This is only a preview of the November 2001 issue of Silicon Chip. You can view 29 of the 96 pages in the full issue, including the advertisments. For full access, purchase the issue for $10.00 or subscribe for access to the latest issues. Items relevant to "100W RMS/Channel Stereo Amplifier; Pt.1":
Items relevant to "A Neon Tube Modulator For Cars":
Articles in this series:
Items relevant to "A Low-Cost Audio/Video Distribution Amplifier":
Items relevant to "Short Message Recorder & Player":
Articles in this series:
Purchase a printed copy of this issue for $10.00. |
Recently Australian speaker
designer and manufacturer
VAF Research introduced a
new high-performance Public
Address speaker to its range
of highly regarded hifi
and home theatre speakers.
And they’ve been run off their
feet ever since! We asked Philip Vafiadis, of VAF Research, to
explain the philosophy behind
the design of the I-201 Public
Address Speakers.
Defining
The Ideal
Public Address
Loudspeaker
L
OUDSPEAKERS play an impor
tant role in a public address sys
tem (or for that matter any sound
reproduction system) as the final link
in the signal processing chain.
They convert electrical energy from
the power amplifier into acoustic
energy in air that travels as sound
waves to the listeners. Regardless of
the quality of the preceding signal
processing chain, if the loudspeakers
are of poor quality or incorrectly connected or operated, the result will be
poor quality sound.
Typically, the performance of loudspeakers is orders of magnitude (ie,
multiples to the power of 10) worse
than what we would accept from other
6 Silicon Chip
audio processing devices.
At some time or other, we have
all struggled to understand a public
speaker in a church or school hall
amplified with the typical “column”
or horn loaded public address loudspeaker system. In fact, most people
have concluded that high-quality
vocal reproduction in a reverberant environment is difficult, if not
impossible.
So what exactly are the requirements for a public address loudspeaker?
Well, it must reproduce acoustically the electrical input signal at an
adequate level to be heard, without
introducing distortion or colouration.
The loudspeaker must accurately
match the ‘timbre’ of the voice or instrument it is reproducing. The sound
should be clear and intelligible for
each listener, even though the listeners may be widely dispersed in three
dimensions. If used inside an enclosed
space, it must do this with the added
encumbrance of the superimposed
room acoustics.
The loudspeaker should not be
prone to feedback or howl-around,
when used with open microphones.
From a practical point of view, it
should be small, light and visually
unobtrusive. It should be physically
constructed in such a way that it can
be installed in optimal positions, both
www.siliconchip.com.au
acoustically and aesthetically. Finally,
it must connect and function reliably.
There are many methods of converting electrical energy into acoustical energy (including some esoteric
ones) but the overwhelming majority
of loudspeakers use electrodynamic
transducers constructed with voice
coils in permanent magnet fields
driving a moving diaphragm. Electrodynamic transducers have so far
proven to offer the best balance of
performance and ruggedness at an
affordable price.
To achieve the performance ideals
listed above, the following areas of
loudspeaker system performance are
important:
♦ Time Alignment and Source
Coincidence,
♦ Controlled Directivity, and
♦ Stored Energy.
Time Alignment and Source
Coincidence
Currently available electrodynamic
transducer technology dictates that a
full bandwidth response requires two
or more drivers to handle high and low
frequency ranges.
Therefore, the acoustic output of
a loudspeaker system is the sum of
the outputs of its individual drivers,
which are in physically different locations on the baffle.
Because of the displacement between drivers, the acoustic transmission path distance from the listener
to each driver’s acoustic centre will
be different for each listening position. Therefore, in-phase arrivals of
two driver’s outputs can only occur
exactly at one point in space for one
listener where the path distances are
equal (or offset as necessary). Typically, this point will be along the
www.siliconchip.com.au
loudspeaker’s main axis.
Thus, a typical loudspeaker can
only be exactly “time aligned” at one
point in space. At all other locations
it will be time misaligned! This misalignment causes ripple in the frequency response for about an octave
either side of the crossover frequency.
Unfortunately, this is typically where
the ear is most sensitive, in the region
from 1-4kHz!
The individual driver outputs from
a 2-way horn-loaded system at best can
only sum properly in one plane and
at worst along only one axis. Above
and below the vertical axis, the path
distances vary significantly between
the horn and bass driver and correct
reconstruction of the audio signal
can not, and does not, occur. Even
side to side, the effective path length
through the horn can vary enough
to cause ripple in off-axis frequency
responses.
In order to prevent the problem of
different path lengths to the listener
from each driver in a multi-way system, the drivers’ effective acoustic
centres must be coincident. This can
be achieved either by coaxial drivers,
or by a mirror image array, with appropriate signal processing delays if necessary. Note that either arrangement
of drivers in itself does not guarantee
coincidence or time alignment!
The high-frequency driver in a coaxial design has its terminating acoustic
load modulated by the position of
the bass driver’s cone, which in turn
leads to high levels of intermodulation
distortion. Unlike coaxial designs,
the high-frequency driver in a mirror
image array is decoupled from displacement modulation effects caused
by the bass drivers and has low levels
of intermodulation distortion.
Controlled Directivity
So called “high Q” or Controlled
Directivity loudspeaker systems are all
the rage today. Unfortunately, there are
a number of false premises on which
this trend has been based. First is the
notion of matching the loudspeaker
coverage pattern to audience area. In
fact, if we reverse engineer the ideal
coverage pattern for a loudspeaker
in a typical application using a CAD
simulation program, it can be shown
that normally a conical or slightly
elliptical (“squashed” conical) polar
November 2001 7
speaker systems is that by definition
they have severe discontinuities in
their power responses. In other words,
they fail to deliver the promise of clearer, more articulate sound. The bottom
line is that a “low Q” loudspeaker
system will sound more musical and
need less equalisation than a “high
Q” system.
Feedback Stability Margin
The I-201 from VAF Research is
ideal for live voice or music and works equally well
as a front-of-house, foldback, monitor, installed, arrayed or portable system.
pattern is normally the best fit.
A second false premise is the notion
of minimising reflections off nearby
walls for improved intelligibility. In
fact, reflections in the first 20 milliseconds or so enhance intelligibility and
reflections from walls much further
away are typically attenuated enough
by inverse square law so as not to be
significant.
A third false premise is the notion
of minimising the reverberant energy field in the room, to maximise
intelligibility and articulation. In
fact, it is generally more significant
that the reverberant sound field has
even energy against frequency. In
other words, articulate, intelligible speech is possible in a highly
reverberant room despite the long
reverberation time.
8 Silicon Chip
For musicality and intelligibility
generally, it is important that a loudspeaker system delivers a smooth power response. Power response is defined
as the total acoustic power output in
all directions against frequency, as
opposed to frequency response, which
only considers one point in space. Of
course, the frequency response is important too but in all directions! Power
response can be considered like an
average of frequency responses taken
in all directions.
A smooth power response will mean
that the decaying reverberant sound
field in a room will more closely match
the direct sound field. This is important to achieve a natural sound quality
for both voice and music, and actually
makes speech easier to understand.
The problem with horn loaded loud-
It is a commonly held belief that
feedback in a sound system is a result
of excessive gain at one frequency and
can be controlled by reducing the gain
at that frequency. And yes, those two
statements are true – but there’s more
to it than that.
Feedback is always occurring in a
public address system with an open
microphone, because sound reproduced by the loudspeakers will be
picked up by the microphone and
reproduced by the loudspeakers and
around the loop it goes again.
Normally the total gain or amplification around the loop is less than
unity, so the system remains stable and
useable. Essentially, this is because the
rate at which sound energy is being
added is less than the rate at which it
is naturally decaying.
If, however, the gain around the
loop exceeds unity, then the sound
level will build up rapidly, causing
the squealing known as feedback. Because of the imperfections of typical
sound equipment, there will be a small
number of frequencies with more gain
around the loop than general and these
frequencies will be the ones which
“take off” when the system goes into
instability.
Perhaps surprisingly, these frequencies do not necessarily relate to peaks
in the frequency response. The time for
www.siliconchip.com.au
SPECIFICATIONS
Frequency Range ����������������45Hz to 25kHz
Frequency Response �����������65Hz to 20kHz (±2.5dB) with grilles
Power Handling �������������������For amplifiers to 200W RMS
Impedance (nominal) ����������4W
Sensitivity ���������������������������95dB (1W <at> 1m)
Enclosure type ��������������������35-litre, vented
Drivers ��������������������������������2 x 210mm fibreglass cone woofers with 37.5mm aluminium
voice coil and synthetic high compliance surround; 1 x 25mm
impedance-matched soft dome tweeter with aluminium voice
coil. Tweeter protection provided.
Crossover ����������������������������Displaced pole third order
Crossover Components �������2% tolerance air-cored heavy-gauge inductors; close-tolerance
metallised polypropylene capacitors.
Crossover Frequency ����������2700Hz
Connectors ��������������������������2 x Neutrik Speakon™ 4-pin connectors (in/out).
Enclosure ����������������������������Arrayable (90° arc) and stackable, 18mm MDF with internal
bracing. Finished in textured black. Black perforated steel front
grille.
Finish ����������������������������������Matte black. Black perforated steel front grille.
Mounting ����������������������������Top hat suitable for 35mm poles. (Other options on request).
Dimensions (mm) ���������������640 (h) x 490 (widest point) x 320 (deepest point)
Weight ��������������������������������18 kg
sound to travel from the loudspeaker
to the microphone is typically 5-25
milliseconds, which is a lot of cycles
at voice frequencies.
In other words, there is plenty of
time for the sound energy level to decay during the propagation around the
loop – unless a resonant mechanism
is at play. Resonances cause energy to
linger at specific frequencies, making
the system prone to feedback at those
frequencies.
Sources of resonance (or energy storage) in loudspeakers include, but are
not limited to, misaligned crossovers,
cabinet panels, dust caps, diaphragm
break-up modes, and cavities including those in the horns of horn-loaded
systems. Unless well damped, these
resonances will be the feedback break
points in the system.
Eliminating or at least severely attenuating system resonances improves
the feedback stability margin to near
the theoretical maximum. In practice,
this provides more than enough gain
before feedback for most applications,
even in difficult acoustic environments such as churches, provided
good quality microphones are used.
Horn-loaded loudspeakers are at
a distinct disadvantage in feedback
stability margin because of the number of resonant modes supported in
a typical horn. Often the assumed
gain-before-feedback advantage of
controlled directivity is more than
traded off because of this.
www.siliconchip.com.au
Kits
without
compromise
VAF
Introducing the VAF Research I-201
Recognising that loudspeakers are
by far the weakest link in the public address audio chain today, VAF
Research has used CAD (Computer
Aided Design) and CAM (Computer
Aided Manufacture) techniques to
develop a new high-performance public address loudspeaker system, the
I-201.
It is intended for use as a general
purpose PA loudspeaker in small to
medium-sized performance spaces. It
is ideal for churches, hotels, restaurants, function centres and other live
performance spaces. The I-201 can be
used for live voice or music, foldback
monitoring or recorded program reproduction.
Multi-purpose cabinet design
The unique cabinet geometry of the
VAF Research I-201 lends itself to almost all applications. The ideal angle
for a foldback wedge, the I-201 also sits
up closer to a side-wall than a traditional trapezoidal design, minimising
aesthetic and sight line problems
in critical applications. Two I-201s
will provide for 180° coverage from
a side-wall position and four I-201s
form a perfect circle for all-round
coverage in a gymnasium, for example.
Where additional reach is required,
I-201s can be stacked vertically and/or
“Sound quality to die for”
Rolling Stone Magazine
“..A new benchmark in
every criteria”
Best Buys Home Theatre
Speaker Kits
without compromise
from $312 pr to $8,863 pr
FreeCall 1800 818882
www.vaf.com.au
vaf<at>vaf.com.au
November 2001 9
I-201 HIGHLIGHTS
♦
♦
♦
♦
♦
♦
♦
♦
♦
♦
♦
♦
♦
Affordable, multi-purpose arrayable, stackable cabinet design.
Works equally well as Front-of-House, Foldback, Monitor, Installed, Arrayed or Portable system.
Full-bandwidth design for music reproduction with reduced need for subwoofers.
Breakthrough low stored energy design for high feedback-stability margin in reverberant environments.
High internal acoustic damping, MDF cabinet construction – not a plastic box!
Powder-coated, anti-resonant, dent resistant, curved metal grille.
True acoustic time-aligned point source – no electronic processing required.
Internal crossover with high-frequency driver protection.
Flush, concealed carry handle positioned on centre of gravity for easy carrying.
Recessed metal terminal plate with dual linked Speakon™ style connectors.
Flush 38mm stand-mount adapter in base of cabinet.
25mm dome high-power tweeter with acoustic dispersion control.
Dual 200mm fibreglass long-throw precision bass drivers.
the angle between pairs reduced from
90° to as little as 0°.
Full bandwidth
audiophile quality
Unlike other small public address
loudspeakers, the VAF Research
I-201’s have been engineered to provide strong usable bass for full-range
music reproduction. What’s more, the
high frequency response extends well
past 20kHz in a typical PA loudspeaker
fashion!
True point source
time alignment
The I-201’s Vertical Mirror Image Array with time-aligned drivers behaves
as a true point source at all frequencies
in all directions. The effective acoustical centre of the two bass drivers
always remains at the same point as
the effective acoustical centre of the
tweeter for waves radiated in any di-
10 Silicon Chip
rection. Acoustical reconstruction is
accurate regardless of polar direction,
giving the I-201 the smoothest total
power response possible.
Feedback
The I-201’s are an ideal choice for a
very reverberant hall, in spite of their
“low Q” design. Combined with reasonable condenser microphones, there
will be plenty of loop gain available,
even for a quiet talker standing behind
a lectern.
There are three secrets to this success. First, I-201’s are a low stored energy system, much lower than typical
PA designs.
Second, typically only one cabinet
a side is needed for coverage, which
eliminates phasing between boxes and
the consequent side lobes. And third,
the polar response of the I-201 is very
smooth, thanks to unique diffraction
control incorporated into the baffle.
Pricing
The new VAF I-201 is available
only direct from VAF Research, in the
following forms:
♦ Fully assembled and tested: $1100
each.
♦ Ready to assemble kits where the
cabinets are fully built and finished
leaving only the final assembly: $940
each.
♦ Ready to assemble kits without
cabinets but including cabinet plans:
$599 each.
Insured freight to most of Australia
$30 each. If you buy the I-201 as a kit
including cabinets, rather than fully
assembled, you will get the same high
level of performance as the factory
finished version. VAF research can be
contacted on FreeCall 1800 818882 or
vaf<at>vaf.com.au More information on
the new I-201 or other VAF models can
be found at www.vaf.com.au
SC
www.siliconchip.com.au
|