here is screenshot from a 32xDMA45 array 60" tall. 10 minutes work cutting an old sim down to size
It shows the vertical dispersion issue. would need shading for seated+standing listening
those 32 1.5" drivers are definitely NOT good down to 30 Hz; didn't check...
It shows the vertical dispersion issue. would need shading for seated+standing listening
those 32 1.5" drivers are definitely NOT good down to 30 Hz; didn't check...
At an 8' listening distance, the angle between sitting height (about 40") versus standing height (about 72") is on the order of 20 degrees. It's hard to see the response along the 20-degree line on the Vituix model, but clearly, there is going to be a noticeable difference in frequency response between sitting and standing at that distance. For a shorter 46" array, the difference will be more pronounced.
Shading results in much a smoother response along the vertical axis, but it doesn't seem to do that much to improve the overall loss of high frequencies at 20 degrees off axis. However, curvature makes a very profound difference. Curvature is hard to implement, but it has benefits. The 4 charts below are for a 44" array with 25 3/4" drivers spaced 1" from each other (1.75" CTC)
1) upper left: no shading
2), upper right: Legendre shading
3) lower left: curvature (50 degree of cap angle, per Keele, but I'm not convinced the model is applying that amount of curvature)
4) lower right: curvature + shading
If you can implement curvature and some shading, you would just be left with the 3dB/octave roll-off, which should be fairly easy to deal with.
Caveat: I think this model calculates the response with curvature correctly, but I'm not convinced that the model is applying the "right amount" of curvature according to the pull-down menu.
Shading results in much a smoother response along the vertical axis, but it doesn't seem to do that much to improve the overall loss of high frequencies at 20 degrees off axis. However, curvature makes a very profound difference. Curvature is hard to implement, but it has benefits. The 4 charts below are for a 44" array with 25 3/4" drivers spaced 1" from each other (1.75" CTC)
1) upper left: no shading
2), upper right: Legendre shading
3) lower left: curvature (50 degree of cap angle, per Keele, but I'm not convinced the model is applying that amount of curvature)
4) lower right: curvature + shading
If you can implement curvature and some shading, you would just be left with the 3dB/octave roll-off, which should be fairly easy to deal with.
Caveat: I think this model calculates the response with curvature correctly, but I'm not convinced that the model is applying the "right amount" of curvature according to the pull-down menu.