1. Impedance and resistance
If we measure a speaker with a multimeter it will give us a different reading, usually lower, than the nominal impedance of the speaker. For example, an 8-ohm loudspeaker will give us a reading of 6 ohm. The reason for these differences is that the multimeter measures resistance, not impedance. Resistance is the opposition to the passage of direct current (DC) and has a single value, while impedance is the opposition to the passage of alternating current, so it is a function of frequency and hence changes with frequency. Therefore, the current delivered by the amplifier to a speaker will be higher at those frequencies where its impedance is lower. Similarly, the current delivered by the amplifier to a speaker will be lower at those frequencies where the impedance is higher.The multimeter is valid for measuring resistors used in passive filters or electronic circuits, since its impedance does not vary with frequency. However, it is not valid for a loudspeaker, since its impedance varies with frequency.
We could say that, in a way, the resistance is the impedance for a frequency of 0 Hz, since 0 Hz corresponds to direct current.
Like a frequency response, an impedance curve has magnitude and phase. The imaginary component of impedance is called reactance. We will avoid complicating this document too much and focus on magnitude. A loudspeaker also has mechanical impedance, but that is a completely different subject.
In the figure we can see the magnitude of an electrical impedance curve of a cone loudspeaker in free air (red curve) and also the curve of a passive two-way system with a bass-reflex type enclosure. Both would have a nominal impedance of 8 ohms. The straight green line represents an 8-ohm resistance. We can see how the impedance varies with frequency, and how it can fall below the nominal impedance at certain frequencies. In the case of the cone speaker with no enclosure (red line), the impedance drops to 6 ohms at 200 Hz.
Also note that, in a practical sound system, the cable also adds its own impedance, which can be significant if the cable is long and/or thin.
2. Impedance measurement
To measure the electrical impedance curve, we need a laboratory analyzer that allows us to do so. These can be either swept sine (they measure the impedance at all desired frequencies with a swept sinusoidal signal) or use noise signal (in which case they measure the spectrum at once).There are also portable impedance meters for installers. These incorporate a frequency generator, usually at 1 KHz, sometimes at more frequencies as well, which allows an impedance reading at those particular frequencies. If the manufacturer gives us the impedance value at that frequency, or we look at it in the impedance curve, we can check if there are irregularities in the speaker line, comparing the value we should get with the one given by the meter.
If we just have a basic electrician-type multimeter, and assuming we are not using transformer lines, we can still check that the loudspeaker impedance is roughly correct, as typically we'll measure a (DC) resistance value that is slightly lower that the nominal impedance (plus the cable, if we are not measuring directly at the speaker terminals).
3. Nominal impedance
Since an impedance curve is not practical for day to day use, manufacturer specify a nominal impedance for their loudspeakers. This is usually given as powers of two, with 2, 4, 8 and 16 ohms being the most common values for sound reinforcement. Some standards specify a percentage that the ratio to the nominal impedance is allowed to fall to the minimum impedance, which means that even if the manufacturer correctly specifies a loudspeaker, different loudspeakers with the same nominal impedance may show substantially different loads for an amplifier, since, for example, a speaker with a 8-ohm nominal impedance could have a minimum impedance of 8 ohms, 6.5 ohms or even as low as 5.5 ohms (the latter would not meet the usual standards, but is not uncommon), so that means very different output power levels from the amplifier (and therefore also different probabilities of thermal protection being triggered when multiple speakers are used).4. Parameters extracted from impedance curves
There are many parameters traditionally calculated using impedance curves. For example, Thiele-Small parameters or low signal parameters - which are used for the design of enclosures - are usually extracted from impedance curves. These days non-electrical means such as lasers are utilized to measure transducer design parameters.The most basic parameter that can be extracted from an cone speaker in free air (red curve) is the resonance frequency (Fs), i.e. the frequency at which the impedance peak occurs. In our loudspeaker this frequency is 34 Hz.
The blue curve shows the impedance response of a two-way passive box with bass reflex enclosure. From it we can calculate the tuning frequency of the box (Fb), which corresponds to the valley between the two bass peaks. In this case, the box is tuned to 50 Hz.
5. Impedance varies with temperature
One aspect that should not be forgotten with regard to impedance is that it varies with temperature. This means that when the amplifier delivers high power levels to a speaker and its voice coil gets hot, impedance increases considerably. This means that the amplifier will deliver less power, a phenomenon called "power compression", which results in a reduction in sound pressure, which is greater the more the coils overheat. This could be thought of as a way of naturally protecting the voice coils from damage, but it reduces the output level, which is why transducer manufacturers prefer to employ a variety of cooling techniques to evacuate heat from voice coils.Related »» Series and parallel impedances