Torq’s Isolation Measurements
Rather than try and guess what level of isolation a given pair of closed-back headphones yields, and how well that matches to the needs of a given individual, situation or environment, I have started using a simple, repeatable, process to measure the isolation provided in a more applicable manner.
There are two types of isolation to consider:
- Leakage: How much sound escapes from the headphones and is audible to an outside listener.
- Attenuation: How much outside noises are attenuated for the wearer of the headphones.
Leakage Measurements
These are presented as the listening volume level, for a given set of frequencies, necessary for an outside listener to discern any sound coming from the headphone.
The listening levels are measured using a miniDSP EARS unit, while the headphones are fed a single-tone sine-wave at the respective frequencies (100 Hz, 300 Hz, 1 kHz, 3 kHz, 6 kHz and 10 kHz - lower frequencies are less audible to humans, and peak sensitivity is typically around 3 kHz).
Audibility is determined by averaging the lowest level at which several listeners first detect any sound coming from the headphone at a distance of 1 meter, in a 40 dB environment (equivalent to a quiet room) for each frequency individually.
The resultant values are graphed. A value of, for example, 85 dB @ 3 kHz means that a 3 kHz tone (roughly the center point for the human voice) could be heard by an outside listener, one meter away form the headphones, in a 40 dB room.
Higher values are better.
Attenuation Measurements
Most hearing-protection devices have a Noise Reduction Rating (NRR), measured in decibels; however this value does not directly reflect the actual reduction in sound level at the ear! This means that the best available in-ear hearing protectors, which have a NRR of 33 dB, would not, in fact, reduce a 100 dB concert to 67 dB.
Instead, the accepted methodology de-rates the quoted NRR protection value by 7 dB (to account for variances in fit/application), and then you divide that result by 2. This means that those NRR 33 dB protectors, in a 100 dB environment, only actually reduce the SPL to 87 dB:
(NRR 33 dB - 7 = 26 dB) / 2 = 13 dB reduction : 100 dB - 13 dB = 87 dB.
Don’t ask me … I didn’t come up with this stuff!
That’s for in-ear isolation. Over-ear (muff) is less effective and consequently the best available over-ear SPL reduction carries an NRR of 30 dB. Such an ear-muff is very thick and deep, and at odds with the design of most audiophile headphones. This is important as it sets an upper limit for what any headphone can potentially do - and since most are not designed to provide protection they will have much lower ratings.
Put another way, headphones are much better at keeping sound in than out, and expecting any closed-back, non-active-noise-cancelling headphone, to provide significant attenuation of outside sounds is likely to prove disappointing. In an outdoor or public environment, such as riding the bus, you’re still going to be able to hear conversation, traffic noise, etc. in any quieter patch of music, in track-breaks, and so on … especially if you listen at lower levels.
Even the best, professional, fully-rated, over-ear hearing protection will not shut environmental noise out entirely!
Note: I am currently working on a measurement methodology that provides a consistent way to assess this NRR in accordance with industry standards, and will begin adding this measurement to reviews as soon as I am happy with the results.