Acoustic Research AR-303 Bookshelf speakers
Have you ever felt nostalgic about a car you bought more than twenty-five years ago, which at the time provided the last word in driving ease, comfort, and performance? Suppose its manufacturer built a 1995 version using an almost identical design with some improvements made possible by technological advances. How would you like to make a side-by-side comparison between the two? I have had the pleasure of doing just that, not with a car but with a loudspeaker, and it was everything I could have hoped for.
In 1954, Acoustic Research (AR) was founded by Edgar Villchur and Henry Kloss. Their initial product, the AR-1 loudspeaker, featured a simple yet revolutionary design consisting of a very compliant, long-throw "acous-tic-suspension" woofer, in a relatively small, tightly sealed enclosure, paired with an 8-inch midrange/tweeter. The woofer cone's restoring force was supplied largely by the compression of the air in the cabinet, instead of an inherently nonlinear physical suspension, eliminating a major source of bass distortion. The AR-1 set new standards for low-distortion, deep-bass reproduction (down to 30 Hz or below) from a "bookshelf" speaker.
Several years later, AR introduced a more advanced acoustic-suspension speaker, the AR-3. Based on an improved bass driver teamed with the world's first dome drivers (a 1-1/2-inch midrange unit and a 3/4-inch tweeter), the AR-3 solved the problem of achieving midrange and treble performance that could match its bass standards. In 1967 the AR-3 was upgraded to the AR-3a, which for many years was a recognized standard for quality sound in the home.
Over the years, AR has changed ownership a couple of times. Today it is a division of IJI (International Jensen, Incorporated), which is seeking to restore the AR name to its early glory with a series of speakers designed in the company's grand tradition of providing true musicality with the deep, low-distortion bass characteristic of a good acoustic-suspension woofer.
The AR 303 is the flagship of that line. It is almost exactly the same size as the AR-3a and has essentially the same driver complement, although the drivers are all newly designed for much better performance than was possible with the technology of the 1960's. It has a 12-inch woofer, a 1-1/2-inch dome midrange driver with the distinctive "fried-egg" appearance of the original (it's protected by a metal grille), and a 3/4-inch dome tweeter. The crossover frequencies have been changed slightly, from 550 Hz and 6.5 kHz to 650 Hz and 5.5 kHz. In the AR-3a the midrange and high-frequency drivers were mounted diagonally across the upper part of the speaker panel, but in the AR 303 they are vertically aligned on one side of the panel (unlike the AR-3a, the AR 303 is made in mirror-image pairs).
The AR-3a had adjustable level controls for its midrange and high-fre-quency drivers, with a range from fully off to maximum. In contrast, the AR 303 has no level adjustments. Although the two speakers are very nearly the same size and weight, the AR 303 has about 15 percent greater internal volume.
There are other differences as well.
The AR 303's sensitivity is rated at 85 dB, about 4 dB less than the AR-3a, and its specified bass response extends a bit lower than the older speaker's. But the real goal of today's AR designers was to provide the essential sound quality of the AR-3a in a modem high-performance speaker system.
We were very interested in determining how close they came to that goal. To that end, we are indebted to collector Thomas Tyson for the loan of a pair of virtually mint-condition AR-3a speakers for comparison with the new models, both in listening and measurement. In addition, I Could not resist bringing my own vintage AR-1 (circa 1955) out of retirement, at least for measurement purposes, even though all it shares with the others is the basic woofer design.
Not surprisingly, there were more similarities than differences between the AR-3a and the AR 303. In fact, their composite response curves (a close-miked bass measurement spliced to a room-response measurement made with a sweeping warble tone and averaged from the two speakers at a single microphone position) were amazingly alike. The AR 303 had significantly flatter response at the high frequencies, however, about 3 to 5 dB stronger than the AR-3a above 6 kHz.
Quasi-anechoic MLS frequency-re-sponse measurements brought the differences between the two speakers into stronger relief. The AR 303 was smoother overall and had stronger output at the highest frequencies (from 10 to 20 kHz). On-axis at 1 meter its output variation was +1.5 dB from 300 Hz to 18 kHz, certainly among the best responses we have yet measured from a speaker.
The AR 303's woofer was true to its heritage, producing only 5.5 percent harmonic distortion at 20 Hz with a 90-dB sound-pressure level (the AR-3 a and my old AR-1 were in the same range, generating around 6 percent distortion). The measured sensitivities of the two speakers were 88 dB for the AR 303 and 90 dB for the AR-3a, reasonably close to their specified values. The AR 303's impedance reached a minimum of about 3.5 ohms at 85 Hz and averaged 5 ohms or higher over most of the audio range. The AR-3a's impedance was a maximum of 13.7 ohms at 37 Hz, with a smaller peak to 8 ohms at 400 Hz, and was about 4 ohms or slightly less over the rest of the audio range.
Interesting as the measurements were, the real test was in the listening. With the speakers side by side and the AR-3a's midrange and treble level controls at maximum (the settings we had used during measurements), we listened to a variety of music while switching between the two speakers. There was no doubt that the two were close relatives in the bottom octaves. In fact, there was little or no audible difference between them in the deep bass. At first, however, the AR-3a was distinctly brighter than the AR 303. Turning the AR-3a's tweeter and midrange controls to their midpoints (probably the recommended "flat" settings) corrected the imbalance and produced a basically similar tonal balance from the two sets of speakers.
Further listening confirmed that the two speakers sounded very much alike. We did not attempt to stress the AR-3a speaker unnecessarily, but it was clearly performing as it did when it left the old AR plant in Cambridge, Massachusetts.
We played some CD's of organ music that included the lowest frequencies (down to the 20- to 30-Hz region) - ideal demo material for evaluating subwoofers. It was not much of a surprise to find that, like its predecessors, the AR-303 does not need a subwoofer at all, unless you are more concerned with shaking the room walls than with what the music sounds like. The AR 303 can deliver a clean, ear-popping output all the way down to 20 Hz; when we added a subwoofer the result was merely a stronger skin massage from the bottom octave.
We used the AR 303 placed on a low stand (as recommended) and several feet from any walls. I suspect that this speaker may require more careful placement than most to avoid undue bass heaviness. Certainly no other current full-range speaker that I know of in the AR 303's size range can match its low-frequency capabilities.
The AR 303, with a pedigreed descent from the earliest acoustic-sus-pension loudspeakers, is not only a fine reproducer, but a bargain as well. The AR-3a sold for $500 a pair in 1969; the AR 303, in 1995, is priced at $1,200 a pair. How many other things (houses, cars, etc.) have only a little more than doubled in price over the past twenty-five years?
