The acoustics of a recording studio are sometimes one of the most neglected aspects, especially in those with a tight budget. The tendency is to invest in mics, compressors, monitors and so on, while one of the factors that most influences the end result of our recordings and mixes, the acoustics of the room where they are done, is not taken into account. Close miking partially mitigates a possible harmful effect of the recording room, but what use are all our processors, plugins and excellent monitors if the control room where the mixing takes place has a series of room modes of its own that distort the frequency response? Hiring the services of a company specializing in acoustics to design and treat the studio is always the best option. However, there are a number of basic measures we can take to improve our listening space without having to invest a lot of money.
In this first article devoted to acoustic treatment, we will present some of the problems inherent to enclosed spaces, and in the next issue we will go deeper into the materials and techniques to solve them.
A bit of theory
Sound is the sensation produced in the organ of hearing by the vibratory movement of a body, transmitted through an elastic medium such as air or water. That vibratory movement causes the particles around the body to compress (pressure increase) and expand (pressure decrease). As this movement passes from one particle to the next, a wave is formed, which finally reaches our ear.
The number of times per second this compression/expansion cycle repeats is called frequency. On average, the human ear perceives waves with frequencies between 20 and 20,000 cycles per second (or hertz).
The period is the time the wave takes to complete one cycle, and is therefore the inverse of the frequency. So a 20 Hz wave has a period of 1/20 = 0.05 seconds (50 ms), and a 20,000 Hz wave has a period of 1/20,000 = 0.00005 seconds (0.05 ms).
Taking the usual 340 m/s for the speed of sound, the physical space occupied by one cycle of the 20 Hz wave is 0.05x340 = 17 meters, while that of 20 kHz occupies just 0.00005x340 = 0.017 meters, that is, 17 mm.
This distance is called wavelength (usually represented by the Greek letter lambda: l), and as we can see, it decreases as frequency increases.
Figure one is a graphic representation of a pure 1000 Hz sine tone. The first cycle of the wave is marked in orange, and we can see that it spans 0.001 seconds (1/1000). If the representation were a function of space, the distance covered by the cycle, or for example the one between the two points marked with the label “Wavelength,” would be 340x0.001 = 0.34 meters.
The following chart is a summary of what has been covered so far:
Room Modes
Imagine a speaker emitting a pure tone (a single frequency) inside a room. The sound will reflect in different ways off the walls, floor and ceiling, and part of that sound will reach the source again. If the path traveled equals the wavelength of the tone, the reflected wave will be in phase with the one coming from the speaker and will add to the one being emitted, reinforcing that frequency. This type of wave is called a standing wave, and when we talk about a room’s acoustics, it makes up one of that room’s modes.
In addition, any multiple of that frequency will also generate a standing wave and therefore another mode, since its wavelength will also fit in the path, except that it will reach the source on the second cycle, third cycle, and so on.
Depending on how the standing waves are generated, we can talk about:
Axial mode: when the wave is generated between two surfaces.
Tangential mode: the result of reflection off four surfaces.
Oblique mode: formed by reflection off all six surfaces.
Since the wave loses part of its energy at each reflection, axial modes are the strongest, followed by tangential modes with approximately half the energy (-3 dB), and finally oblique modes with about 6 dB less than the axial ones.
If instead of a single tone we emit a more complex program (for example music), the number of standing waves increases. The set of all of them is called the room modes of the space, and their presence causes a coloration of the spectrum, since some frequencies are reinforced and others are not.
If we look at the graphic representation of two standing waves between two parallel walls (Figure 2), we will see that another disadvantage is that, for those frequencies, points of maximum amplitude (called antinodes) and points of zero amplitude (nodes) are created in the room. In other words, the coloration also depends on where in the room we are positioned.
At the boundaries of the room there are always amplitude maxima, especially in the corners (again in figure two, we can see that, regardless of the number of antinodes, they are distributed in such a way that there are maxima next to the walls).
That is why, as far as possible, we should avoid placing the monitors in the corners of the room or right up against the wall since, although this will apparently improve the speakers’ response (especially at low frequencies), the reinforcement is not the same for all frequencies, and the result is that the monitors’ response is altered.
Calculating room modes
Let’s suppose a room 3.45 m wide, 2.90 m long and 2.50 m high, such as the control room of a home studio. The first three axial modes would be: (Table 1)
Mode f(100) would form between the left and right walls of the room, f(010) between the front and back walls, and f(001) between the floor and the ceiling. Below we show a table with the first axial modes:
Note: we continue to use c = 340 m/s so as not to complicate the calculations. If we want to be more precise, we should measure the room temperature to calculate the speed of sound, although in reality the small variations we would get in the values above are practically negligible in the case at hand, because of the margins within which acoustic materials operate.
The rest of the modes would be calculated in a similar way; tangential modes are obtained from combinations in which only one of the coefficients is zero (f(110), f(101), f(210), etc.), and oblique modes when none of them is zero (f(111), f(112), f(213), etc.).
We should look for possible problems in the room: isolated frequency reinforcements (room modes very close together, especially axial ones) or large gaps with no reinforcement. In general this room enjoys good spectral spacing, except perhaps for the zone between 70 and 90 Hz, which has only oblique modes, and a small reinforcement around 295 Hz.
An example of a worse distribution would be that of figure 4, corresponding to a room of 2.90 m x 3.10 m x 2.80 m, in which we can have resonances at 60, 120 and 175 Hz, and also significant gaps between these points, especially between the first two.
This is due to the relationship between the room’s dimensions, which form an almost perfect square. The worst possible case would be a space of this shape (for example 3 m x 3 m x 3 m), since the standing waves coincide in frequency, creating what are called degenerate modes, in which two or more modes generate the same frequency, such as f(100) = f(010) = f(001) = 56.57 Hz in the square room. Likewise, if the dimensions are multiples of each other (imagine a room with a floor plan of 2.5 m x 5 m), we will also find degenerate modes.
Problems arising from room modes are usually more evident in small rooms and at low frequencies. This is because, on the one hand, the larger the room, the lower the frequency at which the first modes occur, falling in less sensitive zones of our hearing.
On the other hand, as modes increase their density also increases (as we can see in the previous figures), making the reinforcement at frequencies above 200-300 Hz more even (and therefore less harmful). That is why the study of room modes is often limited to frequencies below these values.
Logically, these calculations are especially useful before building the room, since they let us predict possible problems and thus look for a ratio between the room’s dimensions that provides good spacing of the room modes.
However, there are many cases in which the room is already built and we have to alleviate the problems caused by standing waves. As we will see in the next issue, a series of specific materials is used for this, but before thinking about acoustic treatment we have to decide where to place the monitors, since their position influences the room’s behavior.
Monitor placement
There are a number of considerations worth keeping in mind when placing the monitors:
In a control room, in general, you should aim for symmetry. The monitors and the listening position should form an equilateral triangle, where the distance between the monitors (measured, if possible, from one tweeter to the other) should be identical to the distance from each of them to the listening position. Any shift of the listener off the center plane between the two monitors will make the stereo image move toward the side of the plane we are on.
Likewise, that center plane should be located exactly halfway between the left and right walls, so that the first reflections from them reach the listening position equally.
The tweeters should be at ear height, aimed at the ears. We may need to toe in the monitors and/or tilt them slightly by placing some wedge-shaped material underneath, preferably one that prevents the transmission of sound to the surface the monitors rest on (we will look at some in the next issue).
As we have seen, you should avoid placing the monitors in the corners to avoid altering their response, and for the same reason we should keep some distance from the rear wall (50 cm or more is ideal). Try pulling them out and you will immediately notice the change in their response and an improvement in the sense of depth and stereo image.
Care must be taken when we place the speakers on top of the meter bridge of the mixing console. The first reflections coming from the highly reflective surface of the desk can cause comb filtering at the listening position. It is advisable to place them behind the desk, even if that means buying some stands.
If we have a subwoofer, although ideally the low frequencies it radiates are omnidirectional, in practice that is not the case, and we can almost always tell where the bass is coming from. A good placement, therefore, may again be on the center plane between the L and R speakers, raising it slightly rather than placing it directly on the floor.
Measurements
Once we have decided on the final position of the monitors (it helps to use a tape measure to take exact measurements), we should check the room’s actual response, since the placement of the speakers influences the room’s behavior. In addition, other factors such as the wall material, the layout of the equipment, furniture and decorative items, etc. can make the distribution of the waves not what was expected, solving problems that in theory should have been there (or creating unexpected new ones).
The ideal would be to use an omnidirectional source with a signal generator to emit pure tones and pink noise, and use a measurement microphone along with a spectrum analyzer to check the response at different points in the room.
If we don’t have these tools, we can use a signal generator plugin with our studio monitors as the source, a good condenser mic (in omni pattern) and an RTA (real-time analyzer) plugin for the evaluation. If our editing software doesn’t include an RTA, we can download Inspector from Elemental Audio (www.elementalaudio.com). There is a free version for PC and Mac, in VST, RTAS and Audio Unit formats.
The first place we will put the microphone is, logically, the listening position. Pink noise will give us a general idea of the room’s response, while emitting pure tones in each of the third-octave bands lets us focus on specific parts of the spectrum. The idea is to go through all the third-octave bands and check whether there is any frequency at which the level picked up by the mic is unusually high or low.
If we walk around the room at each tone, we can check the location of the nodes and antinodes, and how their number increases as we go up in frequency.
Because of this dependence of the response on position in the room, it is worth taking measurements at points other than the listening position, to get to know our control room’s behavior more intimately.
NOTE: logically, the response of the monitors and mic we use will itself condition the measurement results. For example, our monitors may not be able to reproduce any tone below 40 Hz. As we said at the start of the article, hiring a professional with the right equipment will always be the best option if we intend to do a serious study of the room.
Conclusions
The placement and arrangement of the monitors are a key point in establishing a reliable listening environment. Even though the calculation of room modes can be carried out theoretically, the place where we put the speakers will determine the room’s final behavior, so it is essential to take empirical measurements to determine its possible acoustic problems.
Once we have checked the room’s response, the next step is to design the appropriate acoustic treatment to correct any deficiencies and improve the sound field. In the next issue we will analyze the different types of acoustic materials and their use, and see how to decide on the appropriate placement. We will also present other problems such as flutter echo (and how to fix it), and, following the same line as this article, we will see that you don’t need to make a big investment to enjoy a good control room.
