Avare suggested this document some time ago. Wyle WR 73-5R.PDF (Prepared by Ben Sharp Wyle Laboratories Research Staff, circa 1973) As much as I dislike spoiling the plot, somewhere between the pages 30 and 36 you can find the answer you seek. If your pursuit is to ask have I seen it personally, no Sir, I have not. But the document above suggests I have no reason to look further into it especially based on my limited ability.Does anyone have data that shows a triple leaf makes the LF worse?
double leaf effect cost more than triple in most situations.
Originally posted at johnlsayers.com, topic 13629.
Page 36, figure 18. "A Comparison of the Transmission Loss Provided by Double and Triple Panel Constructions of Equal Total Mass and Overall Thickness." Also the panels referencedhave no mechanical connections.
This is the heart of all of this confusion. I have not been describing equal mass and equal overall depth. I've mentioned adding a third leaf in front of an existing double leaf. That would mean more mass and more overal wall depth than was originally there. Your STC 30 wall compared to the STC 50 wall scenario for example.
1) picture entered by Stuart: viewtopic.php?p=94540#p94540
2) Document linked by Brien: viewtopic.php?p=94553#p94553
Referring to Brien's link 2):
Ted you like slogans don't you? There is no STC 30 wall nowhere + nobody knows the original measurements of that picture 1). Here some calculations towards the MSM. I stylized picture 1) as being gypsum board panels of 12.5 mm (ca 1/2"). The wool thickness 100 mm ( ca 4"), and that empty cavity 75 mm (ca 3"). That's just a rough visual scaling of picture 1) For every next wall I increased that empty 75 mm (ca 3") cavity with the same amount (doubling, tripling, ....) Hence I compared the principle of that wall 2 (STC 36) versus that wall 4 (STC 50) as suggested previously by Ted. That wall 1 is a sic wall which should be filled with absorption if meant for sound isolation properties (even blow-in if wall exists already). The fo of that picture 18 in link 2) relates to the MSM of a double leaf system. The f+ to the fhigh (upper) MSM of the triple leaf system. These real fo and f+ on that picture 18 are located somewhat lower in frequency and deeper on the y-axis scale. What is noted there is where these TL curves intersect with the mass law (but start deeper at the resonance frequency). The basic problem at such resonance frequencies is just that you don't reach the mass-law TL level (serious dip in insulation), hence these curves dive below that mass-law curve. In these calculations you find all kinds of combinations: Triple & Double leaf calculations What do you conclude from these calculations (within the uncertainties of possible unknown influences)?Page 36, figure 18. "A Comparison of the Transmission Loss Provided by Double and Triple Panel Constructions of Equal Total Mass and Overall Thickness." Also the panels referenced have no mechanical connections. This is the heart of all of this confusion. I have not been describing equal mass and equal overall depth. I've mentioned adding a third leaf in front of an existing double leaf. That would mean more mass and more overall wall depth than was originally there. Your STC 30 wall compared to the STC 50 wall scenario for example.
I'm not sure what you're meaning by slogans, Eric. I'm sorry.
Last night at midnight I was referring to the same STC 36 wall I was referring to several times earlier. Same question, same walls.
Correct me if I'm wrong here. I'd very much like to follow your point:
An insulated 2x4 wall roughly. OKI stylized picture 1) as being gypsum board panels of 12.5 mm (ca 1/2"). The wool thickness 100 mm ( ca 4"), and that empty cavity 75 mm (ca 3"). That's just a rough visual scaling of picture 1)
You take "that empty 75mm cavity" and increased it. This would mean removing the drywall layer?For every next wall I increased that empty 75 mm (ca 3") cavity with the same amount (doubling, tripling, ....)
Ted,
I was still editing my post (crossed).
I compared wall 2 (double leaf) with wall 4 (tripple leaf). I estimated that empty cavity a bit less than the thickness of that wool (visual scaling) estimating that empty part 3" and guestimated the wool at 4".
Then I increased that empty space of 3" in steps of 3". (3, 6, 9, 12, ....) and left that wool of 4" a constant.
And further I want to emphasize that this is all stylized stuff.
In my calculations you can find the buildup of the calculated walls (since that's the base on which I calculate them). I just see the cavity as a cavity with normal absorption filling.
No worries Eric. The stylization is great.
So essentially you're starting Wall #2. Then you add a single leaf insulated 2x4 framed wall to create new wall #4.
You assume a 3" gap between the old leaf and the new frame. Decoupled.
Then you progessively moved that new single leaf wall back in 3" increments.
I believe that's what you are saying. And what did you find?
That what everyone is saying is correct. A triple leaf system varies between bad, risky and inefficient unless the cavities become that large (frequency dependent) that the cavity starts acting as a room in itself. The frequency and wavelength dependency is an complicated business not allowing to put a constant figure on this. And therefore, within normal or traditional space availability, cavity widths, uncertain behavior, it's a good thing that people summarize triple leaf systems as to be avoided whenever possible, and only allowed/accepted when forced by non-acoustic arguments. Just look at these combinations and the resulting resonance frequencies, where for the triple leaf systems the higher resonance is (within traditional common circumstances) is defining. The idea of designing walls is to get this resonance far below the to be insulated frequency range of interest (at least a factor 2 to 3).I believe that's what you are saying. And what did you find?
You're saying that the LF performance of wall system #4 is worse than wall system #2
That's what you like to hear, but not what I'm saying since also depending on the internal damping of the construction. But correct is that with a complete undamped system this is true. And that it certainly doesn't improve that wall in a significant manner making it a risky and to be avoided advice to potential customers. I told that I calculated this in a stylized manner. Nature and structures tend to react a bit more complicated. And I don't gamble, just to make things easier to express.You're saying that the LF performance of wall system #4 is worse than wall system #2
Interesting thread. I have respected your sharing of knowledge, particularly on Avsforum for years. It is difficult to accept things that are contrary to what has been beleived for years, especially if one is an expert in the field. Hows about adding a third leaf reduces LF TL and STC? :D It is all in this thread. http://johnlsayers.com/phpBB2/viewtopic.php?f=3&t=11487&hilit=leaf AndreI don't believe inserting a third leaf makes LF worse than before
Oops! Sorry! :oops: ... OK, I just updated the two-graph post with the correct references. I'm not sure where I got that original "several wall" graph from: perhaps SAE? I'll see if I can find the original source. But the link that I posted to Bob Golds page has roughly the same image. Thanks for keeping me honest! - Stuart -(Stuart can you please enter sources of your pictures?)
Ted, that's because it doesn't........ no question that adding the 3rd leaf - with additional air spring - some more mass and insulation will improve the existing 2 leaf (constructed over a single stud frame)...... it will work every time as long as flanking isn't the weaker link. And that is where the potential problems begin....... you cannot predict exactly how the construction will react without a whole lot of testing...... but (keeping it simple): The following data is from IR-761 Start with this:TestID TL-93-157 STC 34 1 single layer of 16 mm type X gypsum board 2 90 mm wood studs at 406 mm on centre 3 90 mm of mineral fibre insulation in cavity 4 single layer of 16 mm type X gypsum board Now look at the next step (I do not have data with the 3 leaf and mineral wool - but the difference is still large: 1 single layer of 16 mm type X gypsum board 2 90 mm wood studs at 406 mm on centre 3 65 mm of glass fibre insulation in cavity 4 25 mm gap filled with CX 5 90 mm wood studs at 406 mm on centre 6 65 mm of glass fibre insulation in cavity 7 single layer of 16 mm type X gypsum board and you will go from the original to the new - TestID TL-93-297 STC 55...... a gain of 11dB averaged. BUT - the question is what do you give up...... if you simply slip that middle to one of the outsides you reach, TestID TL-93-267 - STC 62 - another 11 dB gain just due to placement of materials. Element Description: 1 single layer of 16 mm type X gypsum board 2 90 mm wood studs at 406 mm on centre 3 90 mm of glass fibre insulation in cavity 4 25 mm gap filled with air 5 90 mm wood studs at 406 mm on centre 6 90 mm of glass fibre insulation in cavity 7 single layer of 16 mm type X gypsum board 8 single layer of 16 mm type X gypsum board And now - I will address the OP in a seperate response....... RodAnd thanks once again! We all have heard the ills of triple leaves and the advantage of double. No question on the data. I don't believe inserting a third leaf makes LF worse than before
Andre, I don't think you can necessasarily equate the reactions of a masonry wall as the starting point with the original question which has to do with a frame wall and drywall on both surfaces..... But it is a good example of how the science is not intuitive - and that the only real bet is tried, true and tested....... RodHows about adding a third leaf reduces LF TL and STC? :D It is all in this thread. http://johnlsayers.com/phpBB2/viewtopic.php?f=3&t=11487&hilit=leaf Andre
Hi Rod, thanks for weighing in on this topic. Thank you for the data.
Andre, thank you also for that data. I'm also wondering if a block wall is reasonable to compare in this instance. For example the question in my mind becomes “does a block wall behave more as a single leaf or as a traditional double leaf with traditional framing?" I ask because if we look at this progression:
4” poured concrete critical frequency of 225Hz
6” poured concrete critical frequency of 150Hz
8” concrete block critical frequency of 80-160Hz
Due to the stiffness, it would seem that the block wall behaves similarly to a massive single leaf, at least in terms of the critical frequencies. Additionally the TL curve is roughly linear through 2500Hz, again perhaps behaving similarly to solid mass up to 2500Hz.
So that causes me to ask, "would the addition of RC channel to a solid concrete slab give us similar data as we saw with the addition of RC to the block wall?" Would the curves look similar?
Here is data from the NRC. This was thrown together quickly, so not precisely apples to apples but essentially we’re comparing a single concrete slab to a concrete slab with RC and drywall attached:
External image, not preserved — original: http://www.soundproofingcompany.com/upload/soundproofingcompany-01.gif
Data from the NRC IR-811.
Next I look at the data provided by Andre:
External image, not preserved — original: http://www.soundproofingcompany.com/upload/soundproofingcompany-02.gif
The curves are similar. Above the concrete (block or slab) critical frequency the added leaf provides advantage. Below that critical frequency the concrete (solid or block) alone is superior. As predicted by Mass Law.
So lastly I ask how does a standard coupled double leaf wood framed ceiling behave when a third leaf is added? Starting on page 70 of the NRC’s IR811 study I find some rough comparisons that might help me.
External image, not preserved — original: http://www.soundproofingcompany.com/upload/soundproofingcompany-03.gif
I see that adding a third leaf to the traditional framed double leaf does not de-rate the system, and does not appear to negatively affect the low frequencies, at least down to the available data limit of 50Hz
So I could infer that the addition of a third leaf to an existing double leaf may not reduce the low frequency transmission loss. I might further infer that due to the stiffness of the hollow block, it does not behave as a double leaf, but rather behaves more like a single leaf.
Clearly these triple leaves are a waste, but that wasn't the point.Andy, if you don't mind I am going to stick my 2 cents in here....... The above is spoken like a true contractor - but not an acoustician....... It is very true that blown cellulose is probably one of the highest rated insulations on the market - but it is NOT one of the highest rated isolations on the market............. Insulation has only to do with temperatures - heat loss - solar gain - R values, U values, etc., etc., etc............ And when it comes to isolation - cellulose isn't anything special.......... as a matter of fact you can do just as good (or better) a job with standard fluffy fiberglass........... especially in the lower frequencies........ A few years back I had an interesting visit from someone in the cellulose industry who needed my approval to bid their product as an alternate for the insulation specified in sound walls for a 21 story hotel project - 440+ hotel rooms..... We visited for about an hour - at which time I finally convinced him to release to me the actual test report/results for their product that he was using to claim a superior product......... (interestingly enough I had just spent some time reviewing their product (along with many others) in IR-761, and they are NOT a superior product in 761........ their numbers are not bad per se - but they are not superior........ I almost immediately figured out why their numbers as reported -(and I do not question even for a moment the honesty of their testing lab) were better than those reported in 761........... Because of what I found - I also have serious concerns regarding THEIR moral compass......... all of the testing in ir-761 is well documents - including the fact that their test samples were all 8' x 10'. This (of course) mean they are dealing with real world conditions - i.e.: 90% off all drywall installed has to deal with joints in the body of the wall. Joint and panel length have a direct effect on how rigid the wall assembly is as a whole - and they also introduce weak points in the mass of the wall.......... The cellulose company (on the other hand) only tested wall systems 4' wide by 8' in height...... Their test sample was design to create the most rigid specimen they could provide......... which would automatically move the numbers more favorably in their direction. In this manner a company can tilt the scale in their favor - and seeing as they weren't releasing the data to the general public - who would know? think about it........... Rodtypically it should only cost around $3 sqft for the blown cellulose and keep in mind that the blown cellulose is adding around 3-4 lbs per sqft to your wall and is the highest rated insulation on the market.
Guys, calm down for a second and follow this in a rational manner - Ted is absolutely correct in what he is saying, and the chart proves it...... We think in terms of what is best - and then view how different from the best degrades a wall - and in that respect we also are correct..... But Ted looks at the chart provided and sees a single wall on the left - add insulation to the wall and see an improvement - add a 2nd wall creating a 4 leaf assembly and see an improvement - remove one leaf inside and see an improvememt - ..... Thus his confusion on the statements being made - the chart clearly shows that going from a single simple wall with insulation to building a 3rd leaf would improve the wall's rating (in a perfect laboratory world) from STC 36 to STC 50 - which is an improvement on the original wall. Granted not the improvement that removing that inner leaf would afford (STC 57) but an improvement none the less....... If you begin with a single leaf wall you cannot run the equation as if you began with an isolated frame double leaf (at which point adding the inner 3 leaf is a 7 dB reduction in isolation)........ It all depends on where the construction begins. Sincerely, RodFortunately for acoustics and studio builders, what you "believe" is irrelevant: all that matters is the science. Science works despite what you believe, not because of it. I can't understand how you can possibly even make such a statement, when scientific proof to the contrary is staring you in the face, in the above graphs and publications. Inserting a third leaf ALWAYS makes LF isolation worse, all other factors being equal, - Stuart -I don't believe inserting a third leaf makes LF worse than before
Eric, he is exactly the same person...... I am quite familiar with Ted's credentials........ By the way Eric........ hello my friend - how are you doing (not meaning to hijack the thread - but I miss my old friend)......... RodJust a question: Are you the former colleague of Brian Ravnaas, of am I mistaken here?
"Ted is absolutely correct in what he is saying, and the chart proves it...."
No he isn't and I haven't seen the chart that proves it.
Initially this is the phrase "Does anyone have data that shows a triple leaf makes the LF worse?"
Home schooled 101, poor question designed to get poor answers.
Rod, Thanks for this lovely inquiry. But as you likely expect that's difficult stuff for me.By the way Eric........ hello my friend - how are you doing (not meaning to hijack the thread - but I miss my old friend).........
Don't confuse critical frequency (referring to coincidence) with Mass-Spring resonance. Indeed concrete blocks and glass bricks behave as single leaf (hollow or not)The curves are similar. Above the concrete (block or slab) critical frequency the added leaf provides advantage. Below that critical frequency the concrete (solid or block) alone is superior. As predicted by Mass Law.
You call it a traditional double leaf, I call it a sick double leaf. That original curve is dominant defined by structural resonance, and the added panel by 2 things: 1) far to the left of the resonance around 200 Hz (you have a cavity of only 13 mm) these 2 panels start moving in phase (restoring themselves to the mass law of both panels together, meaning that these panels further below the MSM start acting as a single panel as if that cavity hardly exists = standard behavior) and 2) the resonance itself is strongly damped to the extreme small cavity (visco-thermal acousto-elastic interaction - see Tom Basten ISBN 90-365-1597-1) Wat I do notice is that on the same pages in this report you refer there are combinations wih more logical cavities where the phenomenon becomes more clearly. These you don't speak about. As long as the others don't get confused here it's OK for me. I also agree with xSpace: wat you see on that typical picture are STC values. A higher STC value is a higher STC value, nothing else. For years people are telling that STC isn't representative for low frequency content, or studio use. For me that picture has no other value than showing the idea of MDOF systems (Mass-Spring systems with Multiple Degrees Of Freedom). That's a typical picture once taken from a UK site (which is afterwards removed by them) and that went into the traditional copy/past net procedure, where giving sources are more of a sin than a virtue. Mass-spring principles applying on STC values based on measurements where structural resonance is, or can be, a dominant factor is like comparing apples with oranges. Ted, why do you think I'm so careful in expressing myself physically, if you try to get me to confirm simplified statements/conclusions?So lastly I ask how does a standard coupled double leaf wood framed ceiling behave when a third leaf is added? Starting on page 70 of the NRC’s IR811 study I find some rough comparisons that might help me.External image, not preserved — original: http://www.soundproofingcompany.com/upload/soundproofingcompany-03.gifI see that adding a third leaf to the traditional framed double leaf does not de-rate the system, and does not appear to negatively affect the low frequencies, at least down to the available data limit of 50Hz So I could infer that the addition of a third leaf to an existing double leaf may not reduce the low frequency transmission loss. I might further infer that due to the stiffness of the hollow block, it does not behave as a double leaf, but rather behaves more like a single leaf. Clearly these triple leaves are a waste, but that wasn't the point.
An Example: http://www.nrc-cnrc.gc.ca/obj/irc/doc/p ... c44692.pdf A Simple model of the sound insulation of gypsum board on resilient supports Bradley, J.S.; Birta, J.A. NRCC-44692 Look at fig. 2 on the 3rd page (p 217) 3 walls are build-up to test possible differences between 3 different RC models. That bad curve is the same wall based on wooden studs without any RC. What you see are structural resonances and flanking. With such a wall you can try to stylize the problem to a mass-spring resonance (MSM). But it has hardly sense since you're looking, from an acoustical point of view, to a sick wall build-up dominated by whatever other phenomena. These 3 RC versions you can apply some simplified reasoning, that acoustically bad version you can, as a matter of speech, make and end-work of it. If Green Glue, Brian tells that doubling the panels on a wooden stud wall only gives a couple of dBs then that is because he compares acoustically 2 bad walls dominated by other phenomena than the doubling of the mass/panels on both sides.Mass-spring principles applying on STC values based on measurements where structural resonance is, or can be, a dominant factor is like comparing apples with oranges.
Thanks Rod. I was giving an example of where adding a third leaf reduces LF TL. As far as the original question, or statement goes, I can not figure out exactly what the OP meant. As far as the suitability of a CMU wall for triple leaf effect goes, mass is mass. CMU units act acoustically for the most part as mass, not a double leaf. This is evident from the TL curve for the bare wall. Here is an example from IR-766 pdf pages 124 and 125. It is a floor system, but the material densities are closer to walls. Examples 1 and 4 use identical material, but in different ways. The design with the third leaf has an STC 17 points lower than the one without. Unfortunately I have not been able to find the TL data for the tests. Of course the usual STC applicability qualifier is emphasized. As an edit because I posted the above earlier, IR-811 has the TL data on pdf page 72. AndreI don't think you can necessasarily equate the reactions of a masonry wall as the starting point with the original question which has to do with a frame wall and drywall on both surfaces.....
xSpace, the chart I am referring to is the one posted by Stuart...... take a look at it please...... In the most simple of forms........ if you begin with a single stud double faced wall - and you build a wall adjacent to it - adding another face - (all things being equal)this will increase your isolation - not decrease it. Which is the point Ted is trying to here here..... The fact that you can add some fairly drastic additional isolation by taking just that 3rd leaf and relocating it to one of the outmost wall surfaces is not relevant to the point he is making. I always tell people to avoid the 3 leaf whenever they can - but when they can't - then just live with what they have - and overcome the obstacle with mass........ 3 leaf walls are not an insurmountable problem, and they cannot ALWAYS be avoided. When they can't be avoided this doesn't mean that the people who own the property are automatically screwed and will never acheive decent (or even great) isolation - but rather, they are just going to have to work a little harder (and smarted) to acheive their goals.. Sincerely, Rod"Ted is absolutely correct in what he is saying, and the chart proves it...." No he isn't and I haven't seen the chart that proves it.
+1. The biggest problem, as I see it, is people taking as absolute what are generalizations, without knowing the factors involved. I think this is about the fourth time I am writing this week that "acoustics is not intuitive and requires detailed knowledge to do properly and cost effectively." AndreI always tell people to avoid the 3 leaf whenever they can - but when they can't - then just live with what they have - and overcome the obstacle with mass........ 3 leaf walls are not an insurmountable problem, and they cannot ALWAYS be avoided. When they can't be avoided this doesn't mean that the people who own the property are automatically screwed and will never acheive decent (or even great) isolation - but rather, they are just going to have to work a little harder (and smarted) to acheive their goals..