Hello everyone!
I was able to find a lot of schematics and thread about driving some LED synced to music, but i need to drive a pretty much high amount of LED's. I want to light up a wall of about 4x4 mt. Also, i need to do a complete design of this circuit (schematics, PCB, safety, heat etc.), so i'm here asking for help. This project will be used in a experimental electronic music show, so the audio signal will be given by an audio mixer. I have some electronics skills, but i have never designed something from scratch to complete build.

BASIC IDEA:

I want to divide the wall in 4 vertical rectangle. Every rectangle (4x1 mt) have 4 mt of LED bars (1 mt each). The circuit will have 4 audio input: every audio input will drive one 4x1mt rectangle (see picture).

I want to use rigid LED bar 1 mt long. I found some of these for cheap on ebay: 12V - 18W ( using 72 x 5630 SMT led, divided in 3 LED group with resistor - see picture )
I can't find any datasheet of these bars, so i'm trusting the seller ratings.

SCHEMATICS:

I found on Instructables this schematic

http://www.instructables.com/id/Sound-Reactive-LEDs-1/?ALLSTEPS

which basically uses an LM324N (two of the four op amps avaible for each channel) connected to the base of a TIP29 transistor.

I don't know if this circuit can damage the mixer or handle the amount of current that bars absobe (1,5A each) and how to connect bars. Also, i want to keep the wiring at minimum ( just four outputs from the circuit ). How would you handle this?

Thanks for the reply.

 

i need to drive ...

What does the word "drive" mean here?

How can others help you if they don't share a common understanding of "drive" with you?

 

What does the word "drive" mean here?

How can others help you if they don't share a common understanding of "drive" with you?

I don't understand the problem about the word "drive" in this context. You can substitute the word "drive" with " switch on or off the LEDs accordingly to the audio at the input". There's also a video in the link i have attached. Did i missed the point of your question?

 

TIP29 are rated for 1A.

How many LED modules will be connected in series for each of the 4 bars you indicate? What is your supply voltage?

Post a schematic or other relevant information from the instructable. I don't generally look at those posts because you have to click through many pages to get the relevant information.

Your usage of the word "drive" was clear to me.

 

Your block diagram leaves a lot to be assumed, so let me try to describe what I think you are trying to do. Correct me where I am wrong.

1. You have 4 vertical columns of LEDs.
2. Each column is made up of 4 1m long LED strips wired in parallel. Each individual strip takes 1.5A at 12V, so a total of 6A per column.
3. Each vertical strip responds in brightness to the level of audio coming from 1 of the 4 outputs from an audio mixer.
4. All of the LEDs in a vertical column respond the same way.
5. There is no frequency dependence on the response of any column (though the 4 different channels from the mixer might very well have different frequency response curves.)

Is that the gist of it?

Bob

 

You can substitute the word "drive" with " switch on or off the LEDs accordingly to the audio at the input".

But what does that mean? Should you switch on the leds when there is a signal? only when the signal is loud enough? how loud is loud enough? ....

"drive" may be clear to you but it may not be clear to others.

 

Your block diagram leaves a lot to be assumed, so let me try to describe what I think you are trying to do. Correct me where I am wrong.

1. You have 4 vertical columns of LEDs.
2. Each column is made up of 4 1m long LED strips wired in parallel. Each individual strip takes 1.5A at 12V, so a total of 6A per column.
3. Each vertical strip responds in brightness to the level of audio coming from 1 of the 4 outputs from an audio mixer.
4. All of the LEDs in a vertical column respond the same way.
5. There is no frequency dependence on the response of any column (though the 4 different channels from the mixer might very well have different frequency response curves.)

Is that the gist of it?

Bob

1. Yes.
2. I still don't know how to wire the LED bars. I suppose the only way to do that is to wire the bars in parallel. Anyway each column is made up of 4 1m long LED strips, as you said.
3. Yes.
4. Yes.
5. Yes. I don't want to differenciate the frequency response of the strips. I just want to have 4 different channels of LED so i can connect 4 different audio sources.

TIP29 are rated for 1A.

How many LED modules will be connected in series for each of the 4 bars you indicate? What is your supply voltage?

Post a schematic or other relevant information from the instructable. I don't generally look at those posts because you have to click through many pages to get the relevant information.

Your usage of the word "drive" was clear to me.

Right! Here's the original schematic: http://cdn.instructables.com/FCC/H1VD/H9G177WX/FCCH1VDH9G177WX.MEDIUM.jpg

The Instructable project uses two separates 12V 2A power supply, because is made for 2 x 5m led strips (2A each strips). I don't have any restriction about supply voltage.

Sorry if i missed some infomation in the description.

 

But what does that mean? Should you switch on the leds when there is a signal? only when the signal is loud enough? how loud is loud enough? ....

I want to make the LEDs proportionally respond in brightness to the level of the audio coming from an input, as BobTPH said. You're right, it can be confusing ( and my second explanation was wrong ). It's not some sort of threshold response.
You can watch this video to understand properly.

 

I understand. it is very cool.

 

The input signal is an ac signal. So which part of it you want to use to drive the led? its frequency? its amplitude? its loudness?

How do you want to drive the led? do you want to link the voltage applied to the led to the input signal? or the current through the led? or its brightness?

And then there is this question of how you envision that relationship to be. should it be linear? or non-linear (like logrithmic)? ...

there is a lot more to the word "drive" than you might think.

A poorly implemented "driver" would be that video: where the leds are basically on / off as the dynamic range is tiny.

If you want to drive a high power led stripe, think about a pwm generator controlled by an input signal. 555 would be one such example. Or a mcu.

 

There is probably more to it like what dannyf have pointed out.

How would you handle this?

Before testing the circuit out using high powered led stripes, why not build a small scale prototype using small LEDs and power supplies not more than 5V. Maybe or maybe not you will get satisfactory results since the instructable only showed for the simple circuit responses to rather 'beats'. Normal music we listen in the radio is more 'mixed or average' in nature that more dynamic filtering of the audio maybe is needed for the underlying beats.
If the lighting is satisfactory for your choices of audio, the relay used for these small LEDs can then further be used to switch another relay handling larger currents.

 

If the LEDs vary in brightness very fast (like about 30 Hz or faster), it will not be seen as moving with the music, more like an annoying flicker. Because of this, if you want the LEDs to respond to the beat and the average volume level, then you are after what is essentially the bass channel of a color organ. The instructables circuit is better than their usual junk because it does have a pseudo-active full-wave rectifier, but its not a good one.

The three stages you need for this are an input buffer stage and lowpass filter, a better active rectifier, and an output driver that relates music volume to LED *current*. None of this is NASA-level science, and while the LM324 is horrible for real audio processing, it is perfect for this. Figure one chip per channel.

ak

 

You can watch this video to understand properly.

that's basically an on/off control scheme. Quite simple to implement: put a resistor from the audio and apply it to a bjt's base, and put the led on the collector. If you want a little bit persistence, add a capacitor from the bjt's base to ground.

No need for opamps / etc.

 

that's basically an on/off control scheme.

No, it's not.

Solid state image sensors (bucket brigade, CMOS, whatever) do not have the persistence a human eye has. Leds turn off in nanoseconds. If they appear to fade out, no matter how quickly, that almost certainly is caused by the power and control system.

ak

 

Wonderful materials out there indeed. Sometimes I wonder how they were first discovered. The ohmeter moved when the lights was turned on? However, it is much simpler in the current case, mixers provide labels about the audio power they provide to the speakers. If its low, we know for sure the mixer is susceptible to harm by the power for the LEDs in case of a failure in the bjt used. And vice versa.

If its an on/off control, why not use optoisolators?

 

You could, and it has the added advantage of providing isolation.

 

ok, analog isolation amplifiers exist beside those digital ones. how convenient.

 

The input signal is an ac signal. So which part of it you want to use to drive the led? its frequency? its amplitude? its loudness?

How do you want to drive the led? do you want to link the voltage applied to the led to the input signal? or the current through the led? or its brightness?

And then there is this question of how you envision that relationship to be. should it be linear? or non-linear (like logrithmic)? ...

A poorly implemented "driver" would be that video: where the leds are basically on / off as the dynamic range is tiny.

I want to link the LED's brightness to the amplitude of the AC signal coming from the mixer. The ideal would be a log response ( as we percive log increase of volume as linear ), but i'm afraid this would add too much complexity to the circuit.

The three stages you need for this are an input buffer stage and lowpass filter, a better active rectifier, and an output driver that relates music volume to LED *current*. None of this is NASA-level science, and while the LM324 is horrible for real audio processing, it is perfect for this. Figure one chip per channel.

ak

I don't need any LPF, because the signal can easily EQ by the mixer. So this would be the block diagram.

For the input buffer and the rectifier no problems. I don't have any info about lin to log converter: it would be ideal adding one but if it's too complex it can be eliminated. For a more smooth on/off behavior a cap will do the trick. I 'd like to increase the dynamic range of led brightness for sure. The part i'm most worried about is the LED's driver. Surely i need another transistor because TIP29 can't handle that amout of current. How can i find a transistor for this application?

 

A power darlington, or power MOSFET, or multiple devices in parallel can handle any current you want. I think you are wrong about not needing a lowpass filter. You basically are building a visual VU meter. There are chips for this that handle the log function for you. Most are designed to run a bar graph, which is constant brightness and variable size rather than what you have, constant size and variable brightness. Again, this is not a difficult adaptation. Search for the LM3915 datasheet.

Lowpass filter is not the correct term. What you actually want is an envelope detector, something than makes an analog voltage that tracks the overall loudness of the audio signal. Again, for the changes to be visible by humans the output has to have a pretty low bandwidth.

ak

 

wow. how does it do that!