All crossovers use capacitors and inductors to direct the correct band of frequencies to each drive unit in the speaker system.

Capacitors reject low frequency signals and pass high frequencies, where as Inductors do the opposite, thus rejecting high frequencies and passing low frequencies.
Thus taking a simple First Order Crossover, (capacitor in series with the tweeter and inductor in series with the bass unit),at a certain frequency the reactance of the capacitor will be equal to the impedance of the bass unit, this is the Crossover Frequency.

Obviously, an inductor in parallel with a tweeter will have a similar effect to a capacitor in series, and likewise a capacitor in parallel will have a similar effect as a series inductor for the bass unit. This enables Second, Third or Fourth Order crossover networks to be designed to give the higher rates of roll-off required by modern high quality designs. Other components, Like Resistors, are often included in the crossover network to compensate for differences in drive unit sensitivity, etc.
 

Hi-fi speakers are designed for correct distribution of power between the different drive units in the system. This assumes the normal power distribution across the frequency range that one might expect in music. This balance can be seriously affected by several factors:-

• Amplifier clipping.
• Indiscriminate use of a graphic equaliser, which can raise the power level in one or more frequency bands.
• Use of the "loudness control" at other than low volume.

Owing to the non-linear response of the ear, our perception of bass & treble notes is much reduced at low volume whilst we can still hear the mid frequencies, usually the speech frequencies. To improve the perceived sound quality at low volume many amplifiers are fitted with a correction control usual called "loudness control".

This boosts the bass and treble frequencies and should only be used at low volume, it is tempting to use this control to give extra bass for rock music at parties, etc. Be warned though, that this puts both treble and bass units at risk from amplifier clipping.
 

This is the most common cause of premature drive unit failure and is a result of excessive use of the volume control. Under normal use an amplifier will produce a flowing wave form.

However, if an amplifier is under powered, then the voltage available at the output stage is unable to maintain the "peaks" & "troughs" in the wave form and produces a square wave which is virtually equivalent to Direct Current rather than the Alternating Current of a normal audio signal. This increases the output from the amplifier , making 40w from your usual 20w amplifier! It is strange but true that you are more likely to damage your speakers with an under powered amp than one that may be too powerful!

Decibels are the units used for sound measurement. The Decibel (dB) is one tenth of a Bel, the unit originally named after Alexander Graham Bell.

A Decibel is about the least perceptible change in sound level.

Doubling the sound power is an increase of 3dB, but due to the non-linear way in which the ear perceives sound, an increase of 10dB seems to us to be a doubling of sound power.

DB(A) is a scale which is adjusted to allow for the fact that we are more sensitive to some frequencies than others.
Doubling the distance from a sound source reduces the sound level by a quarter - a decrease of 6dB (A).

The loudspeakers are the part of the sound system with the fewest specifications, but the greatest effect on the overall sound quality. Most of us can't always differentiate between amplifiers, but can immediately sense a change in the speakers, making this the ideal place to start upgrading.

The basic working parts of a speaker drive unit are the cone, it's suspension, the voice coil and the magnet assembly. When an electric current floes through a wire, it create a magneticfield around that wire, and is increased if the wire is coiled. If this coil of wire is located in an external magnetic field, provided by a magnet, the field of the coilinteracts with that of the magnet to apply force to the coil. An Alternating current causes the field to build up and collapse in response to the frequencies of the current. In a speaker this changing field interacts with the constant field of the magnet, causing the coil to move in response to to the current. As the voice coil moves in and out, it moves the cone which creates pressure waves in the air near the cone, these are heard as sound.

When selecting a speaker for music reproduction, smoothness of response is far more important than extended range alone. A unit with a smooth response of 100Hz to 10000Hz can reproduce music faithfully than one with a range of 50Hz to 15000Hz that has significant peaks within that range. For good performance above 10000Hz the cone must be light with a low mass of no greater than 5grams. However, light cones bend at low frequences, producing distortion. For good bass response the cone should be large to grab enough air to have good radiation resistance. A heavy cone has a lower frequency of resonance and a smoother response than a lighter one, but may have poorer transient response as well as a limited high range.

A speakers ability to handle short pulse without altering their duration is called its transient reponse. For good transient performance, a speaker must start to move almost immediately after receiving the amplifier's signal to do so, then stop promptly when the signal ends. A smooth frequency response is the first requirement for good transient response. A peaky response curve indicates muliple cone resonances, and each one can be started by any signal that contains the resonance frequency. Even the smoothest of speakers will have at least one resonance: the fundamental cone resonance. The prominence of this varies with the system Q; a high Q speaker has low magnetic damping and is prone to peaking at its frequency of resonance. The fundamental frequency of resonance for each speaker is determined by the mass of the cone and the compliance of its suspension. Large cones, having a greater mass, usually have a lower frequency of resonance than small cones. When this is measured on a speaker it is called the free-air resonance(fs).

The specifications of the loudspeaker drive unit that allow us to predict the frequency response of the unit in various enclosures are named after Neville Thiele and Dick Small, the Australian researchers who gave birth to modern loudspeaker theory in the 1960's.

The most important driber parameters are the Free-air resonance (fs),the total Q (Qts),and the Equivalent Volume of the Suspension (Vas). With these 3 parameters you can model the frequency response, phase response and group delay of the driver in various enclosures.

The Free-air Resonance (fs) is the resonant frequency of the driver in free air, that is, not in an enclosure and is generally about the lowest frequency at which the driver can be used.

The Total Q (Qts) of the driver is a measure of the sharpness of the driver's resonance. The Qts is calculated from the mechanical and electrical Q of the driver and indicates the shape of the frequency response curve when placed in an infinite baffle. The higher the Qts the more peaked the response at resonance. The lower the Qts the more round the knee of the response curve will be at the low frequency where it begins to fall off.

The Equivalent Voume of the Suspension (Vas) is that volume of air which would be as stiff as the suspension of the driver when compressed by a piston the same size as the driver's cone. The Vas is expressed as a volume of air, usually in litres or cubic feet.

Besides the Thiele-Small parameters, most manufacturers include a range of other specifications. These often include the magnet weight, voice coil diameter, impedance, sensitivity (SPL),all of which can help with your loudspeaker design.