from long ago - triple leaf data

[Brian Dayton]

in october or so we had discussed some triple leaf tests to be presented to these forums.

most of the test session in october got badly behind schedule due to some problems with installation, but one such comparison was run, and i have been very tardy about entering it here.


The attached graph shows the results:

the walls are:

REFERENCE WALL:  2x4 wood studs, 24" OC, R13 fiberglass insulation, single 5/8" drywall on each side

TRIPLE LEAF WALL:  as above, add soundboard, resilient channel, and 5/8" drywall to one side, giving

drywall/studs+insulation/drywall/soundboard/RC/drywall

DOUBLE LEAF WALL:  same frame and insulation, but a normal resilient channel installation with double drywall on both sides

all drywall was 5/8"
soundboard was 1/2"
channel was 25 gauge and installed 24" OC
screw pattersn, etc., were the same for all

FLANKING LIMIT:  this shows the absolute flanking limit of the facility

the results:

REFERENCE WALL:   STC=40, OITC=29, Tennekes=25
TRIPLE LEAF WALL:   STC=52, OITC=34, Tennekes=28
DOUBLE LEAF WALL:  STC=55, OITC=40, Tennekes=31.5

So, while its clear that the triple leaf performs worse, in this case i'm perplexed by how it didn't perform even more worse, mostly, by how the apparent resonance of this triple leaf system is so low (80=-100hz).

anyway, it does illustrate the basic point, but it doesn't provide the dramatic example that folks seem to desire sometimes.

[Ido]

thanks as always Brian   .

...you might have addressed this before..:

this soundboard, might it soften negative effects? maybe would have been worthwhile to measure same triple leaf without soundboard?

the 2 aircavities in the triple leaf are different thickness's, no?  (less pronounced, more spread out negative effect?)

this I'm sure you talked of once, so pardon me: depending on which type of RC, do the neoprene RC's have some damping advantage?   (same RC's in triple and double? what type?)

shouldn't the triple leaf have also shown some advantages in the HF? might RC soften this effect  as compared to RC in double leaf?

maybe a dry case of triple versus double sans soundboard sans RC would have been more pronounced?

maybe this specific triple leaf was so stressed out by it's bad reputation (some of which is because of you  :mrgreen:  :D !)  it reformed on the spot?

still, the double leaf has the narrowest (not to mention lowest) MSM region, no?

I tell you that in regards to all the types of acoustic resonances, I feel I'm always missing out on the magnitude and significance of these babies (be it "positive" resonances or the"negative" ones").

thanks again Brian,
Ido

[Ido]

Brian, something relevant to the above and to GG, an actual case at hand:
regarding a resiliently mounted 2 layer drywall ceiling to be mounted below the original massive house ceiling.
there are limitations, such as the airspace can be max 1.4 "  /  3.5 cm which only allows Kinetics iso-max AFAIK, and only 2 drywalls.
there will be LF music content.
the purpose is to do the best within the limits.
of course nothing should get worse.
I don't know if you talked of this before, a link would be great too:

1. will GG in such a setup constrain the negative resonances at MSM and such, to an extent that I don't need to worry about making things worse?

2. how do these neoprene resiliient mountings behave regarding damping?

3. another option (or combo with GG?)  would be an interlayer (between the drywalls) of either lead,  or,  of a real cool damping layer we have here(it's like soft soundblock, with an adhesive layer) .

btw, in your graph above, might it be that for the standard single layer wall, the wide & erratic MSM region also shows the non-damped behaviour of single layer drywall,  as opposed 2 layers ?
we never ever do single layers, not for anything.

thanks Brian,
Ido

[Brian Dayton]

Brian, something relevant to the above and to GG, an actual case at hand:
regarding a resiliently mounted 2 layer drywall ceiling to be mounted below the original massive house ceiling.
there are limitations, such as the airspace can be max 1.4 "  /  3.5 cm which only allows Kinetics iso-max AFAIK, and only 2 drywalls.
there will be LF music content.
the purpose is to do the best within the limits.
of course nothing should get worse.
I don't know if you talked of this before, a link would be great too:

1. will GG in such a setup constrain the negative resonances at MSM and such, to an extent that I don't need to worry about making things worse?

well, that question depends on the edge conditions of the ceiling and mechanical stiffness that is present.

sorry to be so slow to reply, BTW, somehow i missed this thread.

if the edges of a wall such as this are perfectly floating, then the impact of Green Glue is at a minimum.  In, as a different example, a double stud wall with its fairly stiff studs to resist motion at MSM and shift system properties towards the mechanical, the GG has been shown to be very effective (typically the dip at MSM is 1-3 dB based on lab tests for a GG double stud wall, with large benefits at higher low-freq's also.  i posted about this at SOS earlier)

2. how do these neoprene resiliient mountings behave regarding damping?

3. another option (or combo with GG?)  would be an interlayer (between the drywalls) of either lead,  or,  of a real cool damping layer we have here(it's like soft soundblock, with an adhesive layer) .

btw, in your graph above, might it be that for the standard single layer wall, the wide & erratic MSM region also shows the non-damped behaviour of single layer drywall,  as opposed 2 layers ?
we never ever do single layers, not for anything.

thanks Brian,
Ido

2 layers of drywall isn't better at MSM than just one.  i have some data if you like on the progressive improvement when going from 1+1 to 1+2 to 2+2 layers of drywall on a single stud wall, other such data is available in IR761.  Generally TL gains at resonance are less than the mass-predicted gains.  resonance beats mass, i guess, as they sometimes say.

which type of mount you utilize (resilient mount) isn't so likely to affect damping,  as the general resilient stiffness is much, much more air cavity than the stiffness of the mounts with such a thin air space.  Because the air cavity is the primary source of stiffness in the "spring" system, it is the damping of the air cavity that is most important.

That is a challenging situation, 1.5" of cavity depth isn't alot, and generally the deeper an air cavity the better the damping...

i guess i don't know much about the soft damping layer you describe.  

let me think it over and i'll get back

[Ido]

Brian, thanks for answering.
I don't get it yet, and I think what's stopping me is the "stiffness" issue.
in theory, we want infinite stiffness, but in lightweight/drywalls, we don't, right?
and this is because..  we want to somehow limit the transfer of energy between the 2 sides?  enable a degree of freedom between the 2 leaves?  so this relates to the "construction"  of the lightweight wall, and not to the leaves/drywall?
are these considerations frequency dependant? for the LF resonances you do want stiffness?
how about stifness in the leaf itself? in a theoretically fully decoupled 2 leaf wall, would you want max stifness on each independant leaf? would you want narrower spacings between studs?
regarding GG, are you saying it's max benefit is in setups that are stiff/rigid to begin with?
so there is a major difference in sonic energy transfer within a 2 leaf wall if it is structure borne or airborne? (as regarding GG?)
you see my confusion    :D ?

...2 layers of drywall isn't better at MSM than just one.  i have some data if you like on the progressive improvement when going from 1+1 to 1+2 to 2+2 layers of drywall on a single stud wall, other such data is available in IR761.  

interesting, I'd love to see that.

[
...  Because the air cavity is the primary source of stiffness in the "spring" system, it is the damping of the air cavity that is most important.
That is a challenging situation, 1.5" of cavity depth isn't alot, and generally the deeper an air cavity the better the damping...

but say the cavity is a given, how about the damping of the drywall itself? if it is indeed an allround  good floating construction, is there no benefot to damping at LF resonance, be it  GG or something else?

i guess i don't know much about the soft damping layer you describe.  

the material is called TECSOUND, the specific type I have is S 35  (3.5 kg per m2, about 2.5 mm thick) . I believe it's Spanish made.
it's like a very pliable MLV, with an adhesive side, I'm pretty sure it is good for damping.
and here it costs about half than the "proper" sound barriers.

http://www2.uah.es/innovaciones/0405/tr ... csound.pdf

I found this link, scroll down and it looks like there is some intereting data in there,
translation anyone? Adore?

Brian, I'd gladly send you a piece.

Ido

[Brian Dayton]

Brian, thanks for answering.
I don't get it yet, and I think what's stopping me is the "stiffness" issue.
in theory, we want infinite stiffness, but in lightweight/drywalls, we don't, right?
and this is because..  we want to somehow limit the transfer of energy between the 2 sides?  enable a degree of freedom between the 2 leaves?  so this relates to the "construction"  of the lightweight wall, and not to the leaves/drywall?
are these considerations frequency dependant? for the LF resonances you do want stiffness?
how about stifness in the leaf itself? in a theoretically fully decoupled 2 leaf wall, would you want max stifness on each independant leaf? would you want narrower spacings between studs?
regarding GG, are you saying it's max benefit is in setups that are stiff/rigid to begin with?
so there is a major difference in sonic energy transfer within a 2 leaf wall if it is structure borne or airborne? (as regarding GG?)
you see my confusion    :D ?

darn it, i somehow missed your reply again.  this must be the invisible thread, probably because i started it (my threds never get alot of hits)

well, when you decouple a wall you create a mass-spring system (as of course you know).  In the case of a system where you have ISOMAX clips and air between the two sides of the wall, the "spring" is the isomax clips + the air

And when i was talking about stiffness above, i just meant the stiffness of the spring.  Perhaps think of it this way:  As the drywall moves forward and back at resonance, what forces are at work on it (the drywall)?

for isomax clips as described above, there are 3 forces

1)  the compressing/rarefying of the air in the cavity
2)  the compression/extension of the rubber in the clips
3)  edge conditions of the drywall, and whatever mechanical stiffness they introduce to the system

So if the air is much stiffer than the rubber in the clips (in this case it will be), then the air dominates and the "spring" is defined basically by the air cavity.  The only mechanical element is introduced by the edge conditions.


Take another example, of a double stud wall, then you have

1)  the compression/rarification of the air in the cavity
2)  the bending stiffness of the drywall/stud assembly

Imaginet hat teh studs were made out of rubber, and the drywall was replaced with MLV.  Then the stiffness of the mechanical part is nearly zero, and the air stiffness would dominate.  But in a real wall, with 5/8" drywall and 2x4 lumber studs, the mechanical stiffness is very high, and the motion of this type of wall at resonance has a considerable (if not outright dominant) mechanical component.



Then take a constrained layer damping material, like Green Glue (or any other), ... it damps the mechanical element of a resonance, and if the mechanical element is dominant, all is well, but if the acoustic (Air) element is dominant, then its overall effectiveness won't be as high.  The attached graph shows a GG double stud wall of about 56 kg/m^2 compared to the flanking limit of the lab and a single stud wall of about 62 kg/m^2.

Now, in theory the single stud wall is defined by mass below its resonance, so at 63, 50, 40, and 31.5hz, this wall is mass defined.  Of course less than perfect measurements/repeatability/certainty/etc. comes with low-freq measurements and all of that usual stuff, but, it seems from these data that the resonance of the double stud wall is extremely well damped by the GG

If this was a 1.5" air cavity with isomax clips, i can't say that my confidence is as high that the damping would be as good, because that type of system is so air-spring dominated.


did that make more sense?

...2 layers of drywall isn't better at MSM than just one.  i have some data if you like on the progressive improvement when going from 1+1 to 1+2 to 2+2 layers of drywall on a single stud wall, other such data is available in IR761.  

interesting, I'd love to see that.

 second picture attached.

[
...  Because the air cavity is the primary source of stiffness in the "spring" system, it is the damping of the air cavity that is most important.
That is a challenging situation, 1.5" of cavity depth isn't alot, and generally the deeper an air cavity the better the damping...

but say the cavity is a given, how about the damping of the drywall itself? if it is indeed an allround  good floating construction, is there no benefot to damping at LF resonance, be it  GG or something else?

i guess i don't know much about the soft damping layer you describe.  

the material is called TECSOUND, the specific type I have is S 35  (3.5 kg per m2, about 2.5 mm thick) . I believe it's Spanish made.
it's like a very pliable MLV, with an adhesive side, I'm pretty sure it is good for damping.
and here it costs about half than the "proper" sound barriers.

http://www2.uah.es/innovaciones/0405/tr ... csound.pdf

I found this link, scroll down and it looks like there is some intereting data in there,
translation anyone? Adore?

Brian, I'd gladly send you a piece.

Ido

unfortunately, it isn't likely that this material is a very effective damping material unless it was engineered specifically for this application.  it's not by chance that CLD systems work well or fail.  i'll check it out.

[Brian Dayton]

i can't understand that .pdf file, ldo

[Brian Dayton]

my comment above might have sounded harsh about this other damping option.  and i do apologize to all, but i'm american, and as a result of this i'm only interested in wood studs and english and i think budweiser is a really good beer and i think we are the source of democracy in the world and i don't care if world history class taught me differently.  

so i'll elaborate a bit.

lots of things seem like good damping materials.  MLV gets recommended as one sometimes.  And that's not illogical, as MLV itself is very well damped, but there is an ocean (a big, big ocean) between a material itself being well damped and that material performing well as a damping material.

Is MLV a good constrained layer damping material?  Nope.  Why?  Well, MLV's biggest use, by far, is to improve sound transmission properties of thin sheet metal.  It works at this because it adds some weight were there really isn't any.  probably quadruples the weight of the sheet metal or something like that.  Its adding mass.  It probably, if adhered to one side of the metal, also increases the damping.  But metal is so incredilby badly damped that ANYTHING can raise its damping quite a bit.

MLV is a material that stays flexible over quite a wide frequency range.  This fact alone reveals to the trained observer that it is not going to be a good damping material.  Asphalt materials are very different.  they melt when hot, they turn brittle when cold, and they are more potent as damping materials than MLV.   The very state of material that is viscoelasticity is a state of transition, not of constancy.


A material like Quiet Coat is a good EXTENSIONAL damping material, but no better than liquid nails as a CLD material.  Green Glue is a good CLD material, but no better than strawberry jelly as a paint on damping material.

I should re-phrase:  Green glue is a great CLD material between drywall or wood or common building materials.  These materials have specific surface properties and traits that are fairly homogenous, and drywall especially doesn't lend itself well to being damped by just anything.  To the best of my knowledge, GG is the only damping material ever developed specifically for this application.  

Is GG a universally good damping material?  of course not, CLD is a really specific thing, and it's good in this really specific application (audio range frqeuencies, potency centered at ~80hz, between the aforementioend building materials in the typical film thickness, around room temp).  For another application it may well be quite poor.

So multi-purpose materials sometimes touted as "and it's a great CLD material"... this is just so, so, so unlikely to be true.

To say Green Glue is the best CLD material in the world is silly.  For its application.  There are many other applications with some other material as best - and always this material was designed for that application.

[Ido]

..... but i'm american, and as a result of this i'm only interested in wood studs and english and i think budweiser is a really good beer and i think we are the source of democracy in the world and i don't care if world history class taught me differently.  

thanks so much Brian. I got me some serious reading to do  :D   (a bit suprised on the Budweiser   , but i'll let it go..).
you see, I was contemplating bringing over some of the green stuff.

[Brian Dayton]

..... but i'm american, and as a result of this i'm only interested in wood studs and english and i think budweiser is a really good beer and i think we are the source of democracy in the world and i don't care if world history class taught me differently.  

thanks so much Brian. I got me some serious reading to do  :D   (a bit suprised on the Budweiser   , but i'll let it go..).
you see, I was contemplating bringing over some of the green stuff.

i was just teasing about that american stuff.  

i just haven't tested the configuration you describe, ldo, so my answer is mostly theory as i just don't have a cocnrete answer based on a test.

[Ido]

Brian, thank you so much.
You explain beautifully with a combo of science and mechanical intuition which is the only way I personally ever understand anything.

I want to ask further clarification, and perhaps this might benefit others too:

To this day I still don’t understand stiffness/TL:
How does the theoretically perfect “infinite stiffness” relate to a practically excellent limp material such as lead?
is this solely relevant to low frequency isolation which is "stiffness" controled?
Is it because the real life stiffness is always compromised/non-infinite?
how does lead behave regarding LF isolation?  only mass?  it benefits from it's "internal damping"?  

BTW, when you say “..edge conditions of the drywall, and whatever mechanical stiffness they introduce to the system”,
is this the basic definition for “stiffness of a panel” (as in the inherent stiffness of the material itself)?
or do you indeed emphasize the significance of the edge conditions (as in, the material at hand is secured firmly, or has freedom of movement)

  ,,,,,,Then take a constrained layer damping material, like Green Glue (or any other), ... it damps the mechanical element of a resonance, and if the mechanical element is dominant, all is well, but if the acoustic (Air) element is dominant, then its overall effectiveness won't be as high.  

Can you please explain the difference in the mechanical behaviour between the energy transfer of vibration/resonance of the mechanical element (stud) and between the energy transfer of vibration/resonance via air stifness?
In the end, isn’t it the drywall itself that is set into vibration and “decides” the outcoming TL?
Does the drywall vibrate differently in the case of mechanical vibration versus the case of the airborne-induced vibration?

Is there a frequency difference between the air-borne vibration transfer and the structure-born (stud) vibration transfer?

Regarding the above mentioned case in hand, of thin arrow air cavity and decoupled studs:
Do you think that in such cases the dedicated decoupling is perhaps an insignificant contributor?

I’m asking a lot here, I know. please don’t feel obliged to answer, but you see, I have to ask….
Thanks Brian,
Ido

[Brian Dayton]

Brian, thank you so much.
You explain beautifully with a combo of science and mechanical intuition which is the only way I personally ever understand anything.

I want to ask further clarification, and perhaps this might benefit others too:

To this day I still don’t understand stiffness/TL:
How does the theoretically perfect “infinite stiffness” relate to a practically excellent limp material such as lead?
is this solely relevant to low frequency isolation which is "stiffness" controled?
Is it because the real life stiffness is always compromised/non-infinite?
how does lead behave regarding LF isolation?  only mass?  it benefits from it's "internal damping"?  

ldo, all:

in the real world, it is not possible to make a wall stiff enough to improve TL.  You cannot brace your walls to stop them from flexing and improve LF performance, youc annot use 2x12's to do better than 2x4's, etc.  This is a common myth among studio folk, but it's not so.

in a panel, increased stiffness/weight ratio results in a lower frequency of coincidence dip.  For a panel of a given weight, the higher the stiffness, the lower the frequency of this dip.  This is very well known, and the reason that people recommend not using rigid glues between drywall layers relates to this lowering of coincidence (in general its better to have that resonance higher in frequency)

One solitary case exists where stiffness is not useless or bad, and that is at the mass-spring resonance of a decoupled wall.  Here you have the wall moving at resonance.  This motion involves the air spring (upon which the two leaves are bouncing at resonance).  And this motion caused by the air spring is resisted by mechanical stiffness, if present.  So the air is trying to "bounce" the layers of the wall, and the mechanical stiffness is resisting this "bounce".  that's not such a bad explanation, maybe.  If mechanical damping is high, like a GG wall, then this is good.  Now, you don't need to brace walls or anything, the graph i posted above with the apparently almost totally controlled MSM is just a normal double stud wall with 2x4 studs 16" OC.  No kinky cross bracing or anything.

BTW, when you say “..edge conditions of the drywall, and whatever mechanical stiffness they introduce to the system”,
is this the basic definition for “stiffness of a panel” (as in the inherent stiffness of the material itself)?
or do you indeed emphasize the significance of the edge conditions (as in, the material at hand is secured firmly, or has freedom of movement)

no, i meant that imagine that the drywall mounted on the ISOMAX clips was totally free at the edges.  its motion at MSM is then almost totally forward/back, like a piston, with mechanical resonances occuring here and there in the panels, but the MSM would almost approach a case where the drywall was moving as a piston.  because it can go forward and back freely at the edges of the panel.  this situation is approached in lab tests for resilient mouts because extremely pliant edge sealants are used.

now imagine a more realistic edge condition in the real world.  USG acoustic sealant is under the drywall, and thoroughly dried (1-2 months).  now the edge cannot move freely, or really at all, and the entire panel must bend during thos motion at resonance.

see?

  ,,,,,,Then take a constrained layer damping material, like Green Glue (or any other), ... it damps the mechanical element of a resonance, and if the mechanical element is dominant, all is well, but if the acoustic (Air) element is dominant, then its overall effectiveness won't be as high.  

Can you please explain the difference in the mechanical behaviour between the energy transfer of vibration/resonance of the mechanical element (stud) and between the energy transfer of vibration/resonance via air stifness?
In the end, isn’t it the drywall itself that is set into vibration and “decides” the outcoming TL?
Does the drywall vibrate differently in the case of mechanical vibration versus the case of the airborne-induced vibration?

like above.  if the panel is acting like a pure piston at resonance (like in the rarified case above), then the MSM has no mechanical element.

if the panel must bend, then it has a mechanical element, and damping materials are maximally effective.  draw a picture of how any given construction has to move at resonance to gain an idea.

Is there a frequency difference between the air-borne vibration transfer and the structure-born (stud) vibration transfer?

Regarding the above mentioned case in hand, of thin arrow air cavity and decoupled studs:
Do you think that in such cases the dedicated decoupling is perhaps an insignificant contributor?

I’m asking a lot here, I know. please don’t feel obliged to answer, but you see, I have to ask….
Thanks Brian,
Ido

no frequency difference.  wherever teh MSM is in a structure, then at that frequency there will be a combination of mechanical (something bending) and air forces.  if nothing bends (perfect piston), then the air is dominant.

when i say air, you have to consider that as meaning air + the spring force of the clips

don't worry about the questions, i just hope my explanations are decent

[andrebrito]

Brian, which part do you want me to translate ?

First page basically says: high density,  high elasticity, high sound insulation at low frequencies (but the charts of page 3 start only at 100 Hz   ), adapts to all surfaces bla bla bla....

second page is the description of their products !

third page some charts

they also say the product is quite heavy and cannot be fully  recycled

[Ido]

Brian, thanks again  :D .

Andre, I think the following graphs seems interesting, what do they say? what is the scenario?

[Brian Dayton]

Brian, thanks again  :D .

Andre, I think the following graphs seems interesting, what do they say? what is the scenario?

hey Andre, thanks for offering.

my biggest questions would be these

A)  is the thickness of the material given
B)  is the density or weight per area given
C)  is the composition given?  acrylic, vinyl, foam,
D)  waht do those graphs ldo showed say.  what are the walls in question?

[andrebrito]

translation:

TOP

"Soundproofing is the main method in noise control. It consists on reducing noise transmission from two place, generally, from one place to another"

BOTTOM

"3  comparing charts of soundproofing"

On the charts... comparing using one of his products and without Tecsound, influence in soundprooginf

the third chart comparing using one layer of their product with two layers I believe

[Eric Desart]

Brian, thanks again  :D .

Andre, I think the following graphs seems interesting, what do they say? what is the scenario?

hey Andre, thanks for offering.

my biggest questions would be these

A)  is the thickness of the material given
B)  is the density or weight per area given
C)  is the composition given?  acrylic, vinyl, foam,
D)  waht do those graphs ldo showed say.  what are the walls in question?

English : 46 pages
http://www.texsa.com/desc/sistemasacustica_eng.pdf

Best regards
Eric
.

[Bob]

Is this
a) mass loaded vinyl hung loosly (with fiberglass on both sides)
b) triple leaf
c) apply only to something non obvious (like single 1/2" gyspum on both sides, or brick pg 26)

from: http://www.texsa.com/desc/sistemasacustica_eng.pdf   page 13
Another effective method, particularly for very rigid walls, is to use the
diaphragm effect. This consists of filling the cavity with a material comprising a thin
membrane with very low f0, positioned between two spring elements, such as felt or
mineral wool. The spring elements prevent movement of the membrane when it is hit
by the sound waves, and this causes greater dissipation of mechanical sound energy
with the consequent increased insulation. It is important not to use sheets of polystyrene
or other rigid foams which worsen the result from the acoustic viewpoint for the air
chamber filling.

[andrebrito]

Perhaps the thin membrane that separates the spring materials is not heavy enough for the system to be a triple leaf ?!?

[Ido]

thanks Eric,
that's one serious brochure, eh?
hadn't looked fully in-depth, but it seems to have with/without tests.
and they do call it visco-elastic..
Ido