[Duke]

For some time I've been experimenting with different techniques that would hopefully give good sound in a fairly small room.  That's an area where I would like to offer something competitive, but so far all of my systems have been more suitable for medium or large rooms (though some people are using the Jazz Modules in fairly small rooms).

Here are the problems inherent in trying to design a speaker to work well in a small room:

1.  Limited real estate available, especially side-to-side.  But not only does a speaker physically take up floor space, it also effectively occupies the space behind it. 

The solution is to use a narrow speaker, preferably one that can be placed close to the wall.

2.  Early onset of reflections.  Because the path lengths to the walls are shorter, reflections begin to arrive sooner than they would in a larger room.  Early-arrival reflections are more likely to be detrimental from an imaging and coloration standpoint than are later-arriving reflections. 

The solution (from a loudspeaker design standpoint) is to have good radiation pattern control as far down into the midrange region as is practical, and aiming this radiation pattern so as to maximize as much as possible the time delay before the onset of reflections.  Note that these characteristics and the narrow speaker requirement are mutually exclusive, so some compromise is inevitable. 

3.  Uneven bass response.  The room-interaction peak-and-dip pattern is usually worse in a small room than in a large room, and contributes to many people not wanting a speaker with significant bass extension in a small room.

The solution is to spread around the bass sources as much as is practical, preferably in all three dimensions.  In my opinion the ideal would be a Geddes-style or Swarm-style distributed multisub system, but if floorspace is at a premium that's out of the question.

4.  Excess room gain at low frequencies.  The proximity of the room boundaries results in more boundary reinforcement than speakers normally get, especially if they have to be placed close to the wall.  So a speaker designed to be approximately flat in the bass region in a medium or large room will have excess bass in a small one, and if that excess bass is also uneven it can be quite unpleasant.

This is really a free lunch in disguise, as the solution is to have a gentle bass rolloff that starts fairly high up.  Because we don't need to generate as much bass energy, we can make the speaker a bit more efficient for a given size.  However if the bass is going to be lumpy anyway, we want to err on the side of having too little bass instead of having too much. 

The amount of room gain varies significantly from one room to another, and also with changes in speaker (or listener) location within the room.   So some adjustability of the speaker's bass response is desirable. 

5.  Poor driver integration at close listening distances.  This isn't an issue in most cases, but at very close listening distances it can be.  The ear is poor at resolving the height of a sound source up to about 1 kHz, and then it improves significantly.   So if there's a large vertical spacing combined with a fairly high crossover frequency, it could cause problems.

Several months ago I was in the home of a Jazz Module owner and he was listening from close enough to the speakers that if he sat forward in his chair he could reach out and touch them both.  I would have expected to hear vertical discontinuity at such a close distance, but he didn't hear it and neither did I.  So as long as I don't get the drivers farther apart and/or raise the crossover frequency significantly relative to the Jazz Modules, we should be okay.

[Duke]

The solution to these sometimes conflicting requirements is inevitably a compromise, but that's okay because so are all competing speakers. 

Briefly, my small-room-friendly speaker will be a floorstander that juggles the above requirements in ways that are at times a bit different from what previous jugglers have done.  It will be physically large for a small-room speaker, but because it will be wall-friendly the actual amount of real estate it occupies won't be too bad.  It will have better radiation pattern control that its competitors, and will incorporate one feature to smoothe out the in-room bass and another to adjust for gain from boundary reinforcement.  It will have good efficiency and will be an easy load for most amps.

The price ballpark I'm shooting for is 2.5 grand, and it can be used in bigger rooms as well (when adjusted accordingly).  It will retain the good power response and dynamic contrast that my other models have, but will have less bass extension in a big room compared to my other models.

I've finished prototyping and my woodworker is building the cabinets, with delivery expected by the end of January. 

[Russell Dawkins]

Hi Duke - it sounds like this is starting to converge with Hsu's thinking, just a little. His systems with the HB-1s, mid bass modules and sub are in a similar ballpark in some ways.
At least his HB-1 could be considered a small room answer to your designs, even if not in the same league in terms of refinement.

[Duke]

Hi Russell,

Thanks!   

Yeah that Hsu system is a lot of bang for the buck.  I could not build an equivalent satellite speaker in its price range.   

I do have a satellite/sealed Swarm system under development that will come in considerably below the Planetarium Alpha in price, and would work great in small rooms.  But any small room that system finds its way into will probably have to be a dedicated mancave, as the multisubs take up a lot of real estate. 

If I can figure out a way to do a cost-effective microSwarm, that might be something that would find a home in a small room system.  But I haven't yet.

To the best of my knowledge I haven't even come close to selling a Swarm to someone who has a small room, so despite the benefits that multisubbing would bring I think two floorstanding speakers are the realistic upper limit in most cases.   I have a 10.5' x 9' x 6.5' spare bedroom in the basement that's my small-room test chamber, and only if that room were re-dedicated as a mancave could I fit a Swarm into it.

[Duke]

Preliminary specs:

Type:  Two-way controlled-pattern floor standing bass reflex system

Radiation pattern:  90 degrees in the horizontal plane from about 1.4 kHz on up

Impedance:  8 ohms nominal, tube-friendly

Efficiency:  93.8 dB/1 watt

Recommended amplifier power:  5 to 500 watts

Thermal compression:  Less than 1 dB at 50 watts (110 dB SPL)

Typical bandwidth, medium room:  45 Hz to 18 kHz

Typical bandwidth, small room:  33 Hz to 18 kHz

Dimensions:  44" tall by 11" wide; depth varies from 20" to 10.5"

Estimated weight:  95 pounds

[timjthomas]

Hi Duke,

Any pictures?

Also, have you thought about doing a "smallish" monitor -- say around the size of the Ellis 1801 or so?

-Tim

[Duke]

Hi Tim,

No pictures yet, sorry.  Still waiting on production-quality enclosures from my woodworker.   They will be a little bit wierd looking, but not like MBL-wierd.  More like "hmmm...".

I have done a LOT of thinking, and some prototyping, of monitors in the size ballpark of the Ellis 1801.   And so far I have not come up with anything that I think breaks new ground or offers a significant improvement over what others are doing.  I do have a high-output satellite speaker in the works, but it only goes down to 75 Hz so it's not competitive as a fullrange monitor type. 

Conceptually I'm not sold on monitor+stand as a cost-effective alternative to a floorstander, but if I figure out a way to build a competitive monitor-type speaker I'll do so.

Duke