John Sayers' Recording Studio Design Forum Archive

Hey guys,

First of all, I want to say thank you John for this amazing Forum. This is an island in the ocean of false information out there!

I am going to build a new Production Studio in Berlin. I am producing music, mostly advertising and movie scoring. So the main purpose is to build a studio for mostly midi-based productions, but I need to be able to record Instrumtents like acoustic guitar, saxophone, small percussions, vocals etc.
The most important is, I want a quite flat frequency response in my studio, as I have been struggling at mixing Kick and Bass in the past. I have been working in a very small room for years and i am so happy to have found a bigger place. I hope to get the room sound really good, so that I would be able to do engineering/mixing jobs too. In January I am going to start the constructions.

As the room is not only for mixing/producing but also a composing place, it's important for me, to keep it enjoyable spending a lot of time in there. So this is why I won't do a artificially ventilated decoupled room. fresh air and natural light is absolutely necessary. It would be too expensive too.
And unfortunately I won't do soffit mounted speakers as can't afford big main speakers yet.
(I am working on Adams A7X, 2" genelecs and a krk 10s Sub and I think to stick with that, as I am used to them)


The room is 441sqft. (27,5' x 16,5') and 10,3' high. It,s not a perfect ratio, I know, but I think I should be able to make this sound good.

The axial modes should be at 20Hz, 40Hz, 80Hz... in in the length axis, 35Hz, 70Hz, 140Hz... in the width and 50Hz, 100Hz... in the height, but I didn't start the measurements yet.

It's an old building (around the 1900's). The 4 walls are quite thick and made of brick. Ceiling (and Floor) are a wooden beam ceilings.
So I thought to lay an extra 1" insulation layer and OSB boards on top over the existing wood floor, in order to get a little more headroom without bothering the neighbours. I don't know how loud I can get, but in the worst case I have to work on a lower volume. This won't be a big problem.

You can see my current layout in the attached pdf file.

I decided to treat the front wall with seven membrane absorbers (39"x39"x19,5"), tuned to 40Hz, 60Hz, 80Hz and 100Hz, combined with 20" deep 5000 Pa·s/mª glass wool panels and ~30" deep super chunks or Hangers in the front corners.

I will build Sidewalls with an angle of 18° (drywall, gypsum boards) which will contain two Helmholtz Resonators tuned to ~140Hz behind a 4" Layer of Basotect (where the early reflection will be) and two further Membrane absorbers tuned to ~70Hz on top of that. And of course openings to the super chunk.

The listening position is placed between 33% and 38% of the room lenght. Using the wall bounce calculator I figured out, that the speakers should be best at around 60" from the front wall.

For the ceiling I planned to put hangers at the space towards the frontwall + towards the rear wall and 8" porous absorber panels where the early reflections are.

The backside of the room will be treated with super chunks in the corners, a 20" glas wool panel in the middle of the rear wall, a super chunk at the floor/rear wall corner, two couches, some bookshelfs and "living roomy" furniture etc. Maybe some diffusor/absorber panels, if necessary.


Now, here my questions:


1. Does the Placement of the Membrane Absorbers and Helmholtz Resonators make sense?

2. Will I have enough Bass absorption in my Layout to reduce the Bass response in my room effectively? Or will I need more Basstraps?

3. Concerning the openings at the sidewalls for the super chunks. Can I use the same Formula as designing Helmholtz Resonators? (My plan is to drill holes of a certain diameter in the drywalls.) How large should the perforated area be? Is there a minimum size? I thought to make the perforated area about 8" wide over the whole room height. Is this wide enough? Will the super chunks have a higher absorption when i make the perforated area larger?

4. What would be the acoustical difference between Hangers and super chunks? (absorption coinficient and absorption frequency) Are Hangers absorbing a broad band of frequencies? How important are the distances between the hangers? Do I can mess up the tuning?

5. Concerning the Helmholtz Calculator post in the acoustic forum. Unfortunately I can't open the file, is it because I am using a Mac? I was using the Allard & Champoux Model in the past. Are they correct?

I did my best to look in previous posts to answer my questions. If nevertheless some of my questions are already answered, please just ignore them and I will try to find previous posts.


Thank you very much in advance!!!
Every type of Comment is more than welcome!

PS: I have to wait for some payments, but as soon as my clients will pay me, I will donate. This forum helped me so much. This shouldn't be for free!

Hi there "Cozy_Coze_Berlin", and welcome to the forum!

... this amazing Forum. This is an island in the ocean of false information out there!

We think so too!

The most important is, I want a quite flat frequency response in my studio,

You mean flat like this? : http://www.johnlsayers.com/phpBB2/viewt ... =2&t=20471 That's about as flat as you can possibly get.

I hope to get the room sound really good,

You seem to be contradicting yourself: you say you want it "flat", but you say you want it to "sound good". You cannot have both. If it sounds flat and neutral, then it sounds like nothing. It has no sound of its own, and does not sound good: it sounds true, and honest, and accurate, and precise, and surgical, and correct but it does not sound good. Some musicians (and even some engineers!) do not like flat rooms, since they have no character. They don't sound like anything, neither good nor bad. They just sound.. flat! So either you room can sound flat, or it can sound good, but it can't be both. That's like saying: "I want my room to be painted perfectly white all over, but I also want it to be colored orange and yellow, with some red shades as well". It is not possible. OR like saying: "My room must not have any smell or odor at all, but I also want it to smell like lavender and roses and a pine forest after the rain".... Either it smells, or it does not. Either it is white, or it is not. Either it is flat, or it is not.

In January I am going to start the constructions.

It's half way through November already, you want a room with perfectly flat acoustics, you have not even started the design process, and you want to build in January? Sorry, but I don't see that happening. That's not realistic. If you start with the design process, you might be ready to build by July. That's realistic. And that assumes you already have the acoustic knowledge and construction knowledge and studio design knowledge that you will need to do it right.

So this is why I won't do a artificially ventilated decoupled room.

So you do NOT want any isolation at all? You are fine with having all the outside noises interfering with your work as you carefully try to balance the mix? And you are fine with disturbing your neighbors as you track sax and percussion, which are rather loud?

fresh air and natural light is absolutely necessary

Correct. But that does not mean that you cannot build it properly, with decoupled walls and good ventilation.

You say you want to track acoustic instruments, but many acoustic instruments are sensitive to humidity and temperature. If you reject controlling the temperature and humidity in your room, the tone and timbre of the instruments will change throughout the session as the air conditions change,
and your musicians will constantly need to re-tune their instruments as well. So the last take will not sound anything like the first take. Are you OK with that? You say you want to produce and engineer high quality: How are you going to do that, if you do not control the conditions in the room? How will you be able to keep the response flat, if you do not control the temperature and humidity? Are you aware that acoustic response of the room itself changes as temperature and humidity change?

And unfortunately I won't do soffit mounted speakers as can't afford big main speakers yet.

Two comments on that:

1) How do you plan to get perfectly flat response if you don't soffit-mount your speakers? If your speakers are not soffited, they will be producing all of the usual artifacts that cannot be dealt with easily in any other way: You will have the SBIR issue, comb filtering, phase cancellations, power imbalance, uneven bass response, edge diffraction, reflections from the front wall, and all the other artifacts that are eliminated by soffit mounting. So how do you plan to get rid of those and make the room response flat, if you refuse to soffit mount your speakers?

2) Why do you think you need "big main speakers" in order to soffit mount? The vast majority of speakers can be soffit-mounted, and reap huge benefits. Even modest, inexpensive studio monitors can get great improvements from being soffited. It's a myth that you must have "really big, powerful, impressive" speakers in your soffits.

I am working on Adams A7X

Why do you NOT want to soffit mount those? They sound MUCH better in soffits:
Is there something wrong with any of those setups?


All of those rooms perform very well. Much better than they would have if the A7X's were just placed on stands, not soffit mounted. There are so many benefits to doing it. Soffit-mounting your speakers is the one single best thing you can do for the room.

The room is 441sqft. (27,5' x 16,5') and 10,3' high. It,s not a perfect ratio, I know,

The ratio is 1 : 1.6 : 2.67, which is outside the Bolt area, but it does pass all three critical tests, and it has reasonably decent low end support. It's large enough that it can still be reasonably decent, but it might not be easy to make it flat.

The axial modes should be at 20Hz, 40Hz, 80Hz... in in the length axis, 35Hz, 70Hz, 140Hz... in the width and 50Hz, 100Hz... in the height

Ummmm... Nope! your first three axial lengthwise are 20.6, 41.2, and 61.8. Your first three width-wise are 34.4, 68.8, and 103.2. And your first three in height are 55.1, 110.2 and 165.3. But assuming correct treatment, the width axials above 0,2,0 and the height axials above 0,0,1 are not very important, because they'll be above the Schroeder frequency for that room.

So I thought to lay an extra 1" insulation layer and OSB boards on top over the existing wood floor, in order to get a little more headroom without bothering the neighbours.

Why would you want to do that? That would likely make the isolation worse, not better: Have you checked what the resonant frequency of that would be? Are you aware of the perils of trying to float a light-weight floor? http://www.johnlsayers.com/phpBB2/viewt ... f=2&t=8173

I don't know how loud I can get,

That should be your very first priority! IT would be sad to invest so much time, effort, materials and money in your studio, only to have the police come shut you down because you exceeded the legal limits. A sax can easily hit 95 dBC, and 100 isn't out of the question. What is the legal limit where your studio is?

You can see my current layout in the attached pdf file.

For a room that size, I still would put the speakers up against the front wall, for several reasons: One is to get the front wall SBIR frequency up high enough that it can be treated. Another is to meet the BS.1116-3 spec for speaker placement. Yet another is to reduce the mid-range comb-filtering and distortion caused by reflections from the desk surface.

I decided to treat the front wall with seven membrane absorbers (39"x39"x19,5"),

?? WHY??? For what purpose? Did you do the math? Are they large enough to have any measurable effect on the decay times for those frequencies? Why are you treating 100 Hz and 140 Hz when they are above the Schroeder frequency for that room?

I decided to treat the front wall with seven membrane absorbers (39"x39"x19,5"), tuned to 40Hz, 60Hz, 80Hz and 100Hz, combined with 20" deep 5000 Pa·s/mª glass wool panels and ~30" deep super chunks or Hangers in the front corners

I don't understand why you would do that: If you are already treating the low end with both thick porous absorption and also superchunks (or hangers), then why do you also have tuned traps? Are you aware that membrane traps (especially low frequency ones) can actually extend the decay times for the frequencies they are tuned to, due to inertia?

I will build Sidewalls with an angle of 18°

Why did you choose 18°?

which will contain two Helmholtz Resonators tuned to ~140Hz behind a 4" Layer of Basotect (where the early reflection will be)

Once again: Why? And how do you plan to tune Helmholtz resonators to 140 Hz? What neck width, cavity depth, and mouth correction factor did you use when you calculated the dimensions? And why would you do that at all, when 140 Hz is far above the Schroeder frequency for the room?

Using the wall bounce calculator I figured out, that the speakers should be best at around 60" from the front wall.

Why do you want to create a front-wall SBIR dip at around 57 Hz? That's right in the middle of the bass guitar range. The general recommendation is to NEVER have speakers at a distance from the front wall of between 18" and about 8 feet, as that places the first SBIR dip in the low end of the spectrum, where it is practically impossible to treat. Either put the speakers up against the front wall, so that the frequency will be high enough in the mid range that it can be treated, or put the far enough away that the dip is lower than the bottom end of the musical spectrum.

For the ceiling I planned to put hangers .... + .... 8" porous absorber panels The backside of the room ... super chunks ... 20" glas wool panel ... a super chunk ... two couches

You are over-absorbing the room. While it's a good idea to have large, thick, deep absorption, that also sucks out the high end, and your decay times will be way too short, and uneven across the spectrum. You will need to cover some (most) of that absorption with reflective or partially reflective materials, carefully tuned to keep the high-end live, and balance the decay times between frequency bands so they all meet BS.1116-3.

1. Does the Placement of the Membrane Absorbers and Helmholtz Resonators make sense?

No. They don't make sense at all.

2. Will I have enough Bass absorption in my Layout to reduce the Bass response in my room effectively? Or will I need more Basstraps?

It's probably about right for that room, and indeed it is hard to have too much bass tapping! But there's always the challenge of also preventing the bass traps from killing the mids and highs. You need to balance the response of the room. Having perfectly flat frequency response is nice, but not all that important. What's far more important is to have flat time-domain response, and to not only have a good ITDG, but also to have a good diffuse-field decay that matches the volume of the room. For a room that size, I'd probably shoot for a decay time of around 300 ms or so, across the entire spectrum.

3. Concerning the openings at the sidewalls for the super chunks. Can I use the same Formula as designing Helmholtz Resonators?

I'm not sure what you mean. You can't combine a superchunk with a Helmholtz resonator. They are two very different devices.

(My plan is to drill holes of a certain diameter in the drywalls.)

Drywall is not a good material for trying to build tuned traps! It is brittle and crumbly. You would not be able to get the accurate, clean holes that a Helmholtz resonator needs. It is also rather thin, so tuning it low is going to be a problem....

How large should the perforated area be?

How much absorption do you need at the tuned frequency? You can't know one without knowing the other. You would first have to calculate what the untreated decay time would be, then calculate what the target decay time should be, the calculate how much absorption you need at that frequency, then you can determine how much total area of "holes" you need in order to proved that amount of absorption at that frequency. And then you have to figure out how you will do the final tuning: since you want to hit room modes, which are very tight, very high Q problems, you will need to tune your devices very accurately to hit the exact center frequency of the mode, and since there's so much variability in building them, you need a method of tuning them after they are built, and a method for measuring the tuning, to make sure that it really does hit the mode that needs treating. There's no point in build a Helmholtz resonator tuned to 40 Hz if the actual mode is at 41.9 Hz: that would be nearly 5% off! Way too far to be any use.

Is there a minimum size?

The minimum size that you need is the exact size predicted by the analysis I outlined above. You can use more than that, sure, but not less. For low frequencies like that, and considering that they are modes, I would estiamte that you will need several hundred square feet of eara for treating each mode.

Will the super chunks have a higher absorption when i make the perforated area larger?

Once again, you cannot combine a superchunk with a highly tuned Helmholtz resonator. I don't even know why you would want to do that. And no, the superchunks will not have greater absorption if you place a perforated panel in front of them. They will have less absorption overall, since the perf panel will be reflecting most of the frequencies that they would have absorbed otherwise.

4. What would be the acoustical difference between Hangers and super chunks?

They are both bass traps. They work on different principles. A hanger trap works basically as a "wave-guide into oblivion", while superchunk is pure porous absorption.

Are Hangers absorbing a broad band of frequencies?

Yes, but with slightly higher peaks at the frequencies related to the wave-guide tuning and efficiency.

How important are the distances between the hangers? Do I can mess up the tuning?

It's hard to mess up a hanger trap! The basic rule is that the bigger you make it, the lower it goes, and the best way to figure spacing is by starting at the center of the room, then continuously sub-dividing and placing a hanger at each half-way point (=half wave, quarter wave, 1/8 wave, etc) until you can't fit in any ore without them touching. Keep them far enough apart that the don't touch, and can swing freely just a bit.

5. Concerning the Helmholtz Calculator post in the acoustic forum. Unfortunately I can't open the file, is it because I am using a Mac? I was using the Allard & Champoux Model in the past. Are they correct?

Be careful! There's an incorrect equation circulating around the world, and it has even been used in some text books. It confuses the "+" sign for the "x" sign! So do make sure you are using the right equation.


- Stuart -

Hi Stuart,

Thank you very much for your detailed reply! Seems like there's is lot of to rethink.
Well, I know that I'm not going to get the room perfectly flat, regarding my budget and my experience. I just want to be able to have a more correct judgement on the low-end, as I had in the past.
I am going to take some time to rethink all of this and I'll post the new Layout.
I am glad to hear that it actually would make sense to soffit mount my Adams. I definitely will think about this!
And Thanks for the Post regarding floated floors.. This actually prevented me of spending money for something which probably would make everything worse!

So do you think I shouldn't work at all with membrane absorbers? What other options do I have? As I know porous absorption doesn't do much under 100Hz... Are hangers absorbing down to like 40Hz?

And no, I wasn't aware, that membrane absorber could extend the reverberation time due to inertia. Where can I get more information about this? Which materials would be bad / good for the membrane?
(Tomorrow I will see if the way I build it actually workout well. I used a 5mm MDF board)

Thank you so much for your time, Stuart! It's a huge gift, getting the thoughts of a professional!
best regards!

Cosimo

Seems like there's is lot of to rethink.

True! There always is. When I'm designing a studio, I seem to spend quite a bit of time re-thinking things, then re-thinking again, to optimize... and also to compromise.

I know that I'm not going to get the room perfectly flat, regarding my budget and my experience.

It's a decently sized room, so it can get reasonably close to flat. I'd guess you should be able to get within +/-6 to 8 dB across most of the spectrum, and decay times within about +/- 80 to 100 ms of each other across the board, between most frequency bands. That's pretty good, for a home studio / project studio. Even some pro studios don't get better than that.

Of course, it would take a lot of work and careful design to get there, plus final tuning, but it's feasible. It all depends on ho far you want to take it, and how good you want your room to be.

I just want to be able to have a more correct judgement on the low-end,

Ahh! The low end. Yup. That's always the biggest problem, in any room. And that's one thing the soffits will do for you, if you do decide to go that way. Soffiting your speakers extends and tightens your low end. And good bass trapping smooths out the decay times. And good room design reduces SBIR from the rear wall. Then there's a couple of other secrets that I have for doing the final tweak, if you want to go that route...

I am glad to hear that it actually would make sense to soffit mount my Adams

Definitely! No doubt about it. Here's an interesting thread from a forum member who started out with his speakers on stands, and then rebuilt his room with soffit mounting: http://www.johnlsayers.com/phpBB2/viewt ... =1&t=20895

And Thanks for the Post regarding floated floors.. This actually prevented me of spending money for something which probably would make everything worse!

There are alternatives, and what you mentioned could be adapted to do a "semi-decoupled damped riser floor", but it probably won't give you as much improvement as you are hoping for if you don't isolate the rest of the room. If you have read many threads where I have commented on isolation, you might have seen my "fish-tank" analogy before: Imagine a guy who wants to build an aquarium in his living room, and buys the metal frame to make it. Then he thinks:" "Well, I only need to see the fish from the front, so I'll only buy one piece of glass for that side"... How well do you think his aquarium will hold water? That's roughly what happens if you try to isolate just one side of a studio. Just like the aquarium guy would have water splashing around all over and flooding the whole place, despite the glass on one side, so too a room will leak sound all over, despite the isolation on one side. In other words, isoalting only your floor is like building a fish tank with glass on only one side...

The thing is, sound is a lot like water in many aspects: once the water is out, it flows wherever it wants, in all direction, including right in front of the glass... Ditto with sound: once it is out, the it is out everywhere, and will get around all over... including on the other side or the "isolated" wall, floor or ceiling. So even though your intentions are good in wanting to prevent sound from annoying your neighbors, isolating your floor will not accomplish that... unless you also isolate the rest of the room. Just like the guy with the aquarium, who really needs glass on all sides if he wants to keep the water in, a studio needs isolation on all sides if you want to keep the sound in.

So do you think I shouldn't work at all with membrane absorbers?

I have used them very occasionally, for very specific issues, but I'm not a big fan. They do work, if you tune them correctly, but tuning them is nowhere near as easy as the textbooks make it sound. You need a method for accurately testing the tuning of your finished device, to find out what it's REAL resonant frequency is (not the predicted one...), and a method for adjusting that to get it where it is supposed to be, because it very probably won't be what you thought it would be.

Another thing the text books also don't tell you, is that membrane traps need to be very large if you want to get a big difference in your acoustic response. Think of it this way: assume you have a problematic axial mode forming between the side walls, and they measure 27 long by 10 feet high. So each side wall has a surface area of 270 square feet. Then you build a membrane trap with a full sheet of plywood (4' x 8'), but you are only affecting 11% of the wall area, so at best you are treating 11% of the problem. If you stacked 5 such devices along each wall, you'd still only be covering 59% of the total wall area, so there would be over 40% of the problem still untreated. But 5 such devices would essentially take up the entire wall, in practical terms (especially if you have doors, windows, and other treatment on that wall, plus a cloud, hanging lights, artwork, gear, etc.), so how would you treat the other two or three modes that are also associated with that wall? There's no space left for that. So even by filling most of the usable area of a wall with membrane traps, you only affect half of one problem, and can't do anything at all about other problems.... That's one of the reasons why I'm not a big fan of tuned devices in general: they take up a lot of space, for little result.

What other options do I have?

Porous absorption...

As I know porous absorption doesn't do much under 100Hz...

That's one of those myths that continues to float around the internet, and get repeatedly chanted religiously, as though it were written in stone by the finger of God himself! People often quote the "half-wave rule" or the "quarter-wave rule", spouting about absorber only being able to affect waves where the thickness of the absorber is at least one quarter wavelength, or more. They say that if you have a 50 Hz tone, the wavelength is 22 feet, so you need an absorber (22/4=) 5' 5" thick.

Garbage.

In reality, a porous absorber can have a very decent effect on a wavelength when the thickness of the absorber is just 3.5% of the wavelength. So that 50 Hz wave can be dealt with by a porous absorber just (22*.035=) 9 1/4" thick. That's reality, and easily shown from simple math and empirical results. Yet, people still bandy around the "It must be quarter wave thick or it wont work" myth.

Not only that, but we are talking about normally-incident sound so far: sound that strikes the absorber head-on, at an angle of 90°. Sound coming into your 9" absorber at that angle will go through 9" of absorber (duh!), but if sound strikes at any other angle, it must go through MORE absorber. Think about it. If the sound wave strikes at 30 instead of 90°, it "sees" the 9" absorber as though it were 18" thick. And since most of the sound waves in your room are NOT arriving normally incident, and ARE arriving at random angles less than 90°, it's safe to assume that your absorbers seem a lot thicker than they really are to most of the sound waves.

The "quarter wave" rule is not wrong: it's just not applicable here. It is NOT talking about thick porous absorbers close to the wall. It was actually meant originally to describe thin absorbers at a large distance from a surface, and it refers to the gap, not the thickness. More specifically, it is useful for describing the absorption of drop-ceiling tiles, which are acoustically very thin (just a few mm) compared to the wavelengths passing through them, but they are spaced a long distance from the ceiling behind them, and there is a measurable effect due to that spacing, not due to the thickness. Drop ceiling tiles do have a small effect on waves that are 4 times longer then the gap size, hence the "quarter wave rule". That's because the tile sits at the point where the particle velocity is highest: the quarter wave distance. But the "rule" has nothing at all to do with the thickness of the tile: it's about the spacing, not the thickness, and refers to acoustically thin absorbers, not thick ones.

Somehow, that got all muddled up into a new mangled, mungled munged "rule" that supposedly requires absorbers to be one quarter wave thick. Not true! You get very good absorption at 7% of the wavelength, and still usable absorption at 3.5% of the wavelength. (Now, if you also move it away from the wall, to get a gap behind it, then you get the best of both worlds...)

Oh, and there's one other thing that the myth-mongers don't take into account: when sound waves move through air, they are attenuated slightly by the air itself: they lose energy. This happens any place where the air pressure increases, such as when a sound wave passes by. Where does that energy go? Converted to low grade heat, which stays in the air, exactly where it was left behind by the passing of the wave. That's called "isothermal" heat dissipation. But that's NOT what happens when a wave loses energy as it moves through a porous absorber! The absorber itself changes the way the air behaves, to "adiabatic". The fibers of the absorber can carry away some of the heat from the places where the temperature has increased due to sound energy being converted to heat. So the heat doesn't stay behind in the air at the point where the pressure hanged as the wave passed: rather, the heat is removed. This is a whole different ball game. To cut a long story short (and a long equation too): Due to adiabatic heat transfer, a porous absorber appears to be about 20% thicker than it actually is. Sounds waves "see" more thickness than is really there, by a factor of roughly 1.2.

There's more to it than that, but even from this simple explanation, you can see that porous absorbers do, in fact, work down to very low frequencies. They still have to be thick to be effective, but not twenty feet thick like the myth-mongers would have you believe!

Here's the real situation, in graphic forum:

This is what you get for a 30cm thick (12") porous absorber right up against a wall, for randomly incident sound:
Not too shabby at all! According to the "quarter wave or die!" folks, that device should have no absorption at all below about 280 Hz (aprox. quarter wave for 30cm), yet the graph says something very different: it shows that at two entire octaves below that (70 Hz.), you STILL have a coefficient of 0.7, and even way down at 40 Hz (nearly three and a half octaves below where they say it doesn't work) it is still giving you better than 0.5.

So lets go a little thicker, and make the absorber 16" thick, and this is what you get:
The entire darn spectrum is covered! Even down at 20 Hz, you are still getting better than 0.4, and everything above about 38 Hz (the bottom end of the bass guitar) is over 0.6 coefficient, which is considered to be very usable.

For both of those graphs, I assumed typical porous absorption with a gas flow resistivity of 9,000 MKS rayls. If you went hunting for lighter stuff that has lower GFR, you could probably improve on that even more.

(Now, that's for an absorber up against the wall: move it away a bit, and things get even better...)

So, consider that a typical superchunk is 36" along the faces, and therefore about 25" thick in the middle of the diagonal, and you can appreciate why they do such a good job as bass traps. This is what a 25" thick superchunk porous absorber would get you:
Well gosh darn my goodness gracious golly me!! It sure looks to me like that would get you a "very usable" 0.6 absorption coefficient for everything above about 33 Hz!.

So much for the "it has to be a quarter wave thick" guys, and the "porous absorption doesn't work below 100 Hz" guys....



Hangers will give you similar performance. I sometimes do superchunks in the rear corners of the room, with hangers between them, across the rest of the rear wall.

However, the danger (as I mentioned yesterday, and as you can see in the graphs), is that even though these thick things are retty good for bass, they are also incredibly good at sucking up everything else too! Even stuff that you did NOT want sucked up. For everything over 1 kHz, these guys have an absorption coefficient of practically 1. Allowing for edge diffraction, it can actually be greater than 1. But most rooms do NOT need absorption in the high end, and even the rooms that do need it, do not ever need as much as you see here. So you need to design your bass traps such that they ONLY trap bass, and reflect back most of the mids and highs.

Are hangers absorbing down to like 40Hz?

Yup. And lower. Just like porous absorbers.

And no, I wasn't aware, that membrane absorber could extend the reverberation time due to inertia.

It's logical when you think about it: A membrane trap has a membrane on it (duh!), and the membrane has mass. It's the mass that determines the resonant frequency: the lower you want to go, the more mass you need. But mass implies inertia: If you have ever tried to push a car by hand, you know that you need to push really hard for quite a while to get the car moving fast enough... because the car has mass. And once it is rolling pretty fast, when you stop pushing it will actually keep on going for quite a long time before coasting to a stop... because it has inertial mass. The exact same thing happens with your membrane trap: The incoming sound wave causes the panel to resonate, yes, but it takes a little while to "get up to speed", because the panel has inertia. So it does not react immediately to the wave you are trying to damp. It takes a few cycles of the wave before the panel can get up to speed, so that it is in sync with the wave and converting some of the acoustic energy to heat energy, in the porous absorption inside it, thus removing energy from the wave. But when the wave stops, the panel doesn't! It has inertia, and just like the car, it continues to "coast" for a while, as that excess energy is absorbed by the porous absorber inside. So even after the wave is no longer exciting it, the panel still carries on for a few cycles, until it eventually stops. Because it has mass, and therefore inertia. The lower the wave it is targeting, the more mass it needs, so the longer it carries on, because it has more inertia.

Now, you might not think that this is such a big deal, but do the math. Let's say that you have a membrane trap tuned to 40 Hz, and that it's a really good one that only continues for 3 cycles after the exciting tone stops. At 40 Hz., three cycles of 40 Hz is 0.075 seconds (do the math). So the device will still be "coasting" for another 75 milliseconds after the triggering wave has gone away. And since a membrane trap is basically just a giant speaker in reverse that absorbs sounds, it also works the other way: If you make it vibrate, it will produce sound. If it is vibrating at 40 Hz for 75 milliseconds when there is no sound in the room, then it is acting like a speaker and creating a 40Hz tone that was not there!! In other words, the vibrating panel has extended the note by an additional 75 ms. So instead of bringing down your decay time at 40 Hz, it can potentially increase the decay time. Yes, it did absorb some energy when the original tone was present, but it only did so starting 75ms after the problematic wave started, and then it carried on "giving back" some extra sound for another 75 ms after the problematic wave went away.

In other words, it extended the decay time, instead of reducing it, which is not what you wanted.

Now, a membrane trap that can stop in just 3 cycles is pretty good. A large, massive membrane with lots of inertia might carry on running for 6 or 7 cycles... at 40 Hz, that would be nearly 200 ms.... which is a LOT! (and audible.......)

Here's a simple exercise to see what I mean. Go get yourself a ruler. I don't know if you ever did this in school to annoy the teacher, but... :
The ruler can continue vibrating for several SECONDS like that, because it is not damped at all: it just resonates at it's natural frequency, just like the membrane panel would, if it was not damped. Now tape a piece of cotton wool to the end of the ruler, and try again: It no longer carries on for several seconds, but it DOES still carry on! It still goes for a few cycles before stopping. Just like a membrane trap does.

So, if you do want to build membrane traps, do design it so that it has as little mass as possible, so that it won't carry on playing tones that you didn't want, after they already stopped.... But in order to have a low mass panel tuned to a low frequency, you need a large cavity depth... Ooops!

Even if you use a limp mass, such as MLV, you still have this problem of the inertial issue, because it's not just the panel that vibrates: the entire thing is a tuned system, and the air inside is vibrating along with the panel...

This is another reason why I'm not a big fan of membrane traps. And the simplified text books don't tell yo about this, but it's very logical and obvious once you think about it...

On the other hand, porous absorbers do not suffer from this problem: they are very light mass, especially when you consider that it is the individual tiny fibers of the absorber that are doing the work, and each little fiber has practically no mass at all! They act individual (each fiber vibrates all by itself, in response to the wave hitting it), and since each fiber has such low mass, it can start and stop instantly....

Which materials would be bad / good for the membrane?

Low mass, and limp (self-damping). I normally use MLV. But that implies a deeeeeep cavity... probably just as deep as it would be for a superchunk... in which case a superchunk would be better, because it absorbs ALL low frequencies, not just the one you tuned your MLV for...

Tomorrow I will see if the way I build it actually workout well. I used a 5mm MDF board

MDF has a density of roughly 750 kg/m3. Fiberglass insulation has a density of roughly 25 kg/m3. Which one has more inertia? a 5mm thick MDF panel has the same surface density as a piece of fiberglass insulation 150mm thick (6", roughly). But once again, the MDF panel acts all at once, as one single solid piece of mass. The insulation does not: it acts as a whole bunch of very tiny, very VERY low mass fibers, each doing it's own thing....

Most of the rooms I design just use porous absorption as the main ingredient. Some of it (or a lot of it) is covered with some type of foil, or wood slats, something like that, which serves several purposes, but mainly to keep the highs in the room and let the lows through to be absorbed. Regarding diffusion, I only use diffusion in very large rooms, and I seldom used membrane traps, unless there's a really stubborn problem that refuses to react the way I want to other treatment. In Studio Three Productions (the link I gave yesterday) there's a HUGE amount of porous absorption in there, as bass traps. Mostly in the ceiling, above the cloud and above the soffits, but also in horizontal superchunks that run across the tops of the walls, just under the false ceiling, and vertical superchunks hidden in the rear corners. There's no hangers at the back of the room (due to the window), but there are hangers under the speakers inside the soffits, and inside the center area between the speakers. That room actually does have a small MLV membrane trap (not visible in the photos), but it turned out to not be as effective as I wanted. There's also other treatment in there, that I'd prefer not to go into, but the vast majority of the treatment in there is plain old porous absorption. Mountains of it. The owner of the studio (Rod) could not get over the truck loads of insulation that went into that place. He thought I was a bit nuts when I kept on telling him to load up more rolls of fiberglass insulation... until we did the initial tests, and he heard (and saw on the graphs) just how well the lows are controlled. Then he was convinced that it was all worth it. That's a big part of the "secret" of how I got the low end of that room so tight, and flat, and smooth: Stacks and stacks of porous absorption. Your room is similar to his in overall volume, but not the same shape or dimensions. So you could theoretical get similar results, if you wanted to go into the same very detailed design and tuning. It's possible. But I would do three things, principally: 1) Soffit mount your speakers. 2) Load in mostly porous absorption, especially on the rear wall, and 3) tune the absorption with foils, slats, and other tricks, to balance the frequency response and time-domain response, such that it gets close to ITU BS.1116-3 specs. That would be my general plan. But of course, there's a lot more to it than that!

Thank you so much for your time, Stuart! It's a huge gift,

!


- Stuart -

Ok, I took everything you said and spend the nights creating a new layout.
I am really happy because after everything you told me, I will get much better results, for less effort and on top of this it even looks way better

So, in order to get rid of the SBIR i will soffit mount the speakers, put a deep (~20") porous absorber on rear wall and 8"or 10" porous absorber panels on the sidewalls.
-I know it would be way better to angle the sidewalls, instead of putting absorbers on them, but this would make me lose a lot more space in the room and would be more more expensive. Though on the wall bounce calculator it looks not bad at all

Behind the baffles of the soffit mount there will be space for 40" deep / 60" wide basstraps. (Think I'll go for Hangers as this will be less expensive at this size and it would provide more air circulation for cooling the speakers) + super chunks at the rear wall corners 36" x 36".
Combined with hangers on the frontside and the backside of the ceiling, this should already do a good job flattening the modes. If this isn't enough i will build additional horizontal chunks on the floor to wall corners.

So my plan is, when everything is build, I'll try to figure out how much and which of the porous absorbers should get a reflective front, so that the room doesn't sound too muffled. I think John got a nice sheet on how to make the reflective absorber panels. I'll just stick with that, so I won't mess it up

I really like the new layout, because it seem to be hard to mess it up.

Nevertheless I still have a few questions:

It seems that nearly 100% of the people are building wood frames for the soffit mount. Is there a problem, building typical drywalls, using metal sheet UW- / CW profiles and gypsum boards? (It would be less expensive and in my case easier to build)
(I though of putting two layers of gypsum board on the frame in order to get the mass and cover it with a nice looking wood finish.)

In most threads about soffit mounting speakers it's mentioned that the baffle and the speaker stands should be as rigid as possible, but it is also mentioned, that the speakers should be decoupled, so that there are no problems with structure-borne sound. Isn't that a contradiction?

The graph of the wall bounce calculator doesn't show me a +6DB Low frequency boost, which I thought should be always the case. Do you have any explanation for this? Can I trust the graph?

Sorry, my question about the soffit mount wasn't very clear and maybe a little bit stupid too.. I could have get more informations in this forum before asking questions

As I understand the issue, the speakers should be mounted rigidly on a as heavy as possible stand or an inner frame, which is decoupled from the ground and decoupled from the outer frame. The outer frame should be as heavy as possible and should be decoupled from the ground and of course decoupled from the speakers. Is this right?

Further I assume that between the speaker and the baffle shouldn't be an air space, so the speaker and the baffle need to be connected with a vibration-absorbing material?

The room behind the speakers has to be absorbing, I assume especialy absorbing for the resonant frequencies of the system?

Would be amazing, if you could tell me, if I my comprehension is right!



Btw. what do you think about those springs, this guy is using?
https://www.youtube.com/watch?v=oZ44hM7sRj8

Thank you very much!

Cosimo

Ok, I took everything you said and spend the nights creating a new layout.

It's looking quite a bit better, but there's still a few issues to fix. To start with, I would suggest NOT raising your speakers above the standard level and tilting them down. There's several reasons for that, but most of them are psycho-acoustic: the way we humans perceive sound, rather than the way it really is. Our ears and brains work together to interpret the sound arriving at our pinna (the fleshy thing that sitcks out the side of your head, that most people call "ear"!). There's a rather complex interaction that goes on there, as the pinna guides the sound waves into the ear hole, and the pinna actually sets ups some conditions that are critical for the ear/brain to be able to determine directionality and frequency response. That complexity depends absolutely on the sound waves arriving from straight ahead, not from above. If the sound comes at you from an angle of more than about 7° above the horizon, then your ears can no longer accurately determine the direction that the sound came from, and as a secondary effect, they also lie to you about the frequency spectrum: In simple terms, you hear tones a bit different from the way they really are, and you can't tell which way the sound came from, accurately.

The standard reference height for speakers is 1.2m above the floor. Most studios are designed to place the speakers at that height. Why? Because that's the average ear height of typical people when seated! The acoustic axis of the speaker should be at ear height, and should be pointing at your ear, or just a little outside of the tip of your pinna. Getting this geometry right is critical to having exceptional sound clarity and accuracy at the mix position.

There's another reason for not raising your speakers: reflections form the desk surface. If you raise your speakers then tilt them down, you greatly increase the amount of reflections from the desk surface to your ears, which results in comb-filtering artifacts, as well as a boost somewhere in the mid range, and general unevenness in the mid range frequency response. What I call "hash". The speakers in Studio Three are raised very slightly, and tilted down at an angle of roughly 4.3°. We had to do that, to get a clear line of sight over the dog box at the back of the console. But it did result in the slight mid-range "hash" that you can see on the graphs. It's one of the decisions I regret from that design, as it is the cause of the only unevenness on the otherwise perfect response of that room. It had to be done, and the angle is minor. The effects are also minor, not really audible at all, but still I would have really liked to avoid that, and have ruler-flat response. Mostly just for boasting purposes! It wouldn't make any difference to the extreme accuracy of that room, but it would have made me happier to see a flatter line!

But anyway, even though theory says you can go to 7° tilt, and some places even say 10° (without providing any justification...), personally I would never tilt them more than 5° down, and I much, MUCH prefer to keep them flat. No tilt.

All of that is apart from the problem of actually designing and building the structure at a downward angle, and preventing it from over-balancing, and doing it accurately! Here's a photo of the early stages of the soffit construction at Studio Three Productions, so you can see what you would be letting yourself in for, of you try raise and tilt your speakers:
It can be done, but it's not easy to calculate, not easy to build, and not easy to tune when built. I would not recommend it for beginners.

put a deep (~20") porous absorber on rear wall

Excellent! I'd suggest doing that with a combination of hangers and pure porous absorber.

and 8"or 10" porous absorber panels on the sidewalls

You don't need that much on the side walls. It's nice if you can afford the space, but you could get by with less. 6" will do the trick, and even then you probably don't need that over the entire wall surface.

-I know it would be way better to angle the sidewalls, instead of putting absorbers on them, but this would make me lose a lot more space in the room

It seems that you are trying to create an RFZ style design, but you can't do that with absorption. You will need at least partial "walls" between the soffits and the side walls, to control first reflections.

Behind the baffles of the soffit mount there will be space for 40" deep / 60" wide basstraps.

You seem to be confusing soffits and bass traps! The area directly behind the baffle, where the speaker is, is all about the speaker, and nothing about bass trapping. That's a "box around a box around a box", and it is all related to controlling the speaker, not related to treating the room. You can do bass traps above and below the soffit section, yes, but not inside it. Here's another photo of the Studio Three soffits under construction a few days after the above photo, with parts of some layers of the baffle in place for initial testing:
You can see hangers installed below the speaker area, and an empty space above the soffit: those are for bass trapping, but not the area between them, which is all about the speaker itself. The entire area above the soffit was later filled with truck loads of porous insulation, as bass trapping.

(Think I'll go for Hangers as this will be less expensive at this size and it would provide more air circulation for cooling the speakers)

Right! You can see the ventilation slots in that photo above towards the top of the baffle. There's a "chimney" make of wire mesh inside the soffit, that prevents the insulation from blocking the air flow path, and the hangers below also do not block air flow.

Combined with hangers on the frontside and the backside of the ceiling, this should already do a good job flattening the modes.

Don't get confused here! You can't flatten modes! You can damp them, yes, but you can't flatten them. They are a fact of life.

So my plan is, when everything is build, I'll try to figure out how much and which of the porous absorbers should get a reflective front, so that the room doesn't sound too muffled.

That's actually not such a good plan. Yes, you certainly will need to test the room multiple times during construction (here's how: http://www.johnlsayers.com/phpBB2/viewt ... =3&t=21122 ), but that's only to check if it really is performing they way it was predicted to perform, so you can tweak the treatment you already designed for it. Waiting until you already built it to test is is not a good idea. You can fix a lot of things in the design itself, then just check them and do slight mods based on the testing.

And once again, the concept here is not to make it so "the room doesn't sound too muffled", because that's basing the design on subjective impressions. The REAL concept is to do the math, and design it so that it has no sound at all! So that it meets ITU BS.1116-3. All of this can be calculated in advance, and designed in, then checked during in construction.

I think John got a nice sheet on how to make the reflective absorber panels.

I think you are talking about "slot walls", but there purpose is not to be "reflective absorber panels". Yes, they do absorb some frequencies, and the do reflect other frequencies, and yes they are very useful devices, but they are also TUNED devices, which implies that you first have to predict what problems the room will have, in order to know hoe to tune them....

It seems that nearly 100% of the people are building wood frames for the soffit mount. Is there a problem, building typical drywalls, using metal sheet UW- / CW profiles and gypsum boards?

You COULD use metal framing instead of wood, it you are really, really good at metal framing, and know how to accurately build compound angled metal framing that can support a heavy load.... but it's far easier to do it with wood. Which is why pretty much everyone does it with wood, as you noticed! And you could use a sheet of drywall as one of the layers that make up your baffle, but personally I prefer to use only wood in the baffle layers. Drywall is messy, brittle, and powdery for this type of work. If you don't like wood, you could do concrete. Of the three images I posted earlier of soffit-mounted A7X's, the first one (white color) is actually a concrete baffle. The owner there cast the baffles in concrete, about an inch thick, if I recall correctly. We got very good results from that one. Even the the room is very small, those concrete baffles really helped to smooth out the bottom end. It's an option, and proven.

In most threads about soffit mounting speakers it's mentioned that the baffle and the speaker stands should be as rigid as possible, but it is also mentioned, that the speakers should be decoupled, so that there are no problems with structure-borne sound. Isn't that a contradiction?

Not really. It just looks that way! There's actually two methods for doing this. Both of them require an extremely rigid, very heavy frame for the baffle, and extremely rigid, very heavy baffle. The difference is in how you mount the speaker behind that baffle. With John's preferred method, he also mounts the speaker very rigidly in a heavy box, so the entire thing is massively heavy, and very, very stiff. The other method, which I prefer, is to "float" the speaker on resilient mounts, inside an massive enclosure box. Both methods work: they achieve the same thing, which is to prevent the speaker from transferring vibrations into the soffit structure. I find it easier to do the calculations for the resilient mounts. John prefers absolutely solid mounts. You can go either way.

The graph of the wall bounce calculator doesn't show me a +6DB Low frequency boost, which I thought should be always the case. Do you have any explanation for this? Can I trust the graph?

Because the wall bounce calculator is not a soffit calculator! Two different things. The wall bounce calculator is showing you how the frequency response will be altered by the reflections form the walls. It is NOT showing you how soffit-mounting affects the frequency response which is an entirely different thing. You can use a baffle step calculator to predict that, then combine the two.

As I understand the issue, the speakers should be mounted rigidly on a as heavy as possible stand or an inner frame,

The enclosure box around the speaker can be mounted like that, with John's method. I use a different method, with resilient mounting.

which is decoupled from the ground and decoupled from the outer frame.

It would be difficult (but not impossible) to decouple the soffit and/or stand from the ground, but it isn't necessary. I decouple the speaker itself from the rest of the soffit.

The outer frame should be as heavy as possible and should be decoupled from the ground and of course decoupled from the speakers. Is this right?

Not necessary, no.

Further I assume that between the speaker and the baffle shouldn't be an air space, so the speaker and the baffle need to be connected with a vibration-absorbing material?

That's the way I do it, yes, but it's not the only way.

The room behind the speakers has to be absorbing, I assume especialy absorbing for the resonant frequencies of the system?

Correct. There will be multiple resonances going in inside there, so it needs to be well damped. I normally pack light weight insulation through the entire cavity, except for the ventilation air flow path.

Btw. what do you think about those springs, this guy is using?

Cute, but no! That framing he shows there is far to light-weight and flimsy. It needs to be heavier than that. And even though resilient mounting is my preferred method, I would not use metal ceiling framing isolation springs to do that! Those springs are NOT meant for that purpose, and they are NOT loading them anywhere near enough. You can't just throw any old resilient device in there and hope it will float! ALL resilient mounts MUST be loaded correctly to get the right resonant frequency. If you don't load it correctly, the resonant frequency will be too high, or it won't float at all if you overload it. There are equations that you must take into account, in order to tune the mount such that the resonant frequency is at least one octave below the lowest frequency that the speaker will be putting out. And there's no way that a ceiling isolation spring is going to work to isolate a speaker. Plus, you can clearly see that the "box" is bouncing around all over the place, and is not stable at all! That just is not going to work. I wish I had a dollar for every incorrect YouTube video on building studios... 'd be rich!

The only thing correct about the video, is two of the three principles he mentions: Mass, and isolation. Those aren't the only principles by any means! They are just two of several. But even though he mentions mass an isolation, he then shows a flimsy, light-weight frame with little mass or rigidity, and an isolation mount that won't actually isolate.

- Stuart -

Thanks a lot Stuart!

You are 100% percent right. As i thought more into the details, I realized how much more difficult it would be to raise the speakers and tilting them down.
I changed the layout, but there is still something that really bothers me. When I put my listening position at around 38% of the roomlenght, I would have a distance around 9' to 10' between me and the speakers. And as my room won't have absolutely perfect acoustics, it would be way better to get closer to the speakers. I just can't think of a satisfying solution. I would love to hear what you think. Should I place the listening position closer to the speakers? I though if I go for something around 1/6 of the room size it shouldn't be too bad..

This is what it would look like. The listening position is now 55" from the front wall. (The room length is 27,5') The good thing in this layout is, that there is space for HUGE basstraps around the soffits.

The listening position is now 55" from the front wall. (The room length is 27,5')

Ummmm... the math doesn't work out at all! 55" into a room that is 330" long, means your mix position is at barely 16% of the room length! Way too close to the front wall, and like will be a source of SBIR from the rear wall too. Your mix position should be more like 125" into the room...

I would have a distance around 9' to 10' between me and the speakers.

Right. What speakers are you using?

And as my room won't have absolutely perfect acoustics, it would be way better to get closer to the speakers.

Two questions: Why won't your room acoustics be good, and why do you think you need to be closer to your speakers? 9' isn't a bad distance for most speakers in most rooms. It would only be bad if the acoustics were terrible in the room... But with soffit mounts, good bass trapping, and reasonable level of treatment, I don't see why it would be a bad thing to have a 9' listening distance.

I just can't think of a satisfying solution.

What angles have you tried for your speakers? It's a large room, so there's plenty of opportunity for adjustment. I'm not understanding why you can't get a more reasonable distance into the room with a usable geometric layout. Studio Three (the example above) is 26'11" feet long, and the speakers-to-mix-position distance is about 8'9" .... Why do you think that is a bad situation?

- Stuart -

Hey Stuart,
Thanks a lot for your reply,

I don't know, why I thought so. Maybe because I always worked in a real nearfield environment and I just wasn't sure if it would workout this way.
So, if you say a speaker-listener-distance of 9' is totally alright, I will go with my previous design, where the listening position is at 38% of the roomlenght.
I am Using Adams A7X and I think I will go for a 60° design.

At the beginning of February I'll finaly get the keys and I can do a first measurement in the empty room.


So glad to have found this forum. Thank you!

So, if you say a speaker-listener-distance of 9' is totally alright,

Just so you feel a bit more comfortable with this, here's a page from the ITU specs for critical listening rooms, about the correct way to set up speakers, distances, angles, etc.:
As you can see, the "base width" (distance between the speakers) should ideally be between 2 and 3m (6'6" to 9'10"), and the distance from the speaker to your ears should be a minimum of "1 times" that distance, and up to as much as "2 to 1.7" times that distance: in other words, between 2m and 6m! That's between 6'6" and 19'8"!!! (Yes, you'd need a rather large room for a 19 foot listening distance... ).

That's also part of my "pet peeve" about so-called "near field" monitors: The "near field" is actually a BAD place to listen to speakers! Less than about 5 feet is no good at all.... yet you see all those photos of supposedly "pro" studios, with speakers sitting on the console meter bride or dog box, just two or three feet away from the engineer's ears.... Sigh! Don't get me started on "near field" monitors... (there is no such thing!)...

- Stuart -