Mass Confusion

[weinstro]

In reading through this forum, I've managed to confuse myself.

Some folks advocate wide dispersion, sometimes through small drivers and baffles, other times through dipole. Others advocate controlled directivity.

So before I run a bunch of wood through the saw, can someone help clarify these for me?

Maybe, let's start with what these terms mean, and then I'd like to try to understand what the relative importance is.

Many thanks!

[badman]

Start with www.linkwitzlab.com, and once you're through that, you can check out http://www.musicanddesign.com/

If you read through their stuff, you'll have enough of a foundation to delve into more specifics. What you've asked is a complex enough subject (dispersion) to fill a book, so you're not likely to get a lot of answers.

[Cal Weldon]

Off topic post removed.

[Drew Eckhardt]

Originally posted by weinstro
[B]In reading through this forum, I've managed to confuse myself.

Some folks advocate wide dispersion, sometimes through small drivers and baffles, other times through dipole. Others advocate controlled directivity.

So before I run a bunch of wood through the saw, can someone help clarify these for me?

Speakers with power response that's uniform with frequency (trending down at high frequencies) sound more natural than ones with less uniform power response. Speakers with off-axis response lacking aberations not present on-axis sound more natural in the presence of nearby objects than ones which don't.

The two ways to get there are to make sure there's nothing reducing off-axis output (omni, with small drivers and low cross-over points so you don't have a big dip in total power at the cross-over point or suffer from beaming as you move higher into the lower frequency drivers' pass bands) or to uniformly reduce it through acoustic cancelation (dipoles) or a wave guide.

More directivity means less sensitivity to nearby objects, better clarity farther in the room, and perhaps a very wide imaging sweet spot (three seats on a couch) because you can aim the speakers so you receive less direct sound from the nearer of the stereo pair which compensates for its sound ariving sooner.

That comes at the expense of more drivers with higher displacement to reach a given SPL (dipoles dump a lot of energy into the acoustic short circuit, with an Orion needing 4X the displacement as a conventional speaker to reach the same SPL at 30Hz) or getting decent waveguides which are harder to make than boxes and can cost as much as drivers to buy.

I own Orions and Plutos. I've heard other dipoles and omni designs like the RAALs. They're more similar than different but the speakers with more directivity work better where walls are nearby and seating distances farther. Some time I'll try a modern waveguide.

[gedlee]

Originally posted by Drew Eckhardt
More directivity means ...

That comes at the expense of more drivers with higher displacement to reach a given SPL

Did I read this wrong? Are you saying that more directivity means more "expense of more drivers with higher displacement". One of the things that I like about waveguides is that they allow fewer drivers and fewer crossovers. I usae two drivers and one crossover. And even though the waveguide can be expensive, the amount that you save on amplifier number and power along with the electronic crossover would more than make up for the waveguide cost.

In the end I find a waveguide system to be far lower cost than the other approaches when all extraneous factors are taken into consideration. For example all my amps and electronic processing cost me $150. Thats a big savings over several seperate power amps and electronic crossovers.

[Drew Eckhardt]

Originally posted by gedlee



Did I read this wrong? Are you saying that more directivity means more "expense of more drivers with higher displacement".

That's the dipole approach, with the "or getting decent waveguides which are harder to make than boxes and can cost as much as drivers to buy" alternative.

Obviously "more directivity" means a directivity index at most 4.8dB better than a conventional speaker.

One of the things that I like about waveguides is that they allow fewer drivers and fewer crossovers. I usae two drivers and one crossover. And even though the waveguide can be expensive, the amount that you save on amplifier number and power along with the electronic crossover would more than make up for the waveguide cost.

Depends on whether you limit your DIY activity to the system as a whole or just the speaker enclosure and what the bandwidth+output level requirements are. An extra channel of power op-amp runs about $15 sharing a common power supply. A pair of three terminal regulators, resistors, and small bulk capacitors drop the power amp rails to usable levels for op-amps with each filter pole running $2-$3 with 1-2% components. More amplifier channels and an active cross-over can be less expensive than the passive solution. For my bedroom system I don't need more than 90dB peak so a 10dB efficiency gain isn't too relevant. With a 120Hz cross-over and peak output not much past 100dB a single 8" mid-bass does fine in a dipole.

A second driver low-passed at 6dB/octave or line-level equalization (especially if you can tolerate some insertion loss) would be alternatives for people who have a strong preference for building with sawdust over silicon.

In the end I find a waveguide system to be far lower cost than the other approaches when all extraneous factors are taken into consideration. For example all my amps and electronic processing cost me $150. Thats a big savings over several seperate power amps and electronic crossovers.

As a guy who doesn't build his own wave guides, I'd be likely to order a pair from you at $300 with foam plugs and as a non-believer in audiophile hype wouldn't hesitate to use an inexpensive receiver. That would get me to the same ~$450 I spent to build my last set of active cross-overs plus amplifiers including commercially produced circuit boards and license to build the Pluto design for personal use. An original design on perf board would have been about $150 cheaper although on the waveguide side I could learn wood turning or generate CAD files and farm the wave guides out to a wood shop with an idle CNC router.

[gedlee]

The more you do yourself the cheaper it is, but the simpler the starting design the lower the cost everything is going to be. It all depends on how much you want to do. Is the goal to DIY or to get a great system and listen to it? There are extremes either way.