It is important to know what load the amp will really see when the system is operating, and this is the load. The amp’s minimum safe load impedance is a concern. When paralleling loudspeakers, you should know what the net resulting impedance will really be.
The maximum phase angle the amp can tolerate can be important, too. The phase angle is an indication of the lag time between the current and voltage when putting energy into the loudspeaker system, and also an indication of the amount of energy coming back from the loudspeaker system.
The reactive part of loudspeaker impedance is an indication of energy storage in the loudspeaker. Luckily, loudspeakers aren’t terribly good at it, rather like the batteries in my laptop, and the energy comes back almost right away.
You probably know that there is resistance to moving a cone when connected to an amp, and you may have noticed that the voltmeter will swing back and forth when you push on the cone while measuring impedance. No surprise here, as a loudspeaker is just another form of motor or generator – they all involve a magnet assembly and a coil of wire, and as such create a voltage when the coil and magnet are moved relative to each other.
The amp has to absorb this energy-return created by the loudspeaker cone returning to its rest position, while trying to put new energy into the loudspeaker. Some amps take exception to this, and although most professional amp designs are by now quite well behaved, some amps from years gone by will generate spectacular fireworks if presented with reactive loads.
High-frequency devices are sensitive to the horns they are mounted on, and the impedance reflects this. For the most part, the differences are less dramatic than LF devices. See Figure 3 for the impedance curves and frequency responses of the same driver mounted on two different horns. Note the behavior out of band in the 200 Hz region. This is important when designing passively crossed-over loudspeaker systems, but thankfully much less so in active systems.
The point of this speedy review of loudspeaker impedance is that impedance is a dynamic thing and definitely frequency variable. It is important that you know what your amp will be seeing, and to this end it is important that you measure the real net load that the amp will see, preferably at the amp end of the cable, to take everything into account.
Such devices as the Techron TEF or MLSSA analyzer can show the phase response of the impedance curves, and display it neatly on the Nyquist diagram, allowing a quick assessment of the reactive character of the load. Measure well beyond the expected operating bandwidth. Strange impedance behavior out of band can still be a problem if there are any spurious signals that happen to infiltrate your signal chain.
Keep in mind, too, that crossovers are not “off” out of band; they are attenuated 18-, 24- or 48 dB/octave, but large signals out of band can still put amplifiers and loudspeakers at risk in nasty situations.
Additional Effects
There are other concerns as well. Cable capacitance is not generally a big problem, but that doesn’t mean it can be ignored. I saw one installation that used cables provided by the original manufacturer of processor-type sound reinforcement loudspeaker systems. The cables were nice, flexible, small diameter, 4-conductor designs that would put the amps into protection without cause.
If the cables were long enough, it wouldn’t even need a signal to create problems – just turning up the amp volume control was enough to do it. In this case it turned out the specific amp model had an inherent design flaw in grounding that caused greater instability, and the manufacturer was happy to repair or update the units under warranty, but problems such as this were initially caused by high capacitance cable. Excessive capacitive reactance can be seen when doing impedance measurements.