JBL S38 loudspeaker Measurements
At first glance, the JBL S38 looks old-fashioned, with its side-by-side midrange unit and woofer. But then you notice the metal-finish cones and the tight vertical array of the midrange unit and the tweeter, and it becomes possible, to echo Bob Reina's phraseology, that this isn't your father's JBL. One area, however, where the S38 is definitely a JBL is its high sensitivity: an estimated 91dB(B)/2.83V/m.
In addition, while the speaker's impedance magnitude drops to around 4.1 ohms in the lower mids and the high treble (fig.1), it is generally an easy load for an amplifier to drive. However, a combination of 5 ohms magnitude and 45 degrees capacitive phase angle at 90Hz will mean that the best amplifier would be one rated into 4 ohms.
Fig.1 JBL S38, electrical impedance (solid) and phase (dashed). (2 ohms/vertical div.)
Though it will be hard to see in fig.1 at the size at which this graph must be printed in the magazine, there is a slight discontinuity in the traces just below 400Hz. This correlates with a major cabinet resonance at 385Hz, which was detectable on all surfaces (fig.2). Other high-level vibrational modes can be seen at around 250Hz, and the JBL's cabinet was generally live, which might be thought to obscure lower-midrange clarity. I do note that BJR was bothered by a lack of clarity on densely orchestrated recordings, which might well correlate with this behavior.
Fig.2 JBL S38, cumulative spectral-decay plot of accelerometer output fastened to center of top panel. (MLS driving voltage to speaker, 7.55V; measurement bandwidth, 2kHz.)
The "saddle" in the impedance magnitude trace at 37Hz (fig.1) indicates the tuning frequency of the twin ports, implying reasonable bass extension. However, the S38's bass alignment has been tuned more for a somewhat exaggerated upper bass than for ultimate extension. This can be seen in fig.3; while it must be noted that some of the broad rise in the bass is due to the nearfield measurement technique, most of what you see is characteristic of the speaker—and BJR did comment on the exaggerated 100Hz region. This bass tuning will be even more exaggerated if these "bookshelf" speakers are actually used on bookshelves.
Higher up in frequency in fig.3, the S38's balance is basically smooth, though with a slight excess of treble energy. Though this, subjectively, will tend to balance the bass, I note that BJR did find the JBL's treble to sound rather forward. There are also a couple of narrow suckouts noticeable, at 1100Hz and 7kHz, though in themselves these might not have much of a subjective effect, suckouts being much harder to hear than peaks of the same height and width.
Fig.3 JBL S38, anechoic response on tweeter axis at 50", averaged across 30 degrees horizontal window and corrected for microphone response, with the nearfield woofer and port responses and their complex sum plotted below 300Hz, 1kHz, and 300Hz, respectively.