Dear Jim and All,

Isn't it sad that there are many people � indeed whole industries � who
choose their units in such a way that the end result is confusion and
conflict.

Could it be that obfuscation is their original goal?

Cheers,

Pat Naughtin LCAMS
Geelong, Australia

on 2003-02-11 13.05, James R. Frysinger at [EMAIL PROTECTED] wrote:

> Interesting. I would have interpreted those figures somewhat
> differently, as noted below...
> 
>> kilopascal wrote:
>> 
>> 2003-02-09
>> 
>> What a weird mess!
>> 
>> I too was confused by the awkward use of decibels until I went to
>> Rowlett's page and found that there are many different decibels.  I
>> have highlighted the definition of the two decibel units seen on the
>> package in red.
>> 
>> I would interpret the "106 dB spl/mW +/-4 dB" to mean that for each
>> milliwatt of power applied to the speaker, there is 4 Pa (2.5 - 6.3
>> Pa) of pressure applied to the ear drum or produced by the speaker
>> diaphragm.  I found this by  setting 106 = 100 + 20�log10(P/2) and
>> solving for P.  The decibels are positive because the pressure is
>> greater then the 20 ?Pa reference.
> 
> Here, John, I would have taken the +/-4 dB as applying to the first
> figure, thus in a slightly different format:
> SPL(re 20 �Pa) = (106  4) dB at 1 mW input power.
> It cannot be SPL/mW = [etc.] since doubling input power would raise the
> SPL (sound pressure level) by 6 dB; it would *not* double it to 212 dB
> of course!
> 
> Likewise, I would interpret the next figure for microphone sensitivity
> as being
> L_V(re 1 V) = (-39 � 5) dB at 1 Pa SPL.
> Of course, that would be better written as
> L_V(re 1 V) = (-19.5 �- 2.5) dNp at 1 Pa SPL
> since voltage comparisons call for the use of napiers while power
> comparisons call for the use of bels. (Read L_V as "L sub V".) There is
> some information on this in IEEE std 260.1 which is about to be put to
> ballot.
> 
> The above comments may not be the most elegant way to put things, but I
> think that they are closer to what was meant. But, of course, I'm
> guessing, so I could be wrong about that.
> 
> Jim

Reply via email to