Shifting ear canal resonance (Patreon)

Lately I experienced some interesting frequency shift effect of my ear canal resonance down by 200hz. I had already experienced this effect before like reported a while back and some slight variation of your ear canal shape etc during the day seems also quite common, but this time I got a shift down to around 5800 from my usual 6000hz. Keep in mind that this is the value I usually get with the Free Pro or Free Pro 2 models, with other IEMs this resonance may be different due to different insertion depth and different tips etc.

Thus from one day to the other my Free Pro 2 tuning I was already listening to for some weeks with the same single setting and couldn’t quite manage to improve it more, all of a sudden it started to sound just bad. I was currently comparing and trying to match and fine tune the spectrums of speaker headphone and IEM to make them all sound more or less the same, therefore I could hear this effect immediately as the Free Pro 2 started to sound just off compared to both speaker and over ear.

Given the fact that exactly this day I had received the Etymotic ERX (more on this model to come soon) and was trying to somehow fit it into my ear but couldn’t quite manage I must have somehow changed some properties of my ear canal, maybe stretched too much like a virgin pussy or maybe pushed some earwax too deep inside thanks to that deep insertion or whatever but it had a significant effect on the perception I am hearing from IEMs. I didn’t notice this effect at all when listening to the KH80 and the HD58X as the altered ear canal properties must have the same influence in both cases and change the sound in a similar way but with a closed ear canal you form a particular resonating chamber which mostly defines the resulting sound through its main resonance and the upper harmonics. Changing the length or diameter of this tube will immediately change its acoustic properties, like with a flute where closing more holes and creating a longer resonance chamber will result in a lower note, nevertheless it’s quite shocking how big the effects on the final sound can be by a just a shift of 200hz. When I noticed that the Free Pro 2 started to sound weird I immediately performed some sine sweeps to better know what’s going on. I heard a drop in amplitude at 6000hz although usually I hear it with a similar amplitude towards around 6800hz and even slightly rising, but this time I heard a drop here. Listening to the raw tuning of the Free Pro 2 I could hear the peak closer to 5800 now and first thought I should immediately visit my otologist to let my ears clean despite having them already cleaned about 3 weeks  ago. But using IEMs from Etymotics always caused similar problems in my case. Instead I decided to use this effect scientifically and tried to make the Free Pro 2 sound normal again by comparing the sound to both the speakers and the overear headphone and finally measure the final graph. In the image you can see what I achieved. In the graphs on top you can see the raw response of the Free Pro 2 measured with my usual resonance of 6000hz and in red with 5800hz. Below you can see my most recent setting for the Free Pro 2 at 6000hz and in red again measured at 5800hz. The differences may appear subtle but they were immediately hearable, it simply sounded way worse and wrong. After correcting the sound to get again something as close sounding to the speakers as possible I came up with the blue graph. To me it now sounds very similar to what I heard before if not even slightly better. Notice that speakers, headphone and the IEM never sounded exactly the same to me, I couldn’t achieve this but I always tried to make them resemble each other and now after readjusting the sound it seems to be OK and similar enough to me again.

You can see that the graphs I measured at 6000hz and 5800hz in blue are nearly the same, they differ slightly in the 6khz region and also in the upper frequencies but these differences are not that disturbing to make it somehow appear off. High frequencies are often more forgiving as mostly you just need to get a correct overall balance of all frequencies but it doesn’t seem to matter that much if 12khz is less if at the same time 14khz is louder. The ear is not that extremely picky with treble and it’s really hard to get it 100% right due to masking effects etc.

With this example I somehow proved again that you need to compensate for different ear canal lengths to achieve the same or at least a similar final response someone else would get with a longer or shorter ear canal. If that final response will still sound correct to  someone else, is another topic as the effects of the individual outer ear will usually put again some own stamp onto what you may hear as natural compared to me. But without compensating for the effects of the ear canal you cannot even expect to get something usable at all in terms of a correct natural frequency response.

Of course I cannot verify how precisely these shifts may still measure with even higher values, and if they will still lead to correct results, probably some errors will be introduced the more our ear canal resonances differ but because of this I try to achieve a setting for myself which sounds as perfect as possible to not introduce further errors that may increase at other ear canal resonances.

I am currently working on finishing all my targets and did some slight corrections recently which I will also use to get a final compensation for my recordings, then I will start to translate these targets either to other headphones or IEMs and hope my settings will also give others a "correct" kind of sound as usually perceived from speakers.