Spendor S3/5R2 loudspeaker Measurements
Sidebar 3: Measurements
I measured the Spendor S3/5R2 loudspeaker with DRA Labs' MLSSA system, using a calibrated DPA 4006 microphone to measure the speaker's frequency response in the farfield, and an Earthworks QTC-40 mike for the nearfield responses. The S3/5R2 is very similar in size and basic topology to the S3/5se, which Art Dudley reviewed for Stereophile in July 2003. You can find my measurements of the earlier speaker here.
The S3/5R2's voltage sensitivity is specified as 84dB/W/m, but my estimate was just 81.7dB(B)/2.83V/m, which is essentially the same as the 2003 version. However, as with the earlier speaker, the S3/5R2's plot of impedance magnitude and phase (fig.1) reveals it to be a very easy load for the partnering amplifier. The magnitude drops below 8 ohms only in the lower midrange, and stays well above that figure for most of the audioband. The single peak in the bass, reaching 32.6 ohms at 76Hz, indicates that this is the tuning frequency of the sealed box, which is where the low-frequency response will be down by 6dB.
A small wrinkle at 200Hz in the impedance traces suggests that there is an enclosure resonance of some kind at that frequency. However, investigating the panels' vibrational behavior with a simple plastic-tape accelerometer revealed nothing untoward at that frequency, though there was a strong resonance at 316Hz on the sidewalls (fig.2). I would have expected this to add a slight degree of midrange congestion.
Though there is a slight rise in the upper bass in the Spendor's nearfield response (fig.3, below 300Hz), this is mostly due to the nearfield measurement technique. The S3/5R2 doesn't have as much of an upper-bass bump as the classic LS3/5a, or the similar-sized KEF LS50, which I reviewed in December. Higher in frequency in fig.3, the S3/5R2's farfield response, averaged across a 30° horizontal window on the tweeter axis, is even, though with a slight lack of energy in the crossover region. It also looks as if the tweeter is balanced a couple of dB too hot. However, this response was taken without the rather clumsy grille; adding the grille pulled down the output between 3 and 10kHz (not shown) by a dB or so, which is presumably why BJR preferred what he described as a slightly more detailed sound with the grilles removed.
The S3/5R2's horizontal dispersion (fig.4), normalized to the on-axis response (which therefore appears as a straight line), indicates that the lack of on-axis energy in the crossover region deepens to the speaker's sides. All things being equal, this might make the speaker sound a little laid-back, though it's more likely that it will leave that slightly hot tweeter sounding a little detached. However, it is fair to note that BJR didn't find the Spendor to sound tipped up toward the high frequencies. In the vertical plane (fig.5), a large suckout in the crossover region develops 5° below the tweeter axis, as well as 10° and more above that axis. More than usual, and especially to a greater degree than its predecessor, this speaker will be very fussy about exact listening axis.
Whereas the 2003 version of this speaker had the tweeter and woofer connected in the same, positive, acoustic polarity, in the 2012 edition the tweeter is connected in inverted polarity, though the decay of the tweeter's step blends smoothly with the start of the woofer's positive-going step (fig.6). At least on the tweeter axis, this will result in a fairly smooth blend of the drive-units' outputs, though I did note a slight lack of energy in the crossover region. The cumulative spectral-decay plot on the tweeter axis (fig.7) is very clean. (Ignore the black ridge of apparent delayed energy just below 16kHz in this graph; this is due to interference from the test computers' video circuitry.)
The Spendor S3/5R2's measured performance indicates fundamentally good engineering, though it is fair to say that this version is not appreciably better than the 2003 iteration, which measured very well.John Atkinson