Music VU bar with 160 led diodes.

[spikker5]

Hello there.

I was woundering if anyone had any schematics laying around for this project. Its my first electrial project so i wanted some help

I am going to have 160 leds on this VU bar:
90 green led's
45 yellow led's
and 25 red led's

I know the basics in circuits, and any help at all help appreciated

 

[canadaelk]

If this is your first project in electrical/electronics may I suggest (cindly) you start with something less ambitious. 10 LEDs driven by a LM3915? E

 

[audioguru]

I made an LED VU-meter with 10 steps but used 20 LEDs.
Each step is 3dB which is barely audible and the total range is 30dB. I extended the range another 20dB by adding an AGC circuit.

160 LEDs is ridiculous. If each step is 3dB then the range will be from no sound to a large meteor hitting the microphone.

60dB is a very low sound level and 120dB is a very high sound level. The range is 60dB and can be done with two LM3915 ICs and 20 LEDs.

 

[ronsimpson]

Most VU meters have 30db range at the most.
I do have some with 60db range but they are hard to find.

Broadcast FM audio, by FCC regulation, only needs to have a signal to noise ratio of 60db. If the loudest sound is 0db then the noise level needs to be under -60db. (from mic to antenna) I just looked up the specifications on a FM transmitter I know of and it is 70db.

On stage a open mic has a noise level of -30 (maybe -40db). So a VU meter with more than 30db does not really give more information.

 

[tcmtech]

Going by 1/8 dB steps or do you just want a big honkin strip of LED's that dance to the music?

 

[audioguru]

Going by 1/8 dB steps or do you just want a big honkin strip of LED's that dance to the music?

Since most modern music is terribly compressed then only a few LEDs will honk and dance but most LEDs will be continuously lighted.

 

[Mr RB]

Agreed! Looking at the spectrum analyser bars on Winamp shows exactly that, when comparing MP3s of modern music compared to classic 70's music etc. It's a shame!

 

[ronsimpson]

Agreed! Looking at the spectrum analyser bars on Winamp shows exactly that, when comparing MP3s of modern music compared to classic 70's music etc. It's a shame!

Music is compressed (in part) because people want it that way. In broadcast and digital recordings it sounds better (louder) that way. With a louder sound the noise level seems lower. Money sounds very loud and money drives most of the audio industry.

Here is where the AudioGuru disagrees.

 

[audioguru]

Listeners do not want the music to sound louder. The music producers and radio stations have a contest to see who makes it sound louder so they use compression.
The noise level from a modern recording or modern receiver is extremely low and does not matter for most of the applications.

If you are very far from an FM radio station then compressed music will have less noise than without compression.

 

[alec_t]

Listeners do not want the music to sound louder.

You and I, maybe, AG. Depends which generation you belong to . Most of the younger generation seem to equate good = loud. Unfortunately, loud > deafness > "louder please" > greater deafness......

 

[ronsimpson]

While <1db louder 'can not be heard', my experiments (at PBS) show that even 0.5db louder sounds 'slightly better' not 'louder'.

Classical has great dynamic range. Perhaps >50db. Getting that through a 60db transmitter and back through a 70db receiver and fighting background noise due to a week signal causes my phone to ring. (back when I worked broadcast) "I can't hear the quite parts." My goal at the classical station was to set on the peaks and boost the quite parts to a point where no one complained about compression and few complained about the quite parts. We told every one there was no audio compression!

Car radio music: The background noise in a car is about 30db down. Non classical audio (radio or CD or tape) should not have anything quite than -30db. I limit and usually compress hard for 'car audio'.

My last CD, the band wanted 'no audio compression'. Against my better judgement I set the loudest part of each song to 100% on the CD. You had to turn the volume all the way up to hear. Switching back to radio or another CD caused a very loud jump in volume. The noise level was real bad. I only made 10 copies. The band rejected them.

Next I, by hand, edited the loud parts and made new CDs that were 10db louder. (no compression) That was better but still rejected.

Next I used the limiter with no compression and got another 10db louder. That was better but still rejected.

At the end we went to press with limiting and less than usual compression plus I expanded the noise down.
...I set on the peaks by 6 to 10db.
...Reduced the dynamic range a little.
...If there was no signal I reduced the gain by 20db to hide noise.
I charged them for "mastering" the CD "with no audio processing". LOL
They are happy.

I used a three band compressor/limiter/expander. Each band (high/center/low) gets processed separately then combined and processed again. By separating the bands, symbol crashes and base drum do not effect the mid band vocals.

I know all this adds distortion. My only goal is to have happy listeners. The loud should not break the speakers (or your ears). The quite needs to be above the noise level by a good amount. The how I do that should be undetectable.

Most of us listen to hard limited, heavy compressed music and speech to the point where uncompressed music does not sound right. (TV, Radio, CD, tape, vinal..... all pressed even if we say it is not) You can get classic with light limiting. With out limiting the loud parts just distorts.

 

[JimB]

Ron,
Interesting insight.
Thank you

JimB

 

[audioguru]

My local TV station turns up the audio gain way too high for the news anchors then compresses it down to a normal level except the compressor is very slow.
The first breath (followed by speech) blasts extremely loud then only muffled vowels are heard. The response might be flat to about 1kHz because most high frequencies are gone.
Then a reporter in the field speaks but does not have the severe compression nor filtering and sounds normal. Commercials also sound normal.

 

[spikker5]

tcmtech Correct

Realy i just want somthing big that bounces
and i wanna use all the leds, so how many ohm do i need on my resistors?

 

[audioguru]

The datasheets for your LEDs show their range of forward voltage. If you take the worst case (maximum current at the lowest voltage drop) then use Ohm's Law to calculate the resistors.
But if you bought cheap Chinese LEDs from E-Bay then you must GUESS and probably burn out some LEDs.

 

[tcmtech]

Realy i just want somthing big that bounces
and i wanna use all the leds, so how many ohm do i need on my resistors?

For that you may looking to just set them up on a bank of comparator IC's that light up a LED for a much smaller voltage change like .05 volts on a 0 - 8 volt input scale rather than having it work as a true VU type meter.

As far as what resistance you need most likely each LED will need its own resistor and that going to have to be calculated by the individual LED's forward voltage drop rating and what voltage you are running the system on.

 

[spikker5]

So i can't realy have a 12v supply and then use the aux to activate the leds with music?

 

[Mr RB]

NO! It needs a TON of parts, or a microcontroller circuit with still a LOT of parts and a heap of clever programming code!

You might be better off to buy one from ebay, here's one ready-made for $50;
**broken link removed**

**broken link removed**

 

[spikker5]

Yeah i know that Mr RB, but i was woundering if i can use a computer psu for 12v and then an aux for signal

 

[audioguru]

White, blue and new very bright green LEDs are about 3.5V. Then you can use a 12V supply if you connect three LEDs in series and in series with a current-limiter.