Stupid baffle question - Woofer beside tweeter, why not again?

[wigginjs]

I know that it is a commonly understood design rule not the place the woofer and tweeter horizontally beside each other in a two-way speaker, but why not again? Is it because it generally makes for a wide and thin baffle that requires more baffle-step compensation? What about in the case of an in-wall speaker, does it matter if the woofer is horizontally beside the tweeter? What is the theory behind this rule?

[cjd]

Uneven lobing - this affects power response more than anything, but also tilts the response to either the "inside" or the "outside" of a speaker pair.

Usable if you understand what it does.

[fbov]

Uneven lobing... Huh? (I'll assume the OP's not up on the jargon.)

If you remember your high school Physics, in wave motion, when you have two souces, you get an interference pattern around the sources. Some areas have no wave motion, others see twice what they should. This is sometimes described by it's appearance parallel to the sources as a "comb" pattern because it has non-zero spikes at consistent intervals, like a hair comb.

http://en.wikipedia.org/wiki/Interference

http://en.wikipedia.org/wiki/File:Two_sources_interference.gif

In a speaker, at the XO frequency, both woofer and tweeter are emitting sound at the same frequency. That results in an interference pattern the same as you'd see in a Physics class demonstration, or on the web. In the latter gif file, you'll note that there are green lines radiating away from the pair of sources. These are regions of zero amplitude - no sound at that frequency. Inbetween are areas of much higher sound level - the lobes. The patterns are predictable, and can be used, as Chris notes, if you understand what's happening. In general, however, one should try to avoid it in speaker design, as several other sources of lobing are unavoidable...

You see, any stereo pair of speakers is a worst case for interference, yet no one complains! That's a clue that our perception of sound is very complex!

Have fun,
Frank

[Kal Rubinson]

You see, any stereo pair of speakers is a worst case for interference, yet no one complains! That's a clue that our perception of sound is very complex!

Yes but you should also note that, in general, stereo pairs are not reproducing identical signals at the same time. This makes any interference less fixed in space and time.

Kal

[cjd]

Ok. We can simplify the problem. And it's not comb filtering specifically (though phase interaction still is the cause).

It doesn't sound the same off-axis on the tweeter side as it does on the mid-woofer side. The reason is because distance to ear changes which un-aligns phase - z-axis IS an important design parameter and a horizontal driver layout means this changes.

Comb issues are present in any multi-driver setup, though you can design around these (sort-of).

Also, a stereo pair (or even a mono pair) generally will NOT have the same kind of issue unless you wire them out of phase with each-other entirely. You can take two speakers (completed, in-box) and face them to each-other near together, wired out of phase, and end up with nearly no output to be heard relative to the input signal level. Spread them apart and that diminishes. Wavelength of distance very quickly gets well outside the problem area. Combine unequal signal and relative phase being identical and it just doesn't work that way.

C