Powered 3-Way Dipole Loudspeaker The Final Mod: Orion 3.3.1SN

My Love Affair with the Orion

It takes a special jolt these days to interrupt my geriatric hibernation and prompt
me to post something new on this website. Audible improvements of the Linkwitz
Lab “Orion” loudspeaker are among the rare stimuli that can still do it.

My obsession with the Orion over the years is due not so much to its specific
characteristics but rather to the generic concept it represents. I can best explain that
with a thought experiment. Pretend that you are brought blindfolded into a room
where you are going to listen to an unknown pair of loudspeakers fed by first-rate
electronics. You sit down and focus on the sound. It is obvious that the speakers
are very large because the soundstage is very wide and very tall. It is an
unprecedentedly open and transparent sound, leading you to believe that new and
unusually sophisticated technologies are being used (floor-to-ceiling ribbons made
of stainless kryptonite?). The localization of instruments is extremely precise—is
there a center-channel speaker they sneaked in there? At last the blindfold comes
off and all you see is a pair of Orions, not at all large and with utterly conventional
drivers. Please explain the magic.

And that’s my point—it’s the concept. No box, just a frame. A dipole, almost
totally symmetrical, front and back. Acoustically small (because, if the size of a
driver approaches a large fraction of the wavelength it must reproduce, all bets are
off). Electronic crossover/processor, with radical equalization of each driver for
optimum linearity. I could go on—add up all the details and it turns out that the
design indeed represents a new generic class and is highly sophisticated, regardless
of its innocent appearance. That’s what gets me.

Now Siegfried Linkwitz has come up with the somewhat similar but still rather
different LX521, of which he owns the so far only extant prototype and which he
claims sounds better than the Orion. Well, I’m perfectly willing to believe it
sounds better than his Orion, but what about my Orion? That’s a different story and
the reason for this review.

The Don Barringer Connection
 

Don Barringer has been Siegfried Linkwitz’s associate and “second pair of ears”
since the late ’70s. He is a former trumpet player, a cutting-edge recording
engineer, and a no-compromise audio fanatic. When SL finalizes one of his
amazing designs and declares victory (which he doesn’t do lightly), Don says wait
a minute, we aren’t done yet, it still needs such and such. SL believes that there
inevitably comes a time “to shoot the designer,” at which point Don cries “don’t
shoot!” Not long ago, their formal collaboration finally came to an end, although
their relationship remains cordial and communicative.

Don was never entirely happy with the circuit values in the incredibly complex
analog filters SL came up with to linearize the Orion’s drivers and overall
response. He did not think the Orion sounded sufficiently neutral, even in its latest
version (3.3.1) and quite aside from its other startlingly superior qualities. When SL revisited
the equalization of the Orion a couple of years ago, he started out with new
measurements of the response of each driver and attempted to flatten out each by
adjusting the circuit values of their respective filters on the printed circuit board of
the crossover/processor. He did not succeed in effecting a significant improvement.
He then instructed his computer to ignore the separate filters and just come up with
an overall circuit that would yield the correct response across the entire audio
range, also including the required HF and LF shelving. That attempt was
successful, at least to the extent possible with standard resistor and capacitor values
of  ±2% tolerance. That’s where Don wasn’t quite happy yet.

Don believed that further fine-tuning of those values could result in even greater
neutrality, which was his only remaining concern regarding the Orion. The
problem was that SL’s incredibly convoluted catchall equalization characteristic
proved to be extremely ticklish to fine-tune. You pushed it in just a tiny bit here,
and it bulged a tiny bit over there. It overreacted all over the place to local
stimulation. It took Don over a year and a half to figure it out, during which he
substituted resistors and capacitors with the minutest changes in value and listened,
over and over again. The evaluation had to be strictly subjective because the
changes he made were much too small to be measurable with a microphone but
still marginally audible. (You can hear a change of, say, 0.2 dB in the electronic
signal path but you can’t reliably chart it on an acoustical response curve.) It was a
desperately laborious process, not unlike picking the fly shit out of the pepper
(1940s GI metaphor). After 20 months (during which, he says, he considered
committing himself), he finally declared victory. He recently posted the results and
DIY instructions on the Orion-Pluto Users Group (where you need to register
before being able to access restricted information).

The remarkable thing is that, with one exception, all the nonstandard and ±0%
tolerance circuit values he came up with were within SL’s ±2% specifications. It is
even imaginable, theoretically, that a random statistical freak would have Don’s
final values in an unmodified Orion 3.3.1! Of course, the bottom-line question is:
does a Don-modified Orion 3.3.1 (he calls it 3.3.1SN, for “subjective neutrality”)
sound different from a plain-vanilla SL-approved 3.3.1? There is only a very small
circle of Don’s followers who had the new resistor and capacitor values installed; I
am one of them; and our answer is a resounding yes!

The Sound of the 3.3.1SN

I have always maintained that the big sonic breakthrough was the original Orion of
a decade ago, because it “blew away,” to use the audiophiles’ favorite expression,
all traditional loudspeakers in sealed or ported enclosures (monkey coffins, in
1970s trade parlance). Subsequent versions manifested incremental improvements,
large and small, but no overall change in gestalt. The same is true of the 3.3.1SN; it
sounds definitely better than the unmodified 3.3.1, but exactly how much better
depends on the importance of sonic nuances to the listener. The best way to
describe the change is to say that the realism is finally complete; the rare reminders
that one is listening to a mere loudspeaker are gone; no more momentary
aggression here or thickness there, just soaring music. I am sufficiently impressed
to have started reassessing all of my favorite recordings. If you’ll allow me a
somewhat farfetched classical analogy, Don Barringer has become the Plato to all
of us faithful disciples of Siegfried Linkwitz’s Socrates.
 

It must be added that all of the above takes on less importance if SL’s new LX521
(also strictly a DIY project) turns out to be as good as he says. As a rule, he is not
in the habit of just whistling Dixie.