Quote:
Originally posted by threers:
Soundhound,

I have been advised by some that the highpass and lowpass poles be staggered a half octave to ameliorate an "in phase notch". I am thinking that there are "real world" inefficiencies given the different drivers and electronics that results in less than a unity in the crossover octave. But I am also thinking that if the frequencies are staggered, there should be an additive amplitudinal change (a spike rather than a notch) in the doubled over region. What is your opinion on this?

I am just bouncing ideas at you before I order more frequency modules. I have not had a chance to measure the speaker output as a whole yet. I've only worked on individual drivers so far with truRTA.



With a 24 db/octave Linkwitz/Riley crossover network, the point between the low pass and the high pass sections is 6db down if both filter sections are at the same frequency. A 6db voltage dropoff, when summed, equals flat response through the crossover point. Assuming that the speaker drivers themselves physically have their voice coils on the same vertical plane, the frequency response in-room should be flat through the crossover region if the filter sections are at the same frequency.

With a Butterworth filter, the point between the crossover sections is only 3db down. When summed, this will yield an excess of energy through the crossover point. This might be what you are hearing about, but it is not an issue with the Linkwitz/Riley crossover network.

[This message has been edited by soundhound (edited July 28, 2004).]