"When two layers of material such as gypsum wallboard are glued firmly together, they behave like a single thick layer with an associated lowering of the coincidence frequency.
If the layers are only held together loosely (with screws for example) so that they can slide over each other to some extent during vibration, the coincidence frequency does not move to lower frequencies, and the friction between the layers can introduce some extra energy losses, thereby transforming more noise energy into heat of friction. "
Somehow the lowering of the frequency after gluing doesn't make sense to me. Frequency is proportional to Sqrt(k/m). Stiffness (k) increases with the cube of the thickness where mass (m) is simply linear. So making a panel twice as thick should make the stiffness 8x as stiff (in a perfect case) while the mass only doubles. Even if the real stiffness was only increased to 4x it still means frequency would increase by around 40%.. not decrease. Anybody figure why it would decrease?
I'm not saying gluing is a good idea... I'm just trying to make sense of the Canadian Wood Councils statement. I only bring this up because I'm trying to understand some of this stuff and this runs the wrong way for me. Feel free to turn me around. Thanks. Creatch
That thought actually crossed my mind before...
Have not sorted it out yet, and am not sure if some other factor(s) is(are) missing...
Sounds worthy of more explanation and/or research...
Of course, I'm not sure; ... if I make a comparison to, say guitar strings, the thicker strings have lower frequencies -- even when at the same tension as thinner strings, which have higher frequencies...
Is my thinking 'wrong' here?
But this is wire, which does not hold a linear shape unless under tension and not a monolith wall panel that 'holds it's own shape' without much (if any) tension needing to be added (neverminding here that this wire is also geometrically much different)...
I had originally thought that not gluing was just mainly based on:
a.) allowance of conversion of sound to friction and heat (i.e., rubbing of the separate panels over each other)
b.) allowing the different panels to have different frequencies
Keeping the frequencies lower would help it seems when the frequency is below audible range.
On the other hand, higher frequencies are supposedly easier to achieve isolation from (I guess the frequencies would have to be very substantially increased; an increase of a low frequency from an inaudible range into the low frequency audible range might not be a good thing).
K
Creature82 wrote:
Somehow the lowering of the frequency after gluing doesn't make sense to me. Frequency is proportional to Sqrt(k/m). Stiffness (k) increases with the cube of the thickness where mass (m) is simply linear. So making a panel twice as thick should make the stiffness 8x as stiff (in a perfect case) while the mass only doubles. Even if the real stiffness was only increased to 4x it still means frequency would increase by around 40%.. not decrease. Anybody figure why it would decrease?...
I'm by no means the expert on the "whys" of this, but I am certain on the "whats" -- you definitely do not want to glue two or more panels together, because (I hope I get this right), yes, it does lower the coincidence frequency a little, but that is not a good thing. If you could lower it way down, it would be a good thing. However, what I think it does is lower it into a range that is audible to humans, and more than likely to a spot that is even harder to control.
I believe that's why Paul Woodlock used 7 layers of 1/2" material (drywall and MDF), definitely not glued, rather than, say, 6 layers of 5/8".
Enough "real" experts around here (and at other places) have stated this enough times that I believe it. :)
I realize you guys aren't being argumentative, and just really want to understand the "whys" behind this stuff. :) Hopefully my lame-o pseudo-expert answer is enough to satisfy your curiosity for the time being... Until a real expert comes along and (1) backs me up, (2) backs me up with corrections and clarifications, or (3) tells me to quit trying to answer questions like this. ;-)
--Keith :mrgreen:
Keith -
You're crackin me up. Please keep answering questions! Don't worry... I won't glue panels (except for using GG). I realize that practical experience is tried and true.
Luftweg -
I had a lot of the same thoughts you did. Thanks for replying so I don't feel quite as far out on a limb.
This is by far the best forum I've been to.
Creatch
Here's a morsel from Trevor Cox that may help a bit -
http://www.acoustics.salford.ac.uk/stud ... /noise.pdf
But don't ask me to help with the math, I'm lucky to handle basic trig much less any of those formulae with "squiggly lines" in 'em :?
BTW, the "frequency" that gets lowered when gluing panels together isn't the mass law freq, it's coincidence which is (for drywall, anyway) much higher - gluing causes this frequency to lower, which makes it more difficult for the mass in the partiition to be able to handle the pass-through as well. (See the graph on page 2 of the pdf... Steve
Now, that is a good article (of course it takes several times reading to even start to 'get it' -- and for me, it's several more times reading than that!... but forget that math for the moment; not much need for differentials and integrals in day to day pharmacy! (though, in pharmaceutics, pharmacokinetics, and pharmacodynamics, it's pretty similar to engineering))...
I can't believe at one time I was doing some of those types of equations when I was in engineering school; it's funny how easy you can 'forget'...
knightfly wrote:
Here's a morsel from Trevor Cox that may help a bit -
http://www.acoustics.salford.ac.uk/stud ... /noise.pdf
But don't ask me to help with the math, I'm lucky to handle basic trig much less any of those formulae with "squiggly lines" in 'em :?
BTW, the "frequency" that gets lowered when gluing panels together isn't the mass law freq, it's coincidence which is (for drywall, anyway) much higher - gluing causes this frequency to lower, which makes it more difficult for the mass in the partiition to be able to handle the pass-through as well. (See the graph on page 2 of the pdf... Steve