[[email protected]]

On Jun 4, 1:32*pm, "Richard Crowley" wrote:
"Harlan Messinger" wrote ...

Don Pearce wrote:
It is not a case of low frequencies being targeted, but really that
dealing with high frequencies is much more technically challenging. If
you really want good HF isolation, forget noise cancelling and buy
some passive ear defender types.

Really, it's more challenging? I would have thought one approach would be
applicable across the spectrum.

The fundamental theory is the same. But reconstructing
a cancellation signal at higher frequencies requires
more processing horespower (i.e. faster processors).

More "horsepower?" In the limiting case, all the
horsepower that's needed is inverting the phase
of the signal.

And at shorter wavelengths it gets trickier
to deliver the exact cancellation waveform
*at your eardrum* from several mm away.

THAT'S the crux of the problem: the fact that
the microphone used to detect the original
noise signal and the transducer used to produce
the cancelling signal can not physically occupy
the same point. Further, the REAL point where
you want the cancellation to occur is in the ear
canal, where it is, at best, very inconvenient to
place either.

It works well at low frequencies because the
wavelengths are large (at 100 Hz, they're 11 feet
long), thus the difference in sound pressure between
the microphone and cancelling speaker (say they're
two inches apart) is small and the phase difference
is also small (on the order of about 6 degrees).

At 10 kHz, those wavelengths are on the order of
1.4 inches of an inches, substantially larger than
our hypothetical 2" separation.