AFMG Network Forum

Ron Sauro · Member

Wow ... I did not know this thread was here.... perhaps I can shed some light based on what we are doing here at the lab.....
1st:... If the material is measured against the floor... the edges must be covered so only the surface is exposed.... Theoretically this is true for any size sample.
2nd.... The sample size has to be a certain size... In ASTM it must total 72 square feet (8ft X 9Ft) rectangle...That said if the baffles are under that size there must be multiple baffles supplied to add up to this area..
3rd If the test is to be ISO then the square area must be 10 square meters
4th... The standard test for ASTM allows the Reverb room to be stimulated by a single full range speaker until the SPL level is constant and when that steady state occurs then the stimulus is removed and the decay time is measured.... This is done numerous times to get a good average...
This RT time is subtracted from the previous measurement of RT done with an empty room.... The difference is the absorption amount

All this said.... in experiments we have done the last 2 weeks comparing numbers from ASTM and ISO measurements we are detecting that the straight edges of a sample have a larger effect than what everybody has thought previously and that the perimeter size has a larger effect on the total absorption than the area has...

The other thing not discussed here was that the sound penetrates the material on the floor and bounces off the hard surface and is absorbed some more on the way out of the material.... This is not true of a baffle hanging in Free space.... The sound goes Through the material and does not reflect back thru the material

end result...Conclusion.... The absorption coeffcient is the same either direction.... If hung in free space you have in your example 20 sq feet of absorption in ONE direction .... that is Thru the material.... but sound is impinging on BOTH surfaces....
in the other example it is against a wall and the sound is going THRU the material and bouncing on the hard surface and exiting back thru the material again.... in this case 10 sq feet exposed but the material is being used twice in this instance so therefore the absorption is done TWICE...
Both have equivalent absorption for those reasons...

What is interesting was my discussions with Gary Mange and Ken Roy about this subject.... NO one ... I repeat... NO one has done direct comparisons of these types of measurements.... Gary and I agree that as part of our qualifying the facilities for scattering that we will do these experiments as well as a few others to prove or disprove these assumptions

Ron Sauro
NWAA Labs

Ron Sauro · Member

Also Intercessor asked about a material for 250 HZ..... Contact MBI and ask for the Bandit material.... If you wish EASE Data contact me or ask Stefan for the data....
it is aimed specifically at lower frequencies and is a very inexpensive product
Ron Sauro
NWAA Labs

Emad El-Saghir · Member

Ron Sauro · Member

To be honest ... I am not sure ..but based on some charts of flow resistance it seems to be true.... The only question I still have is when you put a hard surface behind the sample you are forcing the velocity at the boundry to zero... where when the baffle is in free space it is not changed much .... what effect this has is still open to discussion.... I think that because the noise is diffuse and coming from every angle and freq.. that perhaps phase cancellations and addition happen that make the velocity change moot...The experiments we are setting up will answer some of those questions
Ron

Ron Sauro · Member

I also re-read your question and as an additional answer ... No I do not believe all materials act exactly the same but within that material I think the sound enters and leaves the same
Ron

Jim Mobley · Member

Ron Sauro writes >>> “The absorption coefficient is the same either direction.... If hung in free space you have in your example 20 sq feet of absorption in ONE direction .... that is Thru the material.... but sound is impinging on BOTH surfaces....
in the other example it is against a wall and the sound is going THRU the material and bouncing on the hard surface and exiting back thru the material again.... in this case 10 sq feet exposed but the material is being used twice in this instance so therefore the absorption is done TWICE...
Both have equivalent absorption for those reasons...”

Ron, I don’t think that you can say that the absorption coefficient is the same for both conditions as the exposed surface area is not the same. You could make the argument that the number of Sabins of absorption is the same for both conditions.

Let’s look at a real example from the EASE data base—the MBI Cloud-Lite Baffle, 1” thick with a 3# core and sealed plastic cover. The material in EASE is “BAFL1"3P2M.”

Let’s look at just the 1000 Hz band for simplicity. In EASE the coefficient is .99 but on the manufacturer’s data sheet the coefficient is 1.44. http://www.mbiproducts.com/products/acousticalTestData.aspx?ProductID=2600-1030
No mounting method is specified.

Are we to infer from your post that this material was measured using a 72 square foot sample directly on the floor of the chamber? And that mounted in this manner it produced 104 Sabins of absorption at 1000 Hz? (For those reading this that don’t know, one U.S. Sabin is equal to one square foot of perfect absorber. 72 x 1.44 = 104 Sabins. Thus the material in question is absorbing 44% more energy than a 72 square foot hole in the wall.)

If the above is correct, then in EASE if we use this 72 square feet as a coated face we have less absorption at 1000 Hz than we should—71.3 Sabins rather than 104. (72 x .99 = 71.3) But if we make the face two-fold and hang it in the middle of the model we now have 144 effective square feet of surface area and 142.6 Sabins. But by your logic above, we should still have 104 Sabins of absorption at 1000 Hz—we now have about 37% too much absorption. (And of course with a material that didn’t exceed a coefficient of 1, we’d have 100% too much.)

But let’s run with this idea for a moment. If we “should” see 104 Sabins of absorption, and this is a result of sound going through the material twice, then shouldn’t we consider the surface area of both sides even though it was on the floor when measured? In that case the coefficient would be .72 at 1000 Hz (104 Sabins/144Ft2 = .72) and our Sabine RT times will then be correct for the two-fold application.

Bruce suggested this early on.

But we still have a problem if the material is used single-fold. And now it’s an even bigger problem--assuming the coefficients on the data sheet are correct for an “A” mounting--we now have just half as many Sabins of absorption as we should.

And we’ve not even considered Ray Tracing where rays go through nothing, they are either reflected or they are absorbed. Whole ‘nother can of worms.

It seems from the stand point of prediction software; the best solution might be to have the proper coefficients of both absorption and scattering for the (presumably) ceiling as a whole and then model it as a single face and use AURA for the calculation.

Measuring all of the common permutations should keep you busy for a while Ron. Wouldn’t want you roaming the streets after all. ' Smile

'

Ron Sauro · Member

Jim:

When you read my entire post you'll see that we have found some major problems with how absorption is measured using the ASTM and ISO standard methods. ONE of them is that the perimeter has a larger effect on the coefficient and total absorption than we previously suspected. This we think is going to help explain some of these "edge" effects that get us over the 1.0 mark. The experiments we have already done indicate to us that the absorption measured is very close to the same if you measure a standard sample of 72 sq ft A mounted or standing upright in the room. There seem to be small differences at some lower frequencies but most are close to the same results. The same cannot be said of higher frequencies.

When I speak of both being the same ...what I am speaking of is that the effect in the room is the same for a sample that is 72 sq ft against the floor as it is upright in the room. In one case the sq footage exposed to the sound field is 72 sq ft, in the other it is 144 sq ft. Again, our experiments indicate that the sq footage does not have the expected effect compared to its size. The effect seems to be mostly tied to the perimeter, which is exactly the same here.

As to if the MBI sample was measured A mounted - I do not know . However I can find out from the test data from Riverbank. However, based on what we now know, I do not think it will make very much of a difference measuring it either way as the diffraction effects are affecting the number to an extent that is predominant.

>>>Are we to infer from your post that this material was measured using a 72 square foot sample directly on the floor of the chamber? And that mounted in this manner it produced 104 Sabins of absorption at 1000 Hz? (For those reading this that don’t know, one U.S. Sabin is equal to one square foot of perfect absorber. 72 x 1.44 = 104 Sabins. Thus the material in question is absorbing 44% more energy than a 72 square foot hole in the wall.) <<<

Again you use an equation that is assuming what we have all thought all along -but now may or may not be true. It is possible to prove anything with the wrong assumptions...

On the whole I think we may have to rethink how absorption amounts and coefficients in these programs are used.

This news has been discussed already at ASTM and is causing a row over what other kinds of tests need to be done.

In the rooms I have done this way and then checked with TEF or EASERA we have seen good correlation using the data as it is presented.

Ron