LED respond to sound

[rasco22862]

Hi, i´m looking for a mini-project to add to a Ruby for example. I´m looking for circuit that makes the LED respond to the sound at the input (or output i dont know). This way, the led turns on and off responding to the sound

Thanks.

[jonathan perez]

clipping diode?

[rasco22862]

clipping diode?

i´m not sure. It´s simple, what i´m asking is for a led to turn on very shinny when the sound is high, and turn on normal, when the sound is normal, and when you are not playing, the led is off. The led respond to the input.

[MikeH]

If you look through the gallery there's a schematic for the "Micro-Drive" or something like that.  It's a simple little OD circuit that has led's that respond to input signal.  So if you play soft, they barely light up.  If you strum hard they blink brightly.  If you push a distortion or boost into it, they light up like mad.  That might put you in the right direction.

[The Tone God]

i´m not sure. It´s simple, what i´m asking is for a led to turn on very shinny when the sound is high, and turn on normal, when the sound is normal, and when you are not playing, the led is off. The led respond to the input.

So you are looking for just three settings ?

Andrew

[slacker]

Build the envelope follower part of the Nurse Quacky and hook that up to the input or output, the LED will flash as you play.

[rasco22862]

If you look through the gallery there's a schematic for the "Micro-Drive" or something like that.  It's a simple little OD circuit that has led's that respond to input signal.  So if you play soft, they barely light up.  If you strum hard they blink brightly.  If you push a distortion or boost into it, they light up like mad.  That might put you in the right direction.

this is what im looking for, but i don´t know what part of the schematic is for the overdrive, and what for the blinking leds.
and can i replace the BC547C for a 2N3904?

[rasco22862]

I mean, what part of the schematic, should i look???

[Auke Haarsma]

the blinking led *is* the OD/distortion. Unless you really want the diode to distort your sound, you should check out Slackers advice.

[rasco22862]

i dont know what is an envelope follower?? Can i just build de microdrive and not connect the output effect to the amp?¡

[GibsonGM]

Rasco,

I was going to suggest you use an opamp to boost your input by about 10 times, and feed this to a string of diodes.  I looked at the Quacky envelope section, and that will probably do the job, like Slacker said.

I think the follower section is the 100K pot to set your sensitivity, the opamp stuff connected to it up to the LED on the output.  By omitting the .01uF cap, and adjusting the 3.3M resistor and the 47K, you should be able to get a decent gain to accomplish something like the "bar graph" LED chain effect.   The gain = - R2/R1.  R2 is the 3.3M here, and R1 is the 47K.    If current is an issue, you might have to include that transistor off to the right (I don't know how much current the opamp will let you draw).
If you adjust the input pot and gain right, you should be able to stack a few LED's on the output and have them light up in sequence depending on how hard you hit your strings.  Make R2 a 1M pot, R1 about 18K to start, should get you in the right direction ;o)  You'll have to experiment.   Don't hook the output of this up to an amp! It's a 'sidechain' thing and should terminate in a ground.
Hope this helps!     

[rasco22862]

Ok, im going to try that one, thanks gibson

But can i do this : http://aronnelson.com/gallery/Marcos-Album/microdrive and connet the input to the same input for the amp, and NOT connect the output of the microdrive to the amp output. This way i only have the vissual effect. Is this right?

[MR COFFEE]

yes.

But you may need to be careful about layout and supply bypassing to keep the circuit from introducing some low-level fizz-sounding distortion in the background of your Ruby output even if it's not in the signal chain. Putting a capacitor across the LEDs might help too (1000pf to maybe 0.1 uf) Those high harmonics have a way of infecting nearly circuitry by radiation and through the power supply connections.

[BN]

Hello!

It's friday night and I didn't have anything better to do so I breadboarded the Nurse Quacky envelope follower just for fun and it seems to work pretty fine. But I can't figure out how to stack the diodes at the output in order to get them to light up in sequence as Gibson wrote... Wouldn't that need additional (active) circuitry? I'm tired and thus confused, can someone help me out?

/BN

[GibsonGM]

I'm working on it, BN.  Using a simulation program to see how to do this.  I KNOW how to do it intellectually, but never have, LOLOL.  It's a matter of getting the amplified output in the right range, and having enough current available to drive something like 4 LED's. Probably have to add a BJT. 

Always wondered how to do this, so it'll be a good project - I'm snowed in here in Maine (20+").   I'll be around this evening and let you know what comes of it.   I'm actually making progress....

[Jaicen_solo]

I'd imagine it would be as simple as having parallel strings of LED's. For example, 2 LED's in series in parallel with 3 LED's in series. They 3 LED's will only light up when the voltage has reached a pretty high threshold.

[GibsonGM]

Hmm, I'll try to model that.  Right now I've got 4 1N914's in series, with 1 LED paralleled after each.  The clincher is setting the LED resistors to turn them on where appropriate...trying to get 1st at 100mv, 2nd at 150mV,  3rd at 175mV and 4th at 200mV, as a starting point.     Of course, this has been amplified to fall somewhere in the 6 to 8V range, not a very large zone.  Still trying... 

[BN]

Ahh, of course. That's the way to do it. Why does the solution to so many problems seem so obvious when someone else tells you how to do it? 

Well, I had to take my breadboarded stuff apart before leaving school around 20:00 yesterday, but hopefully I'll try it out in a few days.

Let me know if you make any breakthrough in your simulations Gibson, and thank you all for you input. I'll report back to you when further testing has been performed.

Thanks for now!

/BN

[GibsonGM]

OK, here we go. Spent some time Googling and I found a site that will sell you a kit to put a VU meter together.  I ran a simulation on the schematic using only 1 side (it can be made 2x for stereo), and it worked )   Don't forget that "front end" part, it's the amplifier.  I used 2N3904's and 2N3906's to simulate it, don't have any BC trannies.  Raising the pot value might make it easier to use for smaller signals.

Best thing is, I've seen that design about 4 times by other companies, so it must not be anything "special" within the industry.  Now the VU/led question is solved!   Lemme know if you build it!


http://www.talkingelectronics.com/Projects/StereoVUmeter/StereoVU.html

[zachary vex]

This circuit is a LOT easier to build.

http://www.instructables.com/id/E0K3YAYGI1EP4HD4U4/