Time aligning subs to main speakers

zydeco

New Member
Oct 16, 2010
58
0
0
WA, Australia
#1
I’m after some advice on how to think about t time aligning subwoofers to main speakers. Two papers that have informed my thinking are the http://www.acousticfrontiers.com/guide-to-bass-optimization/?utm_source=CTA and http://www.soundoctor.com/whitepapers/subs.htm. The latter puts forward a case that the main speakers need to be delayed to cater for the delay common to subwoofers and this is a model that I’ve used with some success. My question though is whether, from a theoretical viewpoint, the delay should account for just the delay inherent in subwoofers or also account for the different distances from the listening position. The latter model sees, in the case of a distant subwoofer, the main speakers delay account for the additional travel time to the listening position as well as the inherent sub phase / EQ delay which makes sense. That said, with a set-up that is just focused on delaying the main speakers to account for the inherent phase / EQ delay in subwoofers all the different bass sources would excite the room modes / pressurise the room at the same time which also seems like a reasonable objective. What, from a theoretical viewpoint, is the right answer?
 
Dec 20, 2014
200
1
18
#2
I’m after some advice on how to think about t time aligning subwoofers to main speakers. Two papers that have informed my thinking are the http://www.acousticfrontiers.com/guide-to-bass-optimization/?utm_source=CTA and http://www.soundoctor.com/whitepapers/subs.htm. The latter puts forward a case that the main speakers need to be delayed to cater for the delay common to subwoofers and this is a model that I’ve used with some success. My question though is whether, from a theoretical viewpoint, the delay should account for just the delay inherent in subwoofers or also account for the different distances from the listening position. The latter model sees, in the case of a distant subwoofer, the main speakers delay account for the additional travel time to the listening position as well as the inherent sub phase / EQ delay which makes sense. That said, with a set-up that is just focused on delaying the main speakers to account for the inherent phase / EQ delay in subwoofers all the different bass sources would excite the room modes / pressurise the room at the same time which also seems like a reasonable objective. What, from a theoretical viewpoint, is the right answer?
You may want to speak to the guys at Lyngdorf and DEQX. I have a good friend, and he uses a fully active system using a computer as the crossover and passive subs (AMPS DRIVE THEM SEPARATELY) the software he uses does account for the issues that you raise it is called Accurate
http://www.audiovero.de/en/acourate.php

You really need to be in the digital domain to use this at its best as I understand it - the results are excellent but you really need to read up on the software functions. and 'know your onions' so to speak.
 
#3
In my home theater systems that I have had over time, I discovered a few interesting things:

1) Putting the subs at the same position as the front channels is not always optimum. One of the best results I got was the sub at the listener position or nearby, and inverted phase from the mains. The sub always appeared acoustically as if it was between the front speakers.

2) Directing the sub at an angle into a fireplace hearth was beneficial. It activated the room modes less and integrated into the space better.

Steve N.
Empirical Audio
 

Folsom

VIP/Donor
Oct 26, 2015
2,555
0
36
Eastern WA
#4
The subwoofer frequencies it produces has wavelengths so long that it generally can be nearly anywhere, depending on the crossover point. Here's distances from the main speakers based on frequency.

200hz - 5ft
100hz - 10ft
80hz - 15ft
60hz - 18ft
40hz - 27ft
30hz -36ft

So you can see, it's pretty clear that depending on frequency you got some serious options. Just like with Steve, the best was not near the main speakers nor did it need to be if the frequency is low enough. The most important thing to the ear, once you find a good places that isn't riddled the nodes, is phase. Phase is what you really hear, but the room can suck out certain frequencies so finding the least node-ish spot is the goal. That issue is why Duke came up with the Swarm, which is 4 subs placed about the room to eliminate nodes.
 

Folsom

VIP/Donor
Oct 26, 2015
2,555
0
36
Eastern WA
#6
That is too broad of a question for me to get. Reverberations? Like a picture on the wall rattling? That's an easy fix compared to building a new room.
 

zydeco

New Member
Oct 16, 2010
58
0
0
WA, Australia
#7
The subwoofer frequencies it produces has wavelengths so long that it generally can be nearly anywhere, depending on the crossover point. Here's distances from the main speakers based on frequency.

200hz - 5ft
100hz - 10ft
80hz - 15ft
60hz - 18ft
40hz - 27ft
30hz -36ft

So you can see, it's pretty clear that depending on frequency you got some serious options. Just like with Steve, the best was not near the main speakers nor did it need to be if the frequency is low enough. The most important thing to the ear, once you find a good places that isn't riddled the nodes, is phase. Phase is what you really hear, but the room can suck out certain frequencies so finding the least node-ish spot is the goal. That issue is why Duke came up with the Swarm, which is 4 subs placed about the room to eliminate nodes.
I'm not too sure what, specifically, you're trying to say here (apart from the bit about the swarm which I do get). Is it that travel time to the listening position is unimportant or ...?
 

DaveC

[Industry Expert]
Nov 16, 2014
2,178
0
36
#8
I'm not too sure what, specifically, you're trying to say here (apart from the bit about the swarm which I do get). Is it that travel time to the listening position is unimportant or ...?
At those frequencies it's just not a big deal, we are much less sensitive to time delays at sub frequencies, and you'll do more damage trying to delay the main speakers vs solving a non-existent problem. Adjusting phase is also adjusting time, this is worthwhile to do with the subs but leave the mains alone. IMHO...
 

Folsom

VIP/Donor
Oct 26, 2015
2,555
0
36
Eastern WA
#9
I'm not too sure what, specifically, you're trying to say here (apart from the bit about the swarm which I do get). Is it that travel time to the listening position is unimportant or ...?
Those are distances within the "time alignment" for the frequencies, or rather the spectrum of distance where phase can be set so our ears will percieve it correctly. Past them our ear will hear a seperation in the music, not related to direction. Above 80hz we can hear directionality as frequency goes up, so that is something to account for depending on the range being played. Below 80hz you get vast distance capability for placement because in most rooms there isn't anywhere out of time alignement, and you cannot sense direction of it either. The last bit is phase, which we are super sensitive to, but all subwoofer amps have adjustable phase.
 
May 19, 2014
693
0
16
Round Rock, TX
#10
At those frequencies it's just not a big deal, we are much less sensitive to time delays at sub frequencies, and you'll do more damage trying to delay the main speakers vs solving a non-existent problem. Adjusting phase is also adjusting time, this is worthwhile to do with the subs but leave the mains alone. IMHO...
X2 agree. If space allows another option is to place the sub(s) closer to the sweet spot than the mains and adjust phase (think someone earlier had a similar opinion).
 
Dec 12, 2012
243
0
16
#11
My experience is that physically time-aligning a sub with a main speaker isn't the primary problem. The primary problem tends to either be subs which either output too much upper band content, causing coherency issues with the main speakers, or, subs which cannot seem to integrate with the main speakers through the mid-bass to upper bass region, causing too much or too little warmth.

The warmth perception is usually dependent on the speaker/room interface as much as the sub to mains interface. I've had subs and mains combinations which solidly locked-in together through the mid and upper-bass, and others which I just never could get to sound satisfactorily balanced through that region.
 

DaveC

[Industry Expert]
Nov 16, 2014
2,178
0
36
#12
My experience is that physically time-aligning a sub with a main speaker isn't the primary problem. The primary problem tends to either be subs which either output too much upper band content, causing coherency issues with the main speakers, or, subs which cannot seem to integrate with the main speakers through the mid-bass to upper bass region, causing too much or too little warmth.

The warmth perception is usually dependent on the speaker/room interface as much as the sub to mains interface. I've had subs and mains combinations which solidly locked-in together through the mid and upper-bass, and others which I just never could get to sound satisfactorily balanced through that region.
Harmonic distortion is heard as warmth in bass frequencies and I can see this affecting sub/mains matching, especially a warm sounding sub driver matched with clean sounding mains. The sub will produce output above the xo frequency and cause the coherency issues you mention.

Woofers vary quite a bit in how much distortion they produce so I do agree this can be a potential issue. Most people are used to a fair amount of distortion in the bass, a very clean sounding woofer is rare and takes some getting used to. Low distortion woofers are rare enough the lack of distortion might not be accounted for in the recording, causing some bass to sound bit lean, almost too clean.
 

marty

Active Member
Apr 20, 2010
1,107
1
38
Far Hills, NJ
#13
I read this thread with great interest as getting subs to line up properly in my systems has been the bane of my existence for years. In all honesty, there is no better treatise on this than Soundoctor Barry Ober's excellent whitepaper cited in the OP. It's a dense but masterful summary of the state of the art on the topic. There are a few things however, on which I would like to comment.

First, as Barry says, it's critical to understand that in most systems, the only way to get the subs, which are generally placed behind the main speakers, to time align with the mains, is to retard the time arrival of the mains. And that requires DSP. It's really that simple. Barry cites that in some rare cases one might move the subs forward 9 or 10 feet to achieve this but while that might be possible in some large studio venues, it is an almost impossible request for the home listening room.

Second, Barry says that the next best solution is to use a crossover such as the excellent JL Audio CR-1 and accept the fact that the best one can achieve is to have a situation whereby the bass and mains are phase aligned (by judicious use of the phase control in the crossover) but one needs to accept that at the crossover point, the subs will always be approximately 1 cycle behind in comparison to the the mains. Most importantly, he strongly suggests that such an error will not be objectionable in most systems if the crossover is optimally set-up. That comment is one that almost nobody else makes, but to me, it makes Barry a very wise man indeed.

I have spent many years months trying both of these proposed solutions. I have used DSP systems as well as the superb JL Audio CR-1 crossover. Much to the dismay of some audiophiles, I'm sorry to report there is no "best" or "right" answer. There are always tradeoffs. Some of these have been discussed in depth previously.
http://www.whatsbestforum.com/showthread.php?520-DSP-one-person-s-experience
http://www.whatsbestforum.com/showt...T-digital-crossover&highlight=spectral+jersey

To make matters worse I am currently using a system in which the JL Gotham subs are integrated with my Wilson Alexandria S-2's in the manner that Barry does not recommend which is namely, running the mains full range and augmenting them with subs at a very low frequency (i.e. 30 Hz) though a second set of pre-amp outputs and the integral crossover on the Gothams. All I can tell you about that is that I have tried every other method Barry has suggested, and have come away thinking that my current method is the one I prefer because it offer the best balance of sonic merits/liabilities. It would be a lengthy discourse indeed to go into all the reasons why I have chosen that approach except to cite the obvious- no other method I have tried allows for the same degree of pristine signal from preamp to the mains. As good as the CR-1 is, there is still a sonic degradation through the high pass that is noticeable in comparison to "bypass" which can be done at the push of a button. And for me, once heard, there is no going back. Put another way, my system ain't perfect from a time-alignment perspective as assessed by impulse measurement, but it offers a satisfactory sonic experience nonetheless.

I cannot emphasize enough the wisdom of Barry's final comments in his white paper which is worth repeating here:

"Even if you CAN'T get the timing of your sub to match your mains as closely as it can be done, there IS a saving grace.... Notice that humans actually LIKE the fattening up of the bass loudness envelope in time. Therefore even IF your sub is 12 msec late, and you are one wavelength off, as long as you get that delayed wavelength to line up with the bass coming out of your mains, your frequency response will be pretty good and you won't have any awful objections, again, assuming you get as much else right as possible"

What Barry is saying is something audiophiles have known for years. PERFECT can be the enemy of GOOD. When it comes to adding subs, it's always prudent to keep that in mind. Let your ears be your guide.
 
Last edited:

Mark Seaton

WBF Technical Expert (Speaker & Acoustics)
May 21, 2010
319
2
18
42
Chicago, IL
www.seatonsound.net
#14
When ever concepts of "time alignment" come up with respect to loudspeakers it's a tricky matter, as you are almost never making a single variable adjustment, and few really grasp the time and phase behavior of a conventional loudspeaker, let alone how various filters affect things.

To be sure, it's highly unlikely to get a speaker and subwoofer set at the same distance to the listener to align without some form of delay applied to the speaker. One exception might be speakers with high order 3+ way crossovers where you the low frequencies of the speaker might have enough group delay to line up with the subwoofer.

I would stress that especially at low frequencies, identical physical distance simply makes for easier math, and never makes for acoustic alignment of drivers covering different bandwidths. There are some exciting possibilities with some of the new products I've seen come to market in the past few years where we have DAC/preamps which could eventually offer separate outputs for subwoofers, and easy means to implement delays and high pass filters for main speakers in the digital domain while still using the preferred DAC & preamp. We are already seeing more mid-level preamplifiers offer on-board subwoofer outputs and low pass filters. Even in the analog domain, the most transparent place to do this would be as an integrated part of the preamplifier circuitry. I think we'll see more experiment and add such features, especially with more 2 piece speakers on the market using passive main speakers and separately powered woofers.
 

Folsom

VIP/Donor
Oct 26, 2015
2,555
0
36
Eastern WA
#15
Marty,

Tweeters and midranges are often off by a half wave cycle at the crossover point, and that ONLY GROWS as the frequency goes up.

What makes you think you can hear subwoofer "time alignment" at 40hz, where 10ft behind the mains speakers is only 1/4 wave?????

Your guru found things that work, but have no relevance to his explanation with "time alignment".
 

marty

Active Member
Apr 20, 2010
1,107
1
38
Far Hills, NJ
#16
Marty,

Tweeters and midranges are often off by a half wave cycle at the crossover point, and that ONLY GROWS as the frequency goes up.

What makes you think you can hear subwoofer "time alignment" at 40hz, where 10ft behind the mains speakers is only 1/4 wave?????

Your guru found things that work, but have no relevance to his explanation with "time alignment".
I am obliged to ask a question that I hope you will answer truthfully. Did you read the Barry Ober white paper in its entirety?
 

Folsom

VIP/Donor
Oct 26, 2015
2,555
0
36
Eastern WA
#17
Alright, I went and wasted my time reading it through just to make sure I didn't miss something. Nothing changes about what I said.

There is still a bigger relationship difference between the tweeter and midwoofer than the midwoofer and subwoofer. I don't see what point this guy has... We all know a crossover can be beneficial if needed. Phase cancellation isn't hard to understand, that's why the swarm is 4 subs, not 2. etc etc...

I've made more than one stereo sound good with subs, and I use experience which really helps inform me of how the technical stuff is actually going to happen.
 
May 19, 2014
693
0
16
Round Rock, TX
#18
One segment of this topic that hasn't been touched on is quality subwoofers. You can time align, pick the optimal xover point, level, etc. but if your subwoofer isn't up to the quality (speed, dynamics, articulation) of your mains' woofers and other drivers it's all for naught. I've had Martin Logan, M&K, Dahlquist, JBL and REL. The Rels (~10 year old models) were the first to provide quality bass (which btw makes integration much easier). Then I purchased a JL Audio F112 and realized what quality (and yes quantity) bass is. Net - forget the JL CR-1 xover, worry less about time alignment (I agree with the post above wrt marginally audible low freq time alignment) and ensure you purchase the best sub you can afford. This way you're starting off with an additional speaker(s) that are up to adequate quality level and not a sonic detriment from the get go.
 

Folsom

VIP/Donor
Oct 26, 2015
2,555
0
36
Eastern WA
#19
Indeed, many subwoofers simply lack musical quality. Luckily there are great options out there now, literally for anyones budget.
 

Bruce B

WBF Founding Member, Pro Audio Production Member
Apr 26, 2010
6,558
3
38
Seattle, WA
www.pugetsoundstudios.com
#20
I thought sound travels at exactly the same speed whether it's 40Hz or 4kHz???

If you are trying to align the speakers/subs at the apex of the wave, say 25Hz, then it will ONLY be aligned at that particular freq and its multiples (50, 100, 200........)

Oh crap..... it's not aligned at 63Hz..... what's an audiophile to do?????