LOUDSPEAKER SYSTEM COMPONENT COMPLEMENT  

(4) 2242H subwoofers (2.1% conversion efficiency each or 8.4% in tandem)
(4) 2227H high-efficiency 15" woofers (one pair each side, about 17% efficient each pair) 
(2) 2123H high-efficiency 10" midrange (3.5% efficient each) 
(2) 2382A horn with 2450H compression driver (30% efficient each)

Efficiency or "conversion efficiency" is defined here as the ratio-expressed as a percent-of the acoustical output obtained for a given electrical input. Thus a transducer that delivers 5 acoustical watts to the air when fed a 100-watt electrical input signal is said to be "5% efficient."

In case you're wondering if the poor 10" mid with its mere 3.5% efficiency, can keep up with the horn, rest assured I needed 10 dB of attenuation on the mid to get flat frequency response.

For amplification, I used two BGW SPA-3 triamplifiers. BGW was happy to set up the triamplifiers to provide high-pass filtering for the two 15's at 80 Hz, band pass for the 10" midrange from 300 Hz to 1200 Hz, and band-pass and horn EQ filtering with the lower end of the high frequency band at 1200 Hz. The crossover slopes are Linkwitz-Riley type 24 dB/octave.¹ The amplifier's input section also includes switched attenuation and built-in signal delay to adjust the acoustic time of arrival for acoustically aligning the cones and compression driver. Although the amplifiers are each only 5.25 inches of rack space, they each produce up to a total output of 1000 watts, providing 600 watts for each pair of fifteen's, 200 watts for the mid and 200 watts for the horn. This represents an average of around 30 dB of headroom above normal living room listening levels, which generally range in milliwatts for these speakers. Even though this much headroom and power seem to be overkill, I assure you that is not the case. I include here a table from my Audio Engineering Society workshop on basic audio-it includes the results of about 130 hours of measurements I did of live and recorded material using a $7500 Brüel & Kjær true-RMS voltmeter and a $40,000 Brüel & Kjær audio analyzer to record the peak-to-average power ratio of various signals. (1997 note: these are no longer available except on special order. I recommend using four solid-state amps of your choice-my choice is Adcom-and building a 48 dB/octave (8th order) Linkwitz-Riley crossover).

In order for untrained listeners to perceive no obvious squashing of dynamics, audio reproduction systems should be capable of the following peak-to-average ratios for these stimuli: 

¹ Linkwitz-Riley 24 dB per octave (fourth-order) or 48 dB per octave (eighth-order) crossover filter slopes produce flat acoustic energy summation through the crossover frequency region.

The BGW SPA-3 turned out to be the elegant and simple alternative to a large rack of gear. Having done similar projects many times in the past, I can honestly say I would not go back to the racks and cabling and connectors and ground-loop chasing always necessary as long as the SPA-3 is available to eliminate all the little gremlins that tend to pop up when one builds up complex systems.

Note to builders: I don't recommend trying to build passive crossovers for these units. Acoustic time-of-arrival delay of sufficient time is not practical with passive devices and system performance really suffers without it. Additionally, to obtain anywhere near the performance of the tri-amped system using coils, caps, power resistors and such, the crossover would be far more expensive than a pair of SPA-3's and would probably weigh somewhere in the neighborhood of 100 pounds-because of all the huge coils.

There have been many questions in the letters I've received about doing this project some other way, with modifications, with different components etc., perhaps to accommodate someone's favorite component or fit some particular space or budget. I cannot address these questions individually without addressing the particular goal of the individual asking. I can do this-it is in fact, my livelihood-I will be happy to design a custom system for you, but be aware that this must be done on a consulting time basis. This project however, is what it is, and as such, it pleased all the audio golden ears at Disney enough that they insisted on using my lab pair for the yearly Halloween show with an audience of 2500 people spread over an outdoor area covering 235 degrees, as well as company meetings with Eisner and Wells in huge tents with thousands of people in the audience. This with a sizable company inventory of all manner of THX-rated large theater systems, rock-concert boxes form various manufacturers, etc.

© 1997 Drew Daniels