On Oct 8, 5:35=A0pm, c...@[EMAIL PROTECTED]
wrote:
> On Oct 8, 4:41=A0pm, Randy Yates <ya...@[EMAIL PROTECTED]
> wrote:
>
>
>
>
>
> > c...@[EMAIL PROTECTED]
writes:
> > > [...]
> > > I was trying to get at the fact that we don't put an additional
lette=
r
> > > to denote the reference such as the case with dBm and dBw.
>
> > Hi Clay,
>
> > I don't know why you would assert such a thing. I don't agree.
>
> > There are multiple reference levels, even within the context of
> > acoustics, and even within the context of acoustics for humans.
>
> > In both cases, the expression determining dB is
>
> > =A0 dB =3D 20 * log_10 (p / p_ref),
>
> > The first reference level is dB SPL, which is mentioned in other
posts,
> > in which p_ref =3D 0.0002 microbar (2E-5 newton/m^2). This also
> > corresponds to an intensity of 10^(-12) W/m^2.
>
> > The second does not have a special acronym or identification
> > but it is p_ref =3D 1 microbar.
>
> > I quote these from Baranek.
>
> > Baranek concludes the paragraph with the statement
>
> > =A0 "The reference pressure must always be stated explicitly."
>
> > Unless I'm misunderstanding you, this would seem to diametrically
> > contradict your assertion.
>
> > --Randy
>
> > @[EMAIL PROTECTED]
> > =A0 title =3D "Acoustics",
> > =A0 author =3D "{Leo L. Baranek}",
> > =A0 publisher =3D "McGraw-Hill",
> > =A0 year =3D "1954"}
>
> > --
> > % =A0Randy Yates =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0% "Maybe one day
I'=
ll feel her cold embrace,
> > %% Fuquay-Varina, NC =A0 =A0 =A0 =A0 =A0 =A0% =A0 =A0 =A0 =A0 =A0 =A0
=
=A0 =A0 =A0 =A0and kiss her interface,
> > %%% 919-577-9882 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0% =A0 =A0 =A0 =A0 =A0
=
=A0til then, I'll leave her alone."
> > %%%% <ya...@[EMAIL PROTECTED]
> =A0 =A0 =A0 =A0 =A0 % =A0 =A0 =A0 =A0'Yours
Truly=
, 2095', *Time*, ELO =A0http://www.digitalsignallabs.com
>
> Certainly, the plethora of difference refernce values, would indicate
> one should state which reference he is using. When I get back to the
> farm, I have an old book, "The Science of Measurement," which has a
> couple of chapters on sound measurements. I'll review it and see how
> its author denotes the different reference values. This book is good
> about giving the historical basis behind various units and how they
> sometimes get redefined over time. Sometimes they get rounded to some
> nearest convenient value. A kT TNT energy equivalent is one such
> example. It is taken to be 10^12 calories which is a value very close
> to experimental values for TNT. =A0 I'll re****t what I find out about
> the sound measurements.
>
> Clay- Hide quoted text -
>
> - Show quoted text -
As promised I dug deeper into this acoustic conundrum. Basically you
can use either an intensity (power per area) or a pressure as a
reference. The gotcha is their link:
(assuming air - change for other media as needed)
Intensity =3D (pressure^2)/(density of air)(velocity of sound in air)
This relation is temperature and pressure dependent!
According to Klein's book "Science of Measurement", intensity was
often used as the reference early, but now more often than not
pressure is being used for the reference.
So for a reference pressure of 20 micropascals (newtons per square
meter) we find the equivalent intensity to range from 0.917 up to
0.998 picowatts per square meter over a temperature range of -10C up
to +40C. Some use an intensity of 1.0 picowatt per square meter.
This means then 1 watt per square meter is essentially 120 dB SPL, so
I errored when I gave 0.946 w/m^2 as 100 dB SPL - for that I apologize
for the error in the exponent (20 dB offset). Incorrect data is worse
than no data at all.
For SPL measurements, we may write
dB =3D 10*log(Intensity/Intensity_reference)
from the intensity to pressure relation, we then find
dB =3D 20*log(pressure/pressure_reference)
where the common but not universal references being:
pressure: 20 micropascals
intensity: 1 picowatt/meter^2
I hope this clarifies any issues I created.
Clay
|
