Wow, that surprising. Is it the Rane AC22 or AC23? I think those have 0-2mS delay, which would be about 27 inches. You'd think there would be enough delay difference to cause audible frequency response differences. Strange that you aren't hearing any.
Observations of Zarathu could be correct, yet leading to wrong conclusions (not his fault).
I had the same experience with Rane AC23 x-over delay knobs - I heard no audible difference in any of their positions. While it's overall very good analog crossover, its non-working delay left me puzzling.
Other delays I have used to change relative timing difference aligning HF and midrange drivers in the same baffle always brought very pronounced changes in 'naturalness' as a result of timing incoherence/coherence.
Changes in this subjective quality between mid-bass and midrange were less pronounced as ear gets less sensitive in its localization capability when frequency goes down. Of course, perceived timbre had changed in both cases due different summing in overlap region.
For me there is usually only one delay value for the given ear height that sounds the sweetest for HF/MF alignment. Typically it also measures well as very clean impulse response. But it doesn't necessarily bring flat frequency response - the latter must be corrected independently of timing.
I'd assume that with high probability we can blame Rane x-over for causing disagreement in this thread.
My most interesting find was during experimentation with changes in timing in combination with steeper filter orders. With 8-th order Linkwitz–Riley symmetrical x-over I noticed that changes in timing weren't perceivable anymore. To be more specific - I was crossing over compression tweeter in 18Sound XT-1086 waveguide with 15" vintage driver in OB at about 2KHz.
From this observation I concluded that overlapping bandwidth got so narrow and attenuated enough that each band became perceived by hearing apparatus as a completely separate source, bringing out something like a point-source experience.
This may not hold true at higher x-over frequencies, closer distances between drivers or more overlapping directivity patterns, so do not take this as universally applicable approach. However, It would be interesting to know your experience. I encourage you to try if it a) works the same way for you, b) is repeatable with your different driver setups.
I had the same experience with Rane AC23 x-over delay knobs - I heard no audible difference in any of their positions. While it's overall very good analog crossover, its non-working delay left me puzzling.
Other delays I have used to change relative timing difference aligning HF and midrange drivers in the same baffle always brought very pronounced changes in 'naturalness' as a result of timing incoherence/coherence.
Changes in this subjective quality between mid-bass and midrange were less pronounced as ear gets less sensitive in its localization capability when frequency goes down. Of course, perceived timbre had changed in both cases due different summing in overlap region.
For me there is usually only one delay value for the given ear height that sounds the sweetest for HF/MF alignment. Typically it also measures well as very clean impulse response. But it doesn't necessarily bring flat frequency response - the latter must be corrected independently of timing.
I'd assume that with high probability we can blame Rane x-over for causing disagreement in this thread.
My most interesting find was during experimentation with changes in timing in combination with steeper filter orders. With 8-th order Linkwitz–Riley symmetrical x-over I noticed that changes in timing weren't perceivable anymore. To be more specific - I was crossing over compression tweeter in 18Sound XT-1086 waveguide with 15" vintage driver in OB at about 2KHz.
From this observation I concluded that overlapping bandwidth got so narrow and attenuated enough that each band became perceived by hearing apparatus as a completely separate source, bringing out something like a point-source experience.
This may not hold true at higher x-over frequencies, closer distances between drivers or more overlapping directivity patterns, so do not take this as universally applicable approach. However, It would be interesting to know your experience. I encourage you to try if it a) works the same way for you, b) is repeatable with your different driver setups.
Last edited: