[sdiy] Simple discrete Unity-Gain Follower ?

[Magnus Danielson]

From: "jhaible" <jhaible at debitel.net>
Subject: Re: [sdiy] Simple discrete Unity-Gain Follower ?
Date: Tue, 6 May 2003 09:42:03 +0200

> > This may not be true of new designs with feedback accelerometers on the
> > cones.
> 
> Another servo !!!

My thought exactly.

... and then I didn't point out the builtin servo-loops in the physical systems
of the gain elements (transistors and tubes). We draw funny lines in funny
charts with non-linear curves to find out the operational spot, and all we do
is predicting where the system's servo-system will stabilize itself.

Also, the point about speaker elements having distorsion is indeed true, but I
think you have to reevaluate the situation for each case and there are methods
to keep distorsion within bounds even with speakers.

> JH. (playing the devil's advocate.)

And succseeding. You have missed to point out the inherent distorsion in the
not-so-linear medium of air. Especially in compressor horns etc.

However, you *can* acheive both high levels and low distorsion levels with not
so strange equipment. The system I did much work on where so clean that people
had to dampen out their default equalizing. A fun trick we did was to ask them
to hit the bypass switch on the house graphic equalizer, which always got
people gasping since they where not prepared that their changes would make such
a big impact. Lesson learned: clean up the system and you changes make better
effect. You can do *alot* with just linear thinking which have far greater
impact than people think. Just the linear signal properties of systems is much
more a minefield than people think. Here system-thinking rules, and properties
of specific devices is really meaningless unless you see the bigger picture.
This means that amplifiers, speakers, active filtering, compensations, limiters
etc. all have to work in unison as a system in order to get the optimum
performance. Let's just say IMD and TIM didn't even enter the discussions,
since they in practice where small-potato problems to what a real system design
needs to fight anyway.

Cheers,
Magnus - thinks crystal clear for large audiences is much more of a challenge