Crossover Plugins

[Paul Henry]

The speed of sound through air is relatively slow so human hearing can quite easily hear the difference in arrival times between sources that are at different distances.

There are basically three types of speaker delay functions:

1/ Delaying your Hi, Mid, or Sub drivers so they line up with each other acoustically and keep everything neatly in phase.

2/ Delaying remote or "Delay" stacks so that the people in the back don't hear two different sound arrival times.

3/ Delaying your mains so they line up with the sound from the backline on stage.

[TomyN]

Delay some of your subs to create special coverage patterns

[Donnie Frank]

The speed of sound through air is relatively slow so human hearing can quite easily hear the difference in arrival times between sources that are at different distances.

There are basically three types of speaker delay functions:

1/ Delaying your Hi, Mid, or Sub drivers so they line up with each other acoustically and keep everything neatly in phase.

2/ Delaying remote or "Delay" stacks so that the people in the back don't hear two different sound arrival times.

3/ Delaying your mains so they line up with the sound from the backline on stage.

With all due respect, how does one time align a subwoofer?

Given: Sound travels at 1126 ft./sec.

With this in mind, doing the math, a 50 Hz sine wave is over 20' long (1126/50). 40 Hz is even longer! So where to we time align? At the beginning of the wave? The middle? The end? Which frequency? 50Hz? 60Hz? 100Hz?

I submit that subwoofer frequencies are so broad that there is no way to time align them. You could literally stack your subs 10' from your mains and nobody would hear the difference.

Thoughts?

[davidss1]

phase alignment is only an issue when u have more than one source creating the same frequency,and relative to the position designated for the 'alignment'. u would time align components @ the xover frequency to ensure phase alignment of the overlap.
the 10ft distance u refer to donny is basically one rotation/cycle of a wavelength @100hz (100hz=3.4m),, it may still give u a summation of the overlap frequencies, but the signal is one wavelength early/late. if u had a distance difference of 5-7ft, with no correction would be roughly 180deg relative phase difference giving you infinite cancellation @100hz

[TomyN]

Hi,



red: before delay

yellow: with delay

green: Verification of the delay (polarity inverted on sub).

Pictures taken from: http://www.take-sat.de/download/timealignmentE.pdf


Tomy

[LarryBfaderJockey]

With all due respect, how does one time align a subwoofer?

Given: Sound travels at 1126 ft./sec.

With this in mind, doing the math, a 50 Hz sine wave is over 20' long (1126/50). 40 Hz is even longer! So where to we time align? At the beginning of the wave? The middle? The end? Which frequency? 50Hz? 60Hz? 100Hz?

I submit that subwoofer frequencies are so broad that there is no way to time align them. You could literally stack your subs 10' from your mains and nobody would hear the difference.

Thoughts?

you need to not think of sound waves as a sine wave coming out of a speaker. That is an analogy or representation of the waveform. The sine wave is a visual mathmatical analog of a wave consisting of compression and rarefactions of particles. This is where we get the term "analog" from. There is no way to align individual waves or the begining or end or waves in the time domain. Time align is a bit of a bogus term coined by Don Davis which he later regreted and changed to saying signal alignment in time. You can't align time. Just things in reference to time.
Larry

[Andy Hamm]

You align the system with itself using pink noise or program material. What is important is that crossover frequencies that are produced by 2 different cabinets are aligned so that the common frequency range that both cabs produce align properly.

When you are out of alignment you can get either 'holes', peaks or other weirdness at the crossover points. Any broad band sound will be 'off' a bit and nothing short of physically aligning or delaying a set of cabinets will fix it.

It's one of the biggest problems with a multipoint system with a few factors involved. The physical location of the drivers and phasing issues caused by passive crossovers or poorly designed active ones.

We used to have these little XLR devices that would give you a click and a pink noise burst that we used during line checks. They were great as it made it easy to tell if the system was out of alignment with the noise burst. I haven't seen these in quite a while, but it would be easy enough to just make a wav file to do the same thing.

To make a gross example, imagine if the boom and click of your kick drum were out of time, with the boom arriving sooner than the click. If you put your subs way out in front of your tops, this is what would happen.

With all due respect, how does one time align a subwoofer?

Given: Sound travels at 1126 ft./sec.

With this in mind, doing the math, a 50 Hz sine wave is over 20' long (1126/50). 40 Hz is even longer! So where to we time align? At the beginning of the wave? The middle? The end? Which frequency? 50Hz? 60Hz? 100Hz?

I submit that subwoofer frequencies are so broad that there is no way to time align them. You could literally stack your subs 10' from your mains and nobody would hear the difference.

Thoughts?

[Naturally Digital]

You align the system with itself using pink noise or program material. What is important is that crossover frequencies that are produced by 2 different cabinets are aligned so that the common frequency range that both cabs produce align properly.

When you are out of alignment you can get either 'holes', peaks or other weirdness at the crossover points. Any broad band sound will be 'off' a bit and nothing short of physically aligning or delaying a set of cabinets will fix it.

It's one of the biggest problems with a multipoint system with a few factors involved. The physical location of the drivers and phasing issues caused by passive crossovers or poorly designed active ones.

We used to have these little XLR devices that would give you a click and a pink noise burst that we used during line checks. They were great as it made it easy to tell if the system was out of alignment with the noise burst. I haven't seen these in quite a while, but it would be easy enough to just make a wav file to do the same thing.

To make a gross example, imagine if the boom and click of your kick drum were out of time, with the boom arriving sooner than the click. If you put your subs way out in front of your tops, this is what would happen.

I think it's worth pointing out that any 'alignment' is only effective for a particular point in space. When you align a system, you are really doing so for the listening area (averaging out the alignment) or in some rare cases for a specific sweet spot.

In Andy's example above, the boom would be arriving ahead of the click, but that would be the case *out front* of the system. To compensate, you would delay your subs to align with the mains (assuming you couldn't relocate the subs).

If it were theoretically possible to spin the system around and listen from the 'other side' ie. behind the system, 180 degrees from your out front position, the opposite would be the case where the click would obviously arrive before the boom because the mains would be out front of the subs.

The point is, when you align a system, you are doing it for a particular listening point in 3D space. You may in fact be throwing the system out of alignment at some other point in space (on stage for example)...

One other point to add to Andy's comment is that all filters have some degree of phase shift inherently, whether passive or active, unless they are linear phase filters.

When you are tuning a system the crossover filters interact with the drivers' phase shift through the crossover region and together with the physical relationship, determine how it will sum with the driver it's being crossed over to. Sometimes things align nicely on a first attempt, other times all the various phase shifts interact causing the systems person to use delays to compensate or tweak crossover points, slopes etc.

[TomyN]

Well, linear Phase filter just add 'Pure Delay' so also causing a change in arrival time.
The main use for time adjustment is matching the sub to the tops. Here we have 'long' waves, so that smaller changes in the difference in time do not matter.

In the pre - pc area I was told to use a sine wave with the x-over frequency, invert the polarity of one of the speakers and adjust the delay for minimum audible amplitude. But this would also happen if the difference is just one (or a multiple of ) a wave-length.

So time-adjustment is a area where the current measurement technology shows great advantages.

Tomy