STC Test Data for Partitions

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

STC Test Data for Partitions

Postby jazzbone11 » Tue Feb 17, 2004 5:38 pm

Greetings. I am looking for STC test data out there on the web. Through searching I have found some, but it is hit or miss. I can often find STC values and a description of the partition (which is somewhat useful), but would like to see the actual test data to identify what frequencies are the limiting factor in the rating.

Any sites/links people have would be appreciated (with a preference for metal stud and masonry constructions). As a side note, my pipe dream is to have software which will allow one to build a partition from scratch and approximate STC values based on empirical data (or otherwise)

Thanks.
jazzbone11
 
Posts: 1
Joined: Tue Feb 10, 2004 3:11 pm

Postby Bob » Tue Feb 17, 2004 6:29 pm

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 Savant » Tue Feb 17, 2004 8:52 pm

You can also try:

http://irc.nrc-cnrc.gc.ca/fulltext/ir761/

LOTS of data. (I got this link from our friends over at RO.)

If you're willing to spend some money, "Noise Control in Buildings" by Cyril Harris is a great book. Lots of data and a good read. You can purchase through the Institute for Noise Control Engineering: http://www.inceusa.org
---Σοφός---

10*bLog(P²)

"That buzzing noise means something." - Winnie the Pooh
Savant
 
Posts: 753
Joined: Mon Feb 02, 2004 1:52 pm
Location: Kansas

Postby knightfly » Wed Feb 18, 2004 12:11 am

I've been playing with the freebie calculator from Insul for a while, and it can do quite a bit for a "demo" - won't save, but you can change quite a few of the parameters and get at least a ballpark idea of what to expect. Example - changing the material properties from 690 kG to 2130 kG makes each 1/2" of gypsum work like 1/2" concrete, so you can choose 16 layers of "gypsum" to figure out an 8" concrete wall. Watch for panel size (the double arrow at the top of the page) as it's set to metric odd size. Also, using both layers of each leaf set to different thickness will show you the coincidence effect graphically -

I like this thing enough I'm considering buying the "real" one... Steve
knightfly
 
Posts: 215
Joined: Tue Feb 17, 2004 11:35 pm
Location: West Coast, USA

Postby Eric.Desart » Thu Feb 19, 2004 5:54 am

Jeff,

I know this data (drywalls) already a while.
I should like to get this data in Excel.

If so I can calculate all those walls for different types of sources.
As music, music with very low frequent content, traffic etc.

Only I started once but it's a bit a frusqtrating activity (mass work).

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

Postby Bob » Fri Feb 20, 2004 8:44 am

Hi Eric:

I should like to get this data in Excel. If so I can calculate all those walls for different types of sources. As music, music with very low frequent content, traffic etc.


OK. I've converted
http://irc.nrc-cnrc.gc.ca/fulltext/ir761/
from the PDF into EXCEL. I've also created an SQL file for inserting the data into a database so you can write queries like this

SELECT ElementCount, PageNumber, STC, hz50, hz63, hz100, hz80, hz100, hz125
FROM IRC761
WHERE STC > 50
and hz50 > 20
and ElementCount < 5
ORDER BY STC;

ELEMENTCOUNT PAGENUMBER STC HZ50 HZ63 HZ100 HZ80 HZ100 HZ125
4, 142, 51, 20.1, 14.6, 17, 15.1, 17, 27.3

The WinZip file named "IRC ir761.zip" can be found at
Temporary disabled by E. Desart

It contains 4 files:
1) IRC ir761.sql - which creates a database table and inserts the data into it.
2) IRC ir761 XLS.TXT - a comma separated text file of that data.
3) IRC ir761.xls - an excel 2003 file with the data
4) IRC ir761 excel 4.xls - an excel 4 file with the data

The table definition (and hence the columns in the excel file are)

create table IRC761 (
PageNumber number(3),
TestID VarChar2(10),
STC number(3),
Element1 VarChar2(99),
Element2 VarChar2(99),
Element3 VarChar2(99),
Element4 VarChar2(99),
Element5 VarChar2(99),
Element6 VarChar2(99),
Element7 VarChar2(99),
Element8 VarChar2(99),
Element9 VarChar2(99),
Element10 VarChar2(99),
ElementCount number(2),
Hz50 float,
Hz63 float,
Hz80 float,
Hz100 float,
Hz125 float,
Hz160 float,
Hz200 float,
Hz250 float,
Hz315 float,
Hz400 float,
Hz500 float,
Hz630 float,
Hz800 float,
Hz1000 float,
Hz1250 float,
Hz1600 float,
Hz2000 float,
Hz2500 float,
Hz3150 float,
Hz4000 float,
Hz5000 float,
Hz6300 float
);
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 Bob » Fri Feb 20, 2004 9:03 am

Here's a couple more sample queries

SELECT ElementCount, PageNumber, STC, hz50, hz63, hz100, hz80, hz100, hz125
FROM IRC761
WHERE STC > 50
and hz50 > 20
and ( lower(Element1) like '%fibre board%' OR
lower(Element2) like '%fibre board%' OR
lower(Element3) like '%fibre board%' OR
lower(Element4) like '%fibre board%'
)
ORDER BY STC

ELEMENTCOUNT PAGENUMBER STC HZ50 HZ63 HZ100 HZ80 HZ100 HZ125
7, 251, 51, 24.2, 18.8, 19.5 , 19.9, 19.5, 32.4
8, 253, 52, 26, 21.5 , 19.7, 21.4, 19.7, 32



SELECT ElementCount, PageNumber, STC, hz50, hz63, hz100, hz80, hz100, hz125
FROM IRC761
WHERE STC > 55
and hz50 > 20
and ( lower(Element1) like '%resilient%' OR
lower(Element2) like '%resilient%' OR
lower(Element3) like '%resilient%' OR
lower(Element4) like '%resilient%' OR
lower(Element5) like '%resilient%' OR
lower(Element6) like '%resilient%' OR
lower(Element7) like '%resilient%'
)
ORDER BY STC

ELEMENTCOUNT PAGENUMBER STC HZ50 HZ63 HZ100 HZ80 HZ100 HZ125
7, 225, 57, 22.5, 15.5, 24.5, 21.9, 24.5 , 32.6
7, 220, 58, 22.3, 17 , 26.3 , 23.6 , 26.3 , 34.2
7, 227, 59, 20.2, 15.8 , 27 , 22.6 , 27 , 34.8
7, 229, 59, 20.7, 14.2 , 25.7, 20.7, 25.7 , 34.9
7, 230, 59, 20.9, 15.5 , 27.4, 22 , 27.4 , 34.7


select elementcount, max(hz50)
from irc761
group by elementcount;

ELEMENTCOUNT MAX(HZ50)
3, 21.7
4, 22.7
5, 25.7
6, 26.8
7, 24.5
8, 30
9, 26.7


select pagenumber
from irc761
where elementcount = 8
and hz50 = 30;

PAGENUMBER
349
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 » Fri Feb 20, 2004 12:17 pm

Bob,

I couldn't enter the link (asking for password).

However I disabled this link for now.

This is a protected document which contains incredible expensive test data.
I do not think it's a good idea to put this free accessible on the net.

My intent was, to make somehow an Excel document from it, including much more calculations other than plain STC (little use for music).
This should include sort possibilities graphs comparison possibilities and whatever.

BUT MAINLY:
If it should work out:
Send FIRST a copy to NRC -
Halliwell, R.E.; Nightingale, T.R.T.; Warnock, A.C.C.; Birta, J.A.
Asking for their allowance to make this document public on the Net, with CLEAR reference to the Authors and whatever.

This was in fact an Internal NRC document not that long giving free.
I think without there explicit consent, we shouldn't release anything on the net.

I had this doc already converted to Excel, txt and Word format but the result was really poor, asking for too much subsequent programming to get the data usable.
+ I used some not so official methods.

So I ask you to send me a copy, but do not make it generally accessible.
I'll never publish anything without there explicit consent.

Anyhow many thanks for responding.
It's clear that you know what you're doing

Eric
Last edited by Eric.Desart on Fri Feb 20, 2004 2:05 pm, edited 1 time in total.
Eric.Desart
Moderator
 
Posts: 2461
Joined: Tue Feb 10, 2004 4:29 am
Location: Antwerp, Belgium

Postby Savant » Fri Feb 20, 2004 1:01 pm

Eric,

I think if you ask the folks at NRC, they might be able to supply a lot of the information to you in an Excel format. Most labs have it. Whether or not they release it might be a matter of how nice you ask! :):):):):)
---Σοφός---

10*bLog(P²)

"That buzzing noise means something." - Winnie the Pooh
Savant
 
Posts: 753
Joined: Mon Feb 02, 2004 1:52 pm
Location: Kansas

Postby Eric.Desart » Fri Feb 20, 2004 1:59 pm

Hello Jeff,

I played with the tought, even via the Univ to make it look better.
But I wanted to show them something first, which significantly increased the practical value of their data.
Otherwise I have to explain too much about the purpose.

You know yourself that one can't exclude chauvinisme completely.
And having data enhances their significance.
Still I MUST respect their copyright.
If one recalculates this document in measurement cost even excluding the study in itself that's enormous.

But thanks for the suggestion.
Eric
Eric.Desart
Moderator
 
Posts: 2461
Joined: Tue Feb 10, 2004 4:29 am
Location: Antwerp, Belgium

Postby Bob » Fri Feb 20, 2004 2:06 pm

Hi Eric:

I hoped that password required might only be anyone who is on hotmail. I did set the group to public and the 'pictures' part seemed to work ok.

As for whether it's 'public' or not, I'm of two minds. It can be found from their old public search engine at :
http://serpent.cisti.nrc.ca/dbtw-wpd/textbase/irc/
and typing 'Halliwell' in the author column.
This link says
Disclaimer
This database is for scientific research purposes only, and is not intended as a service to the general public. Due to the highly technical and dynamic nature of the content, expectations of adherence to Government of Canada Common Look and Feel and Official Language requirements may not be met.
The opinions and information presented here do not represent the views of the Government of Canada. The Government of Canada disclaims all liability for any actions taken as a result of information found on this database


But their new search engine at http://irc.nrc-cnrc.gc.ca/ircsearch.html has a link at the bottom that takes one to http://irc.nrc-cnrc.gc.ca/notices.html which says
Non-commercial Reproduction

Information on this site has been posted with the intent that it be readily available for personal and public non-commercial use and may be reproduced, in part or in whole and by any means, without charge or further permission from the Institute for Research in Construction. We ask only that:

Users exercise due diligence in ensuring the accuracy of the materials reproduced;
The Institute for Research in Construction be identified as the source department; and,
The reproduction is not represented as an official version of the materials reproduced, nor as having been made, in affiliation with or with the endorsement of the Institute for Research in Construction.


In my opinion it's ok to post the reformatted data publicly. Although some care should be taken to reference the original authorship by more than the filename.

Nevertheless, it's fine with me to wait until you have asked for permission, since it's more than just 'reproduced', it's reformatted.
If nothing else, perhaps they might enjoy that someone was interested in their work.

Eric, I went looking around for your email address but I can't find it. I'll see if I can find something on this forum, but if not please send me an email to BobGolds at hotmail dot com.
Last edited by Bob on Fri Feb 20, 2004 2:17 pm, edited 1 time in total.
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 Bob » Fri Feb 20, 2004 2:16 pm

Eric

Nope. I can't find your email address. Neither on this forum nor on your website.
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 » Fri Feb 20, 2004 2:51 pm

Hello Bob,

Sent you my Email address, and thanks.

It's correct I'm very sensitive about copyright and all this data distributed without sources or credit.

But again really thanks.

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

Postby Bob » Fri Feb 20, 2004 4:11 pm

Hi Eric:

It only took a couple of hours hours to write the program once I decided to do it.

In your case, if your website is any evidence, you like to manipulate data in Excel.
I prefer a relational database.

In your case you wanted the data to present it in a different form by adding additional data based on the original (musical STC).

In my case I found the document too long to find anything, and I wanted an index. My addition of the ElementCount is sort of simple-new-data as part of that index.

In recording.org Rod recently said he'd never use hemosite. This data shows that for the same number of elements, fiber board instead of gypsum has some advantages in lower frequencies. I know I'd never read anything like that before. For example, here's three systems each with 7 elements:
RC (page 223) reduces 19.5db at 50hz
RC stagared (page 309) reduces 17.9db at 50hz
fiber board (page 251) reduces 24.2 db at 50hz
Anyway, there's too many pages to compare all the combinations by eye. But in the database I might be able to do something. I might need to add some flags and layer type counts.
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 » Fri Feb 20, 2004 4:45 pm

Hi Bob,

It's correct I'm an Excel guy.

My main intent is not only to sort them, but also filter and sort them on overal STC but also versus several other weightings so one can sort and filter them in function of specific purposes.
It should be also easy to add custom sources to result in a custom weighting.

And one should be able to graphically check the weak points in the walls versus specific applications.
It is of little interest that wall x = 10 dB better than wall z at a certain frequency, when this frequency does not or hardly contribute in the overall insulation.

It's good to know exactly what the insulation shortage is per frequency band in order to obtain an overall weighted single number insulation rating of x dB.

In fact often is said that STC represent voices for office purposes.
While partly true, this is not how STC came into being.

I even do believe (almost sure) that STC (and Rw) originally is deviated from German standards via..., via...
In fact the weighting of the ASTM STC and the comparable ISO Rw result from the typical stylized insulation curve of a heavy brick wall (single leaf), which was seen as a good separation between dwellings.

You find the typical behavior of a single leaf wall in the low frequencies, than smoothed towards a flat area (flanking).
The resemblence with a dB(A) weighting and human voices is more accidental.

And the manner it is calculated by shifting a reference curve until certain conditions are met, is not because that should be a good mathematical or physical approach but because those standards are designed in a period that logarithmic calculations where only accessible for few. So accurate energetic (logarithmic) calculations were substituted by arithmetic and graphical alternatives. Those conditions defining the way the curve is shifted are meant to limit errors.
In nowadays terms those standards (STC and RW, but also here in Europe National standards) are a rather primitive calculation approach.

EDITED AND ADDED:
To explain this in simple terms:
Often to STC is referred as some kind of overal quality.

What does those single-number Transmission Loss ratings in fact do, or try to do?

The idea is simple:
One assumes a certain type of sound/noise on the emission (source) side of a partition.
The partition itself has a certain TL curve per frequency band.
On the immission (receiver) side one measures the level per frequency band.
One totals all those bands to get the total immission level.

The difference of the totals between both sides is the overal TL value.
This is corrected with a slight correction factor to normalize this measurement.

So in fact the Single-number rating is nothing else than this overal TL value.

Since this of course depends of the frequency content of the source at the emission side (e.g. music with low frequencies, human voices, whatever) one uses standardized weightings. As such one can see the source for STC close to Human voices or a dB(A) weighting).
This means that an STC just gives the overall TL for this weighting curve representing voices (in fact this heavy brick wall).
But the heck: if you want to compare or rate walls for music with a lot of low frequent content, who then is interested in knowing the overall TL for voices?

Therefore, knowing this limit, and the fact that this is basically senseless, the MTC was designed to cover music. But this was in fact still the STC but with a few corrections to cover the influence of the 2 lowest frequencies.
Since this method is as primitive as the STC itself, this never became an official standard.

The first modern ASTM standard was the OITC.
They introduced this to cover for traffic noise, knowing that STC wasn't good.
The calculation method is simple modern and accurate.
They just integrated a statistical source curve more representative for average Outside-Inside noise.

Thinking this through in fact the correct single-number weighting is this overal TL value which is closed to a source curve resembling the real live source at hand.

As such e.g. in the Netherlands they have several single number ratings, which is in fact nothing else than that they put a representative curve as source.
As such they have a similar complicated old one (say the Netherlands STC version), but they have additional modern ones for:

Train traffic,
Road traffic,
Air traffic,
Pop music,
House music.

Why?
Very simple: if you want to compare isolation of a partition for e.g. train traffic, who the hell is interested what this partition is worth for voices or house music?

Summary:
A single number TL rating represents the Transmission Loss of a wall expressed in one single dB value assuming a specific statistical averaged sound/noise source.
A STC value is a bit a primitive arithmetic graphical version of the same.

Such reference curves to represent certain specific sound/noise are based on enormous amounts of measurements of such related sources and averaged after exclusion of abnormal exeptional measurements. Those reference curves also make it easier to design things: one knows what one works at.

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

Postby Bob » Fri Feb 20, 2004 10:52 pm

Hi Eric:

And here I thought only me and thee would be interested in this stuff.
I've had a couple of requests via my email account for the file already.

Have you sent off a request to make it public to the authors?
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 » Fri Feb 20, 2004 11:45 pm

Bob,

No , And that can easily take a month or two.

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

Postby Rod Gervais » Sun Feb 22, 2004 2:49 am

Bob wrote:
In recording.org Rod recently said he'd never use hemosite. This data shows that for the same number of elements, fiber board instead of gypsum has some advantages in lower frequencies. I know I'd never read anything like that before. For example, here's three systems each with 7 elements:
RC (page 223) reduces 19.5db at 50hz
RC stagared (page 309) reduces 17.9db at 50hz
fiber board (page 251) reduces 24.2 db at 50hz
Anyway, there's too many pages to compare all the combinations by eye. But in the database I might be able to do something. I might need to add some flags and layer type counts.


Bob,

I've had this report for quite a while now - but have not had the time to review it in it's entirety.

I am going to have to make it a point as this question comes up a lot - and i really want to give the most up to date information i can.

I will never reccomend any assembly which hasn't been tested (i guess I'm weird like that) however this does change things in relation to homosote.

Thanks for the heads up.

Rod
Rod Gervais
 
Posts: 1035
Joined: Sun Feb 22, 2004 2:34 am
Location: Central Village CT USA

Postby Bob » Sun Feb 22, 2004 4:48 am

Hi Rod:

I wrote "This data shows that for the same number of elements, fiber board instead of gypsum has some advantages in lower frequencies. "

This report talks about fibre board, not specifically Homosote. The word Homosote/Homasote/Hemosote is not mentioned in the report. Although I mentioned an example, even in that example the values were an odd peak (i.e. 63hz the next frequency up didn't make that much of a difference). I really didn't expect you to ask people to stop building walls on a weekend.

The combination of the difficulty in comparing that PDF, and that it's probably much easier for me to analize it from a numbers only perspective than for anyone else around here, and of course I mentioned it like it was a fact when actually it was my intention merely to say that I think I saw something I wasn't expecting as I flipped through the hundreds of pages at high speed, behooves me to prove (or disprove) my quoted statement above with better examples than the one that I gave.

I'm going to spend a little bit of time tonight trying to improve my database index, so that I can compare fiber board to gypsum, using apples to apples, which in this case means otherwise identical constructions of the same sizes and same insulation, with only the gypsum layer being exchanged for fibre board. With any luck I'll be able to give you examples of pages to compare, and an average of the difference in TL at various frequencies. But we'll see what I can do.
Last edited by Bob on Sun Feb 22, 2004 5:50 am, edited 3 times in total.
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 Bob » Sun Feb 22, 2004 5:12 am

Hi Rod:

I regret there's no way to do the comparison I'd hoped for.

There are only 5 wood fibre wall systems in the report
Page 250 G16_WFB19_WS90(610)_GFB90_WFB19_G16
Page 251 G16_WFB19_WS90(610)_GFB90_WFB19_2G16
Page 252 G16_WFB19_WS90(610)_GFB90_WFB19_2G16
Page 253 2G16_WFB19_WS90(610)_GFB90_WFB19_2G16
Page 364 G16_WS90(406)_GFB65_AIR6_WFB19_WS90(406)_GFB65_G16

For 250 through 252 there's nothing to compare because there are no wood stud walls without RC using double 16mm gypsum in this report. There are double 13mm gypsum layers on wood studs. So I can't compare apples to apples.

The closest I could find was
Page 250 G16_WFB19_WS90(610)_GFB90_WFB19_G16 STC-49 50hz-22.7 63hz-16.5 80hz-18.4 100hz-18.7 125hz-31.1
Page 228 2G16_WS90(610)_GFB90_RC13(406)_2G16 STC-57 50hz-19.6 63hz-16.6 80hz-22.4 100hz-24.9 125hz-33.4
Page 229 2G16_WS90(610)_GFB90_RC13(610)_2G16 STC-59 50hz-20.7 63hz-14.2 80hz-20.7 100hz-25.7 125hz-34.9
which shows a minor benifit around 50hz over RC, but the STC values clearly show that in the higher frequencies (80hz and up) gypsum+gypsum+RC wins over wood fiber + gypsum.

Sliding down to the 13mm gypsum we have
Page 223 2G13_WS90(610)_GFB90_RC13(610)_2G13 STC-60 50hz-19.5 63hz-15.7 80hz-23.9 100hz-29.4 125hz-35.5
which strangely performed better than the Page 229 above.
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

Next

Return to Discussion

Who is online

Users browsing this forum: No registered users and 1 guest