zobel and xover impedance

Archived from groups: rec.audio.tech (More info?)

david h. wrote:
> if a zobel network is used for both tweeter and woofer would the
> impedance become zmin in the speaker's spec sheets?

Not necessarily.

> So the crossover should be designed with this impedance and not zn?

The corssover should be design for whatever the resulting impedance
ends up being at and around the crossover frequency.

The problem comes from the fact that the typical conjugate
compensation network that is to correct for the rising impedance
due to the voice coil inductance can't work perfectly. These
circuits are generally a single resistor/capacitor series circuit.
Their use assumes that the rising impedance is equivalent to
a fixed resistor in series with a fixed inductor, and it is most
decidely anything but. Both the resistance AND the inductance
are frequency dependent, so unless you can find equivalent
resistors and capacitors whose fundamental value are also
frequency dependent, then the best you can do is an approximate
compensation.

For example, the DC resistance of a driver might be 6 ohms. And
a woofer might be spec'ed with a voice coil inductance at 1 kHz
of 0.5 mH. But actually measuring the driver could reveal that
the resistive part of the impedance at your crossover of 3 kHz
might be more like 9 ohms, and the inductance might be only 0.3
mH. If you design your compensation for the DC resistance and
1 kHz inductance, it will give you the wrong results (too little
resistance, too much capacitance) to work well with your 3 kHz
design.

Generally, I design the cmpensator for the most resistive impedance
over the widest range possible, and then use whatever the equivalent
resistance that results for designing the crossover. That result is
seldom if ever the same as either the DC resistance or the minimum
impedance above resonance.