>
It sounds quite optimistic.
Not by a lot. Ceramic-magnet woofers, which don't want absolute max flux, routinely run 8,000 Gauss. Hot woofers 10,000 to 14,000 Gauss. Maximum-output compression drivers, like JBL 2440 (4" coil, 2" exit, too heavy to lift with one hand) do run 20,000 Gauss.
Note that a speaker gap is different from your ribon gaps. The 2440's magnet has an end-area about 7 square inches or 4,500 sq.mm. The coil gap is 4"/100mm in diameter and 1mm tall, or 100 sq.mm area. The magnet flux is concentrated 4,500/100 = 45:1. If you get 1,000 Gauss at the face of Alnico, you (seem to) get 45,000 Gauss in the gap. In fact half your flux will leak, and past ~10,000 Gauss ordinary steel saturates. You go with special steel (Lowther mentions some magic here) and keep the path area large until as close as possible to the gap (tapered pole-piece). Even so, you have to push excess magnetism through the pole against increasing saturation.
Using these obscene magnetic fields lets Lowther use a relatively "heavy" cone, compared to common designs of similar efficiency. Heavy lets them dope-up with wood-pulp or banana-peel or whatever magic stuff they use to take the papery sound out of their cones.
>
the Manger driver looks promising, it's maybe the most promising "full-range" ever:
Hmmmm. As they boast, the idea runs back to Rice and Kellog. Not the breakfast cereal, these are the guys who invented the practical cone loudspeaker. The root math is derived (the hard way) in
these papers.
>
Haven't seen off-axis measurement though.
Note that the paper (they are both the same, one cribs from the other) does not go into the VERY complex (by this analysis) response off-axis except by a general guess:
The time history of the sound pressure at positions off the symmetry axis can be obtained from the above equations only after lengthy numerical evaluations. Presumably, the gradual decay of the function Jo(kra) for increasing arguments causes a "rounding-off of the corners in the time history" for small distances from the symmetry axis, and a completely different time history for points further away. For the ideal case, this looks right. "Completely different time history" is maybe not the result we seek (though for indirect sound, time-history may be meaningless). However for the practical case of 7 inch round disk with felt stuck to it, the actual vibration pattern is far from a theoretical resistive radiator. Those Dopplers sure show very NON-uniform sound distribution at higher frequencies, though they are probably cleaner than paper cones. (Impossible to tell, because you can adjust interferometer scale to emphasize or hide actual amplitudes.)
I doubt their motor can give large amplitude; aside from a rather low-excursion motor and a diaphragm that would become non-linear at high amplitude, any more than around 0.050" excursion in the upper speech range will give large FM intermodulation. Since 0.050" and 7" does not give enough displacement for bass, we are back to a subwoofer.
This is also the objection to high levels from the FR125S. 6mm Xmax in a full-range is insane: the highs will be all IM mush. Note that they show multiples in array, and also a subwoofer. I reject arrays (even MTMs) for accurate monitoring (very useful in hi-fi and commercial sound). If you don't mind a crossover at ~150Hz, the FR125S plus a boomer may be interesting. But at 86dB SPL/W, they are not likely to blow you out of your seat.
>
~2k or even below (if you use the waveguide).
The MCM "waveguide" cited above
looks like a good exponential horn with 3KHz cut-off. If it is a perfect exponential flare, the diaphragm will gain loading below 8KHz but then UN-load below 3KHz. If you want to get to 2KHz, try a 90 degree cone with 1" base and at least 3" mouth on a baffle. A conical horn will not give as much boost ~3.5KHz as the exponential horn, but won't cut-off abruptly, and will give some increase of loading at 2KHz. A clever woodworker can cut a conical horn in a 1" piece of wood, either on a lathe or with clever jigs and a router with a V-bit.
>
main monitors based on the wide range driver...
It is shockingly hard to meet, much less beat, the old Westlake wooden horns. A JBL 2440 driver on a large diffraction horn can give fairly uniform directivity across a control room, 500Hz-9KHz, at VERY high levels. Compared to any cone, it is a much less colored system. It is also nothing like the systems everyday people will play your recordings on.
