subwoofer = red, 21sw115 9 c.f. ported cab, tuned to 22hz, 2nd order high pass at 20, 2nd order low pass at 225hz. 6db of eq added to low end.
midwoofer = blue, 15" high sensivity, in sealed chamber. 2nd order high pass at 200hz and low pass at 900hz.
compression driver = green, ba750(?), seos horn, net 2nd order high pass at 900hz.
black line is net (if I've done this right; more or less)
sensitivity as seen by the avr would be 106db 1w1m. that's 126db 1w1m with 100 watts.
max spl would be dictated by the amp on the subwoofer.
taking a similar approach to what paul did, and I'm not sure if I've done it completely correctly, it would seem that the benefits of MTM for floor reflections may not be as significant as I was thinking.
for a listener sitting 9 feet from the speaker and with the midrange at 3 feet off the floor, the difference distance from direct sound (9 feet) and that of the floor bounce sound (10.81 feet) is 1.8 feet. that is half a wavelength that is 3.6 feet long and corresponds to a cancellation at ~311 hz.
in the picture at 311hz, the midrange is not the only source. it turns out that is about the point where both the subwoofer and the midwoofer are contributing equally to the sound. what is arriving at the listener's ear will be a combination of the midwoofer direct sound, midwoofer floor bounce, subwoofer direct sound and subwoofer floor bounce. this implies that a total cancellation will not occur. it may be simply a small dip in the response.
for many systems, a lot of the cancellations occur from reflections where there is only one source playing on the speaker. but in the case of the big overlapping 3 way, there is a large region of overlap from about 250hz to 750hz where several sources are contributing to the sound. of course, for most of the region above that point the horn controls directivity (more or less).