Gap behind trap vs. thicker trap?

[spm_gl]

Reality is always different from calculation. There are too many effects to be considered. I also believe we have resonance in capillaries of porous material, so we don't only need particle velocity. We were discussing the "ideal" placement.
Your idea of using two different materials is nice.

[Brainchild]

Could someone help out interpreting material's gas flow properties

Originally Posted by Hannes_F

Hi Andre,

sorry, maybe I did not get what you mean.

I am using material with low flow resistivity, around 5000 Pa*s/m². Cheap stuff.

Test these values in the Porous Absorption calculator:

Absorber thickness 100 mm
Air gap 100 mm
Start graph at 30 Hz.
Absorber flow resistivity 5000 or 16800 or 45000 Pa*s/m².



First, we see that the absorption is definetely different in the bass range.

Then, if we compare the 5000 Pa*s/m² with the 16800 Pa*s/m² curve we see that below ca. 135 Hz the absorption is better for the the more resitive material. Tests have shown this as well like Ethan Winer's site shows.

However if this leads to a "more is always better" thinking this could be an error. If you watch the curves closely you will see that there is a frequency range between 135 and 700 Hz where the fluffy 5000 Pa*s/m² material absorbs better. Moreover, the 45000 Pa*s/m² material aborbs worse than the 16800 Pa*s/m² material.

This is definetively counter intuitive. But if you think about it then denser material should be good for insulation, and this can include a good portion of backscattered energy.

Now enter an absorption width of 300 mm and a gap of 200 mm. Surprise - the 5000 Pa*s/m² is best!



If this is true then my conclusions are:

- If you are restricted to panels in the range of 10 cm or 4 " then material in the range of 16000 - 25000 Pa*s/m² is probably your best choice. With other words OC 703 or 705. (But do yourself a favor and don't ignore the demand for an air gap behind them).

- If you have more space then consider thicker absorbers with bigger gaps, filled with fluffy glass wool of 5000 Pa*s/m² or even less.

Now how relevant are those simulated values?

#1 Download the software ZORBA by Marshall Day Acoustics for a double check. It says the same if you activate "alpha normal". If you check their website you see that they verified their simulations with experimental data and the two are very close.

# 2 I am not finished yet with my studio but the traps that I installed so far behave exactly as they should according to these simulations.

My conclusions:
- More is not always better.
- It always depends (i. e. on the frequency range, the absorber width and position, the gap width)
- I try to know what I am doing. Read, read, read (I mean books, scientific articles, papers - not necessarily forum opinions).The links to the literature are everywhere. Start at the bottom of this site: http://www.bobgolds.com/AbsorptionCoefficients.htm . Read all the BBC articles about their acoustics tests.
- I try to understand the theory but also run tests in between.

Somebody who does not know what he/she is really doing may end up spending much money with mediocre results. Cruel but true :-)

Hope this helps
Hannes

Originally Posted by avare

I am the one who recommended, not suggested, using 3.5" material with an airspace in post #3 in this thread, with the technical explanation for the recommendation. When several people remarked upon my recommendation, I clarified it in post#9 with a link to thread that discusses porous insulation and sound absorption. It had nothing to with cost or value point.

The absorption of a porous absorber is dependent upon three things:

Location
Thickness
Gas flow resistance

Location is important because porous absorbers work on impeding the velocity of the air particles caused by the sound waves. If the location is where there is no particle velocity, then the of the absorber is nil.

Thickness affects how wide a frequency range the absorber is effective at. The thicker the absorber, the wider the frequency range. With absorbers against a wall, this being effective down to lower frequencies.

Gas flow resistance is important as for best absorption, the resistance of the absorber should equal the impedance of air (406 Rayls). If the resistance is too low, then the sound travels through the absorber. This is shown in the first drawing in post #29 in the thread I linked. If the resistance too high, then sound will be reflected, as shown in the second and forth drawings. The reflection occurs within the absorber itself. This is shown in the attachment, along with the equations detailing the real and imaginary components, of the sound wave and impedance. The resistance is a function of the gas flow resistivity and thickness of the absorber.

Unfortunately, almost no insulation companies provide gas flow resistivity data on their products, so we (as in studio designers) use material density as an indirect indicator of gas flow resistivity and scrutinize test data for clues.

With the Ultratouch test data (repeated in my previous post), it shows almost no increase in absorption in the 125 Hz band when the material thickness is increased by over 50%. If there is no internal reflection due to high gas flow resistance, then absorption would increase. This is being taken a clue that the gas flow resistance is too high.

As I wrote two paragraphs above " The resistance is a function of the gas flow resistivity and thickness of the absorber. " Since the type of material has been defined, this leaves adjusting spacing and thickness to achieve desired results.

The use of spacing is common in commercial facilities. Eric Desart used it in Galaxy Studios' absorbers. Peutz used it in the wall treatments for the Heineken Music Hall. In the latter case the ratio was 2:1 for space to material thickness) Neither of these are "value point" projects.

If the OP uses thicker Ultratouch instead of spacing with 3.5" material, he will get inferior acoustic results while spending more money. The recommendation for a space is based on physics and the restraints (what material) of this particular project.


The attachment is from the Bruel Acoustics web page. Unfortunately the page uses frames. If you want to read the document, which deals with his Standing Wave Tube apparatus, log onto the web page Bruel Acoustics, click on "Technical review and products information" and then "Technical Review 97-01: Standing Wave Tube SWT". If the name Bruel seems familiar, it could be because he is one of the founders of Bruel and Kjaer.

Andre

Originally Posted by avare

Several things are going on in this thread and rather than quote I will attempt to comment separately.

There is no advanatage to combining, or layering, different densities od acoustic material.

Filling a space (including a corner) is one way to get the required gas flow resistance at the required wavelength away from the wall. Cost becomes the factor and look at the cost of the material! 4" 703 last time I checked was $4/ft^2. Roxul Safe n Sound (general home insulation) at 3.5" thick was 37 CENTS/ft^2. The second product has the required GFR for thicker absorbers. A filled corner with it will cost a fraction of layered/spaced 4" 703! Do the math for prices in your part of the world.

Diffuse field absorption is nominally effective down to about 1/15 of the wavelength of the material thickness. Normal incidence absorption down to about 1/7 of the wavelength. Most peopel do not know the differenc between diffuse field and normal incidence absorption and refer to thte diffuse field, as that is what measured in reverb chambers.

Dig around on STudiotips.com Whealey's gas flow resistance and absorption calculator. It works with normal and specific anges of incidence for sound waves.

I hope this helps.

etc.

[Ben F]

I have another question regarding corner traps.

I am about to build some 4" panels with 703 equivalent Tontine AcoustiSorb (pg6)

If they are going in the corners, is it better to have a closed back on the trap (plywood) or open back and just have them in a frame? It is for absorbing below 150Hz.

[Glenn Kuras]

Originally Posted by Ben F

I have another question regarding corner traps.

I am about to build some 4" panels with 703 equivalent Tontine AcoustiSorb (pg6)

If they are going in the corners, is it better to have a closed back on the trap (plywood) or open back and just have them in a frame? It is for absorbing below 150Hz.

When straddling the corner, you always want to have the backs open.

[Weasel9992]

Originally Posted by Ben F

If they are going in the corners, is it better to have a closed back on the trap (plywood) or open back and just have them in a frame? It is for absorbing below 150Hz.

Backs open in the corners...if you build them right you should have great absorption characteristics below 150Hz.

Frank

[Ben F]

Originally Posted by Weasel9992

Backs open in the corners...if you build them right you should have great absorption characteristics below 150Hz.

Frank

Ok now I'm a bit confused- I'll stick to using this thread.

So it would be better to have 4" thick panels (say with a 1 meter wide surface) with an open back, over making 'super chunk' traps with the same width in the corners?

Is it best to treat all the corners or just the front 2, and have some panels on the side walls to help absorb the speakers early reflections?

Also (sorry for all of these questions) what would be a good width for an 'all-rounder' corner panel trap? Or should I work out the bass frequency problems first using a room mode calculator and the have the distance from the wall 1/4 of the wavelength of the problem frequency?

This is just for a project studio, obviously our mastering studio was professionally designed!

[jayfrigo]

Originally Posted by Lupo

Hi!

Good question! There's probably some subtle caveats to this, but generally speaking, it's the absorber inside the resonator that dampens sound and this works on velocity.

The absorber inside will damp it and, as a practical matter, broaden the Q, but even without porous absorption inside the cavity, a resonator will still absorb at the mouth. In fact, the absorption will be of a greater magnitude without the absorber inside, though of a much narrower Q, and re-emission "ringing" is likely to be more of an issue.

[jayfrigo]

Originally Posted by DanDan

Spm, I am well aware of the classic understanding of such matters. However, to pursue my heresy, 'no particle displacement, no energy'- I just can't accept that. Energy does not go away, it goes to the other vector. Would you put your Bass trap at the room centre, hypothetically of course?
DD

Conservation of matter and energy indeed. However, think about this: for a standing wave, when you are at a node, where does that energy go? Well, a standing wave also has antinodes. You have more energy at the antinode than was emitted by the source. How can that be? Energy can't created from nothing. Nah, the energy is just redistributed, not created or destroyed. You have more than intended at the antinode and less than intended at the node.

Next we can talk about compressions and rarefactions, and at the "zero crossing" there appears to be no energy (matches ambient barometric pressure 14.7 lbs/in^2); and this will lead to the discussion of potential energy vs. kinetic energy, as I think somebody already mentioned. Next we can apply this back to the subject at hand and talk about pressure traps vs. velocity traps. It all turns to heat eventually. Also, if you look closely, you will notice that the spacing of a trap away from a wall will usually exhibit a bump at a certain frequency rather than a simple, even increase in bass absorption. Bonding FRK (FSK) to the front of a porous absorber exhibits a similar bump. Lots in play here... yet all connected.

[DanDan]

Ben,

It is best to treat as many corners as possible, certainly the four obvious one. Wall ceiling junctions are corners also, but at least do the 4.

Superchunks are the best. If you need to save some money, straddle the corners with 4inch thick panels, fill the gap behind with attic insulation. This is very nearly as good.

This is a Broadband technique, don't worry about frequencies, Q, etc.

Due to the size of the Fibre Batts there are two efficient sizes of Superchunk.
The cut patterns are over at studiotips - tips on studio design, acoustics, and wiring
Here's a very clever idea
The wider SSC of course works better, if you have the money and space. Be aware though they do use a lot more fibre, costing more. I reckon a wider straddling panel with backfill would probably outperform the smaller superchunk. So it's a question of distributing the cost and effort wisely. Try to find a good cheap source for your fibre.

Do the RFZ including a Cloud.

This plan is also described at Ethan and Glenn's sites, simply and well. If you do all of it you will get a very 'professional' result.

DD

[DanDan]

Hi Jay, we are in the same choir, it is all connected. I think the one dimensional views, say velocity only, are fine to get some calculations done, but it is healthy to be aware that there are other forces at play. This is why the calculated rarely matches the measured. My understanding of the energy is a little different to yourse, not sure which is a better model, I must look this up. In my mind, the total amount of energy stays the same always. It shifts from velocity to pressure, from kinetic to potential in a cyclical manner with the two vectors out of phase by 90 degrees I think? This seems very analagous to the propogation of Electromagnetic waves, with the Electric (Pressure) and Magnetic (Velocity) fields, again out of phase, rising and falling in strength, as one loses the other gains, thus egging each other along.
But then, I am a Heretic!
DD

[Brainchild]

Again, though, it's important to note that superchunks using highly dense fiberglass such as 705 will give you less absorption at low frequencies than either superchunks using less-dense material (703, 701, etc.) or less-thick (4 to 6 inches or so) panels of 705, etc.

[DanDan]

Densities indeed. Ben, just in case this has reintroduced Confusion-
use Fibre with a density of 3 pounds per cubic foot for SSChunks.
This is 48KG/Cubic metre. There are often cheaper alternatives to 703.
I used Isover Ultimate. If you can only get stuff with foil on it, take it off.
If you go for straddles 3 pcf is fine also. 5-6pcf will work fine too if you need the stiffer, easier to handle panels.
DD

[drbam]

Originally Posted by DanDan

Ben,

It is best to treat as many corners as possible, certainly the four obvious one. Wall ceiling junctions are corners also, but at least do the 4.

Superchunks are the best. If you need to save some money, straddle the corners with 4inch thick panels, fill the gap behind with attic insulation. This is very nearly as good.

This is a Broadband technique, don't worry about frequencies, Q, etc.

Due to the size of the Fibre Batts there are two efficient sizes of Superchunk.
The cut patterns are over at studiotips - tips on studio design, acoustics, and wiring
Here's a very clever idea Attachment 127333
The wider SSC of course works better, if you have the money and space. Be aware though they do use a lot more fibre, costing more. I reckon a wider straddling panel with backfill would probably outperform the smaller superchunk. So it's a question of distributing the cost and effort wisely. Try to find a good cheap source for your fibre.
DD

This morning I completed installing superchunks in the corners behind my monitors (10.5 ft ceiling). I have not taken measurements yet but just playing a few of my favorite reference tracks reveals a radical improvement over the 4" traps I had straddling the corners. I used Roxul (rockwool) which cost me about 1/4th of what 703 would have been. I cut 8 corner sections per panel (17x17x24). I'm very pleased with this and will do the same with the back corners next week.

[Ben F]

Originally Posted by DanDan

Densities indeed. Ben, just in case this has reintroduced Confusion-
use Fibre with a density of 3 pounds per cubic foot for SSChunks.
This is 48KG/Cubic metre. There are often cheaper alternatives to 703.
I used Isover Ultimate. If you can only get stuff with foil on it, take it off.
If you go for straddles 3 pcf is fine also. 5-6pcf will work fine too if you need the stiffer, easier to handle panels.
DD

Cheers mate, I have bought an environmentally friendly version of 703 with the same spec, called Tontine acoustisorb 3. Thanks for your help.

[Keith Moore]

I'm getting ready to build some bass traps for my control room. I want to do super chunks but one of the corners behind my monitors is at a more then 90 degree angle. Would it be alright if the angled side has a 3 or 4" gap or should I just straddle them?

Thanks,

Keith Moore

[DanDan]

Ben- That Tontine looks really good on a brief inspection.
Keith- Bass wavelenghts are huge, so small details won't have any significant effect.
You could presumably plug the gap by squeezing in some more fibre. Take a look at the variations of SuperChunk at studiotips - tips on studio design, acoustics, and wiring. You could straddle the corners with a nice thick 4 inch and fill behind with Attic instulation. Or a straddle panel with the back walls of the triangle lined with batts. It will all work. It' just a matter of sticking to the principles- Use light to medium fibre. 3pcf is often used but you could go lower if you are doing the deeper wider SSC's. Try to distribute evenly. Four corners with smaller chunks are better than two big ones.

DD

[Ethan Winer]

Originally Posted by Brainchild

superchunks using highly dense fiberglass such as 705 will give you less absorption at low frequencies than either superchunks using less-dense material (703, 701, etc.) or less-thick (4 to 6 inches or so) panels of 705, etc.

I'm still waiting for someone to actually test this in a lab, and then we'll all know for sure.

--Ethan

[DanDan]

Yes indeed Ethan, direct compartive tests in a real room would be very welcome.
I would not hold LAB testing in higher esteem than that. LF measurements are notoriously unreliable, even in LABS. However your repeated wishes for such tests kinda cast a shadow on those of us that are happy enough with existing tests showing the trends. e.g. here we see thicker panels and more fill in the corner increasing absorption. Intuitively true too IMHO Acoustics Forum • View topic - Studiotips Corner Absorber
I reduced a 34 Hz mode from 1.3 Seconds to 600mS or so using only three SuperChunks. Real world room, Dr. Dan testing. That must count for something.
DD

[Ethan Winer]

Originally Posted by DanDan

Yes indeed Ethan, direct compartive tests in a real room would be very welcome. I would not hold LAB testing in higher esteem than that.

That's exactly the point I make in my article Alternative Test Methods for Acoustic Treatment Products.

LF measurements are notoriously unreliable, even in LABS.

That's another point I make in that article. I even argue that testing using waterfalls in a "normal" room is more reliable than in a lab - comparatively only of course!

I reduced a 34 Hz mode from 1.3 Seconds to 600mS or so using only three SuperChunks. Real world room, Dr. Dan testing. That must count for something.

Yes, it does account for something, and I trust your results more than some of the data I've seen elsewhere! But what density did you use? And did you compare the results of different densities? That's the part I'm most interested in.

--Ethan

[Glenn Kuras]

Originally Posted by DanDan

Yes indeed Ethan, direct compartive tests in a real room would be very welcome.
I would not hold LAB testing in higher esteem than that. LF measurements are notoriously unreliable, even in LABS. However your repeated wishes for such tests kinda cast a shadow on those of us that are happy enough with existing tests showing the trends. e.g. here we see thicker panels and more fill in the corner increasing absorption. Intuitively true too IMHO Acoustics Forum • View topic - Studiotips Corner Absorber
I reduced a 34 Hz mode from 1.3 Seconds to 600mS or so using only three SuperChunks. Real world room, Dr. Dan testing. That must count for something.
DD

I think there is a lot to say about normal testing but have always felt that lab testing is still the best way to go.

As far as super chunks vs 4" panel our 244 at 50 hz tested at 4.65 sabins while our Tri Trap (kind of a super chunk) tested at 6.53 sabins. Check my math but that is around 30% better. Keep in mind though that our 244 has a 1.5" spacer on the back which that alone will put the face further into the room which will help absorb even better. I have seen the same testing (at riverbank) of just normal 4" 703 straddling and the Tri Trap (super chunk) was around 3 times better. It really is still all up in the air but it is a safe bet that it will work MUCH better on the lower frequencies.

[DanDan]

Good Stuff. Interesting article. Interesting numbers on the TriTrap vs 4inch plus gap panel. That LAB test is of course a compartative one, the kind I like. I find absolutes very confusing, especially when they disagree!
DD

[Ethan Winer]

Originally Posted by Glenn Kuras

Check my math but that is around 30% better ... I have seen the same testing (at riverbank) of just normal 4" 703 straddling and the Tri Trap (super chunk) was around 3 times better.

This is exactly the problem with lab tests at very low frequencies. You test it once and it's 30 percent different, but test it again and now the disparity is a factor of three. This is due to the intentionally random nature of reverb room testing. Having the measuring microphone move constantly in all three planes helps accuracy at mid and high frequencies, but it harms reliability and repeatability at low frequencies. This is why most labs are not certified down to 50 Hz. I've even seen negative absorption values reported at very low frequencies (like 30 Hz).

Most of that uncertainty goes away in a home-made "bedroom" test, as long as you're careful not to change anything between tests, and do the tests one after the other to minimize temperature and humidity changes.

--Ethan

[DanDan]

Ethan, the Rotating Boom is common in the ISO and other Testing Standards. Multiple locations are also used, with limits keeping the mic and speaker away from trouble.
A combination of both would obviously be best, these rotating booms are not long enough to randomise VLF.
I would assume that those GIK panels and TriTraps were tested in the same location in the same manner, probably in similar temperature and humidity conditions? If they were, and if spatial averaging was correctly implemented, I see no reason to doubt the result. Actually, even without spatial averaging of any type, I believe the test would be valid so long as everything remains the same except for the change of test subject.
Same room, same stimulus, same location in the room, etc. etc. One device measures 30% better than the other in the 50Hz band. I can't find a flaw in that.
DD

[Glenn Kuras]

Originally Posted by DanDan

Ethan, the Rotating Boom is common in the ISO and other Testing Standards. Multiple locations are also used, with limits keeping the mic and speaker away from trouble.
A combination of both would obviously be best, these rotating booms are not long enough to randomise VLF.
I would assume that those GIK panels and TriTraps were tested in the same location in the same manner, probably in similar temperature and humidity conditions? If they were, and if spatial averaging was correctly implemented, I see no reason to doubt the result. Actually, even without spatial averaging of any type, I believe the test would be valid so long as everything remains the same except for the change of test subject.
Same room, same stimulus, same location in the room, etc. etc. One device measures 30% better than the other in the 50Hz band. I can't find a flaw in that.
DD

Right. Let me point out again that it was 2 comparisons that I noted. 4" 703 and our 244 which are 2 different panels. Which (not to brag) our 244 is much more effective then just using 4" or 703 at the lower frequencies.

Edit:

Dan and just to add if you compare the Monster and the 244 you will see they both follow each other from 50hz up. The difference between the 2 products are 2" and both where tested on different days. If a lab was that far off you would see a MAJOR swing between the numbers, but if you look they follow each other pretty darn close.
GIK Acoustic Panels are tested and certified.

[DanDan]

Impressive Glenn, although I would rather see the vertical axis run from 0 -3. The non linearities :-) are exaggerated by that scale.
There has been some to-ing and fro-ing about LABS etc. It probably appears confusing and contradictory at this point.
So IMHO
That LAB you use is clearly confident down to the 63Hz band. It is probably very large and certifed. Not many are certified down to this frequency band. Historically, Building Acoustics tests were to 125Hz only. Lately they are being extended downwards, to include even 31Hz in one extreme case (HVAC Noise) Older and smaller LABS, go to 125 only and are therefore not at all suitable for our purposes.
When performing tests in these LABS, RT is measured over a time and spatial average.
i.e. a number of sweeps with the microphone, or multiple microphones, at different locations. The measuring mic(s) are sometimes on a rotating boom, to further randomise the spacial average. Ultimately when the average is calculated, the result will of course be very similar each time. Statistics ensure that. However any individual sweep, even with a fixed mic, seems to give a dfferent number each time. This has been observed by myself and other Slutz here, in LABS and in site test situations. It also happens when evaluating acoustics of a room for sound purposes.
Claro?
DD

[Glenn Kuras]

That LAB you use is clearly confident down to the 63Hz band. It is probably very large and certifed.

It is certified I believe to 125 hz, but I am very confident by just looking at the way our numbers come out. Heck Dr Sabin was the founder.

[Ethan Winer]

Originally Posted by DanDan

the Rotating Boom is common in the ISO and other Testing Standards. Multiple locations are also used, with limits keeping the mic and speaker away from trouble.

A rotating microphone, plus averaging 100 runs, is great at mid and high frequencies. But it is not so great at low frequencies. I'm sure I mentioned this before, but I've seen negative numbers reported for very low frequencies. I've spoken extensively about this with Dr. Matt Nobile who heads IBM's lab, because my first visit I couldn't understand why he was so adamant that testing below the certified frequency is not reliable.

That LAB you use is clearly confident down to the 63Hz band.

As Glenn said, Riverbank is not certified to that low. I assume it is valid down to 100 Hz, not 125 Hz, because 100 Hz is part of the 125 Hz octave band. At least that's what Dr. Matt told me about his IBM lab. And maybe you can sort of trust 80 Hz. But 50 Hz? No way. Though I wish it were so too - it would make our lives so much simpler! Because these labs are not certified below 100 Hz is why they will not list absorption coefficients on the official reports. They'll list sabins as measured if a customer requests it, but they will not list AbsCo.

--Ethan

[Glenn Kuras]

At least that's what Dr. Matt told me about his IBM lab. And maybe you can sort of trust 80 Hz. But 50 Hz? No way.

I think you should talk with David Moyer at Riverbank. You are right about pretty good down to 80hz and have to be careful at 50hz, but our number prove (as I stated above) that if done right you can get pretty darn close.

[DanDan]

Like any tool or instrument, results obtained are dependent on technique and operator skill. A rotating boom cannot randomise VLF. However one can move the boom centre to various different carefully chosen locations.
The use of multiple locations or sometimes even multiple microphones is common I believe. I am not sure that rotating booms are often used in the big rooms, I have seen and used them in smaller semi rev rooms. Multiple locations can extend downwards the validity of the spatial average, plus of course the spectrum of useful results. A room can only be certified to a certain frequency, directly related to it's size. That does not mean the room is useless below that frequency. With ingenuity and skill one can get useful results, certainly in comparative tests.

DD

[Ethan Winer]

Originally Posted by DanDan

I am not sure that rotating booms are often used in the big rooms

The photo below of Riverbank's lab on Glenn's site sure looks like a rotating boom arm.

--Ethan