Paradigm Atom v.3 loudspeaker Measurements
Paradigm specifies the Atom v.3's anechoic sensitivity as 86dB/2.83V/m, which is just below average. My estimate was 86.5dB(B), which is within experimental error of the specified figure. The Atom's electrical impedance (fig.1) changes considerably over the audioband, with a minimum value of 4.4 ohms at 220Hz. The small glitch just above 30kHz indicates the presence of the tweeter's main dome resonance, which is higher in frequency than we usually see. The saddle in the magnitude trace at 66Hz indicates the tuning frequency of the port.
The fig.1 traces are free from wrinkles that would imply the existence of cabinet resonances. Fig.2, a waterfall plot calculated from the output of a simple plastic-tape accelerometer fastened to the center of a side panel, shows that a high-level mode is present at 600Hz. While this mode was detectable on all of the cabinet's surfaces, it might be sufficiently high in frequency not to have too much of a subjective effect. However, the fact that BJR was bothered by an occasional "thick chestiness" might well be associated with this resonant mode.
Fig.1 Paradigm Atom v.3, electrical impedance (solid) and phase (dashed). (2 ohms/vertical div.)
Fig.2 Paradigm Atom v.3, cumulative spectral-decay plot of accelerometer output fastened to center of side panel. (MLS driving voltage to speaker, 7.55V; measurement bandwidth, 2kHz.)
The traces to the left of fig.3 show the responses of the woofer and port, taken in the nearfield, with their complex sum. The woofer's output has the notch at the tuning frequency of 66Hz, as expected, this roughly coincident with the port's maximum output and implying only moderate bass extension. So why was Bob Reina so impressed by the Atom's bass? Part of the broad hump in the upper bass will be due to the nearfield measurement technique, but the rest suggests that Paradigm has pulled the old "LS3/5A trick," which is to underdamp a small speaker's low-frequency tuning in order to give the illusion that it has more low frequencies than it actually does. That the trick can work is shown by BJR noting in his auditioning that he was amazed by the Atom's "reproduction of the pipe organ's nether regions."
Fig.3 Paradigm Atom v.3, anechoic response on tweeter axis at 50", averaged across 30 degrees horizontal window and corrected for microphone response, with the nearfield responses of the woofer, port, and their complex sum plotted below 300Hz, 750Hz, and 300Hz, respectively.
Higher in frequency, there is a broadly even balance, with a suggestion of a little too much presence-region energy, which correlates with BJR finding the speaker's lower high frequencies to be articulate. However, there is a severe notch at 3.2kHz. The primary suspect for this kind of behavior is a crossover topology that swings the main response lobe above or below the tweeter axis. However, the Atom's vertical-dispersion plot (fig.4) indicates that a true crossover notch develops well below the tweeter axis at 2.2kHz (shown by the cursor). The on-axis notch half an octave higher must therefore be due to something else.
Fig.4 Paradigm Atom v.3, vertical response family at 50", from back to front: differences in response 45 degrees-5 degrees above tweeter axis, reference response, differences in response 5 degrees-45 degrees below tweeter axis.
The lateral radiation pattern (fig.5) shows that the notch does tend to fill in to the speaker's sides, which is perhaps why Bob did not comment on a lack of immediacy to the speaker's sound, which is what I would have suspected. While BJR did comment on a lack of air, that is usually associated with a lack of top-octave energy, and the Atom does appear to have wide dispersion above 10kHz. Perhaps it is the slightly shelved-down on-axis behavior in this region in fig.3 that Bob was hearing.
Fig.5 Paradigm Atom v.3, lateral response family at 50", from back to front: differences in response 90 degrees-5 degrees off-axis, reference response on tweeter axis, differences in response 5 degrees-90 degrees off-axis.
The step response (fig.6) gives a clue as to what is going on. The tweeter's output arrives at the microphone slightly ahead of the woofer's, with both units connected in the same, positive acoustic polarity. But the ripples apparent in the decay of the woofer's step are due to a strong reflection about a third of a millisecond later. This must be due to the cone's backwave reflecting from the magnet, or something else behind the cone, and it is clearly evident in the Atom's cumulative spectral-decay plot (fig.7). That it didn't have much of a subjective effect must be due to the fact that the resultant notch is quite narrow, and that the in-room energy in this region doesn't feature a notch. Other than this, the waterfall plot is very clean in the treble, with just a small ridge of delayed energy apparent at 2kHz, possibly indicating the presence of some residual coloration related to the woofer cone.
Fig.6 Paradigm Atom v.3, step response on tweeter axis at 50" (5ms time window, 30kHz bandwidth).
Fig.7 Paradigm Atom v.3, cumulative spectral-decay plot at 50" (0.15ms risetime).
A speaker costing less than $200/pair is usually severely compromised in one or more performance areas. With the Atom, it appears that Paradigm has managed a tricky balancing act so that the overall package still sounds good. However, its bass tuning suggests placement well away from room boundaries if it is not to sound too hooty.—John Atkinson