Spyrow wrote:
For example, this
link from the University of Western Australia which says this about improving TL:
On the other hand, it is possible to create composite or sandwich panels whose total SRI does approach that of a double wall, if the following points are considered.
<snip all other points>
- Use of layers of different thickness can greatly assists in mis-matching resonant and critical frequencies across the panel.
It's nicer that you write: School of Architecture and Fine Arts (somehow a dep. of or related to that Univ. you refer)
I know that site, and it's rather simplified stuff.
Often an architectural education does not go very deep in physics and acoustics. They all have seen it. Their education is very broad but not in-depth in all areas. And they think in standard building practices.
Note that Architects use acousticians for critical projects.
Off-topic
(but not complete)
The most fun experience I had
(on the net, it was around 2000 I guess) with someone proudly emphasizing, that he didn't need too much explanation since he was an architectural designer referring to his related acoustic education.
He only needed answer on a minor question he couldn't find the necessary data for.
Should he better use additional Homasote or gypsum board to obtain his targeted
130 dB isolation (or was it 120, no I think 130, anyhow somewhere there)? He just came still somewhat short in his actual design. Believe it or not, but that was NO typo. He really meant what he said.
:) Hence I lovely asked if he was aiming at the Nobel prize.
But practical:
1) critical frequency = coincidence frequency which I discussed before, and hence hardly matters.
The whole acoustics world and standards refers to STC and Rw or single number ratings derived from these or yielding comparable results.
Even standard weightings for double leaf systems are mainly defined by the lower frequencies, hence the referred phenomenon is even for traditional office or household applications relative little defining.
In our language we speak about "buigslappe" = bending floppy
(AFAIK no English translation available) walls. This refers to panels with a coincidence (depending on author) above 2000 but mostly > 2500 Hz. This covers standard gypsum boards.
Even in traditional building practices one sees panels with a coincidence > 2500 Hz as a panel where this coincidence hardly or limited matters anymore.
Hence one knows the phenomenon, describes it, but simultaneously tells that when high enough it doesn't or hardly matters anymore.
Hence bothering a problem which isn't a problem, sounds learned but only has sense in few occasions. The idea of a good wall is keep coincidence high enough and MSM low enough.
For music and studio applications it even hardly matters at all anymore since it NEVER (unless you build some sick wall) will be defining in whatever music single number rating, not even close.
Have you ever in your live heard that people complained about the highs and harmonics they were disturbed by resulting from their neighbors music? They hate these low frequent beats and stuff, causing the glasses to travel around in their cabinets.
The coincidence frequency becomes important, when one uses material where this coincidence shifts to an area located in the audio range which is defining for overall TL. Hence it can easier be a problem or limit for cellular concrete, thin brick walls, thick glass panes etc.
2) Mismatching resonant frequencies by combining different thicknesses. This is the other non-coincidence part of this sentence of your quote.
I assume they refer to panel resonances here = panel modes.
Unlike the coincidence where indeed the coincidence (critical) frequency shift by altering thickness panel (effect described above and explained here:
Jeff's explanation of critical frequency ), panel modes are defined by the length/width measures
(and speed of sound) of the panel,
NOT by the thickness. This kind of resonance is related to modes caused by reflection of the waves (in the panel itself) at the edges of the panel (comparable with 2D plane room modes). The only manner to shift this is altering the size (length, width or both) of the panel (again like 2D plane room modes = read e.g. axial and tangential room modes in one plane).
This is ALL stylized and rough, and to be honest I don't feel like responding extensively to every related sentence you encounter, but as you can see what you read was superficial, not you but the text (and for me even partly questionable).
And what you also could do is
checking for yourself in the enormous amount of NRC/IRC measurement data to compare stuff,
http://forum.studiotips.com/viewtopic.php?t=1467
or at the Green Glue site which explains a lot of principles in a rather simple but accurate manner.
The main question if you design whatever is: What is important to obtain my goals, what are the main defining factors?
If you have the courage you can play with this MS Excel file which really can give you the feel what the limiting factors are in function of overall insulation related to specific normalized sound spectra (as speech, music, STC, Rw, OITC, Traffic and so on).
It includes > 50 Australian
Boral gysum wall measurements allowing to check the difference between a lot of TL single number ratings and what the limiting bands are to be improved to increase these overall ratings. You can theoretically improve or decrease any single frequency band and check the effect on the overall insulation.
As such you really can remove, improve or worsen that coincidence dip and see how it influences overall insulation. You'll notice that it hardly will influence overall TL. As such you can improve or worsen individual low frequency bands and see how fast/dominant they will translate in the overall TL
Hence it's very interactive.
http://www.fileupyours.com/files/130337 ... ctor03.zip