BEGINNINGS

Beginnings

The horn loudspeaker is the most natural physical concept for sound reproduction.

Even today its characteristics offer undisputable advantages compared to other loudspeaker systems. In order to give you a closer insight into this technology, we have compiled for you a short history as well as the most important principles of a horn in five chapters.

The Very Beginning

It happened more than 100 years ago, when, for the first time, Emil Berliner presented his gramophone to the public. The horn of the gramophoneamplified the mechanical oscillations of a pin running along a groove in a disc producing a sound that could be heard by the human ear. This technological feat was followed by the search for the optimal shape of the horn. This proved to be a bold and complex venture as already hinted at by Harry F. Olson: "The design of a horn loudspeaker is usually a long and tedious task." Similarly, horn pioneers as Gustavus, Webster, Klipsch and Voigt required decades in order to explore the laws of the horn technology. In 1926 Paul Voigt submitted his first tractrix horn to the British Patent Office.

The Golden Age

The heyday of the horn loudspeaker followed. At that time tube amplifiers had only very low power capabilities and required very efficient and powerful loudspeaker systems. The horn loudspeaker was the only existing loudspeaker concept which was able to transfer low electrical power into high sound levels. Famous classics amongst these designs were:

The Horn Principle Common Knowledge

Horns transform pressure by using a sound source, which means a driver. Thus they work on a principle opposite to the human ear. The ear consists of an outer, middle, and inner ear. The outer eargradually decreases in diameter via the auditory canal towards the eardrum. The pressure increases accordingly. Thus, we can perceive infinite pressure variations which enables us to hear better. In former times, when hearing aids were non-existent, simple horns were used as such devices. Try for yourself, by increasing the size of your ear with the palm of your hand, and you'll achieve the same result.

Loudspeakers, however, are supposed to work the other way round. They should improve the emission of sound waves. For this purpose, a sound emitting membrane is placed at the beginning of the horn, which means the horn throat. You'll experience this by cupping your hands around your mouth. Your voice will be louder, however, it will sound slightly different. This can easily be explained.

Your hands form a horn of approx. 1,000 Hz. The sound of your voice is thus being increased within this frequency range which is responsible for the unusual tonality. This effect of the horn was known to even the Neanderthal Man, even though unconsciously.

The Principles of Physics

Horn loudspeakers are provided with a sound reproducing membrane on the horn mouth The operational membrane has to oppose an increased pressure induced by the horn shape. If you compare the excursion of a membrane built into a loudspeaker enclosure with one installed behind a horn and actuated with the same level of power, this will result in a reduced membrane excursion depending, of course, on the shape and size of the horn. Hence, the following principle applies: the more the membrane excursion is reduced, the higher the sound intensity will be.

If the membrane excursion of a loudspeaker is reduced by using a horn, the waste of energy due to friction on the moving elements (voice coil, suspension, fixtures on the rim), as well as due to friction of the flow of air around the voice coil will be reduced at the same time (see Figure 1. "Schematic Design of Driver").

It's as easy as that: the energy, used in order to allow the significantly stronger movement of the operational elements of the loudspeaker, is transferred by the horn with its increased mechanical resistance into increased acoustic output.

The actual sound pressure level gain isn't the maximum sound intensity which can be achieved. When comparing two identical loudspeakers at the same sound level, one on an ordinary loudspeaker built into an acoustic baffle, and the other equipped with a more efficient horn, the membrane excursion in the latter can be reduced by factor of 10.

The reduced membrane excursion causes a membrane velocity increased by the factor 10. If you wish to obtain this velocity within the same time, which means a velocity increased by the factor ten, the distance as well as the acceleration would have to be increased again by the same factor. This implies that an amount of energy by factor 100 would be required.

Figure 1 -- Schematic of a Driver.

Figure 2 -- Slew Rate and Post Oscillations. A square impulse induces an increasing membrane excursion on the loudspeaker. Since the membrane is inert, it cannot follow a sudden electrical impulse directly. The more markedly the excursion between initial- and post-oscillation phase of a membrane, the longer the time required to come back to its original position. New signals cannot be clearly reproduced since the loudspeaker has not yet reached its non-operational position.

Due to the significantly reduced membrane excursion a horn loudspeaker is extremely quick. This not only refers to its acceleration behavior, but also applies to its deceleration capacity, since the acoustical attenuation through the horn causes a considerably reduced amplitude of post-oscillations.

Moreover, the critical partial oscillations of the membranes are avoided by the reduced membrane excursion and the acoustical attenuation through the horn. Thus, the distortions are nearly non-existent and dynamic compression does not occur.

By using a horn, the acceleration of the membrane and as well the air happens nearly without inertia, allowing a speed of sound which wouldn't be feasible without a horn.

Dynamic pressure amplitudes are reproduced extremely fast and the membrane very quickly reaches back its neutral position. That is to say, the critical post oscillations of the membrane are efficiently suppressed (see Figure 2: "Slew Rate and Post Oscillations"). Due to this fact, the quick suppression of post oscillations in horn loudspeakers results in an exceptionally audible resolution of infinite details.

From 0 to 100 dB

What kind of impact does this speed have? Music consists of constantly changing intensities, which means sound levels can vary in their intensity depending on the music. A loudspeaker has to follow these impulses accordingly, and should reproduce them as precisely as possible.

The following example is going to depict the related factors: Just imagine the narrow bends of the Monte Carlo F1 racing circuit in Monaco. No problem for a racing car, but, for an "ordinary" car the narrow bends pose an insurmountable problem. It can easily reach the required speed, however, acceleration and deceleration distances are completely insufficient.

The very same applies to an ordinary loudspeaker. The power and the maximum volume are absolutely irrelevant!! With an "ordinary" speaker enclosure you are able to achieve high sound pressure levels. Much more important is its acceleration potential since every loudspeaker can only process information which is slower than the loudspeaker's potential. Control of the driver both starting (accelerating adequately) and stopping with no overshoot (deceleration) are absolutely critical to the impression of hearing a live musical event. That's the reason why, with our horn systems, you'll hear all musical details, even those you've never heard before!

Measuring Technology

The behavior of a loudspeaker along the time axis (acceleration/deceleration) which has just been described cannot be clearly discerned via frequency response graphs.

Response curves show whether a loudspeaker is balanced over its entire frequency range. This implies that you obtain the value of the amplitude of the membrane excursion and in turn, the induced sound pressure.

From frequency plots we can't see what time the loudspeaker requires in order to reach this level and the time needed to come back into its initial position. The latter can be made visible graphically by means of a so-called waterfall diagram by using a Melissa measuring system.

However, one has to be especially cautious with regard to the evaluation and comparison of the graphs of other loudspeakers, since most of the time the measurements are based on a given initial power input. And the higher the restoring forces on the actuating elements (spider, crimp; crease, fixtures on the rim) the quicker the loudspeaker comes back to its initial position. And that leads us to...

Linear Dynamics and Frequency Response

Linear musical dynamics are essential if music is to be reproduced in such a manner that we can suspend disbelief and accept the sound as "live." First, this means that, for a given electrical input signal of 1 watt, a typical speaker may be specified to output exactly 85.0 decibels at a distance of 1 meter not more and not less.

Next, let's look at a typical musical peak, say just 18 decibels. Suddenly we are asking for significant driver excursions, acceleration and deceleration.

Then, more critically and less likely, for a given peak input of 64 watts, the speaker will output exactly 103.0 peak decibels not more and not less. As the speaker is played, this exact ratio changes, and in all drivers to a slightly different extent.

Each driver in a loudspeaker system has its own unique characteristics. When a subtle change in musical dynamics occurs, generally such loudspeakers respond equally at all frequencies. However, when the system has been playing at more realistic volumes, this ceases to occur.

As the voice coils heat up, each driver (and even crossover components) begins to heat up and ultimately it may saturate (no response to more input other than heat and distortion). It no longer responds linearly to an input signal. How they differ from linear response determines the actual usable frequency response on large-scale peaks.

The loudspeaker industry's emphasis on static frequency response curves is amazingly similar to the electronic industry's long-term resistance to using other than static tests to evaluate amplifiers. Today's best amplifiers are designed with dynamic test standards. Yet no one wants to mention how loudspeakers become very different in their response curves and overall dynamics as they're asked to reproduce more realistic volume levels.

The Solution to the Problem

When we ask a driver membrane to dramatically reduce its excursion requirements, we dramatically reduce the chances for error. A loudspeaker driver that is highly efficient say 94 dB at 1 watt still must move the same great distances. The only way to reduce the opportunity for driver non-linearities is to reduce the driver excursion. For that, you need an acoustic transformer -- a horn.

This results in a vastly superior kind of efficiency. 94 dB with a horn-loaded driver calls for the driver to only reproduce maybe 84 dB. The horn does the heavy lifting.

Distortion is reduced. Maximum power handling is increased. Music's dynamics are reproduced intact. Sudden musical transients are startling. Frequency response instrumental timbres don't change with level. Details emerge from your music that you never heard before.

In short, the sound can be more life-like, but with hornspeakers, our work has just begun...