Document Library

A place for general chit-chat, hanging out, off topic stuff, shooting the breeze, ranting, introducing yourself to the forum, saying "Hi!", etc.
terminator1987
New Member
Posts: 6
Joined: Tue, 2020-Jun-09, 20:36
Location: Poland

Document Library

#1

Postby terminator1987 » Wed, 2020-Jun-10, 11:46

Dear Stuart,

this is my first post, so I would like to thank you for your great commitment to sharing knowledge, myth-busting and explaining extensively to people who ask, so - kudos! I learned a lot from the documents you posted here, as well as from your input in the Gearslutz forum in the topic on the Cardas' "golden trapagon". As the author of the topic and his girlfriend, me and my wife are very grateful for your rational discouragement of the pursuit of "golden trapagon" along with other dragons and unicorns.... :roll:
We are planning to build a house in the near future and I want to design a decent home theater room within, which would be also great for occasional wrestling with my guitars without disturbance to other housemates and neighbors :jammin: :shot: , no matter the time, so highly soundproof, but also acoustically impressive - that's what I'm researching upon.

Now ON THE ACTUAL TOPIC here:
I find one of the documents you posted here redundant and misleading, namely "The Acoustic Role of Glass Wool in Double-Leaf Lightweight walls" by the Acoustic Working Group, 1999 about cavity fill - insulation inside isolation walls.
Problems with this paper are as follows:
1. The most fundamental issue with the article is that it seems to suggest that the best level of sound absorption is achieved with fiberglass batts of air resistivity of ca. 5 kPa*s/m2, which is simply not true, according to other documents I've found and you've posted;
2. Although it refers to some studies such as "Swedish" or "German", the document gives no actual references, and reproduces no graphs with actual measurements;
3. Everything relates to STC rating, which in itself is inadequate to the topic (even the original description of the standard in '1960s recognizes it is only relevant to speech and radio/television - and that of the '60s, mind you), as it doesn't adress transmission of sounds below 125 Hz;
All in all, the conclusion of the article that the full cavity insulation with fiberglass batts is the most cost-effective is somewhat true, meaning that for greedy developers (that care about marketing and meeting the construction code) it is the best way to boost STC, regardless how (un)happy that would make future owners.

IMO, good soundproofing of noise < 125 Hz is of crucial importance in both studio and home theater setups. A much better sense of what's going on (and backed by actual data) can be derived from Appendix C (page 73 and subsequent) of your article IRC-IR-693 "Sound Transmission through Gypsum Board Walls: Sound Transmission Results" by Quirt et al., NRC Canada, 1995.
In Fig. 1 (page 74) you can see that the material marked as GFR4 yields best results both in the critical low-end spectrum < 125 Hz as well as overall, giving it the highest STC rating of the bunch. Further, from Figs. 2a & 2b we can read that GFR4 stands for a Glass Fiber Rigid board with density ca. 80 kg/m3 and airflow resistivity of ca. 50 mks krayl/m (and as far as I could tell, the units in all articles are directly comparable i.e. 1 mks krayl/m = 1 kPa*s/m2 = 1 g/s/cm3). Moreover, we see a clear strong logarhythmic correlation (logarhythmic because dBs are logarhythmic in itself) between airflow resitivity and sound transmission loss (STL) at 1 kHz for all tested materials. However, as we saw in Fig. 1, <125 Hz only GFR4 performs tangibly better than other materials. I still don't fully understand the physics behind it, but the strong statement from the article in question that no increase in airflow resistivity above 5 kPa*s/m2 seems simply false and fictitious in this context. You can also notice that mineral fiber batt performs slightly better than glass fiber batt, yet glass fiber rigid boards were far superior, while mineral fiber rigid boards weren't included (!).

Also to make 3:1 I found two other articles that somewhat confirms my thesis and contradicts that of the Acoustic Working Group.
The first article "Sound Absorption and Insulation Functional Composites" by Peng, Chinese Academy of Forestry, Beijing 2017 (https://doi.org/10.1016/B978-0-08-100411-1.00013-3) deals in mathematical detail with some specific aspects of sound absorbtion and insulation, but also delivers some general rules about porous absorption. On page 344 and subsequent we can learn that:
a) "The sound absorption property is improved with the increasing resistivity of fibrous material, while it decreases when the resistivity gets over a certain value. If the airflow resistance is too small, the acoustic energy attenuation caused by internal friction is minor and the absorption effect is poor. If the resistance is too large, most of the acoustic waves are reflected and the absorption becomes weaker. The sound absorption curves move toward low frequency with increasing resistivity. As for low airflow resistance materials, absorption coefficient at low frequency is low, and it increases sharply at medium and high frequencies. Compared with low airflow resistance materials, the absorption coefficient of high-resistance materials in high frequencies is decreased, and the absorption coefficients at low and medium frequencies are increased. Airflow resistivity of a fibrous porous material is related to the fiber morphology, size, density, porosity, tortuosity, and arrangements."
This is in part congruent with the questioned article (that above certain level of airflow resitivity we get more reflection than absorption), but also clearly states shift of performance from mid-high to low frequencies with increased resistivity, as well as that the parameter depends upon many variables at the microscopic level (I think that might be actually crucial).
b) "The thicker the material, the longer the transmission path, and more acoustic energy is attenuated. The absorption peak value often occurs at the fourth wavelength. With an increase of thickness, the average absorption coefficient is improved and the peak value moves toward low frequency. However, it is impractical to improve the absorption performance by increasing thickness."
I only have problem understanding the last sentence - I mean, of course it's impractical for standard housing, but if it is probably the most powerful tool, then for no-compromise scenario like building an expensive studio that should be the first thing to do, is it not?
c) "At a certain thickness, the increase of density can improve the absorption performance at low frequencies. However, the improvement is less than that by increasing thickness. [...] The influence of density is complex, which is also affected by the morphology of fibers, porosity, and airflow resistance."
d) "Larger porosity means more interconnected pores inside the material and larger specific surface area. There is more internal friction between air and fibers, resulting in higher sound absorption coefficients."
e) "There usually is an air cavity of certain thickness between a porous absorption material and rigid back in applications. The reflected waves by rigid back and the incident wave form a phase difference of 180 degree. The absorption material yields the optimized performance when the thickness of the air cavity is integral in multiples of onefourth incident wavelength. On the contrary, when the air cavity thickness is integral in multiples of one-half incident wavelength, the incident waves are overlaid with reflected waves and the absorption performance is the worst."
And that's a real bummer because as far as I could calculate, the air cavity required to treat noise around 80 Hz requires ca. 1 meter (40"), which in turn would give the worst performance of absorption around 40 Hz, which is already probably satysfyingly low, but it would be best to go down to 60 Hz and 30 Hz respectively, which would require an absurd nearly 1.5 m (60") of air cavity - THAT is impractical! :ahh:

Another paper I found is from Delany and Bazley, British National Physical Laboratory, Teddington, 1970 (https://www.math2market.com/Publication ... Bazley.pdf).
From Figs. 7-9 we can tell that in their set, the material with airflow resistivity od 20 (and not 5) had the best overall performance, however material with the airflow resistivity of 50 (1 g/s/cm3 = 1 mks krayl/m = 1 kPa*s/m2) had double the performance in the lower end (extrapolated) than other tested materials, while compromising some midrange efficiency.
Also, Fig. 5 is very interesting. It shows the normalised relationship of the absorption/reflection ratio with the frequency/airflow resistivity ratio on a logarhythmic scale (but note that's for semi-infinite layer of material). The crossover of the curves occur at ca. 60% absorption and 40% reflection for frequency/airflow resistivity ratio of ca. 10. That would translate to 60% absorption and 40% reflection at 50 Hz for semi-infinite layer of 5 kPa*s/m2 material, while at 500 Hz for semi-infinite later of 50 kPa*s/m2 material. However, (1) semi-infinite layers are definitely impractical, (2) data for lower frequencies are only extrapolated and some other empirical papers you provided indicated that probably different mechanisms govern transmission of soundwaves below and > 200 Hz.
Therefore probably the best practical porous material for low frequency and overall sound absorption has airflow resistivity of somewhere between 20 and 100 (?) kPa*s/m2 and some serious thickness.

Now, an additional remark from me - all those considerations and measurements are actually done with relatively lightweight two leaves of gypsum drywall or in some cases even 3 mm thick plastic. I am not sure how this would translate into a setting of really heavy leaves, because the energy that already passes through such a leaf (a brick wall I mean) would in my (not really substantiated, but intuitive) opinion require some dense and resistive porous material to be stopped with. Additionally, even if for particular frequency a particular material is mostly relfective (lets say 80% reflective, 20% absorptive), then it would still gradually be dealt with by internal reflections within the leaves (brick walls). Furthermore, I don't really think that reflection + absorption = 100% of energy for the low frequencies. It is probably something more like 30% reflection + 20% absorption + 50% passthrough (again, my own unsubstantiated conviction). Would you agree?

To sum up - that one article from AWG, 1999 creates confusion and misinformation, and has no added value here IMO.

P.S.
My idea on near-perfect soundproofing from what I've read so far would be as follows (layers from outside to inside):
1) a 19 cm calcium silicate brick wall (density!) with gypsum board as finishing attached to it without any airspace (to avoid triple- or quadruple-leaf effect)
2) 30 cm airspace filled with non-compressed mineral rock wool rigid boards
3) a 12cm calcium silicate brick wall decoupled from the other (alternatively some more porous bricks or blocks for added absorption) finished with gypsum board without any airspace (as above, alternatively some wet porous finishing directly on bricks, possibly even finished with one of those acoustic paints)
4) drywall ceiling (concrete slab + hanging gypsum, mineral rock wool in between)
5) serious floating floor (high density concrete slab of at least 50 mm floating on a 10mm NPE polyethylene foam, decoupled from the walls, finished with most probalby glued vinyl tiles)
6) acoustic doors on both walls, no windows, separate HVAC airduct and floor heating ducts, insulated ducts
5) treatment: lots of serious bass-traps DIY'd from acoustically certified mineral rock wool, expecially in all of the corners and some strategically placed quadratic diffusers

What you think of such design?

Thanks

Paul



User avatar
Soundman2020
Site Admin
Posts: 889
Joined: Thu, 2019-Sep-19, 22:58
Location: Santiago, Chile
Contact:

Document Library

#2

Postby Soundman2020 » Wed, 2020-Jun-10, 19:07

Hi Paul, and Welcome to the Forum! :thu: :)

I moved your post to the Green Room, since it isn't really part of the document library, even though it is about the document library! So it's here in the Green Room now, which is more appropriate. So I'll just put in a link back to the library, for easy reference, so others can find the document you are talking about: viewtopic.php?f=9&t=4

...in the topic on the Cardas' "golden trapagon".
Ahh yes! One of my favorites acoustic myths!

me and my wife are very grateful for your rational discouragement of the pursuit of "golden trapagon" along with other dragons and unicorns...
:) There certainly are a lot of mythical creatures out there on the dark side of "acoustic" web sites....

We are planning to build a house in the near future and I want to design a decent home theater room within, which would be also great for occasional wrestling with my guitars without disturbance to other housemates and neighbors , no matter the time, so highly soundproof, but also acoustically impressive - that's what I'm researching upon.
Excellent! Then I think you'll find you are in the right place here. It would be great if you could start a thread in the "Recording Studio Design" area, with details about that.

I find one of the documents you posted here redundant and misleading, namely "The Acoustic Role of Glass Wool in Double-Leaf Lightweight walls" by the Acoustic Working Group, 1999 about cavity fill - insulation inside isolation walls.
Problems with this paper are as follows:
To be fair, that is just an extract from the full paper, and is incomplete. It is only "Chapter 1". I wish I still had the complete document to post in full, with all the references and diagrams, but that's all I have right now. If someone else has it, then please do send it to me, so I can post it in full.

1. The most fundamental issue with the article is that it seems to suggest that the best level of sound absorption is achieved with fiberglass batts of air resistivity of ca. 5 kPa*s/m2, which is simply not true, according to other documents I've found and you've posted;
That's only partly right: What they actually said is that there's no significant improvement for higher GFR numbers. Not really that 5k rayls is the optimum.

2. Although it refers to some studies such as "Swedish" or "German", the document gives no actual references, and reproduces no graphs with actual measurements;
See above comment on the full document: I'm pretty sure there were several graphs and a bibliography in the original complete version. The information you are looking for was probably in those.

3. Everything relates to STC rating, which in itself is inadequate to the topic
Yup. I wrote this a while back: Why STC is not a good way of measuring studio isolation. That said, way too many suppliers of acoustic "soundproofing" products do still specify the performance in STC... Go figure! Presumably that's because architects and contractors are more familiar with the STC system... :roll:

All in all, the conclusion of the article that the full cavity insulation with fiberglass batts is the most cost-effective is somewhat true, meaning that for greedy developers (that care about marketing and meeting the construction code) it is the best way to boost STC, regardless how (un)happy that would make future owners.
There's actually very good evidence from many sources that completely filling the wall cavity does, in fact, given optimum isolation... (as long as you don't over-fill it, to the point where the isolation is bridging the leaves, creating flanking paths). The difference, as they point out in that paper (and elsewhere) can be as much as 16 dB better isolation, as compared to an an empty cavity. Partly filling a deep cavity leaves an empty air gap, in which resonances of various types can form, ribbing the wall of isolation. In addition, there is no decrease in the speed of sound for the empty air part of the cavity (that only happens inside the insulation), and there is also no change from adiabatic to isothermal for the "empty air" part: once again, that only happens within the insulation. So there is certainly a real acoustic advantage to completely filling the cavity.... even though it costs more. The real question here, is where to draw the line for cost/benefit. For those who need extreme isolation and have the financial resources, yes, it makes sense to fill every last millimeter of the cavity. If your isolation and/or budget are more modest, then it might be acceptable to go with less than 100% cavity fill. Sometimes, acoustic benefit has to take a back seat to economic reality.

A much better sense of what's going on (and backed by actual data) can be derived from Appendix C (page 73 and subsequent) of your article IRC-IR-693 "Sound Transmission through Gypsum Board Walls: Sound Transmission Results"
That very same page 73 has an interesting statement on it:
absorption-fill-conclusion-from-IR-693.jpg


In Fig. 1 (page 74) you can see that the material marked as GFR4 yields best results both in
Well, yes, but you seem to be comparing apples to oranges. The explanation right above that graph states how it was obtained: "We used a specimen with rather light and flexible panels, to avoid resonance in the frequency range of concern. The specimen had surfaces of 3mm Lexan plastic....". Thus, any conclusions one could draw from that graph would be applicable to walls built with 3mm Lexan. The "frequency range of concern" also seems to be around 1 kHz, which isn't exactly applicable to studio isolation (the only time you might have 1 kHz. isolation issues, is if you maybe had a very serious coincidence dip around there, from poor choice of materials). Even then, there is only 3 dB difference across the entire range of products test like that, and they were all 100mm thick in 150mm cavity. Here's the actual graph and text:
IR569-p74-graph.jpg

...the material marked as GFR4 yields best results both in the critical low-end spectrum < 125 Hz as well as overall, giving it the highest STC rating of the bunch.
True, yes, but I certainly would not design a studio isolation wall that had such lousy performance in the low end! Where there is such a huge MSM resonant dip at 125 Hz, then there's something pretty bad wrong with the wall design. In this case, fully explained by the text itself: 3mm Lexan panels on a 150 mm cavity will, indeed, have MSM resonance at 125 Hz. So I'm not sure that "good" performance in this case is valuable information for extrapolating to typical studio isolation walls, where the MSM resonance should be three or hopefully four octaves lower.

To be honest, I would not try to draw conclusions from that graph, that could be used for studio walls.

"The thicker the material, the longer the transmission path, and more acoustic energy is attenuated. The absorption peak value often occurs at the fourth wavelength. With an increase of thickness, the average absorption coefficient is improved and the peak value moves toward low frequency. However, it is impractical to improve the absorption performance by increasing thickness."
I only have problem understanding the last sentence - I mean, of course it's impractical for standard housing, but if it is probably the most powerful tool, then for no-compromise scenario like building an expensive studio that should be the first thing to do, is it not?
I have not read the entire book, so I can't comment in detail on what the authors might be referring to there, but I'm assuming they are referring to their own comment: that the absorption peaks at the fourth wavelength. For a typical studio isolation wall, where the MSM resonance is hopefully less than 15 Hz, the wavelength is therefore about 23 meters. Four times that is 92 meters (about 300 feet). Building a wall with a cavity 300 feet deep seems a little overboard! :) My assumption is that they are talking about much higher frequencies (several kHz.)... or they are just pointing out the impracticality of using the peak absorption point to determine cavity depth.

The best method for determining wall cavity depth (and thus insulation thickness) is to first determine how much isolation your studio will need, and what frequencies it will need it at. Based on that, there are equations that can be used to predict the performance of any given wall design. The MSM resonance is determined mostly by the mass of the two eaves, the depth of the cavity, and the degree of insulation infill. The simplest equations are fairly accurate for most home studio and project studio design purposes, but there are more precise versions if you need greater accuracy, as they take into account more parameters. To be very honest, I seldom bother with the high precision equations for isolation, as the simpler ones are fine for typical studio needs. The simpler ones don't even take into account the GFR of the infill: all the consider is "infill" or "no infill". Yet the results they produce are valid, and sufficiently accurate for most cases.

From Figs. 7-9 we can tell that in their set, the material with airflow resistivity od 20 (and not 5) had the best overall performance,
Take a look at all of those graphs: the bottom end is at 250 Hz! There is no low-frequency data at all. Not even low mids! The paper itself mentions that these are impedance tube measurements, and thus are only valid for normally incident plane-waves. In typical studios, even small ones, you'll only find normally incident plane waves on the rear wall, and in the very low end. Ditto for wall cavities: some forms of resonance will, indeed, be plane waves, but several others won't. So I'm not sure about how this paper would be applicable to studio isolation walls.

The most interesting graph in that paper, is actually Fig 10, which shows the GFR for several different porous absorbers. I thought it was quite fascinating that the two lines at the far extremes of that graph, are both for fiberglass! "D" is for "fiberglass aircraft insulation", while "B" is for "Fiberglass, resin-bonded". The point being that there's nearly two orders of magnitude difference between the density of those materials, yet both of them have useful GFR properties, in the typical range for both absorption and treatment in studios. So the underlying issue is: you can't generalize about materials, even if they are similar! Those are both "fiberglass" products, yet with rather different characteristics.

Therefore probably the best practical porous material for low frequency and overall sound absorption has airflow resistivity of somewhere between 20 and 100 (?) kPa*s/m2 and some serious thickness.
On the other hand:
OC-703-specs-absorption-coefficient.jpg
GFR for 703 is around 16k rayls.

If you take a look at Bob Gold's page on numerous different types of porous absorption, I think you'll notice that this entire subject is far more complex than can be summarized by a certain range of GFR, or density. There are many, many products with different characteristics that can be used in studios, for different purposes. I sometimes need very light (low density) insulation as part of treatment devices (lower even than 5k rayls), and I sometimes need much heavier stuff for other applications. There's a broad range of products, and a broad range of needs: the trick is to match the best product to each need, and sometimes that's not so easy.

But getting down to brass tacks (as the saying goes): if you are looking for the best insulation for your up-coming studio walls and ceiling, then very probably you would find it hard to beat Rockboard 40, or maybe Rockboard 60, if you can find a source for them. But even that recommendation comes with caveats! :)


I am not sure how this would translate into a setting of really heavy leaves, because the energy that already passes through such a leaf (a brick wall I mean) would in my (not really substantiated, but intuitive) opinion require some dense and resistive porous material to be stopped with. Additionally, even if for particular frequency a particular material is mostly relfective (lets say 80% reflective, 20% absorptive), then it would still gradually be dealt with by internal reflections within the leaves (brick walls). Furthermore, I don't really think that reflection + absorption = 100% of energy for the low frequencies. It is probably something more like 30% reflection + 20% absorption + 50% passthrough (again, my own unsubstantiated conviction). Would you agree?
It seems you are looking at the way walls isolate, from the wrong perspective: it's not about how much each material absorbs, transmits, or reflects. Rather, it is about the wall as a system. It's not so much the behavior of the individual parts that matters, but rather how they behave as a system. In fact, there are at least four key aspects to how a wall isolates, an each one is dominant over different frequency ranges:
four-regions-of-isoaltion--transmission_loss-BGR.jpg

In the very low end, below MSM resonance, isolation is controlled mostly by the stiffness of the wall. Around the resonant frequency, it is the resonance itself that controls isolation. Above resonance, it is mass that controls the isolation, up to the coincidence dip, where it is one again a resonant issue. And beyond that, it is back to mass law again.

Thus, the amount of isolation you get in each part of the spectrum depends on different characteristics of the wall. But the one key factor to all of this, is MSM resonance. If you design your wall such that the MSM resonant frequency is at least an octave below the lowest frequency you need to isolate, then you don't need to worry about the first tow regions, since they will be off the bottom end, and it becomes basically a matter of mass and coincidence. And considering that coincidence isn't usually a problem in typical studio walls (with the possible exception of windows in the walls), it all boils down to one major issue: mass. The more mass you put on the wall, the better it will isolate. So the key issue in isolation wall design, is to drive the MSM resonance down low enough that you can ignore all the other aspects, and just concentrate on mass. If you tune your wall right, with f0 more than an octave down from your lowest isolation need, then you will be fine.

To sum up - that one article from AWG, 1999 creates confusion and misinformation, and has no added value here IMO.
Well, I wouldn't agree with that. The only part I have of the article is incomplete, but the conclusions are valid, and the most important conclusion there is that filling the entire cavity with suitable insulation is very effective. It helps to drive down the MSM resonant frequency in several entire ways at once.

My idea on near-perfect soundproofing from what I've read so far would be as follows (layers from outside to inside):
1) a 19 cm calcium silicate brick wall (density!) with gypsum board as finishing attached to it without any airspace (to avoid triple- or quadruple-leaf effect)
2) 30 cm airspace filled with non-compressed mineral rock wool rigid boards
3) a 12cm calcium silicate brick wall decoupled from the other (alternatively some more porous bricks or blocks for added absorption) finished with gypsum board without any airspace (as above, alternatively some wet porous finishing directly on bricks, possibly even finished with one of those acoustic paints)
4) drywall ceiling (concrete slab + hanging gypsum, mineral rock wool in between)


What you think of such design? [/quote]I didn't do any math on that, so I'll just ask instead: What is the MSM resonant frequency of that system? And how much isolation are you predicting, in decibels? Also, what is the purpose of the drywall over the outer-brick wall? It adds no useful mass, and I can't see what other purpose it would serve. Ditto for point #3: what is the purpose of having a thin sheet of drywall over a high density brick wall?
Also, I'm a little confused by number 2 above: you just spend many paragraphs dissing the concept that filling the entire cavity is a good thing, the your proposed system has the entire cavity filled? :shock:
For point #4, you don't mention how you would prevent the inner-leaf walls from touching the outer-leaf ceiling, and you don't mention how you would get enough mass on the inner-leaf ceiling to match the mass on the inner leaf walls, nor how you would go about decoupling the inner-leaf drywall ceiling from the outer-leaf concrete slab ceiling.

5) serious floating floor (high density concrete slab of at least 50 mm floating on a 10mm NPE polyethylene foam, decoupled from the walls, finished with most probalby glued vinyl tiles)
Why? :) What would the "10mm NPE polyethylene foam" do here, and what would the resonant frequency of that system be? Is it low enough? But an even more basic question is: Do you even NEED a floating floor? You should probably take a look at this before you make that decision: Floating your floor: How and why... and why not.

6) acoustic doors on both walls,
Done like this, maybe: site built door for high isolation

no windows,
You can have windows if you want them: site built windows for high isolation

separate HVAC airduct and floor heating ducts, insulated ducts
Another interesting article that might help you decide how to do the HVAC: why your studio needs proper HVAC.

5) treatment: lots of serious bass-traps DIY'd from acoustically certified mineral rock wool, expecially in all of the corners
I would not use mineral wool for bass trapping. Some people do, but I have a more effective system...

and some strategically placed quadratic diffusers
Why? For what purpose? Where would you place them, and what frequencies would you tune them too?

What you think of such design?
I think you should probably start using some math on your assumptions, to check that you really would get the results you are hoping for, in each case. Often in acoustics things that seem nice at first glance, don't actually work out that way on closer inspection. Many things about acoustics are not intuitive, and work differently from what one would expect, logically. Part of the problem is that we can't see air or sound waves with our eyes, yet our brains are mostly visual, so we try to extrapolate for things we can see, and start imagining how sound behaves based on that... but it doesn't actually behave that way! Things that seem right, or look right, often don't sound right at all. Here's some comments from another new member of the forum who already went down that path, and got burned... viewtopic.php?f=6&t=690 Well worth reading!

But math doesn't lie: the equations will give you the right answers that you are looking for

In your case, your studio is mostly for use as a home theater, so it has some rather different needs than what a typical home recording studio would need, in both isolation and treatment. You should take that into account in your design.

And of course, for the majority of home studios and home theaters, there's an even more important, overriding issue: money! Getting high isolation and fantastic acoustics isn't cheap.... budget often gets in the way...

There's one final thread I'd suggest for you, done by someone who, like you, needed very high isolation for his place. It's the story of how we did that, and how it is going so far (he is nearing completion, but not quite there yet...):
Small studio with high isolation: how to build one.
Tom's thread is an eye-opener. This is what you need for high isolation.


- Stuart -



terminator1987
New Member
Posts: 6
Joined: Tue, 2020-Jun-09, 20:36
Location: Poland

Document Library

#3

Postby terminator1987 » Thu, 2020-Jun-11, 08:21

Thank you for your quick and vast reply.
Before I reply more in detail please let me explain two things:

1) I am at too early stage of my vision to go into that in detail; I am very thankful for your initial thoughts, insights and comments - I've got a lot to learn and think about, when the project will be a little closer to realisation (i.e. actual project) I am going to start a new topic on that in the correct section as you suggested. So I won't be refering much to that now, because you gave me a lot to read, think about and calculate, I will just reply to moments were I feel you misunderstood me.

2) Apparently I was vastly unclear what I am actually critical about the document in question, so I will focus on that in the reply. To begin with, please let me give you the long story short:
100% cavity fill is great, but one can infer that readily from other documents, especially Appendix C from Quirt et al., NRC, 1995. AWG, 1999 gives in my opinion NO ADDED VALUE, seems to target different audience (general developers not studio builders), to focus on inadequate endpoint (STC), and stripped from original references simply looks bad. Yet, being short it is tempting to read it first (as I did), which may give one disputable impression that fiberglass batts are THE BEST way to go, and that's what I was trying to dispute, the whole GFR thing. The only ADDED value of the article seems to be the GFR cutoff value of 5 mks krayls/m and that's at least disputable. Maybe the science is settled on that issue, maybe not - I know what I have found gives no definite answer to that, and I do believe that the microstructure of a particular product (and not one physical property like GFR) determines the acuostical performance. So I would like to see more data on actual products in the interesting range of frequencies <250 Hz, because I feel that above that basically every porous material thicker than a few centimeters performs geat.

Soundman2020 wrote:Source of the post To be fair, that is just an extract from the full paper, and is incomplete. It is only "Chapter 1"

Well, I understand. Still looks bad to me with strong statements, no references and little data.

Soundman2020 wrote:Source of the post That's only partly right: What they actually said is that there's no significant improvement for higher GFR numbers. Not really that 5k rayls is the optimum.


Well, in one sentence they did, and that's were I infer that the target audience of the article are general developers which are of course more interested in price/performance ratio when building a residential area, than studio builders who would often pay 20% more for 10% improvement, even I would in such difficult area to treat as the low end. Maybe the thing is that in Canada fiberglass wool is much cheaper than mineral rock wool ? Well, in my country it is about 10-20% cheaper and actually dependent on the supplier it can even be reverse. So that's also what I am sceptical about - they are giving economical advice to your international audience without any $ they calculated upon. Now, please look on some other exerpts from that text and tell me they don't effectively say fiberglass is THE BEST:
"Glass wool rolls and batts are very effective at causing the resonance frequency of lightweight cavity walls to shift lower, resulting in higher sound insulation."
"Glass wool rolls and batts will give the most economical way of maximising acoustic performance."
"Filling the cavity with glass wool rolls or batts reduces the negative influence of standing waves and results in the best overall acoustic performance."
"The results show that the air flow resistivity has a positive impact on the sound reduction index, but no additional improvement is achieved when the r-value is higher than 5 kPa.s/m2, which corresponds to that of a lightweight glasswool rolls and batts."
"It is imperative that this cavity is completely filled with low density glass wool in order to achieve the maximum benefit from the insulation."
"Increasing the density of the insulation above that achieving an air flow resistivity r ~ 5kPa.s/m2 does not improve the acoustic performance of the double-leaf wall."
"Total filling of the cavity with leightweight glasswool rolls or batts is the most economic way to achieve high sound insulation values compared with larger cavities or extra layers of gypsum board."
"Hence, we can summarise CHAPTER 1 by stating that not filling cavities of lightwieght walls with an adapted glass wool (r ~ 5kPa.s/m2) means not taking the advantage of the most economical way of maximising acoustic performance."

Sorry, but even the marketing leaflets tend to be less explicit about the products they praise, and that looks bad.

Soundman2020 wrote:Source of the post There's actually very good evidence from many sources that completely filling the wall cavity does, in fact, given optimum isolation... (as long as you don't over-fill it, to the point where the isolation is bridging the leaves, creating flanking paths).

I never disputed that ! Completely agreed. What I meant was that noises <125 Hz may (or probably will) still make people unhappy and those were out of focus there.

Soundman2020 wrote:Source of the post That very same page 73 has an interesting statement on it:

Yes, it's all there, hence my comment about no ADDED value of the other paper.

Soundman2020 wrote:Source of the post To be honest, I would not try to draw conclusions from that graph, that could be used for studio walls.

Your points there about that graph are valid. What I was trying to say in general was that there are most possibly insulation compounds with better performance in the low end than 5 k.rayls/m glass wool.

Soundman2020 wrote:Source of the post the absorption peaks at the fourth wavelength. For a typical studio isolation wall, where the MSM resonance is hopefully less than 15 Hz, the wavelength is therefore about 23 meters. Four times that is 92 meters (about 300 feet). Building a wall with a cavity 300 feet deep seems a little overboard! :)

maybe I am missing something but I guess that would be divided by 4 not multiplied by 4?

Soundman2020 wrote:Source of the post Take a look at all of those graphs: the bottom end is at 250 Hz! There is no low-frequency data at all.

True, there are only some hints of what might happen if extrapolated, I simply couldn't find better data, and that is whast we are lacking IMO.

Soundman2020 wrote:Source of the post The most interesting graph in that paper, is actually Fig 10, which shows the GFR for several different porous absorbers. I thought it was quite fascinating that the two lines at the far extremes of that graph, are both for fiberglass! "D" is for "fiberglass aircraft insulation", while "B" is for "Fiberglass, resin-bonded". The point being that there's nearly two orders of magnitude difference between the density of those materials, yet both of them have useful GFR properties, in the typical range for both absorption and treatment in studios. So the underlying issue is: you can't generalize about materials, even if they are similar! Those are both "fiberglass" products, yet with rather different characteristics.

Yes, but this figure shows (lack of) relationship between the GFR and density even in "similar" materials, and nothing about acoustic properties.

Soundman2020 wrote:Source of the post GFR for 703 is around 16k rayls
Soundman2020 wrote:Source of the post if you are looking for the best insulation for your up-coming studio walls and ceiling, then very probably you would find it hard to beat Rockboard 40, or maybe Rockboard 60, if you can find a source for them. But even that recommendation comes with caveats!

Funny you didn't recommend glass wool rolls or batts, as the article you're defending did ;)
Interestingly you can read from manufacturers data sheets on those materials that 703 fiber glass has density of 48kg/m3, while Rockboard 40, 60 and 80 have density of 64, 96 and 128 kg/m3 respectively.
Now the funny thing is with the acoustic performance in the low-end @125Hz - the absorption coefficients for 703, and rockboard 40, 60, 80 respectively are as follows:
0.10 0.26 0.32 0.43 for 2" thick board
0.31 0.63 0.78 0.75 for 3" thick board
So one can see a linear relationship with density (and possibly GFR - couldn't find data on those) for a thickness of 2", but even already at 3" the densiest rockboard 80 starts falling short compared to rockboard 60.
My qeustions are - what are the coefficients of different material at (e.g.) 80 Hz and at what thickness? Which material provides best attenuation of sounds in the range of 30 to 125 Hz for different layer (air cavity) thickness, especially my desired 12"?

Soundman2020 wrote:Source of the post lso, what is the purpose of the drywall over the outer-brick wall?

The finishing purpose only :D And it's technically not a drywall, because in this technique gypsum board would be attached wet to the bricks using a gypsum glue to ensure there are no air cavities creating additional resonances left. It is simply easier to combine gypsum board with a wet compound to get a smooth surface than to shape a smooth surface from a set of wet compounds alone (the traditional technique of finishing brick walls, unless you want to go with unfinished, which I don't).

Soundman2020 wrote:Source of the post you just spend many paragraphs dissing the concept that filling the entire cavity is a good thing, the your proposed system has the entire cavity filled?

As I said before, I never did that :)

Soundman2020 wrote:Source of the post What would the "10mm NPE polyethylene foam" do here, and what would the resonant frequency of that system be? Is it low enough?

Well, 1) it is sold by my local acoustic voodoo supplier as the only suitable material to go under a wet conrete slab and there may be good reasons for it, the most obvious one is that it wouldn't soak wet concrete up, but I don't know really (I know it's not a good enough reason but I'm just responding honestly) 2) the purpose in my mind wasn't to create a low enough resonant frequency, as I am not trying to isolate different stories - the "studio", just as all other rooms in the house, are intended on ground level, so the only purpose of such floating floor would be to decouple it and its vibrations from the side walls and their vibrations (with NPE foam going under AND AROUND the conrete slab of the floor). I will look closer at your post about floating floors and see if it applies to my project, and then maybe we will have that discussion along with other things when I post the actual topic about the project in relevant section. I am not yet prepared to do that.

I hope this clarifies what I meant, and again thank you for the suggestions you provided, I got a lot to read and contemplate upon.

Best regards

Paul



User avatar
Soundman2020
Site Admin
Posts: 889
Joined: Thu, 2019-Sep-19, 22:58
Location: Santiago, Chile
Contact:

Document Library

#4

Postby Soundman2020 » Thu, 2020-Jun-18, 01:31

"Glass wool rolls and batts are very effective at causing the resonance frequency of lightweight cavity walls to shift lower, resulting in higher sound insulation."
True. It changes the way air deals with heat from adiabatic to isothermal, thus greatly increasing efficiency. It also lowers the speed of sound inside the insulation, and the sound waves "see" a longer path. All of those work together to lower the resonant frequency. There is no mistake here

"Glass wool rolls and batts will give the most economical way of maximising acoustic performance."
Also true. In most places I'm aware of, fiberglass insulation is, indeed, the least expensive option, compared to the alternatives mentioned in the paper.

"Filling the cavity with glass wool rolls or batts reduces the negative influence of standing waves and results in the best overall acoustic performance."
Also true. And in fact, that is the main purpose of insulation inside a wall cavity: damping resonances (standing waves). Those resonances do rob the wall of isolation. Filling the cavity completely as opposed to filling it partially (which is what they are talking about here), does indeed provide the best overall performance.

"The results show that the air flow resistivity has a positive impact on the sound reduction index, but no additional improvement is achieved when the r-value is higher than 5 kPa.s/m2, which corresponds to that of a lightweight glasswool rolls and batts."
Also true. And in fact, they didn't go far enough with that. If the GFR gets too high, then isolation suffers. Compress the fiberglass far enough, and isolation drops again, because the GFR is too high.

"It is imperative that this cavity is completely filled with low density glass wool in order to achieve the maximum benefit from the insulation."
Also true. Not in doubt. As I pointed out before, partial filling reduces isolation. Complete filling is the optimum solution.

"Increasing the density of the insulation above that achieving an air flow resistivity r ~ 5kPa.s/m2 does not improve the acoustic performance of the double-leaf wall."
As above...

"Total filling of the cavity with leightweight glasswool rolls or batts is the most economic way to achieve high sound insulation values compared with larger cavities or extra layers of gypsum board."
Correct. Large cavities cost money (they take up real estate), and extra layers of drywall also cost money: completely filling the cavity with fiberglass insulation is cheaper than both.

"Hence, we can summarise CHAPTER 1 by stating that not filling cavities of lightwieght walls with an adapted glass wool (r ~ 5kPa.s/m2) means not taking the advantage of the most economical way of maximising acoustic performance."
Also true. Less than 100% filling means less that 100% of the potential isolation.

What I was trying to say in general was that there are most possibly insulation compounds with better performance in the low end than 5 k.rayls/m glass wool.
That would be comparing apples to oranges! That's not what the paper is about. It is not a comparison of fiberglass vs. mineral wool vs. cellulose. The paper was about fiberglass insulation! Not other insulation... The title is crystal clear, I think: "The acoustic role of glass wool in double-leaf lightweight walls". No mention of other products. The very first paragraph clarifies the purpose of the paper even further:
"Many theoretical and experimental studies have been performed to investigate the effects of porous-mineral wool products in walls. From these studies we can draw the conclusion that there are three basic reasons why the sound reduction index ( R) or Sound Transmission Loss (STL) of a double-leaf wall improves when glass wool is used in the cavity.
1. The resonance frequency of the mass-spring-mass system is shifted to a lower value.
2. The glass wool dampens sound waves transmitted through the wall.
3. The glass wool dampens lateral standing sound waves in the cavity of the wall."

I think it's clear that they are not talking about other products, only fiberglass insulation. That's why they don't mention other products: because that's not what the paper is about. It is about the mechanisms that cause fiberglass to improve isolation, and the comparison of cavity fill to more layers of drywall or cavity depth. The conclusion is therefore valid: Filling the cavity completely with fiberglass insulation provides the best isolation. Filling it less than 100% reduces isolation, and either increasing the cavity depth or adding additional drywall costs more than filling the cavity 100%. Thus, the conclusion is valid: 100% fill is the best, most cost-effective option.

maybe I am missing something but I guess that would be divided by 4 not multiplied by 4?
No. The claim was that "the absorption peaks at the fourth wavelength." Dividing by 4 would give you the quarter wave length, not the fourth wave length. I have no idea where they got that idea from: never seen it before. I have often seen the typical "insulation must be a quarter wave length thick to effectively absorb the wave" claim (which is also completely wrong, of course), but that's the first time I see someone claiming that absorption peaks at four times the wave length.

True, there are only some hints of what might happen if extrapolated, I simply couldn't find better data, and that is whast we are lacking IMO.
Try IR 761. The data in there is very valid, carefully tested, and goes down more than two octaves lower, to 50 Hz. That's about as low as you can expect, and is already into the realm of statistical uncertainty... but still useful, since they are comparing structures all done in the exact same facility, with the exact same procedures. So even if it might not be totally accurate as absolute numbers, it is valid as relative performance. It isn't presented as alpha, of course, but it is more useful in reality. (In fact, alpha has it's own drawbacks, such as over-unity...)

Yes, but this figure shows (lack of) relationship between the GFR and density even in "similar" materials, and nothing about acoustic properties.
Well, it's widely known that different types of porous absorber have very different GFR, even for "similar" materials: manufacturing processes have a lot to do with that. Even the orientation can show large differences. Impedance tube measurements of samples set up face-on and edge-on with some products can produce substantial differences: with other products, they are practically the same. That's why it is important to understand the properties of the specific potential products you have available, in order to determine which one is best for each application, and how you will use it.

Funny you didn't recommend glass wool rolls or batts, as the article you're defending did
Not funny at all! That's because the article in question made no claims at all about mineral wool. Only about fiberglass wool. The products I suggested have better performance than fiberglass for the application you were talking about.

Interestingly you can read from manufacturers data sheets on those materials that 703 fiber glass has density of 48kg/m3, while Rockboard 40, 60 and 80 have density of 64, 96 and 128 kg/m3 respectively.
Yes. But what is your point? Once again, you are comparing apples to oranges. Mineral wool and fiberglass have very different properties: the GFR is quite different. Here's a graph that shows an approximate comparison between the two families:
airflow resistivity glass wool vs mineral wool--GFR.png
You can clearly see the large spread between them, and why mineral wool needs to be more dense to match the GFR (an therefore the performance) of fiberglass. You can't compare them based on density alone. (By the way, the lines in that graph just show TYPICAL performance, for GENERIC products of each type, and should not be taken as Gospel truth for any specific product: There's a broad range of values within each product family, with plenty of overlap. IF you want the date for a specific product, the you'd need to look that up. The graph is only intended to show general trends, and should not be used for predicting performance, since that varies widely by both product and family.)

My qeustions are - what are the coefficients of different material at (e.g.) 80 Hz and at what thickness? Which material provides best attenuation of sounds in the range of 30 to 125 Hz for different layer (air cavity) thickness, especially my desired 12"?
My question would be; why do you need to know the coefficient for 30 Hz? It would not be useful anyway. But if you do want it, you would have to ask the manufacturer for that data, if it isn't on their website. But good luck with that! Manufactures are not particularly helpful with acoustic data on their thermal products.... It took Andre many, many months to get a simple answer from Owens Corning on the actual GFR figure for OC-703: they beat around the bush for quite a while, before eventually coughing it up the data.

Since you'll probably have a hard time getting that data, your best bet is to just use the published GFR numbers, and plug them into a porous absorber prediction tool. It won't be far wrong (except at very low frequencies, of course).

You will also find it practically impossible to get any lab data below about 125 Hz, or maybe 50 Hz. at best. Most reputable labs won't test that low, because of the inherent uncertainties at such low frequencies. Trying to get a true diffuse field at those frequencies requires enormous rooms, which most labs do not have, plus major internal baffling, plus a very careful set of multiple measurements, to average out the noise. Reputable labs won't do that, because the results are not very valid statistically. So you won't find much published data below 125 Hz, or maybe 50 Hz. Even 50 Hz data is hard to find: maybe 63 Hz if you are lucky. Neither is it needed for home studios! The approach to low frequency treatment in home studios is usually sufficient to do the job, since there's no way to measure it accurately if you want absolute values. Especially considering that there's not much going on at 30 Hz in any case (musically speaking) unless you are mixing cathedral pipe organs and canon fire, maybe... And even for that, there's still the equal loudness curves to take into account, for perception. 30 Hz is usually not much of an issue in home studios or home theaters, with few exceptions.

The finishing purpose only :D And it's technically not a drywall, because in this technique gypsum board would be attached wet to the bricks using a gypsum glue to ensure there are no air cavities creating additional resonances left.
Rendering a masonry wall is fairy quick and simple. Painting it is even easier! And faster. and cheaper. There are many good masonry sealants on the market that you can simply paint on, and even ordinary paint will do the job of sealing the wall surface. Also, "gypsum glue" would not help much with gluing things to drywall, sine the surface of drywall is not gypsum; it is paper, cardboard, of foil. They gypsum is inside, yes, but the surface of the board is not gypsum. You would need a good glue that sticks well to cardboard or paper. And even then, you should check your building code: you will likely find that drywall cannot be attached with glue alone. It will likely need to be nailed or screwed, following the approved schedule.

It is simply easier to combine gypsum board with a wet compound to get a smooth surface
Why do you want a smooth surface inside the wall cavity? Nobody is ever going to see that. Just sealing it is fine. I can understand you wanting to get a nice finish on the room interior, but there's no need for that on the outer brick wall, which i what I was asking about. Putting a layer of drywall over that, no matter how it is attached, is a waste of time and money, and won't do anything useful acoustically either. Just seal it.

it is sold by my local acoustic voodoo supplier as the only suitable material to go under a wet conrete slab and there may be good reasons for it, the most obvious one is that it wouldn't soak wet concrete up, but I don't know really (I know it's not a good enough reason but I'm just responding honestly)
Let's back-track here: The original statement was: "5) serious floating floor (high density concrete slab of at least 50 mm floating on a 10mm NPE polyethylene foam, decoupled from the walls,". That is NOT a "serious floating floor"! Not even close. "10mm NPE polyethylene foam" is not going to act as the spring that you NEED in order to make an acoustic floating floor, serious r otherwise. If that's what your "acoustic supplier" told you, then you should probably consider searching for a different supplier!

In fact, you probably don't even need a floating floor at all! Take a close look here: Floating your floor: How and why... and why not. Yes, that's the same link I gave you a coupe of days ago, but you should probably take a closer look, since it seems that you missed a couple of the key points. Floating a floor for a studio is a Big Deal, and it cannot be accomplished with a very thin foam pad. The frequency would be way too high to be useful, and the isolation would not very good. That thin foam stuff might work OK for impact sound, but not airborne sound or vibration. I think your acoustic supplier does not understand the principle of an acoustic floating floor.

the purpose in my mind wasn't to create a low enough resonant frequency, as I am not trying to isolate different stories - the "studio", just as all other rooms in the house, are intended on ground level, so the only purpose of such floating floor would be to decouple it and its vibrations from the side walls and their vibrations
Then there's a major misunderstanding of isolation here! How can you isolate a resonant structure without tuning the resonant frequency low enough? It isn't possible. It is also impossible to isolate a room in only some directions, not others. If the "floating" slab is only decoupled from the side walls, but not from the underlying foundation slab, then that is pretty pointless! The side walls rest on that foundation slab! So any vibration that did get into the foundation slab, will be in the side walls anyway, regardless of what you do around the edges of the slab! Isolation is an "all or nothing" proposition. If the floor needs floating, then either you isolate the entire room, or you don't. There's no half measures, where it would be possible to isolate the ceiling and left wall, but not the right wall or front wall.... that cannot happen. The laws of physics prevent it, unfortunately. So if you do need to float your floor, then you also need to float your inner-leaf walls on top of that floor, and the inner-leaf ceiling on top of those inner-leaf walls: In other words, you float the entire room. That's the normal way it is done for studios (and other structures) that need high isolation. And since most home studios (and home theaters) do not need that much isolation, hence they don't need floating floors.

It seems to me that there is no need at all to do that "floating" floor (which is not even floating to start with, and not isolating), and you would gain 60mm of headroom by not doing that... There's no point, if it isn't floating, and you likely don't need a floating floor anyway: the entire house is "slab on grade", by the sound of it. If your studio has a "slab on grade" floor, then you already have the best possible floor you can get for a studio. It is already very massive, and very well damped, with a flanking limit likely around 70 dB for airborne sound. It's unlikely you'd need more than that, and you'd need an impressive budget to get there! Very deep pockets. The isolation provided by your walls, doors, ceilings, and HVAC system won't get to that level, so you would never actually reach the limit imposed by the "slab on grade" floor. Thus, there is no need for a floating floor. Just build your room on the slab. If you are worried about vibration from speakers getting into the structure, then isolate the speakers, not the entire room! (Sorbothane is your friend here....)

(with NPE foam going under AND AROUND the conrete slab of the floor).
How would you ensure getting the correct deflection on that rubber around the edge of the slab? It does need to be compressed correctly, in order to isolate. It does not matter which direction the foam is intended for: it's the same out to the sides of the floor, as under it. It still has to be compressed correctly, or it will flank. If it is over-compressed, even in a few spots, then you have flanking paths (and no isolation). If it is under-compressed, even in a few spots, then it is not "floating", and thus does not isolate. And we are talking about fractions of a millimeter here: It would be an interesting challenge to install that foam all around the edges of the floor, measuring carefully with a micrometer, to ensure that the deflection is correct all over... I haven't check the elastic properties of that foam, but at a rough estimate, 10mm foam would need to be compressed be about 1.7 to 2.4 millimeters, give or take an even smaller fraction. I'm not sure how you would even go about getting that right, for the entire edge of the slab. How would you even pour the slab with such accuracy? The entire perimeter gap would need to be accurate to within less that half a millimeter...

To be honest, I think your acoustic supplier has not idea what a floating floor is, and just wants to sell you stuff that might work for impact noise, but certainly won't for typical home-theater or studio sound levels and frequencies.

I will look closer at your post about floating floors and see if it applies to my project,
It does, undoubtedly. You want to float the floor of your room, and the article is all about floating the floors of rooms, so it does apply. And the conclusion you should arrive at after reading it, is that the system your supplier proposed would NOT float the floor at all, and you don't need to float it anyway. Attempting to float it on a thin foam pad is doomed to fail, no matter how you do it, and is not necessary anyway. Forget the floating floor, save yourself a lot of money, and gain all that extra headroom, which is far more important, acoustically and aesthetically, for a home theater.


- Stuart -



Avare
Active Member
Posts: 82
Joined: Fri, 2019-Sep-27, 05:36
Location: Hamilton, Ont

Document Library

#5

Postby Avare » Thu, 2020-Jun-18, 03:05

Maybe the thing is that in Canada fiberglass wool is much cheaper than mineral rock wool ? Well, in my country it is about 10-20% cheaper and actually dependent on the supplier it can even be reverse.
It is the same in Canada.

I did a quick search and the document is from Saint-Gobain, a European company, not Canada.

Boh glass wool and Rockwool are mineral wool. Rockwool is also the trademark name of a company.


Good studio building is 90% design and 10% construction.


Who is online

Users browsing this forum: No registered users and 19 guests