success!
...you know, my birthday is coming up...