Wacky dipole idea: use DSP to remove fequal

OllBoll

Member
2010-11-22 11:24 pm
Hello there, while I probably won't try this myself I thought it might lead to an interesting discussion =)

The problem this aims to solve is the fequal at the dipole where both waves are in phase, which is a problem when having a large and more efficient baffle.

So I sat there an thought hey, why can't you like use a dual driver sealed box, say a sphere with one driver in each end. Each driver has it's own channel. The magic would be that we use DSP to change the phase response on each driver such that there is no fequal.

If I'm not mistaken wouldn't that be possible by cascading an all-pass and inverse all-pass?
 

john k...

Member
2004-08-10 2:50 am
US
Hello there, while I probably won't try this myself I thought it might lead to an interesting discussion =)

The problem this aims to solve is the fequal at the dipole where both waves are in phase, which is a problem when having a large and more efficient baffle.

So I sat there an thought hey, why can't you like use a dual driver sealed box, say a sphere with one driver in each end. Each driver has it's own channel. The magic would be that we use DSP to change the phase response on each driver such that there is no fequal.

If I'm not mistaken wouldn't that be possible by cascading an all-pass and inverse all-pass?

Firs, both waves are not in phase at Fequal. There is a net 60 degree phase difference at Fequal. The driver are in phase at the dipole peak. Second, if you change the phase relationship you no longer have dipole response.
 

OllBoll

Member
2010-11-22 11:24 pm
Firs, both waves are not in phase at Fequal. There is a net 60 degree phase difference at Fequal. The driver are in phase at the dipole peak. Second, if you change the phase relationship you no longer have dipole response.

Hmm... I think my ideas got the better of me and you are right =)

So what if you modify it slightly that you use a common channel on both, or a normal single driver dipole and instead add a delayed signal peak at fequal, with phase fixed such that it is the inverse of the rear wave?