crystal clear : was : RE: [sdiy] Simple discrete Unity-GainFollower ?

[harrybissell]

I add a few notes about Magnus' assertions...

I use a BIG amplifier for my speakers... 300W (rms) per channel...which drives
one 15", one midrange horn, and two piezo tweeters.

In this case my driving impedance is quite low...

I spent one month of design and experiment to build the crossovers.  Half of this was
spice design simulation... the other half figuring out why the simulated design did not sound
good.

I did not attempt to do 'high fidelity' design because this system is used exclusively with
"Harry's little band"... I used DAT tapes of my band and tuned the crossovers until it sounded
like my band...  There is a sharp 6dB resonsnt peak at 10KHz... NOT hi-fi at all.

So minimum, for an active system I'd need at least two amplifiers, maybe four. Crossover
networks etc.

So I'm using passive, but I have power to burn and don't care about hi-fi anyway...

Magnus, I'm sorry you don't have the money to send me a sample system. I hope you get
a big raise soon and you'll be able to afford it.  If it would help you can ship freight collect...

(only in YOUR dreams....   ;^)

H^) harry

Magnus Danielson wrote:

> From: "Czech Martin" <Martin.Czech at Micronas.com>
> Subject: RE: crystal clear : was : RE: [sdiy] Simple discrete Unity-Gain Follower ?
> Date: Wed, 7 May 2003 16:23:39 +0200
>
> > >just a messy legacy which is best forgotten. Passive crossovers is just where
> > >you loose out before even turning the rig on. I don't like it in any speakers
> > >actually. Active cross-overs and separate amplifiers is the natural choice here
> > >to get any form of control.
> >
> > What about passive crossovers in the higher frequency range, above 1kHz or so?
> > I think the component values get smaller there, so it's easier to have inductance
> > and capacitance without too much side effects? Most tweeters have a lot of damping built in. And a passive highpass will automatically protect tweeters from DC...
>
> No, you really want to avoid that. You want the time compensation anyway, and
> doing that in passive filters for large scale signals? No thanks. It all goes
> into active filters where it is easy to make well matched and stable
> electronics. Also, I want individual limiters and tossing such in as passive
> components in effective means a non-linear device in there... naaa... active
> cross-over filters it is.
>
> Some may say that this is only large-system design rules, could you then
> explain why my rather small Genelec speakers follow the same principles and do
> a pretty good job (only good for home-sessions for Harrys little band).
>
> Actually, I think you fool yourself on the simplicity of a common amplifier and
> passive cross-over filters. You need a much bigger amplifier than if you had
> two smaller amplifiers. The passive filter naturally needs to use components
> that take much more power than an active counterpart. It is easier to make a
> more complex filter in active electronics. Once you accepted the separate
> amplifiers (which I think is the objection people usually have) the rest will
> follow as really natural.
>
> You gain alot by moving away from passive filters to active. It's really a day
> and night difference at times (like when the Turbosound guys didn't recognice
> the sound of their own speakers, with just a very rudimentary active filter
> setup).
>
> So, no, I pass on passive filters. They are so crude approximations and is not
> as easy to finetune and adapt as needed for a good summing. No, I have
> basically just put them on memory lane and see no real motivation for their use
> anywhere. It may seem like a harsh point of view, but I think people
> overestimates the useability of passives and the cost of active filter
> solutions.
>
> Cheers,
> Magnus