LED VU Circuit

The IC power connections aren't shown, but I assume the low sides are grounded. And LM339's do not like negative inputs.

Maybe the audio was processed into DC elsewhere.

John Larkin Highland Tech Glen Canyon Design Center Lunatic Fringe Electronics

Essentially yes.

The 339 is a comparator. You could think of it as a digital switch - either on or off - depending on the comparison between two levels. All of the even numbered pins on the input side in your schematic have a reference voltage established by the resistor chain. When the input signal on the odd numbered pins varies above and below the reference voltage, the op amp comparator turns on or off.

No.

It can.

Ed

Or the supply is +15 and -15 .

Ed

Also, the circuit relies on the fact that these are open collector comparators. Unlike many op-amps they tolerate a large differential voltage input without the inputs conducting. The output is either open circuit or pulled down to ground to extinguish all the leds below that point. ground. John

There is a problem with that. The comparators could have a

-15V negative supply as they tolerate up to 36V in total. However, if that were done, the leds could be reverse biased by up to 15V which would be likely to destroy them. So I think there is a rectifier and filter circuit first, especially as the VU standard requires a relatively slow time constant. John

Good point! Ed

That would put -15 across the last LED. Some LEDs will zener below that.

John Larkin Highland Tech Glen Canyon Design Center Lunatic Fringe Electronics

But they don't tolerate inputs below their power ground, which audio would be.

John Larkin Highland Tech Glen Canyon Design Center Lunatic Fringe Electronics

Much simpler to use a Microchip PIC 18F14K22 (for example), and its analog input (ADC) and its outputs driving LEDs via resitors. Singe chip solution. And more external control options possible.

Would you digitize the raw audio? You'd have to offset and clip into a unipolar uP ADC. That pic only has a 10-bit ADC.

And then there's the code.

John Larkin Highland Tech Glen Canyon Design Center Lunatic Fringe Electronics

I traced back the input signal seems to be this area

D71 is the output to the other circuit I posted.

The NP cap is prone to fail, not sure if that is critical? I replaced it, but maybe need to recheck it.

Neither circuit shows power supplies.

The 4.7 farad cap might be expensive.

John Larkin Highland Tech Glen Canyon Design Center Lunatic Fringe Electronics

Pin8 is +15 Pin4 is -15

Its a 4.7uF non polarized

The venerable LM3915 is probably a cheaper and better single chip solution. You can still get kits using it even today.

Very strange circuit.

John Larkin Highland Tech Glen Canyon Design Center Lunatic Fringe Electronics

Yes, good old VU meters..

Sure

Look for peak value, do a compare for each level you want a LED to go on Make it log if you want. You can also use the PIC's hardware comparator and drive the reference with a PWM generated DC bias, that PIC has a PWM generator.. fast! And replace the LEDs with a simple OLED. many ways to solve all this

Add a few fast diodes and go Giggle Hertz

Good point, it is absolutely essential for somebody going into electronics to at least have programmed some stuff in ASM.

Talking to chips requires I2C knowledge and knowledge about other interface protocols too. Good question for in a job application interview :-)

But simplest is what Martin suggested: LM3915.pdf

It's probably in error, but it's almost there. The VU conversion entails processing the audio analog with full-wave rectification, followed by a 300ms time constant filter. The circuit shown is half-wave, but the time constant looks about right.

As for the scaling for the bar graph display, they're going for :

Volts, RMS = 10^((10-VU)/20)

Here, RMS is taken in its most literal historical definition as being the equivalent DC.

It's supposed to convey to the viewer a sense of signal loudness (volume) to the human ear.

I did confirm that the input to the op amp array common side was a DC voltage that increased with audio amplitude when a constant signal source was injected and not AC.