Polk Audio LSi7 loudspeaker Measurements
The Polk LSi7 is of slightly above-average voltage sensitivity, at an estimated 88dB/2.83V/m. The impedance is specified as 4 ohms, but as the plot of impedance magnitude and phase against frequency shows (fig.1), the load remains above 6 ohms for much of the audioband, with a minimum value of 4.2 ohms at 212Hz. However, the electrical phase angle is quite severe in the upper bass, which will add to the drive difficulty. A good amplifier or receiver rated into 4 ohms will work best with the speaker.
Fig.1 Polk LSi7, electrical impedance (solid) and phase (dashed). (2 ohms/vertical div.)
A slight discontinuity between 800Hz and 900Hz in this graph's traces implies the existence of some kind of resonant problem. Examining the vibrational behavior of the LSi7's cabinet walls with an accelerometer revealed a strong mode present on the sidewalls at 656Hz (fig.2), with a second mode detectable on the top panel at 1kHz (not shown). Both resonances are of high Q (Quality Factor), which means they need to be hit with music at their specific frequencies for quite a long time to fully develop, which will work against their audibility. In addition, the higher the frequency of a cabinet wall resonance, the quicker it decays, which again works audibility. But as neither mode lies in the range where the impedance plot indicates a problem, something else must be lying in wait.
Fig.2 Polk LSi7, cumulative spectral-decay plot calculated from the output of an accelerometer fastened to the cabinet's side panel. (MLS driving voltage to speaker, 7.55V; measurement bandwidth, 2kHz.)
Fig.3 shows the individual farfield responses of the woofer and tweeter, along with the nearfield responses of the woofer and the two ports. The crossover between the two drive-units clearly lies at 2.2kHz, with approximately third-order acoustic slopes. The tweeter's output is smooth within its passband, though shelving down slightly above 10kHz. However, the woofer's output features a peak just below 1kHz, which coincides with sharp spikes of energy in the outputs of both ports. (These also have a peak apparent in their output at half the frequency of the big peak.) The responses of the low-frequency radiators are plotted in the ratio of their radiating diameters, which means that the 1"-diameter front-facing port (red trace) has a significantly lower output than the 2"-diameter rear port (blue trace). However, it is the front port that features the highest-level peak, at 900Hz, and I must assume that it is the presence of this internal resonance that gives rise to the coloration Bob Reina noted on female vocals and wind instruments.
Fig.3 Polk LSi7, acoustic crossover on tweeter axis at 50", corrected for microphone response, with the nearfield responses of the woofer (black trace), front port (red), and rear port (blue) plotted below 400Hz, 1kHz, and 950Hz, respectively.
The notch at 72Hz in the woofer's output in fig.3 (black trace) coincides with the maximum output from both ports, as expected from the saddle in the fig.1 magnitude trace at the same frequency. This implies only moderate low-frequency extension, which is confirmed by the trace on the left side of fig.4, which shows the sum of the low-frequency nearfield responses, taking into account both acoustic phase and the different distances of the radiators from a nominal farfield point. Though 3dB of the apparent peak in the LSi7's upper-bass output is in fact an artifact of the nearfield measurement technique (which assumes a hemispherical acoustic environment for the radiators), the rest is real, correlating with BJR feeling the speaker to have some "midbass thickness." Speaker designers will very often go for this kind of reflex alignment to give the listener the impression that a small speaker goes deeper in the bass than it actually does. However, my experience has been that getting this right is quite tricky, and whether the excess upper-bass energy will be perceived as power and extension (good) or boom and thickness (bad) will also depend on the speaker's upper-frequency balance.
Fig.4 Polk LSi7, anechoic response on tweeter axis at 50", averaged across 30 degrees horizontal window and corrected for microphone response, with the complex sum of the nearfield woofer and port responses plotted below 300Hz.
The right-hand trace in fig.4 shows that when the LSi7's tweeter-axis output is averaged across a 30 degrees lateral window, its treble balance is quite flat, though the shelved-down top octave can again be seen. This is because the ring-radiator tweeter has limited dispersion above the mid-treble (fig.5). However, this graph also reveals that the Polk has good, even dispersion below that region. In the vertical plane (fig.6), suckouts develop at extreme off-axis angles, but the balance otherwise remains broadly the same over a ±10 degrees range. The speaker will be fairly uncritical regarding listener ear height, therefore. The tweeter's suppressed off-axis output can also be seen in this graph.
Fig.5 Polk LSi7, lateral response family at 50", normalized to response on listening axis, from back to front: differences in response 90 degrees-5 degrees off-axis on port side of baffle, reference response, differences in response 5 degrees-90 degrees off-axis on tweeter side of baffle.
Fig.6 Polk LSi7, vertical response family at 50", normalized to response on tweeter axis, from back to front: differences in response 45 degrees-5 degrees above axis, reference response, differences in response 5 degrees-45 degrees below axis.
In the time domain, the Polk's farfield step response (fig.7) indicates that both drive-units are connected in positive acoustic polarity, with the tweeter's output leading the woofers by a quarter of a millisecond or so. The speaker was raised well away from any reflecting surfaces for the acoustic measurements, meaning that the ripples in the decay of the woofer's step are intrinsic to the speaker. Not coincidentally, their period of just over a millisecond correlates with the frequency of the spike apparent just below 1kHz in fig.4. Peculiarly, however, the LSi7's cumulative spectral-decay plot (fig.8) indicates only a minor amount of delayed energy in this region, and is actually superbly clean over most of the region shown. Undoubtedly, this correlates with BJR's very positive feelings about the speaker's clarity and presentation of detail, particularly in the region covered by the tweeter.
Fig.7 Polk LSi7, step response on tweeter axis at 50" (5ms time window, 30kHz bandwidth).
Fig.8 Polk LSi7, cumulative spectral-decay plot at 50" (0.15ms risetime).
Overall, while the Polk LSi7 does many things well, its low-frequency alignment appears to be balanced a little too much on the "ripe" side, and that resonant behavior just below 1kHz bothered me.—John Atkinson