Monday
Sep152008

How We Engineer Time-coherent Sound

Home > About > Our Difference: What is Time-coherent Sound? - How We Engineer It - AIG Article

 

Tuning_the_Empire_State_Building

Math and physics guarantee results

The image above is of work being done on the antennas at the very top of the Empire State Building, to better direct the flow of radio waves. This is not done willy-nilly, as there is a type of mathematics that shows what must happen at the source (antenna), to get the right result downstream (across town).

Fortunately, this math can also be used to determine in what manner sonic energy must be launched from a speaker. By that, we mean specifically for a speaker designer to use in deciding whether to use one tweeter, 12 woofers, etc. Most speaker designers do not bother with this math, don't even know of its existence, which is why we see so many different designs.

And this is why, if you read on a bit further, we lay out what must happen in order for a speaker designer to make 'the best decisions'. You will then know what separates us from the herd.

To make those decisions about woofers and tweeters-- their number, size, location and much more, the math tells us that we must first know where are your ears? Are they in a stadium, in a living room or perhaps at a computer desk? There is a unique solution for each 'working distance'.

We must also know what must be delivered 'across town', that is, to our ears. What energy is required to appear next to our ears? How loud? What tone range from bass to treble? For that, we must know about the 'sounds' to be reproduced:

You'd think our eardrums would like to receive the natural sounds of live instruments and voices, but that's never going to happen because of how microphones 'hear' and because of their unnatural positions close to each artist, each instrument.

Studio engineers are paid to 'mellow' and otherwise shape those close-up sounds to sound natural, or perhaps unnatural in interesting ways. You can hardly playback or even enjoy a raw recording of close-up sounds. Why not move the mics farther away? Because in a studio, the musicians need to be close enough to see and hear each other. When a mic is moved back, it hears too much of another instrument or voice. This creates sonic interference and usually captures 'too much room echo', as you may know from recording conversations with a tape deck laying on the table.

For an orchestra in a symphony hall, you'd think there would be enough room to space the mics back, but most recording engineers instead place mics above each section of the orchestra, and later "fix it in the mix". It is too time consuming and costly to locate the one place in that hall where just two mics (placed via the ORTF method) can pick up what audience members know as the natural sound of the entire orchestra.

But regardless of the engineering, it can be said with complete assurance that we'd like the recorded waveforms delivered to our ears uncorrupted by the speaker, and overlaid with our room's own echoes (for which any experienced engineer has made allowance).

Looking at the complex waveform of music on a `scope or computer, it's virtually impossible to tell what wiggle belongs to exactly what sound. We can use a computer or spectrum analyzer to break that complex wave into many different pure tones (sine waves) that come and go with different loudnesses and timings.

To re-construct that original wave, both the loudnesses and timings of those individual tones must be reproduced. At your ear. Across the room. Most all speakers scramble the timings because it's something extraordinarily difficult to avoid. Crossover circuits and the limitations of conventional woofers and tweeters and their physical placements all create time delays, delays that are also different at each frequency. Which is the main reason all speakers sound different and none sound like 'the real thing'. They sound "like speakers" or worse, hurt your ears.

Our goal is perfect timing and loudnesses over at your ear. Together, those guarantee you will hear the most emotion, the sharpest images, proper timbres (textures) and best tone balances from any recording.

Given that we hear only when our eardrums move, under the impact of the air molecules right next to them, then our goal really is to move those particular molecules the correct sub-microscopic distances at the right times. Each molecule knows only that it was hit from behind, and so on... all the way back to the speaker. Which is why that math must come into play.

 

The physics of sound

These math equations are called Green's Functions, and can tell us how the speaker must initiate the very first molecular collisions in front of its cones and domes, far from your ears. The math also tells us the size, shape and number of the speaker's drivers, along with their power handling and the tone range each must cover. The math can be adapted to determine the shape, size, and curvatures of the enclosures around each driver. When coupled with the principles of psychoacoustics, the math also tells us where to place the drivers in space.

We also found a way to use this math to guide our Balanced-Phase™ crossover circuit technology, by showing how to compare and contrast close-up microphone readings with what is heard out in the room.

Those measurements only get us close to the required result because they are limited by the resolution of any measuring system and the length of soundwaves. Our designer discovered a way to use the predictions the math makes, regarding what sound should be heard when perfectly combined at the ear, to guide his final listening tests, achieving an audibly singular focus to the sound.

This refinement allowed him to then easily hear the clarity and dynamic accuracies of all the raw parts that make up a speaker. From the capacitors and wires to the drivers and cabinet materials, certain parts clearly stand out from others in their ability to transmit the small inflections that are the essence of musical expression. While those characteristics cannot be easily measured, they are clearly audible from any recording when the other, time-domain aspects of the speaker's design have first been minimized.

While we admit our designer's approach -- using applied physics, his live-recording experience, and psycho-acoustic knowledge -- is unique, it is not unusual. Radio and television engineers use Green's Functions to deliver energy by first mapping the terrain of the area. They know that will alter their signal's strength and other parameters required at your location, and they use that information for their Green's Functions to determine, among other things, the orientation and polarity of the electromagnetic field that must be generated by their antenna.

In speaker design, this same approach results in speakers with predictable behaviors across a range of listener distances in the expected environments.

The result? You hear what we hear -- the smallest inflections of music, the slightest sound effects, the grandest dynamic expressions, the subtle sways and surges, ever-changing timbres and full range of emotions. You will hear all of the messages, all the time.

 

Top