Study Hall

Supported By

Spec Wars: Looking Inside Loudspeaker SPL Specifications

While it might be possible to attain stated max levels from that box, it'll probably only happen once... and it might catch fire.

Revision Of Example System

We’ll set aside the problems that occur at low frequencies that drag the specifications down and focus on getting our cabinet to beat the competition (at least, in theory…). It’s possible to predict where in the frequency range our loudspeaker will be the loudest. There are some options.

Option 1: find a frequency where the loudspeaker exhibits extremely high sensitivity.

Option 2: use the crossover frequency, where both drivers will be working together to generate more output.

Option 3: try to use both.

Earlier I provided the rated sensitivities of each driver. The assumption there is that, yes, there will be peaks and dips, but they’ll average out. Now, we want to make the loudest sound possible with this driver complement, so we’re very interested in those peaks. To find them, let’s visit the published curves for each driver.

Figure 1 shows the 1.4-inch compression driver’s frequency response. We can see there’s an elevated region from 1 kHz to 3 kHz, with a bump around 1.5 kHz.

Figure 1: Frequency response and electrical impedance curve of the compression driver mounted on a 90- by 40-degree horn with input signal of 2.83 volts.

Figure 2 provides the 12-inch cone driver’s response curve. This is actually measured in a bass reflex cabinet, so we can see the sensitivity drop towards the bass. Around 2 kHz, there’s some “peakiness” that represents an area of high sensitivity. This happens to be related to cone break-up, where the cone itself is exhibiting some resonances.

Figure 2: Frequency response of the 12-inch cone driver made in a hemispherical free field. It was mounted in a reflex box with an internal volume of 55 liters and tuned at 60 Hz, applying a sinusoidal signal of 2.83 volts (8 ohms, at 1 meter).

Employing Option 1: the compression driver has it’s highest sensitivity around 1.5 kHz, and it comes in at 113 dB. The 220-watt amplifier means a peak SPL of 136.5 dB. The 12-inch cone driver has it’s highest sensitivity at 1.8 kHz (a narrow peak indicating cone break-up, but that won’t hinder us in the pursuit of the loudest possible sound!), coming in at 104 dB. With 2 KW input, that gives us 137 dB.

Now we’re getting somewhere.

Employing Option 2: the crossover frequency is unique because both drivers are contributing equally to the output of the loudspeaker. Temporarily ignoring the peaks and dips discussed above, there are two bands capable of about 132 dB. Put those two together, and we attain 138 dB, an excellent result.

We can go further, though. Employing Option 3: 1.8 kHz is a good place to start, since the cone driver has it’s highest sensitivity there while the compression driver is still around 112 dB. We’ll get 137 dB from the cone driver and 135.5 dB from the compression driver. When combined coherently, they’ll produce a touch over 142 dB!

(Now that’s a number that will sell. We’ve even beaten Manufacturer A!)

It doesn’t matter that our loudspeaker can do 142 dB only at one carefully-chosen frequency, or that both drivers will be showing signs of extreme stress when pushed that hard. The stated sound level can indeed be produced by this loudspeaker (probably without destroying anything) so that’s what the marketing team is likely to run with.

A brief side-note: for the 12-inch cone driver to hit that SPL at it’s worst-case frequency, the cone would need to move a little over 3 inches one way, where it would likely turn itself inside-out.

Read More
30 For 30: Lessons Learned From (Many) Years Of Tuning Sound Systems

Future & Conclusion

I’d like to see professional audio move towards more sensible numbers. For our example system, we came out with a “peak SPL” rating of 142 dB when the reality would be more like 132 dB, and less when bass is required. That’s a huge discrepancy.

Perhaps we could rate it as follows:

— 142 dB LSP (loudest sound possible) or maybe we could call it WLS (when lightning strikes);

— 132 dB mid-high, not constrained by cone excursion at low-frequencies, i.e., subwoofers are in use;

— 120 dB full-range, worst-case scenario where there’s a lot of bass content.

Before concluding, I submit this for your consideration: let’s say I have a 15-inch subwoofer, housed in a ported box where the response slopes from 97 dB at 1 watt at 100 Hz, down to 93 dB at 40 Hz. The driver also has a cone break-up peak at 1.3 kHz, where sensitivity reaches 103 dB. The amplifier will produce 1 KW.

What’s my maximum SPL? You could say 123 dB or 127 dB within the intended frequency range of the product, or perhaps 133 dB if it’s run up to 1.3 kHz. Care to guess which number many manufacturers will use?

It’s important for potential buyers to be able to compare different loudspeakers and draw meaningful conclusions about whether products are right for them. Currently this can be quite difficult, and at times, pretty much impossible with so many ridiculous claims that are very hard to independently verify.

As a result, we have to make do with hearsay and quick demonstrations, often in sub-optimal conditions. Frankly, we deserve better.

Supported By

Celebrating over 50 years of audio excellence worldwide, Audio-Technica is a leading innovator in transducer technology, renowned for the design and manufacture of microphones, wireless microphones, headphones, mixers, and electronics for the audio industry.