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.

LM3915 VU Meter Help

Status
Not open for further replies.

rekent

New Member
I am using an LM3915 to build a VM meter for monitoring recording audio.

The data sheet shows 4 LEDs light when ≈1.2V is applied. I attached an audio signal from my laptop using a 1/8" stereo to RCA cable that was plugged into panel mount RCA jacks with wire leads coming off. When I attached my voltmeter to the leads, I got an AC that maxed out at ≈481mV. However, when I plugged the lead into my LM3915, I had 3-5 LEDs lighting with varying brightness as the music fluctuated.

How is this occurring when it is well below the data sheet specified 1.2V? Is the LM3915 actually showing the correct dB level even though the voltages vary? Additionally, the audio on my computer is maxed out, and quite loud from its speakers, but I am only getting the 480mV. How can this be the case?
 
The LM3915 is responding to peaks which are 1.414 times the rms value of a sine wave. You meter probably is reading the average or rms value.
 
The LM3915 is responding to peaks which are 1.414 times the rms value of a sine wave. You meter probably is reading the average or rms value.

If it is responding to peaks 1.4 times the true value, why at full laptop volume does it only light 4 LEDs? Is it not actually showing the correct dB value? And if it isn't, would this be solved by simply adding the rectifier circuit before Pin 5 input?
 
Last edited:
Use an oscilloscope to see the same peak values that the LM3915 is seeing. Music waveforms are extremely complex. The momentary peaks are probably a lot more than just 1.4 times the average.

Add a rectifier and moderate (1uF) capacitor on the pin 5 input. Use your multimeter at the same point.
 
If it is responding to peaks 1.4 times the true value, why at full laptop volume does it only light 4 LEDs? Is it not actually showing the correct dB value? And if it isn't, would this be solved by simply adding the rectifier circuit before Pin 5 input?

How do you know what the correct dB value is? are you basing this on a spec sheet provided by the Soundcard, was this spec into a known impedance? Does this specified impedance match that of your LM39xx circuit? You should use some sort of calibrated audio generated from your soundcard. You might look into trying this 30 day demo software, don't know how good it is, first link I found doing a search. PassMark SoundCheck - Download software to test your PC sound card

You can hardly expect to do any valid test using music or the like. Also make sure you understand all the dB specs, mainly dBm.
dB dBu dBFS dBV to volts conversion - calculator volt volts to dBu and dBV dB mW - convert dB volt convertor converter calculation online attenuation loss gain ratio reference audio engineering sound recording dBFS dBVU 0 dB audio logarithm level con
 
Last edited:
It could very well be that your volt meter was not showing the true voltage output from the laptop. You need to get an oscilliscope to measure the actual output, and use a signal that is more stable than the audio from music. Use a program where you can mak a sine wave at a specific frequency and voltage/dB level and use that as your test sample.
 
If you connect pin 8 to ground on an LM3915 then the 10th LED lights when the input has a peak voltage of only 1.25V because then pin 7 is 1.25V. You can make the pin 7 reference voltage anything from 1.25V to a voltage that is 1.5V less than the positive supply voltage by changing the ratio of the two resistors that set the reference voltage.
The datasheet shows an example of the LM3915 with the reference voltage set to 10V. Then the 4th LED lights when the peak input is 1.26V.
 
For the best performance, would you all suggest adding both the rectifier and the half wave peak detector? Placing the rectification first and then the peak detector and then that output into my LM3915?
 

Attachments

  • Screen shot 2010-&#4.jpg
    Screen shot 2010-&#4.jpg
    61.1 KB · Views: 1,091
I use a peak detector for my VU meter because it clearly shows very short duration peaks that are not seen without it. Also it keeps all LEDs at the same brightness.
The transistor/diode version needs an input of 10V peak or more to avoid a huge error.
The opamp version needs a dual-polarity supply.

I use a single-supply inverting opamp circuit for my peak detector because it works with its inputs at 0V and does not need a negative supply.
 

Attachments

  • peak detector..PNG
    peak detector..PNG
    16.2 KB · Views: 1,332
I added a simple automatic gain control circuit to my VU meter so that the range is 50dB instead of only 30dB.
Without the AGC the LM3915 has fairly accurate steps of 3dB.
 
I added a simple automatic gain control circuit to my VU meter so that the range is 50dB instead of only 30dB.
Without the AGC the LM3915 has fairly accurate steps of 3dB.

I guess what I am meaning is: Does your VU meter actually show the true dB value, or is it just a rough approximation to provide visual stimulation to accompany audio?
 
I never borrowed a high quality sound level meter to calibrate mine. I don't know the sensitivity of the electret mic I took from a broken cell phone.
It shows a pin dropped on the floor of the next room when I hold my breath, and it shows very loud music from the TV and stereo.
 
I made my Sound Level Indicator project on Strip-board, not on a pcb.
It looks like this:
 

Attachments

  • Sound Level Indicator 002.jpg
    Sound Level Indicator 002.jpg
    47.1 KB · Views: 450
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top