November 8, 2008
If you stand outside our studio on any given afternoon, you can hear the rumble and roar of the freight trains rolling through an open below-grade pass half a block from our building. This, along with the noise of the buses clambering up Charles Street and the boom of heels on the floor above, presented our first challenges in designing our studio. Not to mention the concerns our neighbors had of a bunch of crazy musicians playing all day and all night right next door. There were many times in the process of planning this renovation when we considered the possibility that we had lost our minds. This was one of them.
So the first thing we did was hit the books. We spent a lot of time researching how to acoustically isolate the live room and control room to get rid of noise both coming in and going out. It turns out that two of the biggest factors in sound isolation are mass and air gaps. So we had a couple things working to our advantage right off the bat. First, the walls separating us from our neighbors and the exterior world were laid about 120 years ago and comprised of 8-10 inches of stone and brick. That’s some serious mass. We also had four-foot hallways between us and our neighbors on either side. Built-in air gaps. Plus, our ceilings were over 10 feet high, which left us with quite a bit of headroom for treatments.
To determine how many decibels (dB) we really needed to get rid of, we conducted a few tests. Our high tech evaluation consisted of one person banging the hell out of a drum kit in what would become the live room while another person recorded dB levels from next door with a $40 sound level meter from Radio Shack. We then compared that to the dB levels of the everyday noise level in that building to determine how much noise we were adding without any sound treatments at all.
We then performed a few variations on the test. We tried it with a keyboard playing various notes through an amp. (This gave us an idea of how the volume changed with different frequencies.) We tried it in different rooms to learn whether the hallways made a difference. (They did.) And we tried it closer to the windows to see how they factored in. (Cold air isn’t the only thing getting in.) And then we did the same thing with the meter in our own space to see how many db the buses and trains sent into our space as well as people walking upstairs.
What we found was that the buses outside actually contributed more noise next-door than our drummer and the weakest link was definitely the old windows. But the prize for most dB went to the people walking upstairs. This impact noise was actually louder than the trains. But while impact noise is notoriously tough to get rid of, so are super-long low frequency soundwaves, which is what the train’s low rumble traveled over.
So after doing some math, our course was set. We had about 40 dB to get rid of as well as some troublesome impact noise. For you acoustical nerds out there, this meant that with the 20 or 25 db our existing structure was already eliminating, we needed to up the Sound Transmission Class (STC) for our new construction to about 60 or 65. Oy!
The key with the live room was going to be building it as a “box within a box” or, in other words, framing and building out four walls, a ceiling and a floor without having any rigid connections between the new inner room and the old external room. Sound travels really well through continuous rigid structures. It does not travel as well through air. (Remember yelling to your parents in the basement through the heat ducts? Your voice was actually resonating through the metal ductwork rather than through the air space inside them.)
So the solution is lots of rubber, lots of air gaps and a type of hat channel called resilient channel (RC) that attaches to the old structure on one side and the new structure on the other and allows give between the two to absorb sound vibrations without passing them through to the rest of the building.
The rest of the construction was fairly traditional, except that we doubled up layers of materials to add mass, packed everything with insulation and allowed for an air gap between walls to allow sound waves to bounce around and diminish. (Ok, not so traditional.) In the live room, we floated the floor on top of the existing floor by resting the new joists on rubber “U-boats” (essentially engineered hockey pucks). This allowed the floor to be anchored without being rigidly attached. Then we packed insulation between the joists and topped it with two layers of plywood, underlayment and a bamboo finished floor.
In the control room and the live room, we floated the ceiling by applying the same concept but in reverse. The framing was attached to the existing ceiling with construction adhesive and long screws and packed with insulation, allowing space for a small air gap. To then separate the rigid attachment from the new ceiling, RC was attached and then a sandwich of drywall and soundboard was attached to the RC. The idea here is that the screws going through the drywall do not penetrate the ceiling joists, only the resilient channel, so there is no rigid attachment. (Note: If you use our plans and the ceiling falls on your head, don’t sue us. We’re neither architects nor engineers. This part we kind of winged. Though we did manage to get all our stamps and permits.)
To take detail to the extreme, the drywall boards were all staggered so there were no consistent seams from one side of the ceiling or floor to the other. And finally, a 1/4″ gap was left between the walls and the floor and between the walls and the ceiling. This was later sealed with a non-hardening acoustical caulk. Fortunately, our contractor was as anal as we were.
We replaced all the old windows with triple-glazed windows and sealed them with caulk. The window in the control room was replaced with glass block, which has a Sound Transmission Coefficient in the 50s. (That’s good.) And we divided the front room into a lounge and a live room, which provided an additional buffer from the traffic outside as well as some chill-out space.
In the name of saving some cash, we decided not to float the floor in the control room. There is very little recording done in there and we agreed it was a realistic compromise. What that means today is that when the train goes by, you can stand in the office in the rear of the building, which has no acoustical treatment and hear the rumble fairly well. Then you can walk into the control room, close the doors, and just feel the bass in your chest. Finally, you can step up into the live room, close the doors and not hear a thing.
It was a good feeling to hear that silence after all the blood, sweat and fears of planning and construction. And it proved that you don’t have to blow your budget on name-brand acoustical materials when layers of plain vanilla drywall, fiberglass insulation and (the cheapest of all) air will suffice.