Noise isolation is the headphones' ability to block ambient noise passively and/or actively.
Passive isolation is a physical contribution to isolation and depends on headphone or earbud design. Typically, more padding on the ear cups or earbuds results in better passive isolation.
Active cancellation refers to the noise cancelling technology that enhances isolation by nullifying sounds that seep into the headphones. Well-isolated headphones provide a noise-free listening experience, even in loud environments or on busy commutes.
For noise cancelling performance, we test the overall attenuation, bass, mid and treble attenuation and self-noise.
When it matters
If you’re often in noisy environments, a well-isolated pair of headphones will keep the ambient noise from distracting you from your listening experience.
Although listening at high volumes naturally overpowers ambient noise of lesser amplitudes, more active and passive isolation is required at lower volumes, or if the headphones are simply used to block sound to make it easier to focus.
Poor noise isolation can be a significant issue, depending on the loudness of your environment. Urban commutes and international travel are examples of when you might require good noise isolation.
Our isolation tests are performed in a 6' x 3' x 3' isolation box. The box is partially treated with bass traps and absorbers, and an M-Audio BX8 Carbon speaker is placed at one end of the box with the HMS (Head Measurement System) being positioned at the other end. The speaker is positioned at an angle in order to create a balance of direct and diffuse sound.
The isolation test is performed at 3 different angles (0, 90, and 180 degrees), once with the headphones on the HMS and once with no headphones on the HMS. This process is repeated 5 times with a full re-seat between each repetition. The isolation curve is derived by deducting the HMS-only measurements from the HMS-with-headphones measurements for each angle, and then averaging the result of each angle to get the final curve. This effectively removes the frequency response of the box from the measurements and flattens the final result which is then smoothed down to 1/3rd of an octave in order to make the curve easier to read. For active noise cancelling headphones, this process is done once with the ANC on and once with the ANC off.
The test signal is a 10Hz-22KHz sine sweep at 90dB SPL. The calibration is done with a Pink noise (random noise with equal energy per octave) limited to 500Hz-2KHz so that the “room modes” of the box don’t interfere with the calibration process. However, since the room modes affect the frequency response of the Iso Box by more than +/-25dB, some equalization is applied pre-sweep using the RME TotalMix’s channel EQ in order to tame a couple of the major room modes. We didn’t find it necessary to completely flatten the response of the box since too much EQ can add unwanted artifacts.
Overall attenuation is calculated by averaging the amount of reduction in noise full-range (20Hz-20KHz). However, this number is not counted towards the noise isolation score, since some in-ears can get very high overall numbers by isolating extremely well in the treble range, even though their bass and mid isolation may be average at best. In the example below, the Jaybird X2 get a better overall attenuation value, but in real-life, the Bose QuietControl 30 provide better isolation. Therefore, only Bass/Mid/Treble values are counted towards the noise isolation score, since the scoring/weight distribution can be adjusted in a way that would make the noise isolation scores more perceptually relevant.
Attenuating bass frequencies are one of the more difficult aspects of noise isolation. Generally, only active noise cancelling headphones are able to achieve good results in this range. Some passive (no ANC) closed-back in-ears are able to achieve decent results in the bass range as well or even better, in some cases, than some mediocre Over-Ears with ANC. But generally, passive headphones will not isolate in this range.
Open-back and semi-open headphones are usually not able to create any significant isolation in the mid range. Some of the better isolating passive closed-back headphones are able to achieve decent results, but like with the Bass Range, only good ANC headphones and a few a passive in-ears are able to achieve good isolation.
All headphones, regardless of type and enclosure have an impact on the treble range isolation. However, only headphones with a closed-back enclosure and a proper fit/seal are able to achieve good results in this range. Active Noise Cancelling is not a factor in the treble range since none of the headphones we have measured so far isolate actively in the treble range and they use their ear cups/tips to isolate passively. As a rule of thumb, and given a proper fit/seal, closed-back in-ears do best in treble isolation, and closed-back over-ears come in second.
For active headphones, self-noise is measured by sending a 0dBFS sweep to the headphones and recording the results between 300Hz-20KHz. This removes bass and sub-bass artifacts from the test and the 0dBFS signal ensures that headphones are not in power-saving mode. For active noise cancelling headphones, self-noise is measured with ANC on.