|
*1. HARMONIC DISTORTION can result when one
or more frequencies are present and is characterized by the generation
of integer multiples, either odd or even, of the original frequencies,
the new frequencies being referred to as "harmonics of the fundamental".
IM (InterModulation) DISTORTION results when
two or more frequencies are present and is characterized by the
generation of sum and difference frequencies that are
not harmonically related to the original frequencies.
Distinct sum / difference frequencies (F2 ± F1) are rarely
generated; instead F2, the higher frequency, will be "smeared" or
"slurred" across the range of frequencies between F2 ± F1,
where F1 is the lower frequency.
Note *7. Dmax - Doppler Distortion
Extensive studies over the last 100 years have consistently revealed
that the human ear is :
(a) Least sensitive
to low-even-order harmonic distortion, possibly because it is most
similar to the harmonics generated by natural musical instruments.
(b) More sensitive
to odd-order harmonic distortion, especially high-order-odd harmonics
in the absence of either odd or even low-order harmonics.
Low-order odd and even harmonics are often referred to as "Soft
Distortion Components" because their contamination of the
original music is least irritating subjectively.
(c)
At least 10 times more sensitive to IM distortion than any form of
harmonic distortion, being the most irritating subjectively.
Unlike the time when only natural musical instruments existed, today's
musicians often make use of the "soft distortion components" produced
by certain types of amplifiers and loudspeakers to create a
particular sound character that is presumed to be pleasing to the
senses; however, the sound system that is used for playback
absolutely should not contaminate the harmonic
structure of the work of music with any NEW distortion
components.
*2.a TIME-ALIGNMENT :
|
Achieving accurate mechanical Time-Alignment for all the drivers in a multi-way
configuration yields a sound character that is closest to the
highly desirable character of a single full-range driver.
|
See Speaker Design Philosophy for an
explanation of single driver limitations.
TIME-ALIGNMENT of speaker drivers is
correct if the energy from the respective drivers for each
frequency range arrives at the listener at exactly the same time.
The vast majority of loudspeaker systems on the market today
have all the drivers mounted on a single flat baffle and thus
suffer from gross mis-alignment; the situation is even worse if a
system with horns has the horn mouth mounted on the baffle forcing
the actual sound source at the driver diaphragm to be pushed back
inside the cabinet. Some manufacturers justify this
configuration by introducing a phase shift into the crossover
network or reversing driver polarity such that a continuous
sine wave test yields nearly flat frequency response.
However, this approach fails when the system is tested with an
impulse function, the result being gross "time-smear" with multiple
arrival peaks which are clearly visible and distinct.
"Music" is not a constant or slowly varying sine wave, but
continuously varying bursts of complex waveforms, and the human
ear can, in some individuals, clearly hear the improvement when the
drivers are correctly aligned mechanically (and correct driver polarity
restored, phase-shift networks removed, in either active or passive
crossover implementations).
*2.b CLARIFICATION OF TRANSIENT / PHASE RESPONSE REQUIREMENTS:
|
It is a common misconception in the industry that an audio crossover
configuration exhibiting a 360° phase shift can be assumed to be
equivalent to a phase shift of 0° and thus be "back in phase"
or have "correct polarity". A simple frequency sweep
will show that the phase starts at 0°, progresses to 90°,
then to 180°, then continues on to 270° (not back to
90°, then back to 0°), ending at 360°.
A calculation of group delay (rate of change in phase vs. frequency)
also verifies this misconception.
|
There has recently been much misplaced emphasis
by certain groups on the ability of a system to accurately
reproduce an impulse, with no regard for what type of inaccuracy
is audible. By definition, any phase shift
across the spectrum implies an imperfect impulse response.
The effect of a gradual phase shift introduced by a crossover
on the impulse response is quite different from the effect due to
mechanical mis-alignment of speaker drivers.
The "Energy-Time Curve", which is calculated from the impulse response,
provides a better way to observe the true nature of the effect.
The gradual phase shift produced by a properly designed crossover
results in a slight broadening of a single
impulse-energy-peak envelope; mechanical mis-alignment of drivers
results in multiple arrival peaks delayed in time.
Extensive psycho-acoustic testing by us as well as many other
reputable engineers and audiophiles has shown that the human ear
cannot detect a gradual phase shift such as that introduced by a
properly implemented 3rd order Butterworth crossover
with correctly Time-Aligned Drivers.
Our crossover and all of our speaker systems meet these
conditions.
*3.a BALANCED I/O - INVERTING GAIN STAGE, CABLE LESS THAN 50 FEET :
It is a popular myth that
Balanced I/O provides superior audio performance;
at some point a gain stage must be operated in non-inverting
mode, thus insuring increased distortion. Since
balanced I/O requires twice as many gain stages, the distortion
problem is further aggravated. The assumption
that Common Mode Rejection provides a useful improvement
is also erroneous since CMRR falls significantly to inconsequential
levels with rising frequency where the supposed improvements
would be most needed; if any common mode noise /
voltage exceeds the input capabilities of the gain stage,
permanent damage will occur. Balanced I/O was
originally conceived to reduce low-frequency hum / noise pickup
for long cable runs where non-zero output impedance in the
driving stage was necessary (as in obsolete 600 Ω
systems where achieving lowest distortion had low
priority and was ignored or assumed to be "good enough").
A properly designed single-ended configuration with
True Virtual Ground at the output stage
and Inverting-Only Gain Stages provides
superior performance. The Input DC blocking capacitor
shown below is not used in any of our equipment, except for
power amps, where it is absolutely necessary for driver protection.
*3.b TRUE VIRTUAL GROUND :
Most audio equipment on the market today requires the output
stage to have an output resistance of 50 Ω to as high as
1 KΩ in order to ensure stability when cables are attached
and these unusually high values of output resistance will most
certainly ensure a high susceptibility to hum / noise contamination.
However, ALL of our equipment has a proprietary configuration
for the output stage which exhibits True Virtual
Ground while maintaining stability, and thus does not
require nor benefit from a balanced I/O configuration.
A Virtual Ground output impedance effectively
short-circuits-to-ground any stray contamination that might
be introduced by the connecting cables.
Note *10. Exotic / Expensive Analog Cable
It is well known in the industry that virtually all op-amps
exhibit higher distortion when operated in non-inverting mode.
The source of the problem is input-capacitance-to-substrate
modulation and the effect is minimized by operating in inverting
mode. It should also be noted that computer models
of discrete amplifier designs also predict lower distortion
when operated in inverting mode, and this has been confirmed
by our own experimental results.
CABLE LENGTH GREATER THAN 50 FEET :
For longer cable lengths, increasing the wire gauge will provide
acceptable hum rejection with the single-ended configuration;
however, for very long cable lengths in noisy environments,
the only viable option for achieving minimum low frequency
hum / noise may be the balanced configuration with it's inherent flaws.
*4. DIELECTRIC ABSORPTION IN CAPACITORS :
The dielectric in a capacitor is the material placed between
the capacitor plates to increase the breakdown voltage and
capacitance for a given plate spacing / surface area.
Unfortunately, dielectric materials that provide the greatest
increase also suffer from the highest dielectric absorption.
Dielectric absorption refers to the tendency of the material
to retain a residual charge, even after the previously
charged capacitor has been discharged. The
deleterious effect is exhibited when the voltage across the
capacitor varies, as it does when a capacitor is in the audio
signal / feedback path. Before the
importance of this phenomenon was acknowledged by the audio
industry, it was widely accepted as a routine fact in the
field of data acquisition, where low DA capacitors were required
in the sample-hold circuitry of early A/D converters.
The dielectric absorption phenomenon is the ultimate source
of distortion in all capacitors; all other capacitor
parameters such as ESR (Equivalent Series Resistance) are
constant and do not introduce non-linearities.
Electrolytics and ceramics are worst-case examples and should
be avoided at all cost; all film capacitors, but especially
polypropylene film, have the lowest coefficient.
|
It is a popular myth that using
back-to-back electrolytic or non-polar electrolytic capacitors
in parallel with a high-quality
polypropylene will improve performance; the extremely high dielectric
absorption of the electrolytics adds to and swamps the extremely
low dielectric absorption of the polypropylene.
|
Conversely, when capacitors are used in power supplies for
filter / bypass functions where the voltage across the capacitor
is essentially constant, the dielectric absorption phenomenon
is not manifested and has no effect on the audio signal.
For this application the relevant
parameters are ESR and leakage, but especially inductance,
which can effectively decouple the capacitor at high
frequencies, thus severely compromising the theoretical
Virtual Ground that supply rails are
supposed to exhibit. Bypassing large electrolytic
capacitors which might have high inductance with a high-frequency
ceramic or film capacitor CAN improve
their filter / bypass effectiveness and possibly result in
improved sonic performance.
*5. DRIVER-BOX TUNING ALIGNMENT : A loudspeaker in a
vented box has amplitude and phase characteristics identical to
an electronic 4th order high-pass filter. For certain
values of speaker Qt, the response acts as quasi-3rd order.
With appropriate active EQ (our -5 EQ), implemented with the
LFEQ module in our
3PX8 Electronic Xover,
the more desirable 2nd order sealed-box response
(i.e., less phase shift at the lowest frequencies) can
be achieved while maintaining vented-box advantages (lower freq.
response, excursion, distortion); the box volume can also be
smaller. This is not one of the classic
Thiele-Small
alignments, but is a proprietary design achieved through extensive
computer modeling and FFT measurement analysis.
Our proprietary sealed-box mid-bass alignment also uses
LFEQ to mirror low-frequency rolloff / phase shift, allowing
virtually ideal frequency response / Time-Alignment*2
in the crossover region to be achieved.
*6. KA=1 : This term applies only to
direct radiators where KA=2π/λ; A=radius,
λ=wavelength. At the frequency where KA = 2.4,
the total power radiated by the driver is half (-3 dB).
On-axis response usually does not drop though, due to "beaming";
the power drop is realized as a loss in dispersion with
increasing frequency.
For frequencies below KA = 1, the driver is essentially a uniform
half-space radiator with near constant dispersion at all
frequencies. It is believed that the narrowing
dispersion with increasing frequency above KA = 1 results in
loss of sound stage width and overall naturalness of music.
*7. Dmax : Dmax is a measure of modulation /
Doppler distortion in loudspeakers (synonymous to IM distortion
in electronic equipment), and decreases as radiating area increases
and diaphragm excursion decreases. The Doppler Effect
should be familiar to anyone who has waited at a railroad crossing
and noticed that the pitch of the horn on the oncoming train increased,
then decreased as the train passed by. During the time a
loudspeaker cone is moving forward producing a low-frequency note,
any high-frequency notes will be increased in pitch; when the cone
moves backward, their pitch will be decreased. This
phenomenon is ALWAYS occurring,
it cannot be eliminated --
the goal is to reduce it's effect to inaudible levels.
1. It defies the laws of physics for a small
diameter driver with "long-throw" large-excursion to produce
acceptable quality sound.
2. It especially defies the laws of physics for
a single small diameter midrange or dome tweeter to
produce acceptable quality sound when operated over typical
bandwidth ratios.
If a 15" low-frequency driver (125" sq.) operating from 20 Hz
to 100 Hz, a bandwidth ratio of 5, is required to achieve an acceptable
value of Dmax, then the frequency range 2 KHz to 10 KHz will
also require the same radiating area for equally low
modulation (IM) distortion.
|
The value for Dmax in our data sheets
is calculated at 1 acoustic watt for the total radiating
area at the specified bandwidth. A loudspeaker with a
reference efficiency of 1%, corresponding to 92 dB/W/M, would
require 100 electrical watts to produce 1 acoustic watt.
In the S18X section above, a 125 Hz tone
would deviate +/- 2 Hz if the diaphragm were simultaneously
radiating 25 Hz @ 1 acoustic watt. Since the reference
efficiency for the S18X is 95 dB/W/M, or
2 %, this distortion level would occur at electrical power of 50
watts.
Horn loading significantly reduces the required diaphragm excursion
to produce a given sound level; thus, horn loaded compression
drivers inherently exhibit very low values of Dmax and IM
distortion.
Psycho-acoustic tests have shown that maximum deviations of 2 Hz
or less are not audible with pure tones; with music, deviations
of 10 Hz and greater are increasingly perceptible and obnoxious.
Modulation distortion is the primary distortion source in
all loudspeakers and a major contributing factor to
"listening fatigue".
The concept of Dmax was
first presented in AUDIO,
Dec. 1977, in an article by Stephen Kurtin.
|
Extensive studies over the last 100 years
have consistently revealed that the human ear is at least 10
times more sensitive to IM distortion than harmonic distortion.
|
*8. FOAM SURROUND is widely accepted as being
by far the most linear material to use where any appreciable cone
excursion will occur, as in low frequency bass drivers; however,
it has a limited life expectancy, even under the best environmental
conditions. Treated Cloth Accordion,
where a proprietary damping compound is applied, is probably the 2nd
best choice and will last virtually forever.
*9. SYMMETRICAL FLUX POLE PIECE with
Flux Stabilizing Ring and
Pole Piece Vent are design features
that significantly reduce a fundamental source of harmonic
distortion that exists in all loudspeaker drivers.
This phenomenon is of consequence for frequencies below 500 Hz
where voice coil / cone excursion is greatest.
The pole piece show below is sometimes referred to as a "T pole" or
"undercut pole". An "extended pole piece" is also
an effective geometry.
Auratron Systems Direct Radiator Driver
|
Conventional Driver
|
*10. EXOTIC / EXPENSIVE ANALOG CABLE :
A straight-forward analysis will show that the best choice for cable
for ANY data transmission will be the one
with the least amount of inductance / capacitance. Since
"Exotic / Expensive Analog Cables" typically have
inordinately higher inductance and especially capacitance, they are
NOT a good choice.
Assuming the equipment has properly designed driving / receiving stages
where the output impedance is Virtual Ground
*3,
the most neutral, accurate sound will result when using the lowest
inductance / capacitance cable.
The change in sound character observed when using different cables
results from unwanted interactions of the driving source and/or the
receiving load with the cable which may result in
an increase OR decrease in high-frequency response,
or at worst, increased IM distortion
due to the onset of oscillations or ringing. This
interaction is unpredictable, and will be perceived differently by
different listeners as sounding "better"
OR "worse", depending on
the listener's personal preferences. It should be noted
that the same effect can be achieved by using the cable with lowest
inductance / capacitance and adding a 50 cent capacitor / inductor
to mimic the "Exotic / Expensive Analog Cable"
loading effects.
For low-level signals with > 10 KΩ
load resistance, "Skin Effects"
are also negligible up to several hundred KHz.
Our analog equipment is designed with a proprietary output stage
that is not sensitive to low-to-moderate values of cable
inductance / capacitance and has an output impedance of
True Virtual Ground;
thus the sound character of our equipment exhibits negligible
change when using different kinds of cable. Also, since our designs are based
on electronic crossover where the power amps are placed close to the
speakers, a typical installation might use 3 or 4 feet of speaker cable
making "Speaker Cable" interaction virtually non-existent.
*11. EXOTIC / EXPENSIVE DIGITAL CABLE :
Since our DA-192X utilizes
Upsampling Sample Rate Conversion,
which re-clocks the incoming digital signal, and the
digital cables for a typical installation are only a few feet long
at most, there is no improvement in performance by using
"Exotic / Expensive Digital Cable".
In any event, as long as there is no impedance mismatch
and excessive noise / jitter does not cause missed / repeated / erroneous
samples, there is no effect on audio quality,
EXCEPT: In cases where
the output clock to the D/A converter is regenerated from a noisy
transmission signal which ultimately causes excessive amounts of
analog IM distortion -- Re-clocking essentially eliminates
this problem.
*12.a CD COPY SOUNDS BETTER THAN ORIGINAL :
|
It is a popular myth that digital audio CD's always provide an
accurate raw digital data stream.
|
Un-correctable errors in the raw digital data stream routinely
occur with many audio CD players. The problem is a
result of varying laser head characteristics in conjunction with
varying CD media properties. Factory-new CD's routinely
contain blemishes in the media substrate which affect the reflectivity
in such a way that many CD players cannot negotiate, resulting in
un-correctable errors. While the error correction
algorithm may accurately deal with "pops, clicks, or scratches"
with accurate data on either side of the blemish,
it is incapable of dealing with overall poor reflectivity errors.
There is a high probability that another CD player/reader, especially
one designed for computer data, may be able to accurately read
the disc. (Especially when
using a superior extraction program such as
EAC.)
If this happens, the copy will contain the
accurate data stream and most likely will not have the substrate
imperfections, will play accurately on the audio CD player, yielding
improved sound quality. By this same reasoning, the
copy containing the accurately extracted data might instead yield
degraded quality if the audio player is
less able to read recordable CD media which has different reflective
properties than factory-processed CD's. The decoder chip
in virtually all CD players handles un-correctable errors by "holding
the last sample", the effect being equivalent to a reduced sample
rate for the duration of the errors.
Thus, the deleterious
effects of un-correctable errors begin at the highest frequencies
and become more audible at progressively lower frequencies as
the errors become more severe. Loss of clarity in the
upper mid-range / high frequencies causes loss of stereo imaging
and the illusion of dimension as well as sounding dull, flat,
and lifeless.
It should be noted that the Error Correction Algorithm used to
encode audio CD's is NOT the same as for digital computer data.
The audio algorithm was chosen from the beginning to give highest
priority to masking un-correctable errors in an attempt to promote
the new technology. Audiophiles immediately recognize
the flaws without necessarily understanding the technical nature
of the degraded performance, as described above.
*12.b LP SOUNDS BETTER THAN CD :
Quite simply, the sample rate of the conventional CD is too low.
Our experiments have shown that increasing the sample rate to
192 KHz results in a sound character that is virtually indistinguishable
from the original source. This is true whether the bit depth
is 16 or 24, with 24 bits yielding only a subtle improvement.
Digital filters only give "the right answer" for continuous
non-varying sine waves, especially at high frequencies where the
number of calculated samples constitute the largest proportion of
samples for a single cycle of the reproduced wave.
Increasing the basic sample rate and using little or no digital
filtering can yield a sound character that rivals or surpasses
conventional analog.
To make matters even worse, most commercially produced CD's have
been dramatically overprocessed by
compressors / limiters in a
misguided attempt to "make the sound fit" into the 16 bits of
resolution. Our experiments in transcribing LP's to CD
have revealed that the best sound results from simple recording at
a reasonable level with no extraneous processing.
While some LP's may probably have been subjected to some compressor /
limiter processing, the best sounding ones have not.
Compressors / limiters are the bane of the audio industry, only
being used by misguided or lazy / incompetent sound engineers.
*12.c SACD, DSD vs DVD-A :
Also, it is theoretically impossible for DSD to rival true DVD-A
(96KHz, 24 bits) because the actual bit rate for DSD is
significantly lower than true DVD-A; DSD tries to compensate by
using "noise-shaping" and "masking", but these techniques ultimately
fail when compared to true DVD-A on an acoustically transparent
playback system. DVD-A at 192 KHz represents the ultimate
level of performance for digital audio and is virtually indistinguishable
from it's analog counterpart.
*12.d Digital Equalizer Distortion :
All of the digital equalizers / software
we have tested introduce an unacceptable "grungy digital sound",
regardless of which band(s) are adjusted or the amount of headroom.
It is our assertion that ALL digital equalizers
are inherently flawed and that it is theoretically impossible to
achieve the level of transparency afforded by properly
implemented analog EQ circuits.
A straightforward analysis of the logical extension of digital interpolation
theory and it's inherent flaws yields this conclusion.
In fact, all of these types of digital manipulations represent
a "worst case situation" because unwanted digital artifacts
are generated in the audible range.
The Remote Unit introduces
no audible / measurable increase in distortion, even when all bands are
set to maximum boost / cut.
Also See Note *13. Room Equalizers
*13. ROOM EQUALIZER :
|
It is a popular myth that "Equalizing The Room" can provide a
useful improvement in the listening experience.
|
While it is possible to set up a measurement / calibration
scheme that will provide reasonably flat average response at
a particular listening position, the result will not correlate
with perceived spectral balance to the listener.
The human ear uses "first sound" to determine frequency
response with much lesser influence from reflections / ambience;
also, the degree of influence is frequency dependent.
A multi-band equalizer or tone controls are far more useful in
correcting the spectral imbalance that occurs in most recordings.
This feature can only be effectively realized if the listener
has instant easy access to the EQ controls, as with the
Remote Unit.
Also See Note *12.d Digital Equalizer Distortion
*14. COMPRESSORS / LIMITERS, VCA's, and VCR's
|
It is a popular myth that the function performed by COMPRESSORS
/ LIMITERS will produce an improvement in recording quality.
Eliminating a few milliseconds of possible overdrive / clipping
at the expense of an overall recording possessing the deleterious
effects of compression plus increased distortion is not a
rational endeavor.
|
All compressors / limiters are realized by a circuit element
referred to as
Voltage-Controlled-Amplifier / gain stage
(VCA's) or
Voltage-Controlled-Resistor (VCR's),
and ALL of these devices exhibit
nonlinearities because, for a given constant control voltage,
their resistance / gain varies with signal
level, the end result
being the introduction of unacceptably high distortion levels.
Once the device is active in the signal path, the distortion occurs
regardless of signal level or the value of the control voltage.
The excuse "Oh well, the element is in the feedback path so it doesn't
count" is also erroneous since, in some topologies, the nonlinearities
can actually be multiplied by the loop gain of the stage.
Even if the circuit were "distortionless", the effect of compressors
/ limiters is to introduce a "canned", lifeless sound character
lacking sparkle and the illusion of dimension, the very attributes
that true audiophiles expect and appreciate.
Compressors / limiters are the bane of the audio industry, only
being used by misguided or lazy / incompetent sound engineers.
*15. Variable Pitch
A 3.3% change in speed will alter pitch one half step;
6% results in a full sharp or flat.
Traditional turntables, market oriented mainly for club DJ's, had
precisely calibrated strobe dots to indicate 0%, ±3.3%, and +6%
(a feature wasted because DJ's could care less, only being
interested in "matching the beats").
However, many true audiophiles appreciated this level of precision
and found that having the instruments "accurately tuned sharp" added sparkle
and a sense of vivacity, regardless of the "kind of music".
Conversely, an indiscriminate change in pitch might cause the
music to "sound weird" or off-key.
Another aspect of altering the speed is a change in tempo;
however, this is of minimal importance since various renditions of
a particular piece of music with various tempos played at "correct
pitch" do not exhibit the effect invoked by varying the pitch.
This was the origin of "true hardware pitch". In the late
1950's and 60's most radio stations routinely played popular music
selections with increased speed (played them "hot") to squeeze in
more commercial time; when fans rushed out to buy their favorite
new song, it actually sounded like a completely different "tired"
rendition than what was heard on the radio.
With digital format music, a change in pitch can be accomplished
by "hardware" or "software". "True hardware pitch"
is accomplished by actually varying the master clock rate by the
required percentage and is equivalent to and has the inherent
high quality of turntable pitch control; all of the original
samples are preserved and passed through. A beneficial
side effect is slightly increased high-frequency bandwidth.
"Software pitch" is accomplished by a DSP algorithm, maintains the
same output sample rate, and suffers from "that grungy
digital sound". While "software pitch"
allows either pitch or tempo to be varied while keeping the other constant,
the diminished fidelity is not a reasonable trade-off, especially
since keeping tempo constant has no benefit.
Misguided engineers who employ
"software pitch" in their equipment espouse the erroneous claim
that outboard D/A converters or other processing steps might not
lock onto the non-standard sample rate. However,
virtually all outboard processors use S/PDIF receivers (as in our
DA-192X) that will reliably lock to any
frequency from 32 to 200 KHz; the non-standard data rate
can then be converted, if necessary, to any new standard rate
using the highly desirable ASRC Upsampler.
Our experiments with a variety of components and software have revealed
that an ASRC Up/Downsampler processing an input
that has been "hardware pitch" altered will
provide far superior performance than a DSP algorithm implementing
"software pitch / tempo".
The argument that the variable clock generator might have high
jitter is also misguided, as the inherent reclocking
of the ASRC upsampling removes virtually all
incoming jitter. Even if the altered master clock is
not standardized, it is our experience that all of the IC's used to
implement A/D and D/A conversion are designed to function properly
with faster or slower MCK's, this fact also being touted as a
marketing plus. Some manufacturers actually
denigrate their equipment by deliberately disabling the
"Digital Out" when pitch is varied.
|
It is a popular myth that each piece of digital processing
equipment should have it's own precision clock.
It defies the laws of probability that
any two crystals / clocks will run at exactly
the same frequency; one will always be faster or slower
and eventually samples will be indeterminate / dropped / repeated.
A thoughtful analysis will show that, in order to ensure
accurate data transmission, there should always be one master
precision clock, with all others being regenerated as slaves
with the master originating in the data source machine.
If the data source uses a non-precision clock as when implementing
vari-pitch, all intermediate devices should still slave from
this clock up to the input of the ASRC Up/Downsampler.
|
|
Loudness Control compensates for Fletcher-Munson
hearing characteristics at soft listening levels.
Also provides The Ultimate In Listening Convenience
by allowing ALL critical parameters to be adjusted instantly at the
listening location without distraction.
Construction techniques for a box to be used as a loudspeaker enclosure
are quite different and more complex than for a box to be used as a
cabinet or piece of furniture. Our enclosures are constructed
from high-density low-resonance MDF board using industrial glue,
supports along all edges, and stiffeners across large panels
to minimize box flexure and spurious resonances. Laminate
finish is applied with rubber-base contact cement resulting in a
TRI-DENSITY sandwich with superior
damping properties; it is impossible to achieve this high degree
of mechanical / acoustic damping with hardwood / veneer construction.
All interior surfaces are lined with an optimum ratio of acoustic
absorption material to dampen the rear pressure wave and further
minimize harmonic resonances.
