Invisible LED VU Meter

[waltk]

This was inspired by the sidewalks at Epcot Center that look like ordinary concrete - until night when they light up with different colors.  So I took a plain old mini-booster, and retrofitted it with an LED VU meter (with much help from Rob's thread here: http://www.diystompboxes.com/smfforum/index.php?topic=98118.0).  I haven't edited a video in so long that I forgot how to do it - so here's a demo - complete with crappy unedited playing.

[toneman]

It's all about the lense and the brightness of the lights behind it.

[waltk]

Good guess, but there are no lenses, and no lights behind it.

[waltk]

It's a little hard to tell in the video, but it's 5 separate LEDS.



When they're not lit, you can hardly see them ('cause they blend with the stone-texture).

[armdnrdy]

I like me the lights! 

[quad]

Hey,

Looking great. I'd like try something like that myself just for the fun of it. I do have one question though. Does one have to split the original signal into two, and what is the best way of doing it? Any loss of quality in the original signal would be... undesirable. 

[waltk]

Does one have to split the original signal into two, and what is the best way of doing it? Any loss of quality in the original signal would be... undesirable.

The input impedance is about 500K, so you don't need to do anything special to split the signal - just tap off the input jack.  It will not load your pickups or tone-suck.  In my first try (video above), I didn't even bypass it along with the pedal circuit.

[mth5044]

That's awesome!! Could be used as a bypass indicator as well.

[Jdansti]

Very cool!!!   

[deadastronaut]

nice one walt...looks cool.

this is going in my next pedal....i hooked up 9 leds to it.

https://dl.dropboxusercontent.com/u/7464107/ledvu.wmv


and yes it can be used as bypass indicator too, but i like it on all the time..

nice and simple 1 wire to audio circuit in, ground,9v...

[waltk]

Thanks for the encouraging comments all.  Nobody asked how it was done, but I'm dying to tell you... so here it is.

The LEDs are tiny 0603 SMD parts, mounted on a super thin PCB, and glued to the top of the box.

Here's the layout.  The one with 5 LEDs is used on the pedal, the other one is the 9-LED layout preferred by DeadAstronaut.



LEDs on...



The LED connections are enamelled wire that go inside the box through one small hole.



The circuit inside the box is a standard through-hole layout of the AN6884 LED driver.



So here's the LED strip glued to the box with a coat of primer on it.



Inside the box.



Then I thought, why not make the driver circuit SMD as well.



Layout closeup.



You can just glue the thin PCB right to the chip.  The pins are just bent up over the top of the PCB.



Here's the LED strip and the AN6884 circuit.  They are small enough to fit in pretty much any box.



So I obviously made a few of these while developing the design.  Anybody want one?  I'm thinking about making a kit with the two PCBs and the SMD parts already soldered on.

[quad]

With all the pedals in the chain cables are running lengthy. Nothing a little Cornish buffer won't fix. It took me a while to fit it all in and a headache in the end, but I had a lot of fun building it. Mine's not that invisible though... 

[waltk]

Quad - looks great.  Nice work.

[quad]

I'm having some noise with the meter. It's slightly crackling. Here's the design I used (Rob's):

[deadastronaut]

hi quad, crackling?.. ive never had a problem with it

what are you going into at the same time with it..?

[quad]

Hello,

A Cornish buffer:

[waltk]

It's slightly crackling.

Do you have it hooked up at the input of the Cornish buffer, or the output?
Does it crackle a little bit all the time, or just occasionally?
Did you use all the same values that Rob suggested (10 uF, 220K, 330K, 2.2k)?
Does it only happen within the first minute or two of turning it on?

I heard a slight crackle with mine once or twice, but it didn't continue to happen.  I was thinking it might have been just until the 10uF cap got charged up or something.

[quad]

It's slightly crackling.

Do you have it hooked up at the input of the Cornish buffer, or the output?
Does it crackle a little bit all the time, or just occasionally?
Did you use all the same values that Rob suggested (10 uF, 220K, 330K, 2.2k)?
Does it only happen within the first minute or two of turning it on?

I heard a slight crackle with mine once or twice, but it didn't continue to happen.  I was thinking it might have been just until the 10uF cap got charged up or something.

I have tried it both ways - IN and OUT with the same result. It starts crackling as the first LED goes out with diminishing signal strength. Yes, I used the same values as depicted in Rob's design. It stays the same (crackle) regardless of duration. There's one more thing I've notice: sometimes the first LED gets stuck (turns on and stays on for a while). It stays on along with a 'hiss' sound, then eventually shuts down.

[waltk]

Have you tried it without the buffer?  I'm thinking the buffer is making the hiss - and keeping the first LED on.  Here are some things I noticed about the circuit that might give you some ideas...

The input impedance is determined by the two input resistors in series to ground.  The datasheet shows a 10K pot forming that input voltage divider, but that would load down guitar pickups and suck tone.  I guess that's why Rob used 220K and 330K resistors to make his input voltage divider.  Are you really sure you didn't accidentally use smaller values for these resistors?  I use 100K and 1M resistors in my circuit - OR a 1M trimpot.  This lets more of the input signal through (more sensitive), and also increases the input impedance.  With a 1M input impedance, there's no need for a buffer at all.

The 10uF  cap determines how long the leds stay lit when the input signal drops to zero.  I accidentally put a 47uF cap there, and it caused the lights to stay on much longer.  They would slowly turn off one-by-one over several seconds when the signal was removed.  The overall effect is that the meter becomes sort of unresponsive.  I've been using 6.8uF or 4.7uF (tantalum) caps in my circuit, and it works fine with those values. (Maybe your 10uF cap is causing the hiss)

[quad]

I appreciate the insight. I'll do some testing right now.