Buffer Audio with an Op-Amp

[menimitz]

I am currently working on a circuit for a light show holiday decoration.  I am trying to get it done with some of the components I have already purchased for other projects but am open o getting some new components if I need to.

The circuit takes an audio signal from any headphone jack source and using a graphic equalizer display IC (MSGEQ7) and an arduino compatible micro-controller drives many different colored LED's to the audio signal. 

My problem is I also want to hear the audio and be able to use different sources for audio (cell phone, laptop, etc).  So I planned to add a speaker and audio amp to the circuit as well.  The problem came when I connected the audio amp AND the msgeq7 to the audio signal from my laptop.  The audio amp made the signal level drop to a point that the msgeq7 was no longer making the light show I desired. 

My question is, can I buffer the signal coming directly from my laptop/cell phone using a unity gain op-amp?  Are there any IC's suited for this job? I have spent the last few hours looking with no success.  I would like to eventually get a PCB ordered so would prefer to use something with a single +5V supply to match the rest of my circuit.

Whenever I get this project working I'll probably take a video and post it on here. 

[kizzap]

TL07x series are quite popular for audio, as are the NE5532 series.

Also yes you can use a buffer on the incoming signal.

-kizzap

[c4757p]

TL07x series are quite popular for audio, as are the NE5532 series.

would prefer to use something with a single +5V supply to match the rest of my circuit.

These requirements aren't quite compatible...

How about RC4580? It goes within about a volt of each rail and is specified down to +/- 2V, so on a single 5V rail you could get an output from 1 to 4V. It's about $1 in single quantity, so not incredibly cheap ($0.50 each amplifier, it's a dual op amp), but has decent characteristics for audio (5-9 nV/rtHz noise, 0.0005% THD, 50mA max output current).

(THD really needs to be specified in dB, I hate counting zeros...)

[CSmith]

TL07x series are quite popular for audio, as are the NE5532 series.

would prefer to use something with a single +5V supply to match the rest of my circuit.

These requirements aren't quite compatible...

How about RC4580? It goes within about a volt of each rail and is specified down to +/- 2V, so on a single 5V rail you could get an output from 1 to 4V. It's about $1 in single quantity, so not incredibly cheap ($0.50 each amplifier, it's a dual op amp), but has decent characteristics for audio (5-9 nV/rtHz noise, 0.0005% THD, 50mA max output current).

(THD really needs to be specified in dB, I hate counting zeros...)

Yes, but they want you to fully appreciate all of those zeros once you're done.

[kizzap]

would prefer to use something with a single +5V supply to match the rest of my circuit.

These requirements aren't quite compatible...

(THD really needs to be specified in dB, I hate counting zeros...)

AFAIK most audio is line level which barely passes +/-1V. If the OP biases the op-amp to mid rail, where lies the problem? OP will have to bias it either way...

[c4757p]

IIRC TL072 and NE5532 have very narrow output swing at such a low voltage. Might not even reach 2Vpp. I could be wrong though. Neither part's datasheet specifies output swing that low, and if you extend the line out, it looks like it barely reaches. Might only work on some units and not others.

[AndyC_772]

Look up "summing amplifier" if you want two inputs to both be able to drive the lights at the same time.

My favourite low-cost op-amp for this type of application is the LMV358, which is a basic op-amp a bit like the LM358 but designed specifically for low voltage operation. The LMV324 is the quad version.

[c4757p]

It does have a good bit of crossover distortion though.

[AndyC_772]

Data sheet specifically states "no crossover distortion":
http://www.ti.com/lit/ds/symlink/lmv358.pdf

It's a much better device than the LM358, which I completely agree, is horrible in that respect.

[c4757p]

  Shame on me for assuming it'd be similar. And shame on them for using such a similar part number!!

It really isn't like the LM358 at all! They really must have just picked that number to make it feel "familiar"... Rail to rail, low voltage, diode clamp to VCC (LM358 can take pretty high voltage on its inputs even when powered down), smaller offset voltage, smaller bias current, worse CMRR, worse PSRR. The only way it's like the LM358 is that it has the same UGBW.....

[dannyf]

The problem came when I connected the audio amp AND the msgeq7 to the audio signal from my laptop.

sounds like the msgeq7's signal is being shorted by the audio amp when the two of them are connected at the same time. Assuming that you want to hear them at the same time, the simplest approach would be to insert a resistor on the input from the audio amp (to artificially increase its output impedance).

[menimitz]

I have the circuit hooked up again and now it seems to be working as intended.  I will need to post a schematic showing what exactly I have done.  I have not been able to look through all the components suggested (I only get about a half hour to work on this every few days, I made the original post from work on my smart phone during lunch).  I will hopefully get a final prototype built this weekend.

Another strange issue I have seen is when I disconnect my audio cable and play with the 5k pot I am using for volume control on the lm386, I seem to pick up local AM 970.  When it first happened I couldn't make out the noise but the more I tuned the pot the better the reception was.  What can cause this?

Also when I get this done, I want to post the project on here so others can check it out.  Would it be more appropriate to post it on this thread or create a new one with a more accurate subject and title?