Stuart's Rules for Soffits / Flush mounting speakers

[endorka]

Here is a list of "rules for soffits" Stuart wrote some time ago. I think he mentioned updating these at some point, so consider it a document in progress.

Cheers! Jennifer

1) The speakers should not go in the corner, nor on the line that divides the corner. In other words, if your walls intersect at  90 °, then draw a line out from each corner at 45°, and stay away from that: don't put your speaker exactly on that line, since it implies that you'll be getting the same artifacts from the side walls as from the front wall. Put your speakers either outside or inside of those lines. More commonly you'll want your speakers "inside" those lines (more towards the center line of the room).

2) The "38% of room depth rule" is not a rule, but it is a useful guideline for a starting point. You'll generally want to have your listening position a bit closer to the front wall than that location, but do be aware that you might be getting into problematic SBIR territory there. (You can treat that, to a certain extent).

3) Keep the mix position away from 25% and 50% of room depth, and try to stay between about 32% and 44%

4) You can angle your speakers differently than the "textbook" 30° angle: Anything in the range 25° to about 35° will work well under most circumstances.

5) Keep the speakers as far apart as possible, while not violating rules 1 and 4.

6) Keep the mix position at a good distance from the speakers, within the range of about 1m to about 5m. Further away is usually better.

7) Don't put the speakers at 25% of the room width: that's a modal null for some frequencies, and a peak for others. Try something more like 28% to 34%.

8 ) Make the front baffle of your  soffit  as wide and tall as you can, within reason. The width should be at least three times the diameter of your low frequency driver. In other words, if you have a speaker with an 8" woofer, then you want the  soffit  baffle to be at least 24" (60cm) wide. Wider is better.

9) Do not put your speaker in the middle of the  soffit  baffle: Offset in both directions. In other words, the distances from the acoustic center of the speaker to each edge of the baffle should be very different, by at least 20%. So for example if your speaker axis is 30cm from one side of the baffle, it should be more than 36cm from the other side, less than 24cm from the bottom edge, and more than 44cm from the top edge. (Rough distances, for illustration only...). Larger differences are generally better.

10) Make the baffle as massively heavy as you can, and as rigid as you can.

11) Make the structure inside the  soffit  (the framing that holds the baffle and speaker in place) as rigid and massive as you can.

12) Mount the speaker inside an enclosure box that is either a very tight fit, in order to keep the speaker rigidly fixed in place, or mount it on suitable rubber pads, to completely decouple it from the the box. Carefully choose the properties and dimensions of that rubber, to make sure the speaker is still decoupled down to at least one octave below the speaker's low cut-off frequency.

13) Take into account that speakers need a lot of space behind them for cooling, and a path through the  soffit  for cooling air to flow.

14) Rear-ported speakers need special attention: Do not overload the rear port, acoustically, with an enclosure box that is too small, or un-ventilated, or un-damped.

15) Damp the hell out of the  soffit  interior! Fill it entirely with suitable damping if you want, except for the cooling path.

[Soundman2020]

I think he mentioned updating these at some point

Yup! Hopefully "some point" will arrive one day soon...


- Stuart -

[gullfo]

here's an example soffit construct. behind this would be a frame attached to the wall to secure it vertically and also sectioned with mesh to support the insulation stack. note the use of duct board to create the passive venting for the speakers. you can also add conduit for routing power and audio cables to powered speakers. also note it's possible to add fans to the duct board venting to force air flow.

[Soundman2020]

Thanks, Glenn!

Note to everyone else: The abundant framing on the front baffles, especially around the speaker cutouts. There's a couple of reasons for that, but one of them is to add additional rigidity to the baffle. Simple rule for building soffits: Beef it up!

- Stuart -

[gullfo]

a couple more notes - each part of the baffles is a flat unit. they are then fastened together using ripped lumber to create the angles. this significantly simplifies the calculations and construction, and forms a solid connection between frames. also note the plywood or mdf, below the front baffles, on the inside of the frames. this allows a soft absorptive area essentially in front of the desk to capture reflections and the hard backing extends the baffle plate surface area for LF.

[DanCostello]

Can someone help me understand this bit:

Mount the speaker inside an enclosure box that is either a very tight fit, in order to keep the speaker rigidly fixed in place, or mount it on suitable rubber pads, to completely decouple it from the the box.

I understand mounting the speaker securely so it doesn't move; and I understand decoupling it from the box. What I don't understand is why this is an either/or situation. Does decoupling keep the speaker from moving? Or am I just getting tripped up by some clumsy wording?

-Dan.

[Soundman2020]

Hi Dan. There's two schools of thought here, and both work.
One is that you mount it so darn rigidly that it cannot possibly move, and therefore cannot transfer vibrations into the soffit structure. To do that, you need a lot of mass and very rigid structure, especially if the speakers are large and you run them at high levels.
The other is that you mount it on tuned resilient pads so that it cannot possibly transfer any vibration to the soffit structure. You can get away with a bit less mass and a bit less overall rigidity this way, but it does require more careful mounting since you need to "tune" the resilient pads such that the resonant frequency is very low. At least an octave lower than the lowest frequency the speaker will be producing at high levels.
In both cases the goal is to prevent the speaker from transferring it's own vibrations to the soffit structure, especially the front baffle. So there's two different ways of achieving that goal: rigidity, or resilience. Both ways work, both have their pros and cons.

- Stuart -

[gullfo]

- Stuart -

welcome back!

[Soundman2020]

- Stuart -

welcome back!

Thanks Glenn!

(Dark times... now behind me hopefully. Slowly getting back into the swing of things.)


- Stuart -

[DanCostello]

Source of the post Hi Dan. There's two schools of thought here, and both work.
One is that you mount it so darn rigidly that it cannot possibly move, and therefore cannot transfer vibrations into the soffit structure. To do that, you need a lot of mass and very rigid structure, especially if the speakers are large and you run them at high levels.
The other is that you mount it on tuned resilient pads so that it cannot possibly transfer any vibration to the soffit structure. You can get away with a bit less mass and a bit less overall rigidity this way, but it does require more careful mounting since you need to "tune" the resilient pads such that the resonant frequency is very low. At least an octave lower than the lowest frequency the speaker will be producing at high levels.
In both cases the goal is to prevent the speaker from transferring it's own vibrations to the soffit structure, especially the front baffle. So there's two different ways of achieving that goal: rigidity, or resilience. Both ways work, both have their pros and cons.

- Stuart -

Ah okay, that makes sense. Thanks.

It also jogs my memory - I seem to recall some discussions in threads years back about designs that had these soffit structures built in two parts: one that was, effectively, the speaker stand and then a second that held the soffit face and was decoupled from the first.

Any thoughts on that approach (assuming I’m remembering it correctly)?

I’m trying to think of ways to do this that keep the equipment easily serviceable while being within my somewhat limited construction skills.

[gullfo]

ultimately what you want from the soffit mounted speakers:

a) extended baffle plate to put the speaker in 2π and eliminate the LF SBIR from the front wall (and preferably extended to some of the side walls to enable a smooth interface)

some info on this:
Full Space = 4π steradians = This represents radiation into free space, that is in the open with no walls, floor or surfaces nearby. 0 dB reference.

Half Space = 2π steradians (commonly specified speaker load) = the speaker on an infinitely large baffle then the front of the speaker would be radiating into half space. The plane divides all of space into two halves. +6dB SPL increase compared to full space.

Quarter Space = π steradians = the speaker placed at the intersection of two infinitely large perpendicular planes. Approximated by the intersection of two walls. The two planes divide all of space into four quarters. +12dB SPL increase compared to full space.

Eighth Space = π/2 steradians (1/8th freespace) = the speaker placed at the intersection of three walls, such as in the corner of a room The three planes divide all of space into eight parts. +18dB SPL increase compared to fullspace (not factoring in ceiling contribution of the room).

b) no extraneous resonances from the assembly (no sense having the assembly radiate unwanted sounds or harmonics)

c) little or no structural transfer to avoid skewing the listening position response (because sound travels through e.g. concrete at about 10x the air, and presuming you have a quiet room, there is potential for those sounds to reach you before the air with enough amplitude to cause your hearing to not properly separate the direct, transferred, and reflected sound as best as it could).

[PLG-88]

Hi, I'm new here and I'm not sure if I'm supposed to ask questions in the reference area, but this relates to the topic and therefore might be useful for others who are gathering information on flush-mounting monitors as well. If this is not the right place to do so, please notify me.

Source of the post 10) Make the baffle as massively heavy as you can, and as rigid as you can.

I'm trying to understand the practical consequences of flush mounting monitor-speakers a little better and therefore I have a few questions in relation to the rigid and heavy baffle: I'm curious how an angled, heavy and rigid partition/leaf will affect room modes and how it possibly might create a triple leaf situation.

1. I'm aware that there are different school of thoughts about the geometry of an inner shell (rectangular or splayed walls), but a lot of books and designers recommend a rectangular shell for the prediction and behavior of the room modes in such a shell. A lot of flush mounted systems I've seen incorporate a baffle from floor to ceiling and obviously under a 30 degree angle. In case of a rectangular shell: introducing such a heavy and rigid baffle (wall) in the room must have an influence on the front to back axial room mode, right? Because you're altering the dimensions and shape of the inner shell? If so, that seems to be in direct conflict with the advise to stick with a rectangular shell.

2. In the case of a two-leaf (MAM/MSM) inner en outer shell, introducing a heavy and rigid baffle would create a local three-leaf system?
Isn't this problematic? Obviously not, because of the examples below by respectable studio designers below, but I'm curious on why this isn't a problem.

Very clear three partitions/leafs (MAM+baffle) can be seen in these examples:
Glass baffle 1 (Diante do Trono studio, Brazil by WSDG): https://wsdg.com/projects-items/diante- ... #iLightbox[gallery386]/3
Glass baffle 2 (Red Bull Studios, Paris by Northward acoustics): https://northwardacoustics.com/wp-conte ... 013-23.jpg

There are other examples with non-glass (stone) baffles as well, (for example in Philip Newell's book Page 437 and page 758) (Yes, I've read the door-stopper comments concerning this book on this forum) but the glass ones obviously show the three leaves better to illustrate my question.

3. In relation to the question above: Is it possible that the baffle for the flush mounted monitors IS the inner-shell of a two-leaf system? Since the mass and dampening(insulation) is already there for sound-isolation purposes? (Obviously taking in account a continuously layer of mass around the recessed monitors in that wall and maintaining an appropriate airspace between the back of that mass and the mass of the outer shell, in relation to the calculated TL of that partition?) If this is possible you would avoid a three-leaf system.

[Soundman2020]

Hi, I'm new here and I'm not sure if I'm supposed to ask questions in the reference area, but this relates to the topic and therefore might be useful for others who are gathering information on flush-mounting monitors as well. If this is not the right place to do so, please notify me.

Hi there, PLG-88, and Welcome to the mad house! I mean forum!

You sure can ask questions in the reference area, or anywhere else. In general it's better to start your own thread and post links referring back to the original post, but you can do it this way too.

I'm aware that there are different school of thoughts about the geometry of an inner shell (rectangular or splayed walls), but a lot of books and designers recommend a rectangular shell for the prediction and behavior of the room modes in such a shell. A lot of flush mounted systems I've seen incorporate a baffle from floor to ceiling and obviously under a 30 degree angle. In case of a rectangular shell: introducing such a heavy and rigid baffle (wall) in the room must have an influence on the front to back axial room mode, right? Because you're altering the dimensions and shape of the inner shell? If so, that seems to be in direct conflict with the advise to stick with a rectangular shell.

Well, to give a crystal clear answer: sort, of yes, but not really. Maybe.

OK, what I mean with that, is that you are correct about room modes being related to the room boundary, yes. And that means the hard, physical inner-leaf wall that separates the room volume from whatever is on the other side. That "shell" has to be sealed, air-tight, in order to act as a true boundary. Soffits are not air-tight, and they don't create an additional boundary, nor do they modify the boundary, so from that point of view, they don't really affect the overall modal response. They are surfaces within the room. And from that point of view, yes, of course, they do have an effect on room acoustics.... but acoustic response in the room, is not the same as modal response. They are somewhat related, yes, but not the same. It's complicated!

In other words, it's a complicated issue. They don't really affect the underlying room behavior, since they are not additional "boundaries" to the room envelope. But they do affect the acoustics inside the room, because they are large, massive surfaces inside the room. Just like a table, sofa, console, or gear rack, they do indeed have an effect on the way sound moves around inside the room. Just like acoustic treatment panels. And for that same reason, yes, they can also have an effect on the modes. Just like you can build a "trap" of some type that is designed to interfere with certain modes, so too your soffits can (and do) interfere with some modes in one way or another. And that's a GOOD thing!

Firstly, just from being so large and massive, the front baffles can have an effect on the modal response. Often, that just means the they "smear" the mode a bit. Instead of being a sharp peak or null (as measured at the listening position), the Q of the peak is a bit lower, and more spread out. That's useful. The overall center frequency doesn't change much, since that is set by the room boundary conditions and absorption, but the peak can be lower and broader.

However, where soffits really shine is with all that empty space inside, above, and below. That can be put to great use for room treatment, notably: bass traps. Bass traps need to be large, deep, and preferably in the room corners.... which is exactly what soffits are! Large, deep, and in the room corners. So they are ideal candidates for bass trapping. In rooms I design with soffits, I use most of the available space inside for exactly that, usually with acoustic "hangars". It can be quite effective. So, from that point of view, yes soffits can have an effect on modal response, just like any other room treatment.

2. In the case of a two-leaf (MAM/MSM) inner en outer shell, introducing a heavy and rigid baffle would create a local three-leaf system?
Isn't this problematic? Obviously not, because of the examples below by respectable studio designers below, but I'm curious on why this isn't a problem.

In theory... partly yes, partly no. Yes, because as you point out, a soffit baffle is a large, massive panel in close proximity to a room wall. But no, for several reasons:

Firstly, the distance are great enough that the frequency is below the audible spectrum. A typical soffit measures maybe 50 or 60 cm deep on the narrow edge, and perhaps a meter on the other edge. That puts the resonant frequency way down. Also, they are massive, rigid surfaces, and there's a huge amount of damping inside, which also moves any resonance way down low. Thirdly, they are angled with respect to the actual leaf, so there isn't one single resonant frequency, but rather a range of frequencies at lower intensities, none of which is dominant. And fourthly they are not sealed cavities. Typically, a soffit has large openings above and below, to allow for air ventilation to flow up past the speaker, for cooling purposes, as well as the opening for the speaker itself, and possible other openings (all inside the soffit, of course). A cavity doesn't have to be sealed to be resonant, but a sealed cavity is far more effective. So the fact that a soffit is not sealed, would make it less effective as a 3rd leaf than it would be otherwise.

(Yes, I've read the door-stopper comments concerning this book on this forum)

It also makes a good riser for supporting a keyboard that's a little too low for comfortable operation. I'm sure it has other uses, too...

but the glass ones obviously show the three leaves better to illustrate my question.

If you do the math (which is a bit convoluted for complex situations, such as that) I'm sure you'll find that the resonant frequency is way too low to be of concern.

3. In relation to the question above: Is it possible that the baffle for the flush mounted monitors IS the inner-shell of a two-leaf system?

[/quote]Not really, not. That would imply that the speaker itself is located in the wall cavity, between the two leaves... so it would not be isolated from whatever is on the other side of the wall. You would have a sound source inside your wall, so it would not benefit from the MSM system at all. You'd just have simple mass-law isolation, in both directions. Plus, there's the issue of how you would cool such a speaker, and how you would seal the wall air-tight for maximum isolation...

There's also the issue that many good studio monitors today have rear-facing reflex ports on them (or other rear-facing acoustic elements). Having those sealed up inside the wall, would be a problem.

- Stuart -

[PLG-88]

Source of the post
Hi there, PLG-88, and Welcome to the mad house! I mean forum!

You sure can ask questions in the reference area, or anywhere else. In general it's better to start your own thread and post links referring back to the original post, but you can do it this way too.



Thank you for the warm welcome and your detailled reply!

Source of the post

They don't really affect the underlying room behavior, since they are not additional "boundaries" to the room envelope. But they do affect the acoustics inside the room, because they are large, massive surfaces inside the room...
...Firstly, just from being so large and massive, the front baffles can have an effect on the modal response. Often, that just means the they "smear" the mode a bit. Instead of being a sharp peak or null (as measured at the listening position), the Q of the peak is a bit lower, and more spread out.



I guess that having a baffle that covers the complete front wall of de room will have a bigger effect on the room modes that something than Glenn detailled in a few post above. I can understand that such a design is just interfering with the acoustics of the room such as furniture as you've explained.

There is one thing I'm not understanding quite right. According to Floyd E. Toole's book 'Sound Reproduction', mounting a speaker inside a wall, flush with the surface, results in a 2Π half-space radiation. As far as I understand it correctly, this means that the low end becomes more directional towards the front because the LF produced by the speaker can't 'fold around' the speaker because of the wall/baffle. This results in less/no SBIR and a +6dB output of LF. Hopefully I've got this right.

My question is: Does a monitor-speaker 'leak' any LF through the back of the speakerhousing/cabinet? And if so, are these amounts of leakage enough to be reflected of the shell-wall, back through the LF trap, back in to the room in case of Glenn's design with a relative large opening below and above the baffle?

From other respectable sources (e.g. this onehttps://gearspace.com/board/showpost.php?p=11805742&postcount=5) I was under the impression that the baffle should be an airtight/closed construction. In this post Thomas further mentioned calculating the resonance frequency of this space behind the baffle to make sure that frequency is very low further implicating a 'closed box'.

Is this just another school of thought achieving the same goals or I am overlooking something fundemental here?

Source of the post
Firstly, the distance are great enough that the frequency is below the audible spectrum. A typical soffit measures maybe 50 or 60 cm deep on the narrow edge, and perhaps a meter on the other edge. That puts the resonant frequency way down. Also, they are massive, rigid surfaces, and there's a huge amount of damping inside, which also moves any resonance way down low. Thirdly, they are angled with respect to the actual leaf, so there isn't one single resonant frequency, but rather a range of frequencies at lower intensities, none of which is dominant. And fourthly they are not sealed cavities.

Typically, a soffit has large openings above and below, to allow for air ventilation to flow up past the speaker, for cooling purposes, as well as the opening for the speaker itself, and possible other openings (all inside the soffit, of course). A cavity doesn't have to be sealed to be resonant, but a sealed cavity is far more effective. So the fact that a soffit is not sealed, would make it less effective as a 3rd leaf than it would be otherwise.

Thank you, this makes a lot of sense. Well, apart from two things, one of them is the sealed cavity issue which I've tried to inquire in the paragraph above, but also the "opening for the speaker" you've mentioned. I was under the assumption, also partly by this document (https://support.genelec.com/hc/en-us/ar ... nstructed-) that the box where the monitor-speaker is positioned should have a closed back. So that it doesn't have any opening towards the space behind the baffle.

Obviously the cooling of the amp of the monitor becomes a problem with a closed box and that's why some sources claim that only monitor-speakers with a detachable/remote amplifier section are suitable for flush-mounting.

Sorry if I come across trying to debate the positions that have come forward in this thread, that's not my intention. I'm not professional acoustician at all, but I'm trying to learn and sort out various viewpoints and (very likely) possible misunderstandings and lack of knowledge from my part.

[gullfo]

the speaker itself is decoupled from the assembly, and sealed to the front. the baffle plate, while not "infinite" is large enough to require the baffle step compensation you would expect in the fully enclosed wall system. bear in mind, the entire back of the assembly is a LF trap so attenuation of LF on that wrapping top and bottom of the baffle plates will be significantly reduced and not cause SBIR.

you'll note in my drawing example the use of duct board to create ducts for convention flows, generally not enough for amp cooling without some forced air flow -- one option to tap off the cooling duct in the wall soffits to put cool air into the speaker box, or best - properly move the amplifiers out and into the center or center cooled cabinet better yet.

if the manufacturer doesn't support removing the amp, then increase the cooling / duct size to ensure sufficient flow. small Genelec etc aren't necessarily an issue and units like ATC and PMC will support remote amping as would fully passive speaker units requiring external amps. however even passive speakers should have venting as speakers can and do generate heat in the course of usage.