Importance of phase and timing optimization when incorporating sub woofers ?


For incorporating a pair of sub woofers into a music system, how important is it to be able to optimize phase ( continuous adjustments and not just 0 or 180) and timing ( optimizing delay ) ? Reading on this subject, some say to get the best seamless integration, the two above factors are essential. REL is frequently mentioned as the most musical/fast/detailed sub woofers for music, however they only allow of 0 or 180 deg phase and no other room correction, as compared to Velodyne DD’s or JL Audio’s. I could utilize something like the JL Audio CR-1 crossover, but adding an additional piece of equipment into the audio chain wouldn’t be my first choice. On the other hand, the added benefit, might make the addition of a well made crossover a net plus ??? Regardless I plan on having my room/system measured to assist with incorporating what ever two subs I go with.
Thoughts from experienced users?
jim94025
Yes, it is very important. In this sense, phase adjustment = delay adjustment and is something you can easily do with a miniDSP unit. Highly recommended.

As you point out, EQ is also important. As is room acoustics, and setting the crossover frequency and slope well.

With no room acoustics, setting the proper delay for correct phase matching plus being able to trim off room modes is still highly desirable.

https://www.minidsp.com/products/minidsp-in-a-box/minidsp-2x4-hd


Best,

E
No it could not possibly matter less.  

This is a hard one for people to grasp, probably because this is so critically important to getting good sound in the midbass, midrange and treble. But the physics and psychoacoustics of really low bass are so much different its important to understand. There’s a lot of guys who should know this but it seems to be one of the hardest lessons in all of audio, as you can already tell. If you really do want to understand there’s great info by Tim and Duke and me, and you can read Geddes, et al for more fundamental research. But I will explain briefly here.

First off, tests show human beings can’t even hear low bass of less than one full wave. The waves at 20-40Hz are as big or bigger than your room. So that right there should make it obvious there’s no point trying to optimize timing.
The fundamental problem with low bass is it bounces around the room creating zones of high and low pressure called modes. Room dimensions and where you put the sub in the room determines where these high pressure zones or modes will be. Moving the sub around will only move the modes around. Changing or shifting the phase is hardly any different than moving the sub. EQ can flatten the volume difference but its important to understand this happens across the whole room, when in reality its only the loud mode area where you want it less.

Two subs helps but not the way most people think. Two subs helps if they go in different locations. Because the more modes the smaller they are and the smoother the bass will be. For this reason four is even better than two.

If you actually do this, as I have, you will find that with 4 (or 5) subs it hardly matters where they go, the bass will be excellent. Superb. Better than you ever heard. In your life. Anywhere. Read the Swarm/DBA threads. You will see.
At low frequencies, speakers + room = a "minimum phase" system. In English, this means that the frequency response tracks the time-domain response. 

So if you have an in-room bass peak in the subwoofer region, you will also have modal ringing at that frequency - the energy will there will take longer to decay into inaudibility. (It is the frequency response peak that we actually hear; the ear as very poor time-domain resolution at low frequencies but arguably has heightened resolution in the loudness domain.) The good news is that when you fix one (by EQ or bass trapping or some other technique), you simultaneously have fixed the other.

Therefore, in my opinion, the phase adjustments on a subwoofer amp (and/or timing adjustments within a bass management system) are tools to be used with the focus on getting the frequency response smooth. Smooth bass = "fast" bass. In other words, don’t sacrifice frequency response in pursuit of what you think would be better phase response. In the bass region when the frequency response is right, the time-domain response (which includes the phase response) is also right. 

Duke
No it could not possibly matter less.  


Man who constantly argues against science completely misunderstands the problem, writes nonsense in long form.
JL Audio has a lengthy white paper on this subject; interesting read. Check out their website for another viewpoint, and one where the author does not insult the audience.
Adjusting phase will smooth the FR around the crossover frequency. The effect is noticeable by ear and it is measurable. (It is best done with a measurement mic.) Without such adjustments, one may have large dips in the FR due to cancellations.

As Eric said, adjusting delay and phase are essentially the same thing.

I used to think and understand phase as delay in a sense but have come to understand that delay is not quite exactly the same thing as adjusting phase. Yes, by adjusting phase you will "match up" two overlapping sources in the crossover region. However, you may be as much as a full wavelength or more BEHIND in timing. So the impulse response and initial attack will be blurred.

That said, I don't understand why many of these EQ and sub management systems that allow you to adjust from your seat remotely do not have a continuously adjustable phase option to adjust from there as well. Nor do the auto EQ room correction devices seem to adjust phase automatically...

Adjusting delay precisely (which will match phase) makes a big difference in my experience. Adjusting only phase isn't equivalent: your subwoofer could be 100 feet away and you can match phase while playing a sine wave but it will be easy to tell the difference in timing.
it seems that subs are generally delayed by positioning distance differences. Add even more delay if they have any servo processing/EQ being done. Seems that there should be the option to delay the mains to adjust phase precisely - rather than subs in some instrances. Do any room correction systems do so utilizing impulse responses and delay mains or subs as needed to match phase?
+1 nekoaudio

@snbeall - In an ideal world, you are right. I mean, DSP itself adds processing time, so even if you originally had subs at the same distance acoustically as your mains, just putting a processor in the bass path introduces 1-2 mSecs of delay (i’m guessing).

In practice though, I find that matching the phase and slopes correctly using the sub alone can be really really good sounding. This keeps your mains "pure" and your sub disappears.
I’ve never seen it mentioned but here’s a thought.  If you're in the phase doesn’t matter camp could you use a variable phase control to nudge your nodes around until they work for you?

@snbeall the miniDSP SHD products allow you to adjust delay for individual channels, so you can adjust main speakers and subwoofers independently. You can use Room EQ Wizard to compute the exact delay value to use.

Trinnov products support this as well, and will perform the exact measurements for you.

Devialet's Expert Pro systems also support delaying the mains or subwoofer output by an exact amount, and Room EQ Wizard can also be used here to compute the delay value.
I started a thread a while back with similar concerns, and opinions were all over the place, as it seems they are here.

Most opinions were against the notion that phase and time were important when setting up a sub, or multiple subs for 2-channel music.
All I can say is that I’ve had the opposite experience using subs with my Dunlavy speakers. I’m guessing this thread will morph into a discussion about using multiple (4) subs in a ‘swarm’ arrangement, as the panacea for most bass evils, and I accept that in principle the swarm should work well. But I haven’t had any success with it in my system and still prefer a single sub placed with attention to phase and time integration to multiple subs placed with the golden ratio principle.
 I wrote about this experience in more detail here https://audioresurgence.com/2020/02/subwoofer-swarm-with-aerial-sw-12s.html

Good luck OP!


Is anyone else just paying attention at this point to see the interaction between erik_squires and millercarbon? It's like waiting for titans to do battle. 
I'm just going to get a subwoofer I can afford and play with the adjustments available until I'm happy with the integration. If I can't get happy, I'll unload the sub and stick with my two speakers. Poverty has its advantages-some things are just simple.
I prefer Larry & Susie.
Yes Duke but FR, Phase and Time interact. In order to get the frequency response right it helps to get the timing right. I have continuous control over the delays of each speaker and subwoofer and can make adjustments by remote. I can hear exactly what happens. When I run an impulse test on the system the computer will automatically set the delays so that the sound from each speaker hits the listening position at exactly the same time in phase. It also calculates filters to make the frequency response dead flat over the entire spectrum. I can overlay target curves on top to give the system the frequency response I like. Also remember like you I use 4 subwoofers. If I bypass the system it is a very dramatic difference. The image blurs, the bass gets funky and there is too much treble. With the system on you can't tell there are subwoofers until a real low note comes along and I cross up high at 125 Hz. 
Jim, having phase controls on subwoofers is not quite a gimmick but close unless you have a method of measuring delays and frequency response. Doing this by ear is next to impossible. Having room control in the sub is a dual edged sword. It does not excuse you from doing proper acoustic management. The bass will only sound right where you place the microphone and off everywhere else. If there is a 6 db null point at say 40 Hz at the listening position your sub's amp will be forced to put out 4 times the power there. If there is a 6 dB peak 8 feet away that peak now becomes 12 dB. The best way to deal with this is by having multiple subs. I find it interesting that both Audiokinesis and I both landed on 4 subs. It is a more effective way of dealing with room acoustics especially if you do not have measuring equipment. 
The Rythmik Audio subs that contain one of the company's "Full" size plate amps (A370, H600, etc.) have a continuously-variable phase control, allowing adjustment anywhere from 0 to 180 degrees (no delay to 16 ms). And if you choose Rythmik's OB/Dipole Sub, it will excite fewer room modes than will ANY single omnipole sub.
Having the luxury of visiting many show rooms in Las Vegas, I learned that sound engineers utilize near-field subwoofer placement to solve a lot of bass problems. Specific show room pending, I’ve seen individual subwoofers serving no more than 12 seats per array. How they solve group delay and echo slab in a 2,000 seat room is another matter and them sound-techs weren’t to quick to divulge.
The advantages of using multiple subwoofers had been promoted to death. Off to the opposite extreme and for the normal one to two center seated listener, I’m an advocate of placing a subwoofer right behind the main seat. With main speakers capable of good 40-50hz extension, the sub can be crossed low enough to avoid any directional clue (provided with low distortion high quality sub and 4th or higher order lowpass)

Advantages:
1. Minimum sq.ft. displacement. (One sub and no clutter)
2. Minimize room node problems for one position.
3. At typical distance of main speaker of 12-15’, phase difference with sub is nominal at these frequencies and easily compensated.
4. Due to close proximity of sub, high power isn’t necessary.
5. Equalizing mainly to extend or smooth sub’s natural respond due to minimum room influence at listening chair.

To address time alignment, I’d learned a trick from videophiles by using digital delay they use to lip sync video/audio. Bit perfect not withstanding, one can delay digital signals going into either the main or subwoofers in 1ms or less increment. (1ms approximates 1.36inches) Yes I’d tried driving the main and sub with a pair identical DACs experimenting with various phase and time alignments. And believe me that it is nearly impossible to identify arrival peaks of the sub using impulse measurements. Main speaker will show distinct first arrival peaks/dips but subwoofers are nothing but a smooth gentle wave that looks almost like a straight line. Best compromise is to physically measure the distance and compensate arrival time.

No I don’t hear much of a difference with sub time alignment but phase i can. It is easier and more impactful to compensate for phase. Oh... sorry I got side track...

Best Regards
audiokinesis2,  Wow! Your post above was a gold mine of information..
Mijostyn wrote: " Yes Duke but FR, Phase and Time interact."

Agreed! Of course they do.

And I bet that when you get the phase and time right, the frequency response is also right. 

Duke
What's funny is I literally just did this last week. I used the OmniMic DVD test to align the subwoofer to the center channel.  My audio processor's distance setting was too coarse, so I got it as close as I could, and watched the frequency response change with each change in sub delay.  First a dip, then, slowly, the dip disappeared.

Physics.

Best,

E