Banner for Audio Level Meter article

Aren’t those blue level meters that you see on McIntosh amplifiers really cool? This article shows you how to build level meters for your system which do far more than even theirs do. The Audio Level Meter project gives you:

Pic of McIntosh MacMC-12Kb amplifier with blue level meter
  • 100dB Logarithmic decibel display so it can show the full audio dynamic range
  • Linear display to show music dynamics similar to a VU meter
  • Fast peak detect mode to identify when an amp will be clipping.

The huge dynamic range of its log decibel scale shows you everything from 500 watts RMS, down to less than one millivolt. It’s so sensitive that, when your amp is off, it shows deflections from conversation in the room as your speakers work like microphones! It shows the whole range, while maintaining a resolution of one dB.

Use peak detect mode to accurately detect clipping of even the shortest peaks found in music. Averaging mode lets you measure continuous power and noise.

The variable decay time control lets you set the meter to capture transients for accurate peak readings or to respond rapidly to show music dynamics.

Variable illumination lets you adjust brightness for varying residential lighting conditions (such as TV viewing).

The supplied files package gives you all of the graphics files needed to print excellent artwork for the PCB, front panel, meter scale and more. We’ve even included the original CorelDraw source files, so you can customize the project, if you like. Get the full article, schematic and files here (free!):

ALM icon

Audio Level Meter article (3MB)

Front view of audio level meter
Custom scale for audio level meter
Inside, behind front panel of audio level meter

Front view of the Audio Level Meter

Custom meter scale

Inside, behind the front panel


Reader Comments

Posted by Terry April 20, 2017 - 10:54 am
Thanks Steve.
My initial decision to use a line level input was because my pre-amp has both balanced and unbalanced outputs. I am using the balanced outputs to drive the power amp, so the unbalanced outputs are available. I thought this is a good way to drive the meter panel. To make it as flexible as possible, I am adding input buffers for either balanced or unbalanced inputs. A couple of additional gain stages are added. One allows scaling the input based on the amplifier gain. This ensures that the expected level at Z2 is present regardless of the gain of the amplifier used. I can select the power amplifier gain in 1 dB increments from about 26dB to 40 dB.
Another change I am making is to the switches. With two meter circuits in one box, the switches start to require too many decks. So I am using additional 4053 switches to take the place of the switch contacts in each of the two circuits. This lets me use a single deck switch (with some diode logic) to control all of the 4053 switches. An additional gain stage at the input selects the meter range (4W, 40W, 400W).
I am considering also adding the option to use the speaker level inputs like you did. My amp uses differential (bridged) outputs (both the high and low side of the speaker lines are driven), so I would need to buffer the input and convert it to a single ended signal. If I can think of a clever way to handle the gain switching, I may add it as an optional configuration. Placing both meters in a single housing means I have plenty of room for circuitry. I will use a full 17 inch panel.
I purchased a couple of IM-17 units to use the panel meters from, but then I came across some interesting looking oval meters. They are surplus removed from Church of Scientology E-meters.
Thanks for a great starting point for the project. The log amp and the extra dynamic range due to the auto gain select are great.


Posted by Steve L. April 12, 2017 - 06:28 pm
Hi Terry,
Thanks for the nice comments. I certainly don't mind your using the info as you wish. All I would ask is for proper attribution where due. That's an interesting application, operating the level meter at line level. Something would need to be done about getting the 50dB gain increase for the low range in log mode.

The existing design has 20dB of attenuation present in the high range for log mode. That accommodates the maximum input of 44.7Vrms (500Wrms into 4ohms). To get a 50dB increase in input gain for the low range, the 20dB attenuator is removed and 30dB of gain is inserted. You might want to operate high range without the attenuator and have the inserted amplifier provide 50dB gain but that's a lot of gain for a single opamp. The OPA2134 opamps I used have a gain-bandwidth of 8MHz. Operated at a gain of 316 (50dB), the 3dB bandwidth of the preamp will be about 25kHz, which is a bit short of ideal but by no means out of the question. Using two, 25dB gain stages would be better though.

Also, the existing design in linear mode expects 4Vrms max at connection Z2, so you would probably want more gain to reduce that. Changing U1B to provide gain could do it. One needs to be careful about DC offset there though, as the Avg/Peak detector which follows must handle low DC voltages with reasonable accuracy in log mode. The minimum DC level fed to the log amp is 11.8mV. So if U1B is given gain, one might want to include AC-coupling. At 25C, the OPA2134 offset spec is 2mV, which is pretty good but one wouldn't want a bunch more.

One issue with AC coupling in the front end is that it can skew peak detection when music signals are present. That's because large, low-frequency transients can leave charge on the coupling caps. Also, low frequency phase shift can upset peak readings. I tried to minimize that by giving the AC couplings one-second time constants (the longest I felt was usable). This corresponds to a 3dB frequency of 0.16Hz, much lower than one would imagine necessary for the instrument's functions.

You can find more info about signal levels on p.10 of the article and I will be happy to help if you have any questions or problems.
----Best wishes for your project!

Posted by Terry April 12, 2017 - 03:25 pm
Great project. I'm really impressed by the documentation. Thanks for taking the time. I am contemplating making a stereo version that connects to the amplifier inputs (pre-amp out), so the gain of the amp will have to be accounted for. But I love the auto ranging idea and will 'steal' that (I hope you don't mind). Thanks for sharing some great ideas.

Posted by Steve L. April 11, 2012 - 07:12 am
Hi Dennis, I really appreciate your kind comments. All the best, Steve

Posted by Dennis April 11, 2012 - 12:16 am
Hi Steve,

Just came across your site while searching for manufacturers of custom VU meters & least to say was blown out with your work on the audio meter, and would like to say, thank you very much for shareing it.
It really goes beyond anything I've seen so far in term of features.

Again, Thanx.

Posted by Steve L. March 02, 2010 - 02:03 pm
For the poster regarding the electrolytic caps (Name?): Actually, two back-to-back polarized caps are the same thing as a so-called nonpolar electrolytic cap, so the circuit is okay as it is. Thank you for your interest, though.

1 - 2 - next> - last>>

Add your comments here...