As sound engineers, we’re all concerned with sound pressure level, a.k.a. SPL. Whenever you measure SPL — whether you’re using a self-contained meter or using measurement software — it’s important to know how the measurement is weighted. To put it another way: What exactly is the measurement microphone listening to, and why? Let’s take a look at weighting and how it affects you.
Why Weight?
SPL measurements typically employ A- or C-weighting curves. A curve is created by adding a filter in between the measurement microphone and the meter display. A “flat” or “unweighted” measurement curve has a linear response: as frequency changes, the amplitude (in this case dB level) remains flat. In the case of an SPL meter, a flat curve means that the meter responds equally to any frequency across the range that the meter is showing (see Fig. 1).
We generally don’t use a flat curve when measuring SPL, because human hearing is non-linear — our ears don’t respond equally to all frequencies. We’re more sensitive to sounds ranging from around 500 Hz to 8 kHz, and (complicating matters) that response changes with volume level. You might recall the Fletcher-Munson curves which demonstrate how our ears are way less sensitive to low and high frequencies at low volumes (thus the “loudness” button on vintage Hi-Fi gear). As volume increases to around 100 dB, hearing response flattens out in the low-end, but it remains non-linear above around 2 kHz. The bottom line is that a flat curve doesn’t mirror our hearing response (though it can be useful for evaluating the linearity of audio gear). Flat curve SPL measurements are usually indicated by “dB SPL,” with no letter following the “dB”.
The A-weighted measurement curve attempts to mimic the response of our hearing at low volume levels. A-weighted SPL measurements — indicated by “dBA” or “dB(A)” — are filtered drastically at the low-end and not-so-drastically at the high end (see Fig. 2).
The A-weighted filter sharply rolls off the low-end below 1 kHz and gently rolls off the high end above 7 kHz. A-weighting best represents human hearing at “normal” volume levels and is often used for measuring noise. As we’ll see in a minute, this is an important point when it comes to measuring the sound level of loud music.
FYI, OSHA (the Occupational Safety and Health Administration) has established legal limits regarding noise exposure in the workplace. The permitted exposure limit or “PEL” for an 8-hour day is 90 dBA, with a 5 dBA “exchange rate” — meaning that when the noise level increases by 5 dB, the amount of exposure time should be cut in half. If we do some math, we can figure out that if your show runs 100 dBA, your exposure time should be limited to two hours to avoid hearing damage.
Just Weight and C
As mentioned earlier, the response of the human ear flattens out somewhat as volume levels increase (particularly in the low-end), and that’s where the C-weighted curve comes into play. Represented as “dBC” or “dB(C),” C-weighting simulates what happens to our hearing response when the volume is turned up. As a result, C-weighting is more sensitive than A-weighting at the extremes, but particularly so in the low-end. Fig. 3 shows the C-weighted curve. Notice that the low-frequency response of the measurement extends down into the lower octaves.
If we overlay the A and C curves, you can see how drastically different the measurements are when it comes to the low frequencies (see Fig. 4).
At 100 Hz, the A-weighted measurement is 20 dB lower than that of the C-weight measurement. This is an important distinction when it comes to measuring the SPL of music because a lot of contemporary music has a ton of energy in the low-end (inexperienced engineers mixing “kick drum plus band” notwithstanding). Given the same source material, an A-weighted measurement will be significantly lower than a C-weighted measurement.
Let’s suppose you’re working a venue where the house rules include a speed limit of 100 dB SPL. You need to know if this measurement is A- or C-weighted. If the measurement is 100 dBA, then you have a fair amount of “wiggle room” in how loud your mix will be perceived because the meter is ignoring a good portion of the low end — thus 100 dBA is fairly loud. But if you’re told that the speed limit is 100 dBC, then the kick drum and bass will contribute significantly to the SPL measurement, and your mix will be perceived to be considerably lower in volume. While we’re on the subject… anytime you are given a venue speed limit, get clarity on where that measurement will be taken. It should be made at the mix position, but you never know for sure until you ask.
Don’t Forget LEQ
Some venues may use an additional measurement known as Equivalent Continuous Noise Level, abbreviated “LEQ.” This is a long-term average measurement which can run as long as 60 minutes (or overnight for environmental noise studies), which means that brief peaks are essentially ignored. In fact, you could probably hit short peaks as high as 110 dBC while still maintaining an acceptable LEQ. Again, it’s important to know the parameters at a venue, because the rule may stipulate something like “do not exceed 100 dBA for more than 10 seconds,” which is much different from “do not exceed 100 dBA, ever.”
I clearly recall a show at the Sweden Rock Festival where I played the intro music for a show, and — not realizing I hit the gas pedal too hard — sent the large SPL display at FOH into a frenzy of flashing red, indicating that I was over the limit. As I turned the level (way) down, the systems tech laughed and informed me that it was a one-hour LEQ measurement, so I’d have nothing to worry about as long as I behaved for the rest of the show. Which I did.
Steve “Woody” La Cerra is the tour manager and front of house engineer for Blue Öyster Cult.
