Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Nathan Riddle]

What is: "Phase-aligned overlap"

Clark also applied a personal engineering touch to the flown arrays. “One of the techniques that I’ve pioneered is what I call a ‘phase-aligned overlap,’ which ends up being a dynamic crossover point between the mains and the subs,” he explained. “What that does is create an overlapped set of frequencies that the mains and the subs work together in. As long as they’re phase-aligned in that overlapped set of frequencies, it doesn’t matter the relative level between the mains and the subs—wherever they are matched in level, it’s phase aligned.”

-http://www.soundandcommunications.com/array-of-hope/

Beyond my desire to throw up [trying to get over that] at the mention of Bose (and yes I've read the threads on their professional products being actually 'professional).

Isn't that what a normal crossover/phase alignment with subs/mains do?

[Will Knight]

Good question.  If it's "aligned", there shouldn't be any overlap.  I would imagine it's "not-aligned" if there is any overlap.   

[Nathan Riddle]

Good question.  If it's "aligned", there shouldn't be any overlap.  I would imagine it's "not-aligned" if there is any overlap.   

Perhaps my semantics is wrong. But I'm thinking the frequency response of both the sub & main isn't a hard wall cuttoff at the xover point which is why he says "overlapped set of frequencies that the mains and the subs work together in."

He then says "As long as they’re phase-aligned in that overlapped set of frequencies, it doesn’t matter the relative level between the mains and the subs" which is correct given my understanding of phase alignment, the level/magnitude of the frequency doesn't affect the phase alignment only the relative phase response can effect the frequency response.

But then he says: "wherever they are matched in level, it’s phase aligned.” I think I get what he's trying to say, but that is flat out wrong. Just because level (magnitude) is the same doesn't mean something is in phase (relative).

Lastly, I'm really confused about the "dynamic crossover point" he refers to. Is there an actual concept for this. The only 'dynamic' xover I know of is improperly done aux subs...? Maybe dynamic xover based upon input level?

[Mac Kerr]

Good question.  If it's "aligned", there shouldn't be any overlap.  I would imagine it's "not-aligned" if there is any overlap.   

How do you have adjacent pass bands that do not overlap?

Mac

[Mark Wilkinson]

Perhaps my semantics is wrong. But I'm thinking the frequency response of both the sub & main isn't a hard wall cuttoff at the xover point which is why he says "overlapped set of frequencies that the mains and the subs work together in."

He then says "As long as they’re phase-aligned in that overlapped set of frequencies, it doesn’t matter the relative level between the mains and the subs" which is correct given my understanding of phase alignment, the level/magnitude of the frequency doesn't affect the phase alignment only the relative phase response can effect the frequency response.

But then he says: "wherever they are matched in level, it’s phase aligned.” I think I get what he's trying to say, but that is flat out wrong. Just because level (magnitude) is the same doesn't mean something is in phase (relative).

Lastly, I'm really confused about the "dynamic crossover point" he refers to. Is there an actual concept for this. The only 'dynamic' xover I know of is improperly done aux subs...? Maybe dynamic xover based upon input level?

I'm guessing the following is the quote you're referring to....yes?

 “One of the techniques that I’ve pioneered is what I call a ‘phase-aligned overlap,’ which ends up being a dynamic crossover point between the mains and the subs,” he explained. “What that does is create an overlapped set of frequencies that the mains and the subs work together in. As long as they’re phase-aligned in that overlapped set of frequencies, it doesn’t matter the relative level between the mains and the subs—wherever they are matched in level, it’s phase aligned.”

My take is that he's just making some of his own definitions.....

and that his idea of a 'dynamic crossover point' might be better described as a 'dynamic crossover range'.
A range of frequencies where phase traces overlap,  within which the x-over freq may be moved up or down, without effecting phase.

And I think his phrase "wherever they are matched in level, it's phase aligned" is meant to say "whatever the relative levels are, subs and mains are still phase aligned"

If my interpretations are correct, I have to agree with him......as long as the phase overlap extends far enough to accommodate summation through the critical region.

This have been a really neat finding for me recently. 
I put a fader on each driver on a 4-way system, to be able to control each driver's level independently.
With the 4-way system fully phase aligned, I was surprised to find I could move  any of the sliders quite a bit, and not change the phase curve.

Best damn eq I've ever found or heard.  Subs span about 2 octaves, Mid about 3 octaves, HF about 3, and VHF about 2.
Sounds so natural moving them up or down, and much easier to dial in than using low and high shelving normally used for system voicing.
I almost can't find a recording that can't be made to sound balanced tonally in pretty short order.
Only caveat I've found so far is use of faders...

[Phil Graham]

What is: "Phase-aligned overlap"

-http://www.soundandcommunications.com/array-of-hope/

Beyond my desire to throw up [trying to get over that] at the mention of Bose (and yes I've read the threads on their professional products being actually 'professional).

Isn't that what a normal crossover/phase alignment with subs/mains do?

Nathan,

Your intuition is correct, that this is "normal" behavior, as there will always be some overlap between bandpasses, and the system should (presuming you're not trying to specifically modify the polar response) aligned phase traces over the overlap region as much as possible.

Here's how I implement this when I'm doing system design and tuning:

• To insure psychacoustically pleasing localization, I aim to have response to 60Hz capable at each main loudspeaker array. This could be from the primary box, or it could be the box plus subwoofer(s) adjacent to mains.
• The filters between these subs and mains are usually 2nd order, so there's overlap for several octaves. I aim to have each individual box to have smooth response through the main bandpass AND the overlap region. typically that means subs are well behaved up to 300hz.
• I don't use the crossovers to shape any of the overall response (e.g. bass hump). The aim is for XOs + delay + allpasses to get the best overlapping phase response between the mains and subs over the relevant range.
• Any tonal shaping eq is done on the "input" side, before the crossover filters.

[Jean-Pierre Coetzee]

Maybe I'm confused but I though this was always the aim with crossovers whether the overlap is 5 octaves or part of the octave.

Our Meyer system has almost always been used without crossovers and its perfectly fine because the boxes have the same phase trace through the range that they output, this should be a goal for any system design to be honest regardless of manufacturer.

Sent from my 2014817 using Tapatalk

[Mac Kerr]

Our Meyer system has almost always been used without crossovers and its perfectly fine because the boxes have the same phase trace through the range that they output, this should be a goal for any system design to be honest regardless of manufacturer.

What Meyer system would this be that does not use crossovers?

Mac

[Mark Wilkinson]

What Meyer system would this be that does not use crossovers?

Mac

Ultra series and their associated subs, yes ?....  I'm thinking subs to mains is the gist of the x-over discussion....

[Mark Wilkinson]

Maybe I'm confused but I though this was always the aim with crossovers whether the overlap is 5 octaves or part of the octave.

I agree, 1/10 octave to all 10 octaves..... there is either phase overlap or there isn't.