Spatially Coherent System (SCS), Cabasse's key principle for many innovations

SPATIALLY COHERENT SYSTEM (SCS), CABASSE’S KEY PRINCIPLE FOR MANY INNOVATIONS.

Coherence of direct sounds

 

Sound is formed by a series of waves of variable length (frequency) and height (intensity), just like waves on the surface of water. To get wide, high waves, you need big stones which could displace a lot of water, whereas very light pebbles will make small waves or ripples which are very short and close to each other. Likewise, in air, you need a loudspeaker with a large diameter for bass frequencies and a small one for trebles. In music, you need a large, heavy cord for the deep tones of a double bass and a small light one for the high notes of a violin.

To faithfully reproduce a sound field from the deepest sounds to the highest notes, the most obvious solution is to use several drivers, each one designed for its own range of use (treble, midrange, bass). However a note, produced by a piano, for instance, is not made up of a pure frequency, but of the note’s own fundamental frequency (440 Hz for the note A used to bring an orchestra into tune) and a host of other multiple frequencies of varying intensities.

The fundamental determines the note, and the other related frequencies for the instrument’s timbre. This is what makes it possible to distinguish between a piano and a trumpet, for starters, then a Stradivarius and a student violin, and finally, a Steinway and its twin. This note, recorded at a single point in space by a microphone, should come to your ear via an acoustic speaker without being distorted.

If some of the harmonics making up this note are delayed because they are coming from another drive unit within the loudspeaker's enclosure located further away from your ear, the note will be modified. If the harmonic is located in the range of frequencies common to both sources, it may vary in intensity due to the time delay.

In the first case, the positioning in space and the filtering will make it possible to sufficiently reduce the delay, at least at a given listening distance. In the second case, the distance between the axes of the two sources must be very small, and all the more so when they share a wide frequency range. In the case of 2-channel stereo, two sets of sources must make the same set of frequencies reach the ears of one or several listeners in order to respect the timbres of each instrument. This homogeneity of direct sound will also influence the quality of the stereophonic image, since the positioning of the instrument in space is determined by the difference in sound levels and the time delay between the left and right speakers. The better the differences present in the recording are respected in the reproduction, the finer and more accurate the sound image will be.

Coherence of direct sounds and reflected sounds

Under home listening conditionings, the sound field in the listening area is made up of only 20 to 30% of the sounds coming directly from the speakers and of 70 to 80% of sounds reflected by the room and its furnishings (indirect sound). Therefore, indirect sounds play an important role in the quality of reproduction. When you move speakers around a room to find the optimal place for them, you can vary the distribution of direct sounds and reflected sounds, in order to obtain the best spectral balance and best stereophonic image at the listening point.

Reflected sounds are mainly emitted by loudspeakers outside of their axis all around the speaker’s cabinet. Depending on the frequencies, the distribution of the power emitted over 360° by any type of transducer will vary, as shown by the diagram above. To make indirect sounds as coherent as possible, you must find off-axis the same lack of delay between sounds emitted by the acoustic speaker’s various transducers, whether these sounds are reflected laterally by the walls or vertically by the floor and ceiling. The balance between direct sounds and reflected sounds must also strike the right proportions, since too many indirect sounds will create extreme spatialization with an image which is unstable and inaccurate.

If there is too much directivity, the timbres may be right, but the image will be flat, with neither substance nor depth at its center. This is where we reach the limits of the analogy with waves or ripples on water, since the role of the speaker is to recreate the wave front emitted by the orchestra towards the listener, without artificially increasing the waves reflected by the wall behind the speakers.

On the polar curves of the diagram here, representing the way the bass, midrange and treble frequencies of one of our speakers are dispersed, the shapes follow physical laws. The absence of irregularity shows exemplary evenness even outside of the axis, which provides optimal coherence between direct sounds and reflected sounds at the listening point. Even though a room’s acoustic characteristics have an influence on the quality of reflected sounds, and therefore color the sound in a way, it is important that the sound emitted off-axis be neutral and transparent. On the one hand, it is practically impossible to obtain total clarity by mixing colors, and on the other, our brains assimilate the coloration of the room we are in. Thanks to this faculty, which microphones lack, we can recognize the timbres of an instrument or a voice in any room.

At Cabasse, in fact, we validate the speakers we produce in several rooms, including a soft room which has been treated to reverberate very little and a modern living room with picture windows and modern furniture, in order to correctly check the coherence between direct and reflected sounds in different acoustic situations.




Coinciding sources to make sounds coherent

Reproducing a wide and deep sound image with a sound source formed by adjacent loudspeakers is like trying to project a 3D video image over 240° using two systems with three tubes side by side. The three colors will coincide on a plane in front of the projector, but the walls will be colored whenever the image moves the axes of the three tubes too far apart. The main frontal image will itself be affected by the changing reflections. Thanks to a single punctual source, the projection system of cinemas like the Géode in Paris can project this sort of image over more than 180° while maintaining shapes and colors. In the same way, coinciding sound sources respect the recording, without adding defects due to the lack of coherence between direct sounds and reflected ones. This is shown in the diagrams comparing the spectral balance of a conventional speaker and a Cabasse speaker at the listening point.




Coaxial coincidence

The Cabasse team’s wager was to put all the drivers on the same axis, or even at a single point, so that they form a portion of pulsing sphere, without getting in each other’s way. They carried the bet in 1992 with the tri-axial TC21 speaker, which was replaced 10 years later by the TC22. This transducer in the up-market Artis range has regular directivity from 80 to 22,000 Hz, with no irregularities in the overlapping zones between the midrange-woofer, midrange and tweeter channels. It is also a vital tool for developing and perfecting measurement protocols and verifying that the principles of spatial coincidence are applied without compromising the quality of Cabasse’s own characteristics. This has enabled our laboratory to create a whole range of satellites with coaxial coincidence, from the big Baltic sphere to the tiny Xo. This coaxial technique is the optimal solution to comply with source coincidence criteria, and will be represented in the very near future by a new benchmark transducer developed by Cabasse’s design department based on the TC22.



The principle of virtual coincidence

The law of spatial sampling makes it possible to virtually recreate a coinciding source with several loudspeakers on different axes, if the distance between the axes of emission do not exceed a half-length of the cutoff frequency between the transducers. In the case of a cutoff frequency of 5,000 Hz which corresponds to a 6.8 cm (2.7”) wavelength and a 17 cm (6.5”)- diameter woofer, in order to comply with the virtual coincidence criterion, the tweeter must be placed within a radius of 8.5 cm (3.3”) from the woofer transducer. If the cutoff frequency is lowered to 2,500 Hz, the maximum optimal distance between the axes of the 2 drivers goes to 13.6 cm (5.4”). According to this law from Nyquist’s studies, the critical distance can be increased by optimizing the loudspeakers’ directivity in their coverage zone. So, recreating a coherent multichannel source based on adjacent sources becomes possible.



Virtual coherent systems

Armed with their know-how in developing new diaphragms and their mastery of the SCS principles in the coaxial ranges, the Cabasse team members have perfected a new series of tweeters which comply with Nyquist's theorem laws and Cabasse's quality criteria. These midrange-tweeters are fitted with lightweight, rigid Kalladex diaphragms made by our automated machines.


Their large emission surfaces allow very low cutoff frequencies. The diaphragm/horn combination has been specially designed for optimized control of directivity outside of the axis and in accordance with Nyquist's formula, to increase the critical distance so that it forms a source of virtual coherence with our woofers. The diagrams below show our midrange-tweeters' superiority in the way directivity develops in the overlapping zone with the woofer, and their close resemblance to the TC22's directivity curve. They are found on the MT3, MT4 and Altura ranges.

Numerous solutions, but only one acoustic signature
With its coaxial and midrange-tweeter applications, the SCS principle can highlight the qualities of each range, without compromising the acoustics:

ease of implementation, power and traditional beauty of the MT3, MT4 and Altura ranges;
compactness and unobtrusiveness of the Xo, Gallia, Cinesound and iO set ups;
elegance and modernity of the Xi and Ki satellites;
the technically and esthetically absolute of the Artis range.

For more technical information: