LABORATORY TEST REPORT

The frequency response of the Dynaudio 60, as measured by Newport Test Labs, was extremely flat, very extended and exceptionally well-balanced across the audio spectrum, with perhaps just a very slight roll-off at the extreme high end of the audio spectrum, and a rather steeper roll-off that only occurs 1kHz above the accepted higher limit of human hearing (20kHz).

The frequency response trace shown in Graph 1 shows the Contour 60’s response as extending from 28Hz to 22kHz ±3dB, which is not only excellent, but also possibly one of the first times Newport Test Labs’ measurement has been almost exactly the same as that of the manufacturer. I say ‘almost’ because Dynaudio specifies the frequency response of the Contour 80 as extending from 28Hz to 23kHz ±3dB—so with an extra 1kHz extension—but that increase in extension is so minor that it’s really to all intents and purposes the same response. (Such small differences could be due to model variations, measurement conditions… many reasons.)

Readers should note that the graph shown is actually the result of two separate measurements that have been spliced together using post-processing. Below 800Hz, the trace is the averaged result of nine individual sweeps measured at a distance of three metres. The trace above 800Hz was measured at a distance of one metre, directly on-axis with the tweeter,

and using a gated test signal that returns the same response that would be returned if the speakers were measured in an anechoic chamber. The resolution of this gated technique is extremely high, so it shows dips and peaks in the response that would not only be inaudible to the ear, but would also not be shown on a trace measured using the less-precise impulse measurement techniques that are so common in the industry (used because they’re cheap, and because they make measurements so fast that less time is required to make them).

Graph 2 shows the high-frequency response of the Dynaudio Contour 80 in even-more detail, but this time also shows the difference in frequency response between using the speakers with the grilles on (red trace) and with the grilles off (black trace). You can see that the two responses are almost identical, but that with the grille on there are very small dips at 7kHz and 9kHz that are not present when the grille is removed—plus the response above 15kHz rolls off just a little earlier with the grille fitted than when it isn’t. I think Dynaudio has done an excellent job with the acoustic transparency of the grilles and I would recommend leaving them in place, even for critical listening sessions, not least because I do not believe the differences shown would be audible.

Low frequency performance (measured using sinus waveforms and standard near-field techniques) is shown in Graph 3. You can see that the output of the two ports is quite different, meaning that you could expect quite substantial differences in sound if you block, say, the upper port, rather than the lower port (and vice versa). The difference in the high-frequency output between the lower of the two bass drivers and the upper driver seems to suggest that Dynaudio is running the upper driver a little ‘hotter’ than the lower one, but I think the slight additional high-frequency extension is simply a result of the different rear loading caused by the different driver position in the cabinet. You can see that without bungs, the drivers roll-off quite steeply below 50Hz to a minima at 28Hz, with the ports taking over the bass duties. With both ports blocked, the bass drivers still roll off below 50Hz, but the slope becomes far more shallow (shown in Graph 4)… just as you’d expect of this alignment.

The different alignments (ports blocked vs. ports open) also affects the impedance of the speakers of course, and you can see the classic ‘bass reflex’ impedance curve in the black trace on Graph 5, which exhibits the two expected low frequency peaks (one at 45Hz, the other at 15Hz) whereas when the bungs are inserted—so the cabinet is essentially sealed—you can see the classic ‘infinite baffle’ impedance curve appear (the red trace).

Both impedance traces converge at around 70Hz, with the peak at 110Hz suggesting that Dynaudio is using the crossover network to do some response compensation to match the response of the midrange driver (which kicks in at 220Hz and is the reason for the dip in the impedance at 200Hz). As you can see, the speakers will present a load of mostly around 8Ω to any driving amplifier, but that dive in the impedance at 200Hz down close to 3Ω means that Dynaudio has to classify the Contour 80 as ‘nominally 4Ω’ under IEC regulations… which it does, as you’ll see if you look at Dynaudio’s own specifications.

While you’re looking at those specifications, you’ll probably notice that Dynaudio rates the Contour 60’s sensitivity at 88dBSPL. Newport Test Labs’ measurement for sensitivity came within a whisker of this figure, with the laboratory recording a figure of 87dBSPL at one metre for a 2.83Veq input under its standard test conditions. This is actually a very good result, because Newport Test Labs uses a more stringent method to measure loudspeaker’s sensitivity than most manufacturers, so almost always comes up with figures lower than the manufacturer’s own specifications… often considerably lower, so a difference of a mere 1dB is a mere bagatelle. At either 87dBSPL (or 88dB) efficiency, the Contour 80s will not demand an overly-powerful amplifier in order for them to perform at their best.

That the Dynaudio’s Contour 80s are superbly designed and engineering was proved in every one of the many tests performed on them by Newport Test Labs: A superior design in every sense of that word. Steve Holding