PBN Montana SP loudspeaker Measurements

Sidebar 2: Measurements

I estimated the Montana SP's B-weighted sensitivity at 89dB/W/m, almost exactly to specification. As BW noted, it doesn't take many amplifier watts to drive this speaker to satisfying levels. Its impedance (fig.1) is moderately demanding, dropping to 4 ohms through the lower midrange. It remains above 8 ohms for almost all frequencies above 800Hz, however. In the bass, the saddle in the magnitude curve centered on 30Hz reveals the tuning of the large 2" port.

Fig.1 PBN Montana SP, electrical impedance (solid) and phase (dashed) (2 ohms/vertical div.).

Note the wrinkle in the traces around 120Hz. This is indicative of some kind of resonant problem. But when I tried to get a handle on the cabinet's resonant behavior using a simple plastic-tape accelerometer, I only confirmed what BW had already found: The PBN's cabinet is quite nonresonant. I could find a mode at 120Hz near the top of the side panel, but it was well down in level. In fact, the only juicy mode I could find was on the front baffle, just above the upper woofer (fig.2). But at a very high 676Hz, its existence is more of a tribute to the efficacy of the cabinet's bracing and construction than something to be criticized. Something else must be contributing to the impedance anomaly in the upper bass, therefore.

Fig.2 PBN Montana SP, cumulative spectral-decay plot of accelerometer output fastened to front baffle above the upper woofer. (MLS driving voltage to speaker, 7.55V; measurement bandwidth, 2kHz.)

Fig.3 shows the individual responses of tweeter and woofers on the tweeter axis, spliced to the nearfield woofer and port responses. The upper crossover point can be seen to lie at 3.2kHz, with the drive-units reasonably well-behaved in their respective passbands. The woofers have a little cone-breakup spike in their output at 9kHz, but this is suppressed by the crossover. The woofers' minimum-motion point can be seen at 30Hz, coinciding with the port's electrical tuning (fig.1). Note, however, that the port has a much wider passband than is usual, its useful output extending from 20Hz up to 120Hz. Note also that it peaks a little at 120Hz, something that is presumably due to an internal airspace resonance, and which correlates with the impedance peak at the same frequency. There is also some higher-frequency hash present in the port's output, but as the port faces to the speaker's rear, this will have minimal subjective consequences.

Fig.3 PBN Montana SP, individual responses of tweeter and woofers on tweeter axis at 50", corrected for microphone response, with nearfield woofer and port responses plotted below 300Hz and 900Hz, respectively.

The overall sum of these individual responses is shown in fig.4. The Montana SP is evenly balanced, with a very slight downward trend with frequency. The bass, however, is more problematic. The sum of the woofer and port nearfield outputs was calculated allowing for the differences in distance between the woofers and the port from a nominal farfield microphone position, and weighting the output of each in the ratio of the square roots of the respective radiating diameters. It can be seen that the 120Hz peak in the port output is actually out of phase with the woofer output at the same frequency, resulting in a small notch. More important, the port output doesn't just reinforce that of the woofers in the region where they are rolling off—the usual reflex paradigm—but adds to it for an entire octave.

Fig.4 PBN Montana SP, anechoic response on tweeter axis at 50", averaged across 30 degrees horizontal window and corrected for microphone response, with the complex sum of nearfield woofer and port responses plotted below 300Hz.

The result—at least in this measurement, which assumes equal contributions to the farfield response from the weighted port and woofer outputs—is a 6dB peak in the 70Hz region. This can be adjusted or even minimized by playing with each speaker's position in the room. But, as BW found, it does make the Montana SP's intrinsic low-frequency response rather overwhelming, with a tendency for the midbass to be emphasized. And, as BW also noted, the ultimate LF extension is rather limited for what is really quite a large speaker.

Vertically (not shown), the PBN's balance doesn't change appreciably as long as the listener sits with his or her ears between the woofer axes. Stand up, however, and the mid-treble is sucked-out. Laterally, the Montana's off-axis output (fig.5) is well-balanced, with a smooth rolloff in the treble with increasing angle. As BW found, this behavior correlates with excellent stereo imaging. Only the changes in response are shown in this graph, which is why the on-axis response is presented as a straight line. But if you look closely at fig.5 and compare it with fig.4, you can see that regions where the on-axis response has small dips are where the lateral radiation pattern has a slight amount of flare. Given the basic flatness and evenness of the tweeter-axis response, this will result in an in-room balance that is smooth and uncolored.

Fig.5 PBN Montana SP, horizontal response family at 50", normalized to response on tweeter axis, from back to front: differences in response 90 degrees-5 degrees off-axis; reference response; differences in response 5 degrees-90 degrees off-axis.

Turning to the time domain, there are no surprises in the step response (fig.6), while the cumulative spectral-decay plot (fig.7) is impressively clean. Yes, there is a slight step just below 5kHz, but its subjective effect should be mild.

Fig.6 PBN Montana SP, step response on tweeter axis at 50" (5ms time window, 30kHz bandwidth).

Fig.7 PBN Montana SP, cumulative spectral-decay plot at 50" (0.15ms risetime).

Overall, other than the finicky nature of its reflex alignment, which has some of the upper-bass idiosyncrasies of a transmission-line design, the PBN Montana SP's measurements reveal it to be an impressively well-engineered loudspeaker.—John Atkinson

1015 La Mesa Avenue
Spring Valley, CA 91977
(619) 465-6450