WO2013001316A1

WO2013001316A1 - Test arrangement for microphones

Info

Publication number: WO2013001316A1
Application number: PCT/GB2012/051536
Authority: WO
Inventors: Andrew Marchant; John BRYAN-MERRETT
Original assignee: Wolfson Microelectronics Plc
Priority date: 2011-06-30
Filing date: 2012-06-29
Publication date: 2013-01-03

Abstract

An acoustic test apparatus, comprises an input, for receiving electrical signals; a first acoustically sealable enclosure; a second acoustically sealable enclosure; an acoustically sealable test volume, located between the first and second acoustically sealed enclosures; a first electroacoustic transducer, mounted between the first acoustically sealable enclosure and the acoustically sealable test volume; and a second electroacoustic transducer, mounted between the first acoustically sealable enclosure and the acoustically sealable test volume. The input is connected to the first and second electroacoustic transducers such that an electrical signal from the input causes each of the first and second electroacoustic transducers to generate variations in sound pressure in the test volume, and wherein said variations in sound pressure generated by the first and second electroacoustic transducers are in phase with each other.

Description

TEST ARRANGEMENT FOR MICROPHONES

This invention relates to acoustic testing, and in particular to an apparatus and a method for testing audio equipment.

In designing or manufacturing audio equipment containing at least one microphone, it is desirable to be able to test the microphone under conditions that reproduce the circumstances in which the microphone will be exposed and used. In particular, many audio devices, such as handsets, including mobile phones, and headsets, for example for personal listening devices, are used in situations in which there is a high level of low frequency noise. For example, a headset that includes active noise cancellation circuitry must include microphones that are used to generate electrical signals that represent the ambient noise, so that these electrical signals can be processed to generate the noise cancellation signal. Such headsets are commonly used in noise environments in which there are relatively high sound pressure levels at low frequencies, for example on buses or aeroplanes. For example, high sound pressure levels even at subsonic frequencies can cause audible distortions due to their effects on the microphones. A practical example, in which there is a possibility of a large subsonic stimulus that may have an effect on handsets and headsets, is in the case of being in a car with the windows closed and a door is slammed shut.

It is therefore desirable to be able to test devices such as microphones in high sound pressure levels at low frequencies. However, it is difficult to generate these high sound pressure levels in a reproducible way, which still provides an acceptable working environment for the testers. Furthermore, the test apparatus should preferably be portable and relatively inexpensive compared to state-of-the art solutions.

According to an aspect of the present invention, there is provided an acoustic test apparatus, comprising:

an input, for receiving electrical signals;

a first acoustically sealable enclosure;

a second acoustically sealable enclosure;

an acoustically sealable test volume, located between the first and second acoustically sealed enclosures; a first electroacoustic transducer, mounted between the first acoustically sealable enclosure and the acoustically sealable test volume;

a second electroacoustic transducer, mounted between the first acoustically sealable enclosure and the acoustically sealable test volume;

wherein the input is connected to the first and second electroacoustic

transducers such that an electrical signal from the input causes each of the first and second electroacoustic transducers to generate variations in sound pressure in the test volume, and wherein said variations in sound pressure generated by the first and second electroacoustic transducers are in phase with each other.

This has the advantage that the apparatus can produce a high sound pressure level, particularly at low frequencies.

For a better understanding of the present invention, and to show how it may be put into effect, reference will now be made, by way of example, to the accompanying drawings, in which:-

Figure 1 shows an apparatus in accordance with an aspect of the invention; Figure 2 is a cross sectional view through the apparatus of Figure 1 ;

Figure 3 is a cross sectional view through an alternative apparatus in accordance with an aspect of the invention; Figure 4 illustrates the form of a further alternative apparatus in accordance with an aspect of the invention;

Figure 5 illustrates the form of a further alternative apparatus in accordance with an aspect of the invention;

Figure 6 illustrates the form of a further alternative apparatus in accordance with an aspect of the invention;

Figure 7 illustrates the form of a further alternative apparatus in accordance with an aspect of the invention; Figure 8 illustrates the form of a further alternative apparatus in accordance with an aspect of the invention;

Figure 9 is a schematic illustration of a test system using the apparatus of Figures 1 and 2;

Figure 10 illustrates the form of a further alternative apparatus; and Figure 1 1 is a schematic illustration of an alternative form of test system.

Figure 1 shows a test apparatus 10, in the form of a rectangular box 12, which, just for illustrative purposes, might for example have a length in the order of 600mm, a depth in the order of 300mm, and a height in the order of 400mm. Other shapes of box can also be used, as described in more detail below.

In this illustrated embodiment, the box 12 is made of a wood material, for example 18mm thick medium-density fibreboard. As alternatives, a suitable box could be made from steel or fibreglass, for example. The box should be sufficiently rigid to withstand the contained sound pressure thus causing minimal acoustic radiation through flexure of the walls. All of the joints 14, 16 etc at the edges of the box are acoustically sealed. Preferably no acoustic wadding should be inserted into the cavities, as this will tend to damp sounds, and therefore reduce the maximum sound pressure levels that can be achieved.

A hole 18 is formed in the upper surface 20 of the box 12. The hole 18 is shown closed by a removable lid 22. The box 12 and lid 22 are formed such that, with the lid 22 in position, the hole 18 is acoustically sealed. In this illustrated embodiment, the upper surface 20 of the box 12 is provided with four toggle clamps 24, 26, 28, 30, so that the lid 22 can be secured in place with a relatively high force, without requiring significant effort by the user.

The lid 22 is also provided with handles 32a, 32b, so that it can be removed, while the box 12 is provided with handles 34a, 34b, so that it can be carried. The lid 22 also has a hole 36 formed in the centre thereof. Figure 2 is a cross-sectional view through the box 12. As can be seen in Figure 2, the box 12 is divided by two vertical walls 40, 42 into three compartments 44, 46, 48 that, in use, are acoustically separated. A first speaker 50 is mounted to the wall 40 so that its sound-producing front surface 52 projects through a hole 54 in the wall 40 into the central compartment 46, while its rear side 56 is in the first end compartment 44.

Similarly, a second speaker 60 is mounted to the wall 42 so that its sound-producing front surface 62 projects through a hole 64 in the wall 42 into the central compartment 46, while its rear side 66 is in the second end compartment 48. The speaker 60 is matched to the speaker 50, in the sense that it is of the same type and model from the same manufacturer, and preferably from the same manufacturing batch, although precise acoustic matching is not required. The two speakers 50 and 60, i.e the two electro-acoustic transducers, are, in this illustrated example, placed opposite each other on the same axis; the lines of their central axes are coincident.

The speakers 50, 60 are electroacoustic transducers, which convert electrical inputs into movement of sound-producing surfaces. In this case, the speakers 50, 60 are moving coil loudspeakers, but pistons driven by hydraulics, pneumatics or mechanically (via a crank from a rotating shaft) are all possible, amongst other things.

A device to be tested, i.e. the test transducer, can be mounted anywhere in the central compartment 46. For example, a device to be tested can be mounted on the base 68 of the central compartment 46. As shown in Figure 2, however, the device to be tested (which in this case is a microphone 80) is mounted to the underside of the lid 22, close to the hole 36, so that wires from the microphone 80 can run through the hole 36, which can then be acoustically sealed. In order to test other types of device, other shaped boxes and/or lids can be used. For example, in order to test an artificial ear device, a lid can be provided with a suitably sized and shaped hole 36, with the artificial ear device mounted to the lid.

The speakers 50, 60, i.e. the transducers, in this embodiment are 25cm speakers designed for use as woofers or subwoofers. That is, they are designed to have a good low frequency response. In addition, they are designed to operate at relatively high powers (over 200W for example), with relatively large cone excursions (more than 25mm peak-peak for example).

In one embodiment, the dimensions of the box 12 are such that the volume of the central compartment 46 is approximately equal to (preferably within +/- 3% of) the total volume of the two end compartments 44, 48. For example, based on the example dimensions of the box given above, the internal volume of the central compartment 46 might be approximately 30 litres, while the two end compartments 44, 48 might each have volumes of approximately 15 litres. The sound pressures intended to be used in the box 12 approach the theoretical maximum, namely 1 atmosphere +/- ~1

atmosphere (i.e. 0 - 2 atmospheres) in the compartments 44, 46, 48. At these pressures, the air in the compartments 44, 46, 48 behaves in a non-linear way. If the volume of air of the central compartment 46 is approximately equal to the total volume of air of the two end compartments 44, 48, the volume of air that is at a negative pressure is approximately equal to the volume of air that is at a positive pressure, and so the effect of this non-linearity will to some extent cancel out.

The acoustic sealing of the box 12 means that the sound pressure within the box will be maximised, thereby allowing the highest possible efficiency, while restricting the amount of sound leakage from the system, and hence the disturbance or even danger to nearby personnel.

Figures 1 and 2 also show an electrical input 90, for example in the form of an electrical socket or other terminal, such that a source of electrical signals can be connected to the input. Figures 1 and 2 then also show wires 92, 94 from the input 90 to the speakers 50, 60 respectively. The speakers are then wired so that an electrical signal supplied to the input 90 causes the speakers to move in phase with each other. That is, the two speakers 50, 60 each move in such a way as to cause alternate

compression and rarefaction of the air in the central compartment 46, and such that the two speakers tend to cause compression of the air in the central compartment at the same time as each other, and tend to cause rarefaction of the air in the central compartment at the same time as each other.

Figure 3 shows an alternative form of test apparatus 1 10. Specifically, Figure 3 is a cross-sectional view through a box 1 12. As in Figure 2, the box 1 12 is divided by two vertical walls 140, 142 into three compartments 144, 146, 148 that, in use, are acoustically separated.

A first speaker 150 is mounted to the wall 140 so that its sound-producing front surface 152 projects through a hole 154 in the wall 140 into one end compartment 144, while its rear side 156 is in the central compartment 146.

Similarly, a second speaker 160 is mounted to the wall 142 so that its sound-producing front surface 162 projects through a hole 164 in the wall 142 into the other end compartment 148, while its rear side 166 is in the central compartment 148. As in

Figure 2, the speaker 160 is matched to the speaker 150, and the lines of their central axes are coincident.

In this embodiment, the magnets 158, 168 of the speakers 150, 160 are attached to one another in a back-to-back manner using, for example, high strength/high rigidity loudspeaker magnet assembly adhesive, such as Loctite™ 326. Other attachment methods may also be used, such as screw-type fixings and/or clamp-type

arrangements, for example. An adhesive-type attachment provides excellent coupling of the equal and opposite mechanical forces generated by the loudspeakers, ensuring that the minimum of mechanical vibration is transmitted from the speakers 150, 160 to the box 1 12.

As in Figure 2, a device to be tested, i.e. the test transducer, can be mounted anywhere in the central compartment 146. For example, a device to be tested can be mounted on the base 178 of the central compartment 146. As shown in Figure 3, however, the device to be tested (which in this case is a microphone 180) is mounted to the underside of the lid 122, close to the hole 136, so that wires from the microphone 180 can run through the hole 136, which can then be acoustically sealed. The speakers 150, 160, i.e. the transducers, in this embodiment are 25cm speakers designed for use as woofers or subwoofers. That is, they are designed to have a good low frequency response. In addition, they are designed to operate at relatively high powers (over 200W for example), with relatively large cone excursions (more than 25mm peak-peak for example). As described above, the dimensions of the box 1 12 are such that the volume of the central compartment 146 is approximately equal to (preferably within +/- 3% of) the total volume of the two end compartments 144, 148, so that the effect of any non- linearity in the behaviour of the air will to some extent cancel out.

Figure 3 also shows an electrical input 190, for example in the form of an electrical socket or other terminal, such that a source of electrical signals can be connected to the input. Figure 3 then also shows wires 192, 194 from the input 190 to the speakers 150, 160 respectively. The speakers are then wired so that an electrical signal supplied to the input 190 causes the speakers to move in phase with each other. That is, the two speakers 150, 160 each move in such a way as to cause alternate compression and rarefaction of the air in the central compartment 146, and such that the two speakers tend to cause compression of the air in the central compartment at the same time as each other, and tend to cause rarefaction of the air in the central compartment at the same time as each other.

Figures 1 , 2 and 3 show arrangements in which there are two speakers, arranged so that the lines of their central axes are coincident. Various other configurations are possible, in which multiple speakers can be arranged such that they receive electrical signals and then act in phase with each other to cause alternate compression and rarefaction of the air, and hence changes in sound pressure, in a test volume of the apparatus. Figure 4 is a cross-sectional view showing one such configuration of test apparatus 200, in which there are three compartments 202, 204, 206, which are acoustically sealed in use, with two speakers 208, 210. The speakers 208, 210 can be arranged so that their front surfaces both face into the central compartment 202, while their rear surfaces are located within respective side compartments 204, 206, or alternatively the speakers 208, 210 can be arranged so that their rear surfaces both face into the central compartment 202, while their front surfaces are located within respective side compartments 204, 206.

This arrangement does not allow the cancellation of the forces generated by the speakers, but represents a simpler alternative that will be acceptable in some situations. Figure 5 is a cross-sectional view showing an alternative configuration of test apparatus 220, having a generally square cross-sectional shape, in which there are five compartments 222, 224, 226, 228, 230 which are acoustically sealed in use, with four speakers 232, 234, 236, 238. The speakers 232, 234, 236, 238 can be arranged so that their front surfaces all face into the central square shaped compartment 222, while their rear surfaces are located within respective outer compartments 224, 226, 228, 230, or alternatively the speakers 232, 234, 236, 238 can be arranged so that their rear surfaces all face into the central compartment 222, while their front surfaces are located within the respective outer compartments 224, 226, 228, 230. In either case, the axes of the speakers 232, 236 are collinear, as are the axes of the speakers 234, 238, with the axes of the speakers 234, 238 being perpendicular to the axes of the speakers 232, 236. The use of four speakers in this way allows higher sound pressures to be generated, while the symmetrical arrangement also allows balancing of the forces generated by the speakers.

Figure 6 is a cross-sectional view showing a further alternative configuration of test apparatus 250, having a generally square cross-sectional shape, in which there are four compartments 252, 254, 256, 258, which are acoustically sealed in use, with three speakers 260, 262, 264. The speakers 260, 262, 264 can be arranged so that their front surfaces all face into the central triangular compartment 252, while their rear surfaces are located within respective outer compartments 254, 256, 258, or alternatively the speakers 260, 262, 264 can be arranged so that their rear surfaces all face into the central compartment 252, while their front surfaces are located within the respective outer compartments 254, 256, 258. In either case, the axes of the speakers 260, 262, 264 are arranged at 120° to each other. The use of three speakers in this way also allows the generation of higher sound pressures than if two speakers are used, while the symmetrical arrangement also allows balancing of the forces generated by the speakers.

Figure 7 is a cross-sectional view showing a further alternative polygonal configuration of test apparatus 280, having a generally triangular cross-sectional shape, in which there are four compartments 282, 284, 286, 288, which are acoustically sealed in use, with three speakers 290, 292, 294. The speakers 290, 292, 294 can be arranged so that their front surfaces all face into the central triangular compartment 282, while their rear surfaces are located within respective outer compartments 284, 286, 288, or alternatively the speakers 290, 292, 294 can be arranged so that their rear surfaces all face into the central compartment 282, while their front surfaces are located within the respective outer compartments 284, 286, 288. In either case, the axes of the speakers 290, 292, 294 are arranged at 120° to each other.

Again, the use of three speakers in this way allows the generation of higher sound pressures than if two speakers are used, while the symmetrical arrangement also allows balancing of the forces generated by the speakers.

Figure 8 is a cross-sectional view showing an alternative configuration of test apparatus 310, having a generally circular cross-sectional shape, in which there are five compartments 312, 314, 316, 318, 320 which are acoustically sealed in use, with four speakers 232, 234, 236, 238. The speakers 322, 324, 326, 328 can be arranged so that their front surfaces all face into the central circular compartment 312, while their rear surfaces are located within respective outer compartments 314, 316, 318, 320, or alternatively the speakers 322, 324, 326, 328 can be arranged so that their rear surfaces all face into the central compartment 312, while their front surfaces are located within the respective outer compartments 314, 316, 318, 320. In either case, the axes of the speakers 322, 326 are collinear, as are the axes of the speakers 324, 328, with the axes of the speakers 324, 328 being perpendicular to the axes of the speakers 322, 326.

Again, the use of four speakers in this way allows higher sound pressures to be generated, while the symmetrical arrangement also allows balancing of the forces generated by the speakers. Figure 9 shows a test system, using the apparatus of Figures 1 and 2. Exactly equivalent test systems can use the apparatus shown in any of the other figures in an equivalent way.

An electrical sound source 400, such as a signal generator (although a recorded sound can be used as an alternative), is connected to an amplifier 402, which in turn is connected through the input 90 to the speakers 50, 60. The electrical connections to the speakers 50, 60 are in parallel, and in phase. Thus the speakers operate in unison with each other. When a positive electrical signal is applied, the cones of both speakers move forward, compressing the air in the central compartment 46, and thus generating a positive peak in the sound pressure.

Conversely, when a negative electrical signal is applied, the speaker cones move backwards, into their respective end compartments, causing rarefaction of the air in the central compartment 46 and a negative peak in the sound pressure. The matching of the speakers 50, 60 means that the cones move equal amounts, but in opposite directions, and thus impart equal but opposite mechanical forces into the structure of the box 12. This minimises vibration in the structure of the box 12, and thus minimises acoustic radiation from the box.

A wire connects the device under test (DUT), namely the microphone 80 to a measurement system 404.

Thus, the frequency response of the microphone 80 can be characterised, by passing an appropriate signal, for example an electrical signal, of known amplitude and of varying frequency from the source 400 through the amplifier 402 to the speakers 50, 60. Thus the amplitude of the signal in the sealable test volume, i.e. the pressure wave detected by the microphone 80 can then be monitored in the measurement system 404, and compared with the frequency response of a previously tested reference microphone, such reference microphone usually being very accurate. The comparison gives a measure of the frequency response of the microphone 80, and each and every other microphone that subsequently needs to be tested against the reference microphone.

Thus, the apparatus 10 can be used as a test device, for example for determining the frequency response of the microphone 80 at subsonic frequencies, or for determining the response of the microphone 80 to very high sound pressure levels at low frequencies.

As an alternative, the apparatus 10 can be used as a specialist high power subwoofer, by removing the lid 22, and allowing the sounds generated by the speakers 50, 60 to exit from the box 12. The amplifier 402 can be chosen depending on the intended use of the apparatus 10. For example, where the apparatus 10 is to be used as a specialist high power subwoofer, the amplifier 402 preferably has a power rating of 300W or more. In this case, the amplifier 402 preferably has a built-in high pass filter, with a cut-off frequency of 15-25 Hz. Otherwise, high power, low frequency signals can result in large excursions when there is no sealed front cavity 46, causing the drivers to bottom out, and be mechanically damaged as a result. Similarly, a low pass filter with a cut-off frequency of 50 - 200Hz is of benefit to restrict the subwoofer bandwidth to the lowest frequencies.

Where the apparatus 10 is to be used for determining the frequency response of the microphone 80 at subsonic frequencies, it is important for the frequency response of the amplifier 402 to have very low degree of roll-off at the subsonic frequencies of interest. Where the apparatus 10 is to be used for determining the response of the microphone 80 to very high sound pressure levels at low frequencies, the amplifier must be able to output the power required to generate the maximum sound pressure level at which the microphone is to be tested, without significant distortion (clipping). For example, while less than 10W might be needed to reach 140dB, 150dB might require 45W, and 160dB might require more than 500W.

Depending on the tests that are to be made, the apparatus 10 can be adapted accordingly. For example, the central compartment 46 might be made large enough to contain an artificial head, in order to be able to make realistic measurements on handsets. Alternatively, the central compartment 46 might be made smaller than described above, in order to be able to generate a higher sound pressure level and/or to allow a higher upper frequency limit. For different sized boxes used in other embodiments, suitably sized speakers should be selected; i.e. the speakers should be matched to the box size. Embodiments are described herein in which there is a central test volume (indicated by reference numerals 46 in Figure 2, 146 in Figure 3, 202 in Figure 4, 222 in Figure 5, 252 in Figure 6, 282 in Figure 7, and 312 in Figure 8), which can be opened to allow the insertion and removal of devices under test, while the other compartments remain effectively acoustically sealed throughout the operation of the test apparatus.

However, it is equally possible for those other compartments to have removable covers, or the like, to allow insertion and removal of additional devices under test, while being acoustically sealed during the test.

In other embodiments, there is a test volume, in which a device under test can be mounted, but there are no sealed enclosures to either side of the test volume.

Figure 10 shows a first such form of test apparatus 1 10. Specifically, the test apparatus 510 shown in Figure 10 is generally similar to that shown in Figure 3, and features that correspond to features of the apparatus shown in Figure 3 are indicated by the same reference numerals, and will not be described further herein.

The test apparatus 510 has a lid 122 in its upper surface, and two vertical walls 140, 142 which define an acoustically sealable compartment 146. Open end compartments 144, 148 are separated from the central compartment 146 by the walls 140, 142 respectively. In further embodiments, there may be no such end compartments.

A first speaker 150 is mounted to the wall 140 so that its sound-producing front surface 152 projects through a hole 154 in the wall 140, while its rear side 156 is in the central compartment 146.

Similarly, a second speaker 160 is mounted to the wall 142 so that its sound-producing front surface 162 projects through a hole 164 in the wall 142, while its rear side 166 is in the central compartment 148. As in Figure 2, the speaker 160 is matched to the speaker 150, and the lines of their central axes are coincident.

A device to be tested, i.e. the test transducer, can be mounted anywhere in the central compartment 146. For example, a device to be tested can be mounted on the base 178 of the central compartment 146. As shown in Figure 10, however, the device to be tested (which in this case is a microphone 180) is mounted to the underside of the lid 122, close to the hole 136, so that wires from the microphone 180 can run through the hole 136, which can then be acoustically sealed.

The speakers 150, 160, i.e. the transducers, in this embodiment are 25cm speakers designed for use as woofers or subwoofers. That is, they are designed to have a good low frequency response. In addition, they are designed to operate at relatively high powers (over 200W for example), with relatively large cone excursions (more than 25mm peak-peak for example).

Figure 10 also shows an electrical input 190, for example in the form of an electrical socket or other terminal, such that a source of electrical signals can be connected to the input. Figure 10 then also shows wires 192, 194 from the input 190 to the speakers 150, 160 respectively. The speakers are then wired so that an electrical signal supplied to the input 190 causes the speakers to move in phase with each other. That is, the two speakers 150, 160 each move in such a way as to cause alternate compression and rarefaction of the air in the central compartment 146, and such that the two speakers tend to cause compression of the air in the central compartment at the same time as each other, and tend to cause rarefaction of the air in the central compartment at the same time as each other.

Figure 1 1 shows an alternative test system 520, using a test apparatus 530 which is generally similar to that shown in Figures 1 and 2, except that there are no sealed end compartments in the apparatus.

The system shown in Figure 1 1 thus has many features in common with the system shown in Figure 9, and those common features are indicated by the same reference numerals, and will not be described further herein. Thus, an electrical sound source 400, such as a signal generator (although a recorded sound can be used as an alternative), is connected to an amplifier 402, which in turn is connected through the input 90 to the speakers 50, 60.

The electrical connections to the speakers 50, 60 are in parallel, and in phase. Thus the speakers operate in unison with each other. When a positive electrical signal is applied, the cones of both speakers move forward, compressing the air in the central compartment 46, and thus generating a positive peak in the sound pressure.

Conversely, when a negative electrical signal is applied, the speaker cones move backwards, causing rarefaction of the air in the central compartment 46 and a negative peak in the sound pressure. The matching of the speakers 50, 60 means that the cones move equal amounts, but in opposite directions, and thus impart equal but opposite mechanical forces into the structure of the box 12. This minimises vibration in the structure of the box 12, and thus minimises acoustic radiation from the box.

There is therefore described an apparatus that can be used as an acoustic source, for example for testing devices to determine their performance at low frequencies and/or high sound pressure levels, or as a specialist high power subwoofer.

Claims

1. An acoustic test apparatus, comprising:

an input, for receiving electrical signals;

a first acoustically sealable enclosure;

a second acoustically sealable enclosure;

an acoustically sealable test volume, located between the first and second acoustically sealed enclosures;

a first electroacoustic transducer, mounted between the first acoustically sealable enclosure and the acoustically sealable test volume;

wherein the input is connected to the first and second electroacoustic

2. An acoustic test apparatus as claimed in claim 1 , further comprising:

a third acoustically sealable enclosure;

a third electroacoustic transducer, mounted between the third acoustically sealable enclosure and the acoustically sealable test volume;

wherein the input is connected to the third electroacoustic transducer such that an electrical signal from the input causes the third electroacoustic transducer to generate variations in sound pressure in the test volume, and wherein the variations in sound pressure generated by the third electroacoustic transducer are in phase with the variations in sound pressure generated by the first and second electroacoustic transducers.

3. An acoustic test apparatus as claimed in claim 1 , further comprising:

a fourth acoustically sealable enclosure;

a fourth electroacoustic transducer, mounted between the fourth acoustically sealable enclosure and the acoustically sealable test volume;

wherein the input is connected to the fourth electroacoustic transducer such that an electrical signal from the source causes the fourth electroacoustic transducer to generate variations in sound pressure in the test volume, and wherein the variations in sound pressure generated by the fourth electroacoustic transducer are in phase with the variations in sound pressure generated by the first, second and third

electroacoustic transducers.

4. An acoustic test apparatus as claimed in claim 1 , wherein respective axes of the first and second electroacoustic transducers are collinear.

5. An acoustic test apparatus as claimed in claim 1 or claim 4, wherein respective sound-producing surfaces of the first and second electroacoustic transducers face each other across the test volume.

6. An acoustic test apparatus as claimed in claim 1 or claim 4, wherein the first and second electroacoustic transducers are moving coil loudspeakers, and wherein magnets of the respective first and second electroacoustic transducers are located in the test volume, and are fixed together, such that the sound-producing surfaces of the first and second electroacoustic transducers face into the first and second acoustically sealable enclosures respectively.

7. An acoustic test apparatus as claimed in any of claims 1 to 6, wherein the test volume has a removable cover.

8. An acoustic test apparatus as claimed in claim 7, wherein the cover has at least one acoustically sealable hole therein.

9. An acoustic test apparatus as claimed in claim 7 or 8, comprising means for mounting a device under test to the cover within the test volume.

10. An acoustic test apparatus as claimed in any of claims 1 to 9, wherein at least one acoustically sealable enclosure has a removable cover.

1 1. An acoustic test apparatus as claimed in any of claims 1 to 9, wherein the acoustically sealable enclosures are acoustically sealed.

12. An acoustic test apparatus as claimed in any of claims 1 to 1 1 , wherein the combined volume of the acoustically sealed enclosures is approximately equal to the volume of the acoustically sealable test volume.

13. An acoustic test apparatus as claimed in claim 12, wherein the combined volume of the acoustically sealed enclosures is equal to the volume of the acoustically sealable test volume to within +/- 3%.

14. An acoustic test apparatus as claimed in any of claims 1 to 13, further comprising a source of electrical signals connected to the input.

15. An acoustic test apparatus as claimed in claim 14, wherein the source of electrical signals comprises a signal generator.

16. An acoustic test apparatus as claimed in claim 14 or 15, wherein the source of electrical signals comprises a power amplifier.

17. A method of testing an acoustic device, the method comprising:

locating the acoustic device in the acoustically sealable test volume of an acoustic test apparatus as claimed in any of claims 1 to 16;

applying an electrical signal to the input of the acoustic test apparatus; and recording measurements made by the acoustic device.

18. An acoustic test apparatus, comprising:

first and second loudspeakers, mounted with a test volume between the first and second loudspeakers;

an input, for receiving electrical signals, wherein the input is connected to the first and second loudspeakers such that an electrical signal from the input causes each of the first and second loudspeakers to generate variations in sound pressure in the test volume, and wherein said variations in sound pressure generated by the first and second loudspeakers are in phase with each other; and

means for mounting a device under test in the test volume between the first and second loudspeakers.

19. An acoustic test apparatus as claimed in claim 18, wherein the test volume is enclosed.

20. An acoustic test apparatus as claimed in claim 18, wherein the test volume is acoustically sealable.

21. An acoustic test apparatus as claimed in any of claims 18 to 20, wherein the test volume has an openable cover.

22. An acoustic test apparatus as claimed in claim 21 , wherein the cover is removable.

23. An acoustic test apparatus as claimed in claim 21 or 22, wherein the cover has at least one acoustically sealable hole therein.

24. An acoustic test apparatus as claimed in claim 21 , 22 or 23, comprising means for mounting a device under test to the cover within the test volume.

25. An acoustic test apparatus as claimed in any of claims 18 to 24, wherein the first and second loudspeakers have respective axes, which are co-linear.

26. An acoustic test apparatus as claimed in any of claims 18 to 25, wherein the first and second loudspeakers are moving coil loudspeakers.

27. An acoustic test apparatus as claimed in of claims 18 to 26, wherein the first and second loudspeakers are connected together, such that said variations in sound pressure generated by the first and second loudspeakers generate opposed forces.

28. An acoustic test apparatus as claimed in any of claims 18 to 27, wherein the first and second loudspeakers are mounted facing each other.

29. A method of testing an acoustic device, the method comprising:

locating the acoustic device in the test volume of an acoustic test apparatus as claimed in any of claims 18 to 28;