[rollei_list] Re: Completely OT- Loudspeaker info
- From: "Richard Knoppow" <dickburk@xxxxxxxxxxxxx>
- To: <rollei_list@xxxxxxxxxxxxx>
- Date: Tue, 15 Feb 2005 18:45:57 -0800
----- Original Message -----
From: "Jerry Lehrer" <jerryleh@xxxxxxxxxxx>
To: <rollei_list@xxxxxxxxxxxxx>
Sent: Tuesday, February 15, 2005 5:39 PM
Subject: [rollei_list] Re: Completely OT- Loudspeaker info
> Eric,
>
> Most crossovers that I know of can only modify phase 180
> degrees.
>
> Jerry
>
> Eric Goldstein wrote:
>
It is possible to introduce delay, which in small amounts
is the same as phase shift, into a crossover network by
using all-pass sections. These are pretty limited when made
of passive components but active ones can be made in the
analogue domain. They tend to be quite complex. Digital
alows much easier achievement of delay networks.
Conventional filters of quite complex nature (lots of poles)
have had wide use in the telephone network in the past. One
problem is that the more complex the filter the less stable
it becomes. Also, simple all pass networks are limited in
frequency range. There is a huge literature on this subject,
mostly couched in mathematical terms which are thoroughly
beyond me.
One aspect of complex filters which is often overlooked
is the transient response. Any network which is capable of
storing energy is capable of distorting transients. This is
a major issue in high quality audio networks of all sorts.
Those who remember what radio and TV network sound was
like, at long distances from the originating point, before
satellites will have a notion of what severe time delay
skewing sounds like. Because AT&T research was on speech
range chanels and not wide range audio, they came to the
conclusion that this kind of distortion was not audible. In
the telephone network it did not affect naturalness or
articulation but in high quality broadcast circuits of any
length it was obvious and sounded awful.
---
Richard Knoppow
Los Angeles, CA, USA
dickburk@xxxxxxxxxxxxx
Other related posts:
