After building a run of nice two-ways in the early 90s I decided to build a three-way of higher performance. After LOTS of work I had ONE mono prototype (8" Peerless +5+0.75" SEAS) voiced really well. After learning that subtle crossover value shifts that I found important measured only 1/3 dB over an octave and a half in the upper mids, and that I could NEVER affordably buy driver pairs that well matched I gave up! Driver manufacturers often custom-match runs or pairs for $$$ to high end speaker producers, as getting cloned response in the midrange and treble is critical. Snell helped pioneer tight driver response QA, and I imagine the Brits too, back in the LS3/5 days for the BBC. After listening to the staging ablity of stereo pairs where the manufacturer matches and catalogs all drivers to a 0.5dB window (my Parsifal Encores, for example), I'm pretty sure I'll never reenter the arena. Even Boston Acoustics uses reasonably-sophisticated QA for its own tweeter production. As driver manufacturers routinely sell off the "outliers" for the DIY market, getting a matched pair of ANYTHING becomes very tough. The 2-3dB sens envelope spec'ed by a tweeter manufacturer can be 5-10 just-noticeable-differences in crossover-tweaking in the lower treble. Too much work to make matched pairs for the little guy...or at least ME!
