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.

Linear Vs. Log LM3914/5

Status
Not open for further replies.
Hello all, Ok, so i'm making a VU meter for a stereo system.

I just bought an LM3915 Log driver as it was reccomended to me on this forum. However, the LEDs are nowhere near responsive enough. It seems life the sound input "cant make the LEDs move" (sound input is good, though).

Only the LED on the end of the lit up string is mimicing the music beat.

Being logarithmic, the input power must double every time to light a new LED, correct?

I have the circuit working fine, and I've adjusted all the controls - it's the chip that I dont think is what I'm after.

Is the LM3914 what I need?

EDIT: I used the attached circuit.
(pic posted...)
 

Attachments

  • Image008.jpg
    Image008.jpg
    449.7 KB · Views: 570
  • audio_level_ckt.png
    audio_level_ckt.png
    10.1 KB · Views: 1,149
Last edited:
I'm not sure- the signal input is straight from my laptop headphone output - which is relitively loud, I have also tried it from other various devices. I have tried both the datasheet reccommended circuit (attached now) and the original one (attached previously). The datasheet one works best, bot not very well. (showing symptoms I described earlier.

I hope i can figure this one out- the chips are expensive from those jaycar crooks..
 

Attachments

  • audio-level-meter-1.gif
    audio-level-meter-1.gif
    15.2 KB · Views: 8,270
yes: heres what i did (its a pot)

|==- positive (12v)
|==->to signal input>
|==- GND

I then measured the voltage at the signal in pin
 
Last edited:
yes- every LED works fine and well. Did you see the youtube videos? you can see how un-responsive mine is. I powered the VU meter straight off my (cheap crappy) 1W stereo at full power. The point is, all it did was turn the higher up LEDs on, with exactly the same un-responsiveness.

Any ideas?
 
You mean this circuit here? can I substitute the 1458's with OPA opamps (Ti's audio op-amps)?

And, can I use a 12-V-12 supply instead of the 15-v-15 one listed?

Also, what do the marking near the pot mean? ("CW")

I'll give this circuit a go when I have the time, and I'll let you know what transpires.
 

Attachments

  • Untitled.jpg
    Untitled.jpg
    22.1 KB · Views: 196
maybe a peak dectector is needed

I myself am helping a boy scout design and build a graphic equalizer using the 3916 chip as well. going with 2-10 x 10 led arrays
Audio guru suggested using a full wave peak detector.Data sheet suggests as well.
Look at the data sheet for the 3915 figure 4
hopefully all the data sheets are similar but there is a schematic of this circuit.
it might help?
 
This circuit is fine and will run fine from a +/-12V supply.
 
Ok- i'll try the one from ESP now, then, if thats no good, well, i dunno if i'd be bothered to try the full wave... but we'll see.

thank you all!
 
The transistor peak detector circuit you have works well only if its input has a DC path to ground. Your lower LEDs are always lighted so your headphones output probably has a DC voltage on it.
Adding an input coupling capacitor to your circuit will fix it. Try a 1uF non-polarized capacitor.

Your 20k pot has two functions:
1) With it turned to ground then the input sensitivity to light the top LED is 1.25V peak. The LEDs will be extremely dimmed.
2) With the pot turned to max then the input sensitivity is horrible. The top LED lights with an input of about 10V peak but the LEDs will be brighter.

Remove the 20k pot and replace it with a 1k ohms resistor from pins 6 and 7 to ground. Connect pins 4 and 8 to ground. Then the sensitivity is 1.25V peak and the LEDs will be pretty bright.

You won't see any difference between half-wave and full-wave with music or speech.
 
Status
Not open for further replies.

New Articles From Microcontroller Tips

Back
Top