[sdiy] "Time Winding" in Audio Cables ???

[Magnus Danielson]

From: Richard Wentk <richard at skydancer.com>
Subject: Re: [sdiy] "Time Winding" in Audio Cables ???
Date: Tue, 12 Jul 2005 01:23:08 +0100
Message-ID: <5.1.0.14.2.20050712010033.042ac668 at mail.skydancer.com>

> At 20:22 11/07/2005, Magnus Danielson wrote:
> 
> >You loose alot of dynamics that way. You need larger amplitude for the same
> >effect etc. If you would run two-way or three-way active crossover with
> >suitable amplifiers you would have less total amplifier effect for the same
> >overall soundlevel than a single amp with passive filter. There are many
> >benefits for going this path, at the cost of spending money on several amps
> >rather than a single amp.
> 
> And... triamping will always sound better for other reasons too.

Indeed. There are many benefits with it, dynamics and all.

> > > But it's interesting that there are so few active designs around, with -
> > > say - the speaker/filter in a feedback loop to try to minimise distortions.
> >
> >????
> >
> >An active filter sits before the amp, dealing with signal-level signals.
> 
> I didn't mean that kind of active. The Philips idea was based on some kind 
> of servo control concept. Now, this was 30-odd years ago so I'm not clear 
> on the details, but it was either based on fairly simple feedback (more 
> likely...) or some kind of active measurement of cone position (less likely).

It is vaugly comming back to me now.

> In fact the latter couldn't possibly work because cones are never perfectly 
> stiff and you'd have to work out some kind of monster transfer function 
> that mapped cone position to actual pressure dynamics under different 
> loading and transient conditions.

There have been systems available where cone positioning have been used in
feedback, however I beleive that it was only for the bas-speaker part of
things, separated and all.

> You could probably do this today with some hefty DSP power, but it's - at 
> the very least, I'd guess - a three year PhD project, just to characterise 
> the details of the problem, never mind solve it.

Actually, DSPs is already doing very neat things to some speaker systems today,
a few PhD projects later. ;O)

> >Actually, I've designed two-way and three-way systems with ampracks well
> >separated from the speakers. Cables included! ;O)
> 
> I'd have been a lot more impressed if you'd made a system that eliminated 
> cables altogether. Now *that* would be cool. :)

Well, call me old-fashioned then! ;O)

Cheers,
Magnus