Impedance Phase Plots

[kelticwizard]

Came across this impedance phase plot on another thread:
http://www.speakerbuilder.net/web_fi...ct/erosmk2.htm

The speaker's impedance pahse plot, including the crossover region, seems to stay between -75º and -120º from 80 Hz all the way up to 20,000 Hz.

Is that a normal phase impedance plot?

[kelticwizard]

Here is the impedance plot.

[Svante]

Originally posted by kelticwizard

The speaker's impedance pahse plot, including the crossover region, seems to stay between -75º and -120º from 80 Hz all the way up to 20,000 Hz.

Is that a normal phase impedance plot?

Nope, this must be the *absolute value* of the impedance and I bet it is a bass-reflex system (dual peak in the bass).

[SY]

What makes you think that this is a plot of phase? I think the phase plot will be a Hilbert transform of that magnitude plot, since there no reason for this to be anything but minimum phase. The builder's claim of a 12.5 degree max phase angle should be easy to check by differentiating the magnitude plot with respect to frequency.

Svante, yes, there is a port in the mechanical drawing.

[sreten]

The plot is the magnitude of the impedance, phase is not shown.

It would be a very unusual phase plot.

Above 80Hz it is +/- 12.5 degrees according to the text.

It is a reflex so goes to 180 degrees in extreme bass.

sreten.

[kelticwizard]

Originally posted by SY
What makes you think that this is a plot of phase?

The reason I thought it was a phase plot was that degree markings are given on the right hand side.

[Ron E]

Phase goes through zero degrees at the peak of "macro" slopes It goes up as the slope of the impedance magnitude curve is increasing and down as the slope of the impedance curve is decreasing. The rate of rise or descent is dependent on the slope change of the impedance magnitude.

[SY]

Originally posted by kelticwizard


The reason I thought it was a phase plot was that degree markings are given on the right hand side. [/B]

Artifact of the software. It probably defaults to that format as "plot paper" capable of plotting two functions, and the guy who posted it didn't do whatever it takes to display the phase curve on the same plot. One tipoff is that there's always peaks in a magnitude plot, and these translate in the phase plot to something that looks like an S turned sideways, with the center of the S crossing zero degrees at the same frequency as the magnitude peak. No S curves crossing zero degrees, it ain't phase.

[kelticwizard]

Originally posted by Ron E
Phase goes through zero degrees at the peak of "macro" slopes It goes up as the slope of the impedance magnitude curve is increasing and down as the slope of the impedance curve is decreasing. The rate of rise or descent is dependent on the slope change of the impedance magnitude.

Are you saying that the impedance phase plot is predictable from a plot of the impedance magnitude?

[kelticwizard]

Originally posted by sreten

Above 80Hz it is +/- 12.5 degrees according to the text.

sreten.

Plus or minus 12 degrees is a very impressive phase response.

I might as well bring this question up.

The major differences between passive and active crossovers, as I understand it, are:

A) The DC resistance of the crossover components, mostly inductors, and their effect on the damping of the drivers;

B) Massive irregularities in phase response compared to active crossovers.

Well, A can be addressed with quality inductors. The cost should not be prohibitive in the 2,000 Hz frequency range. If the crossover was down around 200 or 300 Hz, that would be a different story.

B looks like it has largely been solved. Plus or minus 12 degrees is not that far away from zero.

If that claim is true about the impedance phase curve, would it be fair to say that this speaker, with it's passive crossover, should give overall response very, very close to that of an active crossover?

If not, where would the major differences lie?