Why you should consider ActiveDrive™ for better sound reproduction Active versus Passive A Passive system is one where a network of "Passive" components is used to split the power coming from a single amplifier between two or more drive units. An Active system is one where the audio signal, at line level, is split in two or more signals of limited bandwith, feeding a dedicated amplifier for each drive unit. Let's examine what's happening in a Passive system: An amplifier amplifies the audio signal coming from a CD player (or any other source) to a level high enough to power a loudspeaker. When an amplifier is connected to a passive loudspeaker system, the amplified signal is routed through a Passive crossover network made of inductors, capacitors and resistors. The purpose of the crossover network, is to split the amplified signal in specific frequency bands, and to adjust the relative level of each of those bands to recreate the original sound. For example, in a two way loudspeaker, the low frequencies are sent to the woofer, while the high frequencies are sent to the tweeter. The process is substractive. If the output of the tweeter needs to be reduced to match that of the woofer, power sent to the tweeter will be wasted, transformed into heat by resistors. As the name implies, the modifications to the amplifier's output are done by using passive devices. Passive crossover networks are inefficient. For example, in the low frequency section of even the simplest passive network, the amplified sound may go through tens of metres of copper wire (forming the inductor) before reaching the woofer. For complex networks, the number of passive devices may reach 50 or more components, each of them robbing a little of the signal. What about Active: The signal from the CD player (or any other source) is split into 2 or more dedicated audio signals, and each one being sent to an audio amplifier, driving a specific drive unit. For example, the low frequency amplifier amplifies only low frequencies, and the high frequency amplifier amplifies only high frequencies. There are as

many amplifiers as drive units, and each amplifier is directly attached to the voice coil of each corresponding drive unit. No detrimental passive components between amplifier and driver. The result is pure, transparent sound. Arteluthe DSP processing With Arteluthe DSP - or Digital Signal processing - the signal is transferred to high resolution digital format, then, mathematical operations are applied to the digital file in order to derive a perfect set of signals, frequency and time domain corrected. The signals are then converted to analog and amplified by each of the amplifiers. The signal processing is performed in digital domain, with no loss of information or quality. If the level of one drive unit needs to be raised or reduced, it is only a simple matter of adjusting the gain. Nothing is lost. A closer look at the passive crossover In its simplest form — a first order network —, a two way crossover would have: — an inductor connected between amplifier and woofer; — a capacitor connected between the same amplifier and the tweeter, — and most probably a set of resistors to adjust the relative level of both drive units. It could also include a section made of inductor, capacitor and resistor (LRC) to flatten the impedance of the drive units, or similarly complex baffle step compensation filter. Unfortunately, each time a filter section is added to correct a problem, some of the efficiency is lost. The quality of reproduction is affected, sound becomes dull and lifeless. How all those passive components affect the amplifier? Let's imagine a typical 50 watts amplifier. Amplifiers are power rated (expressed in watts), into a load — or impedance (expressed in ohms). For example, 50 watts into an 8 ohms load. The actual power generated varies depending on the load condition. For example, our amplifier, rated at 50 watts into 8 ohms, will put out 25 watts into a 16 ohms load, 100 watts into 4 ohms and 200 watts into 2 ohms! Obviously, not all amplifiers are able to supply that much power into 2 ohms. Some amplifiers can — some cannot.

Unfortunately, loudspeakers are not constant loads. The impedance of an 8 ohms speaker may vary from 2 ohms (or lower!) to 50 ohms - depending on the frequency. The variation in the actual load, or impedance of the loudspeaker is caused mostly by the crossover filter, the type and number of drive units and by the acoustic loading of the enclosure. From our simple network example above, an inductor connected to a woofer will raise the impedance at higher frequencies, therefore, the amplifier will send less power to the woofer at those frequencies, reducing the sound output. At the same time, a capacitor, connected to the tweeter, raises impedance as frequencies are below a certain frequency. The transition point between the woofer and the tweeter, in this example, is characterised by an impedance peak at the crossover frequency, where the impedance of the woofer section raises until the high impedance of the tweeter section begins to fall. Impedance in active systems The amplifiers in an active system are presented an almost ideal, flat load curve, that is a impedance load that does not vary wildly at different frequencies. The impedance of a drive unit is high at the resonant frequency, then drops to the specified impedance (4 or 8 ohms), and rises again gently, in relation to the inductance of the voice coil. In most active situations, the amplifier will power a driver only in its ideal bandwith, above resonance, where the impedance is flattest. As a result, the amplifiers are always working in ideal conditions. "Hard to drive" passive speakers Some may have heard of "hard to drive" passive speaker systems. These are speakers were the behavior of the load is erratic, with terrifying low impedance, and / or wild variations of resistive / inductive behavior. The lower the load, the closer is the amplifier to being short circuited! In an ideal world, all loudspeakers would have constant, reasonable load. In reality, designers sometimes prefer lower loads as the speakers sound "louder" in the dealer's showroom. Frequency curve The ideal loudspeaker is one that, among other things, can reproduce all frequencies at equal loudness level. Most loudspeaker drivers are not perfectly equal at all frequencies. Some deviation might be acceptable, but

some are not. In a passive system, deviations are generally corrected by adding passive components in the signal path We saw earlier that a passive system works by substraction. If a driver has a frequency peak somewhere, it needs to be corrected by passive components to the filter, removing excess energy. Drops in frequency curve, in passive systems, cannot be compensated for, unless energy is wasted on both sides of the drop. If the designer wants to extend the frequency range of the system at the bottom of the range (i-e better low frequency extension), the designer has no other choice than reducing the output of all frequencies above - further reducing efficiency. In an active DSP system, those corrections are only a matter of raising or reducing output gain, with no detrimental effect on sound and still no components between amplifiers and drivers. The effect of speaker cables The effect of speaker cables on sound is wildly debated. One thing is certain, a speaker cable, no matter how good or expensive, cannot improve the quality of the material being reproduced. It can unpredictably affect the sound reproduction, even act as a way to control the tone of a system. Whether or not we agree on the colateral purpose of loudspeaker cables, we all agree that cables should be kept as short as possible. In all Arteluthe products, the length of wires between amplifier terminal and voice coil is kept as short as can be, well under 2 m long. The effect of the wire has been compensated for in the design of the loudspeaker, so the sound you hear is exactly like it should be, provided high quality source and interconnect are used. We offer interconnect balanced cables especially designed for our products. Directivity considerations Now, let's examine directivity — which affect the soundstage of the system. All loudspeakers do not have perfect sound distribution at all frequencies. For example, a 6 in. woofer will become directive at higher frequencies — typically above 1 500 Hz. At these frequencies, the output radiated on the sides will be far less than the output radiated in the front (on axis) of the driver. Even if a 6 in. woofer can reproduce frequencies up and above 6 000Hz, most of the output will be generated in the front of the driver.

We would prefer to have all drive units radiate broadly in space, with the same sound intensity whether listening from the front, or side. The problem gets more complex as we add drive units. In a two way system, with a 6 in. woofer and a 1 in. tweeter, the woofer will be very directive (beaming) at 3 000 Hz, while the tweeter will have a broad dispersion at that same frequency. The sound field will not be coherent at all frequencies, from all angles. Earlier, we saw that some loudspeakers are hard to drive. In this section we see that some speakers that are hard to position! Suppose our speaker is a two way system, with a crossover frequency of 3 000Hz (where our ears are very sensitive!), the sound may seem agressive when listened to on axis, but if we angle the speaker differently, the same frequencies may be lacking, because of insufficient lateral energy. What results is a tiny sweetspot. That sweetspot is the area in space when the sound is optimally balanced with deep and wide soundstage. Unfortunately, should you move your head just an inch or two, that tiny sweetspot will collapse, resulting in incoherent sound field. With ActiveDrive™, we were able to implement a crossover frequency lower than the frequency at which narrowed woofer directivity occur. So frequency playback are handed out between woofer and tweeter when both units have optimal horozontal dispersion, resulting in a wide, deep and stable soundstage, no matter whether you sit in the sweetspot or not. Timing consideration Music is all about time. Passive speakers are notoriously affected by timing error. When the output of one driver does not exactly departs or arrives at the exact same time as the output from the other driver, time smearing occurs. Timing is especially important at or near crossover frequency where frequencies overlap, causing blurred reproduction or even cancellation due to phase problems. Complex passive crossover topology are affected by severe phase rotation, creating unwanted delays in playback. Simpler topologies are affected by imprecise

frequency overlaps, where the output from one driver does not line up in space with the other driver's output. Some designer rely on "phase alignment", where both drivers are physically aligned on the vertical angle, reducing error, but at only one point in space. An active DSP system is always more coherent, because parts of the signal may be stored in memory, and delayed in order to align frequencies in the time domain. Furthermore, ultra steep crossover slopes minimize overlap, eliminating muddiness. What results is clear, clean, dynamic and precise sound that no passive system can match. Complexity Earlier active systems were difficult to implement, due to complexity. Today, systems like Arteluthe ActiveDrive™ are so small and efficient they can be embeded inside the loudspeaker enclosures, with absolute minimal wiring. Therefore, the complete sound system's complexity is greatly reduced. No more clutter, busy racks and wires everywhere. Not only is the system cleaner and leaner, it is actually cheaper on top of offering superior performance. Sound reproduced the way it should be Perhaps the biggest advantage of Arteluthe ActiveDrive™, is that you always hear sound reproduced the way it was meant to be. No more "hard to drive"and "hard to place" speakers, expensive speaker cables, mismatched and unpredictable components. Just pure sound with no guesswork. By Robert Gaboury  2013 Arteluthe, All rights Reserved. Arteluthe is a registered TradeMark of 7953135 Canada inc. Version1,1 May 1st, 2013.