Phase - the relationship between two waves:
0-120 = additions 12-240 = subtractions 240-360 = additions
- = +6dB 90= +3dB 180= Cancellation 270= +3dB
wavelength (m)= velocity/frequency wl=v/f wl=340/f
100hz wl= 340/100 = 17m
period (for one complete cycle) t(sec)=1/f x1000 for m/s
phase relationship for two sources, different arrivals, offsets etc in m
phase= (distance (m) x frequency (Hz) x 360) / 340 m/s
p= (m.f.360) / 340
phase as a time offset
p = (time (m/s) x frequency x 360) / 1000ms
p = (t x f x 360) /1000
comb filter response
periodicity (distance between cancellation or sums)
pr = 1/time offset pr= 1/o
filters butterworth and linkwitz reiley
butterworth is -3dB at crossover point, with 1, 2, 3 etc order filters lr is -6db at xover point with 2, 4, 6 etc order filters
filters effects 1st -6dB /oct - 45degress phase 2 -12dB - 90 deg 3 - -18db - 135 deg 4- - 24dB - 180deg
I must be taking care with the crossover filters I use, especially as the higher order ones have extreme transiet response characteristics, with ms delays in response. The issue of crossover choice wil take a bit more thinking about but in the case of a dancefloor system, best sound quality will be achieved with farily moderate slopes I think. The ringing effects arent as bad and the cancellation is not so much of an issue with non acoustic information. Its more to do with the tailing off of the slope on top box, and checking polarity as a response to this - mabye not an issue but needs more exploration.