3

Why Flat
Response? Auricle Publishing

Number 13 2005 $35.00 US

4
How Can
You Tell?

8
Speaker Measurements
Importance of
Time & Phase Objective Gauges of Fidelity isfactory conclusions about audio components. I’ll
This Journal will explain how to interpret speaker offer examples in an attempt to explain why.

9 measurements so you’ll know what the test results

mean and what they can tell you about a If you hate listening to a component
Evaluating
product’s potential for accurate per- or system, no group of meas-
Time & Phase
formance. These objective meas- urements can magically
urements can be very valuable to make that experience
11
consumers because they show enjoyable. On the other
Minimizing
which products to avoid there- hand, if a component
Stored Energy
by narrowing the field of con- or system is demon-
tenders and minimizing the strably inaccurate
13
number of components you’ll learn to hate it
Evaluating
you’ll have to audition. sooner or later
Energy Storage
(based on my expe-
Objective measurements rience) when you
16
don’t tell the entire story of learn how to hear
Importance
course, and they raise the the flaw(s) that may
of Bass
age-old questions about initially have gone
objective versus subjective unnoticed.
17
Evaluating Bass product evaluation.
I’m convinced that you
Performance The Debate have a far better chance for
A debate we encounter frequently in audio long-term satisfaction if you fol-
19 involves the validity of “objective” versus “subjective” low the high fidelity approach and
Thiel CS2.4 evaluations. You’ll notice that I have enclosed both choose from those components that
Review terms in parentheses. That’s because neither word accurately reproduce the signal accord-
has an exact definition when it comes to audio, and ing to accepted standards. You can
25 it’s easy to become involved in a semantic argument still rely on your subjective
Interview with that simply diverts attention from the real issue. responses as you choose from
Jim Thiel Objective usually means an evaluation based on those components which are
repeatable “scientific” tests. Subjective usually objectively accurate.
35 means an evaluation based solely on the
Interview with emotional reaction of the reviewer. Ah, We live in a touchy-
Richard if life were only that simple. feely world and many
Vandersteen will try to convince you
In reality neither method can be that “if it sounds good,
46 employed alone in order to come to sat- it is good.” While there is
Engineering
Double Standard
www.audioperfectionist.com
 Intro

an element of truth here, there are objective measurements to tify the brand of piano playing. Doesn’t that make me a skilled
gauge fidelity and many will try to ignore this fact or downplay listener who doesn’t need measurements as a guide? On the
its significance. Why do they do this? Because in a world other hand, couldn’t I be fooled by products with colorations
where no objective standards exist, anybody can be a “design- complementary to my wife’s voice and pianos?
er” and everybody is an “expert” or qualified “reviewer.” Where
would all these pundits be if they were forced to learn some- If a certain coloration reminds me of a certain kind of music
thing about engineering? some of the time, isn’t that enough? However, if I like a certain
coloration won’t my listening be limited to a genre that benefits
from that coloration? If I listen exclusively to that musical genre
Semantics does it matter that measurements may show that the sound I
An argument about the meaning of the words can divert atten- enjoy is not accurate? If I hate listening to an audio system will
tion from the real issue. That issue is deciding which method of my opinion change if I’m presented with a set of impeccable
component evaluation produces the most satisfactory results. measurements?

What’s objective? A conclusion based solely on “facts” per-

haps? Whose facts? Which facts? Can we come to a truly Extremes
objective conclusion based on only a few selected measure- Must we choose between these extremes or is it possible to
ments (facts)? If so, which ones count? Among those that combine the best elements of both objective and subjective
count, which are most important? Do measurements reflect all evaluation and arrive at the most satisfactory conclusion?
the facts we need to examine in order to come to a satisfactory
conclusion? I believe that it is not only possible but absolutely necessary in
order to reach musical listening nirvana. I think we need to
And what’s subjective? A position select those audio components and systems that provide emo-
based purely on emotion? Is tional satisfaction from among those that are demonstrably
the emotional accurate according to accepted standards. Accepted standards
response of the include, but are not limit-
examiner a ed to, reasonably accu-
faulty basis for rate measured perform-
conclu- ance.
sions?
Isn’t
an Yes, Virginia,
audio sys- Facts Exist
tem designed Facts are simplistic, but
solely to pro- they can be very useful.
duce an emo- They don’t necessarily
tional response demonstrate the value
in the listener? of something but they
sure can help to expose
I can always a fraud. This Journal
identify my will explain speaker
wife’s measurements and
voice show their value. This
and knowledge won’t provide a
usual- substitute for personally auditioning
ly iden- components but it will narrow the
field.

2 www.audioperfectionist.com
 Why Flat Response?

Objective measurements can’t tell you everything about how a Why Accurate Speakers Must Have Flat
component or system sounds but they absolutely can tell you
Frequency Response by Richard Hardesty
when the search for accurate reproduction is hopeless.
All audio components, including loudspeakers, must have rea-
sonably flat frequency response in order to accurately repro-
duce the timbre of musical instruments and voices. This fact is
Why Are Certain Facts Important
clearly evident and here’s why.
In this
issue I’ll
Every sound produced by a musical instrument or voice is
explain
made up of a fundamental frequency and a series of harmonic
why cer-
overtones. The fundamental frequency defines the note being
tain per-
played. The harmonic structure allows the listener to identify
formance
the specific instrument playing or voice singing the note.
charac-
teristics
The complete sound of each note the listener hears is a combi-
are
nation of fundamental and harmonics. The fundamental estab-
important
lishes pitch and the harmonic structure defines the unique
for the
characteristics of each instrument or voice based on tone
accurate
rather than pitch or volume. These unique tonal characteristics
repro-
are called timbre.
duction
of
recorded
music.
“...instruments sound different
These
articles
because each has a unique
will offer
my opin-
harmonic structure (timbre)...”
ions and
A piano and a violin can play the same note with the same
illustrate
fundamental frequency. The pitch and volume of the notes
the logic
from the piano and violin may be identical. The instruments
that led me to these conclusions. Then I’ll explain how to
sound different because each has a unique harmonic struc-
tell which transducers provide these characteristics and
ture (timbre). The waveforms from these instruments have dif-
which ones can’t possibly present an accurate reproduc-
ferent shapes when displayed on an oscilloscope and each
tion of the recorded information.
delivers a unique sound to the ear even when playing the
same note at the same volume. The fundamental frequency of
that note is identical but the harmonic structure is different.
How Do You Tell Which Products Work?
The tonal characteristics (timbre) are different and easily
After each of the articles that describes desirable performance
identified by the ear.
attributes there will be an article that explains how these attrib-
utes can be measured and how you can interpret the measure-
The unique timber of each instrument or voice is established by
ments. Yes, objective facts exist and they can be of value to you.
the quantity and ratio of harmonic overtones which determine
the shape of the wave. The exact relationship of these over-
Also In This Issue tones must be preserved in order to reproduce timber accurate-
Shane Buettner reviews the Thiel CS2.4, an outstanding, high- ly. This requires that audio components exhibit flat frequency
value speaker system. And we’ll present my interviews with Jim response and maintain proper timing relationships. Flat fre-
Thiel and Richard Vandersteen. APJ quency response is required in order to preserve the amplitude

www.audioperfectionist.com 3
 How Can You Tell?

sound. A speaker system

that delivers midrange har-
monics in reverse phase
can’t possibly recreate
musical timbre accurately
yet that’s exactly what most
speakers do.

The information that follows

will explain how to tell
which speakers are correct
in the amplitude and time
domains. While the articles
in this Journal will be
specifically about speakers,
the facts apply to all audio
components. Speakers
serve as good examples
because they tend to be the
least accurate of compo-
nents and exhibit the gross-
est deviations from accu-
rate amplitude response
and phase. APJ

relationships between the fundamental and each harmonic

overtone. Proper timing must be maintained in order to pre-
How Can You Tell if
serve the phase relationships between the fundamental and a Speaker System has
harmonic overtones. Flat Frequency Response? by Richard Hardesty
You can test speakers yourself with readily available computer-
If the speaker system or other component deviates from flat ized instruments like Clio (www.mclink.it/com/audiomatica/
response, signals at some frequencies will sound louder or home.htm) and MLSSA (www.mlssa.com/) or you can refer to the
softer than they should. These response deviations may alter measurements printed in trusted publications.
the level of various harmonics and change the timbre of the
sound. Some overtones may be emphasized and/or some may John Atkinson of Stereophile magazine, for instance, has
be diminished in level. A speaker system with response devia- measured hundreds of speakers. He uses methods generally
tions of just a few decibels will alter the timbre of musical accepted by the industry and achieves useful results.
instruments and voices. A speaker system with response that Stereophile speaker reviews are available free on the web at
deviates beyond a window of ±3dB can’t possibly recreate the www.stereophile.com and they can show you that most of the
timbre of instruments and voices accurately. speakers that Stereophile reviewers recommend are incapable
of accurate performance.

If a speaker system or other component alters the phase of any The frequency response graph will tell you what you need to
group of frequencies, the wave shape—and the sound—will be know about the ability of the speaker under test to deliver an
altered. Slight changes in phase will slightly alter the timbre of accurate replica of the recorded information in terms of ampli-
the sound and phase reversals will substantially alter the tude. Flat frequency response is the cornerstone of good

4 www.audioperfectionist.com
 How Can You Tell?

speaker design and must be considered before any other ond are two names for the same thing—100Hz is 100 cycles
parameter but it’s not the only important factor, as we’ll see. per second. One thousand Hertz is often called a kilohertz
(1kHz).
A frequency response graph displays frequencies on the hori-
zontal scale across the bottom, sometimes called the “X” scale, The frequency range of an engineering instrument may be from
and amplitude on the vertical scale at the left side of the graph, 10Hz to 30kHz. The range displayed by instruments often used
sometimes called the “Y” scale. You can use this graph to by reviewers may be limited to 20kHz. The 20kHz upper limit cov-
determine if the speaker under test deviates from flat response, ers the range of human hearing but can’t show ultrasonic tweeter
and to pinpoint the amplitude and frequency of that deviation. resonances, which can have an audible effect on the frequencies
Don’t be intimidated, it’s very easy to do. All you need to know we can hear. The “oil can” resonance of an alloy dome tweeter
is what the gradations on the graph mean and how to interpret will typically occur at frequencies of 25kHz to 27kHz, which won’t
that information. I’ll try to explain that in the simplest terms be displayed by a test instrument with a 20kHz bandwidth but
possible. may affect the sound of the speaker under test.

Frequency Amplitude
The frequency scale across the bottom is usually logarithmic The vertical scale at the left side of the graph shows amplitude
(abbreviated log) rather than linear, which means that the fre- in decibels (abbreviated dB). The decibel scale is logarithmic,
quency goes up in powers of ten—ten, one hundred, one thou- not linear, and this is important to remember. This scale may
sand, ten thousand. show numbers in a range of perhaps 60dB-95dB, or it may
have a point marked zero with ranges graduated in decibels
On a logarithmic scale more space is devoted to octaves at the above and below this point so you can determine how far the
bottom of the scale than to those at the top, which correlates speaker under test deviates from flat response.
well with the actual frequency ranges of musical instruments
This scale indicates the volume of the output from the speaker
under test. We’re not concerned with the actual volume of the
output. What we want to know is the relative volume across the
frequency scale so we can see which frequencies are empha-
sized and which are diminished and by how much.

The Stimulus
The test signal is called a stimulus and speakers are typically
stimulated to produce an output of 80-85dB SPL at 1kHz but
that doesn’t matter much. As noted above, what we’re interest-
ed in is how much louder or softer the speaker plays at various
frequencies with an input signal at a constant level.

Most speaker measurements use quasi-anechoic techniques to

eliminate the effects of the room in which the speaker is meas-
ured. The test signal for a quasi-anechoic measurement is
called a Maximum Length Sequence (MLS), sometimes
referred to as pseudo-random noise, which is converted into an
(see Journal #12). Frequency is shown in cycles per second, impulse by a computerized test instrument. The impulse is then
now called Hertz (abbreviated Hz). Hertz and cycles per sec- converted by the computer into frequency and phase informa-

www.audioperfectionist.com 5
 How Can You Tell?

tion utilizing the Fast Fourier Transform (FFT), which is a math- been spliced together with the frequencies above about 500Hz
ematical method of converting between time domain and fre- measured at a distance of three or four feet using quasi-ane-
quency domain information. choic techniques, and the frequencies below 500Hz created by
combining a close-microphone measurement of the woofer(s)
You don’t have to understand how this works in order to realize and a close-microphone measurement of the vent(s) using
the limitations and benefit from the results. (If you do want to mathematical techniques to add the outputs at appropriate fre-
understand read Testing Loudspeakers by Joseph D’Appolito quencies and calculate the difference at frequencies where the
from Audio Amateur Press.) driver(s) and vent(s) would be out-of-phase.

Quasi-anechoic measurements minimize the effects of ambient

noise and reflections from the room boundaries and allow The Graph
speakers to be tested in regular rooms rather than anechoic The “X” and “Y” scales together show how loud the speaker
chambers. The measurement window starts when the first plays at each frequency in response to a test signal with a con-
sound arrives at the speaker and ends with the arrival of the stant volume or one that can be correlated to a constant vol-
first sound reflected from the room boundaries. The room ume. A perfect speaker would produce a straight line across
boundary nearest to the test microphone will usually be the the frequency range at 80-85dB on a scale marked in dB, or at
floor, which limits the accuracy of low frequency measure- the zero line on a scale that shows zero with ranges above and
ments. below that point graduated in decibels. An imperfect speaker
will produce a horizontal line with small deviations above and
If the speaker under test is standing on the floor the accurate below the zero line (or a horizontal line drawn through the
lower limit of the measurement will be about 500Hz. If the amplitude point at 1kHz).
speaker is elevated that limit may be extended to 300Hz or
slightly below. The measurements you see in magazines have A deviation above the zero line means that the speaker under

6 www.audioperfectionist.com
 How Can You Tell?

test will emphasize signals at that frequency. A deviation below

the line means that signals at that frequency will be diminished
by the speaker under test. The Real World
Speakers that utilize D’Appolito arrays (two midrange drivers,
You can determine the amplitude of the deviation by following a one above and one below the tweeter) will measure better than
horizontal line from the tip of the deviation to the scale at the they sound because the microphone will be exactly centered
left side of the graph. Be sure to note whether the scale is between the midranges and the response differences of the
graduated in 5dB increments (typical) or other increments. You two drivers will produce a smoothing effect on the frequency
can determine the frequency of the deviation by following a response. In the real world a listener will never be exactly cen-
vertical line down to the scale at the bottom of the graph. (See tered and the resulting time smear will simply blur definition.
the illustration.)
Speakers with dipole radiation patterns will sound better than
Deviations are usually referred to by the difference in amplitude they measure because the horrendous response deviations in
between the level at 1kHz (1,000Hz) and the level at the devia- the direct signal will be smoothed by the rear-wave energy
tion. If the specification says frequency response is 40Hz- reflected off the wall behind the speakers. In the real world this
18kHz ±3dB, that means that when tested the response of the time smear will prevent any real definition but will provide an
speaker never deviated more than 3dB from flat response artificial ambience effect that some find pleasing.
between the speaker’s bass limit at 40Hz and the highest tre-
ble it could reproduce at 18kHz. In other words no output peak Speakers with bipolar radiation will create even more time smear
was more than 3dB louder than the signal at 1kHz and no out- than those with dipole radiation in most rooms because they radi-
put dip was more than 3dB softer than the output at 1kHz. This ate substantial energy to the sides increasing the amount of
6dB window of error over this bandwidth is the minimum per- reflected sound that arrives at the listener. Some people like this
formance that I would consider acceptable for high fidelity sound but it’s not related to the recorded signal.
music reproduction.

www.audioperfectionist.com 7
 Importance of Time & Phase

There’s More to Accuracy Arguing about whether an alteration to the shape of the
A speaker that can’t produce a fairly accurate frequency resulting waveform is audible is like arguing about whether
response graph can’t reproduce an accurate replica of the a frog is waterproof. Observation provides irrefutable evi-
recorded information but there’s much more. Phase and energy dence. You can observe frogs in a pond and you can
storage are separate issues that will be considered in the arti- observe waveforms on an oscilloscope. A frog emerges
cles that follow. APJ from water completely unscathed and musical instruments
produce different waveforms on an oscilloscope. Changing
the phase of any wave component alters the shape of that
waveform. Period.

Why Must an Accurate Speaker System be Of course we can distinguish between waveform shapes with
Time- and Phase-Correct? by Richard Hardesty our ears. How else could we hear timbre and distinguish
As described in the articles about frequency response, correct between various voices and instruments? Helmholtz demon-
timbre cannot be reproduced by a speaker system that alters strated this fact 150 years ago and modern day musicians are
the amplitude of any harmonic overtone because that will sure- well aware of it even if they can’t necessarily provide an expla-
ly alter the shape of the resulting waveform. The same is true nation. Piano tone can be altered by changing the point where
of phase. Reversing the phase of portions of the frequency the hammer strikes the string because this alters the relation-
spectrum changes the shape of the waveform. No audio com- ship between the harmonics that string produces. If the piano is
ponent—except the loudspeaker—does this. in tune the fundamental frequency will not change, regardless
of where the hammer strikes the string. If the piano is not in
Stereo imaging is of great importance to some (like me) and tune it can be adjusted with an electronic tuning device that
couldn’t matter less to others. This effect has a more subtle displays the fundamental frequency (pitch) of each string. The
influence on musical realism than the accurate reproduction of same tuning device can be used with other instruments even
timbre but is significantly impacted by the time domain perform- though they have unique timbre.
ance of the speakers.

Guitar players can alter the tone they produce by changing the
Timbre place where the string is plucked or altering the way in which
The harmonics of real instruments or voices add to or subtract
from the fundamental frequency which creates pitch. This
results in unique tone called timbre that allows us to identify
the instrument or voice. A piano doesn’t sound like a violin and
Sheryl Crow doesn’t sound like Ray Charles even when they
each play or sing the same note. The pitch is identical but the
timbre—and the waveform—is different.

the string is set in motion (nail, pick or flesh). If the guitar is in

tune the pitch stays the same. The switch at the bottom of a
Fender Stratocaster changes the electrical phase relationship

8 www.audioperfectionist.com
 Evaluating Time & Phase

of the three pickups positioned beneath the strings. This 5- ed to appropriate drive elements. Phase shift varies with fre-
position switch alters the tone of the instrument, not the pitch, quency and can’t be corrected by changing the physical posi-
which is established by the fundamental frequency of the tion of the drivers.
vibrating string.
Driver position can affect the time relationship between drivers
Phase does matter and those who suggest that it doesn’t prob- because each has a slightly different rise-time and driver posi-
ably make speakers which are not time coherent. tion may have a minor effect on cancellation in the overlap
region where two drivers are reproducing part of the same sig-
nal which has been phase-shifted by the crossover filters.
Imaging
The effect of phase on stereo imaging is more subtle but can Speakers with steep filter slopes can’t be made time- and
be observed with experience. Time- and phase-accurate speak- phase-correct. Speakers with first-order acoustic slopes will
ers produce a stage that is rectangular (as viewed from above) also require that drivers be carefully positioned to compensate
with depth that is apparent well to the sides of that stage and for differing rise-times. Read this paragraph again because it’s
not only at the center. Stage width can extend well beyond the very important.
loudspeakers.
A time- and phase-accurate speaker will have gentle filter
Typical speakers produce a stage that is triangular (as viewed slopes and physically aligned drive elements. One without the
from above) with an illusion of depth only at the center and a other won’t do and all talk to the contrary is simply rhetorical.
reduced sense of depth toward the speakers (sides of the Time and phase do matter and there is a sure way to tell when
stage). Time incoherent speakers will usually define the outer these timing relationships are correct.
edges of the soundstage. APJ

When the Green Flag Drops the Bullshit Stops

The step response graph tells whether a speaker system is
How Can You Tell time- and phase-accurate, clearly and without ambiguity. If
if a Speaker System is some drivers pull while others push, the speaker is not phase
Time- and Phase-Accurate? by Richard Hardesty coherent and this will be revealed by the speaker’s step
Speaker manufacturers and reviewers have presented loads of response.
meaningless misinformation about the importance of phase
and you’ve probably seen dealers or manufacturers meticulous- A speaker that is not phase coherent cannot accurately repro-
ly adjusting the position of drivers or speaker tilt-back, as if this duce timbre or image correctly. The shape of the step response
has an effect on phase coherence. It doesn’t. graph tells all. Here’s why.

Phase shift is primarily the result of crossover filters which

The Step Response Graph
divide the spectrum into ranges of frequencies which are direct-
A step response graph displays time on the horizontal scale

www.audioperfectionist.com 9
 Evaluating Time & Phase

across the bottom, sometimes called the “X” scale, and ampli- Sound occurs as a drive element moves and stops when the
tude on the vertical scale at the left side of the graph, some- drive element stops, even if it stops at the end of a long excur-
times called the “Y” scale. You can use this graph to determine sion. Why? Because a stationary driver diaphragm can’t move
if the speaker under test is time coherent or if some drivers air and create sound.
push while other pull.
If you apply direct current (DC) to a loudspeaker the woofer will
The factor important to us is the shape of the output signal, not be displaced in direct proportion to the amplitude of the cur-
the increments which may be significant to designers but not to rent—and stay there until the current is removed. The capaci-
listeners. The step stimulus is like the top half of a square tors that act as high-pass filters for the other drivers will block
wave or an impulse with extended duration. I’ll try to explain DC allowing only the upper harmonics to pass. The stimulus
that in the simplest terms possible. exercises the entire speaker just like music, but only momen-
tarily. With computerized test instruments that’s all we need to
determine which drivers move to create positive displacement
The Square Wave (outward) and which ones are out-of-phase (move inward).
I’ve heard it said that we don’t listen to square waves but the These facts are important to know if we want the output wave-
lowly square wave is the test signal that most resembles music. form to mimic the input signal, which is absolutely necessary
A square wave, like a musical wave form, is composed of many for accurate reproduction of the recording.
sign waves—in fact an analog square wave generator creates
a square wave by generating a sine wave at the fundamental
frequency and adding sine waves for all the odd harmonics. Step Response
The resulting square wave is a combination of many sine The step response graph shows only the output of the speaker
waves with precise amplitude and phase relationships, just like under test. Think of the stimulus like the top half of a square
music. If everything is right the result looks square. Deformation wave. The output from a time- and phase-accurate speaker
indicates deviation—showing that some part of the signal has should look like a triangle above the reference line with a sharp
been altered. rise and a slow decay. The beginning rise will slope slightly
because the step stimulus rises almost instantly (straight up)
You can learn almost everything about amplifier performance but the speaker has limited bandwidth and takes some time to
by observing the square wave response. Most speakers can’t rise. (Bandwidth and rise-time are corollaries.)
even come close to reproducing a square wave because
some drivers push while others pull and there can be no out- The speaker makes sound as the drivers respond to the stimu-
put when a drive element is stationary. To gauge the time lus and then output ceases so the signal on the graph decays
domain performance of loudspeakers we use the step back to zero, and maybe a little beyond due to inertia
response instead. (rebound), over a period of a few milliseconds.

The test signal never goes negative so any significant output

The Test Signal for the Step Response that extends in a negative direction (below zero) is out-of-
The step response is actually derived from an impulse correlat- phase. You can’t tell for sure which part of the frequency spec-
ed from an MLS stimulus but for understanding think of it like trum is out-of-phase but the output from the tweeter usually
an impulse with extended duration or the top half of a square arrives first followed by the output from the midrange (if there is
wave. The signal starts at zero, rises quickly to a level that is one) and then the woofer.
maintained for a time at a positive DC potential and then
returns to zero. The signal never goes negative, which is A speaker with a third-order crossover will typically have the
important to remember. This signal tells you which drivers midrange driver wired out-of-phase. A speaker with a fourth-
move in which direction and when, but the speaker can’t pro- order crossover will typically have the drivers wired in phase
duce a step that exactly matches the input signal because but their output will be smeared over time and a portion may
there is no sound when the drivers reach maximum excursion still go negative due to crossover phase shift.
in response to a constant current (DC).

10 www.audioperfectionist.com
 Minimizing Stored Energy

Examples Why Must Energy Storage

The following examples show an ideal step response and the be Minimized? by Richard Hardesty
step response from a typical speaker. A step response graph Speakers contain electrical and mechanical components that
that is not a positive (upward) triangle indicates a speaker that tend to store and release energy after a short delay in time.
is not time- and phase-accurate. Trying to extract additional This tendency must be minimized if the speaker is to reproduce
an accurate replica of the recorded infor-
Step Response – volts (ideal) mation and nothing else. What we want
0.66 – to hear is the recorded information. We
0.56 – don’t want to hear sounds created by the
0.46 – speakers or sounds that originated from
0.36 – the recording but have been delayed and
0.26 – reradiated by the speakers.
0.16 –
0.06 –
-0.06 – Musical instruments are designed to
-0.04 –
deliver rich and pleasing resonances.
-0.14 –
Accurate loudspeakers are designed to
-0.24 –
reproduce recorded sounds rather than
-0.34 –
create sounds of their own. A recording
auto
9.0 10.0 11.0 12.0 13.0 14.0
should capture resonances produced by
Time – msec real musical instruments and speakers
should accurately reproduce the record-
ing and nothing else.
Step Response – volts (out of phase)

0.10 –
0.08 – The audible output from stored energy
0.06 – can be attenuated by damping and what
0.04 – remains will be released after a delay.
0.02 – The amount of attenuation will be deter-
0.00 – mined by the damping qualities of the
-0.02 – resonating material. The delay will be
-0.04 – determined by the amplitude and fre-
-0.06 – quency of the stored energy. This reradi-
-0.08 – ated (stored) energy will produce inap-
-0.10 – propriate sounds that may blur or obfus-
fixed cate the actual signal.
9.0 10.0 11.0 12.0 13.0 14.0
Time – msec
Perfect speakers will deliver the recorded
data from the graph is not within the scope of this article or the information at the right time and then stop producing sound.
experience of most readers. Imperfect speakers will deliver the recorded information and
store some energy that will be released later when there should
Don’t be fooled by the appearance of a speaker or rhetoric be silence or when other recorded information should be pro-
from the manufacturer. The step response graph shows which duced. Nothing is perfect, of course, but well-designed speak-
speakers are time- and phase-correct and which ones aren’t. ers will reduce stored energy to an absolute minimum.
Most of them aren’t. APJ

www.audioperfectionist.com 11
 Minimizing Stored Energy

Sources of Stored Energy Electrical Resonances

Resonant energy can be stored in enclosure panels and driver Capacitors and inductors are used as high-pass and low-pass
diaphragms. Electrical energy can be stored in crossover com- filters in speaker crossover networks. Capacitors resist a
ponents. Reflected energy can be radiated by enclosure baf- change in voltage and inductors resist a change in current. Both
fles, driver frames and internal enclosure structures. Some of elements cause a phase shift between voltage and current.
these sources will subtly degrade the signal and some will have
a profound negative impact on sound quality. In an inductive circuit voltage leads current by about 90
degrees. In a capacitive circuit current leads voltage by about
90 degrees. Certain values can result in resonant electrical cir-
“A good speaker design will cuits. Generally, a crossover network with fewer components is
better but many crossover components may be used to com-
address all these sources of pensate for driver anomalies.

stored energy and reduce The crossover networks in well-designed speaker systems
have been carefully engineered to divide the frequency spec-
each one to inaudibility.” trum and direct the divided bands to the appropriate drivers;
compensate for driver anomalies in frequency and phase; and
A good speaker design will address all these sources of stored provide a constant and acceptable load to the amplifier. This is
energy and reduce each one to inaudibility. A poorly designed a very tall order.
speaker will deliver a signal that combines part of the recorded
information with the stored energy from the speaker itself. The If you think you can build a speaker system in your garage that
frequency response graph may appear to be flat with dips in is competitive with today’s finest designs, think again.
the signal reproduction filled by resonant peaks created by
speaker components and vice versa. That’s why no single
measurement can be trusted and even a group of measure- Reflected Energy
ments must be viewed with skepticism. The measurements can The most prominent source of reflected energy is the baffle that
be useful to narrow the field of contenders but must not be surrounds the speaker drive elements but there are others.
used as a substitute for careful listening evaluations. Energy radiated from the back of a drive element can be
reflected by cabinet structures and reenter the room through a
woofer port. This rearward radiation can be reflected by the
Mechanical Resonances magnet structure and frame of the drive element and reenter
Enclosure panels can resonate and this tendency can be mini- the room through the diaphragm that produced it, delayed by
mized by making the panels thicker; constructing them from the additional path length from the diaphragm to the structure
dense, well-damped material; adding cross braces to larger and back.
panels to raise the frequency and reduce the amplitude of the
resonance; and/or constructing the panels from exotic materi- Regardless of the source, reflected energy travels farther than
als, or laminates that contain exotic materials, with high internal direct energy and smears the signal over time, blurring defini-
damping characteristics. tion and degrading imaging.

Driver diaphragms can be constructed from materials with high

internal damping and/or drive elements can be used only to Stored Energy is Bad
reproduce frequencies that are substantially below (or above) Some speakers are designed to create artificial sounds by stor-
the diaphragm resonant frequency. This is particularly difficult ing energy in shimmering Mylar™ panels or purposely directing
with first-order speaker systems where each driver must cover energy toward the walls where it will be reflected back to the
a wider frequency range than would be necessary in a speaker listener. Some people enjoy this artificial “ambience” but it’s not
with steeper crossover slopes. (See Journal #12). related to the recording and can’t be called high fidelity repro-
duction by any stretch of the imagination. APJ

12 www.audioperfectionist.com
 Evaluating Stored Energy

How Can You Tell if

a Speaker Stores Energy? by Richard Hardesty
I use the phrase “energy storage” to describe a variety of flaws
which result in time smear. Resonances and reflections fall into
this category. Stored energy is a major source of coloration
and a primary reason why speakers that measure well in other
areas sound different.

It’s easy to hear these flaws but the exact source of the prob-
lem is often difficult to pinpoint and correct. That’s why we still
get a broad range of sound from speakers which appear similar
in design and have reasonably accurate frequency response.

An amateur builder can order excellent drivers from Denmark

and use a computer program to design a crossover network
and still end up with a loudspeaker with substantial colorations.
There are some very vocal “geniuses” on the internet who
would have you believe that speaker design is a simple matter
that has been reduced to mathematics. They’re wrong. A good
engineer will consider all aspects of his design and energy
storage encompasses several important factors. You’ll want to
consider only those speakers which were designed by good
engineers.

Measuring Stored Energy

Energy storage is not as simple to measure as other parame-
ters but you can get some idea from spectral decay plots and
accelerometer measurements of cabinet resonances. The
spectral decay plot shows which frequencies linger and for how
long but the source of these lingering frequencies is not easy
to identify. An accelerometer attached to a cabinet panel can
show the resonant energy stored in that panel, identify the fre-
quency of that resonance and show if it’s well controlled or
potentially audible.

“The spectral decay plot shows

which frequencies linger
and for how long...”
An accelerometer may spotlight a panel resonance but can
only suggest how deleterious this resonance may be to the
sound from the speaker so you’ll have to use common sense

www.audioperfectionist.com 13
 Evaluating Stored Energy

(uppermost trace).
Instruments frequently
“autorange” (self adjust
levels) so pay attention
to the amplitude levels
displayed. They won’t
necessarily be the same
for all graphs.

The graph consists of a

series of wiggly horizontal
lines. The one at the top
is the first one, which
shows the response to
the test signal, and each
line below that is the
response at a slightly
later time, as indicated on
the time scale at the right.
This is an example of the poor cumulative spectral decay plot produced by a planar-magnetic speaker. The graph starts
The wiggly horizontal
with the line at the top. Each line below that shows the spectrum of response at a later time. This graph clearly displays
critical midrange energy that persists for several milliseconds after the signal ceases. Some listeners will mistakenly
interpret this as “ambience” but the source of this energy is the speaker, not the recording. line at the rear (top)
shows what is essential-
during final evaluation by listening. Designers have to try to ly the frequency response of the speaker under test and each
solve problems. Consumers can simply avoid them. successive line toward the foreground (bottom) is the spectrum
of response after a period of time has elapsed. These traces
Problems at the front of the speaker will tend to be more audi- show which frequencies emanate after the signal has stopped
ble than problems at the sides and rear. Stored energy at lower and how long they persist. A ridge indicates substantial stored
frequencies will tend to blur details more and stored energy at energy, probably from a resonance.
higher frequencies will tend to be more directly audible. Here’s
an explanation of the graphs in the simplest terms possible.
“A real speaker will deviate
The Spectral Decay Plot from flat response and
The spectral decay plot is frequently called a “waterfall”
because of its appearance. It displays three parameters—time, keep singing after the
frequency and amplitude—in a two-dimensional representation
of a three-dimensional graph. This graph is slightly more com- song has ended...”
plicated than the ones that show just two variables but not as
difficult to interpret as it first seems. A perfect speaker would produce a nearly horizontal line at the
top and nothing else because it would store and release no ener-
Frequency is displayed along the bottom usually in log form. gy after the signal ceases. A real speaker will deviate from flat
Time is displayed at the right side usually in milliseconds. response and keep singing after the song has ended—continuing
Amplitude is displayed on the left side of the graph usually in to produce sound for a period of time after the signal has
dB. An amplitude level like -24dB means that the signal at this stopped. A well-designed real speaker will produce little sound
point is 24 decibels softer than the original signal at the top after the signal stops and that sound will be well down in ampli-

14 www.audioperfectionist.com
 Evaluating Stored Energy

tude so it won’t intrude on the signal from the recording. So how has diminished since the initial trace. Sounds that linger after
do you gauge the performance of the speaker under test? the signal stops should be well down in level when compared
Examine the spectral decay plot. to the original trace, perhaps reduced in level by 24 to 30dB
after a millisecond or so has elapsed.

Interpreting the Spectral Decay Plot In a typical quasi-anechoic measurement setup, frequencies
The top of the graph is offset slightly to the right to show the below 400Hz or so will not be accurate. If the measurements
three variables in a simulation of three dimensions. The test were taken at a greater distance in a large anechoic chamber
instrument will allow the tester to place a curser at the point of accurate bandwidth may be extended down to 200Hz or so.
Anything below this fre-
quency should be
ignored.

Accelerometer
A cumulative spectral
decay plot (waterfall)
can be made using an
accelerometer instead
of a microphone as a
sensing device.
Inexpensive accelerom-
eters are readily avail-
able and can be
attached to cabinet
panels. The output from
the accelerometer is
fed to the test instru-
ment instead of the out-
This is an example of a pretty good cumulative spectral decay plot. With the exception of a fairly prominent resonance at
about 10kHz, residual energy is greatly reduced in level and virtually gone after an interval of less than 2.5ms. put from a microphone,
as in most tests. The
a deviation and read the frequency directly but you’ll have to resulting measurement will display any prominent panel reso-
follow a vertical line to the bottom of the graph (moving slightly nances, identify the frequency and suggest the extent of the
to the left) to determine the frequency that is being sustained. problem.

You can determine the duration of the sustained frequency by fol- Determining how this will affect the sound from the speaker
lowing a horizontal line to the time scale on the right side of the under test is a subjective matter best left to experienced
graph. The graduations here show the amount of time which has testers. You can see from the graph when there may be a prob-
elapsed since the initial trace. Two milliseconds is a long time in lem and listen to determine if that problem is sufficiently audi-
speaker terms and a substantial output after that length of time indi- ble to be problematic to you. APJ
cates an audible coloration, which would be unacceptable to me.

You can determine the amplitude of the sustained frequency by

following a horizontal line to the amplitude scale at the left. The
graduations here show how much the amplitude of a deviation

www.audioperfectionist.com 15
 Importance of Bass

The Importance Real Bass

of Bass by Richard Hardesty We’re not talking about “car stereo” bass, which is really
You may have heard that clear midrange is all that’s necessary emphasized mid-bass. We’re talking about real low frequency
for a satisfying experience from recorded music. If you believe fundamentals below about 80Hz. These bass frequencies are
this I can make you a great deal on a large bridge in San an important part of a live musical performance and must be
Francisco. Midrange isn’t an entity unto itself. It exists partly as present for a reproduced performance to sound like a reason-
a result of bass and treble. The idea that bass plays a small able facsimile of the original. You can hear music without bass
roll in music is simply not true. Bass is more than a minor over the telephone but it’s only a vague suggestion of the real
attribute of music—it’s a vitally important component. thing.

There is a small kernel of truth in the big bass lie: you probably To fully convey the musical message, bass must be present
are better off without the poorly defined and badly timed bass and it must be tightly controlled and presented with impeccable
timing. (Refer to Journal #12 for
the actual frequency ranges of
real musical instruments.) I can’t
be satisfied by an audio system
that lacks outstanding bass per-
formance and once you’ve heard
a system that presents bass cor-
rectly you probably won’t be satis-
fied with less either.

You can subjectively evaluate

bass performance by observing
your rhythmic involvement with
the music. Good bass makes it
easier to follow the beat and it
makes the band seem to be play-
ing together in tighter unison.
Percussive elements define and
punctuate the experience.
produced by most speakers. But a true aficionado will never be
satisfied by reproduced music with poor bass. Bass is the foun- You can subjectively evaluate the contribution of bass frequen-
dation of all music and a major component in both of the fac- cies to the overall presentation. With good bass, music sounds
tors that separate music from noise: the rhythm (beat) and the fuller and more balanced. Bass harmonies add complexity and
melody (tune). interest to the musical score. Deep bass notes can deliver a
thrilling comment to an otherwise subtle tune.
A percussion instrument like a drum often sets the rhythm and
establishes the pace of a piece of music. Bass instruments add
body to this rhythm while playing a tune in harmony with the It’s more difficult to objectively evaluate bass performance but it
instruments that create the melody. The upper harmonics pro- is possible to gain some vital knowledge by examining meas-
duced by bass instruments provide a substantial contribution to urements. Let’s learn how. APJ
the melodic content of the music, which is typically perceived
as “just” midrange.

16 www.audioperfectionist.com
 Evaluating Bass

Evaluating Bass
Performance
by Richard Hardesty
It is possible to accurately measure
low frequency performance out-
doors (even very large anechoic
chambers are too small) but this is
seldom practical for magazine
reviewers. What you’ll see in most
reviews is a measurement made by
placing the microphone very close
to the radiating diaphragm or vent
opening and splicing this measure-
ment, or the complex sum of these
measurements if there is a vent, to
the quasi-anechoic graph that
shows frequency response from
400Hz or 500Hz on up. The blue line is frequency and the gray line is impedance.

which will coincide with the –6dB point or low frequency limit of
This low frequency graph gives a pretty good idea of the the speaker under test. Output from the speaker will roll off at
speaker’s response with no indication of the room’s effects. I’ve about 12dB/octave below this point. A well-designed transmis-
found that these measurements correlate well with what I hear sion line will behave more like a sealed enclosure than a vent-
when auditioning speakers tested this way. You can get addi- ed design. It will probably have a single resonant peak with
tional information by examining the impedance plot. reduced amplitude as compared to a sealed enclosure. The
resonant peak will probably occur at a slightly lower frequency
as compared to a sealed enclosure.
Impedance Plot
The impedance plot tells us a great deal about the design of
the woofer section of the speaker and reveals clues about tran- Vented Enclosures
sient response. Of course it shows a number of other interest- A vented enclosure will have two resonant peaks and the bot-
ing things that are beyond the scope of this article, but we’ll tom of the trough between them will define the resonant fre-
use it as part of our bass performance evaluation. quency of the air mass in the vent. The resonant frequency of
the vent will coincide with the minimum excursion point of the
The number of resonant peaks in the impedance plot and the
frequency of these peaks indicate the type of bass loading, as
does the rate of low frequency roll-off. The speakers we are “...an impedance plot with
likely to encounter will have sealed enclosures or vented enclo-
sures. two peaks is the least
The vented designs will probably have tuned ports, passive desirable indicator of
radiators or transmission line loading.
bass performance.”
Sealed Enclosures bass driver demonstrating that what you hear at this frequency
A sealed enclosure will typically have a single resonant peak, is the vent resonance not the output from the bass driver.

www.audioperfectionist.com 17
 Evaluating Bass

the filter “rings” (oscillates after the

signal stops). Ringing in a woofer is
often referred to as “overhang.”
Overhang is an undesirable charac-
teristic that muddies bass definition
and makes the rhythm and pace of
music more difficult to follow.

A sealed enclosure typically delivers a

second-order high-pass characteristic.
It rolls off at 12dB/octave with the
least oscillation (overhang). A vented
enclosure typically delivers a fourth-
order high-pass characteristic. It rolls
off twice as fast (24dB/octave) and
oscillates twice as much after the sig-
nal stops. A passive radiator design is
a vented enclosure with the passive
The blue line is frequency and the gray line is impedance. radiator replacing the volume of air in
the vent. They work the same way
Output from the speaker will roll off at about 24dB/octave below and perform pretty much the same except that a passive radiator
this point. can be weighted, allowing it to work in a much smaller box.
Midrange frequencies, which might find their way into the room
through a bass vent, will be blocked by a passive radiator.
Passive Radiators
A design with a passive radiator will behave very much like a A vented enclosure reduces driver excursion near the resonant
vented design. A band-pass design will behave like two filters— frequency of the vent. This allows the bass/mid-driver in a full-
one at the bottom, which could perform like a sealed enclosure range speaker to deliver reasonable midrange performance
if one side of the band-pass woofer is sealed, and one at the and go a little lower than it otherwise would. A vented sub-
top, which will perform like a vented enclosure in reverse, pro- woofer will play louder than a sealed design while sacrificing
viding a fourth-order low-pass filter. bass extension and transient response (acceptable for home
theater but not the best choice for music).

Impedance Peaks
Two impedance peaks are the least desirable indicator of bass per- My Observations
formance. Two peaks are produced by a vented design with the I don’t like the sound of speakers that use vented bass loading.
steepest roll-off and the poorest transient response. One peak is Transmission-line loading sounds a little better to me but not
better, indicating a sealed design with more extended bass enough to justify the added cost and complexity. Sealed enclo-
response and improved transient performance. No peak is best, sures require equalization or enormous size for real bass
indicating a design with little or no resonance in the pass band. This extension. These are definitely potential drawbacks but EQ is
may be achieved by an ideal transmission line or a sealed enclo- the compromise that I choose to make in the real world.
sure operating below the fundamental resonance of the system.
These are generalizations and there are near-infinite possibili-
ties for design compromises. Use my opinions as guidelines as
Filter Analogies
you listen and make your own determinations about what
A woofer is a mechanical high-pass filter. The design determines
sounds best to you. APJ
how steeply the response rolls off at the bottom and how much

18 www.audioperfectionist.com
 CS2.4 Review

THIEL time- and phase-accurate loudspeakers. Since the demise of

Dunlavy, Vandersteen and Meadowlark are the only others I’m
CS2.4 aware of.

Speaker System Review by Shane Buettner I share Richard Hardesty’s opinion that time- and phase-correct
There are few companies in high-end audio with the well- speakers represent a higher standard of performance and a
deserved reputation for innovation and performance that Thiel more sophisticated level of connoisseurship compared to con-
Audio enjoys. Thiel has been manufacturing high performance ventional loudspeaker designs. For those who value convincing
loudspeakers in the U.S. since the late 1970s. The Thiel 2.4 is dimensionality in imaging and soundstaging I don’t believe
one of the company’s latest designs, and is a relatively diminu- there’s a substitute.
tive three-way floor-stander that sells for just $4,200/pair in
satin black. (Thiel’s typically lavish finishes can be applied for By their very nature, time- and phase-accurate speakers
an additional charge.) require more work and ingenuity on the part of the designer,
and very high quality parts in the drivers and crossovers. It’s no
I’m very excit- wonder most designers prefer to tell you that time-domain per-
ed to be formance isn’t audible—it’s very difficult and expensive to prop-
reporting on erly execute a time- and phase-correct design.
Thiel’s 2.4
loudspeakers, If you’re a new subscriber I heartily recommend Richard
which are truly Hardesty’s article Time and Phase, Not Just a Craze from the
high-end, high Audio Perfectionist Journal combined issues #6&7 for an out-
performance, standing primer on the importance of time-domain perform-
time- and ance. That issue also contains an in-depth look at Thiel’s
phase-correct design philosophies and an account of Richard Hardesty’s visit
and hand- to the Thiel facilities in Lexington, Kentucky.
crafted in the
USA. In an
industry rid- Design and Construction
dled with over- Many loudspeaker designs that sell for ludicrous sums are sold
priced and on the alleged integrity of construction in their cabinets and/or
under-perform- the rarity, quality and expense of the materials used in the cabi-
ing compo- nets, drivers and crossovers. After reading about the design
nents, a techniques and materials used to create this $4,200/pair speak-
speaker like er from Thiel, I hope you’ll cast the same jaundiced eye that I
this is a breath do toward the companies selling 7” two-way speakers (or a 7”
of fresh air. two-way on top of a vented woofer box) for over twenty thou-
sand dollars per pair based on spurious claims of construc-
While many magazine reviewers would have you believe that tion/parts quality.
the carriage trade products they have on long-term loan are
truly reference-quality, here’s a speaker that many people can
actually afford (without a second mortgage) that’s demonstrably Cabinets
superior to those pretenders in a number of respects. Perhaps the most unique appearance aspect of the Thiel 2.4 is
the sloped and sculpted front baffle. The slope is used to main-
Regular readers also know that the Journal is a big proponent tain physical temporal alignment of the drivers, which, along
of time-domain fidelity in loudspeakers and that Thiel Audio is with the coaxially mounted midrange/tweeter, obviates adjust-
distinguished as one of the few remaining manufacturers of ment of the speaker’s tilt in order to maintain optimal image

www.audioperfectionist.com 19
 CS2.4 Review

focus or tonal balance. The

contoured edges of the baf-
fle are designed to break
up any early reflections
from the drivers that
could arrive at the listen-
er’s ears close enough
in time to the primary
signal to cause degrad-
ing time smear. Thiel
believes the contoured
baffle results in a
more open and
focused soundstage
and my listening
experience with the
2.4 backs up that
assertion.

The 3”-thick MDF

front baffle mate-
rial is robust and
rigid. The rest of
the cabinet is that are mirror-matched on each speaker pair. You simply won’t
1”-thick MDF, find finer craftsmanship or greater aesthetic appeal in a speak-
and is braced er cabinet no matter how much more you spend.
to a truly
extreme
degree. The Drivers and Crossovers
Thiel 2.4 cabi- The CS2.4 uses a newly developed coincident aluminum tweet-
net is inert. er/midrange driver (1” dome tweeter, 3.5” midrange) that’s
Rapping the cab-
CS2.4 inet with your knuck-
les produces the dullest of
thuds, with no sensation of resonance whatsoever.

I assisted Richard Hardesty a few years back when he updated

the drivers in Widescreen Review’s Thiel CS6 speakers to a
newer model. I haven’t seen any speaker at any price that sur-
passes Thiel CS6 structural integrity. Although the CS6 front
baffle is made of a more advanced material, the 2.4 cabinet
seems to be constructed to similarly high standards in spite of
the price differential.

I had Thiel send the review speakers in satin black, but for an
additional charge Thiel offers a number of exquisite veneers

20 www.audioperfectionist.com
 CS2.4 Review

A potential drawback of using a coaxially mounted

midrange/tweeter is whether the drivers stay effectively decou-
pled from one another and avoid intermodulation distortion. In
other words, does movement from one driver cause unwanted
movement in the other, which results in distortion? Thiel com-
bats this by shaping shallow midrange cones to form a proper
surround for the tweeter.

“...the midrange and tweeter

are temporally aligned by
sharing the same space...”
The 2.4 uses a single-layer aluminum midrange material in its
unicoil driver, where Thiel’s more costly designs (such as the
CS6 and CS7.2) use a three-layer sandwich material for
unique among coaxial drivers in that it uses a single voice increased rigidity and damping.
coil—the so-called “unicoil” design. This coaxially mounted
midrange/tweeter relies on a mechanical crossover (no electri- Using aluminum for the midrange driver material has tradeoffs.
cal network) between the two drivers that Thiel claims con- Aluminum is lightweight but very stiff and can operate over a
forms to a phase-coherent 6dB/octave acoustical slope. broad range of frequencies with high resolution and low distor-
tion, which is a prerequisite for drivers in a speaker with first-
As with any high quality speaker there is substantial cost tied order crossovers. Aluminum drivers exhibit a so-called “oil can”
up in a Thiel speaker in its drivers and crossovers. The unicoil resonance at certain frequencies, a malady Jim Thiel engineers
system offers a very high performance-to-price ratio by using around by making sure the resonances occur beyond the fre-
only one motor/magnet system and eliminating the electrical quency range at which a given aluminum driver operates in his
network. Jim Thiel told me that, while this driver is more expen- speakers.
sive than a separate midrange and tweeter, it’s not as costly to
implement as a typical dual motor midrange/tweeter, especially Objectively, an impulse response test will show aluminum driv-
with the crossover eliminated. He essentially considers this ers as prone to ringing. For example, the driver resonates over
design to offer “three-way performance at a two-way price.” a longer period of time after a transient than drivers made of
softer materials. On the other hand, softer driver materials
Coaxial drivers in general have advantages in a time- and don’t operate in linear, pistonic fashion over as broad a range,
phase-correct speaker. One is that the midrange and tweeter and tend to absorb energy, either of which can result in lower
are temporally aligned by sharing the same space. A gentle resolution and compressed dynamics.
slope of the cabinet is all that’s required to align the coincident
driver’s output with that of the woofer, and the typically narrow
vertical listening window of a time- and phase-correct speaker Subjectively, the most obvious drawback of aluminum in a
is expanded greatly. I’ve never found the narrow vertical win- midrange driver is that attentive listeners will hear the distinct
dow troublesome as my critical listening is done from the same sonic signature of the metal cone in this critical band.
chair at the same height every time. But some people may Aluminum midrange drivers have a sound that will appeal to
appreciate being freed from the perceived “head in a vice” con- some and turn others off. It’s my opinion that Thiel gets the
straint. most out of this design choice, absolutely minimizing, if not
negating entirely, the potential pitfalls.

www.audioperfectionist.com 21
 CS2.4 Review

The 2.4 uses an 8” aluminum woofer with a 7.5” x 11” passive hung” voice coils—short voice coils in a long magnetic gap. A
radiator. Vented/ported woofers resonate the mass of air in the typical long voice coil/short gap motor system produces distor-
port to increase the low frequency output of the driver. A pas- tion in bass drivers because the power of the magnetic field
sive radiator does the same thing by using the mass of the acting on the coil and the amount of iron in the coil vary as the
passive radiator instead of the mass of the air column of the voice coil moves back and forth, toward and away from the
magnet structure. With a short coil in a long gap, even when
the coil has moved a long way, it’s still in a uniform magnetic
field within the gap. Thiel further eliminates these distortions by
using copper sleeves over the pole piece and copper shorting
rings around the pole base to stabilize the magnetic field acting
on the coil.

Thiel’s crossover networks are all hard-wired (no circuit

boards) using the finest quality parts, including polypropylene
and polystyrene capacitors, along with very pure, low oxygen
copper, and air core inductors. The network in the 2.4 con-
forms to a first-order, 6dB/octave acoustic slope between the
woofer and the coaxial midrange/tweeter. Thiel’s networks are
complex as they compensate for impedance and phase devia-
tions between the driver elements as well as damping driver
resonances.

All in all, there’s enough engineering innovation and quality

parts and construction in the Thiel 2.4 to flat-out embarrass
many of the ultra expensive designs out there.

Adding Multichannel Capability to a Thiel

Speaker System
Another consideration that may be enticing to some is found in
Thiel’s unique options for adding multichannel based around
their stereo speaker pairs. Their PowerPoint® surround speak-
port but has the same time-domain characteristics as a vented ers are attractive, versatile beyond belief (they can be easily
box, which means more phase shift and group delay than a installed on floors, ceilings or front, side or rear walls!) and
sealed box. When listening, that translates to bass that plays a constructed to Thiel’s exacting standards using high quality
little slower than the band is actually playing, which has obvi- drivers similar to those in their floor standing designs.
ous repercussions to rhythm and pace.
Thiel’s MCS1 is designed for center channel use and employs
Passive radiators, however, are superior to vents in eliminating a coaxial midrange/tweeter very similar to the one used in the
port chuffing (air moving through the port that’s annoyingly 2.4. The long-awaited SmartSub® system is among the most
audible at the listening position), and eliminating any potential innovative subwoofer/room correction system designs around,
for the backwave from other drivers in the enclosure coming to which I hope will be enough to one day lure Dr. Boom himself
the listener through the port too close in time to the primary (Richard Hardesty) out of retirement from the field of reviewing
signal. subwoofers.

Another unique design choice by Thiel is the use of “under-

22 www.audioperfectionist.com
 CS2.4 Review

Setup and Reference System reviewing them for the Journal), and Ayre’s P-5xe phono
I achieved the sound I liked best with the 2.4s just under 7 feet stage. Source components included Ayre’s D-1xe CD/DVD
apart (center to center), with a distance of just over 9 feet from player, and a Linn LP12 turntable with all the latest accou-
trements: Cirkus bearing, Lingo power supply, Ekos tonearm,
Akiva cartridge, and the “Speed” carbon fiber mat from
Extreme Phono.

My system cables are all AudioQuest’s battery-biased lines:

Sky and Cheetah interconnects and Kilimanjaro speaker cable.
My power cords are AudioQuest NRG-5 and Richard Gray’s
Power Company High Tension Wire power cords. I also use
RGPC 400S power line conditioners.

Listening
As soon as the Thiel 2.4s were set up in my room, they made
great sound, immediately exhibiting the expansive soundstage
and convincing image focus that sets time- and phase-coherent
designs completely apart from conventional speakers.

Conventional speakers tend to create a strong image deep in

the center of the soundstage that compresses sharply front to
back at both sides, rather like a triangle. The Thiel 2.4s main-
tained layered image depth and excellent focus far out to both
sides, and far behind the speakers. The imaging capability of this
speaker can’t be described as anything less than spectacular.

“The Thiel 2.4’s soundstage

was completely and utterly free
each speaker to the listening position (an included angle
between the speakers and listener of considerably less than 60
of the physical boundaries
degrees, which is less than Thiel’s user manual recommends).
of the speakers…”
In my 24.5' x 17.5' room the distance from the coaxial
The Thiel 2.4’s soundstage was completely and utterly free of
mid/tweeter to the respective side wall was just over 5 feet and
the physical boundaries of the speakers, and very open sound-
each speaker was more than 5 feet from the front wall. This did
ing. Not only did this speaker never sound boxy, it never
not optimize bass performance in the room as there was little if
sounded like it was there at all, even in the bass.
any boundary reinforcement from this position, but produced
Vandersteen’s 3A Signature, which I owned for nearly three
the best soundstage with the most coherent, convincing, pre-
years, has better bass extension, but that comes at the price of
cise imaging and best soundstage depth.
sounding a tad boxy in the lower registers.

The amplification components in my system during the Thiel

The Thiel 2.4 doesn’t go as low, but you never hear the box
review consisted of VTL’s TL7.5 preamp, Theta Citadel
either. I suspect this is a tribute not only to the cabinet con-
monoblock power amplifiers (both of which I bought after

www.audioperfectionist.com 23
 CS2.4 Review

Low bass (50Hz and below) is where I’d describe this speaker
having a slight subtractive coloration. Although its -3dB point is
specified at 33Hz, low bass lacked size and impact in my room,
but also sounded just a bit loose at the same time.

Big acoustic bass sounds, for example, were a little thin on

body sound and the strings didn’t snap quite as tautly as I’m
used to. I’m certain I could have improved the bass extension
by moving the speakers closer to the room boundaries, but my
biases are such that I’d prefer to maintain the spatial perform-
ance derived from having the speakers out in the room.

If you want more bass from this speaker it’s my opinion that
you should add a quality subwoofer to your system and leave
the speakers out in the room where they image best.

What about that aluminum midrange driver? Does it have a

sound? Yes, in my opinion, it does. The Thiel 2.4 unequivocally
does NOT have anything resembling the harsh, metallic sound
of other speakers I’ve heard using aluminum midranges.
(Monitor Audio and RBH are two examples of speakers I’ve
heard using aluminum midranges that sound just plain nasty.)
There is a damped, restrained coolness in these Thiels, and
struction but also to Thiel’s distortion-minimizing woofer the midrange of the 2.4 is certainly not as relaxed as other
designs, explained in detail earlier. Superior components like speakers that don’t use aluminum midrange drivers, including
the Thiel 2.4 make your system sound less mechanical, less the Vandersteens I own. But there is a big difference between
like a system and more like music naturally occurring in space. sounding cool and sounding metallic, and in my system these
To that point, “free” and “open” are two words that pepper my Thiels never crossed that line.
notes on the listening sessions I spent with this speaker.
Occasionally I felt this speaker gave up a rather teensy bit of On harder recordings there’s definitely less forgiveness than
ground to the dense, fully rounded 3-D imaging I get from my some are used to hearing, but there was no bite or glare either
reference speakers. But take this nit-picking in its proper con- and this sound didn’t prevent me from getting deeply involved
text. No conventional [phase incoherant] design I’ve heard at with the music and thoroughly enjoying this loudspeaker. So
any price is even worthy of comparison to the spatial precision long as you stay away from components with a pushed, hard
and dimension of the Thiel’s soundstage. midrange you’ll hear open, highly resolved, slightly-on-the-cool-
side-of-neutral sound from these speakers.
Time is not the only domain in which the Thiels are coherent.
Tonally the 2.4 is exceptionally balanced from top to bottom, as Further expanding the 2.4’s charm is a very engaging sense of
even and neutral a presentation as I’ve heard from a speaker. transient speed, dynamics and lifelike snap. Drum kits had out-
The top is airy and extended without being zingy or calling standing pop and excellent dynamic contrast. Micro changes in
undue attention to itself, and the midrange is resolved, if a little voice level or the intensity of plucked strings were clearly
cool (more on that in a minute). The mid-bass (50Hz-100Hz) apparent.
sounds noticeably quicker and cleaner than typical vented
designs, if just a tad over-damped. Combined with the alu- While the 2.4s were in my system they pulled double-duty for
minum midrange, this gives the 2.4 the cool signature Thiels home theater playback and did an excellent job. I used Theta’s
are known for. powerful Citadel amplifiers (400W per channel into 8 ohms),

24 www.audioperfectionist.com
 Interview with Jim Thiel

and was more than pleased with how the Thiels performed in match its high standard of performance is the icing on the
my system even when driven at demanding levels. cake. The Thiel 2.4 is the kind of product that high-end enthusi-
asts should celebrate! APJ

“I think at $4,200 a pair Thiel Audio

1026 Nandino Blvd., Lexington, KY 40511-1207
this speaker is one of the (859) 254 9427 www.thielaudio.com

finest values in audio.”

In terms of overall resolution this speaker outclasses nearly
everything I’ve heard at or near its price point, and is frankly An Interview with…
superior to many speakers I’ve heard that cost multiples of its
price. If pushed I’d admit that I believe the Vandersteen 3A JIM THIEL by Richard Hardesty
Signature can take you farther into the recording space, and I’ve known Jim Thiel (cofounder and chief designer) and Kathy
reveals more low level detail, but I also think a lot of listeners Gornik (co-owner and company president) since the late ‘70s
will gravitate to Thiel’s clean, open and lively sound. And if form when I became a dealer for their speaker products. I sold,
factor becomes an issue, the Thiel has scoreboard, although
that comes at a price.

Another factor in the valuation of this speaker is what’s hap-

pening with the dollar. It is not my intent to get jingoistic here,
but current events dictate that buying speakers made right here
in the US of A has never been a better deal. With the poor
position of the dollar in the world economy, and the fact that
importers and distributors have to add a good percentage to
the price of their products to cover their expenses and make a
profit themselves, the odds are you’ll pay more and get less
from an imported speaker.

Conclusion
The Thiel 2.4 is simply superb. It occupies a small footprint in-
room for a full-range, floor standing loudspeaker and can be installed and repaired Thiel speakers for many years and am
purchased with a gorgeous furniture-grade finish that makes it quite familiar with the quality of their construction and the peo-
an attractive and practical speaker to share your living space ple who make them.
with. I feel compelled to mention the price because I think at
$4,200/pair this speaker is one of the finest values in audio, but Jim, Kathy and I have maintained our personal friendship and
I also feel that mentioning the price denigrates this speaker in Jim has been one of the truly knowledgeable engineers upon
some respect. The Thiel 2.4 is not a terrific speaker at this whom I have relied for education over the years. If I needed
price—it’s a terrific speaker in its own right, regardless of all business advice I’d call Kathy and she was always ready to
other factors, and I thoroughly enjoyed listening to it. help. When I wanted to know about the function of an audio
component I’d call Jim and he’d always share his experience
The Thiel 2.4 is a hand-crafted, high resolution, time- and graciously while supplying concise answers to my questions.
phase-coherent speaker that will simply embarrass many con-
ventional designs costing much, much more. That its looks A few years ago my wife Paula and I visited the Thiel factory in

www.audioperfectionist.com 25
 Interview with Jim Thiel

beautiful Lexington, Kentucky. This was an enjoyable experi- of magical. So this work suits my personality well and I really
ence that provided an indispensable glimpse into the sophisti- like it.
cation of the current state of the art in speaker manufacturing.
The value of Thiel products can best be realized when you see Had you been experimenting with making audio components
what goes into them before and during construction. Thiel for a long time, before you got into the business?
speakers are among the most thoroughly engineered products
available and an interview with Jim Thiel is sure to provide Yes, electronics was actually what I knew much better than
valuable information to Journal readers. acoustics and speakers and I had built amplifiers and pre-
amplifiers and also band equipment, including guitar and PA
Jim, tell us a little about yourself and how you got into the amplifiers.
loudspeaker business.
When I decided I wanted to start my own business I considered
I’ve liked music a lot since I was young. I took piano lessons doing electronics like amplifiers and other electrical compo-
when I was 6, 7 and 8 years old. I played in my high school nents. But I thought, rightly or wrongly, that there was more
band. So that’s part of it. I’m also a technical person who likes room for improvement in loudspeakers—and particularly more
to work on challenging problems and great sound reproduction room to make improvements that people could appreciate. I
is, I think, a challenging problem. Nobody has succeeded in thought that I could make a better amplifier but it might not be
doing it perfectly, so we can always strive to get better results. obvious, or appreciated by that many people, that it was a bet-
ter amplifier. I thought that I might be able to make speakers
It’s interesting to me to apply technical effort toward music that were enough better that a lot of people, just on hearing the
reproduction and to think that when you’re all finished what product, would realize that it was better. So I decided to build
you really have is an audio product that’s made out of metal my business around speakers, even though when I began I
and wood and plastic—and music comes out! And that’s kind didn’t really have any professional experience in loudspeaker
design, other than having built my own
speakers as a hobby through high
school.

I’ve been happy with speakers because

electronic engineering is involved,
mechanics and acoustics are involved—a
lot of different things, including material
properties. Also, certain aspects of speak-
er designs don’t readily lend themselves
to measurements or pure simulations or
solutions by mathematics alone. So you
also have to bring a fair amount of intu-
ition into the mix.

Time domain performance has been

ignored by most of the major loudspeak-
er manufacturers and the magazines that
review their products. Why are Thiel
speakers time- and phase-accurate?

The approach I’ve taken to speaker

design from the beginning is to consider it

26 www.audioperfectionist.com
 Interview with Jim Thiel

Here is, I think, one

of the interesting
things about loud-
speakers. This was
not a scientifically
controlled experiment
in the sense that the
only difference
between these two
speakers was that
one was time-coher-
ent and the other
was not, because it
was not possible for
me to make products
that were identical
except for that factor.
So the test probably
wouldn’t convince a
skeptical person
beyond a shadow of
a doubt that time-
a problem with solutions and my job is to find the solutions. coherence makes an audible difference. But fortunately I don’t
Loudspeakers don’t sound like a live musical performance. Why have to convince everybody else—I only have to convince
is it that when you close your eyes you can tell whether you’re myself that it’s worthwhile. Nobody has to agree with me, but if
listening to the sound from a speaker or an actual musician? I’m going to make speakers I want to know for myself whether
something is worthwhile. And I do think it’s worthwhile and I
I continually ask myself this question: what are loudspeakers can do it whether it’s been objectively proven or not.
doing that makes them sound like loudspeakers? And I’ve come
up with a lot of ideas. For example, each note—from a piano or Anyway, listening to these two speakers, there was actually
a guitar or whatever—is composed of a series of harmonics more of a difference than I expected, almost a jaw-dropping
that, in real life, arrive at your ear at precisely the same time experience! The phase- and time-coherent speaker seemed
with precise phase relationships. A typical speaker takes that much more realistic to me. There was much more sense of
signal and breaks it down into different frequency bands, each space and depth and clarity.
of which gets to your ear at a different time with the phase rela-
tionships between them obliterated. So that is something that a There were other tonal differences that probably could have
speaker does to alter the signal and that would be one reason thrown off an inexperienced listener, but when you have some
speakers don’t sound like live musical performances. experience with what tonal differences sound like you can hear
through that and appreciate the difference that the other char-
I set up an experiment to test that theory. I built a loudspeaker acteristics are making. So I became quite convinced that this
that was phase- and time-coherent and did not have any of was an important aspect of speaker performance—to make
those distortions. I compared that speaker, when it was wired loudspeakers that sound more realistic.
up to be time- and phase-coherent, with the same type of
speaker wired up not to be phase- and time-coherent to see if I Many of the diaphragms in Thiel speakers are made from alu-
could tell the difference. minum. Why did you choose this material and what advantages
does it provide?

www.audioperfectionist.com 27
 Interview with Jim Thiel

because,
although it’s
much stiffer and
offers a wider
bandwidth that’s
free of distortions
produced by the
diaphragm, at
some high fre-
quencies there
are resonances
that need to be
compensated for
or corrected. And
some manufac-
turers aren’t
compensating or
correcting these
resonances,
which create
audible prob-
lems. These res-

Ideally we want a driver diaphragm to be infinitely stiff and light onances can make the diaphragm sound like a metal
enough so that we can maintain reasonably high efficiencies. diaphragm, which of course you don’t want. So it becomes a lit-
The benefit of an infinitely stiff diaphragm is that it could move tle bit more involved to use a metal diaphragm effectively. The
as one piece at all frequencies and not introduce any distor- reason I consider it a better material is that we get drivers that
tions or colorations that would result if it were internally res- perform well over a wider range of frequencies using aluminum
onating. We can’t make a material that is infinitely stiff so we diaphragms.
use a material that is as stiff as possible for it’s weight and alu-
minum performs very well in that regard. As the voice coil in a conventional drive element moves inward
and outward it encounters a varying amount of ferrous material.
There are also other considerations and those have to do with This can produce variations in the inductance of the coil and is
the material being affordable, formable, and able to give very a mechanism for distortion. You use underhung voice coils in
consistent results from unit to unit. Aluminum is very good in all many drive units to eliminate this nonlinearity. Can you tell us
these areas. You can easily form it into all kinds of shapes, it why you chose this unusual construction method and how lis-
gives very consistent results and it’s a very practical material. teners benefit?

The old paper diaphragms were really not bad at all. The wood One of the reasons is exactly what you describe. Normally, as
fiber material that they’re made of was in nature’s R&D depart- you said, the voice coil has varying amounts of iron inside the
ment for a billion years to evolve into strands with very high coil, depending on whether it’s moving inward or outward while
strength-to-weight ratios and they actually do a very good job. producing the sound and that changes the inductance of the
The biggest problems with the paper diaphragms had to do coil and, therefore, changes the frequency response of the
with inconsistencies from unit to unit, and from batch to batch. speaker.

Metal diaphragms are much more consistent but there’s a com- So every time the cone moves in and out the frequency
plication with metal. You can’t use aluminum indiscriminately

28 www.audioperfectionist.com
 Interview with Jim Thiel

response is changing, as you described. By using the short coil I didn’t invent this. It’s like phase- and time-coherence that
and a long magnetic gap instead, all of the coil is always within have existed in text books for many decades. Often, engineers
the gap and, therefore, the amount of iron in the coil does not don’t choose to execute things because of cost or engineering
change and the amount of inductance doesn’t change and, difficulty. And I think there’s some cynicism too, that nobody will
therefore, the frequency response does not change. But that’s hear the difference anyway.
actually what I consider the second most important reason.
In the early days of my business, I used to wonder, Should I be
The larger benefit is that, because the short coil is always spending all this money and time doing this? Will anybody hear
entirely within the magnetic gap, it always experiences a non- it? My answer to myself was, If nobody hears it and nobody
changing magnetic field strength. Therefore, it’s able to produce cares, that doesn’t mean I want to design speakers the normal
way anyway. It really
means that I would
want to find another
line of work. If I’m going
to design speakers I
want to try to design
them well.

You want to design

speakers that you can
take personal pride in.

Exactly!

That’s how everything

good gets made, isn’t
it?

I guess so. But it’s a

pleasant surprise that it
turns out there are peo-
ple out there that can
a force that is proportional to the input signal from the amplifier. hear the difference—maybe not “most” people, maybe not a
large enough group to allow me to buy a great mansion. But
Normally a voice coil is longer than the magnetic gap and, there are people who appreciate it a lot. We get letters all the
therefore, as it moves in and out, the amount of magnetic field time from customers telling us they’ve gotten so much enjoy-
acting on it changes. And that’s the major distortion-producing ment from our speakers, even the ones they bought 15 years
mechanism in a loudspeaker. Usually, 90% of all the distortion ago, and thanking us for doing what we did. And that’s pretty
produced in a loudspeaker comes from that mechanism within neat!
the driver motor systems. We can reduce that to a tenth or less
of what it would otherwise be by using the short coil and a long Many Thiel speakers use coaxially mounted midrange and
gap. The only problem is that then you need a much larger tweeter driver units. Why did you go to the trouble of develop-
magnet than you otherwise would and you also need bigger ing the special drive elements required to accomplish this phys-
front plates and the cost of the driver is higher, but the distor- ical arrangement?
tion is much less so you get a much cleaner, purer sound.
That way, we can insure that the sound from those coaxially

www.audioperfectionist.com 29
 Interview with Jim Thiel

coaxial drive units that share

the same motor system and
use a mechanical crossover.
There’s only one magnet sys-
tem and one voice coil, but
that voice coil is connected to
a mechanical crossover sys-
mount- tem that allows only the lower
ed drivers gets frequencies to reach the
to the listener’s ears at midrange diaphragm, whereas
exactly the same time. I consid- all of the frequencies go to the
er that important, as we talked about ear- tweeter diaphragm, including
lier. The other methods (positioning the drivers on a sloped baf- the very high frequencies.
fle) that we originally used, and still use, to achieve that time-
coherence work very well in most cases, but not necessarily in So that compliant ring that
all cases. A slopped baffle puts a limitation on the speakers separates the two decouples
that they can’t be mounted on the ceiling or put up on a shelf them at higher frequencies?
because then you lose control over how far away the listener is
going to be from each of the drivers. By mounting them coaxi- That’s exactly right. The cou-
ally, we can insure that no matter where the speaker is placed pling suspension, as we call it,
and no matter where the listener is, he will always hear the decouples the midrange
sound from those two drivers at exactly the same time. diaphragm at higher frequen-
cies so it stops moving, leaving only the tweeter to move exact-
Like a lot of things in loudspeaker design, solving one problem ly the way a tweeter would normally move. The benefit of this
can create other problems. In this case, we found that you can’t system is that the costs for three-way speaker performance are
simply mount a tweeter in the center of a woofer because that really not much more than they’d be for a normal two-way sys-
affects the response of both drivers. Even though you can do tem where you have just a woofer and a tweeter.
that fairly easily to get correct time-coherence, other problems
are created. For example, the tweeter really does not like being Here, we have a woofer and this mechanically coupled coax,
in the acoustic environment of the throat of a horn, which is the but there is no additional electrical crossover network between
shape of a normal woofer cone. Even if you have a perfect the midrange and the tweeter and there’s no third magnet sys-
tweeter, when you mount it in the center of a deep cone woofer tem required for a midrange driver so the costs are not much
its response gets screwed up. So we had to do something about higher than they would be for a two-way system, but you get the
that problem before we could use the coaxial mounting method. performance of a three-way system because the tweeter doesn’t
With attention to other factors, it does give you a great and pre- have to come all the way down to meet the woofer. You have a
cisely accurate way of achieving time-coherence. larger diaphragm than the tweeter operating in the midrange so
I consider it a way to get much better performance for a small
Some of your coaxial elements have separate voice coils and price than you can with a normal two-way speaker system.
use conventional crossover networks, and some use a single
voice coil and a mechanical crossover to drive both I’m really happy about how this has worked out. This is an
diaphragms. Can you explain this for us? example, by the way, of one of the ideas we use that I did think
was original but on investigation it turns out that this concept
As you said, many of our coaxial drivers are more or less con- also was patented and used back in the 1930s.
ventional in that there is a self-contained tweeter with its own
motor system that is mounted within the self-contained Really? I’d never encountered that before. I’ve seen “wizzer”
midrange with its own motor system. But we do make some cones used in attempts to extend high frequency response of

30 www.audioperfectionist.com
 Interview with Jim Thiel

can hold them in the positions that cause the sound to reach the
listener at the same time. Another task the baffle needs to per-
form is to not reflect or diffract any of the energy from the drivers
so the driver is putting out its energy into the room unaltered by
the edges or steps or angles in the baffle itself. That’s the reason
we round the edges of our baffles—to achieve as little interfer-
ence as possible with the sound of the drivers.

The baffle, and the rest of the cabinet, should not generate any
sound of its own and the only way to accomplish that is for it to
be inert and not vibrating at all. The best loudspeaker enclo-
sure would be made with thick, reinforced concrete or some-
thing that’s extremely strong and won’t vibrate at all—but that’s
impractical. Although we did use concrete in some of our big
speakers. We approach this problem in the more moderately
priced products by using very thick materials in our baffles.
Depending on the model, our baffle material is 2 inches or 3
inches or 4 inches thick to minimize vibration. So if the baffle
holds the drivers in position, does not reflect or diffract or inter-
fere with the energy radiated by the drivers into the room, and
does not vibrate itself then we consider that it’s doing its job
very well.

Besides the beautiful finishes what other special characteristics

are incorporated into Thiel cabinets?
CS7.2

midranges but I’ve never seen one with a mechanically decou- Stressing what we just talked about, ideally the whole cabinet
pling crossover… I looked at that and thought, Gee, if he tells needs to be so rigid that it will not vibrate. You have these driv-
me how this works he’ll probably have to kill me! This has got ers that are producing sound by vibrating in and out and sound
to be secret. is radiated from the front of the diaphragm and that’s what we
hear. But sound is also radiated from the rear of the diaphragm
No, it’s not secret. The difficulty is in making it work in practice. and goes into the speaker cabinet. The job of the enclosure is
We literally had to make over 100 experimental drivers before to completely contain that energy so it doesn’t get out into the
we worked out the exact material and geometry needed for this room to distort the sound that you’re hearing—and that
coupling system in order to make the thing work well so it was becomes very difficult.
a long project. It’s kind of simple in concept but to have it work
properly in practice was not so easy. Like I said, if we had foot-thick concrete cabinet walls it would
be a lot better but that isn’t practical. We use 1 inch-thick mate-
Many of your front baffles are sloped to temporally align the rial for our cabinet walls and a lot of bracing inside the cabinet
drivers and they are exceptionally thick and contoured to mini- to reinforce the strength of the cabinet walls so that they vibrate
mize stored and diffracted energy. Can you elaborate on how much less and radiate much less distortion to the listener.
you arrived at your baffle configurations?
Your speakers are time- and phase-accurate so I assume that
We’ve pretty much covered the high points. The baffle has sever- they use first-order acoustic slopes to integrate drive units.
al tasks to perform. The basic one is to hold the drivers in the What other special design features do your crossovers
position that you want. In our case, the baffles are sloped so we employ?

www.audioperfectionist.com 31
 Interview with Jim Thiel

crossover. If you were a great enough engineer to develop driv-

ers that had first-order acoustic roll-offs in themselves, then
you wouldn’t need an electrical crossover network except, pos-
sibly, to keep low frequencies from passing to the tweeter.

So designing a first-order crossover is much more complicated

than designing an electrical first-order network.

What the crossover network has to do is whatever is required

so that when you add its response to the response of the driver
the net result will be a first-order acoustic roll-off. The
crossover network needs to work with the driver so you can’t
generalize about what such a crossover needs to do because
the response that it needs to provide is different depending on
which drivers it’s working with.

In addition to the general requirement of working with the dri-

ver’s response to produce the first-order roll-off characteristics,
the crossover also has some other things that it must do. It
needs to correct the response irregularities of the drivers them-
selves. And it must correct for alterations in response that the
cabinet contributes. And, of course, it must precisely match the
Your assumption is correct; we do use first-order acoustic
levels of all the drivers.
slopes to integrate the drive units. As you implied, that is the
only way to achieve true time- and phase-accuracy. First-order
Another thing that we do, that usually isn’t done, is to include in
crossover systems achieve not only accurate frequency
the crossover network additional compensation for the imped-
response but also completely accurate time response and
ance changes of the speaker. That allows the speaker to pres-
phase response and energy response. So they’re completely
ent a much more uniform and resistive load to the amplifier.
accurate in every way and it’s the only type of crossover that
Now, technically that’s not changing the sound of the speaker
does have those characteristics.
at all because these circuits are not in the signal path. But cre-
ating a consistent load allows some amplifiers to sound better
“What the crossover network than they otherwise would. It really depends on the amplifier

has to do is whatever is
required so that when you add
its response to the response of
the driver the net result will be
a first-order acoustic roll-off.”
You’re also correct in pointing out that it’s the acoustic slopes
that are important, and not the electrical characteristics of the

32 www.audioperfectionist.com
 Interview with Jim Thiel

but some sound better—sometimes significantly better—by

working into a resistive load rather than a reactive load.

So a crossover network has a lot to do and I think it’s the most

important element of the speaker. All the basic elements of a
loudspeaker—the drivers, the enclosure, the crossover net-
work—are very important, obviously. But I think the crossover
network is perhaps the most critical.

I believe that speakers should be demonstrably accurate and

provide good sound. What measurements do you find the most
informative? Do you measure in any unusual ways?

Measurements are absolutely necessary but are not sufficient

to evaluate loudspeaker performance. The one I consider to be
most valuable—and the oldest one probably—is frequency
response. I think that the only frequency response measure-
ment that’s of any value is a true anechoic frequency response.
To my way of thinking, it’s of no value to measure the so-called
“in room” response of the speaker because you’re really just
measuring the response to the room and…well, that’s another
story.
ments are by far the most useful. There are some caveats
I think you really need to know what the response of the speak-
about that. What would look good on paper might not sound
er is independent of the room and, therefore, your frequency
like a good frequency response measurement.
response measurements should be anechoic measurements.
Your ear is so fantastically sensitive that, I believe, you can
hear frequency response deviations of as little as one tenth of
“For lower frequencies I a decibel if they’re across a whole octave. It’s practically

need to measure outside.” impossible to take measurements that accurately, and even if
you could there are all kinds of things going on in the perform-
ance of the speaker that result in errors greater than .1dB.
Unlike 20 or 30 years ago, the equipment now is low in cost
because we use computers to achieve anechoic measurements
Even a great speaker will have frequency response errors on
without actually building an anechoic environment. But we still
the order of ±1dB, which is audible. But complications arise
need the same size and space that was needed with the old
because it depends on what causes those irregularities. If you
anechoic chambers in order to get resolution down to workable
have a minor diffraction mechanism that’s causing a response
low frequencies. So even today, you can’t make useful ane-
irregularity of 1dB, I’ve found that it’s usually not very notice-
choic frequency response measurements of a speaker at low
able. But if you have a high-Q resonance that’s causing a
frequencies in a normal room. For example, the room that I use
response irregularity of 1dB it can be quite audible and irritat-
has a 20 foot-high ceiling and the speaker is suspended
ing. So you can’t just say that if the speaker measures flat with-
halfway up and even that only allows me to get accurate read-
in ±2dB it’s good—and if it doesn’t it’s bad. One speaker could
ings down to about 200 cycles.
measure flat within ±2dB and not sound good at all, and anoth-
er might not measure any better in terms of frequency
For lower frequencies I need to measure outside. It’s tricky to
response but sound quite good.
get such measurements but the frequency response measure-

www.audioperfectionist.com 33
 Interview with Jim Thiel

It would be a boring world if we could measure everything, woofer in the system that’s integrated well. So you could make
wouldn’t it? speakers that would be good for movie systems that have a

Yes, and we wouldn’t be having this conversation and all

speakers would be perfect and we’d be doing something else. “A lot of people want the
But that’s not the way it is. There are other measurements that
can be taken—but I find that I don’t take very many of the time speakers to play excruciatingly
response measurements or even phase measurements
because all of that can be either calculated or is connected loud with movies.”
with the frequency response anyway.
subwoofer and would not necessarily be particularly applicable
Especially if you use an impulse as a stimulus... to a music system that didn’t have a subwoofer. So most of the
differences have to do with how loud they’ll play and how low
Yes, that’s a good excitation signal so that you can theoretically they’ll go and if there’s a subwoofer or not.
get all the information that is to be gotten in every domain from
that signal. The real question about measurements becomes Tell us about your subwoofers.
not which measurements to use—and you said this almost in
these same words a few minutes ago—I think great measured I’ve been working on subwoofers a lot longer than I planned!
performance is absolutely necessary but it is in no way suffi- It’s been five years now and we finally have several products
cient. It’s a starting point. on the market.

The Audio Perfectionist Journal is primarily about accurate The idea of great deep bass has always been very appeal-
music reproduction but many of our readers also watch movies. ing to audiophiles. That’s one of the things we love to have
Thiel makes a variety of home theater products. Do you do in our music reproduction. Unfortunately, great deep bass is
anything differently when you design a speaker for film sound? difficult; it tends to make the speakers large and inefficient
and expensive. Subwoofers came along and now we think
Well, yes and no. In terms of the basic performance values of we can have great deep bass. But in musical terms, most of
accuracy, time-coherence, low distortion and wide dispersion them don’t really sound very good at all. And they can sound
and on and on, we don’t do anything different at all and the kind of horrible! So many audiophiles think that a good
reason, of course, is that an accurate speaker will reproduce all music system shouldn’t include subwoofers because they
signals accurately and it doesn’t make any difference to the don’t sound good. So what we’ve tried to do is to develop
speaker if the signal comes from a music source or a movie subwoofers that sound really great in musical terms—and I
source. So a speaker that is accurate enough to be great for think we’ve succeeded.
musical enjoyment will also be perfectly accurate for movies.
So, in those ways we don’t do anything differently. However, There are a couple of problems that subwoofers have that
there are some practical differences and most of them have to other speakers don’t have. Probably the simplest issue is that
do with how loud the speaker can play. subwoofers are usually placed against walls and in corners and
those placement positions really mess up the bass response of
A lot of people want the speakers to play excruciatingly loud any speaker—not just of subwoofers.
with movies. So something had to be done about this because
speakers we traditionally make for listening to music won’t play You’d never take your great two-channel full-range speaker and
that loud. So what can be done is to make a speaker that’s not put it back into a corner because, in addition to the other prob-
nearly as extended in bass response. That doesn’t directly lems that would cause, it would really screw up the bass
compromise any musical values in terms of accuracy but it lim- response. But that’s where people often put subwoofers and
its the range that the speaker is trying to reproduce. when they get screwed up bass response they wonder why.
That works well for movie systems if you have a good sub-

34 www.audioperfectionist.com
 Interview with Richard Vandersteen

So we’ve developed a really nice way of correcting that prob- This allows the addition of a subwoofer—which can provide
lem right at the source, right at the subwoofer. You tell it where deep and powerful bass that is up to the standards of the most
it is positioned in the room, how far away from the side wall it finicky audience, in my opinion.
is and how far away from the front wall it is. Then it can know
what effect that placement will have on its response and pre- Thank you Jim for an informative and enlightening interview!
correct its response internally so that what it puts out in that
position will give you the sound that would’ve been produced if Richard I really appreciate your interest and willingness to do
it weren’t near any walls. this. I think what you’re doing is great—not only for your read-
ers and the industry but for me and our company. APJ
And you do that with that “smart” crossover.

This part of it we actually do inside the subwoofer itself where

you tell the subwoofer on the back panel what the side wall dis- An Interview with…
tance is and what the distance is to the wall behind the sub-
woofer. One reason this is done inside the subwoofer is that RICHARD VANDERSTEEN by Richard Hardesty
you may have multiple subs in different locations and may In the nearly thirty years I have known Richard Vandersteen we
need different compensations. Another reason is that you might have become close personal friends. My retail store was one of
not have our smart crossover. the very first dealers for Vandersteen speakers and I have first-
hand knowledge about the engineering behind each of the
We’ve developed a new type of audio equipment. We call it the
SmartSub® integrator. It’s similar to an electronic crossover and
its purpose is to blend the subwoofer with the main speakers and
that’s the other big problem. Even if subwoofers do produce high
quality bass, which is a big if, their output is usually not well
blended with the main speakers and what you end up with is a
very separate sounding, disconnected bass with a different char-
acter, which sounds unnatural compared with the midrange.

So with the SmartSub integrator, you tell the unit about the char-
acteristics of the main speakers and then it calculates the shapes
of the crossover filters that will give you perfect results with that
speaker. All the settings—what you want the low-pass frequency
to be, what you want the phase characteristics to be—are auto-
matically calculated for you to make the subwoofer blend with your
main speakers, which it knows about because you have told it.
products in the line. I sold, installed and repaired Vandersteen
That sounds like a very complicated device. speakers and offered my opinions about design and compo-
nent choices for decades. My prejudices lean toward
It was complicated to design and yet it’s very simple to explain. Vandersteen speakers because they embody the engineering
And I can tell you about it because we have a patent on it. choices that I have found to be important.

S+M=1 and so 1–M=S. So that’s what this unit does. It takes Richard and I don’t see each other frequently but we communi-
the input signal, 1, calculates what the main speaker is going cate by phone several times a week and share ideas and
to be contributing, M, and subtracts that from the input. The observations about audio components and systems as well as
result is the subwoofer output, S. a myriad of other subjects. We certainly don’t agree on every-
thing but we are in close agreement about how live music

www.audioperfectionist.com 35
 Interview with Richard Vandersteen

sounds and about which audio components best replicate sion set but there was always music in our home. My sisters
recordings of this sound. I believe that each of us has helped played piano and my father was an accomplished singer. Music
to mold the opinions of the other and I hold his engineering was always very important to us.
prowess in the highest esteem. He answers my questions
about audio engineering and I tell him what I hear when these We had a console with two turntables, one of which was capa-
engineering concepts are applied to audio components. ble of recording by actually cutting lacquer discs, which we
used to communicate with the family in Holland. We sent these
Richard Vandersteen and I have arrived at many mutually discs back and forth by mail so that the family could hear the
agreed upon opinions. We have reached these conclusions kids and also for basic communication. This device couldn’t
individually and together. We are predisposed to the ideas provide high resolution but it was an excellent introduction into
which we believe to be correct yet eager to adopt new positions recording technology.
when they are proven to provide a better explanation of the
reality we observe. I own Vandersteen speakers. In my youth I started experimenting trying to improve the sound
of the family audio system—which of course was mono in those
Richard tell us about your personal background. days—on a budget. I built some amplifiers and then some
speakers, which I was continually improving as time went on. In
My parents came to the United States from Holland, the the Air Force I studied electronics and after my stint in the mili-
Netherlands. I was born into an immigrant family here in tary, I continued experimenting with loudspeakers as a hobby.
Hanford, California [between Visalia and Fresno], and like Eventually that hobby became a business called Vandersteen
many families from Europe and Scandinavia we had no televi- Audio.

Isn’t it amazing that vinyl discs are still the highest resolution
source we have!

I’d agree, but I don’t know that there’s anything amazing about
it. It’s a simple system; it has more resolution than anything
else we typically play in the home and I think it’s charming and
there’s a little bit of a ritual to it but that’s part of the enjoyment.

How did you actually get into the speaker business?

Havens & Hardesty, in southern California, and California Audio

Systems, in Visalia, were the first dealers to place orders for
the early Vandersteen products. I was reluctantly dragged to
the Consumer Electronics Show in Chicago in 1977…

…Where you displayed speakers with response measurements

that I made …

Yeah, and we ended up with a bunch of dealers and a bunch of

orders—and that’s what I’ve been doing ever since… Almost 30
years now!

Many manufacturers would have us believe that loudspeaker

time domain performance is inaudible. Why are your speakers
time- and phase-correct?

36 www.audioperfectionist.com
 Interview with Richard Vandersteen

The very first speakers that I experimented with weren’t time- chain, just all different kinds of noises and different sounds, try-
and phase-accurate. I started by making speakers that provid- ing to determine which of these speaker designs most accu-
ed different “flavors” of sound. At that time there was the “east rately replicated the actual sound. We could determine that the
coast sound” and the “west coast sound” and few talked about signal passed through the playback chain including the amplifi-
accurate sound. er with low distortion. All that was left was the loudspeakers
and we’d listen to what came out of those and compare it to
When I considered building speakers professionally I gathered the actual sound.
together a group of friends with high-end audio experience and
asked which sound would be the most commercially success- During that process, we came to some conclusions about what
ful. There was no unanimous agreement. After a few beers, I was required for accurate reproduction. We decided that
said, “Now wait a minute. What does the signal really sound speakers needed only one driver per frequency range, first-
like? What is coming from the amplifier? Shouldn’t we replicate order filters, minimum sized baffles and flat frequency
that as best we can?” response.

Accurately reproducing the signal from the amplifier became “We decided that speakers
the new standard for this group. We would have long listening
sessions in which we’d compare real sounds to recorded needed only one driver
sounds. This predates our early experimentation with FFT com-
puter analysis of speaker performance and we had to gather per frequency range”
information empirically—by listening.
The original speaker models had multiple drivers and large baf-
Using a Revox tape machine we’d record all kinds of things: fles with frequency response tailored to deliver a certain
various musical instruments, including piano, shovels scraping “sound.” We ended up removing redundant drivers and the baf-
on the concrete, piling cardboard boxes, keys jingling on a key- fles in order to get rid of reflections from the box and repro-
duced sound became
more like the source.
Flat frequency response
was necessary in order
to accurately reproduce
a variety of recordings.
First-order filters
unquestionably provided
better replication than
filters with steeper
slopes.

And this just evolved

over time until the very
first commercial model
that we made was time-
and phase-correct, with
first-order filters and
temporally aligned driv-
ers mounted on mini-
mum-sized baffles. In

www.audioperfectionist.com 37
 Interview with Richard Vandersteen

the opinion of this listening panel, that’s what it took to most It’s expensive to do that engineering correctly and to manufac-
accurately replicate these different sounds that we’d recorded. ture these products. And although our cabinets are sophisticat-
ed, they can be made with automated machinery and the finish
So, your speakers are time- and phase-correct because you is not important. So you end up with about 16% of the cost
think that sounds better. invested in the cabinets as opposed to 50-80%, which is typi-
cally the case with loudspeakers. Creating speakers with sim-
Whether something sounds better or not is a matter of opinion. ple exteriors wasn’t really a deliberate thing, but in order to do
What I was trying to do was to replicate the sound we’d record- what we wanted to do at these different price points it was pure
ed when playing it back through speakers right after having economics.
recorded it.
So, you’re providing more performance for the money by not
Why don’t more manufacturers make time- and phase-accurate providing that fancy furniture.
speakers?
Well look at it this way. First-order time- and phase-correct
It’s a very, very tedious process. It requires high quality drivers, loudspeakers wouldn’t otherwise have been possible because it
most of which must be custom-designed in order to provide a couldn’t have been successfully done. We couldn’t have afford-
wide linear operating range. You cannot just design one and ed enough high quality drivers and enough time to do the hand
then duplicate it en masse or buy an off-the-shelf product. tuning of the crossovers if we had invested instead in cabinetry.

It’s important that the drive elements be reasonably flat in Why do Vandersteen speakers have a “family” sound?
amplitude response and reasonably low in distortion and the
crossover networks must be adjusted for each speaker to
ensure that. So basically, what you’re talking about is custom
designing, in an anechoic chamber, a crossover for each and
every loudspeaker made. That’s very time-consuming and not
really conducive to mass production.

Why do many of your products have simple cloth-wrapped

enclosures?

The high labor quotient required for tuning and hand-making

the crossover and the high quality drivers that are necessary—
and the fact that we use more of them to produce four-way
designs—leaves little money for fancy furniture, which really
doesn’t contribute to sound quality.

You can’t really get satisfactory time- and phase-correct per-

formance with only a two-way speaker. All of the criticisms that
we hear about first-order filters are relevant only when there
are too few drivers to share the spectrum. As the critics say,
the IM distortion and power handling would be seriously com-
promised in a two-way speaker. So that means you’re automat-
ically relegated to using at least a three-way—we select four-
ways in most models—and to divide the spectrum sufficiently
so that each driver can handle the range in a linear way.

38 www.audioperfectionist.com
 Interview with Richard Vandersteen

We do use slightly modified OEM units where that is adequate

to the task because it’s the most economical thing to do. The
best examples would be woofers or subwoofer drivers. There
are many commercially available woofers, for instance, which
are perfectly adequate to operate up to 500 or 600Hz.
Midranges and tweeters present a different situation. These
drivers must be more sophisticated than what’s available off-
the-shelf. In many cases it became necessary to design special
drive elements when no acceptable commercial units were
available.

Our midrange unit, for instance, has a unique construction that

Vandersteen Audio is all about preserving the waveform as eliminates reflections from the frame. That’s a design I came
passed to the loudspeaker from the amplifier. If speakers are up with and received a patent on. Because nothing like that
reasonably flat in amplitude, and reasonably low in distortion, was commercially available, we were forced to design and
and if they are time- and phase-correct, and if they’re replicat- source our own components and have them assembled. The
ing the waveform as accurately as possible then they’re all push-pull subwoofer driver used in the Model 5 is another
going to have similar sound. example of a completely unique drive element. I designed it
and made the tooling to manufacture it. It’s actually assembled
I’ve never understood these designers who create a line of from my parts by another company that specializes in this type
speakers which all sound totally different from one another. of work.
They don’t seem to have an actual opinion. All our speakers,
on the other hand, are based on the same design principles Your crossover networks appear complex but the signal path is
because those are what I think are most important for repro- often very simple. Can you explain this for us?
ducing live music, or repro-
ducing music in a way that
would remind you of hear-
ing it live—in a piano bar,
for example. So they
should all sound very simi-
lar to one another. For
more money, you’re either
getting lower distortion or
better power handling or
retrieval of more informa-
tion—but the basic sound
should be the same.

Many of your drive ele-

ments are uniquely con-
structed while other speak-
er manufacturers use off-
the-shelf drivers or slightly
modified OEM units. Why
was this added expense
necessary?

www.audioperfectionist.com 39
 Interview with Richard Vandersteen

Because each filter is first-order the signal path is inherently found that it made crossover networks using these capacitors
simple This is the only crossover type that has the potential to sound a lot better.
be—but isn’t necessarily—time- and phase-correct. However,
for speakers using these filters to function acoustically all those
extra crossover components are required to manipulate the dri- ...a properly designed PC board
ver’s frequency and phase characteristics so that everything
truly comes together at the listening position. The individual always sounds better than a
response characteristics of the driver and the crossover are
less important than the acoustic signal that actually arrives at hard-wired crossover...
the listener’s ear. All of our research considers the combined
results at the normal listening distance. I did some experimenting at the time and found that the same
was true with film capacitors. Most of us have noted that
So then, many of those components are to compensate for there’s a kind of warm-up that occurs in the first half-hour or
driver aberrations and aren’t technically in the signal path. hour of listening to even an excellent amplifier or an outstand-
ing and well broken-in pair of loudspeakers. It is my opinion
None of the compensating components are in the signal path. that the audible change that occurs in that first 30 minutes to
All of them are in parallel circuits used to correct the frequency an hour or so is due to dielectrics forming. The whole idea of
and phase response of the driver so that the speaker does battery biasing is to keep the capacitors in the same state
what we want it to do at the listening position. they’d be in if they’d been played for 24-hours or so. There’s
nothing mystical about this and it certainly does work.
Doesn’t that kind of debunk some of the naysayers who say,
“Look at how complex these networks are”? Speaking of dielectric materials, what capacitors are used in
Vandersteen crossover networks? What resistors and inductors
Well, the naysayers will always have something to say. do you prefer?

You pioneered the use of

batteries to provide a con-
stant bias voltage for
capacitor dielectrics in
crossover networks. Many
people have called this a
“snake-oil” approach. Is
battery bias simply hype
or does it really work?

Battery biasing was actu-

ally not my idea. Many
years ago, before every-
one became aware of how
important it is to use film
capacitors in the signal
path, there was rampant
use of electolytics. A cer-
tain listening panel applied
a bias voltage to elec-
trolytic capacitors and

40 www.audioperfectionist.com
 Interview with Richard Vandersteen

We wind our own inductors; we use only air core inductors in those with even slight differences between left and right.
the signal path. The capacitor dielectrics vary. They do sound We’ve recently released a new model called Quatro, for
different. We haven’t found one particular dielectric to be a instance. We first brought the prototypes to the CES show two
favorite in all instances; we often use combinations of three dif- years ago. Everyone at this year’s show commented about how
ferent ones for any capacitance that is in the signal path. We much better the Quatros sound now. There was no design
use styrene, polypropylene, and in some cases even Teflon™. change; the only difference between these two products is the
crossover, which went from point-to-point wiring on the proto-
We use non-inductive resistors that are made without any steel types to PC boards on the production units. They both have
parts and have copper leads and we’ve found that makes an exactly the same drivers and design. Many of our dealers
audible difference. They’re only available up to 5 watts so we asked, “What did you change? It sounds so much better.”
often have to use multiple units.
Vandersteen speakers have used minimum sized baffles to
Hard-wired crossovers are heavily advertised by some manu- support drive elements since the beginning. How does this
facturers and touted as superior. Why do you use supposedly affect sound?
inferior circuit boards?
Speakers with large baffles are easier to make because the
Over the last 25 years I’ve found that a properly designed PC baffle reflections randomize and “smooth” the response from
board always sounds better than the hard-wired crossover from the drivers, especially in the crossover region. Time smear is
which the board was designed because all the electrical char- the negative result of this smoothing.
acteristics are far more predictable and uniform. Exact compo-
nent positioning and consistent electrical impedance assures Obviously we have baffles, but very small ones to hold the driv-
uniformity and helps to ensure that the left and right speakers ers. If you minimize the size of the baffles you no longer have
will be identical. this reradiation support and you have to achieve flat response
by using better drivers and more closely matching thecrossover
Stereo imaging is enhanced when the left and right speakers networks. By eliminating this time smear, Vandersteen speak-
are exactly the same. Even with speakers that aren’t time- and ers achieve far better transient performance and provide more
phase-correct, closely matched speakers will image better than detail while remaining smooth.

What other special

construction tech-
niques are used in
your enclosures?

Loudspeakers create
a tremendous
amount of energy
and that can be a
big problem. We do
lots of different
things at different
price points to elimi-
nate or compensate
for the energy that
vibrates the enclo-
sure rather than pro-
ducing sound.

www.audioperfectionist.com 41
 Interview with Richard Vandersteen

It’s very expensive to make an enclosure that’s totally inert and cynic and I set one of the brooms right smack in front of the
impossible to completely achieve that goal. We attempt to driver—not 120 degrees out to the right or left as the actual
make speakers that are acoustically inert by using different dowel supports are. The other speaker didn’t have a broom.
sizes and thicknesses of different materials. If there is some
secondary radiation it’s spread out so that there is less energy We used a mono source and compared A versus B. The guy
over a wider range of frequencies rather than more energy con- agreed that he couldn’t distinguish between A and B and when
centrated at a couple of frequencies. Also our Models 1, 2 and we removed the blindfold he was quite shocked to see this
3 are tuned to make sure that that residual energy is in an up- broom handle, which is about the same diameter as the dowels
and-down (vertical) motion not a fore-and-aft (horizontal) in our loudspeakers, sitting right smack in front with no perceiv-
motion, which could affect the performance of the drivers. able effect on the sound. Now if that is true, and we proved
that it is, then moving those dowels far away from the natural
Do the dowels that support the cloth wraps on Models 2 and 3 radiation pattern should be even better.
have a negative sonic impact?
Some Vandersteen
speakers are spe-
cially designed for
home theater use.
How do these
designs differ from
your purist audio
products?

Well, they don’t

really. They’re all
time- and phase-
correct and have
flat frequency
response. If there is
a need for accurate
waveform replica-
tion it may be more
noted by those
who’re into home
theater because,
unfortunately, most
people are more
That’s a good question and we’ve been criticized about that a familiar with the sound of car doors slamming and phones ring-
lot over the years. I have proven by actual demonstration that ing than they are with live music.
the effect of the dowels is completely inaudible.
They may not be familiar with what a piano sounds like without
I love to give seminars because it puts me in touch with con- amplification in a little nightclub but they’re certainly familiar
sumers. I can learn what’s on their minds and how they’re with the everyday sounds that they hear in the movies. I know
using the stuff; it makes me more aware of their situations. these sounds are done with foley effects and are not necessari-
There’s always a cynic in every group and at one particular ly accurate but the foley artists are very good and they create
seminar he just kept coming after me about the dowels. I bor- sounds that are remarkably lifelike.
rowed a couple of brooms from the dealer. We blindfolded the

42 www.audioperfectionist.com
 Interview with Richard Vandersteen

The first year we introduced our center channel at CES in

Chicago, my wife and I and a sales manager from another
company were in the back of our long, skinny room. On the
right wall of the room was a counter with a phone above it.
Over the three or four days of the show we were showing the
film “Always” and during a tense control tower scene, a tele-
phone rings in the tower (on screen). Everyone viewing the film
in our room would look over to the right side wall and wonder
who was going to answer the real phone. It’s an interesting
effect and I think that replicating the waveform is every bit as
important in home theater as it is in audio. To do home theater
properly you need a very fine audio system.

“…most people are more

familiar with the sound of car
to the screen for listeners sitting off-axis. The VCC-1 does that.
doors slamming and phones Our challenge was to make a center channel that played louder
and went deeper but still worked as it was supposed to.
ringing than they are with We needed to find a way to add two woofers in order to get

live music.” higher levels with lower bass and lower distortion and to blend
them physically with a coax to make the speaker sound as if it
were a triax, so that all of the sound comes from the same
You have a unique center channel speaker, the VCC-5. Can
point. This would provide a center channel speaker that was
you tell us about it?
time- and phase-correct in all positions in the room—not just
for the guy sitting on the crack of the couch but the person on
The VCC-5 is a point source that uses four drivers. Because
the right and the person on the left and the person standing up
you don’t know where it’s going to be located it needs to sound
or sitting on the floor.
the same no matter where you are, or it is, in the room—right,
left, up or down. This was a real challenge.
All this work culminated in a patent-pending design that’s called
the VCC-5. It took us a year to accomplish that and we’re very
There was demand from the dealers for a bigger center chan-
proud of it even if few people understand how it works.
nel speaker that could play louder. Our small VCC-1 is a coaxi-
al design, point-source radiation, time- and phase-correct cen-
Vandersteen makes some unique subwoofers. How do the 2Wq
ter channel speaker. So we built a prototype of a larger center
and V2W designs differ?
channel speaker carefully adding two woofers—one to the
right, one to the left—and set the crossover very low at 100Hz
Well, the 2Wq is based on the 2W that’s been in the line now
in order to get more dynamic range and more low frequency
for about 20 years. Many things have changed over those
reach, and also lower distortion, because that’s what a bigger
decades and the V2W provides an answer for some of the new
center channel should do.
requirements that have arisen.

With these added drivers we encountered a comb filter effect

Twenty years ago, sub-woofers had a very bad reputation
and the speaker sounded different to each listener in the room.
because people couldn’t get them to blend, they stuck out like
This is exactly what a center channel speaker is not supposed
a sore thumb and there was a lot of negativity about them. We
to do! A center channel speaker is supposed to anchor dialog
spent many years perfecting a subwoofer that would blend

www.audioperfectionist.com 43
 Interview with Richard Vandersteen

seamlessly with the main speakers. We wanted to make it are using one of each. The 2Wq and the V2W subwoofers
sound like your main speakers got bigger and gained authority. look identical so they set a 2Wq in one corner so they can run
We found that the mids and highs were improved and imaging their main speakers full range while they’re listening to music
got better when the low frequencies were removed from the and have the benefit of a subwoofer. In the other corner they
main speaker structure and from the amplifier driving those put a V2W that is fed by the LFE output from their surround
speakers. We found that the character of the main amplifier processor. When they’re watching movies both the 2Wq and
could be retained into the deep bass with a passive, first- V2W are working and when they’re playing music they’re
order high-pass filter system that also helped provide a seam- using the 2Wq only.
less blend.
When you select “small” for the main speakers with surround
We still get calls every now and then from customers who say, processors you’re introducing at least a second-order (if not
“I bought one of your 2Wq sub-woofers for $1,300 and I can’t higher) high-pass filter. If you have all Vandersteens you have
hear it.” To me, that’s a compliment not a complaint! But for a time- and phase-correct loudspeaker system but you do not
some people, when they spend $1,300 they want to hear it, have a time- and phase-correct signal chain.
they want to know they got their money’s worth. That’s not
what we were attempting to achieve with the 2Wq. The advantage of using the 2Wq for the main speakers is that
it allows you to designate the main speakers as “large” thereby
2Wq getting a first-order transfer function and avoiding that second-
or higher-order filter that’s in the processor. You can experi-
ment with that just by switching between “large” and “small”
and listening to what happens to the naturalness in the mids
and highs.

The passive high-pass filter used for integration of the 2Wq

subwoofer and the subwoofer sections of the Quatro and
Model 5A speaker systems is claimed to offer some unique
advantages. How does this system work and why is it better?

Audiophiles get very nervous when you put an extra compo-

nent in the signal path because we know that any time you add
an extra component there’s going to be some degradation. This
is simply a fact. Although the battery-biased high-pass, which
now has a biased cable on its output, is very neutral and is
made with extremely high quality parts, we know that it still has
some negative impact.
The V2W, on the other hand, uses the same three drivers, slot-
loaded and everything, but it is a little more traditional although However, high-passing the amplifier driving a loudspeaker makes
it still has good transient behavior. It uses a large passive radi- such a stunning positive difference that it’s like taking one step
ator to allow it to play substantially louder and accepts a line- back and four steps forward. That’s still a net gain of three steps.
level input. In home theater, people require and enjoy more
energy. They don’t quite want to hear it the way it really is. The Years ago, a company came out with a thing called a “warp
V2W doesn’t have the ability to mimic the amplifier driving it knot” that was a device that plugged into your amplifier input. It
because it’s driven line-level not speaker-level by the cus- was simply a subsonic filter to get rid of noise from record
tomer’s amplifier. But it’s got more punch. warps and other deep energy that you can’t hear anyway, with
a first-order transfer function. Even though it was made with a
In systems doing both music and home theater, some people very low quality capacitor and it had a fairly significant sonic

44 www.audioperfectionist.com
 Interview with Richard Vandersteen

character, it was still deemed by many people to provide a pos- flat frequency response. The only penalty is noise because
itive sonic contribution. we’re building back on top of something that’s been rolled off.
However, with modern solid-state technology the noise floor is
Vandersteen high-pass filters perform a similar function at an so low that that’s not a factor.
even higher frequency allowing the amplifier to have an easier
life. Even “arc welder” amplifiers sound better when they’re not Tell us about the unusual subwoofer driver in the Model 5A
having to work hard at infrasonic frequencies. So do speakers. speakers.

If you can make a high-pass filter so transparent that the step Around the factory here,
backwards is very small, one has to focus on the fact that it’s commonly called the
you’ve moved four steps forward in all other areas. Your ampli- jackhammer. It’s a com-
fier and main speakers will sound better and you’ll have better pletely unique driver that
bass from the subwoofer(s) you’ve added. continues our commit-
ment to reduced bass
To demonstrate these facts at seminars we’ve misadjusted the distortion.
high-pass frequency to below 30Hz so the speakers were run-
ning pretty much full range. Then we’ve compared the high- We’ve always done
passed speakers to speakers that were actually running full proprietary things in
range. Virtually everybody has preferred the sound of the sys- our low-frequency
tem with the high-pass filters in the chain because, even drivers, which are
though they’re hearing all the information that’s on the record- required for long
ing, the subsonics and so forth have been attenuated. In all the excursion. A woofer
times I’ve done that experiment, I’ve never had anyone prefer does not behave lin-
the system without the high-pass filters in the signal path. early on its inward
stroke versus its
Then what happens? After you’ve high-passed the main speak- outward stroke and
ers, the woofers replace that information by… that’s partly
because of nonlin-
The woofer sections in the 2Wq subwoofer, and the Quatro and earities in the sus-
the Model 5A speakers, apply the inverse of the roll-off created by pension and part-
the high-pass filter. Additional equalization is applied to adjust for ly because when
the falling bass response of the sealed enclosures and the result you drive the
is fed to an amplifier(s) specifically designed to drive the load of voice coil inward there’s
Model 5A
the driver(s) used. more iron in the coil versus when
you drive it outward there’s less iron. In all of our speakers, we
Many people think that if you set a high-pass filter at 100Hz, extend that iron out and add faraday rings to minimize the
99Hz disappears. Actually it’s only reduced slightly in amplitude inductance change. Even with these additions some slight non-
and the signal level continues to fall as it goes lower in fre- linearities remain.
quency at 6dB per octave. The entire signal is still there; its
amplitude is simply reduced. It’s like the RIAA curve for a The push-pull subwoofer driver in the Model 5 has a sandwich
phonograph record. That’s why the amplifier likes it. And the cone attached to a voice coil former that goes all the way
main speakers like it. through it from one motor to another. The cone is shaped like a
very thin flying saucer and there’s a motor on both ends of the
All that low-frequency information is dramatically diminished voice coil so it’s truly driven push-pull and it’s truly linear over
from the amplifier circuits and the loudspeaker structure. The its one and a quarter-inch stroke. Motion is always identical in
subwoofer amplifiers have the inverse of that built in, to restore positive and negative directions.

www.audioperfectionist.com 45
 Engineering Double Standard

Thank you Richard. This has been an enlightening interview

and I’m sure the information it contains will prove useful to our
readers. APJ

The Engineering
Double Standard by Richard Hardesty
The magazines and the reviewers who influence the high-end
audio industry have adopted a double standard for engineering
competence that has had a negative impact on all of us. This
double standard is pervasive. They expect all audio compo-
nents other than speakers to meet certain minimum standards
of objective performance. At the same time they treat speakers
as if they were works of art to which no objective standards
apply. This position is both false and misleading.

There are objective tests to gauge the fidelity of speakers and

these tests are frequently more comprehensive than the recog-
nized tests that gauge the fidelity of other components.
Objective tests can’t tell the entire story but they certainly can
expose those products which are poorly engineered and can’t
possibly reproduce the recording accurately.
So all the nonlinearities would be cancelled because you’ve
got everything duplicated and the two systems are operating in
opposite directions? Amplification Components
Preamplifiers and amplifiers are expected to have wide band-
Exactly. There are four amplifiers with a bridged pair driving width and low distortion and they certainly should. An amplifica-
each of two voice coils wound on the opposite ends of a single tion component with more than 1% total harmonic distortion is
former that goes all the way through the cone. Each voice coil looked at with suspicion and a product with bandwidth that is
has its own motor system and magnet with the whole thing limited to below 200kHz is often viewed critically. An amplifica-
operating push-pull. tion component with frequency response variations of ±2dB
would be severely criticized.
So everything is push-pull all the way, electrically and mechani-
cally. I’ve read comments like “this product measures like it’s broken”
in reviews of amplification components but never in reviews of
Right. The Model 5A and the Model 5 have four amplifiers in loudspeakers, many of which measure far worse!
each speaker. There are two balanced bridged pairs, with one
pair to drive each voice coil. Even digital switching amplifiers, which exhibit high frequencies
that are audibly soft and “closed down,” typically have band-
The Quatro uses two 8-inch drivers (the same ones that are width to 60kHz or above, which is three times the generally
used in the 2Wq) and two amplifiers that are paired into a fully accepted limit of audibility. No one would recommend that you
balanced bridge to drive them. Using multiple small drivers, buy an amplifier that delivered the midrange frequencies out-of-
we’re not trying to hit as high a peak level and linearity is not phase from the rest of the spectrum and no amplifier that I’ve
as much of a problem. Of course the Quatros are half the seen does this.
money of the 5As.

46 www.audioperfectionist.com
 Engineering Double Standard

Magazine reviewers hold amplification components to strict page after page of accolades yet produced a frequency
standards of objective performance. Contrast these standards response graph that resembled a profile of the Grand Tetons.
for performance with the rave reviews you’ll read about speak- These measurements were described as “enigmatic.” What’s
ers with frequency response variations of ±10dB, high frequen- truly enigmatic is the response of the reviewer. Couldn’t he
cy response that falls like a stone above 10kHz and midrange hear these gross flaws?
drivers wired out-of-phase with the woofer and tweeter.

Some magazines publish these speaker measurements, show- Analog Source Components
ing that the products are poorly engineered but the reviewers Turntables are expected to turn at the correct speed within nar-
row limits and exhibit low levels of wow, flutter and rumble and
these are reasonable standards for good audible performance.
“No one would recommend These factors are much more difficult to measure so you’ll hear
far more about picoseconds of jitter from CD players and sig-
that you buy an amplifier nal-to-noise ratio misrepresented as dynamic range from com-
pact discs.
that delivered the
Vinyl records clearly have greater dynamic range and wider
midrange frequencies bandwidth than compact discs and this fact is easily heard
when the caparison is made using accurate loudspeakers.
out-of-phase from Could the popularity of the compact disc be partly due to fac-
tors other than convenience? When the magazines discarded
the rest of the spectrum.” the notion of high fidelity in loudspeakers did they doom a gen-
eration to a life of musical dissatisfaction?
often love them anyway. Other magazines use these conflicts
to “prove” that measurements don’t mean anything. They’re
wrong because the measurements do mean something—the Speakers
reviewers can’t hear! I think all high-end audio
components should be held
to high standards of objec-
Disc Players tive performance.
Digital disc players are expected to have flat frequency Loudspeakers are an impor-
response and low jitter and these are necessary requisites for tant component in an audio
good performance. Jitter is measured in picoseconds, which system. As this Journal has
are incredibly short periods of time. Do you think these errors explained, speakers—like
will be audible on a loudspeaker that is totally incoherent in the all audio components—
time domain? If the high frequencies arrive at the listener sev- should have flat response
eral milliseconds before the fundamentals will a few picosec- with minimal energy stor-
onds of jitter be detectable? age and they should be
time- and phase-accurate.
I recently read a review of an SACD player that was criticized It’s time to hold speakers to
for frequency response that didn’t exceed 30kHz. In the same the same objective stan-
magazine there was a rave review of a loudspeaker system dards as other audio com-
with a ribbon tweeter that was described as “the star of this ponents. Let’s expose the
show.” It had virtually no measurable output above 12kHz pretenders even if it costs
where a sharp resonance appeared. the magazines some adver-
tising revenue. APJ
Another speaker system with an exorbitant price tag received

www.audioperfectionist.com 47
 APJ
Paula T. Hardesty, Publisher
Richard L. Hardesty, Editor
Edith Hardesty, Copy Editor
Shane Beuttner, Equipment Review Editor
Rick Johnson, Art Director

If you’ve got money to burn, you can probably afford not to

read the Audio Perfectionist Journal.
The Audio Perfectionist Journal is not another magazine.
Magazines are supported by advertising and are written for the support of advertisers. The Journal delivers demonstrable facts along with
candid opinions free from bias. We provide expert advice based on real-world experience. You can’t get this information anywhere else.

Visit our web site at www.audioperfectionist.com to understand the differences. Read about
the Audio Perfectionist Journal and Audio Magazines.

How Much Did Your Last Audio Mistake Cost?

The Journal isn’t biased toward expensive products and big advertisers. We don’t accept
that products costing more perform better. In fact, we frequently find the opposite to be
true. Journal content is produced for the benefit of readers not advertisers. We under-
stand how audio components work.

We want exactly what you want: the best sound for our money.
Subscribe TODAY with our personal guarantee of complete satisfaction.

Audio Perfectionist JournalTM (ISSN 1525-3392) is published quarterly by Auricle Publishing Inc., 17141 Los Robles, Fountain Valley, CA. 92708-2027,
1-714-968-9405, email: info@audioperfectionist.com. Bulk Postage Paid at Houston, TX. and At Additional Mailing Offices. Postmaster: Send address
changes to: Auricle Publishing Inc, 17141 Los Robles, Fountain View, CA. 92708-2027. The subscription fee for Audio Perfectionist JournalTM is $160
for 4 issues. International subscribers please add US $20. Please remit payment in U.S. funds only. Audio Perfectionist JournalTM accepts no advertising
and is published solely for the benefit of its subscribers. Copyright© 2005 by Auricle Publishing Inc. All rights reserved. No part of this publication may
be reproduced in any form or incorporated into any information retrieval system without the written permission of the copyright holder. Please forward all
subscription requests, comments, questions and other inquiries to the above address or contact the publisher at info@audioperfectionist.com.
Opinions expressed in Audio Perfectionist JournalTM are not necessarily those of this publication. Mention of specific products, services or technical
advice does not constitute an endorsement. Readers are advised to exercise extreme caution in handling electronic devices.

17141 Los Robles Circle, Fountain Valley, CA. 92708

www.audioperfectionist.com