DeVore Fidelity Orangutan O/96 loudspeaker Measurements

Sidebar 3: Measurements

I used DRA Labs' MLSSA system and a calibrated DPA 4006 microphone to measure the DeVore Fidelity Orangutan O/96's frequency response in the farfield, and an Earthworks QTC-40 for the nearfield and spatially averaged room responses.

The Orangutan O/96 has a very high specified voltage sensitivity of 96dB/w/m. This is both unusual and means that the speaker will play very loudly with very low-powered amplifiers. My estimate of the DeVore's sensitivity was somewhat lower, at 91dB(B)/2.83V/m, though this is still usefully high. Concerned that I had mischaracterized the sensitivity, I checked my estimate by comparing the SPL produced by the O/96 with that from a BBC LS3/5a at the same drive level. The O/96 was 9dB more sensitive than the LS3/5a, which is within experimental error of the original 91dB figure (B-weighted).

As can be seen in fig.1, the O/96 has an unusually high impedance, conforming to the specified 10 ohms. At the 2.83V level used to calculate the speaker's sensitivity, the O/96 will therefore draw less than 1W from the amplifier, which will have a negative impact on the measurement. The impedance magnitude (solid trace) drops below 8 ohms only in the lower midrange and the mid-treble, reaching respective minimum values of 7.2 and 7.8 ohms. While the electrical phase angle (dotted trace) reaches extreme values in the bass, the magnitude is very high at these frequencies. Overall, the DeVore O/96 is one of the easiest speakers for an amplifier to drive that I have encountered.

Fig.1 DeVore Fidelity Orangutan O/96, electrical impedance (solid) and phase (dashed) (2 ohms/vertical div.).

The traces in fig.1 are disturbed by some discontinuities in the midrange, suggesting the presence of cabinet resonances of some kind. Listening to the cabinet walls with a stethoscope while I played the half-step–spaced toneburst track from my Editor's Choice CD (Stereophile STPH016-2), the side and rear walls were very lively between 130 and 230Hz, and some lower-level modes were audible higher in frequency. Investigating the cabinet's vibrational behavior with a simple plastic-tape accelerometer revealed very strong modes at 148 and 219Hz (fig.2), these coincident with the frequencies of two of the wrinkles in the impedance traces. These modes might have lent the speaker the richness on voices noted by AD.

Fig.2 DeVore Fidelity Orangutan O/96, cumulative spectral-decay plot calculated from output of accelerometer fastened to center of rear panel (MLS driving voltage to speaker, 7.55V; measurement bandwidth, 2kHz).

There were also peaks at these two frequencies in the output of the two ports on the rear panel when measured in the nearfield (fig.3, red trace). Though these are down in level, they do result in discontinuities in the nearfield woofer response (blue trace) and the overall low-frequency response (black trace below 300Hz). Other than that behavior, however, the ports' output peaks sharply in classic manner between 35 and 55Hz, its maximum level coinciding with the minimum-motion notch in the woofer's response at 43Hz, somewhat higher than the "mid 30s" mentioned by AD. (At this frequency, the back pressure from the port resonance holds the woofer cone still.)

Fig.3 DeVore Fidelity Orangutan O/96, anechoic response on HF axis at 50", averaged across 30° horizontal window and corrected for microphone response, with nearfield responses of woofer (blue) and port (red) and their complex sum (black), respectively plotted below 350Hz, 750Hz, 300Hz.

The black trace above 300Hz in fig.3 shows the O/96's farfield response averaged across a 30° horizontal window centered on the tweeter axis. Overall it is relatively flat, with small peaks balanced by small dips in the response. The region covered by the tweeter is very slightly less sensitive than the woofer's range; of more concern is the sharply defined peak just below 2kHz. This coincides with a wrinkle at the same frequency in the impedance graph ; this behavior might be due to a termination problem with the woofer cone and its surround at this frequency. However, this peak coincides with a lack of energy to the speaker's sides (fig.4), and so may well not result in coloration. What appears to be an off-axis "flare" between 3 and 4kHz in this graph is actually due to the on-axis suckout in this region filling in to the speaker's sides. Taking this into account, while the DeVore O/96's horizontal dispersion is narrower than usual above 1kHz, it is actually much better controlled and consistent with frequency than I was expecting from a design using a large woofer and a wide baffle.

Fig.4 DeVore Fidelity Orangutan O/96, lateral response family at 50", normalized to response on HF axis, from back to front: differences in response 90–5° off axis, reference response, differences in response 5–90° off axis.

In the vertical plane (fig.5), the optimal response is obtained just above the tweeter axis. This is sensible, given that the tweeter is just 32" from the floor when the speaker sits on its matching stand.

Fig.5 DeVore Fidelity Orangutan O/96, vertical response family at 50", normalized to response on HF axis, from back to front: differences in response 15–5° above axis, reference response, differences in response 5–10° below axis.

I visited Art Dudley to give a listen to the O/96s in his system and to measure their in-room response before bringing them back to my place for further measurements. I averaged 20 responses taken in a rectangular grid 36" wide by 18" deep, and centered on a position 36" from the floor at AD's listening position. (We have found that 36" is the height of a typical seated listener's ears; AD actually sits a little higher.) The speakers were driven by AD's Shindo Corton Charlemagne amplifiers; the result is the red trace in fig.6. The treble slopes down smoothly, if a little more than I would have liked—I found the balance in Art's room somewhat lacking in top-octave air, though the low frequencies were well extended.

Fig.6 DeVore Fidelity Orangutan O/96, spatially averaged, 1/6-octave response in AD's listening room (red); and of Audio Note AN-E/SPe HE (green).

What is fascinating about the O/96's in-room response is that it is almost identical to that of AD's reference speaker, the Audio Note AN-E/SPe HE (green trace). The Audio Notes were placed in the room corners for this measurement, with a severe toe-in; the DeVores were well away from the wall behind them but fairly close to the sidewalls. As with the Audio Notes, there is a peak between 500Hz and 2kHz, and a lack of energy between 100 and 200Hz. I suspect that, with both speakers, the latter is due to destructive interference between the direct sound from the woofer and the reflections from the walls and floor. In the case of the O/96s, these reflections would be reinforced by the fact that the distances between each woofer and the two closest room boundaries were very similar.

I was aware of the lack of lower-midrange energy when I first started listening to the DeVores in AD's room, though that receded as I continued listening. I suspect that this kind of interference is something we accommodate to. I investigated further when I got home, setting up the O/96s in my own listening room and driving them with the Devialet D-Premier integrated amplifier. The red trace in fig.7 again shows the Devore's spatially averaged response in AD's room; the blue trace shows the O/96's response in my room, measured in identical manner. Because I could place the speakers farther away from the sidewalls in my room, which is wider than AD's, their lack of energy in the lower midrange has filled in nicely. Without the boundary reinforcement in my room, the bass is slightly shelved down, though it still extends down to below 25Hz. Though there is still a bit too much energy in the upper midrange, with the solid-state amplifier the treble is considerably more extended in my room, if with the same smooth characteristic measured in AD's room.

Fig.7 DeVore Fidelity Orangutan O/96, spatially averaged, 1/6-octave response in: AD's listening room (red), JA's listening room (blue).

The O/96's step response (fig.8) indicates that both drive-units are connected in positive acoustic polarity, and the smooth integration of the decay of the tweeter step into the start of the woofer step confirms optimal crossover design. The cumulative spectral-decay plot (fig.9) reveals a generally clean decay in the treble and midrange, but with some low-level hash evident in the low treble and a prominent ridge of resonant energy coincident with the on-axis peak at 1730Hz.

Fig.8 DeVore Fidelity Orangutan O/96, step response on HF axis at 50" (5ms time window, 30kHz bandwidth).

Fig.9 DeVore Fidelity Orangutan O/96, cumulative spectral-decay plot on HF axis at 50" (0.15ms risetime).

After I'd finished measuring them, I spent a day listening to the Orangutan O/96s in my room. Even though I knew about the low-treble resonance and the lively enclosure, these problems were considerably less audible than I was expecting. Only with recordings of solo acoustic piano did they get in the way of the music by producing noticeable coloration, the piano's midrange sounding uneven, with some notes obscured. But with well-recorded rock and classical vocal recordings, the measured problems seemed to step into the background, letting me appreciate the O/96's full-range, evenly balanced sound and superb clarity. It looks as if John DeVore has accepted the inevitability of the problems caused by the O/96's physical concept and had carefully worked around them to produce a well-balanced speaker that is also drop-dead gorgeous.—John Atkinson

DeVore Fidelity
Brooklyn Navy Yard, 63 Flushing Ave., Unit 259
Brooklyn, NY 11205
(718) 855-9999
Share | |

Enter your username.
Enter the password that accompanies your username.