[[email protected]]

On Wednesday, October 16, 2019 at 12:43:04 PM UTC-4, Peter Wieck wrote:
a) Argument-by-exception is, generally, fallacious.

Argument-by-exceptionally-rare-exception: more fallacious? :-)

b) Anything including a coil that carries current will be inductive.

Not necessarily true, especially over the bandwidth of interest.
I have, for example, measured transformers intended for wide-
band audio use (Jensen comes to mind) that, when e secondary is
loaded with it's intended resistive load, exhibits completely
resistive impedance with NO sign of an indictive component well
beyond the audio bandwith.

Further, even considering the lowly typical "high-end tweeter",
as one sweeps upward in frequency, one observes first primarily
resistive impedance, then resistive-inductive, then resistive,
then resistive-capactive, then resistive and then, finally, as
one approaches the high-frequency cutoff, resistive-inductive. And
even at that, the impedance is dominated by, in the vast majority
of cases (both tweeters and frequencies) the resistive component
of the impedance.

But, really, to be technically accurate, one should say that when
current is being asked to move across a subtended area, it will
exhibit inductive behavior. It has nothing, per se, to do with
"coils": the effective inductance of the prior circular particle
accelerators was a rather thorny problem. It's not as much of an
issue for the LHC, simple because you really have two "coils"
occupying the same physical space (to a reasonable approximation),
running 180 degrees out of phase, thus cancelling the inductance.

c) Most well-designed speakers using conventional drivers that include voice-coils will account for this in their design.

I might be inclined to emphasize "most" in this context". The two
exception noted were, in one case, a proprietary driver not
widely available, the second being a very interesting driver
that never achieved real production and distribution.

d) Many crossover designs include inductors of various natures typed.

Yes, but they, in and of themselves, do not necessarily
contribute to the effective inductive behavior of
the impedance curve.

This last point is THE crucial one: regardless of what's under
the hood, it's the resulting impedance curve that's the issue
at hand. One can have parts that are inductive (or even inductors)
in a circuit) that, overall, does NOT exhibit an inductive
component to the impedance, and one can have a circuit that has
NO inductors whatsoever whose impedance looks all the world
like an inductor (the classic gyrator is one example).

e) And those well-designed speakers that incorporate the exceptions
will also account for those option.

Yes, but the details of such may well be irrelevant in
the current scheme of things.

Comes down to the question of: Does driver/speaker inductance
in *this* particular speaker coupled with *that* particular
amplifier matter at *this* range of frequencies and volumes?

Theory is all well and good, but how things operate in the real
world at the living/listening room level are, or at least should,
be the primary issue.

Entirely agreed, and I would only emphasize the point by saying
that it is certainly possible for one to find a exceptionally
rare combination of things that support a particular thesis:
such exceptions, contrary to the popular idiom, do NOT prove
the rule: they simply demonstrate the ability to concoct a
completely pathological exception.

In following this thread, I did review Trevor's list of
"pathological" (my use of the term) loudspeaker impedance
curves to be found at Stereophile, and my overall reaction is
that I find the situation they (these manufacturers) enumerate
to be disturbingly irresponsible. With few exception, when I
have been asked to consult on projects which exhibited these
sorts of pathological; impedance curves, almost without
exception, are the result of design imcompetence. There are
ways of designing passive crossovers with the same electro-acoustic
transfer functions that DO NOT have these same gross impedance
anomalies. This is especially true of three-way, multi-order
passive parallel ladder-type networks (which are the vast majority
of such found in such multi-way systems).

Designing a system that exhibits the kinds of impedance
properties that lead to the sorts of issues discussed here
is highly irresponsible and yet another sign of the technically
insular nature of the high-end world.

[Peter Wieck[_2_]]

Mpffffff....

Now, we need to get into the difference between "accuracy" and "precision" - which for the record, are not even a little bit the same thing.

Accuracy: A digital thermometer calibrated in two (2) degree Celsius increments, that is always with two degrees of the actual temperature is accurate - but not necessarily precise.

Precision: A digital thermometer that is calibrated to six (6) decimal places of a degree, but is randomly somewhere between 4 and 8 degrees off the actual temperature is very precise. Not hardly accurate.

Point being that good speaker design covers inductance, whether intentional or accidental (peculiar to the nature of the elements in use).

DO, PLEASE look up the original definition of Peculiar.

Peter Wieck
Melrose Park, PA