MikeC I've done the analysis again and it's more like +23dB. See http://bingouille.ifrance.com/
(of course it's best viewing the jpg in 100% size; many browsers will badly resize the 1280*999 picture down, if so, find the no-resize option in your browser's picture display parameters, or simply download a copy of the 412 KB jpg file and view it at 100% size).
Also you can see on this graph that 450-750 Hz frequency range has similar amplitude to 1560 Hz pike; however human ear is particularly able to differentiate between two sound samples in the 200-2000 Hz frequency range so the 1560 Hz is easily audible apart from anything else.
I can extract the 1560 Hz pike sound from the rest of the frequency range (and keep amplitude unchanged) and post a small wav or mp3 to help identify what is a 1560 Hz sound.
Really hope your 7200.9 sample was bad // Seagate's nice reputation in hard disk noise aspects would hardly survive to such a bad thing as this 1.5kHz pike...
many thanks for your site, reviews, forums; it gave me great help a few days ago to determine what hard drives (which I seem to consume like cherries) to buy; after examining all those great mp3 samples in a multitrack audio mixer (to easily compare and switch from one to another) I just ordered 4 Samsung SpinPoint P120 (250 GB ultraATA version)
OK, I see what you have done. But to say it is a +23 peak is still quite misleading. It is certainly not pereceived in this way. You are measuring it relative to the level of the freq immediately adjacent. The reality is that the overall perceived level is dominated by the broader peak at 600Hz, as well as the lower frq noise. It's the total area under the curve that determines total SPL. In this context it is easy to see that the 1.5kHz peak is a tiny portion of the total energy.
You also need to take into account the fact that the conditions in which the recording was made were designed to deliberately accentuate the peak artificially
, as much as we could do -- just to illustrate what we're talking about.
Plewase don't take this info out of context!