Post by Ken PlotkinA tremendous problem with STC for this kind of question is that STC is
intended for evaluating interior-to-interior sound insulation, not
exterior-to-interior sound.
Agreed
Post by Ken PlotkinSTC is based on transmission of typical
indoor noise spectra,
It's not even that elegant. I have observed, and seen and heard the
arguments for, the development of the STC 'rating curve'. It's history
is long and anarchic. It seems that before and during WWII, the means of
deriving a single number for transmission loss test results was to
average tl values from maybe 100-500 Hz (exact range not certain,
probably pure tone or warble tone tests from 128cps through 4096cps.
This was practiced by NBS personnel and others at the time.
After WWII, German rebuilding and testing of bombed out building in many
German cities brought the need for more definitive test analyses. The
observed that a definitive concern was the transmission of amplified
voice signals, typically emitted from radio sets of that period, and the
100-4,000 Hz range was a reasonable way to represent it, but some
frequency weighting ws in order to make the results more meaningful.
Since masonry has a coincidence dip around 300 Hz, accuracy in that area
was unimportant, Since radio sounds and voices have relatively little
energy below 100 Hz (then), there was no need to evaluate their walls
and floors below that frequency. Finally since masonry also has good
noise isolation at lower frequencies, there really was no motivation to
devote efforts to lower frequency sounds. A rating scheme involving 16
(a nice round and square number) 1/3 octave band center frequencies plus
a nice 0, 1 and 3- slope dogleg "curve" was evolved and everybody was as
pleased as plum for decades. The US used 125 Hz and 4,000 Hz as terminal
frequencies (ASTM E413), but the European rating scheme (ISO 717) used
100Hz and 3125 Hz as terminal frequencies, paying tribute to the fact
that lower frequencies needed more attention. The US stuck doggedly to
125 Hz as the lower limit, adding an 8 dB limit for the greatest
deviation of test data from that dog-leg curve to cover for any really
bad coincidence dip (drywall and window glass) or low frequency (double
leaf walls) transmission loss deficiencies.
The "Significance and Use" statement for ASTM E413, the US dog-leg
rating curve scheme clearly states in 4.2 that "...This classification
method is not appropriate for sound sources with spectra significantly
different from (speech, radio, and television). Such source include
machinery, industrial processes, bowling alleys, power transformers,
musical instruments, many music systems and and transportation noises
such as motor vehicles, aircraft and trains."
And yet STC remains prominent today. A few brave new souls have ventured
to practice ASTM E1332. This uses a weighted 'transportation noise'
spectrum from 80 Hz up) as an hypothetical impingement, then tallies the
A-weighted result of the purveyed noise. The result is "OITC"
(Outside-Inside Transmission Class"). The EWR is similar.
Post by Ken Plotkinwhich are higher frequency than the
transportation noises that are usually of interest for exterior walls.
Exterior Wall Rating (EWR) was developed to account for this. It's an
STC-like composite, but based on road, rail and aircraft spectra.
Unfortunately, it never caught on, so the mis-use of STC continues.
We in ASTM E33 strongly (severely?) recommend the application of E1380
(OITC) for the rating of all facade elements including doors, windows,
walls and roofs.
Angelo Campanella