ABSTRACT

When the diaphragm of an open-framed driver moves forwards, the compression of the air at the face of the diaphragm is accompanied by a rarefaction at the other side of the diaphragm, and the natural tendency is for the pressure difference to equalise itself by a movement of air around the sides of the driver. At frequencies whose wavelengths are large compared to the circumference of the diaphragm, the equalisation is almost perfectly accomplished, and so almost no sound is radiated. It is therefore necessary to discourage this pressure equalisation if low frequencies are to be radiated. The simplest means of accomplishing this is to mount the loudspeaker in a large, rigid board, or baffle, as shown in Figure 3.1. If the board were to extend in all directions to infinity, it would be a true infinite baffle. It would cause no change in the air loading on each side of the diaphragm, it would exhibit no resonances, it could cause no diffraction, and, with a good quality driver (or drivers) would sound excellent. Unfortunately, its great drawback is that it is a rather impractical concept. The two practical realisations of this idea are the finite baffle, where a

baffle of perhaps a metre square is employed, or the so-called infinite baffle, which is, in fact, a sealed box. The radiation pattern of the finite baffle is shown in Figure 3.2(a). The cancellation around the sides of the extended plane of the driver cause response nulls to the sides, in the direction of the plane of the baffle, resulting in a three-dimensional figure-of-eight pattern in free space. The low frequency cut-off is determined by the size of the baffle. The final rate of low frequency roll-off is 18 dB per octave, but some measures can affect the nature of the entry to the roll-off. Varying the Q of the driver resonance, by mechanical and/or magnetic changes, can yield response shapes such as those shown in Figure 3.2(b). By placing the driver off-centre, the cut-off can be made more gradual due to the distance from the driver to each edge of the baffle being different. Open baffles are rarely used in recording studio control rooms because of the problems of where to site them and how to control the rear radiation, but they find use in listening rooms and domestic high-fidelity systems. In these instances the baffles can be sited somewhat more flexibly than in an equipment-loaded control room, and the loudspeaker and listener positions can usually be found which give good results. Subjectively, open baffles tend to sound very clean and, not surprisingly, open. They are largely free of resonances, so their time-domain responses are limited only by the drivers and the

rooms in which they are placed. The ways in which they couple to the rooms will be discussed in Chapter 7. When mounted on the floor, the solid surface below the open baffle

acts like an acoustic mirror, so a baffle of one square metre placed on the floor behaves like a baffle of two square metres in free space. This enables baffles of practical size to be useful down to frequencies of 40 Hz or below, but the lack of anything other than atmospheric loading on the rear of the diaphragms and poor efficiency of radiation may lead to over-excursion problems with high sound pressure levels at low frequencies. The resonance

frequency of the driver on an open baffle will be that of its free-air resonance. Because the open baffle mounting does not push up the free air resonance of the driver, and the back-pressures are not augmented by any constraint of the air behind the diaphragm, lighter moving assemblies may be used. Driver cooling is also something that poses no problem with open baffles, so power compression problems are rarely encountered. The open baffle, in the hi-fi world, still enjoys a devoted following of aficionados.