simplest VU meter

Thread Starter

mike_canada

Joined Feb 21, 2020
239
I'm trying to come up with the simplest VU meter possible using the fewest parts because I want to detect different frequencies of sound as opposed to different volumes.

I already made a board of LED's with anodes individually connected to pins of a 3.3K SIP resistor so if possible, I want to use my existing LED arrangement.

So a circuit I thought of with sortof minimal components is circled in red. and if I'm not mistaken, C1, and R5 form a high-pass filter. Which means I might be able to get away with copying this arrangement (and modifying C1 and/or R5) for each led but that involves drilling 12 holes per LED and since I have 10 leds, that's 120 holes.

I want it so that when no sound is present, all LEDs are off.

Is there a simpler circuit I can use for a volume meter (preferably using only one transistor, a capacitor, 2 resistors and an LED)

vu.png
 

Audioguru again

Joined Oct 21, 2019
6,692
When the amplifier output is loud at +6V peak then the signal through C1 will completely destroy the base-emitter junction of Q5.
Then C1 charges to +5V and the negative swing signal at the base of Q5 goes to -11V that seriously damages it since its max allowed reverse base voltage is -6V.

To fix it you need to attenuate the signal that feeds C1.
But since Q5 has no base bias voltage then it does nothing with signals less than 0.65V peak. If 6V peak is loud then signals less than 0.3V are common.

This CR filter is a very simple highpass with a very gradual slope. It gradually reduces the level of frequencies below 2.9Hz (earthquake sounds) but passes all audio frequencies.

Look in Google for Color Organ Circuits from years ago. Here is one but its bass and treble parts are swapped:
 

Attachments

Last edited:

Thread Starter

mike_canada

Joined Feb 21, 2020
239
ok so it seems if I crank the input volume too loud, the voltage measured from C1 to Q5 can sometimes go lower than -6, but at normal level (0.47V amplitude with 0V offset), the voltage doesn't go past -2 as shown.

I am ok with the LED going on when the volume is loud. If the volume is too quiet then I'd rather have the LED off.

I also added another capacitor and through experimentation it seems that R5+C1 is a high-pass filter and R3+C4 is a low-pass filter which means I have made myself a band-pass filter with lots of parts.

I did add an attenuation resistor R6 but I noticed if the value is too high, the LED would glow dim at best.
 

Attachments

Audioguru again

Joined Oct 21, 2019
6,692
Sorry, I was wrong. R5 turns on Q5 almost all the time and C1 turns it on with much more (too much) base voltage and base current to destroy it.
 
Last edited:

Ramussons

Joined May 3, 2013
1,409
I'm trying to come up with the simplest VU meter possible using the fewest parts because I want to detect different frequencies of sound as opposed to different volumes.

I already made a board of LED's with anodes individually connected to pins of a 3.3K SIP resistor so if possible, I want to use my existing LED arrangement.

So a circuit I thought of with sortof minimal components is circled in red. and if I'm not mistaken, C1, and R5 form a high-pass filter. Which means I might be able to get away with copying this arrangement (and modifying C1 and/or R5) for each led but that involves drilling 12 holes per LED and since I have 10 leds, that's 120 holes.

I want it so that when no sound is present, all LEDs are off.

Is there a simpler circuit I can use for a volume meter (preferably using only one transistor, a capacitor, 2 resistors and an LED)

View attachment 276122
The simplest will call for a series of Band Pass filters for selecting the frequency band(s) of interest. You will need to design Active Filters, keeping in mind the Cutoff Frequencies and the "order" of the filter.
Maybe the same can be performed by Sound Processing software in a good processor.
 

Audioguru again

Joined Oct 21, 2019
6,692
Your first-order RC filters are so simple that the bandpass includes all audio frequencies when I reduced the capacitor values so that it peaks at 850Hz.
I show an active Butterworth bandpass filter using four RC parts that has sides that are twice as steep as yours. Third-order and fourth-order filters have steeper sides.
 

Attachments

Top