Originally Posted by brandonnash
How would I have a polar response mess? I don't know why that's the only reason I'm asking.
Acoustic dipoles have a nice cosine-alpha polar response (-3dB at 45 degrees, -6dB at 60 degrees, -12dB at 75 degrees, etc.) with 4.8dB of directivity through about .25 v/D Hz with v the speed of sound and D the path length separation which approximately equals baffle width on rectangular baffles.
Above that point the response broadens (reaching a 6dB on-axis peak at .5v/D Hz) and breaks down into a multi-lobed pattern approaching the dipole's first on axis null at v/D Hz.
At .17 v/D the dipole's output matches a monopole's at the same excursion.
Solving for a domestically friendly foot wide baffle we find this occurs at 192Hz.
The "simple" solution is to use a wider baffle, with a two foot wide baffle pushing that point down to 96Hz.
That produces a first dipole null at 565 Hz which is not the sort of thing you want to hear in the midrange.
The right solution is to use narrow baffles, boost the drivers' low ends electronically to get more bandwidth in fewer ways, and cross-them over to the next higher frequency driver before the polar response deteriorates.
You get more wiggle room where the drivers are close to baffle width and are becoming directional on their own.
You want to read Siegfried Linkwitz's web sitehttp://www.linkwitzlab.com/models.htm#Ahttp://www.linkwitzlab.com
(note the flower-shaped polar response at the first null) and John Krevosky'shttp://www.musicanddesign.com
where he puts the upper limit of dipole usability somewhere between d/w (d path length separation, w wave length) = .5 and d/w = 1.
It would seem that (without seeing an impedance graph) the response overall would be prett linear and placement or acoustic treatments should be key in defeating any dipole reflections from walls. Like I said, I don't know and this is just for trying out more than anything. I'm not looking for an end all speaker system, but if there is potential I'll further experiment.
I rank every open baffle speaker I've heard with an acoustically large baffle somewhere between bad and mediocre. As much as I like Nelson Pass (he's a great lecturer, really supportive of the DIY community, and does fine things with analog electronics) I include his big baffles (Bob?) in this category even when fitted with the $40,000 Feastrex field coil drivers run both full-range and crossed to dipole sub-woofers.
If you want to use full-range drivers, try narrow back-loaded horns. That can net subjectively great, natural sounding results (especially with the smaller drivers) and some bass.
If you want to build dipoles (I built/own Orions, have heard v4, and listened to other similar speakers - all have been "excellent" or better) accept acoustically narrow baffles, multiple drivers, and low-frequency equalization.
I speculate that my subjective impressions come from what's going on with polar response (narrow baffle = excellent, acoustically large = bad; same driver in a box without the dipole issues = better; hearing of other speakers with more uniform polar response)