A V A L O N A C O U S T I C S
T E S S E R A C T S I G N A T U R E
Our Stealth Technology, the redirecting and diffusion of ambient room energy, makes the disappearing act possible. The use of tensor field mathematics, assigning a value and direction to expanding harmonic waves, along with enormous amounts of number crunching, has given us the ability to make reliable predictions regarding which physical forms will best disappear in space. Making an array which is over 7 feet tall disappear is quite a feat of practical mathematical prediction and actual form fabrication.
Sonic Holography is accomplished through meticulousness in recovery and preservation of low level signal information which is encoded into any well made recording. Whenever a microphone is open it is collecting phase and spatial information which is rarely fully decoded by a playback system. This fragile information may be destroyed at many stages of the recording/playback process. Our goal is to preserve as much of the truth that inhabits each recording without putting our own fingerprints on the final presentation.
Most importantly, we never apply technique without considering the effect it will have on the emotional content of the presentation. For us technique is always in service to the music. There must be an aesthetic or emotional result for each intellectual or technical choice we make, one that speaks to our souls.
Selected Design Highlights
Sonic Holography: Superlative 3 dimensional imaging capability with two-channel source material
Proprietary Quad driver internally powered Isobaric transmission line system
Symmetrical impedance characteristics
Constant driver to driver Q
Purely Pistonic driver behavior throughout the bandwidth
Point source focus
Phase linear filter topologies
Fully damped non-reactive amplifier load
Stealth Technology: Acoustic radiation control for launch surface and room ambience return energy
Linear octave to octave radiation balance
The contextual mindset we have followed in arriving at a solution to each aspect of our investigation is that physical wave phenomena, whether optical, magnetic, or gravitational, all follow the same laws. For example, Maxwell’s partial differential equations are not only the foundation of electrodynamics, but also of fluid dynamics, and in the case of tensor fields also govern gravitation and special relativity. In other words, it is possible to find a solution to each of our technical challenges through research into immutable physical principles. WIDEST POSSIBLE BANDWIDTH WITHOUT DISTORTION A Flat response between 16Hz and 70 kHz. (‐3db points 13Hz and 70 kHz). NO PHASE ERRORSA new family of filter topologies developed exclusively for this project based on classical all‐pass phase networks. These proprietary networks are more phase coherent in the cross band between drivers (minimum time ripple) with lower output in the stop‐band. Transducer phase coherence is vastly more complex than the elementary alignment of drivers in space according to their acoustic centers. Generating smooth lobing patterns in 3‐dimensional space requires blending of 2π to 4π radiation patterns with the intrinsic lobe output of the electrical network. Our proprietary networks maximize phase coherence by taking into account the variable radiation area of the piston, inter‐driver distance, and frequency at which the drivers cross. These new formulas dictate that driver filters will be idiotypically different within the array. Phase in this regard includes all distortive effects that compromise the presentation of time in a recording. Digital audio engineers are acutely aware of the bandwidth and transient response compromises inherent in brick‐wall filters vs. more sophisticated methods. The passage of time is an absolute constant; loudspeakers distort or destroy this invariant quantity more than any other component in the audio chain. Most loudspeakers do this poorly while some ignore the existence of time entirely. NO GROUP DELAY DISTORTION No phase errors are acceptable in the last octave between 32Hz and 16 Hz. The goal of no low frequency phase errors and bandwidth below 16 Hz is not achievable with a vented box (phase deviation below cutoff frequency of the port) or a sealed box (impossible to generate low extension at full power without excessive rear cavity size). The only viable solution given the previous design constraints is a transmission line, bringing its own set of problems to be solved. Ripples in the pass‐band that obscure midrange harmonic structure must be eliminated. Aperiodic tapering and internal standing wave control are critical. The best musical blending of the TL with mid‐bass frequencies was achieved through our implementation of tempered scale tuning. The twelfth root of two or 12√2 is an algebraic irrational number which circumscribes the frequency ratio of a semitone in equal temperament in the Western musical scale. This principle was essential in separating standing nodes within the transmission line. In order to reach the low Q levels of the mid‐bass and midrange a high Bl product system had to be created. This was achieved by designing a custom driver with specific characteristics and utilizing it in an isobaric configuration. System Q is 0.5 with electronic manipulation of the lowest octave to give flat frequency and phase response. Electronic low pass filtering extinguishes any ripple in the pass‐band before it interferes with the output of the mid‐bass section. A 3.2 Kw amplifier is directly connected to the TL of each channel within the loudspeaker. All analog circuitry and automatic level sensing retain the sonic character of the amplifier utilized in driving the main frequency bandwidth. MINIMAL INTERMODULATION DISTORTIONS This refers to the distortion between adjacent drivers in the array. Any vibrating surface adjacent to another vibrating surface will induce distortions in the output pattern of that waveform. The solution is to use only the output of each plane in phase, extract any low level vibration from a driver not engaged in the full power output of the system and add specific physical planes to redirect uncorrelated energy. Inert physical cabinet structures are essential in achieving this goal; however no material or combination of materials is immune from vibrating. Excessive reverberations must be eliminated, while tertiary low level vibrations are harmonically integrated into the signal.
SHARP IMAGE EDGE DEFINITION WITHOUT FREQUENCY RESPONSE EXAGGERATION
Analog domain anti‐aliasing filtration eliminates excessive bandwidth from related overlapping wave generators. Noise outside the stop‐band must be specifically limited in order to eliminate the Moiré blurring effect at the edges of images. Proprietary analog filter technologies are applied to eliminate this effect.
MAXIMUM AMBIENCE RETRIEVAL
Most hall ambience is ‐20 db from the fundamental tonal structure and often as low as ‐60 db. Any phase noise or intermodulation at high frequencies will obliterate this information. No break‐up modes are permitted in any driver, especially the tweeter.
LOW NOISE FLOOR
Lost information within the filter network is common in all loudspeakers, hence the misguided idea that simpler networks are better. Poor components will exacerbate these losses in complex circuits. Our solution has been to create proprietary precision elements. These are low‐loss components with specific time constants (known number of micro seconds of storage time), precision resistance (no temperature dependent changes in resistance), and low hysteresis (minimized self‐inductance and low DC resistance through a wide bandwidth). With these component elements it is possible to design complexity into the circuit, conferring significant electrical control over driver output; this would be impossible in a simplistic crossover.
MAXIMIZED DYNAMIC HEADROOM BY ELIMINATION OF CURRENT SATURATION EFFECTS
Pioneering technology in inductor and core design makes possible 10x the current spike within the device before saturation effects are present.
MAXIMIZED TRANSIENT SPEED THROUGH CONSTANT Q AND VANISHINGLY LOW HYSTERESIS Electrical signal contouring is combined through the previously enumerated techniques with careful matching of driver Vas, Bl product, and specifically shaped and damped driver cavities to produce a constant Q design. In this instance constant Q refers to minimal transient ringing at all frequencies within the array as well as keeping the Q consistent from driver to driver. This requires a sealed enclosure in some cases and a vented or transmission line enclosure in others. A new ceramic/honeycomb mid‐bass driver was custom designed for this purpose with an extremely high Bl magnetic structure in an underhung gap for the continuation of the 0.5 Q into the warmth zone of the transducer. New radial magnetic structures are employed in the midrange and tweeter with a combination of sealed transmission lines to continue the minimal ripple transient response concept.
POINT SOURCE FOCUS FOR REALISTIC IMAGE SIZING WITHOUT FREQUENCY DEPENDENT EXAGGERATION
SMOOTH AND EVEN POLAR RESPONSE
CONSTANT FIRST DERIVATIVE OF CROSS SECTIONAL RADIATION AREA
These three criteria are intertwined. In a multi‐driver array only a line source or point source driver arrangement are viable options in maintaining faithfulness to the original recorded information. The same is true for polar response. Without an open omni‐directional or open‐cardioid response, phase errors will enter the presentation post driver/cabinet interface. This problem is not resolvable through room treatment. The most common solution in elementary loudspeaker design is to use first order filters without any equalization in a large complex array. This non point source approach will inevitably produce huge overblown midrange images at the front of the stage and tiny anemic images at the back of the stage. THIS IS THE MOST IGNORED LOUDSPEAKER DISTORTION and one of the most obvious.
A smooth polar response brings evenness of image presentation across the soundstage, while careful attention to radiation area is an added component in the previous discussion of image size. If an array is incorrectly balancing its surface area output, instruments or voices which should be small will appear large and vice versa.
PURELY PISTONIC MOTION THROUGHOUT THE PASS BAND
No driver break‐up modes in the passband. No soft diaphragms that intrinsically distort phase information while coloring harmonic structures with a specific artifact. This is a holdover from incorrectly equating high‐end transduction with instrument amplification and sound reinforcement. In the “players world” we are always searching for a specific sound we have in mind for our instruments and amps. The best playback transducers should be faithfully reproducing the sound (from an amplified instrument, e.g. guitar) that the musician has determined through his choices, not by adding a specific sound of our own. The colorations of various instrument amps (Marshall vs. Fender vs. Hiwatt for example) are significantly due to the exitation of specific cone modes. This is unacceptable in high‐end reproduction. We will allow no phase error or break‐up within ‐60 db of the signal. SYMMETRICAL OVERALL IMPEDANCE CURVE FOR EASE OF DRIVABILITY; DRIVER DAMPING CIRCUITRY FOR PROPER LOAD INTERFACE
These criteria are addressed through a combination of physical driver damping and passive electrical elements which cancel voice coil inductance as well as back EMF. All drivers are considered equal in this, not just the woofers. These issues are essential to the loudspeaker/amplifier interface. By providing a benign load to the amplifier the best (and worst) qualities of the amplifier will be easily heard. We are all aware of loudspeakers that sound quite the same regardless of the amplifier or front‐end driving them. This blunt instrument approach is antithetical to our goal. TOTAL ENERGY RADIATION IS CALCULATED OVER THE ENTIRE SURFACE OF THE LOBE NOT AT ONE MICROPHONE POSITION
Loudspeaker radiation patterns have intrinsic effects on distortion of the waveform being generated as well as in‐room distortion effects from the loudspeaker interacting with the room itself. Judgment of waveform spread should be conducted as though measured from an infinite number of points around the transducer and integrated over the entire radiational area. Measuring from any single point (including the listening position) will not give an accurate representation of the actual signals that impinge upon the listener. Physically large loudspeaker arrays add another problem which is rarely considered by designers; how do we make such an imposing structure sonically disappear? Regardless of how well the initial signal from the transducer integrates with the listening environment, when energy returns from room boundaries to the loudspeaker cabinet, it will be “seen.” If a refrigerator shaped box is added to a room, even without making a sound of its own, it will be heard. Only by carefully designing a “stealth cabinet structure” through reversal of the radiation concepts discussed above will a large array sonically disappear in a listening environment. More than 25 years of dedicated research into the specific physical and electrical phenomena of transduction has provided the foundation for our incomparable line of loudspeakers. Never before have all of the previously discussed qualities been incorporated into a high-energy audio transducer. We have named this culmination of research, design, and aesthetic balance: Tesseract Signature.
Tesseract Signature SPECIFICATIONS
- Response: 16 Hz to 70 kHz (+/- 0.5 dB anechoic, +/- 1 dB in room response typical)
- In room low frequency response: (-1.5 dB @ 16Hz, -3 dB @ 13Hz typical, dependent upon room volume)
- Efficiency: 93.5 dB (1 watt @ 1 meter)
- Harmonic distortion: 0.30 % throughout the bandwidth at rated efficiency
- Inter-driver phase deviation: 3 degrees within all cross bands
- Noise floor: 68 dB from signal level (maximum noise level), 76dB Typical
- Group delay distortion: 3 milliseconds 100 Hz to 16 Hz. 1 millisecond 100 Hz to 20 KHz.
- Impedance: 6 ohms nominal (4 ohms minimum @ 80 Hz)
- Recommended amplifier power: 15 to 500 Watts
- Power: 3.2 Kilowatts per channel
- Automatic signal sensing and gain matching
- Master Gain: +/- 2.5 dB (fixed resistor elements of 0.25 dB increments)
- Tempered-scale aperiodic tuning
- Fully Analog symmetrical circuitry
- No digital signal conversion or manipulation
- 4 x 15" low Q high force factor sub-woofers
- 1 x 11" ceramic/honeycomb underhung voice-coil midbass
- 1 x 4.5" ceramic underhung voice-coil midrange
- 1 x .78 " diamond underhung voice-coil tweeter
- Four independent cabinets, two per channel vertically integrated
- Overall Height: 87" (221 cm)
- Width: 29" (74 cm)
- Depth: 32" (82 cm)
- Weight: 827 lbs. per channel (375 kilos), 1653 lbs. total (750 Kilos)
- Shipping weight: 2002 lbs. (908 kilos)