Factory Direct: To B-DEAP Or Not To Be Deep
Employing new concepts in subwoofer design

By Tom Danley

This Factory Direct was submitted by Sound Physics Labs. Live Sound makes every effort to eliminate any use of marketing inspired hyperbole.




The B-DEAP32 takes advantage of boundaries to extend bass response.

Some of my recent loudspeaker designs are partially driven by the need to be able to operate (with minimal interference) on a physical boundary like a wall or an acoustic boundary, such as an identical loudspeaker. This is the property that allows the Sound Physics Labs (SPL) boundary compliant Unity loudspeakers to be arrayable acoustically, not just physically.

The Boundary Dependent Ext-ernal Air Path (B-DEAP) concept is a continuation of this direction. The primary thing that keeps most sound folks from considering a bass horn in most situations is the “well-known rule” that says “a 30 Hz horn has to have a 10- to 12-foot diameter mo-uth in order to work properly”. This makes for a rather “large footprint” right off the bat ­ even if set up as a multi-box, segmented horn system.

A less considered side of this “well-known rule” is that the design of a good horn is much more difficult to achieve than a vented box, and some equipment developers who point out the problems with bass horns also happen to make vented subs and/or vented box drivers.

As a result, bass horns are often thought of as specialized things generally to be used only when their considerably greater output (in relation to their size) is the driving concern. Even with this in mind, there are still some horns sold that are far from what would be possible in their box size if properly designed. (At least in my humble opinion.)


The two-over-two configuration used to determine some of the TEF data that follows.

BIGGER MOUTH?

The approach behind B-DEAP: yes, one does need a minimum mouth size for the horn to work properly, but what if one were to minimize the amount of wood and air needed to attain that appropriate mouth size? We believe our approach indeed achieves this.

First, consider what is really needed to make a horn work properly. A large mouth-size requirement for low-frequency cutoff is a fact. However, what may not obvious is that if the horn is placed on the ground, the mouth area needed is cut in half because of the “acoustic mirror image” that the solid boundary (the ground) produces.

Each time one cuts the radiation space in half (from full space, like hanging in the air, to half space, like sitting on the ground) the needed horn mouth area is halved. And in a corner, for example (1/8 space if unbounded in the other directions), only 1/8 of the full space mouth area is needed.


Figure 1: Response in 1/2 space of four units measured at 10 meters.

Looking out from that room corner from a sound wave’s point of view, one sees the corner is also a conical horn. The concept of “corner loading” like that of the famous Klipsch horn was an attempt to utilize this conceptually.

Follow the rate of expansion of the cross sectional areas, and you see the dimensions of the enclosure (and the space it takes up) do not couple well with the boundary area expansion, as evidenced in its response and impedance curve. That is to say, the horn does not tie in well with the natural horn formed by the walls and floor boundary.

When the Unity horn was being developed, it was also necessary to deal with the low-frequency loading of a conical horn. With a corner, or any space for that matter, essentially all one needs to do is couple into the horn at the right place for the desired frequency range.

EXPANDING SPACES

The idea of B-DEAP is to quite exactly consider the expanding space outside the subwoofer. This includes the shape of the radiation space around the speaker, the air space be-ween the speaker and the room boundary, the speaker’s external dim-ensions and internal horn structure. These are all coordinated to make the horn flare and room boundary expansion not only ties in, but also does it as seamlessly as possible.


Figure 2: The horizontal polar plot for the two-over-two array.

The executive summary of this concept would read something like “since the enclosure has ridged walls on the outside as well, the idea is to use the physical boundaries, enclosure walls and geometry to form the large end of the horn, outside of the traditional enclosure”.

Specifically, if a B-DEAP32 subwoofer is placed at the right location and distance from a room corner, the exterior part of the horn path is about 75 inches long, Imagine not having to move or buy the last 75 inches (at the “big end”) of a large bass horn. That’s a lot of wood, air and space.

The B-DEAP32 measures 42 inches by 42 inches by 18 inches, and in a corner, we’ve measure it to provide efficient bass response down to the mid-30 Hz region, at very high power levels, and this can be equalized to a lower low-end cutoff at lower levels. No 10-foot -diameter horn mouth needed; in fact, the horn of the B-DEAP32 is not visible.

“What if I don’t have a corner, just a back wall?” Good question! An acoustic mirror can also be attained with a wall, surface as well as another B-DEAP unit. Two of these subwoofers, coupled together, provide the acoustic equivalent to a single unit in a corner, with twice the power handling.

One thing I didn’t expect is from this design is that it is very forgiving in relation to needing to be placed in exactly the “right space”. The 1/8-space horn has proven to work well on walls (1/4 space), it’s just down some at 32 Hz. Clearly there is some unintentional flexibility here, but I’m not complaining!

TAKE IT OUTSIDE

Because room measurements in rooms always depend on a lot of geometry and variable “room stuff,” I thought outdoor measurements (1/2 space) would better provide an idea of the capability of multiple B-DEAP units to form acoustic boundaries, even though the low-end cutoff is not as low as when used/measured indoors, particularly in a corner.


Figure 3: Two subs measured on the ground plain, with the ground used as a mirror image to the horn outlets.

Note Figure 1, which shows the response in 1/2 space of four units (stacked two over two), measured at 10 meters (about 30 feet), a distance that minimizes the fact that the speakers are large and take up some of the space at 1 meter. Thus, the data reflects a more realistic measure of acoustic power.

Further, this was taken with 1 watt of total input power supplied, and at a distance which is -20 dB down from a 1 meter “point source”.

That is to say, the equivalent of 1 watt at 1 meter (1W/1m) sensitivity is 96.5 dB (average), plus 20 dB (for 10 meters distance), equals 116.5 dB. (Again, 1W/1m)

I was taken back by the high sensitivity. A true point source that was 100-percent efficient would only be 112 dB, and to be totally honest, my best horn designs to this point are only about 50 percent efficient. Clearly there was an explanation hiding somewhere.

The answer lies in the fact that the array of these four subwoofers has significant directivity. In this configuration, the final horn angle is set by the wall angle between boxes. This array was set to be flat across the front (180-degree wall angle). In other words the final expansion is a 180-degree wide horn, with the pattern set by the boundaries and acoustic mirror images.

Figure 2 shows the horizontal polar plot for this array. As you can see, there is considerable directive gain over a point source; the “DI” is about 7.5 dB (average) across the operating band. In fact the “impossible” sensitivity is de-mystified when one subtracts the 7.5 dB gain caused by directivity from the 116.5 dB, leaving us with 109 dB or about 50 percent efficiency.


What the rear of the subs look like. Note the horn opening.

Note too, there are no big lobes or interferences, nor were any processors or racks of amplifiers, just bass directivity. I also measured two of these subs on the ground plane, using the ground as a mirror image for the horn outlets. (Some day I’m going to try playing bass guitar through this set up just for fun!)

Again measured at 10 meters and with 1 watt (total) of drive, Figure 3 shows an average sensitivity of 93.5 dB, which is equivalent to 113.5 dB (1W/1m). All of these measurements were taken with an Earthworks M-5 microphone and TEF-20, without whose faithful and unbiased guidance all these years, many of my “speaker things” wouldn’t have happened.

I hope this helps create some mental images of what the product is, and what it does. Anyway, until you hear one, this is about as good a description as I can come up with. You probably have the “big end” of bass horns lying around nearly everywhere you look, but remember, you just need the “smart end”.

 

Tom Danley is director of engineering for Sound Physics Labs/Servodrive, located in Glenview, Illinois. He is holder of several patents in loudspeaker design.

November 2003 Live Sound International

Email this story to a friend.