Manufacturer's Specifications:
Enclosure Type: Sealed.
Drivers: 6 ½” (170-mm) long-throw woofer; 3/4-in. (19-mm) ferrofluid
Frequency Response: 68 Hz to 20 kHz, ±3 dB.
Nominal Impedance: 8 ohms.
Crossover Frequency: 3.5 kHz.
Recommended Amplifier Power: 5 to 40 watts.
Cabinet Finish: Wood-grain vinyl.
Dimensions: 13 1/2 in. (343 mm) x 814 in. (210 mm) x 73/4 in. (197 mm).
Weight: 9 pounds (4.1 kg).
Price: $75.00 each.
Company Address: 130 Condor St., East Boston, Mass. 02128.
The Boston Acoustics Model A40 is a diminutive loud speaker system capable of high performance at a very low cost. A two-way system, the A40 uses a 170-mm (6 1/2-inch) long-throw woofer and a 19-mm (3/4-inch) Ferrofluid-cooled tweeter, packaged in a box whose longest dimension is only a bit over a foot, at 343 mm.
Weighing a mere 4.1 kg (9 pounds), the cabinet is finished on four sides in wood-grain vinyl. A snap-on grille protects the loudspeakers from inquisitive fingers, and electrical connections are made to spring-loaded clips mounted in a recessed cavity on the rear of the enclosure. Polarity is clearly indicated, both by color coding and symbols adjacent to the connectors. No difficulty should be experienced in hookup, since there are no equalizer controls on the loudspeaker and its small size makes experimentation for optimum location within a room much easier than with a larger system.
The sides of the enclosure are smooth, and, if placed on overhead shelves, care should be exercised so that inquisitive toddlers do not harm themselves by pulling these light weight boxes down. Boston supplies a set of adhesive backed rubber feet or pads to prevent the speakers from sliding off the shelf, though (as with all small speakers) a lip on the shelf is advisable. Tipping is not a problem if the speakers are sandwiched closely between shelves; other wise, a wire down the shelf back will help.
Measurements
The impedance which the Boston Acoustics A40 presents to a power amplifier is shown in Fig. 1. Although rated as a nominal 8-ohm system, the measurement indicates impedance dips as low as 5 ohms. Since this occurs in the octave around Middle C, I considered this system to be a 4-ohm load. This is borne out in the complex impedance plot, Fig. 2, in which a maximum lagging phase occurs at around 100 Hz with a magnitude of 9 ohms and angle of 23°. Treating the A40 as a 4-ohm system, and using larger size hookup wire in short runs from the amplifier, will also prevent the impedance variations from modifying the frequency response due to line drop. With that in mind, the A40 presents no unusual impedance load to the power amplifier. The complex impedance plot is devoid of spurious terms, indicating a tight and well-constructed enclosure.
The anechoic frequency response is plotted in Figs. 3 and 4. Taken at one meter, directly on axis, these measurements are made using a drive voltage corresponding to one aver age watt into 4 ohms, or 2 V rms. Boston Acoustics rates the sensitivity as 88.5 dB at one meter and 2.83 V rms. On average, I measure 0.5 dB higher SPL, correcting for the difference in drive voltage. The A40 clearly meets specification in this regard.
The frequency-amplitude response, Fig. 3, shows an unusually smooth on-axis response, commencing at a low-frequency cutoff of 75 Hz and progressing to 20 kHz.
Boston Acoustics rates the response at 68 Hz to 20 kHz, ±3 dB. Again, they are a bit conservative relative to my measurements, which put the range as 65 Hz to 20 kHz, ±3 dB.
The frequency-phase response, Fig. 4, is corrected for two effective acoustic positions. The mean average delay, at a one-meter on-axis location, for the woofer is 3.1501 mS, while the delay for the tweeter is 2.9725 mS. The two drivers differ in arrival time by 0.1776 mS (that's the spirit), and the actual acoustic crossover occurs around 5 kHz. The woofer is polarized at 00; a positive-going voltage applied to the " +" terminal produces a positive increase in sound pres sure (when the speed of sound is corrected from the measurement). The tweeter has a 90° phase shift (when corrected for time delay). The audibility of this phase shift has not yet been established, but the effect is to produce a sound pulse which, for the tweeter, is the Hilbert transform of that for a zero-phase reproducer. In this case, the first 0.15 mS of impulse arrival (corresponding to the response above 7 kHz) has a sine shape, as distinct from the raised cosine which a 0° phase tweeter might have.
The three-meter room test, Fig. 5, also indicates an extremely smooth response. In this case the A40 was mounted 790 mm (31 inches) above a carpeted floor and 75 mm (3 inches) in front of a hard wall. The microphone was positioned at normal listening distance, three meters away from the speaker and one meter above the floor (standard in this test). No articles of furniture, with the exception of the speaker stand, were placed nearer than one meter on either side of the A40. Figure 5 is the measured frequency spectrum of the first 13 mS of sound which arrives at the listening position. Two measurements were performed, and the plots displaced 10 dB on this scale in order to provide clarity of presentation.
Fig. 1--Impedance.
Fig. 2--Complex or polar impedance.
Fig. 3--Frequency amplitude response measured at one meter directly on axis
with a constant drive voltage corresponding to one average watt into 4 ohms.
Sitting right in front of the A40 produces the upper plot, while sitting 30° off axis, as if the system were the left channel of a stereo pair, produces the lower plot in Fig. 5. If the two curves had not been displaced 10 dB, they would virtually fall on top of each other. This speaker has an off-axis response that is almost the same as the on-axis response, decibel-for-decibel. The dips at 300 Hz, 900 Hz, 1.5 kHz, etc. are due to sound reflecting off the carpet and ceiling and arriving about 0.6 mS after the direct sound.
These reflections disappear above 4 kHz, where the slightly higher-directivity tweeter carries the information. As a lesson in physics, a perfect omnidirectional loudspeaker would reinforce at d.c., 600 Hz, and 1.2 kHz under this geometry and dip at 300 Hz, 900 Hz and 1.5 kHz. Taking this into account, the room response of the A40 is exceedingly good.
The horizontal and vertical polar energy responses for this speaker are plotted in Figs. 6 and 7. These are true energy responses in which the information plotted is the integral of the square of the amplitude of the impulse response for a perfect band-limited signal extending from 20 Hz to 20 kHz.
The horizontal polar energy dispersion is uniform within ±30° from the front axis position, and so is the vertical response, with the exception of the slight upward projection of sound energy which is characteristic of speakers with this frontal configuration of drivers. The wide lateral dispersion indicates excellent stereo imaging capability. The wide vertical dispersion, on the other hand, can be a mixed blessing. Being uniform in the vertical plane means that one gets the same sense of stereo illusion whether standing or sitting, an extremely good feature, but it does cause significant energy to be radiated toward the floor and ceiling. Thus the A40s should never be placed where any hard reflecting surface, such as an overhanging shelf, could cast early reflections back into the principal listening area. If that is not possible, then such surfaces should be covered with acoustically absorbent materials in order to preserve the excellent imaging properties of this speaker (even a doubled-up towel will do in a pinch).
Fig. 4-Phase-frequency response, measured under the same conditions as Fig.
3, with the tweeter corrected for a time delay of 2.9725 mS and the woofer
3.1501 mS.
Fig. 5-Three-meter room response.
Harmonic distortion measurements for the tones of E1 (41 Hz), A2 (110 Hz), and A4 (440 Hz) are plotted in Fig. 8. Not surprisingly, E1, which lies an octave below cutoff, is difficult for the A40 but only at high drive levels. During the earlier listening test, I was struck by the fact that the A40 did an excellent job on wide-range material reproduced at brisk level. To be sure, the deep bass is not there, but when I began to drive it really hard, the A40 did not cave in where I knew deep bass to be present. The data of Fig. 8 show why.
The distortion for E1 rises uniformly with drive level, and the system does not show acoustic distress until levels of about 10 average watts (into the assumed 4 ohms). The combination of a uniform increase in distortion with drive level, and virtually the same ratio of second to third harmonic over most of the usable range, tends to prevent the system from becoming annoying in its harmonic structure. By all means, the cleanest reproduction will result when deep bass notes are kept away from the A40, but, as this measurement shows, the system does a creditable job of handling what ever you give it. Tones of A2 and A4 produce moderately low distortion, below 1.5%, up to an acoustic overload point of around 80 average watts. It must be remembered that the data of Fig. 8 are from a burst measurement technique intended to determine how the loudspeaker reproduces momentary bursts of high-energy signal. Any steady-state measurements at these levels would soon fry the driver.
The intermodulation distortion produced on a tone of A4 (440 Hz) by a low E1 (41.2 Hz), mixed one-to-one, is plotted in Fig. 9. Again, although the low-frequency cutoff is well above 41 Hz, the listening test indicated that low bass simply did not muddy up orchestral passages which were reproduced at high level. I would normally have expected a significant amount of mud due to low bass, which was moving the same cone that reproduces everything below about 5 kHz. It did not seem to happen, and, again, the measurement confirms the ears. The test signal has an equal mix of E1 and A4, and the indicated power level is such that a peak drive voltage of 2.83 V corresponds to one average watt into the assumed 4-ohm load. The IM is computed as the combined energy in the sidebands around A4, caused by E1, and plotted as a percentage of the A4 level.
The values of IM produced by the A40 would be acceptable for some of the better and more-ambitious wide-range loud- speaker systems. Considering the fact that the A40 uses a 170-mm (61/2-inch) woofer, the values are startlingly good.
The nature of IM is principally phase modulation at lower levels, with only a small amount of amplitude modulation.
This begins to change above 10 average watts. At 40 average watts, the IM consists of 8° peak-to-peak phase modulation of 440 Hz by 41 Hz and 3% peak-to-peak amplitude modulation. At 100 average watts (remember, this is a short-duration burst test), there is 9° peak-to-peak phase modulation and 8% peak-to-peak amplitude modulation. There does exist a measurable effect which I did not pick up during the earlier listening test. As power level is increased, the mean average acoustic center of the A4 tone moves toward the listener; this amounts to 4.5° phase shift of the A4 tone at 100 average watts, or about 1 centimeter of spatial advance.
The test for acoustic transfer gain and the crescendo test both show the A40 to be an excellent performer. The indication is that the stereo illusion should remain stable under wide surges in orchestral dynamics. Up to 10 average watts, the tones of Middle C and A4 remain within 0.05 dB of their proper level, while A2 dropped 0.2 dB. Inner musical voices of A2 and A4 remained within 0.1 dB of their proper level when wideband noise of 20 dB higher energy level was superimposed, even up to peak levels corresponding to 60 watts.
The energy-time curve for the A40 is plotted in Fig. 10.
The initial peak of energy, due to a Hamming-weighted 20 Hz to 20 kHz band-limited pulse, arrives at 2.97 mS. Following an initial reverberation decay at a rate of about 110 dB per millisecond, subsequent arrivals stay at least 30 dB below the main peak. All in all, this indicates an excellent impulse response, with the majority of energy arriving within 0.3 mS.
Fig. 6-Horizontal-plane polar-energy response.
Fig. 7-Vertical-plane polar-energy response.
Fig. 8-Harmonic distortion for the tones E1 (41.2 Hz), A2 (110 Hz), and
A4 (440 Hz).
Fig. 9--Intermodulation by E1 (41.2 Hz) when distortion of A4 (440 Hz) mixed
in one-to-one ratio
Fig. 10--Energy-time curve.
Use and Listening Tests
The information supplied with the Boston Acoustics A40 needs more recommendations for achieving optimum placement. As I soon found out, the A40 has a very wide dispersion of sound which produces unpleasant tonal notches and nasal properties when placed next to reflecting surfaces. When positioned away from such surfaces, the sound is clear and remarkably wide in frequency range for such a diminutive enclosure. Most of the listening tests were done with the A40s raised 780 mm (31 inches) above the floor, which placed the center of the enclosure at about ear level when seated.
As claimed, the dispersion is quite good, and there is no essential audible difference produced when the speakers are rotated directly toward the listening area or pointed forward, which places the stereo listening position about 30° off axis. All of the listening tests were done, however, with the speakers canted toward the listening area.
As might be expected, the super low bass is absent. But bass balance is such that response is uniform down to about two octaves below Middle C, then it drops smoothly with no obvious bumps or dips. This is not a loudspeaker whose low end should be brought up with tone controls. It should be played flat. The midrange has a few dips (which the three-meter room test later showed to be due to interference reflections), and the top end is quite smooth and extended in its frequency range. I could not hear any difference with or without the grille assembly, so I left the grille in place. As far as wide-range tonal balance is concerned, this is a prime candidate for augmented bass using a sub-woofer. Because the top end goes right on out, the lack of low bass seemed, to me, to create a tonal imbalance on some wide-range material. This subjective effect was some what relieved by actually pulling down the extreme high end with equalization. Normally, I preferred the flat position for most material.
Stereo imaging is excellent. The illusions of both lateralization and depth are accurately conveyed, particularly at higher levels. I was pleasantly surprised to find that, al though rated for a maximum of only 40 watts, the system can handle very brisk levels without stress or audible break up. At higher SPLs, the A40's ability to handle peaks gave the impression of a much wider frequency range than it was actually producing. In other words, after long periods of listening only to the A40s, I accommodated to the response and did not sense the lack of low frequencies on orchestral peaks. Only when I switched to a wider range system was the loss of bass apparent. This is a good example of Catastrophe Theory in action. That the A40 did not crunch or evidence other problems in the bass region under wide dynamic conditions allows this accommodation to occur.
Piano was moderately good, though not the very best I have heard from loudspeakers. And while the A40s do an excellent job of reproducing male voices, female voices, particularly choral groups, were not as realistically reproduced to my ears.
If this loudspeaker were to sell for a few hundred dollars, I would be tempted to say that it is a moderately good bargain; considering its actual price, I think it is an extraordinarily good one.
-Richard C. Heyser
(Audio magazine, July, 1983)
Also see:
Boston Acoustics T1030 Speaker (Jan. 1991)
Avid Model 102 Speaker System (Sept. 1975)
Cambridge Physics Model 310 Loudspeaker (Nov. 1982)
Black Acoustics Night Speaker (Aug. 1982)
Boston Acoustics MC-1 vdH Phono Cartridge (Feb. 1984)
= = = =