When setting up an audio system, or perhaps installing a new speaker system, anticipation of the pleasing results to come sometimes precludes proper setup diligence. When this happens it is perhaps traceable to unfamiliarity with what is needed, and it can persist years after the errors are made, covertly robbing the listener of all the potential his system actually has. The exacting demands of proper speaker and listening position geometry within the room would surprise many audiophiles who are used to treating this subject in approximations. The concerns of this paper include speaker and listening position placement as well as some of the reasons for their criticality. There are no formulas for getting this exactly right the first time, but there are some truly objective rules that, if followed carefully, will yield the best results for a given experiment. These rules will also provide a decent starting point around which experimentation can proceed more closely to the final optimal solution. Speakers are often very heavy objects that make experimentation painful and inconvenient. Not only do we want to get the job done right with a minimum of effort, we want to avoid never achieving success because the experiments are becoming prohibitively odious.
We will start by considering some boundary conditions that are most common to the problem. Later-on we will add commentary on some less common conditions and how they affect the final solution.
Listening Arrangement
We begin by defining the Listening Arrangement as the combination of two speakers and one listening seat. Let us assume that the listening room is a rectangle with dimensions that necessitate that the Listening Arrangement be on-axis with the room axis, i.e., square with the room, whereas a sufficiently large room could facilitate a listening arrangement at any arbitrary alignment angle. The first decision to make is whether to align the Listening Arrangement to the major axis or to the minor axis of the room. For instance, if aligned to the major axis, the room is relatively deep and narrow. Otherwise, it is relatively shallow and wide. These two scenarios are depicted below:
Major Axis Alignment
The key concern here is reflections off the Front Wall and the Rear Wall. These have everything to do with bass response. The Minor Axis Alignment has the drawback of intensifying these reflections and limiting the distances we can get things from these walls, so that generally the bass response in this alignment is worse. There is arguably an advantage to soundstage because the Right Wall and Left Wall are further out of the picture, but Rear Wall/Front Wall reflections also degrade sound stage. Usually, the price to bass is too severe to make this a favored alignment. Hence, most people tend to select the Major Axis Alignment. Again, we are assuming a room of ordinary, not extra-ordinary, dimensions.
Minor Axis Alignment
Distance
The next question has to do with the distance of the speakers from the Front Wall, and the distance of the listener from the Rear Wall. We will begin with the assumption that the speakers are traditional box variety and not of the dipole (planar) variety. The Front Wall reflection will produce a second wavefront that is delayed by 2df/v, where df is the distance from the speaker front surface to the Front Wall, and v is the propagation velocity of sound (about 1100 ft/sec). The result of this is a dip in response, sometimes referred to as a suckout, at a particular frequency Dip Freq = v/(4df). This is the frequency at which the sound wave reflected by the Front Wall is exactly 180 degrees out of phase with the incident wave emerging from the speaker, resulting in cancellation of sound pressure. Additionally, there is a peak frequency at twice the dip frequency (360 degrees phase with constructive interference). The same arguments can be made regarding reflection from the Rear Wall. The table below shows the dip and peak frequencies as a function of various distances (whether speaker or listener) from a front or rear wall, respectively. A brief look at the table will convince you that something strategic is possible. If the distance from the listener to the Rear Wall (dr) is twice the distance of the speakers to the Front Wall (df), then the peak from the Rear Wall will mitigate the dip from the Front Wall. Alternatively stated, the loss of sound pressure due to the Front Wall reflection is partially mitigated by a gain in sound pressure due to Rear Wall reflection. Another observation is that when these two distances are identical then the dips from each of the two walls are mutually reenforcing, a very bad situation. Part of the art of speaker placement is using the room to cancel its own evils.
| d | Dip Freq | Peak Freq |
| 2’ | 138 Hz | 276 Hz |
| 2.5’ | 110 Hz | 220 Hz |
| 3’ | 92 Hz | 184 Hz |
| 3.5’ | 79 Hz | 158 Hz |
| 4’ | 69 Hz | 138 Hz |
| 4.5’ | 61 Hz | 122 Hz |
| 5’ | 55 Hz | 110 Hz |
| 5.5’ | 50 Hz | 100 Hz |
| 6’ | 46 Hz | 92 Hz |
| 6.5’ | 42 Hz | 84 Hz |
| 7’ | 39 Hz | 78 Hz |
| 7.5’ | 37 Hz | 74 Hz |
| 8’ | 34 Hz | 68 Hz |
| 8.5’ | 32 Hz | 64 Hz |
| 9’ | 31 Hz | 62 Hz |
| 9.5’ | 29 Hz | 58 Hz |
| 10’ | 28 Hz | 56 Hz |
Using the table for a case-in-point example, one might set the speakers 5’ in front of the Front Wall, and place the listening position 10’ in front of the Rear Wall. The Front Wall dip occurs at 55 Hz, while the Rear Wall peak occurs at 56 Hz, nearly the same frequency. The cancellation technique referred to above is only partial. For this reason, it is best not to allow a bass dip that is too low in frequency because it may give the impression of thin bass, even if the rear wall distance is used strategically to help this. It turns out to be the case that for modest room dimensions, a good starting point for speaker to Front Wall distance is about 5’. Smaller distances can take their toll on sound stage depth, while longer distances can result in thin bass. Of course, experimentation must be performed, but this is a good place to start. Short distances from the listener to the Rear Wall are usually not favored because the Rear Wall reflection can disturb the development of the sound stage.
Putting this together, the distance df is limited by room size, as it requires dr = 2df to control bass response. On the other hand, if df is too small, the sound stage suffers. That is why a happy medium in the neighborhood of 5 feet makes a smart starting point in most rooms of ordinary size. When the room is larger than ordinary, these distances can both be enlarged, and the sound stage will usually thank you for it. It may be tempting to think that all of this can be disregarded by using room treatment to kill reflections. While treatment devices can work very well for midrange and high frequencies, they tend to be very ineffective in the bass region. However well one might be able to manage reflections, he still does very well to make the room his friend and not his enemy.
Speaker-Listener Positioning
Now, a word about speaker-listener positioning is in order. Ideally, the speakers are placed so that they fall on a circle whose center is the listening position, meaning that the speakers are equidistant from the listener. This is because the concept of stereophonic sound requires that the amplitude (loudness) and time delay from each speaker to each respective ear of the listener be the same. This makes intuitive sense. It is also imperative that the listener is seated directly between the speakers, without any lateral offset. Unless this condition is met, the center stage, where a singer or featured instrument is located, will seem to be shifted to the side. Any compromises to these assumed conditions will result in sound stage skewing and shrinkage. The diagram below depicts this ideal scenario:
The following shows an error where the speakers are equidistant from the listener, but not centered around the listener. This will move the center stage toward the right:
The next diagram shows the speakers centered around the listener, but not equidistant. This will move the center stage toward the left and will shrink sound stage size as well:
The next diagram shows errors in lateral positioning and distance from listener. The sound stage is both compressed and offset:
The next diagram shows correct lateral position and equidistance, but is skewed within the listening room. The sound stage is asymmetrical, developing more fully on the left side than on the right side, because inadvertent side wall reflections tend to stop the development of sound stage “bloom”:
Another factor to consider is distance of listening position from speakers. The speakers and the listening position form a triangle (see diagram below).
The closest position for the listener is often taken to be that which forms an equilateral triangle, whereas the furthest position is taken to be twice that which results in an equilateral triangle. When the listener is too close, the center stage is often diffused or seems to be located within the listener’s head (the headphone effect). When the listener is too far away, the sound stage shrinks width. It is easy to hear this effect just by moving forward and away in a swatted position, or by sliding a chair around. Clearly, there is constraint between this consideration and bass response spoken of in the preceding section. The art of getting all of this right is an art of compromise. The smaller the listening room, the more constrained these competing factors are, and the more demanding it is to make very wise choices in achieving this compromise.
A further consideration with regards to listening position proximity to the speakers, is that the closer this distance, the less emphasis the room reflections have. They are overpowered by the direct incident waves from the speakers. Therefore, the farther one wants to sit from the speakers, for any given room, the greater the onus to properly treat reflecting surfaces to keep them from damaging the sound stage. When large room dimensions exist, say in excess of 30 feet, the need to treat reflections is greatly reduced in most cases.
Speaker Toe-In
The next subject to discuss is speaker toe-in. The following diagrams show less toe-in and more toe-in than the first diagram:
Less Toe-In
More Toe-In
A good starting point with toe-in is to adjust the angle of the speaker front plate to aim directly at the respective ears of the listener. From there, begin to increase the toe-in, paying attention to possible loss of center stage fill. Alternatively, begin decreasing the toe-in, paying attention to sound stage width collapse. There is a matter of listening room Left Wall and Right Wall interaction here. Increasing the toe-in reduces interaction with the side walls, whereas reducing toe-in increases side wall interaction. This adjustment should be optimized only after sound treating the side walls to absorb reflections, either by hanging sheep skins or placing other room treatment absorbing products. The place to do the treatment on side walls is at the position where a billiard ball struck from the position of the speaker would bounce off the side wall and end up at the listener, and at the elevation of the listener’s ears. This treatment location is shown in the diagrams as green rectangles. The larger the area covered by the treatment, the more complete the benefit will be. Usually, when the side wall reflections have been treated, the optimal toe-in will be very close, if not identically, at the angle at which the speaker directly faces the respective ear. An example to the contrary occurs when the speaker has a “hot” treble, and turning the speaker away from the ear a bit helps soften the problem. This type of correction is usually done by toeing the speakers out a bit, i.e., less toe-in than simply aiming directly at the ears.
Exceptions
This is a good time to mention some of the exceptions to the norm, and how they can be taken into account. The first exception is the case of a small room or a room of more nearly square dimensions. In these cases, wall reflections are prominent and it is hard to develop a large sound stage. One possible solution is to run the Listening System at an off-angle with the room. This helps keep the listener from being in the path of standing wave reflections between opposite walls. Admittedly, the cost to décor might end a long-standing relationship. Where décor cannot be sacrificed, choosing a listening position that is near the speakers, i.e., shrinking the whole Listening Arrangement, can help suppress the acoustic reflections of the walls that are otherwise too close to silence.
Another exception is the planar or dipole speaker. Such speakers radiate out the back as well as the front, and the rear radiation is 180 degrees out-of-phase with the front radiations. In this case the Front Wall reflection dip and peak frequencies are interchanged in the table above. Where a traditional speaker has a Front Wall dip, the dipole has a Front Wall peak, and for the Rear Wall reflections the dip and peak frequencies are unchanged in the table. Since the rear radiation is out-of-phase with the front radiation, one cannot run dipoles very near the Front Wall under any circumstances, without killing the bass. The entire bass frequency range is cancelled under these circumstances. This is because the wall reflections are close in time with the front emanating wavefront, and they are polarized to cancel each other. On the other hand, if the speakers are placed too far from the Front Wall, a cancellation or dip can occur right in the middle of the bass spectrum. For instance, referring to the table, if the dipole speaker is placed 10’ from the Front Wall, you have a dip at 56 Hz. You may want to place the listening seat 10’ in front of the Rear Wall to get a compensating Rear Wall peak. The help from this rear wall peak is not always complete enough to fix the problem. Furthermore, it is a large room that will afford the speakers to be placed 10’ in front of the Front Wall. Such a large room may invite placement of the listening seat much farther from the rear wall, perhaps 20’. This will result in very thin bass. A much better result occurs when the speakers are moved to about 5’ from the front wall. The resultant dip frequency is now 100 Hz, which is high enough not to sound as much like a loss of bass. As it turns out, 5’ is a magic distance for speaker placement in front of the Front Wall, for both standard box and dipole speakers. Of course, experimentation around this starting point is required in each case to get to the optimum arrangement, but a good starting point saves a great deal of labor and travail.
Yet another exception is the spacious room with a low ceiling. Rooms of this ilk invite ceiling reflections, which are predominant problems in the treble. There are essentially three ways of treating this problem. The first is to avoid the dome tweeter because of its wide projection angle. The second method is to treat the ceiling with very good sound absorbing material, such as sheep skin or acoustic absorption tiles or wedges. The third method is to layout a small Listening Arrangement. The closer the listener is seated to the speakers, the more favorably the incident radiation from the speakers drowns out the ceiling reflection, and the more acute the projection angle from the speaker that would reflect sound off the ceiling and back to the listener. Here is where the dipole speaker design really shines, because the tweeter portion of a dipole is very often a line speaker (narrow and several feet vertically). Line drivers have inherently small vertical projection angles. Hence, they do not throw much sound at the ceiling anyway.
Perhaps the difficulties involved in optimizing the whole listening setup are a bit repulsive, inviting a nonchalant approach and a forgiving attitude about the whole thing. That is certainly understandable, except that it is not consistent with the vigor in general with which high-end audio is pursued. At the risk of stating what is more easily said than done, if the subject of this paper is treated with the same energy and enthusiasm as went into the curation of all of the playback hardware, the results will be found equally befitting the effort. And, what is more, the impact on the bank account for so doing will be a great deal less painful.
The information contained herein is presented in good faith and with all hopes of helping the reader to find the best equipment for an audio system. Neither the author, nor Bella Sound LLC, assumes any responsibility or liability due to loss on the part of the reader, owing to or believed to be owing to the content of this article, or resulting from or believed to be resulting from an application of the information contained herein, regardless the level of expertise engaged thereto.
