Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Digital VU Meter

Status
Not open for further replies.

Pepsiiuk

New Member
Hi all,

I just want to run a project idea past you.

As I understand it, VU is just an average measure of amplitude of an audio signal, correct?

In that case, why can you not have a microcontroller sample an audio signal several times and calculate the average; the result of which is translated into the usual -20dB to +3dB scale?

In my mind this would be possible. Is it? Or am I missing something?

Thanks
 
You can, and it's commonly done - although there's no particular need to calculate the average in the processor, simply use the processor to display the input. The input needs to come from a rectifier and suitable attack decay circuit, just as with any other VU.

I did this years ago (2003 actually!), using a 16F876 to create a stereo 16 LED version of the LM3914/5 - with switching for lin/log and dot/bar. To do the log scale I used a log lookup table, created using a little Delphi program.
 
Last edited:
You can, and it's commonly done - although there's no particular need to calculate the average in the processor, simply use the processor to display the input. The input needs to come from a rectifier and suitable attack decay circuit, just as with any other VU.

I did this years ago (2003 actually!), using a 16F876 to create a stereo 16 LED version of the LM3914/5 - with switching for lin/log and dot/bar. To do the log scale I used a log lookup table, created using a little Delphi program.

Thanks for the reply.

I haven’t looked into this much so what is an 'attack decay circuit' and why is it needed?

I was planning on doing a 3 band (bass, mid and treble) version using three active filters, each connected to an ADC. I haven’t really thought about the output yet, although it will probably be three arrays of Charlieplexed LEDs.

Any hints or tips?

Thanks again.
 
VU meters need fast attack and slow decay, there are specific rates used in professional equipment, the specification for which was set by the BBC.

You might try looking at:

**broken link removed**
 
Hi again.

I am really stuck on the analogue front end.

I can’t think how to convert the single +/- voltage input into the three, 0-5v signals needed, all with only a single rail supply.

The input put is from any headphone out so I assume some attenuation may be needed.
This then has to be rectified and then filtered, or do filters not work with rectified signals?
For this purpose, is it better to use passive or active filters?

I also saw the half wave peak detector (figure 1) in the LM3915 data sheet. Would this be a better option than full wave rectification?

Sorry about all the questions. I normally deal with digital systems and this is my first time designing an analogue one from scratch.

Thank for any help!
 
It all boils down to cost versus performance tradeoffs.

Full wave rectification is better, but the circuit is more complex, and a bipolar supply is required.
For non-critical apps, a half wave rectifier may be enough.

Now, if you only have +5 volts available, but you would still want a full wave rectifier, you could use an ICL7660 or similar to generate the -5volt. Then you may use the circuit shown by Nigel.

One last ward of caution....the LM1458 opamps in that circuit operate best with +/-15 volts......for a lower voltage of +/-5 volts, you would be better off with a rail to rail opamp, like the TLC2272.
 
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top