Fiberglass insulation around false wall instead of grn glue?

[joenobody]

I'm going to make a false wall to help soundproof an 11'x11' room from the outside. Here's my plan:

1.) I will mount steel studs vertically to the existing wall

2.) r13 fiberglass will be mounted between the steel studs

3.) 5/8" sheet rock panels will be mounted to the steel studs

4.) The sheet rock gaps will be taped and sealed with drywall compound

My questions are:

1.) Are there any problems with what I have in steps 1-4 so far?

2.) For sealing around the edge of the false wall, I would like to avoid using acoustical caulk and so forth because I'm planning on taking the wall down when I move out and don't want a lot of marks that I have to clean up. I would like to do is pack fiberglass insulation in the gaps around the edges of the false wall. Will this work close to as well as using acoustical caulk?

[Soundman2020]

Hi "joenobody", and welcome to the forum!

1.) Are there any problems with what I have in steps 1-4 so far?

You've described what you plan to do for only one wall: we'd need to know the plans for the other 3 walls, and the ceiling, as well as the doors, windows, HVAC system and electrical system, in order to be able to say if it will work or not. Trying to isolate a room by beefing up just one wall is sort of like trying to build an aquarium with only one piece of glass in it.... Unless you put glass on all sides, it won't hold water. Same with a room: unless you beef up all walls to the same level, the room won't "hold sound".

Also, your proposed method would not isolate well in any case, and especially not for low frequencies. You are planning to do this: "... mount steel studs vertically to the existing wall ... 5/8" sheet rock panels will be mounted to the steel studs..." That implies that the studs would be attached to the existing wall, and therefore would not be decoupled, so there would be major flanking from the existing wall to the new drywall. Also, you would only have a thin air gap there (the depth of the studs), so the MSM resonant frequency would be fairly high, thus limiting the isolation for low frequencies (drums, bass, keyboards, etc.). And finally, assuming that the existing wall already has a frame of some type with sheathing on each side, your proposed procedure would turn that 2-leaf wall into a 3-leaf wall, thus limiting isolation even more, and perhaps even making it worse than it is right now for very low frequencies. A 3-leaf wall will always have worse isolation than a 2-leaf wall, all factors being equal.

2.) For sealing around the edge of the false wall, I would like to avoid using acoustical caulk and so forth because I'm planning on taking the wall down when I move out and don't want a lot of marks that I have to clean up. I would like to do is pack fiberglass insulation in the gaps around the edges of the false wall. Will this work close to as well as using acoustical caulk?

Unfortunately, that won't work. In order to isolate a room acoustically, it must be sealed airtight. Fiberglass insulation will not accomplish that, since it is, well, "fibrous": it consists of numerous fibers with air between them. Compressing it greatly does not make the air gaps go away completely: they just get smaller, but they are still there, so it won't seal.

You absolutely do need to seal all four walls, and the ceiling, and the floor, and the doors, and the windows, using a suitable sealing compound, such as acoustic caulk or even common bathroom caulk. If not, then you won't get isolation.

Try this experiment: drive your car to a really noisy location, and close all the windows fully. Listen carefully to the sound level inside the car. Now open one window just a tiny fraction, maybe a 1/16", and listen again: notice the huge increase in sound level inside the car? That's what you'll get if you do not seal your room properly.

One other point: Green Glue is not acoustic caulk. The Green Glue Company makes acoustic caulk, yes, but their flagship product (called simply "Green Glue" by most people) is not acoustic caulk. It is an entirely different compound, meant for a very different purpose. It cannot be used as caulk, and also isn't glue! Maybe you already knew that, but I wanted to add the comment to clarify for others who might be reading your thread.


- Stuart -

[joenobody]

Thanks for your help. I was under the impression that steel studs decouple about the same amount as resilient channels, but apparently not?

In that case it would be more effective to build a freestanding frame that's sealed with caulk to the existing walls and have it supported with legs on the inside of the room?

On the low frequencies and other 3 walls, I really only need to isolate for midrange and highs in order to keep out the noise of my neighbor's yappy pugs and to keep in the noise of my vocals. All the drums and so forth will be done electronically and never played very loud. The vocals that need to be contained are the higher pitched chorus parts since it's difficult to sing low and loud anyway, so probably it's in the 4k-8k range that's most important.

With respect to the other 3 walls, the room I'm isolating is sandwiched between the rooms of the rest of the house and I've found that noise coming out of the room is quite directional in that if you're standing in the back yard, the sound coming from it is very dependent on the angle and distance you are from that room, and if you go around the corner of the house the sound is gone completely.

I really only have space in the room to isolate one wall anyway and if I isolate two or three it would not be a project worth doing since I'd have no room to work.

Regarding the thicknesses, if I want something that will reduce 4k vibrations, can I plug 4k into a wavelength-in-air calculator to get an idea for about how thick the wall needs to be?

For instance, using 4k on this page: http://www.sengpielaudio.com/calculator-wavelength.htm I get 3.37", is that at all relevant?

[joenobody]

Also, here's the page I was thinking of with the idea that 25 ga steel studs act like resilient channels:

http://www.gearslutz.com/board/studio-b ... annel.html

metal studs do the same as wood studs and resilient channel. Resilient channel was invented to effect the same acosutic properties with wood studs as metal studs have.

Are they wrong about that or am I misinterpreting it?

[Soundman2020]

I was under the impression that steel studs decouple about the same amount as resilient channels, but apparently not?

Light gauge steel studs, yes, when used correctly: ie. fully decoupled. But that's not what you were planning to do.

In that case it would be more effective to build a freestanding frame that's sealed with caulk to the existing walls and have it supported with legs on the inside of the room?

I'm not sure I understand: If you build a proper decoupled frame, why would you need to "support it on legs"?

Also, here's the page I was thinking of with the idea that 25 ga steel studs act like resilient channels: ... Are they wrong about that or am I misinterpreting it?

Andre and Rod are talking about using light-gauge metal framing to build the inner-leaf of a typical decoupled two-leaf "room-in-a-room" studio. That's not what you are talking about. You were planning to attach metal studs to the existing wall (not decoupled), so therefore there would be benefit. In order to act as RC does, they need to be light weight, and also decoupled.

If you take a look at IR761, it should be a lot clearer.

To be very honest, I think it would be a lot easier, neater and more effective if you were to just take off the drywall from the inner surfaces of the room, put up proper resilient channel on the studs (whatever they are: either wood or metal), then hang new drywall on top of that. Two layers of 5/8" would get you pretty good isolation.

On the low frequencies and other 3 walls, I really only need to isolate for midrange and highs in order to keep out the noise of my neighbor's yappy pugs and to keep in the noise of my vocals. All the drums and so forth will be done electronically and never played very loud.

It would really help if you could explain what you are trying to accomplish, and why! And also the level of isolation you need for your room. Designing the isolation system for a room can't be done by guesswork: it should be calculated, or at least done empirically, based on known situations, methods, and materials. It's not clear at all if this is just a rehearsal room where you need to isolate your voice gong out, or a recording area, where you also need to isolate outside sounds coming in. It's also not clear what level of isolation you need.

The vocals that need to be contained are the higher pitched chorus parts since it's difficult to sing low and loud anyway, so probably it's in the 4k-8k range that's most important.

That's way beyond the top end even of the piano! A full grand piano covers range from 27.5 Hz to 4186 Hz, so in order to sing in the 4k to 8k range, you'd be singing an octave higher than the highest note on a piano...!

The range for male and female voices (fundamental frequencies) is between about 70 Hz for a very deep male bass voice, up to about 1,300 Hz for a female soprano. Yes, there are harmonics that go quite a bit higher, but most of the energy is in the fundamentals and one octave above, which is much, much lower than you imagine.

A basic tenet of isolation is that the MSM resonant frequency of the wall should be tuned to at least one octave lower than the lowest tone you need to isolate, so for that very low male vocal, you'd need to tune your wall no higher than 35 Hz. Your needs might not be as extreme (depends on your own vocal range), but you do need to isolate down to pretty low frequencies.

The graph below shows the approximate frequency ranges of music and singing:

human-speech--formant_singer.jpg

noise coming out of the room is quite directional in that if you're standing in the back yard, the sound coming from it is very dependent on the angle and distance you are from that room, and if you go around the corner of the house the sound is gone completely.

So it seems that your main concern is to isolate sound leaving the room, so that you don't disturb the neighbors? (And the dog! Can't forget him... )

I really only have space in the room to isolate one wall anyway and if I isolate two or three it would not be a project worth doing since I'd have no room to work.

Why? Resilient channel is only about a half inch deep. Two layers of 5/8" drywall plus Resilient Channel adds up to less than 2". Is the room really so tiny that it would become unusable if you lost 4" of width, 4" of length, and 2" of height?

Regarding the thicknesses, if I want something that will reduce 4k vibrations, can I plug 4k into a wavelength-in-air calculator to get an idea for about how thick the wall needs to be?

Unfortunately, it's a whole lot more complex than that! If all you want is to isolate 4k, then a sheet of cardboard would do a pretty decent job... Isolating a room requires knowing the lowest frequency that you need to isolate, not the highest frequency. You then need to calculate the MSM resonant frequency of the wall to be at least an octave lower (half the frequency), and you do that using this equation:

MSM-resonance-equation.jpg

... if it is a two-leaf wall, or this one below if it is a 3-leaf wall:

TRIPLE-LEAF-EQUATION.jpg

That will give you the masses and air gaps that you need in order to obtain the correct tuned frequency. Then you use the following equations to plot a graph that shows the full predicted isolation spectrum and levels for that wall, so you can check if it covers the range and levels that you need:

Full-spectrum-TL-isoaltion-equations.jpg

Where m1 and m2 are the surface masses of each panel and d is the depth of the airspace between, f0 is the mass-air-mass resonance frequency and f1 is equal to (55/d) Hz. R1 and R2 are the transmission losses for the individual panels.

That's the procedure if you want a wall that does what you want it to do, isolating to the level you need at the frequencies you need.

If you don't feel like doing all that math, then the only other method is to examine documents like IR761, with hundreds of empirical measurements done on real walls, to find one that fits your needs.

So there are ways of figuring out how your isolation system can be attained, but they have to be based on realistic numbers and full understanding o the situation. So when I say that using Resilient Channel all around your room with 2 layers of 5/8" drywall on it, that's based on empirical experience: That will get you really good isolation for the majority of cases, assuming that your needs are typical and your situation is typical. But in order to be sure, you'll need to tell us a lot more about what you are trying to achieve, what the current levels are, what levels you need to attain, etc.


- Stuart -

[joenobody]

Sorry for not replying back sooner, thanks very much for your very detailed response. It's extremely enlightening and clears up many things for me.

I had planned on getting some test data before replying but I have not been able to due to various things so I decided to just reply back to make sure I thanked you. Thanks!

On comparing steel studs and resilient channels, is this what is meant when they say they're equivalent? In ir761 there is:

Wood+RC STC 43:

Page 166

Element Description:
1
single layer of 16 mm type X gypsum board
2
90 mm wood studs at 406 mm on centre
3
90 mm of glass fibre insulation in cavity
4
resilient channels at 406 mm on centre
5
single layer of 16 mm type X gypsum board

Steel Studs STC 46:

Page 74

Element Description:
1
single layer of 16 mm type X gypsum board
2
90 mm steel studs at 406 mm on centre
3
90 mm of glass fibre insulation in cavity
4
single layer of 16 mm type X gypsum board

They have similar STC so is that where the equivalence comes from?

What I still don't understand is this part:

"...if you were to just take off the drywall from the inner surfaces of the room..."

I realize that the inner walls as they are now are not decoupled, but if I were to do resilient channels all the way around the room, wouldn't it make more sense to add the RC on to the current drywall? It seems like losing all that mass would not help things.

I haven't collected the decibel information I wanted to send to illustrate this, but the informal testing I've done suggests that I probably don't need to worry about flanking, so redoing the other three walls would not be needed. If I were able to spend that much time and money in the room, it would be better spent making that outside wall very soundproof rather than redoing the other 3 which already are soundproofed since they're surrounded by the rest of the house.

The main reason I wanted to use steel studs against the wall rather than RC is that with steel studs you have space to add insulation to add to the STC. Assuming that I don't actually have any flanking problems, what about using both steel studs and RC for the outside facing wall?

Steel Studs+RC STC 54

Page 237

Element Description:
1
single layer of 13 mm type X gypsum board
2
single layer of 13 mm type X gypsum board
3
90 mm steel studs at 406 mm on centre
4
90 mm of glass fibre insulation in cavity
5
resilient channels at 610 mm on centre
6
single layer of 13 mm type X gypsum board

[Soundman2020]

They have similar STC so is that where the equivalence comes from?

There's a difference of 3 points in the STC rating: that is significant, but not huge. So yes, from that point of view, they are similar. However, you should not be looking at the STC rating at all: you should be looking at the TL curve. STC does not tell you anything useful about how well a wall will work for isolating a studio.

wouldn't it make more sense to add the RC on to the current drywall?

No, because then you would have a 3-leaf system, and you'd also have a very small air gap on the third leaf (only the depth of the RC). So the MSMSM resonant frequency of the entire wall would be rather high, and the isolation for low frequencies would be worse than if you took off the drywall, added the RC, then put it back again. A 3-leaf system will always give you worse low frequency isolation than the equivalent 2-leaf system, all other factors being equal.

but the informal testing I've done suggests that I probably don't need to worry about flanking, so redoing the other three walls would not be needed. If I were able to spend that much time and money in the room, it would be better spent making that outside wall very soundproof rather than redoing the other 3 which already are soundproofed since they're surrounded by the rest of the house.

Isolating a room works as a complete system: ALL of the parts of the room must be done to the same level. You can't do just one or two sides really well, and not do the other sides. All sides of the room must be isolated to the same degree. If not, then sound will simply take the easiest path out of the room, through the "weakest" side, and once it is out, it will spread out all over.

The main reason I wanted to use steel studs against the wall rather than RC is that with steel studs you have space to add insulation to add to the STC.

If you do it properly, by taking the drywall off, adding the RC then putting it back again, you will have even greater depth for insulation inside the wall.

Also, the insulation does not increase the STC rating directly: that's a side effect of how isolation works as a system. Insulation by itself is really lousy at stopping sound. The purpose of insulation inside a wall cavity is not to stop more sound, but rather to damp the resonance inside the wall cavity, and to change the way the sound waves lose energy as they move through the cavity (adiabatic/isothermal). The increase in isolation is because the combination of the above drives the resonant frequency of the wall down to a lower frequency, and therefore the wall will provide better isolation at all higher frequencies. You can think of it as sliding the entire TL curve across to the left.

Figuring out isolation is not as simple as it looks at first glance: there are a large number of factors involved, and they all interact with each others, sometimes in unexpected ways.


- Stuart -