Loudspeaker Basics --
        Part 4
        We'll wrap up our series on loudspeaker basics with a brief
        discussion on crossover slopes/orders and speaker sensitivity. In the last part, we
        discovered that the hand-over between two adjacent loudspeaker drivers is similar to a
        disc jockey's cross-fade between two songs: while one song/driver fades, the other
        song/driver enters. A DJ may manipulate his fader controls slowly, for a very gradual,
        subtle and drawn-out transition. Or, he may opt for a very rapid progression between songs
        and create a sudden, very obvious and shorter passage.  
        The DJ works with time. He takes one or three or seven
        seconds to fade from one song into the next. A crossover works with frequency on either
        side of the specified crossover point. A sudden, or "hard," transition occurs
        over a smaller area of overlap, while a gradual, or "soft," transition may take
        place over several octaves. 
        How much roll-off (fade) is accomplished per octave
        determines the order of a crossover. The minimum amount is set at 6dB, which gives
        us what is called a first-order network. (It uses a 6dB per octave roll-off or
        filter slope: 6dB x 1 = first-order). A second-order network employs
        12dB/octave (6dB x 2) while a third-order network implements the even
        faster-acting severity of 18dB/octave (6dB x 3). Low-order crossovers (first
        and second) have gentler filter slopes and create more overlap between their drivers. High-order
        crossovers (third and fourth) act more aggressively, with steeper slopes that minimize
        such overlap. 
        And that's all you really need to know, although there's a
        lot more to this subject. There is considerable debate as to which crossover slope is best
        and entire books have been devoted to the subject. For now, rest assured that your ears
        will let you know without a doubt which loudspeakers you like and which ones don't make your
        grade, regardless of their (filter) slope.  
        This now brings us to loudspeaker sensitivity,
        sometimes also erroneously called efficiency. Sensitivity is simply a spec that tells us
        how loud a particular speaker will play given a standardized amount of amplifier power.
        The industry standard is to present an 8-ohm speaker with a 1 watt (2.83 volts) input
        signal and measure its output in an anechoic chamber with a microphone at a one-meter
        distance (an anechoic chamber is a specially designed room that  completely absorbs
        sound). The resultant spec will read something like 89dB @ 1W/1m. This means that for 1
        watt of input of amplifier power the speaker will output a sound pressure level
        (SPL) of 89dB heard at a distance of one meter in front of the speaker. Speakers that
        output lower SPL levels for a given input have lower sensitivity, while speakers that
        output higher SPL levels have higher sensitivity. 
        However, not everybody has an anechoic chamber, so some
        companies give you an in-room measurement (marketing people also like to do this
        because it can make the sensitivity specification "look" better). If you measure
        a speaker in a room, the boundaries (the walls, floor and ceiling) reinforce the
        speakers' output by 3dB -- thus that loudspeaker that measured 89dB in an anechoic chamber
        will measure 92dB in a room. This is called room gain and you must be sure you
        are comparing apples to apples when comparing speaker sensitivity.  
        A 3dB increase doubles the sound output of a speaker, so
        you can see why you must be careful to account for room gain when comparing loudspeakers.
        (Also, be sure that the loudspeakers you are comparing are both the same impedance. If one
        is an 8-ohm load and the other is 4 ohms, you need to subtract 3dB from the 4-ohm
        loudspeaker's sensitivity rating to get an equivalent measurement.) 
        Now, let's compare speakers with 89dB and 92dB sensitivity
        (assuming they are both 8-ohm speakers). What these specs tell us is that the second
        speaker, with the same input signal strength, will play 3dB louder than the first. It will
        play twice as loud with the same input power.  
        In real-world terms, a 3dB loudness differential doesn't
        mean that the 92dB speaker is better than the 89dB loudspeaker (a sensitivity
        measurement does not necessarily correlate with sound quality), but it does mean you may
        not have to spend as much on amplification. For example, an 89dB speaker may require a
        100W amplifier for a satisfying listening volume for a certain listener. All things being
        equal except the speaker's sensitivity (room size, speaker placement, etc.), the more
        sensitive 92dB loudspeaker will require only 50W of amplifier power to play to the same
        volume level. Remember, for only a 3dB increase in volume you require a doubling of power.
        Higher sensitivity speakers need less power all other things being equal. 
        Here's a surprising observation -- not only are
        lower-powered amplifiers generally more affordable, they frequently sound more refined.
        Some manufacturers produce variations on a general design theme where the more powerful
        designs simply add output devices and power supply storage capacity to arrive at the
        higher wattage ratings. In some cases, you don't get better sound as you get more power,
        you just get more power. And in some cases, the added power actually comes at a price in
        "sweetness" and "purity" (although exceptions always exist). 
        The moral of this story: Considering a loudspeaker's
        sensitivity rating can not only save money (because less powerful amps will prove
        adequate), it might even help you buy a better-sounding amplifier. 
         
        
        
         |