Burmester B25 loudspeaker Measurements
The Burmester B25's voltage sensitivity is specified as 88dB/2.83V/m. However, my B-weighted estimate on the speaker's tweeter axis came in a little higher, at 89dB(B)/2.83V/m. The speaker's impedance magnitude and electrical phase with the port open are shown in fig.1. Though the magnitude remains above 6 ohms in the upper midrange and treble, it drops to 3 ohms between 75 and 250Hz, with a high capacitive phase angle through much of the bass region, making the speaker a difficult load for the partnering amplifier to drive. The saddle at 41Hz in the magnitude trace suggests that this is the tuning frequency of the rear-firing port; with this blocked with the supplied foam plug, the impedance becomes that of a typical sealed-box design tuned to a low 32Hz (fig.2). All things being equal, leaving the port open will give better upper-bass weight but a faster rate of low-frequency rolloff in-room. Which is preferable will depend on the owner's musical tastes and room acoustics.
Fig.1 Burmester B25, electrical impedance (solid) and phase (dashed) with port open (2 ohms/vertical div.).
Fig.2 Burmester B25, electrical impedance (solid) and phase (dashed) with port plugged (2 ohms/vertical div.).
There is a discontinuity at around 170Hz in the impedance traces, and investigating the enclosure's vibrational behavior with an accelerometer did uncover a fairly strong mode on all surfaces at 172Hz, as well as one a little lower in frequency and level on the front baffle (fig.3). The side and rear walls also flexed a little between 50Hz and 70Hz. I would have expected this behavior to have led to some lower-midrange congestion, and it may well have contributed to LG's finding Chris Jones' Fender bass on Live at Merkin Hall to sound muddy.
Fig.3 Burmester B25, cumulative spectral-decay plot calculated from output of accelerometer fastened to center of front baffle level with woofer (MLS driving voltage to speaker, 7.55V; measurement bandwidth, 2kHz).
The B25's quasi-anechoic farfield response, averaged across a 30° horizontal window on the tweeter axis, is shown in the right portion of fig.4. The regions covered by the midrange unit and tweeter are both extremely flat in their respective passbands, but the tweeter's output is 3dB higher than the midrange's, which explains why my B-weighted estimate of the speaker's sensitivity was slightly higher than specified. The complicated set of traces on the left side of fig.4 shows the individual responses of the midrange unit (blue trace), woofer (blue), and port (green), measured in the nearfield; ie, with the tip of the microphone capsule in close proximity to the radiating surface. Also shown are the complex sums of the individual responses with the port open (black trace below 300Hz) and plugged (red). The crossover between the midrange unit and woofer appears to be set just below 200Hz, with steep rolloffs. With the port open, the woofer's minimum-motion notch occurs at 45Hz, with the port's output peaking in the same region and smoothly rolling off above it. Commendably, no higher-frequency resonances are visible in the port's midrange output. The nearfield measurement technique results in an apparent boost in the summed woofer and port midbass response (black trace), with then a sharp rolloff below 40Hz. With the port plugged, the summed response (red) rolls off at 12dB/octave below 65Hz, with no trace of any nearfield boost, which suggests that the sealed woofer alignment is somewhat overdamped.
Fig.4 Burmester B25, anechoic response on tweeter axis at 50" without grille (black), averaged across 30° horizontal window and corrected for microphone response, with the nearfield responses of midrange unit (magenta), woofer with port open (blue), and port (green), plotted in the ratios of the square roots of their radiating areas below 400Hz, 900Hz, and 300Hz, respectively, and the complex sum of their nearfield responses with port open (black) and plugged (red), plotted below 300Hz.
Turning to the B25's horizontal dispersion (fig.5), the contour lines are smooth and evenly spaced. Interestingly, because of the low crossover frequency between the woofer and midrange unit, the fact that the woofer is mounted on the side of the speaker has no effect on the lower-frequency radiation pattern. (The wavelength of the sound the woofer emits is many times larger than its diameter, meaning that its radiation is omnidirectional within its passband.) The AMT tweeter becomes increasingly directional throughout its passband, presumably because it is recessed in the front baffle, though its dispersion does suddenly broaden again around 15kHz. This increased directivity between 4 and 15kHz will, particularly in smaller and more reverberant rooms, work to some extent against the tweeter level being theoretically too high compared with that of the midrange unit. In the vertical plane (fig.6), the B25 maintains its response over a wide window, with just a hint of a suckout in the upper crossover region appearing in the response 15° below the tweeter axis. The AMT tweeter's vertical dispersion is, limited, however, which is why LG found the sound to change quite considerably when he stood up.
Fig.5 Burmester B25, lateral response family at 50", normalized to response on tweeter axis, from back to front: differences in response 905° off axis on woofer side, reference response, differences in response 590° off axis on side opposite woofer.
Fig.6 Burmester B25, vertical response family at 50", normalized to response on tweeter axis, from back to front: differences in response 155° above axis, reference response, differences in response 515° below axis.
Fig.7 shows how this quasi-anechoic measured performance added up in Larry Greenhill's listening room. The ports were open, and I averaged 20 measurements taken for each speaker individually in a vertical rectangular grid centered on the position of LG's ears in his listening chair, 7' from the speakers. Below 400Hz, the measured response is affected by residual room effects that have not been completely eliminated by the spatial averaging, but you can see that the midbass boost in the summed nearfield response does translate to an excess of in-room energy in the same region. The speakers' output drops rapidly below the port's resonant frequency. Above 400Hz, the B25s' spatially averaged output is superbly flat through 20kHz, not even showing the usual rolloff above 5kHz that results from the increasing absorption of the room furnishings at high frequencies. In LG's room, the AMT tweeter's increasing directivity at higher frequencies doesn't compensate for the slightly exaggerated level on-axis, giving rise to the airy, spacious presentation that so impressed LG.
Fig.7 Burmester B25, spatially averaged, 1/6-octave response in LG's listening room.
Turning to the time domain, the B25's step response (fig.8) indicates that all three drive-units are connected with positive acoustic polarity, this confirmed by looking at their individual outputs. More important, each one's step smoothly leads into that of the next lower in frequency, correlating with the good integration of their outputs seen in the frequency domain. The Burmester's farfield cumulative spectral-decay plot (fig.9) is superbly clean.
Fig.8 Burmester B25, step response on tweeter axis at 50" (5ms time window, 30kHz bandwidth).
Fig.9 Burmester B25, cumulative spectral-decay plot on tweeter axis at 50" (0.15ms risetime).
Like the Burmester B99, reviewed in the June 2002 issue, the B25 offers excellent speaker engineering and equally excellent sound quality. But its cabinet is a little more lively than I would have liked.John Atkinson