| Intersubjectivity and Audion:
Run for Cover Venturing afield to share my preferences for quasi-complementary
topologies with other like-minded folk, I discovered an Australian fellow named
Rod Elliott who offers one of the most concise explanations of what a quasi-complementary
circuit is versus complementary, how a quasi-complementary can be converted to
complementary, and so on. Rod
disagrees with me about dielectric absorption, which I feel is a significant factor
in the performance of audions. He holds that dielectric absorption isn't detectable,
even if it is measurable, and that I should try an ABX test. I have never cared
for ABX or 'blind comparisons.' They are never blind, since I hear a difference,
even slight and don't need to be blinded in order to do so, and moreover, I'm
like flint onto the sound that I hear to be superior. As noted audiophile Gulley
Jimson was heard to remark to a colleague at an Audion League rally held last
century: 'Have you even a grasp what it is like mixing three colours. Three, for
a thousand shades of grey?' 
ABX tests are silly. The necessity for switchboxes, the ordeal of enduring
proper focus, and a slavish devotion to the 'sweet spot' which may or may not
be your own exclusive domain but rather a zone that you have to share with Harry
Pearson;.ugh, I'm running away. I did, however, have occasion to subject myself
to mind-numbing combinations of loudspeakers, amplifiers and, further and with
trepidation, hookup configurations of my multi-unit dynamic driver system. All
of the above was precipitated by a recent acquisition: a pair of Linaeum speakers
from Radio Shack. Mahvelous. End of the line clearance saw the pair in
my hands for a c-note in Canada or, maybe, a few twenties for my American friends.
When the Linaeum came out in the mid 1990's, a combo cost $280 or thereabouts.
A Swap Test My cone driver system
is comprised of a 10" driver in an enclosure size of one cubic foot, in acoustic
suspension, and a second 6" driver in an enclosure size of one third cubic
foot that is in parallel with a 2" driver that acts as a stiff port for the
bass from the 6" driver. A series connected pair, I dubbed it the SCCS (serial
connected cone system). The swap being contemplated in this instance was between
the Linaeum speaker and the small 2-way, unconventional monitor that I consider
to be useful from 80 Hertz up that makes up the midbass-treble portion of the
SCCS. 
In order to make the test a fair one, I made a quick modification to the
Linaeum My first impression of the loudspeaker was that it was somewhat over
the top in the midbass, and that the whole thing doubtless had something to do
with the port that the pretty speaker mouths off with. I popped a sock into Linaeum's
mouth and then extracted it again, to taste. Sock installed, the Linaeum's
voice can be damped to austerity with an attenuation that extends into the lower
midrange. Somewhere in between is the ideal which, while lean, is 'clean.' The
Linaeum speaker is of limited value as a standalone loudspeaker unless you own
a Scottie, and I can't imagine
how it came to be praised by the high end press, even if only in a milquetoast
kind of way in its earliest inception with the wobbly woofer. The Linaeum tweeter,
although interesting, isn't as authentic as the cone driver that I use as a tweeter
in my existing cone reference system, and having listened to the unit for some
time now, I have to believe that what Peter Walker says about treble frequencies
it true: Dipolar radiation doesn't yield as good a result as direct, unidirectional
dispersion. Having said that, I should point out that the Linaeum is far from
being bad news, does some things marvelously well, and praise the designer! The
Linaeum's drivers are aligned. Brilliant! But there is more: I may seem hard on
a modest speaker, but a loudspeaker must have deliverables as a standalone. The
Linaeum works well with a Scottie, but that's because the Scottie is a 3 watt
amplifier that has little power to give at low frequencies. Any extra synergy
works very well with small amplifiers like the Scottie or the Realistic Champ
Amp. The Linaeum's dipole is a must-have. At least once it would be useful
to listen to how a system images through these drivers. I have been surprised
more than once. The Linaeum doesn't sound like an ESL, but then again, imaging
isn't one of the Quad's specialties. Both dipoles, but upside-down, so to speak,
with the Linaeum on top and the Quad having dipolar woofers. That a dipolar
radiation pattern would yield universally superior results seems reasonable, particularly
since the instruments of an orchestra demonstrate an entire range of dispersion
patterns, from the directional horn and woodwind, to the dissipative sounds of
percussive instruments, to the bi-directional radiation of a harp. Dipoles have
a huge advantage over pistons since energy gets released in all directions. But
surprisingly, the ESL has padding behind the treble panels. Some listeners have
their speakers modified to restore the dipole effect to the high frequencies.
I expect that Walker preferred to enclose the rear of the panels to add directionality
to an already dissipated emitter, and to minimize diffraction at high frequencies
behind the loudspeaker. I have listened to two Quad ELS systems that have received
the same treatments by the same craftsman (he worked from an inventory of panels,
matching sets carefully), powered by audion amplifiers and preamplifiers designed
and built one at a time by a much maligned chap named Joe
Rosen, driven by the best ancillary components extant. Both ELS systems sound
superior to my own Quads, which are entirely stock. Ancillary amplification is
also a factor here, as both of the 'mod Quad' systems benefit from Rosen's fuss:
The amplification is of his head and hand, built by aforementioned craftsman.
One system has a one of a two of a kind air bearing platter and tone arm gramophone
built to 'better' specifications with materials further into the sphere of arcana
than few manufacturers delve. For instance, Rosen prefers to smoothen the flow
of air in his own air bearing turntable by utilizing two air pumps with one phase-reversed.
Although this method does not eliminate laminar flow, only a small pressure bladder
is needed to produce an airflow that is completely free of turbulence. Some argue
that an air-bearing system works satisfactorily with one pump alone. Rosen begs
to differ (in ignominy, of course).
| At some point, I decided that a comparison between the
Linaeum and the custom driver jobbie that I use with the bass system was a waste
of time. The two systems announced themselves and altered their tune little, when
placed under similar tests. All of these tests were performed with the sock lightly
but securely stuffed into the port. Out of frustration, I pulled a 7868 based
amp out of its rack, tossed it on the floor, and tried out each speaker as a standalone
and when integrated with the bass driver. I initially used a preamp that supplied
me with a set of passive outputs. Later I switched to a Radio Shack SQ decoder
that I raved about as a TV sound improver in the Audio Ideas Guide. What was
I thinking? However, by then I had formed any impression that I was going to
make vis. the speakers and so with a shudder I pulled the SQ decoder out and for
a while, listened to the CD player through the DCC's D/A converters and first
class line stage, into the 7868 tube amplifier whose outputs I terminated on the
front panel 'Yes we have no Mananas, so get your Mananas today' banana jacks,
effectively shorting out the 8 ohm taps. The amplifier is very quiet so set up,
but buried 8 dB down, there it is. Music. | |
All of my long-running impressions of the sound of my 7868 amplifier
(and I say my because it is so. The 7868 is one of several amplifiers and
pre amplifiers that I have designed and constructed) came to the fore. It is a
depth laden amplifier that is capable of etching detail. But it is bass deficient,
a quality that makes it a useful amplifier with Quad/Gradient since the amplifier's
character is a close equivalent to the Gradient's crossover to the ELS, thereby
allowing me to bypass the crossover on the Quads, using it for the subwoofers
alone. Eminently preferable when audions are used, since the Gradient box is truly
the equivalent of sacrilege, having as its contents the transistors and integrated
circuits that turn the electrons into mutants. Who can say? A bass deficient
amplifier is a bad choice to evaluate bass-deficient loudspeakers such as the
Linaeum. Since most vintage audion amplifiers such as those from Scott, Fisher,
etc., are in a large measure worse than my 7868 amplifier when it comes to bass
performance, audion enthusiasts should consider small driver loudspeakers like
the Linaeum to be a bad match for music that demands good bass control. Both loudspeaker
and amplifier will disappoint, so combined. Stick to low power solid state.
| I did more
tests, using a Scott 260B as the amplifier for the Linaeum and a Scott 367 as
the amplifier for the series configuration cone system (SCCS), and for the dual
driver unit as a standalone speaker. |  |
Since the amplifier stages in the 260B and the 367 are identical, I
decided to try bi-amplifying the SCCS. I was disappointed to discover that there
was no way to integrate the components into any configuration that offered an
improvement. I can think of several reasons, all crossover related, that might
account for this, but in view of the chaotic manner in which the SCCS came about,
I will have to save that story from another day. In a nutshell, the SCCS enclosures,
on their own are useless. The bass driver is out of the game around 200 Hz and
the resonant frequency of the midbass-treble unit is 100 Hz. Together, and series
connected, the drivers work well, the loudspeaker is efficient and it presents
16 ohms to its amplifier, an easy load. The
SCCS is Audio Negentropy The Linaeum integrated well into SCCS with
the alterations described above, with the usual caveats. The sock mod. is important
enough to the proper performance of the loudspeaker. The caveat applies because
there is no way to measure and compare a subjective experience of sound in any
manner than to say 'I like it.' Perhaps I overstate: There is some jargon to describe
sound, but there is no yardstick to measure understanding in subjective audio,
so what are you going to do? Plus, people hear differently. 'I like it' may be
an appealing describer because it is universally understandable and so preferred
because there is no other means of ascertaining how to obtain a tractable result
with only a description that goes something like 'stick a sock in it..'
Mere Confusion Upon further listening
I decided that the sock-equipped Linaeum was likable and sufficiently different
as a separate package in comparison with the SCCS that another cone driver loudspeaker
would be helpful as a object of comparison. I pulled out of storage a pair of
Wharfedale W3's that were first manufactured in the late 1950's. My pair came
out of the factory in May of 1961. Bass reflex with a 12" driver with a separately
mounted 5" midrange (squawker) and 3" tweeter; both open air. This configuration
is very much in evidence in the Linaeum speaker. Surprisingly, considering their
differences in size, the sock-equipped linaeum sounded every bit as lively as
the W3, and there was a deep similarly to the voices of the two designs, separated
as it were, by 40 years and Kevlar. Neither have the taut bass that I enjoy
with the SCCS, both edge the SCCS in the mids, but after extended listening, I'm
not certain which I prefer. I briefly experimented with the Linaem and the W3
in an 'equivalent' configuration as the SCCS with no success. The systems integrated
poorly, but the Linaeum perched on top of the W3 with its piano black finish was
a source of some amusement , as the assemblage looked fetching, and very modern.
Using the W3 as a subwoofer may do the trick, I thought. Grabbing the Gradient
crossover, I hooked the W3s up as a sub, using the sock loaded Linaeum without
a crossover. My first impression was, Gradient. Woe is me. Having said that,
it is noteworthy that the Gradient's deficiencies seemed less apparent with the
cone driver W3 to the cone driver Linaeum. No dipolar dynamic driver mating to
an electrostatic panel here, but such advantage notwithstanding, there is a tenuous
match between the sub frequencies and the merely low. I have never liked the sound
of subwoofer based systems for this reason. Why I believed that Gradient and Company
would do better than anybody else with a difficult transition from sub to ELS
is beyond me. In experiencing the vagaries of the Gradient, I learned quite
a bit about the importance of 'sameness to secure sane-ness' when it comes to
loudspeaker system implementation. It is safest to use identical amplifiers for
the subwoofer and the loudspeaker portions. Phase coherence is an obvious reason.
How can a subwoofer be integrated with a loudspeaker if the speakers are out of
phase? A difficult proposition unless the amplifiers are identical. This, in plain
English, means the same number of phase inversions. Two amplifiers of different
power output can utilize the same design. Such is often the case with the amplifiers
that are comprised of as a series, where the only source of distinction is the
B+ voltage. Mating amplifier to loudspeaker and to subwoofer is a matter of hit
and miss to the following degree (here insert difference in circuit stages:
the one with the fewest is the fastest. One or the other may be inverting but
the fastest becomes the primary wave, oh and try to keep the topologies similar.
Sameness...). All considered, subwoofers are to be eschewed at best, but tolerated
if you went out and bought a set and therefore have them around. Smoke 'em, if
you can't stand 'em. In the case of a Quad ELS, you should adhere to the sameness
rule with some passion. For those who have a single monophonic subwoofer: You
are truly lost in space for having bought into the concept that low frequencies
are non-directional. The foundation of stereo sound extends to the lowest registers,
ahem. So, none of this matters, since you are not trying to create an authentic
stereo field.. Don't worry about speaker placement, nothing matters. Certainly
not the design of the amplifier to be used to power the subwoofer, certainly not
phase, for how can monaural phase be applied in any way to a complex stereo field
that relies on two channels to assemble phase information? All considered, subwoofers
are to be reviled or at least held in contempt for being the fudge caramel of
audio, and hey, the manufacturers don't bother to do it in stereo. A coffee table
that vibrates. |
