STC Test Data for Partitions

Post and discuss acoustic topics, Studio design, construction, and soundproofing here

Postby Eric.Desart » Sun Feb 22, 2004 8:45 am

Hi all

I think the BBC repports on the net are still the best.
I have old internal BBC reports (books).

They started very long ago with their Camden walls (drywalls)
They standard used such types of fibreboards as homasote or comparable (often called Soundboard).
In fact they used it back then as a kind of decoupling from the studs. This name Camden orginate from a UK theater where those walls first were used very long ago.

More recent studies showed they don't use those boards anymore, but gypsum board (different types).

INC investigates in function of the influence of parameters. This is very important.
So I want to do something with this data in order to analyze this better.
But this is not specific targetted towards high insulations or low frequent insulation.

The BBC studies are targetted towards high insulations in function of their broadcoasting activities and studios. And those guys are good, really good.

Best regards
Eric
Eric.Desart
Moderator
 
Posts: 2461
Joined: Tue Feb 10, 2004 4:29 am
Location: Antwerp, Belgium

Postby Bob » Sun Feb 22, 2004 10:05 am

Hi Eric:

I did a google search for "BBC flexible walls sound" and the only reference I found was a thread you'd written in April 2001. http://tinyurl.com/34yrp

There's a bit of a debate going on in home theater circles about whether to follow Russ Herschelmann (the home theater guy) and Dr. Floyd Toole's advice that small rooms with rigid walls sound boomy but with a single layer of gypsum on the inner wall the wall is flexible enough to avoid the 'boomy' problem, vs it doesn't matter if the wall is rigid or not for a variety of reasons (starting with it's really hard to design such a wall, and it's easier and cheaper to treat the room after it's built, etc.).

Certainly it's easier to build a rigid wall when one is interested in having STC-60 or more walls.
In my case I'm defining rigid as 2 layers of gypsum, whereas I belive you were defining rigid as brick/concrete.

Since the purpose of the wall is for isolation, I'm leaning towards 'rigid is ok'. If I didn't need the isolation, I wouldn't bother building the wall.

On 2001-04-02 01:30:03 PST in alt.sci.physics.acoustics you wrote that you were still studying the phenomenon.

Do you have any thoughts on the matter? Or would you like me to elaborate a bit more before you comment?

This is the formula which I use for now (deviated from Gösele approach) for a wall existing out of 2 stiff leaves, with an air cavity standard filled (> 2/3 of air cavity) with a fibreglass layer:

f_d = (240.42/sqrt(d)) * sqrt(1/M_1+1/M_2)

Translated to a more general approach:
f_d = 4000*sqrt(s*((M_1+M_2)/(M_1*M_2)))/(2*pi)

M_1, M_2 kg/m²
f_d resonance frequency
d thickness cavity in meter
s dynamic stiffness in MN/m³ divided by thickness in meter.

Since I can't find the original article back, I'm not 100 % sure that the formula is build-up for random incidence, since studying my own made documents, it's somewhere between straight incidence and random incidence. Therefor (until find originals back) I assume it's meant for random incidence. Anyhow those formulas fit my own tests for random incidence.

... such papers can be the cause of problems, since lots of people will apply blindly, the procedures described.

Gösele, K. Berechnung der luftschalldämmung von doppelschaligen Bauteilen" Acoustica 45, 1980, S. 208

Sound insulation of partitions in Broadcasting Studio Centres: field measurement data.
It is based on sound insulation data for hundreds of partitions used by the BBC over a period of 15 years. My copy dates from Oct 1986 (BBC Engineering, Research Department) and has a reference number UDC 534.833.522 ... "Sound insulation of partitions in broadcasting studio centres: field measurement data" by K. Randall, D. Meares and K. Rose, October 1986. ...the sound insulation was severely limited by flanking that could not be quantified,
Regards
Bob Golds
"The only thing we regret in life is the love we failed to give."
"Be a rapturist -- the backward of a terrorist. Commit random acts of senseless kindness, whenever possible" - Jake Stonebender
Bob
 
Posts: 4360
Joined: Sun Feb 08, 2004 4:37 am
Location: Oakville, Ontario, Canada

Postby Eric.Desart » Sun Feb 22, 2004 2:53 pm

Bob,

Can you give the link where I find this quote.
I've written so many things.

And since a couple of months you must be one of the most energetic guys on the net.
I really can't follow anymore.

Time...................................................

Anyhow Gösele is good, really good.
It's probably the only guy in the world that discovered that the traditional formulas, which are educated in ANY acoustic course to calculate drywalls do NOT apply for stiff walls.
And stiff/flexible is defined in function of coincidence frequency.

Only the problem is the gray area below and above coincidence.
Also Gösele didn't solve that.
Gösele sais: that's how it is for flexible walls. This resembles the known theories.
Gösele sais: that's how it is for stiff walls. This is new and EXTREMELY important.

But lots of walls are in-between (say coincidence between 100 and 2000 Hz).
There Gösele sais: Nothing

Also Gösele ignores angle of incidence.
His findings are based on an empirical approach not an analityc solution.

EXTREMELY IMPORTANT HOWEVER is that Gösele sais to the acoustic world (even academic) Hey guys you're ignoring something extremely significant here.
And still the same mistake is educated anywhere ........

I've been involved in legal proceedings as expert for the governement with an estimated damage of over 7.5 million dollars which can directly be assigned to this error + the use of a wrong single number rating as engineering target. (an Eric not known here).

And I know there must be an analityc solution, I'm still working on that + needing help from cleverder guys (but that's when I feel like it, and get an idea in bed, if it doesn't work out it's stuck again, so can take some while).

This means that this gray area isn't solved yet.
And I do have more. But I told before I just don't put everything in newsgroups.
I'm busy with those things for many years. For me this is an asset.
Put this in a group and tomorrow it's all over the net and most feel almost guilty when referring to sources.

But I need time to put some things on web pages.
All those in-depth questions via newsgroups just overwhelms me.

I type with one finger, in a strange language, with a slow mind.

Best regards.
Eric
Last edited by Eric.Desart on Mon Feb 23, 2004 4:56 pm, edited 2 times in total.
Eric.Desart
Moderator
 
Posts: 2461
Joined: Tue Feb 10, 2004 4:29 am
Location: Antwerp, Belgium

Postby Eric.Desart » Sun Feb 22, 2004 5:02 pm

Maybe something more:

When I speak about stiff and flexible walls this relates to the coincidence frequency, in US documents also referred to as the critical frequency.

In my language in acoustics we distinguish between:
"buigslappe wanden" and "buigstijve wanden"

literally translated this becomes:
"bending soft wall (separation)" and "bending stiff walls (separation)"

So we, but also the German and other languages have specific acoustic notions for those concepts.
I asked that long ago in the science group if there are English equivalents for those specific acoustic notions.
As far as I could gather from US acousticians there aren't.
So the closest is a flexible (bending soft) and a stiff (bending stiff) separation.

In general for normal building practices (but that can deviate from the author, since it is a bit subjective) one calls a separation with a coincidence (critical) frequency <= 100 Hz a stiff (buigstijve) wall, and a separation with a coincidence >= 2500 Hz a flexible (buigslappe) wall.

Acoustically this is important:

Above coincidence a wall act as a piston. The wavelenght of the incident sound is smaller than the bending waves in the wall. This causes every bit of vibration to be converted again on the receiving site to sound radiation. One speaks about a radiation degree of 1. In fact one can measure the sound radiated by the wall by vibration measurements and recalculate this vibration in intensity and sound radiation.
Tipical examples of walls with coincidence below 100 Hz are:
Brique walls, concrete walls etc.
So in fact one uses such a separation in the insulation range above coincidence.

Below coincidence the wavelength of the incident sound is larger than the bendingwaves in the panel. This causes the radiation degree of such a panel to become smaller then 1. In fact only part of the vibration is converted to radiated sound, and part is cancelled directly in front of the panel. What happens is that the air directly in front of the panel toggles between the crests and the troughs of those bending waves in the panel. As such overpressure neutralizes underpressure. You can call this natural anti-noise.
Defining the sound radiation on the receiver side with aid of vibration measurements becomes a complex and easily inaccurate business. For any practical purpose one can say that this vibration technique is almost useless. For such type of wall one must resort to intensity measurements to define the relative contribution in the total receiving pressure level (in situ situations, not in labs since they are designed to make sure that all sound passes the "to be measured" partition).
Typical examples are: Gypsum board, wood panels, particle board etc.

This lower radiation degree is part of the reason those lightweight panels are so efficient in function of mass.
So in fact one uses such a separation in the insulation range below coincidence.

At coincidence the wavelength of the incident sound equals the bendingwaves in the panel. Here one get reenforcement. This causes this dip at 2000 Hz to 3000 Hz in gypsum panels (depends on thickness + properties).

However if one uses gypsum blocks, or cellular concrete or thin brique walls or concrete panels, then this coincidence comes in the low-mid to mid frequent area.
The negative effect (depending on material properties) can extend over 2 octaves and more. We call that the insulation plateau. Such a wall is neither one nor the other principle, with a weak area separating both principles.
That's also why you can't glue gypsum panels together because it causes this coincidence to lower to a more significant area in the frequency spectrum. Doubling the thickess of a gypsum board (using 2 panels glued together) will HALVE the coincidence. Not gluing them will cause two panels to give you the benifit of the increased mass without the disadvantage of the lowered coincidence frequency.

So and that's what I referred to with Gösele:
He tells how to calculate the mass-spring resonance for heavy brique walls and for flexible walls, but what with this gray area?

Is this a limitation: Yes it is.
But all the rest are completely wrong in this sence that they calculate the mass-spring for double leaf heavy brique or concrete walls in exactly the same manner as a drywall. Standard acoustic books and courses just ignore the influence of this coincidence effect (stiff versus flexible).

Eric
Last edited by Eric.Desart on Sun Feb 22, 2004 6:36 pm, edited 8 times in total.
Eric.Desart
Moderator
 
Posts: 2461
Joined: Tue Feb 10, 2004 4:29 am
Location: Antwerp, Belgium

Postby Bob » Sun Feb 22, 2004 5:06 pm

Hi Eric:

Can you give the link where I find this quote.


http://tinyurl.com/34yrp

The quote is assembled from several posts in that thread. Most of it is yours, and some of it is written by others, but it's all from that thread.

As for my question de jour re flexible/rigid walls, don't waste your time researching it on my account. But if you stumble across something in the coming weeks or months on the topic while you're doing other things, please know that I'd be interested in it.

Hey, that last message of yours wasn't there a second ago! I'll just go and read it now.
Regards
Bob Golds
"The only thing we regret in life is the love we failed to give."
"Be a rapturist -- the backward of a terrorist. Commit random acts of senseless kindness, whenever possible" - Jake Stonebender
Bob
 
Posts: 4360
Joined: Sun Feb 08, 2004 4:37 am
Location: Oakville, Ontario, Canada

Postby Eric.Desart » Sun Feb 22, 2004 5:11 pm

Yes, I saw, we crossposted :):):) and after writing such things I start editing.

Eric
Eric.Desart
Moderator
 
Posts: 2461
Joined: Tue Feb 10, 2004 4:29 am
Location: Antwerp, Belgium

Postby Bob » Sun Feb 22, 2004 9:45 pm

Hi Eric:

Each paragraph in this post is an unrelated thought.

It took me a couple of reads to realize you were not talking about the wall system resonance (e.g. wall system = 2 layers gypsum + 2x4 + insulation + 2 layers gypsum = 5 inches ) which should be about 65hz (mass-air-mass), but rather the effect on a single panel (half an inch of gypsum) and that dip in the wall system TL chart around 3khz.

I would have thought that although the real TL with two unglued panels of gypsum would be higher, given
C:\Temp\Home Theatre\Loudness Phons Threshold Fletch-Munson Robinson-Dadson.htm
that the perceived TL would be lower. i.e. If they are not glued together then the coincidence dip in the TL chart (as shown by insul48SA.exe ) at 2k to 3k hz, is the same frequency range that is at the low point in the sensitivity region for the Fletch-Munson curve, and halving the coincidence to 1k to 2k would move the perceived TL up a couple of db in the Fletch-Munson curve, giving a human ear perceived TL benefit.

Just before yahoo went offline I posted a note about using undrying glue between layers of drywall, for a constrained layer damping effect.

The question I was asking about flexible/rigid was about 'boomyness', which to me meant it's effects in the bass region of 15hz to 80hz. So I guess we were talking about different effects.
Regards
Bob Golds
"The only thing we regret in life is the love we failed to give."
"Be a rapturist -- the backward of a terrorist. Commit random acts of senseless kindness, whenever possible" - Jake Stonebender
Bob
 
Posts: 4360
Joined: Sun Feb 08, 2004 4:37 am
Location: Oakville, Ontario, Canada

Postby Eric.Desart » Sun Feb 22, 2004 9:53 pm

I'll come back on this later:

Just short:

mass-spring resonance: indeed resonance caused by two leafs connected by a spring (air layer).
coincidence frequency: property of the single individual leaves = resonance where the wavelength of the incident soundwaves equal the wavelength of the bending waves in the panel itself.

BUT THIS INDIVIDUAL LEAF BEHAVIOR ALSO INFLUENCES THE MASS-SPRING FREQUENCY OF THE 2 LEAVES COMBINED AS A DOUBLE WALL.

THE MASS-SPRING CALCULATION FOR FLEXIBLE DOUBLE LEAF WALLS IS DIFFERENT THAN THE MASS-SPRING CALCULATION FOR STIFF DOUBLE LEAF WALLS.
And the notions flexible and stiff must be interpreted in function of the coincidence of the single leafs (not how stiff it feels).

And it is Gösele that braught the fact that this individual leaf property (coincidence: stiff/flexible) influences the mass-spring (double leaf property) to the attention of the acoustic world.
That's why you see my previous message as unrelated, but it isn't:
To understand what Gösele tells in function of a double leaf behavior (mass-spring) you must understand how he distinguishes the individual leaves (stiff/flexible).

Does this sound better?

Eric.

PS: This becomes difficult.
It's as giving acoustics without a logical sort order.

One should start:
1) How does a single leaf behaves. What parameters influences this?
2) How does a double leaf system behaves?
3) How do the properties of the single leafs influence this double leaf behavior?

But then you start better with reading books covering architectural or building acoustics.

I've read your question further.

In fact it has no sense to continue until you understand my other messages about single-number ratings.
In fact to do acoustics you first start to learn how to calculate logaritmically.
Then you can see the influence of different sources on the overal TL.

Bob you want to know a lot. And I really do appreciate that.
But in order to get a good understanding you start with step 1, then 2, then3 and so on.
You pick arbitrary things from the net trying to get a picture.
It becomes as reading a book starting at chapter 17, then 10, then 23, then 2 and so on.
If you really want to understand in-depth you just have no other choise than learning systematically.

Summary:
As long as you continue talking about STC and do not understand why I say that this had nothing to do with music.
As long as you can't calculate the overall TL from a wall in function of a specific source.
Then in fact we will continue walking in circles.

So get my Acoustic selector Excel file from my site, and only when you understand what this file does, we can build further.
Last edited by Eric.Desart on Mon Feb 23, 2004 4:45 pm, edited 1 time in total.
Eric.Desart
Moderator
 
Posts: 2461
Joined: Tue Feb 10, 2004 4:29 am
Location: Antwerp, Belgium

Postby Eric.Desart » Mon Feb 23, 2004 9:34 am

Bob,

What I meant with learning to calculate the overal TL of a wall in function of the spectrum of a specific source.

Suppose you have a coincidence dip of 15 dB at 2500 Hz, but you have still 40 dB TL left.
Suppose the low frequencies give only 18 dB TL.
What does bother me this coincidence dip at 2500 Hz when the level on the receiving site is completely dominated by the basses coming through the wall.

What sense does it have to concentrate on this 15 dB dip, when this only influences this overal TL level with minor fraction of a dB.

I indeed almost get nervous from all those special things.
If you want to solve the worst part of this dip, combine different types of boards or thicknesses. They will have different bending wavelengths, as such damping one another.
1) Even when doing this it will have minor impact on the overal insulation for music, but you could do it because it a cheap solution, to surpress this phenomenon.

Anything better or more special is almost nonsens.
The idea is keep the coincidence high enough that it remains in a range were little sound energy is. And use the cheapest solution if needed to surpress this dip somewhat.

And all the magic is only possible and attractive as long as you can't calculate the relative impact/contribution of this dip in the overall insulation.
Try first to get this basic picture, then you see what frequencies are important.

If you check my Excel file, you'll notice that I allow user to alter TL values of individual frequency bands of existing measurements.
I've made this file like that, just to allow users to get the feel what happens if they could improve certain frequencies (or worsen them).

I doubt, if you should damp ALL coincidence frequencies by improving TL at those very frequencies as if this phenomenon even doesn't exist, that the overal TL for music will improve more than 0.5 dB even for the worst case of those 50 measurements.

A lots of those sites concentrate on individual phenomena in order to sell products or solutions, counting on the fact that the visitor is not able to rate the relative importance of the described phenomenon in the larger picture.

The problem with music is that the glasses start traveling in the cupboard at the neighbors, that the water in the fish tank starts wrinkling, in other words: the basses are.
I never in my live heard somebody saying: Damned I'm disturbed by the high frequent harmonics of my neighbor's music (if there are no holes of course).

Eric
Eric.Desart
Moderator
 
Posts: 2461
Joined: Tue Feb 10, 2004 4:29 am
Location: Antwerp, Belgium

Postby Eric.Desart » Mon Feb 23, 2004 3:48 pm

Bob,

In fact you touched several topics simultaniously.

You referred to Gösele:
I think I explained that.
Summary:
Gösele tells that the mass-spring calculation for stiff walls is different than for flexible walls.
He is right, I tested that. The whole acoustic world still ignores this and calculates mass-spring without distinguishing between stiff and flexible, using an algorithm (calculation method) mainly valid for flexible walls.

You referred to the Coïncidence and the Loudness Curves:
There I said first try to check the relative impact of this dip in the overal TL before even being interested in whatever magic. You will notice that for music this phenomenon is really mostly insignificant.
More important: Almost without extra cost you can easily damp this phenomenon by using boards showing different natural bending waves. Just adding a thin board or using different thicknesses will do the job.

You referred to boominess etc.
I explained how a double leaf system works and made the analogy with silencers before.
A double leaf wall is a panel trap. So the absorption increases downwards towards the mass-spring resonance where it is at it's maximum. Going further down in frequency, the absorption diminishes again. The wall will start behaving as a single leave wall with a mass equaling the total mass of both leaves together.
A wall TL curve is really an upside-down flipped graph you can find for silencers as I explained already some 2 months back (I know you saw those Sylomer threads).

Very roughly: Where the absorption is highest the TL value is lowest and vice versa. So you can say that both behave opposite: > TL is < absorption; < TL is > absorption.

Warm regards
Eric
Eric.Desart
Moderator
 
Posts: 2461
Joined: Tue Feb 10, 2004 4:29 am
Location: Antwerp, Belgium

Postby Bob » Mon Feb 23, 2004 7:12 pm

Hi Bob,

I figure that if one solves bass TL problems, the higher frequencies will take care of themselves because the the high frequency TL is always so much larger than the low frequency TL in any wall system. I believe this is similar to your attention to the Music-STC.


We have a dialectict saying here:
"Ha uwe frank is gevallen"

Literally translated:
"Ha your dollar is dropped", meaning: finally you got it:)

But not yet complete:
You can have a higher STC of wall x versus z, while simultaniously having a lower music insulation.
And that's what those single-number ratings are all about.
Calculating the overal TL value weighted for a noise/sound source with a specific spectral distribution.

Regards
Eric
Regards
Bob Golds
"The only thing we regret in life is the love we failed to give."
"Be a rapturist -- the backward of a terrorist. Commit random acts of senseless kindness, whenever possible" - Jake Stonebender
Bob
 
Posts: 4360
Joined: Sun Feb 08, 2004 4:37 am
Location: Oakville, Ontario, Canada

Postby Eric.Desart » Mon Feb 23, 2004 7:34 pm

BOB

A MILLION TIMES SORRY,
BY ACCIDENT I EDITED YOUR MESSAGE.
SO IT's LOST.

I WANTED TO QUOTE BUT HAVE DONE SOMETHING WRONG HERE.
PLEASE FORGIVE ME................ PLEASE
I REALLY DIDN'T MEAN TO.

A VERY GUILTY GUY

SORRY, SORRY, SORRY
Eric
Eric.Desart
Moderator
 
Posts: 2461
Joined: Tue Feb 10, 2004 4:29 am
Location: Antwerp, Belgium

Postby Bob » Mon Feb 23, 2004 7:38 pm

Hi Eric:

ha ha ha ha :-)
Don't worry about deleting the post. (You're totally forgiven)

You'd have to delete the whole site before I'd get mad. (Been there, done that). VBG

Besides, although it might have contained something relevant to others, it was addressed to you, and at least you read it.
Regards
Bob Golds
"The only thing we regret in life is the love we failed to give."
"Be a rapturist -- the backward of a terrorist. Commit random acts of senseless kindness, whenever possible" - Jake Stonebender
Bob
 
Posts: 4360
Joined: Sun Feb 08, 2004 4:37 am
Location: Oakville, Ontario, Canada

Previous

Return to Discussion

Who is online

Users browsing this forum: No registered users and 5 guests

cron