Active Crossovers - A different perspective

[Guest srobak]

In the flat, 2-D drawings - what you guys say make sense. However - it only identifies a single plane. I'd like to see and better understand what the behaviour of the lobe is in 3-D space, when all axis are taken into account...

If what you say is true - then that pretty much throws pink noise sensitivity and response metering right out the window, as the effective lobe can be in a completely different place from amp to amp, or speaker to speaker, or even driver to driver...

[Deang]

then that pretty much throws pink noise sensitivity and response metering right out the window

mas tosses everything through the looking glass.

[Guest srobak]

Or at least through distorted fiber

[mas]

I don't have them handy for all of the passive crossovers, especially as actives with signal alignment offer greater flexibility.

But yes, depending upon the signal alignment the polar lobing and associated comb filtering can indeed be different. And the relationships can easily exist between any combination of the following (and thus present various orders of magnitude that one must address, be it in speaker design or room analysis and design.).

The typical sources of such error ( in an order of increasing ordes of magnitude) are due to phase relationships of the acoustical origins/physical offsets between.individual drivers, multiple speakers, and speakers and virtual sources such as difractive surfaces and room boundary reflections. (I would tend to not worry about the amp.)

And no, there is no 'one size fits all' model. It depends upon the topology and signal offsets and polar patterns of each conributing real and/or virtual source..

Attached is a disagram that shows the replationship for 2 ideal speakers oriented horizontally. (For a vertical orientation simply rotate the diagram 90 degrees - for a 45 degree orientation, rotate it 45 degrees.

[Coytee]

Go MAS go!!

(good thread Doc Who!!)

[]

[Guest srobak]

Ok - so that demonstrates dispersion pattern depending on distance between identical drivers playing the same sound(s)....

You are trying to say that two different drivers - whatever distance apart - when playing over-lapping frequencies and cancelling each other out (why would they do that exactly?) will actually tilt that dispersion pattern in relation to the pole? Instead of appearing as a couple of unified drivers doing the same thing at the same time (which theoretically should only make the center lobe larger)?

[DrWho]

Ok - so that demonstrates dispersion pattern depending on distance between identical drivers playing the same sound(s)....

You are trying to say that two different drivers - whatever distance apart - when playing over-lapping frequencies and cancelling each other out (why would they do that exactly?) will actually tilt that dispersion pattern in relation to the pole? Instead of appearing as a couple of unified drivers doing the same thing at the same time (which theoretically should only make the center lobe larger)?

Notice how the 3D pictures provided are almost like a bunch of vertical discs? If you rotate them 90 degrees so that the discs are horizontal instead of vertical, you will have the same thing my first pictures are showing, except my pictures are just a 2-D representation (a cutout from the side if you will).

The reason you get cancellation is because the phase from each source isn't the same at those positions. In this case, the phase is different because the distance to each driver is different. This article will describe phase cancellation, but isn't specifically talking about speakers:

http://www.glenbrook.k12.il.us/gbssci/phys/Class/waves/u10l3c.html