On the acoustics advice from the world’s leading cat audiologist, Dr. Snicker’s McFeline, it became apparent that what is seen in the sound field, such as my living room, does not necessarily occur in the small occluded volume of an ear canal plugged with an earphone or in-ear monitor. This was discussed in last week’s blog.
Specifically, even though higher frequency sound energy emanates from a loud speaker in a room in an almost laser like directional beam, this is not the case for in-ear monitors or in the ear earphones that a musician, or anyone who likes to listen to portable music, may wear. Sound does not possess the same directional characteristics as a function of frequency in the occluded small volume ear- at least not in the normal bandwidths that are typically found for in-ear monitors and earphones.
The idea that an in-ear monitor should have an ear mold impression whose most medial end aims at the center of the eardrum is false.
So given this improved understanding of in-ear monitors and earphones, what are some optimal parameters of custom made earphones? Specifically, what should we be asking those hearing health care professionals who make earmold impressions for custom earphones or in-ear monitors?
Well, it turns out, its much the same thing that we have always been doing.
An earmold impression should have two characteristics: (1) made with a long canal past the second bend and (2) made with the mouth open. Part 1 of this blog series will appear today with Part 2 appearing next week.
Earmold made with a long canal past the second bend:
Historically the reasons given for making long earmold impressions were to aim the sound, especially the higher frequency sounds, towards the center of the eardrum, and to reduce or at least minimize the occlusion effect. This first reason is incorrect. The second reason however is quite correct.
The occlusion effect is a measurable increase in low frequency sound from a person’s own voice. One can measure this low frequency increase in sound level using any clinical probe tube microphone. For the audiologist readers of this blog, calibrate the probe tube system in the routine fashion; perform an “unaided” measure in the routine fashion (this step is optional); disable the reference microphone and the loudspeaker output- if using the Audioscan set the stimulus level to 0 dB, and if using the Frye system set the stimulus level to “off”. You are now ready to measure the sound level of any stimulus in the person’s ear canal.
Have the person utter the vowel [i] as in ‘beat’ with their ear canal unoccluded and then do the very same thing with the earcanal occluded. This difference is a measure of the occlusion effect and can be as great as a 20 dB low frequency increase in sound level. This can also be done with the vowel [u] as in ‘boot’ but it must be a high vowel. For those of us who remember our speech acoustics classes, high vowels have very low frequency first formants F1 (around 125-150 Hz) and it is this low frequency energy that is enhanced with the occlusion effect. In contrast if a low vowel as used such as [a] as in ‘father’, there would be minimal occlusion effect since the first formant F1 would be at around 500 Hz. Some of this can be found in Attenuation variables in earmolds for hearing aids by Pirzanski, myself, Klenk, May, and Purdy- sounds like a law firm, eh?
Try it for yourself, even if you don’t have a probe tube measurement device. Utter the sound [i] with your ear occluded and unoccluded and you should notice a significant difference.; the same cannot be said of the low vowel [a].
Alternatively, this is a quick and easy trick to verify that any hearing aid is not occluding. Since the sound level of [i] and [a] are almost the same, have your clients say these two vowels when fit with a hearing aid, in-ear monitor, earphone, or hearing protector. If the loudness of [i] = [a], then there is no appreciable occlusion effect and you should be confident that this will not be an issue in the fitting. If however, the loudness of [i] > [a], then a significant occlusion effect is apparent. There is a high probability that this client will be back in your office for a remake (or return of device) in short order.
The clinical strategies to reduce the occlusion effect are to either extend the device (and impression) significantly into the boney portion of the ear canal (past the second bend) or alternatively to make it shorter but with an air vent placed in the device. For long bore impressions if there is still some occlusion effect (i.e., if [i] is slightly greater than [a]), all that may be required would be a narrow 1.4 mm diameter vent. There would be some compromise with the bass response of the device, but like many things in clinical practice, a proper balance needs to be achieved.
Earmold impressions should be made as long as possible, preferably past the second bend in the ear canal. This has everything to do with minimizing the occlusion effect and nothing to do with ensuring that higher frequency sounds are optimally transduced.
