No matter what gear you use to capture an instrument’s sound, if the acoustics of the space are poor you will always get a skewed sonic picture of that instrument. Likewise if the control room itself accentuates various frequencies, minimizes others, and adds excessive echo, you won’t really know what your mix sounds like. Many unfortunate surprises can creep up when you take your final mix to another listening environment.
When I recently decided to take my own home studio to the next level, I spent months considering possible floor plans, analyzing acoustical issues, reading up on others’ approaches, and surveying my friends and colleagues in the industry. What I quickly discovered was that there is a great wealth of information available on the subject, but a lot of it is quite confusing and often even conflicting.
To make matters worse, most of the solutions offered proved to be too expensive for my rather meager budget. Without batting an eye, one person told me that I absolutely needed “X” to do it right—which would put me back at least $80,000. It was somewhere around then that I learned to take everything with a grain of salt.
Like so many musicians, I have limited funds but lofty dreams, so I decided to get the most for my dollar by doing all of the work myself and finding inexpensive solutions to address my acoustical concerns. What I ultimately came up with was a montage of simple yet effective design concepts that in the end turned out better than I had dared hope.
This article is about that other room in project studios, the one that’s most often sonically abused: the isolation booth.
When deciding how to approach building a booth, there are many factors to consider:
• How dead or neutral do you want it?
• How much room do you have?
• What will it be used for?
• How much money and time do you have?
For my studio I wanted an isolated, sonically neutral, medium sized space with a fair amount of broadband frequency attenuation. It needed to be large enough to isolate a full drum kit as well as accommodate vocal overdubs. Part of the reason for this choice was that I was also creating a second, larger room that would be much more live and complex acoustically. Both spaces would be created by dividing a very large preexisting basement playroom.
If you have the luxury, the first and foremost concern to address is the room dimensions. The spatial relationships of the walls, floor, and ceiling will determine what fundamental frequencies, or modes, the room will accentuate.
Since I would rather my room have an even response across the sound spectrum, I worked to find dimensions between opposite surfaces that would not be too closely related. The primary rule to keep in mind is that these measurements should never be multiples of one another (e.g. the room’s length and width are the same, or the length is twice the height). If they are, the emphasized frequencies will be the same and become overpowering. Modes can overlap and become cumbersome even if one follows this basic principle—it just happens less often.
To determine the most basic of these modes, called axial, that occur between two parallel surfaces, divide 172 by the length of the room (in meters) to obtain the primary mode or fundamental frequency. This can then be multiplied by whole numbers to calculate upper harmonics of this primary mode.
By figuring these out to at least the tenth harmonic for all three dimensions you can compare to see which frequencies are shared. While there are also modes that occur between 4 or 6 surfaces, called tangential and oblique, we will ignore these for now to keep things simple.
The table below shows the original modes of my room in Hertz. The dimensions were based on three preexisting walls, floor and ceiling, but constructing a fourth wall whose placement was suggested by the shape of the original room and around the houses support structure. Although the dimensions already avoided multiples, there were still issues at 183 Hz, 366 Hz, 412 Hz, 549 Hz, and possibly also around 246 Hz.
To avoid these problems, I tried adding and/or subtracting a few inches from either the length or width to see what would happen. In this particular case the height could not feasibly be changed. If I kept the original length and made the width 115", the number of problem frequencies was only 1—but it occurred on all three planes! Ultimately, I chose to make the length of the room 144" so that the instances of overlapping modes fell from 5 to 2.
Using these new room dimensions I built a wall across the middle of the room at slightly over 6 degrees from parallel with the far wall in back. The 144" length was measured from the center of the wall so that would be the average distance. The angle (roughly equivalent to one inch drift for every foot) also helps to reduce the strength of axial stand ing waves in each new room, thereby serving to smooth out some notches in both of their frequency responses, and specifically addresses one of the two remaining problematic modes in my booth. Even more importantly, an angle of this nature helps prevent flutter echo, sometimes called “pinging,” in the higher frequencies.
The dividing wall itself was built using a double wall construction, with two isolated walls built one inch apart and with alternating 2x4 support beams. Each wall was then insulated with regular Kraft faced R-13 fiberglass and a triple layer of 5/8" drywall attached to each outside surface.
The drywall was cut so that the seams of each layer would not overlap and all seams were taped and plastered before more drywall went up. This helped to further minimize possible leakage between the two rooms. Silicone caulk was also used across the bottom of each side to make sure that there were no gaps between the walls and the concrete floor for sound to sneak through.
In order for ensemble members to get visual cues from one another, I decided to put a large double-paned window in the new wall. It was critical that this be done properly since windows and doors located directly between two recording spaces are often the greatest sources of sound leakage in any studio. To minimize this I needed to address three factors: physical coupling between the panes or frames, getting an air tight seal on b oth sides, and minimizing acoustical transmission between the panes of glass.
Decoupling was accomplished by creating two separate frames, one for each side of the double wall. Each frame was made from solid 2" thick oak slabs that were separated from each other by a 1" neoprene gasket.
While the mass of the oak makes it more difficult for the frame to vibrate, the neoprene gasket assures that the frames do not touch but will still form an air-tight seal. The panes themselves were cut from solid laminated or glazed glass, 1/2" and 5/8" thick respectively. Again, it is the mass of such thick glass that makes it more difficult to vibrate and the two different thicknesses mean that each piece will have different resonance frequencies.
The panes were tilted outward toward the top at a 4 degree angle, making an 8 degree total, again slightly more than that magic 6 mentioned earlier with the wall. The sills were lined with 1/4" thick neoprene, which decouples the glass from the frame and holds each pane snugly. Half-inch neoprene is often used instead, and it would have served even better, but the cost was prohibitive. The sills and liner were also caulked to ensure an airtight seal.
While installing two high quality prefabricated and insulated windows would have been an easier and reasonable alternate solution, I found that they would cost about the same but look and perform less professionally than if I did the job right and built my own. Considering the time and energy that this installation took, however, you may find this other approach to be a mental and physical godsend.
Installing the extra thick, solid core Luan double doors required precision and care, but it was a relatively simple task after installing the window. The door frame was created from scratch, since most inexpensive prefabricated frames do not have the close tolerances and larger contact area to form as tight a fit and seal as necessary.
To ensure this type of fit the outside edge of the frame was not secured into place until the door itself was already hung. The frame was made from doubled 2x4 beams and I created a sill around the interior with 1.5x.75-inch pine slats, which were caulked and later fitted with a wide foam seal so that there were no openings for sound to get through when the door is shut. Finally, a number of long metal strips attached to the exposed wooden edges of the frame keep it from being torn up when artists load gear in and out.
Earlier in my decision making process, it was suggested that I could eliminate doors from that wall altogether and make a new entrance to the booth through a pre-existing side wall. It was a very good idea that would have eliminated further possible leakage points, but it was not really feasible for me. That plan would mean that the drummer would have to load in and out through the only other possible entry point: the bathroom! While this would certainly provide a “conversation piece” aspect to the studio, I ultimately decided against it.
With all this accomplished, there were a few last basic details that needed to be taken care of so as not to undermine all the previous hard work. Transmission of acoustical energy between the two spaces could still occur through gaps between the floor and preexisting walls, so these were all filled with a hefty bead of caulk.
Next, I opened up the ceiling in about an 18-inch strip along the double wall just inside the booth. This was to insert insulation (unfaced R-13) into the ceiling in both directions to minimize that possible egress for the sound.
Finally, so as not to bother the neighbors or get unwanted outside noises in my recordings, the two deep window openings were filled using a combination of some cheap pillows, insulation, 3-inch thick foam, and felt covering. I also discovered that I had to caulk the outside perimeter of my house up under the siding where it meets the foundation. While working on the control room one day, it was a persistent cricket that convinced me of the absolute need to fill this gap.
Then it was time to address interior acoustical treatments. While the booth was now reasonably quiet and isolated, it was still haunted by flutter echoes and, yes, a few nagging modal issues. The thick carpet that was to be put down would certainly help reduce this, but not enough to fix these problems.
Broadband absorption was called for. The constraints were that it had to be done within my limited budget and the physical space available for the booth.
There are quite a number of different absorption products available these days that are very good. While I used some of these in other parts of the studio, it was not economically feasible to treat the entire booth with them, and only the most expensive of these products would address low frequency issues. To deal with this I used ideas from several industry sources.
One frugal solution for mid and high frequencies was to hang fabric-covered, unfaced, standard pink insulation on the walls. While it is not as uniform in its frequency absorption as some of the shaped foam products available, there is an averaging effect that occurs when it is used over greater areas due to its rather chaotic tonal character.
The range of frequencies absorbed by this method, however, can be determined by one of the most fundamental physics principals of acoustic design: the interrelationship of frequency, wavelength, and the speed of sound itself. This relationship is expressed by the mathematical formula f=v/l, where f is frequency, v is the speed of sound (approx. 344 meters/second), and l (lambda) stands for wavelength.
If at least a quarter wavelength of any sound is interrupted at the surface of reflection (i.e. the wall), then it can’t reflect. Therefore it is the thickness of the insulation that will determine what the low-end cutoff frequency of absorption will be. By modifying the formula to l=v/f and then dividing the result by four (or l=0.25v/f), we can see that a 6-inch sheet of insulation will absorb from 564 Hz and above (1 inch = 0.0254 m).
(0.25 x 344) / (0.0254 x 6) = >564 Hz absorbed For 9 inches of insulation:
(0.25 x 344) / (0.0254 x 9) = >376 Hz absorbed
While this works pretty well for both mid and high frequencies it is not really feasible to use for lows. If we wanted to absorb down to 20 Hz, for instance, the formula tells us that it would take around 4.3 meters (over 14 feet) of insulation to do it! Since this would more than fill my entire booth, I decided to find other solutions.
To aid in the absorption of low frequencies I incorporated a second general principle into my fiberglass absorbers: harmonic resonance through sympathetic vibration. Basically, any object will vibrate at specific frequencies that are based on its dimensions and other physical characteristics. Since an acoustic energy is converted to motion (vibration) in that object, it is therefore lost (absorbed).
While this again means greater dimensions for low frequencies, if we take advantage of the space along the wall rather than out from it we now have larger areas to work with that will not detract from performance space significantly. So I glued the insulation to large sheets of pegboard, selectively suspended by hooks from the ceiling and walls so it would be as free to move as possible. The exact modes also depend on just how the pieces are suspended, which affects the ways in which they can vibrate.
To further the broadband effects of these absorbers, it is good to follow two more basic rules. First, make sure that the dimensions of any particular pegboard slab are as unrelated as possible. Second, use as many variously-sized and shaped large absorbers as your room will accommodate. Variety is not just the spice of life—here it is key.
To enhance the aesthetic visual quality of the room I covered my absorbers in various fabric animal prints. Though this looks great and clients seem to love it, I do have a problem keeping people from petting it. Warning them that they might get uncontrollably itchy from the fiberglass (which has not yet actually happened) helps keep the pawing to a minimum.
That brings me to one last point about this particular type of absorber; it must be bound so that fibers cannot be emitted into the air. This is also true if rockwool is used instead of fiberglass. US Gypsum makes a rockwool and sells a very thin plastic sheathing material that won’t significantly affect the frequency response but will keep the fibers out of the air. Rigid fiberglass, such as Owens Corning 703, does not shed fibers as easily and is a better broadband absorber, but is a bit more expensive.
As final measures to attempt further bass absorption, I created two different bass traps. By building a non-parallel closet in the back corner and leaving the top open, I hoped to accomplish a sort of sonic roach motel: certain frequencies check in but not out.
A similar thing happens in another corner where I hung a large vertical absorber an inch from the floor and the converging walls, but kitty-corner to leave a fair sized gap behind it. After these measures, and hanging the animal-themed absorbers, the usable room dimensions are about 10 feet by 9 feet, still big enough for a drum kit.
One last design approach that I should certainly mention (even though I didn’t use it) is diffusion, where both direction and timing of reflections are scattered off a complex surface. Diffusers help alleviate standing waves and flutter echo but do so without damping or absorbing sound. In fact, they can give the sense that the room is larger than it is. Since that was not the sound I wanted for this room, and diffusers are both complex and expensive, I decided to leave them out of my design.
The underlying concepts and designs introduced here are based on the most fundamental acoustic principles. Good isolation is achieved through mass, acoustical decoupling, and an air tight room. (Ventilation issues are worthy of another article in and of themselves!) Sound quality in a room is addressed through a combination of proper dimensions, angles, absorption, and diffusion.
While some of the methods used were rather crude, they are all quite functional and work well for those on a budget who are not put off by the do-it-yourself approach. This sort of project is best undertaken only if you have the time, patience, and skill... but may lack the funds to hire contractors and purchase prefabricated parts.
By starting with these ideas and modifying them for the specific needs of your own studio, you can create much better recording environments in which to work. When facing the challenges of your individual space, remember to design and be as musicians are supposed to be: creative.
Most of all, have fun.
John Shirley, Ph.D. can be reached via firstname.lastname@example.org.