"Wave Launch" and Subwoofer Placement?

[LL21]

Mark - I have emailed you at sales@seatonsound.net

 

[DaveC]

Good thread.

Lots of good info, but I think Mike L got it right from the start with an easy to understand explanation.

Frequency response matters more than placement but it's nice to have the sub's "wave launch" arrive at the LP at the same time. I think the importance of this has been overemphasized as a marketing gimmick though, I'm not sure many can really hear phase at subwoofer frequencies and those who claim to are most often confusing it with changes in frequency response. Mark's characterization of the direct sound vs room contributions as being fuzzy are so because of this difficulty of truly separating variables, making it difficult to experiment with and making it easy to arrive at faulty conclusions.

Also agree there's a lot to be said for a truly full range speaker properly setup. Adding subs, especially the tower arrangements, is like adding another "way" to your speakers, but without the proper crossover to rolloff the main speakers.

On impact, impact comes from the full frequency range so an overall larger speaker, especially a horn as it's design lends it's self to efficiently moving air, has an advantage. In terms of subs and impact, if they negatively effect frequency response they can harm dynamics and make things all muddy, so I think some perception of improved impact with subs is a result of smoother frequency response as well as an increase in driver surface area, which is usually good. IMO, the more the better as far as woofer surface area!

 

[LL21]

Thank, DaveC...i was thinking about this overnight and realized there is an obvious non-techie question in here from me:

In the same way we have all seen the presentation by standalone mini-bookshelves be transformed by adding a massive, high quality sub, what happens if we add proportionately massive sub towers to truly full-range speakers (so you are talking truly massive if your speakers are big Wilsons for example)?

(I am aware of the flaws in these mini-monitor/sub systems...but I suspect some of those go away when the main speakers are actually full range.)

So if we added, for sake of argument, 2 towers of 6-10 x 18" cones per tower (to keep life simple, assume expertly designed and extremely well set up)...do we think that those full range speakers take just as great a leap forward as those mini-monitors?

I remember reading that Dan D'Agostino had a 'wall of Krell MRS Master Reference Subs' with his Wilson X1s many years ago. (Each Master Ref Sub is 2 x 15" cones, weighs 450lbs...and a wall of them was apparently something like 6 of them).

 

[Mark Seaton]

Thank, DaveC...i was thinking about this overnight and realized there is an obvious non-techie question in here from me:

In the same way we have all seen the presentation by standalone mini-bookshelves be transformed by adding a massive, high quality sub, what happens if we add proportionately massive sub towers to truly full-range speakers (so you are talking truly massive if your speakers are big Wilsons for example)?

(I am aware of the flaws in these mini-monitor/sub systems...but I suspect some of those go away when the main speakers are actually full range.)

So if we added, for sake of argument, 2 towers of 6-10 x 18" cones per tower (to keep life simple, assume expertly designed and extremely well set up)...do we think that those full range speakers take just as great a leap forward as those mini-monitors?

I remember reading that Dan D'Agostino had a 'wall of Krell MRS Master Reference Subs' with his Wilson X1s many years ago. (Each Master Ref Sub is 2 x 15" cones, weighs 450lbs...and a wall of them was apparently something like 6 of them).

The catch and difficulty with any speaker designed with more limited bass extension is that unless it is expected to be used with a subwoofer, they almost always have some elevation of the bass range they do cover to enhance the harmonics of the range they are missing and not sound overly bright, or light on the bass. If you think of some of the old Linn micro/mini-monitors and compact ProAc speakers they subjectively sounded rather balanced on their own despite the obvious limitation in low frequency extension. That subjective balance is largely formed by a relative balance of the upper and lower range energy. In overly simplified terms, if you drop off an octave of bass energy with a speaker that extends to 60Hz vs 30Hz, you have to subjectively make up for it some with elevated response at and above 60Hz. It is an imprecise adjustment which works better on some material than others.

The problem comes in when you have a speaker voiced to sound full with limited extension and you then extend the bottom to 15-20Hz vs 60-80Hz of the speaker itself. It can sound very good, but it often becomes a tenuous balance of just enough vs too much. This is where a more full range speaker will have a balance that may be more complimentary to adding deep reaching subwoofers. Of course if you work some form of electronic correction into the mix like a Trinnov, Lyngdorf, Dirac enabled component, or similar into the mix, this mostly becomes a non-issue, while adding the ability to usually add a high pass for the speakers and delay to match the bulk energy arrival and phase of the speakers with the subwoofers.

 

[musicfirst1]

Thank, DaveC...i was thinking about this overnight and realized there is an obvious non-techie question in here from me:

In the same way we have all seen the presentation by standalone mini-bookshelves be transformed by adding a massive, high quality sub, what happens if we add proportionately massive sub towers to truly full-range speakers (so you are talking truly massive if your speakers are big Wilsons for example)?

(I am aware of the flaws in these mini-monitor/sub systems...but I suspect some of those go away when the main speakers are actually full range.)

So if we added, for sake of argument, 2 towers of 6-10 x 18" cones per tower (to keep life simple, assume expertly designed and extremely well set up)...do we think that those full range speakers take just as great a leap forward as those mini-monitors?

I remember reading that Dan D'Agostino had a 'wall of Krell MRS Master Reference Subs' with his Wilson X1s many years ago. (Each Master Ref Sub is 2 x 15" cones, weighs 450lbs...and a wall of them was apparently something like 6 of them).

I think of the effects in terms of car motors and the adage, 'there is no substitute for displacement' when it comes to bottom octave response. My previous Rockport Cygnus had usable response to 20hz, supplied by 2 x 10" woofers for a total radiating area of 157 in2 per channel. When I add 3 x 12" Rel subwoofers per channel my useable low frequency radiating area increases by 339 in2. Is it meaningful? IMO most definitely.

 

[DaveC]

The catch and difficulty with any speaker designed with more limited bass extension is that unless it is expected to be used with a subwoofer, they almost always have some elevation of the bass range they do cover to enhance the harmonics of the range they are missing and not sound overly bright, or light on the bass. If you think of some of the old Linn micro/mini-monitors and compact ProAc speakers they subjectively sounded rather balanced on their own despite the obvious limitation in low frequency extension. That subjective balance is largely formed by a relative balance of the upper and lower range energy. In overly simplified terms, if you drop off an octave of bass energy with a speaker that extends to 60Hz vs 30Hz, you have to subjectively make up for it some with elevated response at and above 60Hz. It is an imprecise adjustment which works better on some material than others.

The problem comes in when you have a speaker voiced to sound full with limited extension and you then extend the bottom to 15-20Hz vs 60-80Hz of the speaker itself. It can sound very good, but it often becomes a tenuous balance of just enough vs too much. This is where a more full range speaker will have a balance that may be more complimentary to adding deep reaching subwoofers. Of course if you work some form of electronic correction into the mix like a Trinnov, Lyngdorf, Dirac enabled component, or similar into the mix, this mostly becomes a non-issue, while adding the ability to usually add a high pass for the speakers and delay to match the bulk energy arrival and phase of the speakers with the subwoofers.

Yes, speakers are also often rolled off up top to match the bottom end so when you extend the bass the the tonal balance is off.

I think of the effects in terms of car motors and the adage, 'there is no substitute for displacement' when it comes to bottom octave response. My previous Rockport Cygnus had usable response to 20hz, supplied by 2 x 10" woofers for a total radiating area of 157 in2 per channel. When I add 3 x 12" Rel subwoofers per channel my useable low frequency radiating area increases by 339 in2. Is it meaningful? Most definitely.

Yup, I think it's "there's no replacement for displacement".

In speakers it's more surface area, as adding more bass via excursion is not as good as more surface area. It is hard to put my finger on exactly why it's so much better. I have my own horn speakers with 15" woofers and also a pair of 3-way towers with 2x7" woofers per side and the 15s are simply in another league. I have measured less distortion and lower extension in the 15s, but there's more to it...

 

[LL21]

Thanks, gents! In the case I am specifically asking about...having big Wilsons...it probably means that I do have some ability here to really 'play' with 'serious surface area options' which would likely be cut off above 38hz-40hz or so. In nearly every case, big Wilson owners I know have calibrated their subs to be cut off above 38hz-40hz.

But whether it is the REL '6-pack', Seaton Sub Special, Thor Hammers, etc...it seems that there is another world yet to explore...

 

[LL21]

Mark Seaton, Duke LeJeune and Others, here is a direct question:

- If we look at how transformational adding a Velodyne DD18+ is to a smaller speaker (as per professional reviews, WBF member experiences including my own) we have a speaker like B&W 805s or the SF Guarneri which has a cone surface area of around 29 square inches per speaker...vs a DD18+ with a cone surface area of 254 square inches (divide by 2 to reflect that the Velodyne is 'split' between the 2 speakers, so 127 square inches)
- This means that this transformation is created by adding a sub that has 4.4x more cone surface area than the main speaker

DIRECT QUESTION: IS IT LOGICAL THEN TO TAKE THIS RATIO AND APPLY IT TO FULL-RANGE SPEAKERS AND "KEEP GOING"? (To avoid room dynamics being a constraint, assume a large room 40' x 20' x 12')

- in other words, taking this ratio to the next logical step, we have a Wilson XLF which is 211 square inches per channel (2 x 1" tweeters, 2 x 7" midranges and 1 x 13" midbass woofer...i assume only the 15" woofer really takes lower bass registers.)
- This means a subwoofer with cone surface area of 924 square inches PER CHANNEL.
- That is basically 4 x 18" woofers per channel or 1018 square inches. (Not far off 6 x dual 15" subs...3 per side (1060 square inches per side)...which legend has it, was something Dan D'Agostino did with his own Wilson X1s years ago)

We are starting to read about REL 6-Packs (509 square inches/channel) and Dual Thors (353 sq inches/channel), as well as YG Acoustics massive dual 21" subs (693 sq inches/channel), never mind the 4-tower speaker systems from Gryphon, Evolution Acoustics, Genesis/Infinity, YG Acoustics, Stenheim, Tidal, Goebel, Alsyvox and Verity Audio (Monsalvat)...and I am convinced it is NOT because everyone watches Jurassic Park and Edge of Tomorrow all the time...but because there is something tangible (at nearly any volume, low or high) in the spacial cues, sense of space, ambience, hall that comes with effortless, low distortion sub-bass information. And having spoken with an owner of the REL 6-pack, the low-pass filters are set at varying levels between 24hz and 28hz as I recall...so this is truly sub-terranean cues not 'music'.

Reviewers like Roy Gregory and Jeff Dorgay as well as our very own Ron Resnick (of Gryphon Pendragon fame), Dato Han, Chris Martens, Marc Phillips (Parttime Audiophile), Jason Victor, Rick Becker (Enjoythemusic.com) who have been painstakingly writing about these massive 4-tower speakers and/or subs have described a sense of realism (not only scale-related) in the surrounding sense of expanse of the original venue, as well as details coming forth higher up in the register which had been heretofore obscured.

Thanks for your thoughts on this.

 

[musicfirst1]

In speakers it's more surface area, as adding more bass via excursion is not as good as more surface area. It is hard to put my finger on exactly why it's so much better. I have my own horn speakers with 15" woofers and also a pair of 3-way towers with 2x7" woofers per side and the 15s are simply in another league. I have measured less distortion and lower extension in the 15s, but there's more to it...

Isn't this an interesting observation, as it very much mirrors my experience. I used to consider 'displacement' as merely the driver surface area multiplied by maximum displacement, but like DaveC, I found that more driver surface area worked much better than, for example, 1/2 the area at twice the excursion. I often thought it might have to do with the smaller excursions being more in the linear region of the driver, or perhaps how much easier larger drivers couple with the room.

 

[Duke LeJeune]

Mark Seaton, Duke LeJeune and Others, here is a direct question:

- If we look at how transformational adding a Velodyne DD18+ is to a smaller speaker...

- This means that this transformation is created by adding a sub that has 4.4x more cone surface area than the main speaker

DIRECT QUESTION: IS IT LOGICAL THEN TO TAKE THIS RATIO AND APPLY IT TO FULL-RANGE SPEAKERS AND "KEEP GOING"?

I don't use cone area ratios as a primary metric for system design. I DO take thermal and mechanical compression as well as SPL targets into account, to make sure one part of the system doesn't prematurely become the limiting factor. So cone area does get factored in, but not specifically as a target ratio.

That being said, what I do isn't normal... I'm inclined to use fairly large midwoofer cones, for the sake of radiation pattern control. So I tend to end up with lower cone area ratios than what you described.

When pairing more "conventional" main speakers with subs I'd probably still pay more attention to uncompressed SPL capabilities than to cone area ratios. And adding subs to main speakers which are already full-range raises issues that would probably have to be resolved on a case-by-case basis.

 

[LL21]

Yes, speakers are also often rolled off up top to match the bottom end so when you extend the bass the the tonal balance is off.




Yup, I think it's "there's no replacement for displacement".

In speakers it's more surface area, as adding more bass via excursion is not as good as more surface area. It is hard to put my finger on exactly why it's so much better. I have my own horn speakers with 15" woofers and also a pair of 3-way towers with 2x7" woofers per side and the 15s are simply in another league. I have measured less distortion and lower extension in the 15s, but there's more to it...

Interesting...the 15" woofer has a surface area of 177 square inches, vs the 2 x 7" woofers combined of 77 square inches. Plus, (all else being equal), i would have thought the forward/backward excursion might need to be greater as well in order to match the excursion of the bigger 15" cone (leading to 'potential' for greater distortion). So I wonder whether even 4 x 7" woofers (154 sq inches) would still be close (again, all else being equal which invariably it never is.)

 

[LL21]

I don't use cone area ratios as a primary metric for system design. I DO take thermal and mechanical compression as well as SPL targets into account, to make sure one part of the system doesn't prematurely become the limiting factor. So cone area does get factored in, but not specifically as a target ratio.

That being said, what I do isn't normal... I'm inclined to use fairly large midwoofer cones, for the sake of radiation pattern control. So I tend to end up with lower cone area ratios than what you described.

When pairing more "conventional" main speakers with subs I'd probably still pay more attention to uncompressed SPL capabilities than to cone area ratios. And adding subs to main speakers which are already full-range raises issues that would probably have to be resolved on a case-by-case basis.

Thank you! All else being equal, are SPLs logarithmic to cone size or linear? Just curious is my cone surface area approach (clearly assuming linear progression) is missing a logarithmic increase in scale.

That being said, i wonder how the Velodyne DD18+ SPLs at sub 48hz compares with the SF Guarneri...ie, what is THAT ratio?...and therefore once again, whether that ratio can be applied to the big Wilson's capability and the level of SPLs one would seek for matching subs to produce to get that same level of uplift in scale/performance/presentation.

 

[Solypsa]

...
That being said, what I do isn't normal... I'm inclined to use fairly large midwoofer cones, for the sake of radiation pattern control. So I tend to end up with lower cone area ratios than what you described.
...

I think thats only 'not normal' in hifi pretty typical in pro/theater/touring applications.

 

[Duke LeJeune]

All else being equal, are SPLs logarithmic to cone size or linear?

For a doubling of cone area we theoretically gain 3 dB in SPL, but in the real world of course all else is never equal. Cone mass increases, which changes the frequency response curve and lowers efficiency (but extends the low end deeper, assuming box size is appropriately increased). The larger woofer might have a more powerful motor which would bring the efficiency back up, possibly accompanied by an increase in voice coil diameter which would increase the power handing, which would also tend to increase the max SPL capability, assuming amplifier power is scaled up. And so forth.

That being said, i wonder how the Velodyne DD18+ SPLs at sub 48hz compares with the SF Guarneri...ie, what is THAT ratio?

Making guesses based on the published specs, I estimate that a single DD18+ will go ballpark 6 to 8 dB louder than a pair of Guarneri Tradition speakers.

...and therefore once again, whether that ratio can be applied to the big Wilson's capability and the level of SPLs one would seek for matching subs to produce to get that same level of uplift in scale/performance/presentation.

Extrapolating from the Guarneri Tradition to the Wilson Alexandria XLF, I guesstimate that it would take about 6 DD18+'s to "match up" in roughly the same way with a pair of the big Wilsons.

"On paper" that looks to me like overkill... NOT that "overkill" is necessarily a bad thing when it comes to subwoofers... or high-end audio in general...

 

[LL21]

For a doubling of cone area we theoretically gain 3 dB in SPL, but in the real world of course all else is never equal. Cone mass increases, which changes the frequency response curve and lowers efficiency (but extends the low end deeper, assuming box size is appropriately increased). The larger woofer might have a more powerful motor which would bring the efficiency back up, possibly accompanied by an increase in voice coil diameter which would increase the power handing, which would also tend to increase the max SPL capability, assuming amplifier power is scaled up. And so forth.



Making guesses based on the published specs, I estimate that a single DD18+ will go ballpark 6 to 8 dB louder than a pair of Guarneri Tradition speakers.



Extrapolating from the Guarneri Tradition to the Wilson Alexandria XLF, I guesstimate that it would take about 6 DD18+'s to "match up" in roughly the same way with a pair of the big Wilsons.

"On paper" that looks to me like overkill... NOT that "overkill" is necessarily a bad thing when it comes to subwoofers... or high-end audio in general...

Hi Duke - Thank you! That is extremely helpful guidance.

 

[Klonk]

Any experience with this device, Anti-Mode 8033 Automatic Subwoofer Equalizer?

 

[Mark Seaton]

Thank you! All else being equal, are SPLs logarithmic to cone size or linear? Just curious is my cone surface area approach (clearly assuming linear progression) is missing a logarithmic increase in scale.

That being said, i wonder how the Velodyne DD18+ SPLs at sub 48hz compares with the SF Guarneri...ie, what is THAT ratio?...and therefore once again, whether that ratio can be applied to the big Wilson's capability and the level of SPLs one would seek for matching subs to produce to get that same level of uplift in scale/performance/presentation.

Hi Lloyd,

I would first caution that comparisons of cone area will make for reasonable sanity checks and some quick litmus testing, but there are many variables involved in overall headroom across the spectrum, and most often the correct answer is some version of "it depends." Comparisons are more direct and easier to conceptualize with fully sealed systems, while ports and other acoustic loading augment, extend, or compliment output in more frequency dependent ways, and are harder to dimensionally equate to cone area.

One important understanding to start with is that maintaining the same SPL output at lower and lower frequencies requires 4x the volume displacement for every octave. The relationship is a squared function. Here are two useful practical applications of this reality:

1. A sealed speaker requires 4x as much excursion to produce 30Hz as it does 60Hz, and 16x more excursion at 20Hz than 80Hz.
2. In order to produce the same SPL under the same conidtions, an 18" woofer producing 20Hz and a ~5" woofer producing 80Hz will move the same distance in/out.

While surface area is a useful figure to understand, ultimately we care about headroom, distortions, and efficiency to the degree it limits output. In an extreme example we could consider the case of a 6" woofer with 100dB@1W/1m sensitivity and a 15" cone that is only 86dB@1W/1m. The woofer needs 25x the power to match the SPL of the 6", despite the cone area advantage. We really care about the combination of sensitivity at the frequency range of interest (dictated by box size, type, and the woofer), the power available or needed, and what distortions and compression might occur due to limits in power, driver excursion limits, and distortion from the woofer as each are increased. Different approaches with size of woofers, size of enclosures, power, and enclosure type will tend toward specific strengths and characteristics, but there are always outliers and a wide range of performance trade offs.

 

[Mark Seaton]

Any experience with this device, Anti-Mode 8033 Automatic Subwoofer Equalizer?

While the point-n-shoot qualities of the Anti-Mode products make it attractive, there are others such as MiniDSP which offer much more control and setup options. For high performance systems where budget isn't as sensitive, some of their more advanced products can actually be used as just subwoofer control, despite many other features and capabilities being packed into the product. One example would be their SDH series. At $1200 with a microphone, the SDH could easily handle 1-4 subwoofer channels in a system on its own, or the SDH Studio would be enough to do it in the digital domain if you have digital volume control upstream, or linked volume control of more than 2 channels using DACs of your own choosing for the output of the subwoofer channel.

 

[DaveC]

While surface area is a useful figure to understand, ultimately we care about headroom, distortions, and efficiency to the degree it limits output.

Surface area is directly related to all of these things, so when you increase surface area you will also raise headroom, lower distortion, increase efficiency and have greater SPL capabilities.

I'm not sure if these measurements covey the entirety of the subjective improvements that having larger woofer surface area bring to the table, but I don't think it's a coincidence that the great majority of speaker manufacturers increase woofer surface area to a massive degree when you look at their cost-no-object systems.

 

[DaveC]

I think thats only 'not normal' in hifi pretty typical in pro/theater/touring applications.

Most amplified music venues I've been to around here are using massive numbers of subwoofers combined with line arrays.