Plausible solution for a programmable pedal (analog, but digital control)?

[stormbringer]

Hey.

Preparing for a few gigs and the very late album release, i've noticed that even though i'm down with a 4 button pedal i've built (distortion, gain boost, delay and volume boost), there's just too much stomping. I would like to mod my pedal to be programmable with presets instead. using an mcu, display and momentary switches instead to switch on/off multiple effects, but also change tone, delay and volume.

I can get the digital part off easily, and i already got the analog part. Now interfacing them in a convenient way is where i've hit the brick wall.

I know about digital pots, but i've also heard lots of arguments against them. i dont really need insane resolution, but it wouldnt hurt to get atleast a few possible variations, which kinda rules out the string resistor solution i experimented with, as i need 1 resistor per setting. i also found a forum thread about using LDR's, the way mesa boogie does in the triaxis amp.

But i'm really not sure what route to go here, something that wont enlarge the casing by a factor of 10 would be nice.

i did connect a jfet shunt to GND with a R2R ladder to set the level for the gain and volume, and well, it seems to work. But not tested enough yet. The tone stack though is where i'm entirely lost. Any suggestions are welcome

[Roly]

You just need more feet.   :lmao:



I'm quite partial to the DAC0800 myself.  This is actually an 8-bit current source/sink; add an op-amp and you get voltage, but (and I haven't tried this) it looks to me that if you injected a signal into the ref point it would act as a digital volume control to the output.  They normally require bi-polar supplies but you may be able to busk it somehow, and they are cheep-cheep and common as dirt.

Getting an isolated digitally variable resistance for tone controls could be a bit more interesting, but there are digitally controlled vol/treb/bass chips around.  I haven't used any, and they seem to be ghetto-blaster quality, but they still might be worth a look.

A DAC0800 driving a green LED into a LDR might be a simple way to go, and the DAC0800 has I and NOT-I outputs making two LED's and LDR's a possible way to make a digitally settable "pot" (say for EQ).

It's a bit runcible, but there are motor-driven pots around for remote control (but no position feedback AFAIK).  Model control servo + pot?

I assume you are thinking of something like one of the Arduino's as the brains.

[stormbringer]

Thanks Roly, great answer as always

I will look into the LDR solution, it seems to be the most viable solution to achieve a somewhat wide range of variable resistance.

I got the most experince with Pic Mcu's, and got loads of them laying around here, so i will probably use one of those. I did design a "multi effect control" way back that had a bunch of send/returns to patch in different pedals in presets based on a pic with a 4*20 alphanumeric display. I never built it, but had it running in sim, i might base everything off that one.

I was also experimenting a little today with running the signal through 3 parallel high pass/low pass filters, and then attenuate each one separately and mixing them back together. But i dont know. felt a bit akward, and not sure if it would even work.

[stormbringer]

Alright, drawing stuff here. Noticed i had some 100k digital pots, so will try them out while i wait for my LDR order to deliver.

If i want to "emulate" a gain pot for example that is 1M, but the digital pots and ldrs are only up to 100k, and jfets even lower. the pot is located between 2 gain stages, 1 to GND, 2 to stage 2, 3 to stage 1.

The only solution i came up with (and i'm not even sure that it's a usable one) was to connect a buffer in place of 2 with a 1M resistor to GND, and another in place of 3 with a 100k attenuator after it, in order not to mess up the impedance of the circuit. seems a bit clumsy... How would you guys that actually know what you're doing go about this?

[J M Fahey]

An LDR can go to several Megohms, just keep it *really* in the dark.
You can excite it with a current controlled LED, as within an Op Amp feedback loop.
You'll need a Log current source for a Log response.
So 0 ohms (really a few tens of ohms) to 1M is doable.
If you want to emulate a regular pot and not just a variable resistor, you'll need 2 LDRs, one for each half of the pot.

[stormbringer]

Thanx JM! i totally misinterpreted the data provided for them, had another look after reading your post and realized how it should be read instead.

Went ahead and ordered 80 pieces of 4 different values.

The log current source, i think will be easiest for this build to be generated in software for this project. but will take a look.

[J M Fahey]

4 different values

I buy them in bulk (100 or 200) and just order "5mm LDRs"  to match my 5mm LEDs.
Just make sure you nip the rounded tip of the LED , tape it to the LDR (I use paper painter's tape) and seal the whole shebang in heat shring tube.
I heat it with a Bic lighter and press the ends with flat nose pliers while still hot, so I completely seal them.

[Roly]

Thanks for the Hi-Tec manufacturing tip JM.   :lmao:   :dbtu:

BTW, what colour LED's do you use with your LDR's?  I see that CdS has it's peak sensitivity in the green, but people have been using them in "roaches" with neons since Noah was gigging, so have you found any significant difference?

[J M Fahey]

I made exactly the same mistake as you  :lmao:
I also read the datasheet and specifically used green LEDs (contary to what everybody does: red), with not that good results.
I looked for stronger and stronger green LEDs, paying quite a bit extra $$$.
Speaking about this with an Engineer at ARROW, he told me that he had personally measured the same I was finding and finally came to his personal conclusion, which I quote:

*Human eye* also peaks in Green, so a Green Led does not have to put out that many photons to be seen.
Compared to it, our sensitivity to Red (specially deep Red, bordering Infrared) is *much* poorer, so Red Leds are "doped to death" to produce Tons of Photons.
Even if CDS cells are *somewhat* less sensitive to Red, that is more than compensated by the flood they receive .
So in practice, any Red Led we *see* as "bright" is flooding the LDR and making it react by brute force.

Now I use "High Intensity Diffuse" Red Leds which are still inexpensive but work very well.
I *always* used these, but now are shown as obsolete: http://www.mouser.com/ProductDetail/Kingbright/L53SRD-F/?qs=sGAEpiMZZMs4quMj8r4lmsxifWSRNhRYAsWO8BwY0%252bY%3d
700MCd (maximum for a much more expensive one is 950 MCd, so here we have 75% the power for 5% the cost )
A "modern" one should be:
http://www.mouser.com/ProductDetail/Kingbright/WP7113SRD-E/?qs=sGAEpiMZZMs4quMj8r4lmiytw%252bIhznruipCf1oABWag%3d (9 cents) , how's that?
Cheap generic Leds don't cut it, but this one offers maximum bang for the buck.
Buy 50 or 100 and use them for *everything*.

[stormbringer]

Thanx for your tips guys, received my LDR's yesterday, so will start experimenting.

[Roly]

Thanks very much for that JM, it happens to be very topical here at the moment.  Certainly a zillion amps used a neon very effectively, and not much green in that light.  As you guessed I was going by the published response curves of CdS LDR's and hadn't thought of the green LED/eye sensitivity trap you found.  You have saved me from a lot of unnecessary fooling around.   :dbtu:

[J M Fahey]

Thanks but I didn't discover it.
To give due credit, it was my friend Miguel Maciel, EE first at Arrow (Argentina and Brazil) , then at Avnet .

[Roly]

Indeed, credit where credit is due*, a big hat tip to Miguel for working it out, and thanks to you JM for passing it on just before I wasted a whole bunch of time chasing my tail.   :dbtu:

{* just had a local web site lift a whole bunch of text, images, and circuits wholesale from AVA without a word of credit or attribution, and has so far ignored my e-mails.  A number of my contributors have their noses out of joint.  I've only asked for suitable attribution but if he continues to ignore me the only apparent option is to have the site taken down.  If you look at AVA you will see that I try to credit all sources, which may be why I haven't had a single complaint in eight or nine years.  Credit is the difference between flattery and plagiarism.}