From the earliest work in sound reproduction, it has been known that horn loaded transducers provide the most efficient means of coupling the listening environment to the sound source. The horn acts as a transformer, converting high pressure, low velocity acoustic air movements into low pressure, high velocity sound radiation. Horns can be seen way back to earlier civilisations being used amplify and broadcast sounds. Conch shells and ram’s horns, later, Tibetan horns and alpenhorns are all based on the same principle as their modern counterparts. They convert a sound at the small throat into a broad, fast moving integrated wave front leaving the horn mouth that travels long distances to the ear of the listener.
The most familiar image of a horn in use is the “His Master’s Voice” painting employed by the Gramophone Company at the beginning of the 20th century to popularise their new product over the older phonograph, invented by Thomas Edison.
Before the First World War, records were played on gramophones with a clockwork motor and an acoustic membrane vibrated by a stylus needle in the record groove. This energy was captured and amplified by the attached horn. The efficiency of these horns was such that a reasonable Sound Pressure Level (SPL) could be achieved so that a small group of listeners could experience a performance that had been recorded live as though they were at the original recording. Of course in comparison to the live event the quality was severely compromised but, never-the-less, there was an aspect of realism that was very satisfying. This has much to do with the dynamic qualities of wave front produced and how the horn converts energy with such small losses.
In the 1920’s and 30’s things started to change dramatically. With the advent of electricity, radio and the thermionic valve a whole new approach was made to sound reproduction. Rice and Kellogg invented the moving coil loudspeaker at General Electric and Western Electric (a branch of Bell Labs) produced many innovations. In Britain Paul Voigt developed wide frequency drivers that would eventually become Lowthers and loaded them into his Tractrix horns. The cinema encouraged a speedy development of horns as the only way of reproducing realistic sound at the movies. Western Electric made wonderful exponential horns and later Vitavox, Tannoy, Altec Lansing and Electrovoice amongst others followed suit.
A few leading designers studied the subject in depth and produced patents and classic speakers for the new era of hi-fi that developed after WW2. Paul Klipsch developed his ground breaking Klipschorn. Guy Fountain produced the Westminster for his Tannoy company. Both remain classic designs and their contemporaries like Olsen at RCA worked in depth to try out many different expansion curves from conic through exponential, parabolic and tractrix to multiple horns.
More recently Professor Jack Dinsdale consolidated much of this work in his UK Wireless World articles in the 1970’s and Bruce Edgar did similar in the US in his speaker builder magazine.
A few enthusiasts kept the flame alive for horns and single full range drivers with valve amps alive for the last 40 years in the face of the age of cheap transistor technology that allowed very large output amplifiers. The days of valve amps with just a few watts output were numbered, and the use of horns almost relegated to history except perhaps in big PA systems where efficiency was still a big factor. Most modern sound equipment had multiple drivers and big amps. However it was always clear that there was something that horns did that was special. It is hard to define because our measurements tend to focus on frequency response rather than dynamic response. We obsess about total harmonic distortion and ignore the intrusive inter-modulation distortion produced by multiple drivers and crossovers. It is the immediacy, the “being there”, the depth of field that gives horns that special quality.
Of course there are compromises as with everything. Most horns have compression loading to the rear of the driver and a front throated horn. This works well for treble tweeters but becomes increasingly a problem as the frequency is lowered. Horns are ideal at the mid frequencies, but because of their efficiency in this range (analogous to the way the human ear works being more sensitive to the vocal range), they have to be balanced with other frequencies.
Larger horns are not better they just produce lower frequencies. To this day the laws of physics dictate the size of the bass horn and all horns must compromise or be extremely large to produce really low notes. If horns are arranged in bandwidths with top, mid and bass or even sub, one arrives at similar problems to any multiple driver … crossovers! Hi-end products like Avant Garde use this arrangement. Even with PA style active crossover and separate amps the problem of uniform power band arises. At one level everything may sound fine but turn it up and the levels have to change uniformly or it sounds like there are steps in the filter and the flat response fails.
Added to these problems come the other factors, if everything is compression loaded there is a sort of pneumatic spring behind the driver that loads the back of the cone. To stop resonance and reflections affecting the cone it needs more mass and heavier power. This produces the desired energy levels but simultaneously slows the driver down, defeating one of the main advantages of the horn. For a long time, the holy grail has been a compound horn with front and rear loading. This coupled with a single full-range driver as per early Lowther designs was regarded as the epitome of dynamic realism.
Unfortunately this is very hard to achieve, and many people succumb to a sub-woofer or other two-way compromise. There are many lovely sounding speakers that very nearly achieve these goals but few quite get there. With the AXJET we have attempted to do just this and have come very close to an ideal horn speaker with that vibrancy which is not available elsewhere. For details see AXJETS.