Totem Acoustic Forest loudspeaker 2010 Measurements
Unable to do any listening, I continued with the measurements, using the other sample of the pair (SN PM4003). I performed my usual quasi-anechoic measurements using DRA Labs' MLSSA system and a calibrated DPA 4006 microphone. I used my vintage Audio Precision System One for the impedance testing; the magnitude and electrical phase are shown in fig.1.
Fig.1 Totem Acoustic Forest, 2009 sample, electrical impedance (solid) and phase (dashed). (2 ohms/vertical div.)
The traces in this graph are identical to those in the graphs for the 2001 and 2005 samples of the Forest. The overall impedance remains above 8 ohms at almost all frequencies, meaning that the Forest will be very easy for the partnering amplifier to drive, but the traces are disturbed in the midrange by small, sharply defined glitches that indicate the presence of cabinet resonances of some kind. The low-frequency saddle in the impedance-magnitude trace indicates that the 2"-diameter port on the rear panel is tuned to 42Hz, the lowest note of the four-string electric or acoustic bass.
Fig.2 Totem Acoustic Forest, 2009 sample, acoustic crossover on tweeter axis at 50", corrected for microphone response, with nearfield responses of woofer (blue) and port (red) plotted below 300Hz and 800Hz, respectively.
My estimate of the Forest's voltage sensitivity came in at 85dB(B)/2.83V/m, which is within experimental error of the 2005 sample's sensitivity. Fig.2 shows the individual responses of the tweeter (green trace), woofer (blue), and port (red). As with the earlier samples, the tweeter is a couple of dB too hot in absolute level, but also note the enormous peak at 28kHz. This is due to the metal dome's primary "oil-can" resonance, where the center of the dome is moving in the opposite polarity to the edges. While this resonance will not be excited with CD playbackthere is no energy above 20kHz to set it offit might lead to problems with LP playback or with hi-rez digital, both of which contain ultrasonic signal energy. The tweeter crosses over to the woofer at 3.5kHz, with asymmetrical slopes evident. The woofer is basically well behaved within its passband, but a sharp discontinuity can be seen at 280Hz, suggesting that a resonance is present at that frequency. Indeed, the port's output has a sharply defined spike at that frequency, as well as four others higher in frequency. This measurement was taken without the lower cavity filled with sand, but, as both Erick and I strongly recommend to Forest owners, you must fill that cavity to try to kill these resonances.
Fig.3 Totem Acoustic Forest, 2009 sample, anechoic response on tweeter axis at 50", averaged across 30° horizontal window and corrected for microphone response, with complex sum of woofer and port plotted below 300Hz.
Fig.4 Totem Acoustic Forest, anechoic response on tweeter axis at 50", averaged across 30° horizontal window and corrected for microphone response, with complex sum of woofer and port plotted below 300Hz, of: 2001 sample (green), 2005 sample (blue), 2009 sample (red).
Though its blown woofer meant that I couldn't measure SN PM4002's overall response, I did look at its tweeter. Commendably, its output differed from that of SN PM4003 by less than 0.25dB above 4.5kHz, and by less than 0.6dB at lower frequencies. The on-axis full-range frequency response of the working 2009 sample of the Forest was basically identical to that of the earlier samples (fig.3), while fig.4 shows the frequency response of the 2001 (green trace), 2005 (blue), and 2009 (red) samples. Erick's Forest doesn't differ significantly from those that Larry Greenhill and I auditioned, and this graph is a tribute to the consistency of Totem's manufacturing. (See also fig.3 in my Follow-Up on the Totem Mani-2.)
So what to conclude from this Follow-Up to a Follow-Up? First, Totem's consistency in manufacturing the Forest to a uniform standard over an eight-year period is impressive.
Second, from my conversations with Erick Lichte, it is clear that the woofer of SN PM4002 failed after he had concluded his critical listening to the Forests. However, if the woofer had been abused by one of the dealers who had used the review samples before they were sent to Stereophile for review, it's possible that this either made it more vulnerable, or contributed to EL's finding the Forests' imaging to be not as good as he had been led to expect.
But if the woofer was fine before its catastrophic failure, how to explain the differences between EL's findings and those arrived at by LG and myself. Having worked with Erick on recording projects for nine years, I know he is a careful, skilled listener. In addition, at half my age, his high-frequency hearing extends up to 19kHz, whereas mine cuts off above 15kHz. It's possible, therefore, that he is less tolerant of high-treble anomalies, which would include that slightly too-hot tweeter. Remember, too, that in addition to filling the speakers' bottom cavities with dry sand, I also ended up placing bags of lead shot atop the Forests to try to eliminate those high-Q cabinet resonances.
To give Erick the final word, he pointed out via email that, considered dispassionately, his review was actually a continuation of the same observations LG and I had made but amplified slightly toward the negative by two things: his own hearing and auditioning, (with his own sonic biases and preferences) and the continued progress of loudspeakers since the Forest first appeared. "I feel I observed and said many of the same things you and LG said," Erick concluded, "but my tone in writing was a bit stronger."
We will continue to list the Totem Acoustic Forest in "Recommended Components," but with the proviso, as always, that potential customers audition the speaker properly mass-loaded before purchase.John Atkinson