The No.1 Website for Pro Audio
basic question regarding sound absorption coefficients
Old 20th May 2010
  #1
Lives for gear
 
audiothings's Avatar
 
🎧 15 years
basic question regarding sound absorption coefficients

Going by the following info on sound absorption coefficients which I found online, 6" of OC703 directly on the wall absorbs everything above 125 Hz. This doesn't make sense. What am I getting wrong here?

Thanks,

125 Hz - 1.19

250 Hz - 1.21

500 Hz - 1.13

1000 Hz - 1.05

2000 Hz - 1.04

4000 Hz - 1.04
Old 20th May 2010
  #2
Lives for gear
 
avare's Avatar
 
🎧 15 years
Quote:
Originally Posted by audiothings ➑️
What am I getting wrong here?
Nothing. What is the point of confusion?

Andre
Old 20th May 2010 | Show parent
  #3
Lives for gear
 
Dange's Avatar
 
🎧 10 years
Andre,

Whenever I see absorption coefficients greater than 1, I wonder if it's the measurement method in the lab or the fact that the standards seem to require the use of Sabine's RT formula, rather than Eyring's or Millington-Sette. (Sabine's does not give accurate results for high absorption). Maybe it's both....what do you reckon?
Old 20th May 2010 | Show parent
  #4
Gear Guru
 
Ethan Winer's Avatar
 
🎧 15 years
Lightbulb

Quote:
Originally Posted by Dange ➑️
Whenever I see absorption coefficients greater than 1...
Maybe this will help:

Alternative Test Methods for Acoustic Treatment Products

--Ethan

________________
The Acoustic Treatment Experts
Old 20th May 2010 | Show parent
  #5
Lives for gear
 
avare's Avatar
 
🎧 15 years
Quote:
Originally Posted by Dange ➑️
Andre,Whenever I see absorption coefficients greater than 1, I wonder if it's the measurement method in the lab or the fact that the standards seem to require the use of Sabine's RT formula, rather than Eyring's or Millington-Sette. (Sabine's does not give accurate results for high absorption). Maybe it's both....what do you reckon?
Great post summing up a lot of confusing things to people learning acoustics.

The biggest factor involved is that people confuse Sabines,which are units of measurement with absolutes of the absorption of an open space.

Second is confusion of edge diffusion with edge absorption. BBC tests of raw material absorption are with the edges covered. There are still absorption co-efficients greater than one. At dimensions similar to and less then the wavelength, the sound wave sees a hole where the absorbent is, and sound around it is diverted towards it. Somewhat how a whirpool forms around the top of the drain in a sink. That is a poor technical explanation. I someone can give a better one explaining edge difraction.

Third is as you described the limitation of the calculation equation itself.

I hope this helped somewhat.

Andre
Old 20th May 2010 | Show parent
  #6
Lives for gear
 
audiothings's Avatar
 
🎧 15 years
Quote:
What is the point of confusion?
i thought it would have been harder to deal with 125 Hz...

it seems, and also the link that ethan pointed to seems to suggest that - below 125 Hz is where the challenge is at.

so... above 125 Hz can be handled by simply putting 6" of OC703 against the wall? (of course, there are other considerations... diffusion, liveliness etc.)

if this were true (as it probably is), all a small voice recording booth needs is 6" of insulation on the walls?

Thanks,
Old 20th May 2010 | Show parent
  #7
Lives for gear
 
avare's Avatar
 
🎧 15 years
Quote:
Originally Posted by audiothings ➑️
i thought it would have been harder to deal with 125 Hz...

it seems, and also the link that ethan pointed to seems to suggest that - below 125 Hz is where the challenge is at.

so... above 125 Hz can be handled by simply putting 6" of OC703 against the wall? (of course, there are other considerations... diffusion, liveliness etc.)

if this were true (as it probably is), all a small voice recording booth needs is 6" of insulation on the walls?
Thanks for the clarification of the confusion.

There is no cut off at 125 Hz. Just a lack of widespread test results. Adding to the lack data at those frequencies, is the lack of knowledge of teh interaction of the room and the absorbers.

Getting to you vocal booth questio, which I think is your main query, 6" of 703 like material would be great for vocal booth walls. What many people are not aware of is that the absorption data normally reported is for diffuse field conditions. That means sound coming from directions towards the test material. When sound comes only directly at the material, also called normal incidence, the absorption effectiveness of the material is reduced. For practical purposes, the absorption starts to reduce about an octave higher than for diffuse field. So if one takes 4" of 703 as being flat in absorption down to 125 Hz in diffuse field, then it will be flat down to 250 Hz with normal incidence. Using 6" instead of 4" will shift this down a little under half an octave. With a female voice, this is an anechoic space for the entire vocal range. For a male voice it is somewhat reflective for the lowest octave.

HOWEVER, just about nonw of the sound in a vocal booth will be at normal incidence, so for practical purposes the booth would indeed be completely anechoic, that is echo free.

I hope the long answer helped provide usefull information for you.

Andre
Old 20th May 2010 | Show parent
  #8
Gear Nut
 
Insulation Man's Avatar
 
🎧 10 years
If you put 6" of 703 board on the wall, you won't hear a thing on the other side!

With Sound Absorption Coefficients, you basically want to get close to 1.0 on whatever frequencie band you are trying to absorb. If you find yourself using products that have over 1.0, its basically over kill. Although there is no problem with having 6" of insulation on your wall. Well, if space is an issue, then I guess that is a problem and the additional cost of 6" rather than 4" or 3".
Old 20th May 2010 | Show parent
  #9
Lives for gear
 
audiothings's Avatar
 
🎧 15 years
Quote:
If you put 6" of 703 board on the wall, you won't hear a thing on the other side!
not sure what you mean, but i think we are all agreed that it is mass that stops sound from getting to the other side, and OC703 is not known for its mass.
Quote:
I hope the long answer helped provide usefull information for you.
Andre, any answer you give is treated with respect and gratitude, from here.
Old 20th May 2010 | Show parent
  #10
Lives for gear
 
fabricaudio's Avatar
 
🎧 10 years
Quote:
Originally Posted by audiothings ➑️
not sure what you mean, but i think we are all agreed that it is mass that stops sound from getting to the other side, and OC703 is not known for its mass.

Correct and decoupling too but

YouTube - fabricaudio test #1


If you have space you can use rockwool too.



Nikolas
Old 20th May 2010 | Show parent
  #11
Lives for gear
 
Dange's Avatar
 
🎧 10 years
Quote:
Originally Posted by Insulation Man ➑️
With Sound Absorption Coefficients, you basically want to get close to 1.0 on whatever frequencie band you are trying to absorb. If you find yourself using products that have over 1.0, its basically over kill.
This is not correct. Absorption coefficients go from 0-1. Anything over 1 is actually an error in measurement. This has been explained earlier this thread by Andre (Avare).
Old 20th May 2010 | Show parent
  #12
Lives for gear
 
Dange's Avatar
 
🎧 10 years
Quote:
Originally Posted by avare ➑️
Second is confusion of edge diffusion with edge absorption. BBC tests of raw material absorption are with the edges covered. There are still absorption co-efficients greater than one. At dimensions similar to and less then the wavelength, the sound wave sees a hole where the absorbent is, and sound around it is diverted towards it. Somewhat how a whirpool forms around the top of the drain in a sink. That is a poor technical explanation. I someone can give a better one explaining edge difraction.
Another way is to say it is that areas of the test sample appear bigger than they actually are when tested. The sample absorbs more than it should for it's size due to diffraction at the edges.

Say you tested a 5m^2 sample. Then used the results for predicting a 1m^2 sample size's performance. You'll see differences. The 1m^2 will absorb more than expected....I think I've got the right way round....
Old 21st May 2010 | Show parent
  #13
Lives for gear
 
avare's Avatar
 
🎧 15 years
Quote:
Originally Posted by Dange ➑️
You'll see differences. The 1m^2 will absorb more than expected....I think I've got the right way round....
You have it. It is a difficult concept to explain.

Andre
Old 21st May 2010 | Show parent
  #14
Lives for gear
 
Rod Gervais's Avatar
 
🎧 15 years
Quote:
Originally Posted by Dange ➑️
This is not correct. Absorption coefficients go from 0-1. Anything over 1 is actually an error in measurement. This has been explained earlier this thread by Andre (Avare).
You must be talking about a different avare then the Andre I know - because he never said any such thing - perhaps you should reread his post........

And I do know the real Andre...........

Rod
Old 21st May 2010 | Show parent
  #15
Lives for gear
 
Dange's Avatar
 
🎧 10 years
Quote:
Originally Posted by Rod Gervais ➑️
You must be talking about a different avare then the Andre I know - because he never said any such thing - perhaps you should reread his post........

And I do know the real Andre...........

Rod
Post #5?

On reflection, I think 'anomaly in measurement' is better than 'error in measurement'.

Absorption coefficients have to go from 0-1. In calculations any quoted as greater than 1 should be inputted as 1. For example if you used the values in image source or ray tracing software.
Old 21st May 2010 | Show parent
  #16
Lives for gear
 
🎧 10 years
Does anyone know about any absorbers that operate exclusively on the basis of diffractive absorption? A wooden panel (or three-dimensional matrix) riddled with voids in the appropriate configuration would be "pretty awesome."
Old 21st May 2010 | Show parent
  #17
Lives for gear
 
avare's Avatar
 
🎧 15 years
Quote:
Originally Posted by Dange ➑️
Absorption coefficients have to go from 0-1. In calculations any quoted as greater than 1 should be inputted as 1. For example if you used the values in image source or ray tracing software.
It is not an easy reality to comprehend. Absorption coefficents greater than 1 are real. Where the problem occurs, beyond understanding it, is when using calculation methods based on the reflective component of what an absorber does. This is what much software does. This is obvious looking at the equations if they have terms using the absorption values where the absorption is subtracted from one, and/or a conversion to the natural log of that value.

Absorption of sound is too complex to be accurately determined by first priniciples, which is why actual test results are so important in acoustics.

Andre
Old 21st May 2010 | Show parent
  #18
Lives for gear
 
Rod Gervais's Avatar
 
🎧 15 years
Quote:
Originally Posted by Dange ➑️
Post #5?

On reflection, I think 'anomaly in measurement' is better than 'error in measurement'.

Absorption coefficients have to go from 0-1. In calculations any quoted as greater than 1 should be inputted as 1. For example if you used the values in image source or ray tracing software.
Sorry,

perhaps you should read it again - he never stated there was either an error or an anomaly - in fact he stated that the BBC reports with the edges covered also reported numbers greater than 1.0

As he tried to explain - he stated:

Quote:
The biggest factor involved is that people confuse Sabines,which are units of measurement with absolutes of the absorption of an open space.
Now - in layman's terms: a Sabine reflects the amount of sound that would pass through an opening of 1' square foot - however - suppose the material (when tested) exhibits properties that would indicate it would absorb, diffuse - or a combination of the 2 - GREATER than the sound that would pass through a 1 square foot opening - does that suddenly mean that the opening got larger when all things are measured according to the standards?

No it does not - the reported figures are the reported figures.

They are not an anomaly - they are not a mistake - they simply are what they are............

Material "A" at frequency "C" attenuates the equivalent of .91 Sabine -

Material "B: at frequency "C" attenuates the equivalent of 1.1 Sabine.

Everything is measure equivalent to - not in absolutes.......

Now, what you use for calculation is a different story altogether - but that has nothing to do with your claim...

Test 2 rooms of equal volume but different dimension - or even 2 rooms of equal volume and exactly the same dimension - but with slightly different construction configurations - place exactly the same room treatments in the room and measure - and you will get 2 completely different results.....

In fact - even in 2 rooms with exactly the same construction and volume - you will still get different results.

Want one better? - take 1 room - put the treatments in it - set everything up and test the room - test it again the next day (without moving or adjusting the gear or treatments) - and the hands down odds are that you will not measure the same results as you did on the first day.

This is the reason that the standards for reverb room construction are so tight - not only must the room be constructed painstakingly to the standards - but the tracking of temperature - barometric pressure - humidity levels - exact location of the materials being tested, microphone and speaker locations, etc., etc., etc. all have to be taken into account in order for the data to be meaningful and reproducible.

This so that you could test in another lab and achieve the same results.

You will never reproduce the lab results outside of the lab.

The numbers are a guideline to allow you to compare results from different products fairly - they are not intended for you to use with the intent that you will achieve those numbers......

So it would not make any sense to bother with a starting point of anything over 1............

Also - just so we're clear on this - even with the most sophisticated sound analysis software - and here I am talking programs running in the 10's of thousands of dollars - the analysis is an approximation of what to expect in the room - not an absolute...... the programs cannot take into account any of the natural anomalies that will take place during construction - which is why there is always a need for room testing after the fact........

I hope this helped..........

Andre can always correct me if I am mistaken of course............ in fact I would appreciate his reply on this subject.......

Rod
Old 21st May 2010 | Show parent
  #19
Lives for gear
 
avare's Avatar
 
🎧 15 years
Quote:
Originally Posted by Rod Gervais ➑️
Andre can always correct me if I am mistaken of course............ in fact I would appreciate his reply on this subject.......
Thanks for the complement Rod. There is nothing to correct in what you wrote. It is one of parts of acoustics that is difficult to grasp. What is consistent with many aspects of acoustics, is that because acoustics is so complex, an integral part of building something to acoustic standards includes as an integral part the testing, analyzing, modifying and repeating process.

Well tested theoretically,
Andre
Old 21st May 2010 | Show parent
  #20
Lives for gear
 
avare's Avatar
 
🎧 15 years
Quote:
Originally Posted by Brainchild ➑️
Does anyone know about any absorbers that operate exclusively on the basis of diffractive absorption? A wooden panel (or three-dimensional matrix) riddled with voids in the appropriate configuration would be "pretty awesome."
That is easy to answer, but the answer is not very satisfying.

Reflection grating (QRD, PRD, etc) diffusers are exactly what you described. If you look at RPG's website for the technical data on their diffusers, they include absorption data. Totally honest and not very helpful answer.

Andre
Old 21st May 2010 | Show parent
  #21
Lives for gear
 
PaulP's Avatar
 
🎧 10 years
Cox and D'Antonio seem to think that absorption coefficients greater than
one are due to a few things but mostly edge diffraction. The fact that the
sound fields in test rooms are not entirely diffuse and that the rooms don't
have reverberation times of zero also play a role.

Their book Acoustic Absorbers and Diffusers deals with the issue in several
places. See : excerpts. You could do a lot worse than buy this book. It is
sometimes on sale, I got mine for about 100$ delivered a couple of months
ago.

Floyd Toole's book Sound Reproduction (probably the best book I've read
to date on room acoustics) says that the usual coefficients are not terribly
useful since they were measured in circumstances not applicable to small
rooms (they don't have diffuse sound fields, instead the sound arrives at
the absorber at a definite angle). Google Books won't show the page in
question. Floyd would like to see coefficients that took the angle of incidence
into consideration. He also gives the example of covering an absorber with
Guilford of Maine which has a great effect on the reflective properties of
the absorber when sound hits it at an angle.

Paul P
Old 21st May 2010 | Show parent
  #22
Lives for gear
 
Weasel9992's Avatar
 
🎧 10 years
Quote:
Originally Posted by Insulation Man ➑️
If you put 6" of 703 board on the wall, you won't hear a thing on the other side!
100% incorrect.

Frank
Old 22nd May 2010 | Show parent
  #23
Lives for gear
 
Dange's Avatar
 
🎧 10 years
I can see where it's all gone wrong! Thanks for the input Rod, Andre and Paul heh

My post that caused this was a hasty, not well thought through, reply to the claim that 'using products that have over 1.0, its basically over kill'. That just seems bizarre.

I am very well aware of the inherant inaccuracies in acoustic testing both in laboratories and in the field. Rod, I may well refer to builders as 'natural anomalies' in the future as they seem to have the greatest effect!

You could regard Sabines as being anomalous as a measurement value. That fact that the measurement method is deeply routed in international standards for testing and Sabines are very easy for non-acoustians to use (e.g. Architects) is why they're commonly used.
Old 22nd May 2010 | Show parent
  #24
Lives for gear
 
Dange's Avatar
 
🎧 10 years
Also to go back to the OP's first post, those absorption coefficients, where did you get them from?
Old 22nd May 2010 | Show parent
  #25
Lives for gear
 
ciro's Avatar
 
🎧 10 years
Quote:
Originally Posted by Dange ➑️
Also to go back to the OP's first post, those absorption coefficients, where did you get them from?
http://www.bobgolds.com/AbsorptionCoefficients.htm

Ciro
Old 23rd May 2010 | Show parent
  #26
Lives for gear
 
jhbrandt's Avatar
 
🎧 10 years
To the OP,

One thing that is often overlooked in small booth construction is the LF trapping needed to attenuate or eliminate resonances. In larger spaces we deal with room modes and in booths such as a vocal booth, we deal with resonances within the space. Remember that the energy in the LF can be 10 to 100 times the power of the energy in the mid and HF. One may wish to place a guitar amp in the 'dead' vocal booth. The room resonances may be nice for the guitar - and they might not, as this is a subjective issue. They also might (probably will) affect the accuracy of the vocal recording.

So, if you have the space, I would recommend trapping the ceiling as much as possible, ie; ceiling dropped and trapped down 12 to 24 inches & filled half-way with attic blanket type fiberglass insulation. This will usually take care of those pesky resonances.

Cheers,
John
Old 28th May 2010 | Show parent
  #27
Lives for gear
 
Dange's Avatar
 
🎧 10 years
Just to round off the absorption coefficients greater than 1 thing, there is a statement at the top of that page from the link above ^ which some, who have delved deeper into the maths that goes with acoustics, may find confusing. Mainly that random incidence absorption coefficients shouldn't be used as percentages, but if you open an acoustics textbook there are established formulae staring back at you which do.

It is possible to measure an absorption coefficient greater than 1, certainly when using Sabine's Formula. Another factor to consider is edge diffraction.

I've looked into this and am not going to attempt a full explanation in this thread. heh
For interested parties, I have found the best explanation is to be found in Cox and D'Antonio's Acoustic Absorbers and Diffusers Book, Chapter 12, as originally recommended by PaulP in an earlier post. Follow the link to Google books and can get most, if not all, of the detail
Old 24th June 2010 | Show parent
  #28
Gear Guru
 
Glenn Kuras's Avatar
 
🎧 15 years
Honestly people should be looking at sabin numbers. That truly is going to give you the absorption of the given product. heh
Old 24th June 2010 | Show parent
  #29
Lives for gear
 
Dange's Avatar
 
🎧 10 years
Quote:
Originally Posted by Glenn Kuras ➑️
Honestly people should be looking at sabin numbers. That truly is going to give you the absorption of the given product. heh
Yes, as normally that's all you can get from the manufacturer!

But using those values to compare products isn't always straight forward.

You can use them to compare a manufacturer's range of products against each other, as one would assume they were tested in the same lab.

You can use them to compare products from different manufacturer's tested in the same lab.

Comparing different manufacturer's products tested in different labs is the tricky one. I'm not sure what limit I'd used to say whether differences in performance were significant, Cox and D'Antonio noted there were 0.4 differences seen for the same sample in different labs.....

Also there are Imperial Sabins and Metric Sabins heh You could argue that edge diffraction could cause the smaller Imperial Sabin to show better results than the Metric Sabin for the same material, if testing only a 1ft^2 and a 1m^2 piece.....

Then there are also other absorption units possible, Millington-Sette and Eyring....doubt you'd ever see this in manufacturer's data though.

So the only thing really to be careful about is small differences in performance between different manufacturers. Small differences should be ignored, but would be good if someone could chime in with a limit, 0.2, 0.3?

There is also a difference between Imperial Sabins and Metric Sabins, but again that should be small
Old 24th June 2010 | Show parent
  #30
Lives for gear
 
Weasel9992's Avatar
 
🎧 10 years
Quote:
Originally Posted by Dange ➑️
There is also a difference between Imperial Sabins and Metric Sabins, but again that should be small
.15 or smaller is within the statistical error range, isn't it? Ethan, Glenn or somebody can correct me if I'm wrong.

Frank
πŸ“ Reply
Post Reply

Welcome to the Gearspace Pro Audio Community!

Registration benefits include:
  • The ability to reply to and create new discussions
  • Access to members-only giveaways & competitions
  • Interact with VIP industry experts in our guest Q&As
  • Access to members-only sub forum discussions
  • Access to members-only Chat Room
  • Get INSTANT ACCESS to the world's best private pro audio Classifieds for only USD $20/year
  • Promote your eBay auctions and Reverb.com listings for free
  • Remove this message!
You need an account to post a reply. Create a username and password below and an account will be created and your post entered.


 
 
Slide to join now Processing…

Forum Jump
Forum Jump