John Sayers' Recording Studio Design Forum

Hi there, I'm new to the forum so I'll try to keep this short ans sweet. I'm a sound engineer / mixer with some experience of professional acoustic treatment, but I don't have direct experience in new builds.

I'm helping a client to get a new private studio literally off the ground - completely new building, 2 storeys high. We're not trying to build a commercial studio, but the guy would like to hire white rooms to local producers, hence the need to create as much separation between the rooms and a shared live area, which might also be used for media and live music purposes - don't ask!

Design drafted by architect features a main recording room, approx. 13m x 10m and reaching the ceiling approx. 5.5m high. Behind one of the narrow sides there's a corridor running the whole length opening onto 2 rooms approx. 4m x 5m separated by a stairwell, with another 2 similarly sized rooms on the first floor.

The idea is to use double skin concrete blockwork throughout: 140mm blocks laid on their side to achieve 215mm. with an air gap of approx. 100mm. so total wall thickness (no braces) would be approx. 530mm. for the double-height room.
The same principle, but with the blocks laid on their regular 140mm. base, applies for the double skin of the 4 white rooms - hence a wall thickness of 380mm. Second storey's rooms will rest their floor (likely a concrete slab poured over corrugated steel) over the outer skin of the rooms on the ground floor, so no room will share any wall with any other. Corridors and stairwell should also help with decoupling the spaces.

My question to the community here is this: assuming standard concrete 7N/mm2 blocks, what level of sound abatement should I expect from each double skin wall? I cannot find any information to calculate the effect of the air gap, while blockwork usually rates around 42-48dB between 250Hz-4KHz depending on rendering. Also, no information on improvement for cavity infill with - say - RW3 rockwool or other material (kiln-dried sand?)

I'm told that using rodded hollow blockwork with concrete poured into the cavities achieves values equal to a poured cast concrete wall, approx. 6dB better than standard blocks. Can anyone confirm that?

All the walls would be unbraced but sharing the foundation's concrete raft, which is standard construction practice (except the lack of bracing) but I'm not sure if there's any technique to further decouple each wall from the foundation without actually building separate foundations for each room - anything that doesn't involve massive springs to float the concrete floor obviously!

To complete the picture, I was planning to add floating lamellar floors on the first floor white rooms to reduce the amount of sound transmitted through the speaker plynths, but not to bother on the ground floor.

unfortunately I cannot disclose any drawings, but more than the layout of the place I'd be interested to hear about ways to calculate the sound reduction of dual skin concrete walls, the contribution of the air gap (which size is better) and if engineered blocks exceeding 7N should be used to get us an ideal rejection of 80dB between rooms in the range 250Hz-20KHz and maybe 50dB at 50Hz. Please bear in mind that the sound would have to travel through 2 sets of double skin blockwork separated by a 1.2m corridor.

Feel free to call me a madman if you must!

Marco Migliari

Hi Marco, and Welcome!

OK; so you're a madman if you must! But then again, if you were not a madman, you wouldn't be building a studio! I think that's a prerequisite. All of us here are quite bonkers.

Anyway, from your description it seems like you are planning to build a bomb-proof bunker, not just a studio. The isolation you'll get from that is pretty darn good, if done right.

Let me see if I can answer some of your questions, but in random order:

Yes and no: Yes, that's what it looks like on paper, but sound doesn't always behave the way you expect it to. To start with, each "double skin blockwork" wall is actually a tuned system, that will resonate at a specific frequency set by the mass on each of the two "leaves" and the size of the air gap in between. There are equations for calculating that frequency. The two leaves don't act individually very much, but rather they both act together, to create this tuned system. Technically, it is an MSM system, sometimes also called an MAM system. MAM = "Mass-Air-Mass". MSM = "Mass-Spring-Mass". (Air is a spring, as far as sound wave are concerned.)

So each wall is tuned to a resonant frequency, called "F0": It isolates terribly at that frequency, and passes sound wonderfully. But at all higher frequencies, it isolates. At 1.4 x F0 it starts to isolate. At 2 x F0 it isolates reasonably. At 3 x F0 it isolates well, and it just gets better and better the further you go up the spectrum. Obviously, the trick is to tune it such that F0 is as low as possible, and no more than 0.5 x the lowest frequency that you need to isolate.

So how do you "tune" a wall? You change the mass on each leaf, or the depth of the cavity, or both. More mass = lower frequency. More cavity depth = lower frequency. More insulation in the cavity = lower frequency.

In your case, you have PLENTY of mass on each leaf, and plenty of air gap, so don't sweat it. IN fact, I think the term I would use here is "overkill". You have way more mass than you need to get good isolation and a very low F0, and as long as you can maintain that in the build you'll be fine.

One possible issue is that you will have two such tuned MSM systems right next to each other, separated by the corridor: Theoretically, the two systems can interact with each other, each changing the tuning of the other, and acting together as a single 4-leaf MSMSMSM system. However, the masses and dimensions are so large as to simply not be an issue. No problem.

Your challenge here will be to supervise the builders like a hawk, to make sure that they do NOT tie the leaves together across the air gap inside each wall: not intentionally, and not accidentally either, by dropping mortar, broken brick chunks, or any other kind of debris in the gap. Anything at all that bridges between the two leaves can seriously degrade the isolation, as it destroys the MSM system. To give you an idea: your car shock absorber is an MSM system. Imagine how well it would work if you were to weld the two ends of the shock absorber together with a steel bar... that's how well your wall will work if the builders manage to create even a single bridge between the two leaves of your wall. So this is a critical issue, and you need to think very carefully about how to prevent that.


Close to one of Sepmeyer's ratios. That's fine. Nice large room, high ceilings: bodes well for a great room, if treated properly!

Perhaps... and perhaps not. They might be tuned systems themselves, if not designed right, and could interact with the tuned walls, in theory.

The walls will, theoretically, get you better than 70 dB of isolation (TL, not STC). However, that's theory, not reality, and you will have many other weak points in the system that will most likely prevent you from getting there. For example, getting doors, windows, HVAC and electrical systems to isolate to 70 dB is a tall order, so whatever you manage to get there is the limiting factor for the entire build. In other words, if your walls do in fact get you 70 dB of isolation, but your doors only get 60, your windows 50 and your HVAC 45, then the total isolation is 45: It is only as good as the weakest link. So I would be concentrating more on those things than on the walls.

There's another issue here: flanking limit through the building structure itself. You say that the rooms are going to share a common slab, common foundation, and common structure. So whatever the flanking limit is for that structure, well, that's going to be the isolation limit for the entire building.

It depends on the mass more than the rendering, since you seem to be talking about single-leaf walls, which are limited by mass law. The density of your blocks is probably somewhere in the region of 2,000 to 2,500 kg/m3, so at 215mm thick you are taking about a surface density of maybe 400 kg/m2 to 500 kg/m2. That should give you around 40 dB TL at 125 Hz, rising to 65 dB at 4kHz. But that's for single-leaf mass law. You are not building single leaf, but rather MSM, which is an entire different ball game.

The air gap is what makes it an MSM system. The air is the spring, and the insulation in the gap is the damper on the spring. The blocks on each side are the two masses. Together, they form the entire tuned system, and the whole is much greater than the sum of its parts. You get much better isolation than mass law alone predicts... but only above F0. Mass law predicts 6 dB extra isolation per doubling of frequency or mass, MSM law predicts 18 dB per doubling. Guess which one you want?

Porous absorption, such as mineral wool or fiberglass, in the cavity between the two leaves, will gain you about 10 dB of isolation, and also lowers F0. It can gain you as much as 16 dB, theoretically, if you fill the entire cavity, but that's theory. 10 dB is more realistic. And 10 dB is huge, so it is well worth doing, for a normal wall. However, yours is so massive already that the difference will not be quite so spectacular. But still worth doing, as it damps all the nasty resonances going on inside the cavity.

But not sand! Sand will turn your beautiful double-leaf MSM wall back into a single-leaf mass-law wall. Not a good idea. It adds mass, sure, but also couples. You could use sand in the hollows of the blocks themselves, but not in the wall cavity

Only for single-leaf mass-law governed walls, and you'll only get 6 dB improvement if you double the total mass of the wall. For MSM walls, it adds mass to each leaf (obviously), which re-tunes the wall to an even lower F0, an that's good. It improves isolation even better than 6 dB per mass doubling. How much depends on other factors, but it will always be better than mass law.

So there is your flanking limit. It doesn't matter how great the rest of the build is, the isolation limit will be set by flanking through the slab.

Not for each room, but rather for each leaf. And yes, that is one common way of getting very high levels of isolation: the outer leaf sits on one foundation, and the inner leaf on another.

Actually, in your case you would probably want to float the entire room, not just the floor, since you seem to be aiming for really high levels of isolation.

That would be a possibility, but floating the floors is not an easy thing to do right. This thread might help you understand the issues a bit better:

viewtopic.php?f=2&t=8173

And if you want more technical details on various ways of doing it right, this is a good paper:

http://www.nrc-cnrc.gc.ca/obj/irc/doc/p ... /ir802.pdf

The equation is simple:

F0=c[(m1 + m2)]^.5 / [(m1 x m2 x d)]^.5

Where:
C is a constant: 43 for imperial, 1897 for metric
m1, m2 = surface density of leaf #1, leaf #2 respectively
d = depth of the air gap.

(The constant assumes insulation in the gap: if you don't have insulation then it changes to 60 for metric, 2647 for imperial).

That gives you the resonant frequency of the wall, at which it does NOT isolate at all. It starts isolating at 1.4 x F0, and the isolation increases at a rate of 18 dB per octave, except at the coincidence dip, where it falls again, before continuing to rise.

That's the theory, at least: Actual mileage may vary. Void where prohibited by law. Not responsible for anything at all. And all the other usual caveats...

The biggest air gap you can afford is the best!

You are talking about STC, not TL: STC is a lousy system for measuring studio isolation, since your real problem will NOT be in that range. Your real problems will be in the range below 300 Hz, and especially below 125 Hz, where things like drums, bass guitars, and keyboards live. Those are by far the hardest instruments to isolate. And they are also the loudest. A wall that rated at STC-80 could very well be really lousy at isolating drums and bass guitar, so you should be looking at the low end of the spectrum too. Use TL, not STC.

And isolating to 80 dB TL is a really tall order. You might be able to do that with your walls, but your floor will be flanking way lower than that. You definitely would have to go to separate foundations, or floated rooms, to get to that kind of level.

And building doors to get you to 80 dB TL... ... well, lets just say that you are talking about dual (back to back) huge, massive, monster doors, hung on enormous hinges, and with multiple seals around the entire perimeter of each door. Same for windows: you are talking inch-thick (25 cm) laminated glass in massive frames, with huge air gaps. Then there's the HVAC system: Getting 80 dB of isolation through the HVAC ducts is a major design issue. Not easy at all.

To do that purely with mass law, you would need walls with a surface density of about 2000 kg/m2. That implies reinforced concrete walls, one meter thick... With MSM you could probably do it with a pair of reinforced concrete walls, each 25 cm thick, and an insulation filled cavity of 30 cm depth (theoretically). And, of course, the leaves would have to be on separate foundations, or the entire inner-leaf would have to be floated on springs, with F0 tuned well below 10 Hz.

I guess the big question here is: Why on earth do you need such major isolation? Are you planning to explode hand grenades in there, with sleeping babies across the hall? Seriously, getting 80 dB of isolation is a major undertaking, so you are in for some huge expenses. Once again, the issue is not going to be the walls, but rather flanking through the building structure. Everything will have to be floated on springs, to get that kind of isolation. The issue is also going to be the doors, windows, HVAC and electrical systems. Isolating those to 80 dB is hard. I truly hope your budget has several zeros in it, since you are going to need lots of those!

So please satisfy my curiosity: WHY do you need such major isolation? Torture chambers? Testing explosives? Test bed for jet engines? Shotgun shooting range?

You sure did get me curious, Mr. Madman!

- Stuart -

_________________
I want this studio to amaze people. "That'll do" doesn't amaze people.

Hi Stuart,

I'm very grateful for your reply - very detailed indeed! First, the reason for attempting to get such a high isolation is due to 2 factors:

1 - If the structure is designed to just get what we need, if we then realise that we needed more, we won't be able to improve it easily.

2 - We don't really know how this media complex is going to work out. The worst case scenario is that we might have a band rehearsing with a PA in the double-height room and 3 producers all trying to work on their own projects in the other rooms. I'm PRETTY sure that, no matter how hard we try, we won't get satisfactory separation (unless we spend £1 million - which we don't have) and a degree of thumping will inevitably leak from one room into the others. Maybe the doors won't seal so well, maybe the ceiling won't be as substantial as it should, maybe the windows won't be thick enough. But if we get the structure right, we can work on these problems over time.

The weakest link, unfortunately, is the roof. According to the current design, the building is covered by a 300mm timber roof supporting roof panels. A freestanding steel frame will hold the timber up, and it won't be connected to the blockwork. The thickness of the rafters is just about enough to stuff it with rockwool and so treat the ceiling of the 2 white rooms on the first floor - this won't decouple the rooms, because sound will travel through the roof, but if we float the room with the higher ceiling of the two (the roof is slanted across the building) we might stand a chance of getting enough isolation.

The tricky part is the isolation of the double-height room ceiling. Obviously, it needs its own ceiling resting on the inner skin blockwork, otherwise it'll drum any sound into a huge section of the roof - or collect it from the rooms on the first floor in much the same way. The ceiling could rest on steel beams or engineered wooden beams, but either of them will require blockwork columns to hold them up, a single skin of 215mm blocks 5m high just won't be enough I don't think!
My idea would be to leave a 30-40cm. air gap between roof and ceiling - I was planning on a sandwich structure similar to that of a floating floor, but then I thought that corrugated steel covered in 1 foot of sand could do the trick and it would be easier to install - I have seen and used a rehearsal room built like that and it worked rather well. Mind you, the beams were big!

Having beams on the ceiling means that we could then hang whatever acoustic treatment (diffusors, perforated panels, etc.) from them at a later stage, without trying too hard to incorporate them into the ceiling structure.

Still a lot of thinking to do, but if I can crack the roof problem, I feel I can relax a little.

Thanks,