Acoustical Treatments Master Thread - Page 221

I bought the 24" x 48" x 2" 703 at about $75 for TWELVE sheets. From each sheet, I made eight 17" x 17" x 24" triangles.

That was 3-4 years ago...

Jeff

Wouldn't that (whether to remove the kraft paper or not) depend on whether your room needs additional mid/high absorption or not?

Crosssectionally .. is that a word? .., as SSC bass traps are constructed, the paper won't make any difference in mid- or high-frequency absorption.

Of course, I might have incorrectly inferred that the context *was* SSC bass traps. If the OP was referring to a sheet straddling the corner, then you are correct.

Jeff

Ah gotcha, I wasn't thinking superchunk.

Yes, exactly. For corner bass traps, having a facing is very useful. It increases LF absorption substantially while reducing mid and high frequency absorption to avoid making a room totally dead sounding. Of course, at reflection points and a few other key places, no facing is needed.

--Ethan

In a couple of weeks I will have 6 acoustic panels - CSR Ultratel glass fibre Acoustic Panels. These are made from double layer 48kg fibreglass batts and are 100mm thick, 600mm wide and 1800mm high and there are 6 of them. They work as broadband absorbers and can also be stacked for corner bass traps.

There will be two Bass Boxes in each corner (beside the kitchen opening)
Bass box = Width: 60 cm X Depth: 70 cm X Height: 70 cm
Dual opposed 15" drivers so one end is firing into the corner.
Frequency up to 320Hz from the bass boxes. Higher frequencies from a horn.

There is 180+ cm between the front 'wall' (opening to kitchen) and mantlepiece so I can lay them on their side there. The ceiling is pretty high! (~11')

The room will be used for HT and music. The fixed Xscreen (200cm) drops down from above the kitchen opening (projector on the rear wall above the sofa).

Any suggestions on where best to place them in my room?

 

Room Acoustics dimensions.pdf 13.7685546875k . file

Ethan,

I downloaded and ran your test tone cd yesterday.

I set pink noise to 70 prior to the test.

Here are some findings:

A. I can get fairly flat from 24Hz to 56Hz at 73db.

B. Then a slow decline at 1Hz intervals from 57Hz to 72Hz a drop of about 13db to 60db.

C. From 73Hz an incline up to flat at 80hz, it stays that way to about 115Hz bouncing between 70-72 db.

D. From 116HZ there is a steady decline until it bottoms out at 50db at 158Hz.

E. Then I have a steady incline up to 170Hz..then flat until 210Hz.

F. at 211Hz it starts to decline down to 52db at the 221Hz.

G. It then zig zags with slow inclines and declines at these levels:

62db at 230Hz
52db at 233Hz
64db at 242Hz
52db at 255Hz
72db at 262Hz all the way to 300Hz

Sorry I dont have a picture of the plots but what is the best way to handle some of these valleys?

My room is roughly 17.5 x 12 x 7.5

Thank you for your help and the test tone cd is great.

K.

See this:

Acoustic Basics

--Ethan

That's quite typical, and the solution is bass traps. The more traps you have, the closer you'll get to a flat response. Yes, it's that simple.

--Ethan

A different and perhaps a more sightly and economical approach would be to use an appropriate parametric EQ. Also, you need to determine just what is causing the "dip" to occur? Is it modes (unlikely if it is that wide) or is it something in the room that is causing a suckout, or is it something else, SBIR, speaker/listener positioning? This should be determined first if you want to treat it correctly. Best wishes!

Thanks, I have your site bookmarked.

I hope I can quote from a non-subscription site. The fallacy of this very well explained by Ethan:

"Trying to use an equalizer to fix room acoustics problems does not work very well. Every location in a room has a different response, so no single EQ curve can help everywhere. Even if your goal is to correct the response only where you sit, it's impossible to counter nulls. If you have a 25 dB dip at 60 Hz, adding that much boost with EQ will increase low frequency distortion in the loudspeakers. And at other places where 60 Hz is too loud, EQ makes the problem worse. EQ can reduce peaks a little, but it does not reduce the extended decay time that accompanies most peaks. Our Audyssey Report article explains why EQ is not a suitable substitute for bass traps and other treatment."

Not quoting a site...but of my own volition...I hope you recognize that the description you provided is not entirely correct. Ring time is a function of the peak itself. The additional energy of the peak provides a longer ring time (check out a waterfall plot of your room if you don't believe me) due to the fact...well...it takes more time for that energy to decay to an inaudible level. By reducing the energy of a peak via a PEQ, you thereby also reduce the ring time. I certainly hope you did not come away with the idea that bass traps strung throughout the room can reduce a null by increasing that null's energy. Bass traps work the same way as a PEQ just less targeted (unless you are using a Hemholtz resonator)...they reduce the peaks...which thereby reduces the depth of the null. Believe it or not though, you can sometimes increase a null's energy via a PEQ (generally not a lot but you can) without causing audible damage or audible distortion...but I've never heard of a bass trap increasing a null's energy. I am not saying that I don't incorporate bass traps...I do through design...but I certainly do not rely on them to deal with frequency problems below 100 Hz. It's just not practical or the best method IMHO. Best wishes!

Ring time's relation to peak energy is linear. Of course.
You seem to be a knowledgeable fellow, so your characterization of a null as absence of energy and insistence that it can't be treated successfully with a bass trap is puzzling. It all depends on how you define "energy". A null is in fact the meeting of two (or more) sound waves 180 degrees out of phase with each other. One large cause of this is a sound wave reflected off the back wall meeting incoming waves from the speaker. If bass traps can get rid of, or ameliorate, this reflection, the null can be reduced considerably. In this case "reduction" of a null equates to bringing the energy back up at the null's frequency.

But I'm sure you know this... that's why it's puzzling.

PS I don't quote what I don't understand.

It's not really puzzling. It's just a relation of the surrounding frequencies to the null. If you drop the peaks...you eliminate the effects of the null, but what you are really eliminating is the harmful effects of the longer ring times. Yes, your analysis is correct, but in order to really effectively treat that peak....it would need to be approximately 1/4 the wavelength of that peak you are trying to reduce. In the case of 50 Hz...it would need to be 5-6 feet thick. 1/10 the wavelength is also possible provided certain specifics are adhered to...but the trap would still be 2.25 feet thick. Now, another factor to consider, bass traps as the main stream thinks of them, are made of fiberglass which is a frictional absorber. Frictional absorbers work best when something is moving against them. Along the walls...the velocity of the wave is near zero...but the pressure is much higher (boundary gain anyone) especially for low frequency waves that we are referring to. Now, since the velocity is near zero, the frictional absorber doesn't work as well as it would say at the point of lowest wave pressure and highest velocity of that wave...which is not along the walls but out in the room itself. So, I am not saying you can't use bass traps to deal with freqs below 100 Hz, but they would be impractical and obtrusive to be truely effective. Engineering and PEQ are the only way to go IMHO. BTW...you don't seem too shabby yourself in the knowledge department! Best wishes!

So my Main issues are in these areas:

A. 57Hz to 79Hz
B. 116Hz to 168Hz
C. 211Hz to 221Hz
D. 230Hz to 259 Hz

For the most part the rest of the test tones were +/- 4db

Sure, it is linear but the actual slope depends on the Q of the peak and can be compensated for with a filter of appropriate Q.
[/quote]

Yes. But Ethan's point is that Q (and the amplitude of the peak or null) exists in one place in the room. What about the other places? How do you use an EQ to correct the many different response curves a room has in different locations? It seems it's better to try to stop the waves from bouncing around and interacting all over the room as much as possible first and then apply EQ. I do have a Behringer FDP EQ and I use it.

So: SierraMikeBravo, I thought the bass traps were triangular or put across corners so they could interact with the higher velocity movement. And, I thought as the energy of the sound was absorbed and converted to heat, the wavelength no longer mattered for the portion absorbed because the pressure goes away with the absorption of the energy. This, of course, is only true for the portion of the energy that the trap can absorb.

So I have a question: when the energy of a sound wave is absorbed, does the distance between the absorbing material and the wall still matter other than getting the material into the high velocity area?

IMHO, the question should really be, what's the goal?

All of my comments below refer only to absorption of sound energy in air.

There are two commonly used mechanisms to absorb sound energy: (1) velocity type absorption; and, (2) diaphragmatic or pressure absorbers. For a velocity type absorber to function effectively, that device must be placed in a location where the velocity of the air molecules are at their maximum velocity. For a pressure type absorber to function effectively, that surface must be placed at a location where air molecule velocity is zero (or near zero) and pressure is at its maximum.

If we imagine an air molecule striking a wall, at the instant it strikes the wall, its velocity is zero and pressure is high. However, sound propagation is cyclic in nature and cycles through rarefication and compression. At maximum rarefication and maximum compression, velocity is zero and pressure at its maximum.

To achieve maximum effectiveness in a velocity absorber, that absorber should be placed such that it is located at the 1/4 wavelength of the frequency(ies) for which we wish to reduce energy. Velocity absorbers are characterized by fiberglass batts or fiberglass panels. At 80Hz, the wavelength is 14' and the quarter wavelength is therefore 3.5'. If a dimension in your room is any multiple of 14', then a velocity absorber would need to be placed 3.5' away from a wall surface to be effective at 80Hz.

A simple form of pressure absorber is a fiberglass batt with paper backing. If you imagine a trampoline, the paper backing is the surface of the trampoline and the fiberglass is the spring. Now, imagine jumping off the roof of your house onto your trampoline. The trampoline will indeed absorb some of the energy from your fall; but, the spring is going to release some of that energy back throwing you back into the air. The spring action is an issue in the design of diaphragmatic type absorbers since they can become "speakers".

One of the characteristics of modal frequencies is they, by definition, are always at their minimum velocity and at maximum pressure at the wall surface. In this situation, a pressure type absorber is more effective at the wall surface. A velocity type absorber would need to be placed at a 1/4 wave length distance from the wall surface. (At 80Hz, that's 3.5'.)

In the low frequency arena, the bigger (not only) sound quality problems are with modal frequencies (and some would argue that in a small room, it's the first three axial modes which are the real killers). [Note: in small rooms modal problems dominate from approximately 300Hz downward but generally present the biggest audible problems below 100Hz.]

The reason we typically find vendors suggesting their bass control devices be placed in room corners is because it is at the room corners where all the axial modes are at their highest pressure/lowest velocity.

The statement made by SMB is true in that general purpose bass traps/absorbers are not very effective below 100Hz. For them to work at the velocity level, they'd have to be 14' from a wall (assuming a 20Hz modal frequency). For them to be effective as pressure absorbers, they'd need to be at the wall surface (for modal frequencies). Neither is very practical. When placed any distance from a wall, their effectiveness would vary by frequency based almost entirely on their placement (a 6" difference can radically change the frequencies upon which you'd see the greatest effectiveness).

Multiple tools need be utilized to resolve low frequency sound quality issues in small rooms. These tools would include electrical as well as mechanical.

As to the question about nulls/peaks existing in multiple areas of the room, a couple of points. First, we don't give a tinker's damn about sound quality in areas of the room where no one is sitting. Secondly, when it comes to LF and modal issues, it is the amplitude of the peaks/nulls which are audible and cause us grief. As the energy to that frequency increases, the delta between the peak and null gets larger and more audible. The converse is true as well. As energy is absorbed, that delta decreases and you have more consistent response (I didn't say good, I said "more consistent"). The use of differential parametric EQ, is one means be which modal frequencies can be 'resolved' without sucking the life or energy out of a room.

Among the challenges for a well performing room is to have all frequencies not only decay at the same rate (in the seating locations), but all frequencies to have the same relative SPL. This is a significant challenge since any form of pressure, or velocity type absorber will affect different frequencies differently just based on their position in the room. Adding bass trapping devices to a small room is very helpful. It cannot, however, be a helter skelter placement nor simply putting fuzzy stuff in the corners. There will be a point at which the treatment ceases to be helpful and begins to work against your objective.

It may seem like an impossible task and there are theoretical issues. Nonetheless, there are several systems that do achieve significant improvements.


I completely agree that physical room treatment is a better approach but most of us find that we are limited in how much of that can be accommodated. So, we do that first and finish up with EQ.


Yes, it will influence the frequencies that can be efficiently absorbed. (I leave the rest for the more techy experts.)

To those who have built/used Helmholtz resonators: can these things effectively help tame peaks (and presumably therefore the corresponding nulls) in a typical small room HT? By "effectively," I mean "a real improvement in SQ."

I ask, because I had an idea that would combine the use of these resonators and the use of corner spanning rigid fiberglass. Why not build resonators with triangular boxes (tuned to whatever problem frequencies you may encounter in the sub 300hz range) and place them in corners behind a span of say, 4" rigid fiberglass? If built and positioned correctly, such a combo could treat frequencies lower than 100hz all the way up, right?

One would have to take lots of measurements to find out which corners best support which peaks and then place the resonators accordingly. If all wall/wall, wall/floor, and wall/ceiling corners were employed, you could have a room filled with multiple resonators hidden behind fiberglass.

I plan to do this in my HT (as soon as I get to that stage- still a way to go). If no one has experience or advice, then I'll post results later... at my rate, probably much later .

Thanks,
John

That simply is not true. The relative decay time remains the same. Now, you could argue that once a ringing peak is reduced in amplitude that the ringing will be less objectionable. And this is true. But the slope of the decay remains the same. Versus bass traps that really do reduce the decay times. Another huge advantage of bass traps over EQ is that bass traps reduce the Q of peaks. This has a marked affect on clarity, and goes a long way toward eliminating the problem known euphemistically as "one note bass." EQ cannot alter the Q of a room's resonant peaks.

The following graphs were measured in a small room, 16 by 11.5 by 8 feet. It's the same room, and same data, as shown in my company's Hearing is Believing video.

Here are Before / After graphs showing the reduction in decay time, and the reduction in Q of the peaks, after adding bass traps to the room;





The graphs above also prove that conventional "porous" type bass traps can indeed be highly effective to as low as 40 Hz and even lower. In this case the traps are only six inches thick, and most are not even mounted straddling corners.


That's a joke, right? The waterfalls above prove the point very well, and the response-only graph below shows Before and After overlaid to make the difference even easier to spot. I'm sure I don't have to state which line color is with bass traps and which is without.

--Ethan

Not really. A null has low energy because of the cancellation occurring there. If you add treatment to mitigate the modal energy and smooth the response, there will be more energy at that point/frequency (as you indeed show). Otherwise, it would still be a black hole.

Now don't get all technical on me Kal!

Using common acoustic-ese, the "energy" of a null relates to the SPL at that location in the room. A lot of people wrongly believe that bass traps reduce the level of bass in a room. If a room is dominated by peaks, such as is common in a square or cube shaped room, the overpowering bass at select frequencies will indeed by reduced. But in my experience, the larger problem in most rooms is nulls. So usually, after adding bass traps the perception is that of more bass. Not sure if that's also the case in your 2:2:1 room though.

--Ethan

It is and that is exactly the point I was trying to make.

Hi Dennis,

Great post--well, all you posts are darned good, but this is a keeper! One little question, though, when you said:

What do you mean by 'differential' EQ?

Tnx!

Man, I tell ya, Dennis gets all the love!

Heck if I know ... made use of a big word though.

When you have multiple subs in the room, the first step in calibration is to have them all level matched and at the same phase. You can then calibrate in real time by adjusting phase, PEQ, and SPL of each sub (individually) to deliberately create constructive/destructive interference to further deal with modal issues. That is differential EQ. Currently (at least for me) this is pretty much trial and error along with knowing from experience what worked and what did not. Differential EQ would be an excellent PhD thesis for someone working toward a PhD in acoustics. (Hey, Todd. You busy? )

Example with two subs. We know when you have two subs in a room, the summed behavior of those two subs is identical to having a single sub mid-way between the two physical subs (virtual subwoofer). By tweaking the SPL of the physical subwoofer you can "move" the location of the virtual sub along that line between the two physical subs. You can do that with four subs (much more a challenge and time consuming). As well if some of those multiple subs are close to the ceiling you can manage the height of the "virtual sub" as well.

Fascinating!