Re: [AUDITORY] Tweeters for noise trauma

Hello Boris,

I would add some to what Bob mentioned, but also ask some questions:

In the studies you refer to, are the noise trauma stimuli delivered in the free field or in a closed acoustic system attached to the ear? If in a closed system, it is quite easy to develop 130 dB SPL.

The reason has to do with the acoustic load (impedance) that the speaker sees. As Bob mentions, most speakers are specified to deliver a certain sound pressure level at a certain distance from the speaker - because the sound radiates into a 3-D space, the sound power per unit area decreases as the square of the distance from the speaker, and therefore the sound pressure decreases linearly with the distance from the speaker. (In a reverberant space, the echoes mean that there is more sound power, because it does not radiate away.). Speakers are specified in terms of sound pressure, because that’s easy to measure. In fact, what limits them is the electric current you can send to them without overheating, and the output related to the current is the sound volume velocity.

In a closed acoustic system, for a given electrical drive, the same volume velocity will produce a higher sound pressure because the acoustic impedance that the speaker sees is much higher. At our lab, we can generate 130 dB SPL or so in a mouse ear from an old Radio Shack speaker (40-1310 or 40-1377) or a standard Beyer DT-48 audiometric earphone, because the sound is delivered from the speaker directly to the ear through a tube fit closely to the ear, and the small tube and ear have a much higher impedance than the open field. These tweeters were marketed to the amateur audiophile and have a frequency response far beyond the range of human hearing - but ideal for small animals. The electrodynamic tweeters actually have a fairly flat frequency response; the piezoelectric tweeters, not so much.

More recently we have used a Skanspeak (? I’m not at my office and don’t have that info handy) or Fostex speaker (<https://www.madisoundspeakerstore.com/bullet-tweeters/fostex-ft17h-horn-super-tweeter/> - looks like a copy of the old Radio Shack 40-1310) for midrange sounds and a small concave dome tweeter (originally through Parts Express, now available through Digi-Key, I believe) for higher-frequency sounds (~2-50 kHz).

As Bob brings up, what is the frequency range you’re interested in? As he also mentions, speakers are usually rated for “music power”, which assumes that the sound frequency and level vary with time. Using a steady-state noise delivers much more actual power of the same combination of current and voltage. To find your speaker’s effective power handling capability, you may just need to play your stimulus for awhile and monitor the sound output level and the temperature at the back of the speaker: If the former goes down or the latter goes up, you’ll need to turn down the voltage.

A last thought: As you have probably discovered, few tweeters seem to stay on the market for very long. Until a few years ago, one could still get some of these old tweeters on eBay, but they have been discovered - but there’s still this: <https://www.ebay.com/itm/Radio-Shack-40-1310B-Horn-Super-Tweeter/174607236018?hash=item28a766fbb2:g:UDkAAOSwtrNgCzmP>. Best just to spend a couple hundred dollars/euros and get half a dozen different models and try them out; and once you find one that will work, buy several spares.

Hope this is helpful - Mike

External Email - Use Caution

How are the ethical concerns in animal testing at such sound pressure levels considered?

Henrik Sundt

> 11. feb. 2021 kl. 11:12 skrev Boris Gourévitch <boris@xxxxxxxxx>:
>
> Dear auditory list,
>
> I was looking for a speaker (probably a tweeter) that would reach 120 to maybe 130 dB SPL in a high frequency range (for mice).
>
> I am a bit puzzled with the many papers inducing noise trauma (i.e. Warren, Fenton et al, J Neuroscience 2020; Amanipour et al, 2018; Chen, Sheppar and Salvi 2016; Novak, Zelenka et al, 2016) which actually used pretty common tweeters to do that. Technical specifications for these tweeters do not really indicate that they could handle such high SPL (and therefore such high voltage). How did they determine that ? How can we know then ? Only by sensitivity measure adjusted for the watt power of the speaker ?
>
> Let's take an example: in the Warren et al above cited, the Visaton speaker from RadioSpare is used to deliver 3kHz at 126 dB SPL. The technical specifications (https://docs.rs-online.com/170c/0900766b802e5b23.pdf) indicate 30W max and a sensitivity of 86 dB. Thus, it seems that this speaker could not deliver above 101 dB SPL. How did the authors do to avoid damaging their speaker ?
>
> Same question with other speakers as well. I'm not an acoustician/electronician and I am probably missing something here.
>
> Many thanks for your help. Also, many of the tweeters from the papers above are discontinued so if you know a pretty robust one that you are safely using at high SPLs, please do not hesitate to tell me.
>
> Best regards
>
> Boris
>
> --
> Dr Boris Gourévitch
> CNRS Researcher
>
> Institut de l'Audition
> UMR1120 Inserm, Institut Pasteur
> 63 rue de Charenton
> 75012 Paris, France
>
> Tel: (+33) 1 76 53 50 41
>
> Personal Web page: http://www.pi314.net/
>
>
> --
> L'absence de virus dans ce courrier électronique a été vérifiée par le logiciel antivirus Avast.
> https://www.avast.com/antivirus