Use of color-changing LED's in pedal projects

[Strategy]

Tonight as I was attempting to debug my 2nd of two 4ms Phaseur Fleur builds-- ONLY the LED doesn't work-- my wife asked the interesting question:
"instead of alternating between light and dark for modulation, why not an LED that changes from one color to another"? ....awesome idea. A local surplus shop (since closed) had a lot of multi color LEDs and I never bought any for lack of idea how to use them. thinking to experiment.

What sort of implementation would be needed to do the color-alternating-as-modulation-indicator? Anyone have examples of multicolor LED's used for anything at all in your projects?

Strategy

[merlinb]

I once tried to put a colour changing LED into a pedal, simply as an indicator. I got annoying interference from it, like a swirling, buzzing, like from an internal oscillator.

[Rixen]

2 lead bi colour LEDs are red when passing current one way and green the other. If you connect one lead to the output of the LFO and the other to 1/2 supply or voltage divider it will change colour when voltage goes + to -

note that the LED will go off briefly as the voltage across it goes below 2 v, so it might be most effective on the square wave output, if available.

I once made a VU 'meter' that illuminated green, then changed though yellow about -6dB and red at +3dB..

Julian
www.rixenpedals.com

[kaycee]

I don't think they work on the changing current feed from an LFO, IIRC it messes with the timer that changes the colours. Might be wrong, it's easy enough to try. I've used them as indicator LEDs, sometimes you get a tick as the colour changes so you might need some power filtering.

I did use a couple inside a pedal with a window(a tagboard fuzz factory), they change out of sync giving an interesting effect. You can also use a trim with a lowered CLR and adjust it so that they glitch as they change which is fun too.

[italianguy63]

I use these in some of my builds.. and they DO create noise.  I found that you can add an electrolytic from the LED voltage to ground and it will usually eliminate the noise.  You have to play with the value, buy I recall 47uF was a reasonable value.

MC

[Rixen]

I mean't the LED's without the built in chip, like this one shown connected to a 3PDT switch:



Other bi colour LED's are two diodes with common cathode. If you run one diode off the LFO and the other off an inverted LFO voltage it will span the colours between red and green (or blue if you can find a red/blue one)

http://www.digikey.com/catalog/en/partgroup/3mm-and-5mm-bi-color-led-lamps/34242

[greaser_au]

I made s slightly  tongue-in-cheek suggestion about this sort of thing here...  http://www.diystompboxes.com/smfforum/index.php?topic=110664.msg1016974

david

[samhay]

I use 3-lead bicolour LEDs for LFO and envelope indicators.
The easist way to do it is the continually run current through e.g. the green LED and modulate the red LED current - red always seem to be brighter/mA than green when I play with them, so it doesn't work as well the other way round.

[cloudscapes]

I usually control bicolor leds with microcontrollers, thoguh I realize that isn't an option for most here.

I guess one way to do it is with the 3-lead leds. Drive one color with your regular LFO signal (buffered), and the other color with an inverted version of that (loop up opamp inverter, it's really easy). So you have two versions of the LFO signal, out of phase with each-other.

I really like red/blue bicolor leds, over the red/green ones. Harder to find, but the contrast is cooler!

[Mark Hammer]

RGB LEDs are relatively cheap these day, compared to 10 years ago.  They have 4 pins - one for each LED element in the unit, plus a common.  A colour-changing indicator entails more effort, parts, and cost than is worth it for me.  If you were going to do it, I suspect you'd want to clock a 4017 or other counter chip and, via diodes connected to each of the counter outputs, turn on combinations of RGB elements as you step through the sequence, to produce a rainbow.  I suspect someone will say it could be easily done with the right PIC, and they're probably right.  But like I say, seems like a lot of trouble for not much gain.

One effective use of multi-colour LEDs can be for modulation pedals.  So, you may want to be able to see the LFO rate visually.  The stompswitch can be used to change which colour is illuminated for the LFO indicator, such that you only have one LED on the control panel.  For example, when engaged, the LFO indicator flashes bright green, and when in bypass it flashes a lower-illumination red.  You get to see whether it's on or in bypass with a single LED and you can still adjust LFO rate to taste in bypass.

[Strategy]

Thanks for the answers, all. This will be fun to try.
the Phaseur Fleur's LED is both effect-on indicator AND modulation indicator, so not sure yet whether I'll fuss with implementing that in this particular pedal. However I was thinking about getting one or another of the 'LFO add on' daughter board projects currently available and adding this to a project, that seems like an ideal way to experiment with the modulation indicator ideas.

Thanks!!!
Strategy

[karbomusic]

A colour-changing indicator entails more effort, parts, and cost that is worth it for me.

Exactly. Additionally, the initial excitement is it is "RGB" and you can create any color you wish, which sort of falls short in practice not to mention the pain to get it there. I usually end up with one trim pot per color pin and tweaking those until I realize it's more about finding two colors vs the exact ones I wanted - and as you said, feels like more parts/count trouble that it is usually worth unless you are simply switching between R, G or B.

[merlinb]

"instead of alternating between light and dark for modulation, why not an LED that changes from one color to another"?

An LDR is more sensitive to one wavelength than any other. By changing the LED colour all you're really doing is changing the coupling between the LED and LDR, i.e. it's the same result as when you use a fixed LED but vary its brightness. OK, the variation may be more 'unusual' than with a fixed LED and a simple sine oscillator, but ultimately you could get the same effect with a fixed LED and a funny-waveform or pseudo-random oscillator. Two or more oscillators both feeding the LED could do this, and give you more freedom of speed/depth/shape control than a self-changing LED that has a fixed pattern.

[samhay]

A colour-changing indicator entails more effort, parts, and cost that is worth it for me.

Exactly....

In my example, the additional part(s) amount to 1 resistor and an additional link/wire to V+. I don't see that as being a significant deterrent, but each to their own.

The main issue I have with multi-colour LEDs is that I can't find super-bright versions.

[karbomusic]

A colour-changing indicator entails more effort, parts, and cost that is worth it for me.

Exactly....

In my example, the additional part(s) amount to 1 resistor and an additional link/wire to V+. I don't see that as being a significant deterrent, but each to their own.

The main issue I have with multi-colour LEDs is that I can't find super-bright versions.

Even when you want to choose your own two colors say orange and purple?

[samhay]

I don't know - haven't played with RGB LEDs. How do you get orange?

[midwayfair]

I used a red and green dual anode LED in the Tap Tempo Cardinal as the rate/depth indicator.

You need to make tiny adjustments to the current draw but that was easy enough in my build.

I didn't really notice any noise with it ... it's just two LEDs inside with the anodes tied together, so I'm not sure how that would be different from two discrete LEDs with a connection made on the PCB.

Agreed that it will be ze goggles with an LDR, though.

[karbomusic]

I don't know - haven't played with RGB LEDs. How do you get orange?

The proper mixture of green + red should give orange (using more red than green IIRC because equal amounts of R+G would be yellow so we dial the green back some to get orange). To get "RGB" colors we have to mix the individual LEDs at different brightness levels which results in less than satisfying results because the mixing doesn't work that well for certain combinations, especially where you might need 20% of one LED and 90% of another. That's sort of what I was getting at, if I choose RGB it's more from the idea of what RGB is normally used for, mixing colors, to do that seems to take more parts and/or trouble than it's worth. Because, if I'm just going to keep it simple by turning on red or green, it would be easier to just use a two color LED or just switch pins on an RGB but neither of those is mixing colors in the traditional RGB sense.

All that being said, I have a drawer full of RGB LEDs both clear and frosted. The original idea was to mix colors to fit the visual scheme of the pedal. For example, I have an OD with a white LED and wanted it to turn purple when the 'push' switch was on. Turned out to be too much fiddling (white isn't a very good white R+G+B and purple was a little unsatisfying) and less than ideal results to make that happen consistently across builds.

[slacker]

You could use a stripped down version of this to drive an RGB LED http://www.cgs.synth.net/modules/cgs62_sd.html this would give different colours depending on what the LFO wave was doing.
The first circuit in this thread would also produce an interesting light show http://www.diystompboxes.com/smfforum/index.php?topic=89977.msg765149#msg765149

[PEPPER!]

I came up with this as 2nd gang of a gain pot for a fuzz I built for a friend and have used it a few more times:



It sweeps blue/purple to yellow/orange and while the gradation across the sweep is far from even it's pretty consistent from build to build; ie. magenta or pale purple "sound" about the same on different builds.  I hadn't seen this before on a commercial pedal or "in the wild." I guess I made this up . . . I am using 3-in-1 SMD LEDs like the ones that are used in rope lights.