Is there more to perceived loudness than just the sensitivity rating?
Feb 15, 2011 at 4:59 PM Thread Starter Post #1 of 6

caimanjosh

New Head-Fier
Joined
Feb 4, 2011
Posts
20
Likes
0
There's something that's been mildly bugging me for a couple of weeks now.  I have a relatively low-power input portable mp3 player, and as such I wanted some "loud" earphones to go with it.  As a result, I looked for earphones with a high sensitivity rating, believing that this would result in the most "loudness".  I eventually settled on a pair of Sennheiser CX281's, which claim they have a sensitivity rating of 120 db (outrageously loud, I figured).  However, after trying them out, I found that they seemed like they were actually not as loud as my old Creative EP-630's...which were only rated as having sensitivity of 106 db.  Both sets have an impedance of 16 ohms. 
 
So, my question is, what gives?  Why would the earphones with the lower sensitivity rating actually end up seeming louder?  I note that the EP630's had significantly more mid-bass...could this be why my ears think they're louder?  What other factors may be coming into play? 
 
Feb 15, 2011 at 5:09 PM Post #2 of 6
Sensitivity is always measured at a specific frequency - usually 1,000 hz.  If that frequency happens to occur at the top of an enormous peak, then overall the phone might sound quieter than a phone where 1,000 hz occurred at the bottom of a deep trough.
 
That's an exaggerated example, but I hope you see the point.
 
Also, even though the impedance numbers are the same, the same principle applies ... the numbers are nominal.  Impedance can vary with frequency.  One might present very different impedance over a broad span than the other.  If it goes generally lower, it will ask for more current, which your player might not be providing.  If it goes higher, the player might not swing enough voltage.
 
There's no substitute for real-world audition, as you have found.
 
Feb 15, 2011 at 5:14 PM Post #3 of 6
search for Fletcher Munson Loudness curve
 
also you now have to check the reference level for sensitivity - some manufacturer's are using 1 Vrms instead of the previous 1 mW, at 16 Ohms the difference is ~ 18 dB
 
Feb 16, 2011 at 10:27 AM Post #4 of 6
I think bassier headphones will be perceived louder.
 
I thought I was running my K702 at very moderate levels (though the bass still felt very full), but I noticed the next day that my ears were ringing just from listening for a few hours. It seems like I listen to my Ultrasone HFI-580 a lot louder, but they don't give me any ringing whatsoever.
 
Feb 16, 2011 at 10:42 AM Post #5 of 6
I see...so it may well be that my CX281's have a more powerful mid-range (I'm pretty sure they do), but less powerful bass.  So measuring sensitivity at 1 kHz results in a higher number for them than for the EP-630's, even though to my ear, that doesn't seem to be the case.  That makes sense now.  Regarding the difference about Vrms vs. mW, do you mean that the measurements made using mW are louder?  Or the other way around? 
 
Feb 16, 2011 at 11:05 AM Post #6 of 6
Quote:
also you now have to check the reference level for sensitivity - some manufacturer's are using 1 Vrms instead of the previous 1 mW, at 16 Ohms the difference is ~ 18 dB


This is the crux. Sennheiser often uses db SPL @ 1 Vrms instead of @ 1 mW, so especially with low impedance headphones the sensitivity number will be a lot bigger.
(1 V into 16 ohm = ~63 mW!)
 
Like jcx wrote, just substract 18 from 120 dB and you'll get 102 dB @ 1 mW ... this is not as loud as the EP-630 with 106 dB as you wrote.
 
(Of course the frequency response curve also plays a role.)
 

Users who are viewing this thread

Back
Top