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.

40 led vu meter

Status
Not open for further replies.
The LM3914 has linear steps, so tiny volume changes at the high volume end would change many LEDs. At the low volume end, very large volume changes are necessary to change only 1 or 2 LEDs.

The LM3915 is logarithmic and the LEDs change with steps of 3dB. Therefore with a range of 60dB, from very loud to nearly the background noise level, only 2 IC's are needed with 20 LEDs. The circuit is in the datasheet. I built it and it works great. :lol:
 
Is there (was there) not also an LM3916 or some such variant that was actually a VU version of the series? I'm sure I've seen those back in the 1980s and maybe had a few that I never actually used.

Dean
 
I Believe I have a couple of 20 LED boards, Using the LM3915's, Completely assembled. I built them, Tested them, but Never went any farther. They Consisted of a Couple of other IC's as well and were also dot or bar.

Could probably give them to you for free.
Could also send you a picture of them.

If your interested, send me an email and I'll have a look for them.
and check them out.

I Think My Electronics Days are Coming to an End and My Best Friend/Parts supplier Died this morning. Sad day for me.
 
hy sorry for reply so late.

I have builded a 40 led's vu-meter based on the schematic on this website

It's not with LM3915 or with other version of display driver from NSC.
The entire circuit is build around LM339 comparators.I'ts a little bit complicate to build because is a lot components & the control board + display LCD can reach substancial dimension it but it worth the effort because it's cheap,it's 40 LED's per channel and is logaritmic
The signal response is very acurate and the vu meter speed is also good.

I have build this circuit & it works 100%

All the tehnical here:

http://www.tekniikka.oamk.fi/~archy/vu-meter/index.html

Again sorry my late reply but i wasn't active on this forum since march 2005.Now I'm back
If u have any question i'll be glad to answer them

Hope it help & good luck

Best Regards, Adrian
 
Hi Adrian,
You designed a very complicated circuit. I don't know why you selected only 1dB steps when nobody can hear such a small change in level. I have a 3dB-steps meter made with two cascaded LM3915 ICs that accurately shows on its 20 LEDs a range of levels much greater than yours.

On your video, frequently the peaks aren't seen. Then the peaks are immediately followed by a reduction in overall level which is odd. For most of the video I didn't see the peaks, saw the reductions and thought the displays were upside-down.

I am used to seeing a VU-meter that holds the peaks long enough to see them, about 40ms. Then an RC discharge time fast enough so that 50Hz is still accurate.

It would be good to make a VU-meter like those on stereos. One LED holds the peak for a while and the other LEDs are instantaneous. :lol:
 
audioguru said:
It would be good to make a VU-meter like those on stereos. One LED holds the peak for a while and the other LEDs are instantaneous. :lol:

It's called a PPM (Peak Program Meter), or at least a proper one is, the specs were designed by the BBC a great many years ago on analogue meters (and copied widely ever since).
 
Hi Nigel,
When I was young, I visited the BBC engineering lab and learned a few tricks. :lol:

Do you also think this LED meter appears to operate "upside-down" with a reduction immediately following a peak? :?:
 
audioguru said:
Hi Nigel,
When I was young, I visited the BBC engineering lab and learned a few tricks. :lol:

Do you also think this LED meter appears to operate "upside-down" with a reduction immediately following a peak? :?:

I don't quite know what you mean?, if you're holding the peak (so you have time to see it), then the only way it can go from there is down.

With an analogue meter you don't see the signal drop rapidly away (because the peak holds it), but with an LED display it's usual to not only hold the peak, but also to show the lower LED's falling away as well.
 
Nigel Goodwin said:
I don't quite know what you mean?
I played the video and sounds again and saw it again.
In the beginning and near the end when the entire "orchestra" is playing, the overall level appears to drop about 8 LEDs for each beat. I don't see the instantaneous peaks, maybe because they are so fast, or maybe because the circuit messes-up and reduces the level of the display when it shouldn't during each beat.
In the middle of the video during the orphan drum-beats (without any other sounds) then the display appears to operate normally showing very fast rise and fall swings including the duration of the peaks.

You know what?! The faulty "compressor" in the DSP of my old car radio behaved like that. It over-compressed for each beat of the music. It drove me nuts because it messed-up the sound, not just the display.
It messed-up only the sounds of its FM radio, its CD player sounded perfect. I nearly called my favourite FM station to complain about their "over-compression" then I heard the station playing perfectly on my sound system at home. :lol:

You'll know what I mean when you see my sketch:
 

Attachments

  • posted_led_vu_meter.png
    posted_led_vu_meter.png
    10.7 KB · Views: 2,018
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top