On Sun, 09 Oct 2005 11:34:26 -0700, hoarse with no name wrote:

If programs implementing the T-S laws are so good and allow such a good
understanding of how the enclosure and related issues affect sound, then
should not the determining factor differentiating speaker designs be the
driver design itself?

The equations dealing with the T-S parameters do nothing more than deal
with the way the fundamental spring/mass resonance of the driver interacts
with the enclosure to form a high pass filter. They have no relationship to
the sound of the drivers outside this region. All of that is still a matter
of manufacturers making good, flat drivers.

d

hoarse with no name wrote:
If programs implementing the T-S laws are so good and
allow such a good understanding of how the enclosure
and related issues affect sound, then should not the
determining factor differentiating speaker designs be
the driver design itself?

A couple of points.

First, the Thiele-Small MODEL (it's not a set of laws,
it's a model) is accurate within the domain in which the
model is defined. And that, you will find from reading
every article authored by the two gentlemen on the topic,
has specific limits imposed on it. Relevant to your question
is that the model is defined for operation within what's
called the "piston" band of the driver's operation. By
"piston band" is meant that the wavelengths of the radiated
sound is substantially long enough relative to the driver
radiating diameter that the radiation impedance of the
driver is behaving in a specific and very predictable
fashion. And, for the most part, that means low frequency
behavior only.

For example, the piston band for a typical 8" driver is
considered to go no higher than about 600 Hz. (A reasonable
approximation for the upper limit of the piston band of a
circular driver is the speed of sound divided by the
circumference of the driver. Since an 8" driver has an
effect radiating diameter about 7", that means the
circumference of the driver is pi * 7", 21" or about 1.83
feet. Speed of sound at STP is 1129 ft/sec so 1129 ft/sec
divided by 1.83 ft gives you about 617 Hz). Above that
region, other factors begin to dominate, and the Thiele-
Small model no longer accurately describes behavior.

Another limit of the model is that it assumes time-invariant,
linear behavior. Drivers, at least the ones we can put our
hands on, don't meet these requirements to a greater or lesser
extent.

A second point: the Thiele-Small model is not about drivers,
it's about SYSTEMS. The useful feature, especially of Small's
work, is that it casts the system analysis in terms of the
integrated behavior of the driver electrical, mechanical and
acoustical elements and the enclosure's mechanical and acoustical
elements and, what makes it both simplifying and powerful, is
that it relates them all in parameters of compatible units.
One could, for example, look at the electrical resistance in
MECHANICAL units, just as easily as one can view mechanical
friction in electrical units. This makes the analysis and
synthesis of systems amenable to extant tools (for example,
electrical circuit mesh analyzers such as SPICE).

For the most part, the most powerful determining factor of a commercial
speaker design is economics and marketing.

For the end buyer, a lot depends on the intended application of the speaker
and personal taste.

Do you like trumpets, violins or clarinets? They all can play roughly in the
same range.

James. )


"hoarse with no name" wrote in message
...

If programs implementing the T-S laws are so good and allow such a good
understanding of how the enclosure and related issues affect sound, then
should not the determining factor differentiating speaker designs be the
driver design itself?

On Sun, 9 Oct 2005 20:00:04 +0100, Don Pearce
wrote:

On Sun, 09 Oct 2005 11:34:26 -0700, hoarse with no name wrote:

If programs implementing the T-S laws are so good and allow such a good
understanding of how the enclosure and related issues affect sound, then
should not the determining factor differentiating speaker designs be the
driver design itself?

The equations dealing with the T-S parameters do nothing more than deal
with the way the fundamental spring/mass resonance of the driver interacts
with the enclosure to form a high pass filter. They have no relationship to
the sound of the drivers outside this region. All of that is still a matter
of manufacturers making good, flat drivers.

The T-S parameters can tell you the required volume of the enclosure
and what shape should be the port, but they tell you nothing about the
internal dimensions, shape or resonant qualities of the enclsure, all
of which are just as important as the drivers.
--

Stewart Pinkerton | Music is Art - Audio is Engineering