bill@verdix.com (William Spencer) (06/27/91)
msc@alessia.dei.unipd.it (Matteo Santinato 228156): > while high frequencies are very directional. I assume that > the low frequency response of the M1 crossover unit is the > same for both the front and rear woofers and that the response > of both woofers is added in the room. On the contrary, the > tweeter response is not added because it is directional. The directivity of the woofer increases with frequency so you shouldn't assume that only the tweeter is directional. > In order to have a flat frequency response from the system, > assuming the same sensitivity from all of the drivers, I > must provide more drive to the tweeter than to the woofer. > Am I right? No. Consider what happens with a conventional speaker. At frequencies that are low enough to be non-directional there is "diffraction loss". The loss of direction means that the forward frequency response droops in the bass. The transition corresponds to a wavelength about equal to half the baffle width. So this needs to be compensated for. With the bidirect design this is not needed. There could be a more subtle effect that needs compensation however. The Mirage enclosure is shallow and wide so the wavelength of the sound coming around the sides from the back is long compared to the front-back delay. Now, the "same sensitivity from all the drivers" would be assuming the rated sensitivity of the woofer had no diffraction loss figured in. The latest issue of Speaker Builder includes on article on Ralph Gonzalez' DELAC speaker which uses a rear firing woofer to compensate for diffraction loss. This may be somewhat helpful information. In my opinion, adding drivers on the rear for the mid and high ranges is also desirable, even without the rear woofer (these rear drivers could be low level or full level like the Mirage). Otherwise, the reflected energy in the room is bass heavy due to high frequency directivity. However, a rear woofer doesn't necessarily increase the imbalance in the reflected sound. With forward-firing systems the baffle width should not be a small as possible without regard to diffraction loss, contrary to popular opinion. Choosing baffle characteristics to coincide with some convenient EQ method is better. There are many ways to provide this EQ. Bill Spencer