A good hi-fi system reproduces an incredible amount of detail: a ringing here, a deep impulse there. But sometimes, what’s missing is even more decisive for our perception of tonality and imaging.
Sometimes our senses need a jump start in the form of a rich contrast. A counterweight that helps us to correctly classify all impressions. A pinch of sugar that makes the pasta sauce taste even richer, the movie theater whose formerly inconspicuous escape route lighting mutates into floodlights once the main lighting goes dark, or the bold hammer stroke hitting the thumb rather than the nail that makes us realize how beautiful the world was just five seconds ago … In isolation, many things lose their intensity, and this results in an intriguing inconvenience: Often, without comparison to a “reference value,” we don’t even notice that something is wrong. We can encounter the same phenomenon when listening to a loudspeaker.
At least in high-fidelity matters, the FIDELITY team is in a luxurious bubble of exception. In the listening room, several CD players, turntables, streamers, amplifiers and loudspeakers are available at all times. Cable sets are also plentiful. So it’s no problem to find out by a quick A/B comparison what sounds more right or (should that be the case) where exactly the shoe pinches. Nevertheless, we also fall into the trap now and then. A few weeks ago, a colleague rejoiced about the breathtaking imaging of a freshly unpacked loudspeaker. And indeed, the proud floorstanding speaker played so large and spherical that its music literally embraced and enclosed us. After a brief moment with our jaws dropped, initial skepticism set in. Somehow the playback lacked the usual power and dynamics. A walk around the speakers revealed the mystery – one of the speakers was reversed in polarity. The plus and minus plugs of the loudspeaker cable sat in the wrong sockets of its terminal. Of course, the problem could be solved in two easy steps. And yet, this hardly noteworthy incident brought up an interesting question: Why does a pair of speakers with reversed polarity sound so much more spatious than a correctly wired pair?
Phase, I can hear you flounder
The electrotechnically educated elite among the FIDELITY readers have long known the answer, of course: The incorrectly connected loudspeaker duo does not play remotely larger, deeper, wider or broader than a correctly wired one. The contrast is simply greater because something quite decisive is missing in the middle. Therefore, the now completely isolated fanning out of the right/left components seems more extreme to us and at times even more interesting.
In order to fathom the background of this phenomenon, we should mentally take a seat at the mixing desk of a recording studio. When mixing, an experienced sound engineer will always try to distribute the instruments of the band or ensemble homogeneously in the mix. In the case of a large orchestra, he will use the seating arrangement of the musicians as a guide. With pop, rock and electronics, more creativity is allowed: He will keep adjusting the panorama controls on his mixing console until the individual elements no longer get in each other’s way and the stage has the desired width. He usually leaves high-energy elements like bass and kick drum in the middle. First, their low frequencies can hardly be located anyway. Second, they are reproduced equally by both amplifier channels and speakers – the energy-hungry low frequencies are shouldered by several powerhouses. Often, vocals (or at least the most important vocal track) also remain in the center – giving voices more presence and focus.
And now to the acoustic component of the phenomenon: the membranes of a loudspeaker convert the electrical energy of the amplifier into sound. They accomplish this by making a forward movement in the event of a positive excursion of the driving current or by retracting into their speaker chassis in the event of a negative excursion. The electrical waveforms are thus converted into rapid air pressure fluctuations, which we – a flourish please! – perceive as sound. The propagation of these sound waves in space follows a fixed set of rules. Low frequencies spread out spherically around the loudspeakers and thus cover the entire room. Towards the mids, the speaker’s radiation becomes more and more directional. Finally, the highs are radiated by the tweeter directly to the listening position without any significant scattering. The analogy with a laser pointer is exaggerated and at this point overused to the point of yawning, but it points in the correct direction.
This leads to the following consequential conclusions: Bass and fundamental energy of both speakers has large overlaps in space. The sound waves overlap in complex patterns, and whenever two pressure highs or lows meet, they reinforce each other. This is true to a limited extent for the midrange. Here, there is simply less overlapping. The highs are completely out of the picture, although I should emphasize that we deliberately exclude aspects such as wall reflections, diffusion and reverberation for the sake of “de-complication”.
If you reverse the polarity of one of the speakers, the following happens: Bass components and fundamental, which are quasi-mono in the center of the imaging, no longer amplify, but are almost or completely cancelled out by the opposing phases. This fate also befalls the voice(s) and all instruments that were placed exactly in the center during mixing. What remains in the image are only those frequency components that are supposed to provide the stereo impression on both speakers, are different and thus cannot be cancelled out. These are – due to the mixing logic described above – especially the upper midrange and the treble. Or extreme stereo effects like the reverb trail of vocals, whose spherical depth impression reinforces the imaging size. The large (bass) hole in the center of the stage enormously enhances our perception of depth and width. All quite logical, isn’t it?
So this means … what, exactly?
Well, first of all, you should connect your speakers correctly! This can be a bit of a challenge at times due to the color coding of cables and connectors (white/red, black/red, black/white) not always being very clear. Furthermore, if you’ve ever wondered how reverb effects or DSP processors manage to tickle room-filling 3D sound out of two stereo speakers, you now have an answer: such systems work – albeit noticeably more homeopathically – with exactly the same mechanics. The most extreme example here is probably the Auro-3D concept, which creates reasonably vivid 360-degree images from mono signals through lively phase rotations.
And as surprising as it may seem, there are some quite interesting and helpful lessons to be learned from our wiring error. For one, it reinforces my personal credo when commissioning speakers: For years, I’ve used the same song when starting my tests. It’s conditioning. If I had done it again this time, I would have noticed the mistake immediately. On the other hand, the deliberate reversal of polarity can of course be used to bring in fresh speakers and test their “matching qualities”. If you place a pair of speakers with a phase-reversed loudspeaker close to each other and play them with a broadband mono signal (noise, sine sweeps etc.), almost no sound should leak out. This way you can burn in the speakers for days without disturbing anyone. If you still hear a noticeable amount of the signals, either the pair matching of the speakers is not good (in plain terms: the speakers differ), or you didn’t put enough effort into matching them as exactly as possible.
And finally, you can take advantage of the phase cancellations of a reversed loudspeaker duo to achieve a near-perfect setup. In theory, a stereo system is supposed to do just that: the mono signals, i.e. those impulses and frequencies reproduced by both speakers, are supposed to be rock-solid in the center, while the stereo differences (also known as “sides”) are supposed to be further out and define the stage width. In practice, this works only to a limited extent, since wall, floor and ceiling reflections as well as room modes interfere with the imaging. Apply a mono signal (preferably “pink noise”) and observe how well bass and lower mids are cancelled with the speakers in reverse polarity. If you are unsure about evaluating frequencies, you can consult a frequency analyzer, such as those available as free apps for any smartphone platform. You will probably find that something still comes through here and there. Now comes the Sisyphean task: by moving and angling the speakers, you can try to see if you can find a positioning variant on which the speakers illuminate the listening room more symmetrically, and the cancellations are correspondingly more “gap-free”. This will also have a reverse effect on “in-phase” listening: The imaging becomes more precise and sharper, the stage seems wider – as we said, the contrast makes the music.