Whereas I often wear a set of noise-canceling headphones on a plane, I have always wondered how effective they are. This was brought home recently when a friend let me listen to his new ones and asked me what I thought of them. Wow! It was like walking around with your head in an anechoic chamber. Actually, they were too good, but they were better than mine. I have two sets because I forgot my original set on a flight one day and at the layover point decided I needed a quieter life so bought another pair at the airport. Although fairly cheap, I reckoned that they were better than my original ones and were certainly more comfortable because they sat around the ear rather than on the ear. I am not sure if the sound quality was any better, but it was certainly better than the usual aircraft headphones.
Then, on a recent 10-hour flight back from the US, I got bumped up to business class and, apart from the additional leg room and other niceties, that also came with a set of noise-canceling headphones. I was hooked. The cans weren’t hugely comfortable, but the noise reduction was pretty good. So, before my next flight, I treated myself to a new set of noise-canceling headphones. These also go around the ear and are extremely comfortable. A five-hour flight almost became a joy; for the first time ever, I was quite impressed with the sound quality of the onboard audio entertainment. Now, I wouldn’t want to use (listen to) these units for too long because they are too screechy (and a bit bassy), but onboard a plane, they are just great. I happened to have them with me the other day when onsite and trying to judge the quality of an audio signal under far-from-ideal noisy conditions. They worked a treat and turned what could have been a frustrating half hour into a productive session.
The trouble with being an audio engineer with an inquisitive mind is that I wanted to know what the performance of my headphones is. Considering that these days you can pay more than $1000 for a set of headphones (well more than that in quite a few cases), there is a remarkable paucity of information that accompanies them. I have yet to see a manufacturer provide any real data. This is particularly true of the noise-canceling units. Surely, that is exactly what you need to know: How much do the headphones reduce the noise? If you buy a car, you want to know how fast it will go and how many miles to the gallon you might get.
So, I don’t think it is unreasonable that a headphone manufacturer should tell you what the performance of a given model is. For example, the technical spec of a well-known brand leader lists the following under “technical specification”: Dimensions, Weight Without Cable, Length of Cable, Type of Battery (for noise-canceling model) and Battery Life if you really dig deep.
But technical performance? Not a jot!
However, my studio-quality reference headphones aren’t much better, although, initially, it appears that there is something more than just lowest common denominator marketing stuff: Frequency Response: 10Hz-41kHz (no tolerance or limits), so pretty worthless info really; Sound Pressure Level: xxxdB (no indication if this is max, long term or instantaneous before self destruction or what); THD: < 0.0x%, but at what level?; Contact Pressure: ~2.5N (now that’s interesting, but difficult to know how much pressure that really is (e.g., a vice grip or a weak stiction that only just overcomes gravity?); Ear Coupling: circumaural (i.e., round the ear); Cable Length: xm; Jack Plug: 6.35mm; Weight Without Cable: xg; Nominal Impedance: xΩ; Transducer Matching Tolerance: +/-1dB.
So back to my noise-canceling inquisitiveness…. It was no good; I had to get the binaural measurement head out and hook up the analyzer.
Now, when I put on headphones, I always listen to the effect that they have on the ambient noise I am hearing (passive, as well as active models). Although, overall there is generally a reduction in noise level, I am often acutely aware of an increase in some frequency bands. Well, it turns out that my ears haven’t been deceiving me! Feast your eyes on the sound reduction (and amplification) shown in Figure 2. The noise level was around 80dBA, but for ease of viewing, I have normalized the readings to show the measured level difference. Figure 2 shows the attenuation and gain that some typical passive headphones may give.
Looking at Figure 2, it is quite clear that some passive headphones can actually make matters worse when wearing them in a noisy environment. The top blue curve (1) for example is for an “on-the-ear” open-back unit that clearly can be seen to be increasing the noise level by more than 5dB at around 4kHz. The red curve (2) is for my studio reference units that have a pretty neutral effect, except above about 2kHz, when they provide some useful attenuation. I find the green curve (3) slightly ironic because this is for a pair of high-quality “business class” airplane passenger headphones that clearly make matters 5dB worse at around 500Hz. The purple curve is for a set of good noise-canceling headphones, when switched off. These provide really useful attenuation above 600 to 800Hz, but provide some gain (i.e., passively amplify the sound) in the 160 and 400Hz bands, although only by around 3dB.
Figure 3 shows that my original “on-ear” noise-canceling headphones are pretty useless, increasing the noise level by 4 to 5dB at around 1kHz, even when working(!), with only a 2dB improvement at low frequencies. The high-frequency improvement above ~2kHz is a purely passive effect due to the construction of the units. The circumaural headphones do a slightly better job, providing around 2 to 4dB of active attenuation, which is certainly quite noticeable and an improvement for aircraft listening.
Figure 4 shows the performance I measured for my new set of noise-canceling headphones. Now that’s more like it, with 10 to 15dB of attenuation across the audio band. It is interesting to see that above ~1kHz, it is the shell of the headphone that is passively reducing the noise rather than the electronic circuitry. Also, I found a difference in performance between the left and right ears above ~3kHz. However, at these frequencies, just taking a set of headphones on and off can produce significant variations in performance.
Now, what is the take away from all of this? Well, (1) is to believe your ears and (2) I can listen in a far more relaxed manner in a noisy environment these days. Pity that I had to use more than $15,000 of test equipment to find out! But engine noise is now a thing of the past. Ah, the silence of the cams.