NASA, LF Absorption and Gas Flow Resistivity

[avare]

This linked document is a study commisioned to determine the best sound treatment for a NASA wind tunnel. Some interesting points in the study are that 703 is assigned a gas flow resistivity value of 27 000 mks Rayls instead of ~ 16 000 Rayls, which we have accepted. I find the difference intriguing as the source that we, or at I, have used is form Owens-Corning. However, I find it hard to beleive that NASA would get something like that wrong.

Figure 7 shows the difference in predicted and measured (mean) values for absorption. Measured values extend effective absorpttion down about an octave lower than predicted. As Ethan says, measurement beats theory.

Enjoy!

Andre

[DanDan]

Andre your personal library is priceless. This is a little gem full of great short punchy facts. I particularly noted the 45 degree incidence bit.
DD

[Ethan Winer]

Originally Posted by avare

Measured values extend effective absorpttion down about an octave lower than predicted. As Ethan says, measurement beats theory.

I do indeed say that - all the time!

--Ethan

________________
The Acoustic Treatment Experts

[avare]

Originally Posted by DanDan

Andre your personal library is priceless. This is a little gem full of great short punchy facts.

Thanks Dan for the compliment. Sometimes people get frustrated when working on their studio design (others:see my tagline if you are not familiar with it) that their are no single books that answer all their quesitons about their studio. It may seem esoteric to some that we (the pros, or commited) have a wide variety of resources in our library.

BBC Research reports and other documents that delve in great detail into certain aspects. For sound isolation of many walls and other related data, reports and studies from the Canadian National Research Council. No doubt some would be incredulous to find out that our studio acoustics knowledge base includes NASA reports!

Yes it is a great document. Sort of like a Sharp* light.

*Ben sharp published in 1973 THE text on improving the transmission loss of light weight partitions. It makes the NASA text seem like senior public school (grades 5-8) reading. It is here for those who do not fear to tread.

Andre

[Hannes_F]

Originally Posted by avare

Figure 7 shows the difference in predicted and measured (mean) values for absorption. Measured values extend effective absorpttion down about an octave lower than predicted. As Ethan says, measurement beats theory.

It should be noted that the conclusion of the paper is that the theory was confirmed within the error margin, and that thick OC3350 with a gas flow resistance of 4100 rayls/m absorbs better in the bass than OC703.

I say that since years, now it is finally true .

[Jens Eklund]

Originally Posted by Hannes_F

It should be noted that the conclusion of the paper is that the theory was confirmed within the error margin, and that thick OC3350 with a gas flow resistance of 4100 rayls/m absorbs better in the bass than OC703.

I say that since years, now it is finally true .

Not too far from what Soundflow predicts:




PAROC WAS 50 sound absorption?

[avare]

Originally Posted by Hannes_F

It should be noted that the conclusion of the paper is that the theory was confirmed within the error margin, and that thick OC3350 with a gas flow resistance of 4100 rayls/m absorbs better in the bass than OC703.

I say that since years, now it is finally true .

Correct. The conclusions are in agreement with physics. What you and I have been writing for years.

Andre

[avare]

Originally Posted by Jens Eklund

Not too far from what Soundflow predicts:

+1. Experimental validation of software developed independently. Isn't acoustics wonderful when no black magic is applied?

Andre

[silverdisk]

Originally Posted by avare

Correct. The conclusions are in agreement with physics. What you and I have been writing for years.

Andre

Andre and Hans, for the layman... like me ....... what OC product is 3350.... is that just regular pink insulation? thanks in advance

[DanDan]

703 is assigned a gas flow resistivity value of 27 000 mks Rayls instead of ~ 16 000 Rayls, which we have accepted. I find the difference intriguing as the source that we, or at I, have used is form Owens-Corning. However, I find it hard to beleive that NASA would get something like that wrong.

Intriguing indeed. There is a lot based on the accepted 16,000.
NASA have gotten a few things wrong, the flag waving in the breeze in that televised moon landing......

So which is it 16K or 27K? Does anyone have any way of confirming the real figure here?

DD

[Sebg]

703 is (I assume) 3x16=48kg/m^3.

The best I can do is this:

CSR Bradford (Australia) has published flow resistivity figures for some of its products.

Building Blanket (11kg/m^3): .. 5,600 mks Rayls/m
Multitel (18kg/m^3) .............. 15,300
Flexitel (24kg/m^3) .............. 16,200
Supertel (32kg/m^3) ............ 18,200
Ultratel (48kg/m^3) .............. 31,500

This suggests that the 16,000 figure for 703 is a bit low. Unless it has a real lot of binder in it and not so many fibres/m^3.

Unless of course the Units are different... Which is of course, one way to navigate a spaceship so that it misses Mars ... (I think it was a NASA probe that was given imperial rather than metric instructions; could have been the Europeans)

[PaulRain]

I remove a little bit of dust on this thread
I didn't find any confirmed datas about the OC 703 but, in Europe at least, Isover is more common and they do have on the tech datas of few of their products the Flow resistivity.....
Isover E100 S (50Kg/m^3)
Flow resisivity 44 kPas/m^2 - which is 44000 mks Rayls/m
sover E60 S (30Kg/m^3)
Flow resisivity 22 kPas/m^2 - which is 22000 mks Rayls/m
I think this two product could be very similar to the OC and in my opinion the 703 is more likely close to 27k than 16k
Another brand, easy to find in Italy but probably in all Europe is RockWool...
on the Italian site there is an interesting software:
http://www.rockwool.it/download/software
this is in italian but I think you can manage it....it's based on test measures made by the Engineering Department of the university of Ferrara.
Sadly it stops at 100 and only calculate the absorbtrion coefficients at the normal incidence, but it mention the specific Flow Resistivity datas of many of their products.....

[mhch]

Originally Posted by PaulRain

I remove a little bit of dust on this thread
I didn't find any confirmed datas about the OC 703 but, in Europe at least, Isover is more common and they do have on the tech datas of few of their products the Flow resistivity.....
Isover E100 S (50Kg/m^3)
Flow resisivity 44 kPas/m^2 - which is 44000 mks Rayls/m
sover E60 S (30Kg/m^3)
Flow resisivity 22 kPas/m^2 - which is 22000 mks Rayls/m
I think this two product could be very similar to the OC and in my opinion the 703 is more likely close to 27k than 16k
Another brand, easy to find in Italy but probably in all Europe is RockWool...
on the Italian site there is an interesting software:
Software | Rockwool Italia S.p.A.
this is in italian but I think you can manage it....it's based on test measures made by the Engineering Department of the university of Ferrara.
Sadly it stops at 100 and only calculate the absorbtrion coefficients at the normal incidence, but it mention the specific Flow Resistivity datas of many of their products.....

Best is to extract the data for the Rockwool products from the software (or an associated file) and plug that into more sophisticated porous absortion calculator. Chris Whealy's one uses the same models as the software you indicated, but it is more general and can be easily adapted (well ! for software folks) since this is an Excel spreadsheet.

[PaulRain]

actually....Whealy's is theorical, the results coming from the rockwool software are based on real tests, and if you compare the two there is a small difference, probably because the Flow Resistivity is not the only crucial data....
Anyway the real tests gives a slightly better curve.
But I agree the most important datas can be taken from there and used somewhere else....
What is interesting for me is to see the relation between the density and the Flow Resistivity of the same kind of material by the same producer.....
for example, a 40 Kg7m^3 has less than hals the Flow Resistivity of a 80 Kg7m^3.....and the more you rise the Density the more the proportion change....so an half denser material has less than half Flow Resistivity.
and, even more important, less Flow Resistivity not always gets a better low frequency absorbtion.....it depends on the thickness....but I think everybody agree that for real LF thickness is the only cure if we talk about porous absorbers

[mhch]

Both are theoretical, using the same models according to the related web pages. Whealy's has no data, and the Rockwool software has, which is great value and may be can be extracted ...

So the differences are in the implementation of the models, may be Rockwool software has been tweaked and tuned, but which is correct (whatever that means !!) is hard to say !! None can be claimed to be bug free. Actually Whealy's has been subject to a number of fixes along time ....

During my whole professional career, I observed that silicon chip manufacturers (Intel, AMD, ...) always compared several different software tools meant for the same task, and spent a lot of time validating each one of them as a way to make sure the results are correct, since no tool can be guaranteed perfect and 100% correct. Even though problems occured regularly (cf the Pentium and the recent SATA bugs for Intel!!)

Can you show us an example of the difference ?

[Hannes_F]

Originally Posted by mhch

Both are theoretical

Not wanting to take anything away from what you say, but I would like to add one thought that seems to be largely unknown.

People refer to the Whealey calculator and other formulas as being 'theoretical', indicating that these would be originated from some sort of physical model. However this is not true.

Maybe you know what physical modeling means for sounds: The properties of e. g. a trumpet are fed into a computer model which then tries to calculate the sound propagation inside and outside the trumpet and generates a sound signal from that. In a similar way one could try to create a model of a fibrous absorber and then calculate the chaotic air flow in it.

In that case one could argue that any physical model has its restrictions and there will always be a difference between theory and practise. Right so, a physically modeled trumpet still does not sound very much like the real thing.

In contrast to that the formula the Whealey calculator uses is based on a large number of measurements. Scientists (Delany and Bazley) simply fitted a mathematical curve to their measured results. These are not really theoretical values, these are values abstracted and calculated from practical experiments.

I see this distinction constantly blurred in this forum. Some people try to discredit calculated predictions indicating that would 'only' be theoretical ideas (meaning pipe dreams) while they are doing real measurements. No.

[mhch]

Originally Posted by Hannes_F

Not wanting to take anything away from what you say, but I would like to add one thought that seems to be largely unknown.

People refer to the Whealey calculator and other formulas as being 'theoretical', indicating that these would be originated from some sort of physical model. However this is not true.

Maybe you know what physical modeling means for sounds: The properties of e. g. a trumpet are fed into a computer model which then tries to calculate the sound propagation inside and outside the trumpet and generates a sound signal from that. In a similar way one could try to create a model of a fibrous absorber and then calculate the chaotic air flow in it.

In that case one could argue that any physical model has its restrictions and there will always be a difference between theory and practise. Right so, a physically modeled trumpet still does not sound very much like the real thing.

In contrast to that the formula the Whealey calculator uses is based on a large number of measurements. Scientists (Delany and Bazley) simply fitted a mathematical curve to their measured results. These are not really theoretical values, these are values abstracted and calculated from practical experiments.

I see this distinction constantly blurred in this forum. Some people try to discredit calculated predictions indicating that would 'only' be theoretical ideas (meaning pipe dreams) while they are doing real measurements. No.

You're 100% right. These are fitted and not theoretical models, which of course doesn't mean they are 100% exact for all possible cases. They are as correct as a fitted model can be.

And for sure I'm not discrediting any, as they are very good tools to help understanding what can happen in a particular case.

[PaulRain]

sorry I really tought it was only theoretical....I should have read more carefully on the website...always too lazy..... please fell free to hit me
anyway, here the differences I find:
https://picasaweb.google.com/piovesa...QE&feat=email#
I do believe both could have mistakes and tollerances, and they are more or less similar...
I find the Rockwool very usefull because I can actually but exactly that material which have been tested....

[mhch]

Not a really big difference, unless I misread/misinterpreted your images.
The 20% or so difference at 100Hz isn't a huge difference for such a model, at least for my own taste

[PaulRain]

20% is not much and as I said it could be a result of several aspects.......it could have been nice to see other materials tested at the same time.....
anyway with Wealey calculator is very difficult to get over 0.5 at 100Hz with that Flow Resistivity (15000)...
anyway, my intent was just to make this Rookwool software available to the people interested

[avare]

Originally Posted by PaulRain

my intent was just to make this Rookwool software available to the people interested

Thank you. It is appreciated.

Andre

[PaulRain]

I think is kind of self explaining, most of the words are very similar in english but if anybody needs a translation of something just let me know
P

[johndykstra]

'Last edited by avare; Today at 09:54 AM.. Reason: Corrected spelling. (It does have five words)'

classic

[Hannes_F]

Originally Posted by PaulRain

on the Italian site there is an interesting software:
Software | Rockwool Italia S.p.A.
this is in italian but I think you can manage it....it's based on test measures made by the Engineering Department of the university of Ferrara.
Sadly it stops at 100 and only calculate the absorbtrion coefficients at the normal incidence, but it mention the specific Flow Resistivity datas of many of their products.....

Thanks Paul, very interesting!

[johndykstra]

in other NASA news:


NASA Announces Results of Epic Space-Time Experiment - NASA Science

pretty crazy

[Jens Eklund]

Originally Posted by johndykstra

in other NASA news:


NASA Announces Results of Epic Space-Time Experiment - NASA Science

pretty crazy

Soon we will have not only pressure and velocity based absorbers, but also space and time based. You simply send the excess bass to another dimension or time.

[johndykstra]

good one

[avare]

Originally Posted by johndykstra

in other NASA news:


NASA Announces Results of Epic Space-Time Experiment - NASA Science

Great stuff. Related to this, physicists are starting to reconsider the forces in the universe and drop gravity form the list, as gravity being a product of distortion of space-time, instead of a basic force.

FYI the three forces in the universe are now considered to be:

strong nuclear force
weak nuclear force
electro-magnetic force.

Forcefully,
Andre

[Jens Eklund]

Originally Posted by avare

Great stuff. Related to this, physicists are starting to reconsider the forces in the universe and drop gravity form the list, as gravity being a product of distortion of space-time, instead of a basic force.

FYI the three forces in the universe are now considered to be:

strong nuclear force
weak nuclear force
electro-magnetic force.

Forcefully,
Andre

I bet you when they’re done, there will only be one force.


... The resonant force of strings perhaps?

[Brainchild]

Of course, it depends on how you look at it. The graviton is still a force-carrying particle when the situation is viewed from a QD standpoint. Although how a particle is the very warping of spacetime has always been a little beyond me [just a little].

Maybe I'm jaded about relativity, but the craziest thing in that article to me was

"These ping pong-sized balls of fused quartz and silicon are 1.5 inches across and never vary from a perfect sphere by more than 40 atomic layers." That's outlandish.