Nousaine wrote:

But what happen to the old concept that bass is omnidirectionnal? Or
more exactly, frequencies whose wavelenght are over three time the
diameter of the reproducing driver are non directionnal?

They still hold true. The main difference is that in your home 80 Hz is 2
standard deviations below the frequency at which people can generally
locate a subwooer. In your car the people at Ford Audio found that 150 Hz
(an octave up) was the typical frequency where low frequencies began to
become directional.

So, a typical in-car driver reproducing frequencies below 80 Hz in a
linear fashion *is* omni-directional and humans cannot locate it either.

It would seem to render moot any sub "aiming" concept...

No "near-field" has a specific meaning that in engineering terms means
that the microphone is within a half inch of the diaphram at low
frequencies and sees basically the anechoic response of the speaker.

I stand corrected. I was under the impression that multi-way systems
(at least home speakers) were measured at 1 meter so to acount for the
inter-driver interaction.

At the critical distance where the reflected sound is equal in intensity
to the direct sound we enter the far-field. In your car the radiating
surfaces are so close to the interior surfaces the listener is always in
the far field. (the amount of reflected sound is equal to or greater than
direct sound.)

OK

Because there is no 'direct field' in the car and the space is much
smaller. In a similar way the acoustics of the living room differ from a
concert hall.

But the effect on system tuning is that many people become irritated with
the equalization process because the try to make the system "flat" in the
far field without taking the slope into account and find it doesn't sound
good.

Indeed. Any attemps to have 16 kHz at the mid range level proved to be
ear tearing!

But in a car using multiple drivers (ie sub in back, low-mid in doors
and tweeter on dash, how youd you approach frequency response measuring?
Adjust each driver individually to be flat in near-field or measure at
the listening position and try for the 3 dB per octave slope?

-- Eric (Dero) Desrochers

Hiroshima 45, Tchernobyl 86, Windows 95