The Fifth Element #85 Sealed Boxes

Sidebar: Sealed Boxes

I'm convinced that Villchur's acoustic-suspension "bookshelf" loudspeakers, beginning with the Acoustic Research AR-1 of 1955, were the key innovation that opened up hi-fi to apartment dwellers and to buyers of suburban tract houses who lacked the floor space required by the previous, larger mechanical-suspension loudspeaker designs. That floor-space factor became even more important with the widespread acceptance of stereo after 1956.

There being no free lunch, the tradeoff inherent in acoustic-suspension loudspeakers was and is low efficiency. Getting deep bass out of cabinets much smaller than before required amplifiers much more powerful than before. The march to market domination of the acoustic-suspension loudspeaker advanced in inverse proportion to (and was the proximate cause of) the declining market share of vacuum-tube amplifiers. Solid-state amplifiers could drive acoustic-suspension loudspeakers to room-filling volumes.

However, many early transistor amplifier designs did not sound very musical. The "specification wars"—ie, the marketing of amplifiers by advertising their output-power and harmonic-distortion specifications rather than how they sounded—only led the industry even further astray. Therefore, I am well aware that some thoughtful observers (Art Dudley, perhaps?) might view the entire era of acoustic-suspension speakers as the wrong road taken.—John Marks


Catch22's picture

Well done my man!

remlab's picture

Very enjoyable! And after 30 years, I finally know how to pronounce "Kloss". Hah!

Mikeymort's picture

I loved this article. When I served in the Army during the middle seventies, the barracks where I lived was filled with post exchange acquired gear. Hi wattage receivers and acoustic suspension speakers. I had large Advents and a Marantz receiver. Luckily, one of my roommates had a tubed Dynaco system with A-50 speakers. (What is aperiodic anyway?) We could listen to his system hour after hour with no listening fatigue. I have a similar system today.

John Marks's picture

And so by implication does not have a primary resonant frequency.

Specifically, an "aperiodic" loudspeaker enclosure design can be looked at either as a ported enclosure that has resistance in the port, or, a sealed box that leaks.

While the Dynaco Variovent is the most familiar application, I believe that stuffing the port in the ProAc Tablette with cocktail stirrer straws accomplished the same thing.

Thanks for the kind words. It was fun to ferret out all those old stories. BTW, I asked Winslow Burhoe, who was Edgar Villchur's research assistant on the AR4, what it was like to have Villchur as a boss, and he impishly replied, "I don't know. He was never there."

A priceless moment.


tmsorosk's picture

A very entertaining and informative article.

Well done Mark.

John Marks's picture

I'm John.

corrective_unconscious's picture

It's a coarses for hoarses thing.

ganyc's picture

Hi John,

Nice review. In this case do you think that "fullness" and "presence" equate to warmth? My only experience with ATC speakers was infrequent appearances at RMAF. I don't think I would have described what I heard as warm.



hnipen's picture

Thanks for a great history lesson John :-)

Timbo in Oz's picture

system as a resonant circuit. No?

It was able to help us build good sealed boxes, too. It is possible to build boomy/heavy bass sealed speakers.

JBTWay the KEF b110 has been used in Rb boxes, in the bass-extension stand/tower for the LS3/5A. But even that didn't go particularly deep.

It is possible to get similar results to a sealed enclosure with room-gain, (still with some cone-bounce issues,) by using a critically damped QB3 alignment or even an over-damped alignment.

IME the best kind of Rb enclosure is an 'active' or 'assisted' 5th or 6th-order alignment using an active 2nd order HP filter in the line to the power amp with a little gain. Boost |\__ This almost eliminates the distortion that is visible as cone-bounce, while also limiting motor excursion, and can provide another half octave of extension.

This idea could also help with TL designs which are damned close to Rb's when modelled.

Timothy Bailey