Time alignment vs Phase alignment?

[Harrycr]

As title suggests, please can someone explain simple difference between the two.

 

[Harrycr]

Are we phase aligning the sub or time aligning?

 

[jtalden]

It is hard to know the exact intent when these terms are used. They are often used interchangeably with respect to crossover timing. There are often 2 or 3 favorable alignments at the listening position and one could contend that each of these are time and phase aligned.

A conventional crossover target alignment is the one that provides the closest phase tracking for the direct sound throughout the crossover range. It is both time and phase aligned. In the case of SW crossover timing, room effects often make this target unfavorable for SPL response at the listening position and thus we adjust the timing as needed for SPL favorability.

 

[Harrycr]

It is hard to know the exact intent when these terms are used. They are often used interchangeably with respect to crossover timing. There are often 2 or 3 favorable alignments at the listening position and one could contend that each of these are time and phase aligned.

A conventional crossover target alignment is the one that provides the closest phase tracking for the direct sound throughout the crossover range. It is both time and phase aligned. In the case of SW crossover timing, room effects often make this target unfavorable for SPL response at the listening position and thus we adjust the timing as needed for SPL favorability.

Thanks
So accurate time alignment should have good phase?
Time align first and then check phase tracking to fine tune?

 

[keantoken]

I suggest you use the spectrogram on Wavelet mode at lower freq resolution numbers (=higher time resolution) to see where the spectrum peak is in time and then adjust the ref time to match that to zero seconds (you may not be able to reach perfectly zero results). Then the phase will be pretty accurate. I have done this when measuring woofer and tweeter separately so that I can tune the crossover for equal acoustic distances.

 

[jtalden]

So accurate time alignment should have good phase?

Yes but I wince a little at the word 'accurate'.
Smooth SPL is demonstrably the most important characteristic of good bass response. Good phase tracking is required in order to provide generally good SPL support across the XO range that can be more effectively EQ'ed.

Time align first and then check phase tracking to fine tune?

Yes, that is my approach. I time align to the direct sound and then adjust timing as needed for favorable SPL at the LP position.

See here If interested, run the exercises to get an understanding of the process and then apply it to your situation.

There are all sorts of other DIY methods and numerous automated methods that lead to favorable results. Good SPL support being the main indicator. It is not necessary to use this method. Arguably, the only advantage to this method is that we can more easily find the direct sound timing. We can then understand why and how far we chose to deviated from that initial setting to provide the most favorable SPL result at the LP.

 

[Harrycr]

I suggest you use the spectrogram on Wavelet mode at lower freq resolution numbers (=higher time resolution) to see where the spectrum peak is in time and then adjust the ref time to match that to zero seconds (you may not be able to reach perfectly zero results). Then the phase will be pretty accurate. I have done this when measuring woofer and tweeter separately so that I can tune the crossover for equal acoustic distances.

Ok thanks, so same principle.
Did you measure at the MLP?

 

[Harrycr]

Yes but I wince a little at the word 'accurate'.
Smooth SPL is demonstrably the most important characteristic of good bass response. Good phase tracking is required in order to provide generally good SPL support across the XO range that can be more effectively EQ'ed.

Yes, that is my approach. I time align to the direct sound and then adjust timing as needed for favorable SPL at the LP position.

See here If interested, run the exercises to get an understanding of the process and then apply it to your situation.

There are all sorts of other DIY methods and numerous automated methods that lead to favorable results. Good SPL support being the main indicator. It is not necessary to use this method. Arguably, the only advantage to this method is that we can more easily find the direct sound timing. We can then understand why and how far we chose to deviated from that initial setting to provide the most favorable SPL result at the LP.

Thanks.
There is no "absolute" timing.
Ok so it's the "smoothest" Spl through the xover not the "most" Spl.

I will try the sample once I am able to download the latest update. Only downloads the REW icon (0 bytes) not the actual program.

On the subs (Velodyne DD) they have polarity as well as phase variable 0-180. Why both because isn't polarity basically 0 or 180 phase?

 

[jtalden]

There is no "absolute" timing.
Ok so it's the "smoothest" Spl through the xover not the "most" Spl.

For the final timing adjustment at the LP I typically choose the maximum SPL support across the XO range. Any setting with good SPL support is favorable however.

On the subs (Velodyne DD) they have polarity as well as phase variable 0-180. Why both because isn't polarity basically 0 or 180 phase?

I don't know for sure. I don't have a typical active SW to experiment with. The limited measurement data I have seen here suggests that the phase control operates differently, but there is little if any practical difference between using those 2 settings and arrive at similar SPL support. The phase control does provide another useful method to time/phase align the XO. It is also much easier to do for those that want to get up and running.

 

[Harrycr]

For the final timing adjustment at the LP I typically choose the maximum SPL support across the XO range. Any setting with good SPL support is favorable however.

I don't know for sure. I don't have a typical active SW to experiment with. The limited measurement data I have seen here suggests that the phase control operates differently, but there is little if any practical difference between using those 2 settings and arrive at similar SPL support. The phase control does provide another useful method to time/phase align the XO. It is also much easier to do for those that want to get up and running.

Thanks appreciated.

Right so with a normal xover of say 80hz how far do you typically go either side of 80hz for max spl?

 

[jtalden]

The minimum amount needed to get strong SPL support in hte XO range.

 

[Harrycr]

Right maybe 10hz either side?

Is it best to measure Left and right main seperate and the sub seperate with all xovers (eg. xover 80hz receiver, sub xover inactive) active?
Is there any need to measure left and right combined?

With automated Audyssey, YPAO etc. they say keep phase/polarity switch to 0° pre process. How can that work when maybe preferable might be 180°?

 

[jtalden]

Right maybe 10hz either side?

Hz?? No, we change the ms delay or the AVR distance not the XO frequency.

Is it best to measure Left and right main seperate and the sub seperate with all xovers (eg. xover 80hz receiver, sub xover inactive) active?

Yes, but we could measure FL+FR instead of measuring them separately. We would just not be able to analyze the impact on the channels separately within REW. We could instead make the change and measure FL+SW and FR+SW to see the effect on the individual channels.

Is there any need to measure left and right combined?

They are combined within REW in the method suggested in the other thread. There are lots of other methods that work just fine.

With automated Audyssey, YPAO etc. they say keep phase/polarity switch to 0° pre process. How can that work when maybe preferable might be 180°?

They adjust the timing by distance changes in the AVR. It can work just fine.

 

[DanDan]

I often wonder about such time alignment ventures, in Control Rooms, HT, and Venues. Surely it is listener position dependent? Also a 50Hz component does not achieve a full wavelength until 20mS has passed.

 

[jtalden]

We are really only trying to get good SPL support in the XO range in the listening area. It's easy to get too distracted in thinking of timing/phase as a quality factor. Those are optional measurement factors that can be used to achieve a favorable SPL. We can all agree that poor timing/phase causes a large SPL dip in the XO range that can be a noticeable sound quality problem. We can easily differ however on how carefully we need to adjust the timing/phase to obtain favorable SPL support. Interestingly, we can also differ on which one of the 2 or 3 favorable SPL solutions sounds the best for any given setup. Most listeners seem to be able to hear a clear difference.
The REW alignment tool method helps those that are interested in inspecting what is happening in the XO range in more detail in and deciding what adjustments are likely to be helpful. It also is the most reliable way to confirm which of 2 or 3 favorable SPL solutions is nearest to the conventional direct sound target alignment. It is not discounting any of the other setup methods available.

 

[DanDan]

Tx, great answer.

 

[Harrycr]

Hz?? No, we change the ms delay or the AVR distance not the XO frequency.

Yes, but we could measure FL+FR instead of measuring them separately. We would just not be able to analyze the impact on the channels separately within REW. We could instead make the change and measure FL+SW and FR+SW to see the effect on the individual channels.

They are combined within REW in the method suggested in the other thread. There are lots of other methods that work just fine.

They adjust the timing by distance changes in the AVR. It can work just fine.

Sorry the hz either side I meant for example 80hz xover, we should look at best spl through the 60-100hz region?

With AVRs they sometimes suggest to leave the phase/polarity set to 0° degrees and suggest a unusual figure for the delay/distance. Is this because of the phase has done a full cycle, or have I got it all wrong?

Is it best to time align the rear surrounds with the sub, and if so shouldn't the mic be at 90° position?

Thank you

 

[jtalden]

we should look at best spl through the 60-100hz region?

Yes but my rule of thumb would be ±1 octave (40-160 Hz). The measured XO range should be confirmed See here.

With AVRs they sometimes suggest to leave the phase/polarity set to 0° degrees and suggest a unusual figure for the delay/distance. Is this because of the phase has done a full cycle,

By 'unusual distance' I assume you mean one that does not match well with the physical measured distance. Yes, that may the reason as there is normally favorable SPL timing that falls roughly at 1/2, 1, etc., wavelength distances from the conventional target timing. If the SPL support is strong that is a favorable setting. If there is a large dip it may, or may not be a good setting. It may just be room effects that cannot be corrected via timing.

Is it best to time align the rear surrounds with the sub, and if so shouldn't the mic be at 90°

The mic orientation doesn't matter in the bass range. It only starts to make a difference above about 2 kHz.
Surround speakers can reasonably be timed to the SW. I have no strong opinion and have never seen an authoritative discussion/recommendation based on the various sound modes we may decide to use.

 

[Harrycr]

Yes but my rule of thumb would be ±1 octave (40-160 Hz). The measured XO range should be confirmed See here.

By 'unusual distance' I assume you mean one that does not match well with the physical measured distance. Yes, that may the reason as there is normally favorable SPL timing that falls roughly at 1/2, 1, etc., wavelength distances from the conventional target timing. If the SPL support is strong that is a favorable setting. If there is a large dip it may, or may not be a good setting. It may just be room effects that cannot be corrected via timing.

The mic orientation doesn't matter in the bass range. It only starts to make a difference above about 2 kHz.
Surround speakers can reasonably be timed to the SW. I have no strong opinion and have never seen an authoritative discussion/recommendation based on the various sound modes we may decide to use.

Thank you

What determines which speaker or sub the delay value is applied too?

In theory shouldn't the value be altered in the mains, because the mains sound is going to be faster than subwoofer sound?

What measurements are required for a dual mono subwoofer and mains time alignment?

With the single sub 80hz xover it is:
Left main only (with receiver 80hz xover set)
Right main only (with receiver 80hz xover set)
Sub only (with sub xover disabled, controlled by the receiver 80hz )
What I do is for the sub I turn off speaker A on the receiver that leaves the sub playing on its own, with only the receiver xover frequency in my case <80hz.

 

[jtalden]

What determines which speaker or sub the delay value is applied too?

It's the relative timing that is important. Adjust the one(s) that fits the setup situation.

In theory shouldn't the value be altered in the mains, because the mains sound is going to be faster than subwoofer sound?

It makes no difference. The speed of sound is similar for all frequencies.

What measurements are required for a dual mono subwoofer and mains time alignment?

With the single sub 80hz xover it is:
Left main only (with receiver 80hz xover set)
Right main only (with receiver 80hz xover set)
Sub only (with sub xover disabled, controlled by the receiver 80hz )
What I do is for the sub I turn off speaker A on the receiver that leaves the sub playing on its own, with only the receiver xover frequency in my case <80hz.

That works for the the method being discussed here.