Car Audio Central - Ultimate Car Audio Forum and Community

I don't know about you guys, but I prefer Icecream Cones the best

"superiority of...
...the cone? Or the speaker?
(how ever did you isolate "the cone", I ask myself... )"

SEAS implementation of all these elements :D

I can't agree with you more on any of the above, Excels were just a specific example after the damping comment (choosing damped material every time). Really, though, I don't see this going much further, unless you'd like it to, since I feel we all can agree that it is the engineering of a driver that produces the end result, and not generalizations on materials technology, motor topology, crossover type, and so forth. We're kind of starting to get nit-picky now :)

I do feel that as we become more familiar with driver design, implementation, and overall engineering of the components, we become more able to give an unbiased opinion of certain aspects of a driver regardless of our preference. Or at least respect the benefits and tradeoffs and designer has chosen in order to accomodate the intended audience. I can respect electrostats, cone drivers, horns, ribbons, in all their pros and cons. Maybe even SPL competitors, but I'm getting ahead of myself ;)

:cheers:

Holy post overload!

[quote="Jonathan"]Take a look at SEAS Excels/Lotus, some of the most highly renowned reference drivers (home and car audio) in the world. Pure magnesium, inherently poor (probably the worst of all metals) damping characteristics. [/quote]
Before we go down a road that's not quite on-target, first let's not forget that we were originally talking about subwoofers... lower frequencies, higher excursion levels, different resonances... fundamentally lower resonant frequency. ]The secret is that the density of magnesium allows enough material to be cast into the cone to get an ultra high rigidity for the mass, pushing resonances to a single peak around 5khz for a 7" midbass driver and eliminating the edge resonances that ALL paper, poly, carbon fiber, and other cones experience throughout their intended frequency range.[/quote]
The secret is the density of ANYTHING. :D
You can execute anything poorly, or well.
Cone materials don't contribute with such stereotypes as you list here...
I could produce for you a carbon/kevlar example that had good stiffness, lightness, and damping... and a poorly made one that wasn't so stiff, had resonances, and maybe even wasn't so light.
I could produce for you a metal cone that was stiff, light, and well damped...
...well, you know where I'm going here. ;)

Depending on the design attributes of the OTHER components in the sub (or speaker), harmonics and resonances might appear that make one cone material more logical than another.
Or, a need might arise (such as a lower Fs) that dictates the use of other types of cone materials.
Or, other design issues and interactions between components (let's see... I have [i]this[/i] motor, which will maybe not have as much strength as usual, but more excursion... and I have [i]this[/i] suspension, which has an excellent damping affect on cone motion but brings my resonant frequency up... so let's pick a cone material that I may not necessarily need to worry as much about damping, and lightness becomes my first priority... just for example).

If there was one "good" cone material for all applications...
...every speaker would use it.
Human nature - we evolve towards the truth.
In this case, we're "evolving" and still keeping the same mix of paper, poly, metals, and resin-impregnated wovens. Each has it's benefit - but you really can't discuss the benefit without looking at the whole picture. ;)
[quote] Then, a simple notch filter erases this peak, and a crossover around 1.5-2 octaves below this point to avoid 2nd and 3rd order distortion from exciting this resonance. Effectively, this driver sees far less resonance than even the best paper cones could even dream of. [/quote]
You just mentioned one resonance that wouldn't show up with a paper cone.
And the specific resonance you mention is certainly not attributable to "every metal (or even magnesium) cone"! It's that ONE cone, and in large part because it's assembled onto a structure having THAT surround, THAT spider, THAT dustcap, and THAT motor...change those elements, and you'll move that resonance all over... and that's not discounting the cone's contributions to resonance, but you can manipulate those cone properties (thickness, shape, size) and move the resonances around also.
You can do this with ANY cone material. ;)
[quote]This shows itself in a waterfall plot. [/quote]
And it's good that you mention a waterfall plot...
Once again, something where resonances show up, possibly ring longer than other frequencies - but ultimately it's not the fault of "the cone" - it's the fault of the speaker designer.
...maybe for the speaker designer's goals, that resonance was an acceptible trade off, given whatever his ultimate goals were (and he as well may have rationalized it away as "oh, they can just add a notch filter and it'll be smoothed out"...
...but if his priorities were to smooth it out, he could have done that alternatively as well, by manipulating the very things that gave him the resonance. :fing02:

Or, in a LOT of cases, it's not so much what the engineer wanted... but what he didn't care about addressing.
Either way - you've got a renegade resonance.
Other speakers have other ones.
Don't blame the cone - blame the speaker. Er, the designer. ;)

[quote]And I haven't seen a single person that DIDN'T admit their superiority in clarity, detail, and neutrality over some of the other highly renowned competitors such as Scan-Speak Revelators and their Carbon Fiber versions using the SD-1 motor, Vifa PLs, XTs, Accutons ceramic drivers, and many others. [/quote]superiority of...
...the cone? Or the speaker?
(how ever did you isolate "the cone", I ask myself... :D )

And how many people DID you ask, in this unanimous vote, when you were having this great double-blind A/B/C/etc. comparison between Scan-Speak Revelators, Scan Speak CF drivers, Vifa PL's, Accutons.. and "many others"?
And who were these audio experts anyway?
What were the other conditions of the test?
How were their opinions normalized against one another?
:p

It may SEEM like "everyone" has a unanimous opinion...
...in reality, I don't think I know three audiophiles who I can even put in a room together and not have a discussion involving why each one thinks their setup is superior... not that they don't like the other's setup, but for their personal tastes, it suits their needs.

Complicating things further, you've got the "just buy the highest end, latest and greatest technology" guy, together with the "just use analog sources and tube amplification for the signal depth" guy, together with the "I'm a DIY experimenter guy", and maybe the "I run Electrostats / magnetic planars / dipole arrays" guy to throw a completely new curve at you. :fing02:


Hell, I work in the audio industry (manufacturing/design, retail brick and mortar consulting, and marketing both on the internet and in retail), and
1) if I counted the people I knew locally who even had enough technical expertise to understand what components in a speaker controlled what parameters on a spec sheet, I wouldnt' end up with enough people to have a good game of poker... :p
2) the more respectable people I meet, the more varied and diverse the directions and approaches become - the less from a "unanimous" outcome on [i]anything[/i] - theoretical conversation, listening evaluation, or anything. :D

[quote]I used to prefer paper cones, but when I first heard the Excels when they came out, I quickly changed my mind.[/quote]
Again - how can you say it's "the cone" that makes the difference here? ;)

Lots of speakers have different tonal characteristics...
It's not just the cone controlling that. ;)

I'd never stereotype against any particular material (not even poly), because not only can any one of them be well engineered or poorly engineered - but any one of them might just happen to be "the perfect fit" for a particular design, due to any of numerous other design decisions that ultimately make up the entire puzzle that is the finished product - the speaker.

[quote] In the end, I'd put paper as second best if you want clarity and accuracy, but I haven't heard any that could touch SEAS drivers. And it proves itself in testing:
[url]http://www.linkwitzlab.com/x-mid_dist.htm[/url]
[/quote]Remember, sound quality is subjective, not always objective.
Solid state amplification consistently "tests" superior to tube amplification.
Yet a huge portion of the home audiophile world prefers tube amplification.
You can't account for intentional dulling of response...
If you could, there would be no soft dome tweeters in the world, after all. And a great many people prefer [i]those[/i] as well. ;)

[quote]The only thing I'd want excellent damping with is a tweeter, namely because it is near impossible to keep harmonics from exciting the resonant peak of a metal dome tweeter. Focals Beryllium tweeter is close, but it still suffers the characteristics of an inverted dome Focal tweet (poor off axis response, bad low freq. power handling and harshness at that point, only achieves good results directly on axis and VERY hard to sound natural).[/quote]I'd rather say that you can do different things with a Focal tweeter.
If you are trying to put a Focal tweeter down in a kick panel - 90% of the time, it's not going to be a good fit there.
And that definitely has to do with the directionality of the tweeter.
(but with the exception of the Beryllium tweeter, you are aware of the waveguides, which let you steer the vertical dispersion of the tweeter, right? ;) )
Try putting it up in an A-pillar. It's a location I generally try to avoid - and most often, the Focal TN52's, and even the TN47's just sing in that location, without the typical harshness - because the off-axis energy isn't playing roller-derby off the windshield as much, if you will. ;)
And that allows you to take advantage very often of a higher, flatter soundstage - even wider, quite often. :cool:

On the other hand, starting with vehicles that had stock locations up there (or near there - such as the dash-top at the base of the A-pillar), soft dome tweets like the DLS Magnesiums, I've had less than great luck with those A-pillar locations.

[b]The bottom line is:[/b] every driver has an intended purpose, application, or goal.
The designer chooses his components to suit that goal.
One subwoofer design is built for sealed compatibility, for a vehicle... another is designed for vented compatibility, for a home ESL, another is designed for vented compatibility in an SPL vehicle tuned to 55hz.

These speakers all have different applications. They'll all find themselves in different enclosures. In those different enclosures, they'll find themselves contending with different air pressures (and pressure forces of different phase relationships relative to cone motion, in fact ;) ), at different frequencies. Even if you take it no further, this [i]alone[/i] can account for differences in cone material.
But if you get into the reasons why the engineer designed the other parts of the sub, for the applications it's been built for, you'll find even more reasons.
And possibly alternatives that would have worked just as well - but maybe didn't suit the designer's [i]other[/i] goals - possibly appearance, or marketing to the intended target audience? ;)

Just some things to ponder. :cheers:

So, who is sinister a new dealer for again? ;)

"There is no perfect damping solution - but if you compared the sound of a cone with poor damping characteristics to one with good damping characteristics, you'd pick the cone with good damping characteristics every single time, believe me. ;) "

Not really, see below.

"Even look at high-end speakers if you will. Midrange drivers, where cone resonances are a much more serious (much more audible) issue:
DLS is a well known high-end manufacturer.
Look at their "Ultimate" series offerings. In there you'll find poly cones, and even an exotic carbon/kevlar honeycomb structure...
...and what does their Iridium set use? Their "Best of the best" flagship?
A heavy, course, unpressed paper cone. "

Take a look at SEAS Excels/Lotus, some of the most highly renowned reference drivers (home and car audio) in the world. Pure magnesium, inherently poor (probably the worst of all metals) damping characteristics. The secret is that the density of magnesium allows enough material to be cast into the cone to get an ultra high rigidity for the mass, pushing resonances to a single peak around 5khz for a 7" midbass driver and eliminating the edge resonances that ALL paper, poly, carbon fiber, and other cones experience throughout their intended frequency range. Then, a simple notch filter erases this peak, and a crossover around 1.5-2 octaves below this point to avoid 2nd and 3rd order distortion from exciting this resonance. Effectively, this driver sees far less resonance than even the best paper cones could even dream of. This shows itself in a waterfall plot. And I haven't seen a single person that DIDN'T admit their superiority in clarity, detail, and neutrality over some of the other highly renowned competitors such as Scan-Speak Revelators and their Carbon Fiber versions using the SD-1 motor, Vifa PLs, XTs, Accutons ceramic drivers, and many others. Far above DLS Iridiums, which I wasn't terribly impressed with other than the inherent characteristics that make them more easy to tailor to a vehicle (namely the midrange driver, which can cover ~300hz up to 6-7khz). The Excel/Lotus drivers do tend to have a more cold tonality, but they're much more neutral and accurate. If you want the "romantic" (resonant) sound of a paper or poly driver, it's not for you, but if you want clarity, transparency, and natural sound, it is definately a world beater.

I used to prefer paper cones, but when I first heard the Excels when they came out, I quickly changed my mind. In the end, I'd put paper as second best if you want clarity and accuracy, but I haven't heard any that could touch SEAS drivers. And it proves itself in testing:
[url]http://www.linkwitzlab.com/x-mid_dist.htm[/url]

The only thing I'd want excellent damping with is a tweeter, namely because it is near impossible to keep harmonics from exciting the resonant peak of a metal dome tweeter. Focals Beryllium tweeter is close, but it still suffers the characteristics of an inverted dome Focal tweet (poor off axis response, bad low freq. power handling and harshness at that point, only achieves good results directly on axis and VERY hard to sound natural).

[quote="Jonathan"]I've never ever seen a dead flat XBL^2 motored anything. ...Depends on your definition of flat, though, whether at the absolute limits or right past center.[/quote]
It really doesn't matter truthfully - BL intermodulation (fluctuating BL levels as the cone moves) is what causes intermodulation distortion.
And typically, we're talking about some significant fluctuations here... on the order of 10%, maybe more.
The ripples you are looking at there - they are somewhere on the order of a 1% fluctuation.That's not going to be audible.

ALL subs will have some fluctuations like this - with XBL^2 it's predictable, with other subs it's not - take another look at that W7 DUMAX.
If you aren't calling the Brahma flat, you aren't calling the W7 flat. ]Not arguing that, but there are other topologies that give a flat BL to a certain point. Namely underhung.[/quote]
No more so than XBL^2 - in fact less. Just like with overhung, there's an inherent curve to the BL curve.
With underhung, as those short windings move closer to the edges of the top plate, the magnetic flux density is progressively less and less - and as a result, BL drops off.
This is similar to overhung, but the rolloff is slower.
Only dual-gap technologies really directly address this, by setting up a scenario where inherently, as the sub moves, as windings move out of one gap, an equal number are taken up in the other gap - it's not until you start to move out of that gap's field that BL begins to fall - creating a true FLAT BL between those two points. ;)
...and technically, theres' no limitation to two gaps, one coil... Adire had a demo unit at CES two years ago, using three gaps and two coils - and it had something like a 4" one-way throw to it. They had to custom-build an experimental suspension for it.. the Arachnid. It was cool to see in action though. :cool:

[quote]Show me a perfectly flat XBL^2 or DDD driver and I'll take a picture of myself in a tutu and post it as my avatar. It doesn't exist. [/quote]
Show me a picture of a "perfectly flat" underhung... same thing.
Even if you accept ironing out 1% measurement ripples as nominal, XBL^2 subs will still yield a flatter BL curve, even if you limit the excursion.

It's just like the contour of the earth...
To our perpsective, the ground looks flat, but really it's a progressive curve.
The same is true for an underhung sub... where you might say the overhung is more like the smaller diameter of the moon - a steeper curve. :D
Yes, underhung has a gentler curve than overhung - but XBL^2 is the concept of the truly "flat earth" that Christopher Columbus was going to sail off the edge of. :D
(which we didn't think could exist, until a couple years ago... a true "flat" BL curve that is, not a flat earth :p )

[quote]W7 is a completely different beast from a typical LGLC design. Can't generalize them, as you can't generalize XBL^2. W7 stays flat to its limits then experiences the dramatic BL rolloff that XBL^2 does past its limits. [/quote]Officially, it's a LGLC sub. Unofficially, you could consider it an LGLC / XBL^2 hybrid, due to that pole-piece cross drilling that they really don't talk much about... :bigwink:

[/quote]"The 10W7 isn't flatter - you can even see with the 10W7 that the BL curve has a different shape on the in-stroke, compared to the out-stroke."
As does a Brahma.[/quote]
Well, not really. You can tell by the DUMAX report that the Brahma that was measured had a slight bit of an "off-center" problem... which could have fully been the fault of the DUMAX measurement, since we're manipulating the cone fully by air pressure in the chamber, and the cone position is measured by positioning lasers - the operator tells the machine when the cone is at the "at rest" centered position. ;)

Conversely, the anomolies you see with the W7 aren't issues that could be attributed to a centering issue.

[quote]"A heavier cone would tend to experience resonances later than a lighter competitor, because it will inherently have greater damping properties - [b]pushing those resonances down to a lower frequency, as well[/b]."
That's the problem.[/quote]
Well, just like damping materials - it's not simply an issue of "intensity" or "frequency" - it's both.
So the resonances not only will be a lower frequency, but much less in intensity. The lower frequency also helps them become inaudible - our ears are much less capable of picking up distortions at lower frequencies. ;)

There is no perfect damping solution - but if you compared the sound of a cone with poor damping characteristics to one with good damping characteristics, you'd pick the cone with good damping characteristics every single time, believe me. ;)

It's not so much a problem, as "it's as good as it gets".
The same holds true for any [i]reference[/i] product out there - the boundaries can only be pushed so far, at any given point in time - until new innovations are discovered.
What cone material did JL Audio use?
A poly cone.
Now ask yourself "why"? Acoustics?
No - the complicated W-cone they designed mostly required it, due to the complexity and molding required. And even Poly is good enough, when you are talking about subs... even "reference" subs.

What cone material does JBL use on their flagship subs - the GTi's, with the DDD motors?
A VERY coarse, very heavy paper, that makes the Brahma look smooth as a baby's bottom, by comparison. Why? Pure acoustics.

Even look at high-end speakers if you will. Midrange drivers, where cone resonances are a much more serious (much more audible) issue:
DLS is a well known high-end manufacturer.
Look at their "Ultimate" series offerings. In there you'll find poly cones, and even an exotic carbon/kevlar honeycomb structure...
...and what does their Iridium set use? Their "Best of the best" flagship?
A heavy, course, unpressed paper cone. :fing02:

It may seem intuitively "low tech" - but that doesn't make it sonically inferior. ;)

"It's not the flattest. Dual gap technologies are... and since XBL^2 is scalable beyond dual-gap in theory, theoretically XBL^2 would be the flattest. "

I've never ever seen a dead flat XBL^2 motored anything. Not theory, I mean in a true speaker application, such as comparing the Brahma with the Audiomobile MASS 12 (underhung) I was referring to above. Brahma has a longer stroke, but BL isn't as flat or symmetrical until you get close to 15mm each way or so. Depends on your definition of flat, though, whether at the absolute limits or right past center. I'd prefer the latter with my intended use, but I'm not a bass junkie.

"Flat BL is going to help you out in SQ, at any excursion level. Elimination of intermodulation distortion - because there is no modulation of BL!
And none at 20mm means none at 15mm... at 10mm... at 5mm. "

Not arguing that, but there are other topologies that give a flat BL to a certain point. Namely underhung.

"Yes, underhung has a gentler BL curve shape than overhung, but it's not perfectly flat - where XBL^2 and DDD truly are."

Show me a perfectly flat XBL^2 or DDD driver and I'll take a picture of myself in a tutu and post it as my avatar. It doesn't exist.

"You see that out to about 10mm in either direction, that the underhung design is the closest match to the XBL^2 architecture in terms of BL linearity, but once that short coil even gets near either edge of the magnetic gap, since the magnetic field becomes increasingly weaker, BL does roll off - just like with an overhung sub, even LGLC overhung designs like the W7."

W7 is a completely different beast from a typical LGLC design. Can't generalize them, as you can't generalize XBL^2. W7 stays flat to its limits then experiences the dramatic BL rolloff that XBL^2 does past its limits.

"The 10W7 isn't flatter - you can even see with the 10W7 that the BL curve has a different shape on the in-stroke, compared to the out-stroke."

As does a Brahma.

"A heavier cone would tend to experience resonances later than a lighter competitor, because it will inherently have greater damping properties - [b]pushing those resonances down to a lower frequency, as well[/b]."

That's the problem.

BTW, I'm really enjoying this. Don't take it like I'm angrily arguing with you or anything, just sharing opinions :bigwink:

yea us spl guys have it easy!!! mmm hmmm!

[quote="Jonathan"]Really, though, lets say you set up your system power to allow a Brahma to reach full x-max at 20hz. This can hit 140s in the right application, but we'll say 135 because we know a Brahma is capable of this. Let's take it down to 123db, 12db down, which is still pretty loud, the excursion is going to reduce by around 8x (sticking to theory). So 3.5mm at 20hz, 1/4 that at 40hz, and so forth. Not a lot of excursion, and many overhungs can match it at that level. Remember we're talking SQ here :) Not that XBL^2 isn't incredible in it's ability to keep a flat BL, but you have to consider your application and whether you really need it. [/quote]
Flat BL is going to help you out in SQ, at any excursion level. Elimination of intermodulation distortion - because there is no modulation of BL! :D
And none at 20mm means none at 15mm... at 10mm... at 5mm. ]XBL^2 has inherent advantages that surpass many motor topologies overall. For SQ reasons I'd really prefer a good underhung design, but it would be insanely expensive if it were built like any of these subwoofers.[quote]Not so much expensive - well, it would be - but I think there are other physical limitations that are even [i]more[/i] prohibitive:
Underhung subs by nature are big, heavy beasts.
They require a massively thick top plate - because the excursion is dictated basically by the distance between where the voice coil windings end (when held "at rest" at the center of the gap) and where they would reach the end of that top plate when in motion.

To increase excursion, you must increase top plate thickness - and pole piece length - and since the top plate and T-yoke are the heaviest components on the sub, you are increasing the mass of the sub dramatically.
You could use a machined top plate, so that it increased in height only [i]inside[/i] the sub (TC sounds has done this for years), but then you have to start worrying about another issue...

...the flow of magnetic flux.
It's bad enough having a tremendously thick top plate, because the magnetic flux available will dissipate across that increased gap width, resulting in a weaker magnetic field as measured at any single point in that gap... and it's difficult to increase the magnetic energy to compensate for it... stacking magnets only increases magnetic energy around 10% per each you add - you need to increase the surface area of the magnet, meaning larger diameter (read: heavier, and more expensive) magnets, as well as larger diameter top plate and back plate... and you can only go as large as the cutout for the sub, ultimatelly... ruh, roh...

Not to mention, with a machined top plate, you have to worry about creating "elbows", or joints where magnetic flux is artificially constrained. The pole piece / backplate joint is always an issue on almost every sub - the most flux-flow resistive element in the circuit. With the machined top plate, you can create the same sort of situation - and [i]less[/i] flux isn't the goal!

[quote]underhung is flattest of all, just not cost effective since it doesn't integrate flux efficiently. Main benefit to XBL^2 is that it can be implemented for low cost compared to more exotic designs like the W7 architecture, underhung, and JBLs DDD technology.[/quote]
It's not the flattest. Dual gap technologies are... and since XBL^2 is scalable beyond dual-gap in theory, theoretically XBL^2 would be the flattest. ;)

Yes, underhung has a gentler BL curve shape than overhung, but it's not perfectly flat - where XBL^2 and DDD truly are. W7 gets close, thanks to it's XBL^2-esque pole-piece cross drilling - slightly changing the density of flux in that area of the motor, a mild version of what XBL^2 does [i]more effectively[/i] - inherently. ;)

Since we're speaking in generalizations, look at the conceptual comparison at the top of this tech paper on XBL^2:
[url]http://www.adireaudio.com/Files/XBL2TechPaper.pdf[/url]
You see in the comparison, that all subs have been "normalized" to all "be 15mm Xmax subs", to be fair.
You see that out to about 10mm in either direction, that the underhung design is the closest match to the XBL^2 architecture in terms of BL linearity, but once that short coil even gets near either edge of the magnetic gap, since the magnetic field becomes increasingly weaker, BL does roll off - just like with an overhung sub, even LGLC overhung designs like the W7. ;)

[quote]The 10W7 is on there, not 12W7. And on the dumax, it's still a flatter BL curve than a Brahma. I doubt Adire did the favor of smoothing it for them. ;) The most impressive shown on there IMO is the Audiomobile MASS 12. Benefits of an underhung design, tends to have flatter BL than all of them.[/quote][/quote]It used to be, I have it somewhere, unfortunately not on hand at the moment... I'll have to remember to look for it when I can.
The 10W7 isn't flatter - you can even see with the 10W7 that the BL curve has a different shape on the in-stroke, compared to the out-stroke. And the ripples in the 10W7's BL curve are every bit as prominent as the Brahma - and look, we're taking about a fraction of 1 BL value here... totally ignorable in either case (you had mentioned it previously as a concern). ;)

Also take into account the differences in the scale of both the X-axis (excursion) and Y-axis (BL). This somewhat magnifies the ripples that you see, compared to the more compressed W7 plot.

Ultimately, you are talking about two subs with a flat BL and the same excursion level... the Brahma is simply more symmetrical in stroke, with it's more direct implementation. The W7 uses a similar approach - not technically two gaps, but setting up the magnetic flux in the same way by cross-drilling the pole vent... different technology? Not really. ;)

[quote]No, not transient response. Any heavier cone is going to experience resonances before a lighter competitor, and it doesn't give you a lot of flexibility in design. A heavier cone would [i]tend[/i] to experience resonances [i]later[/i] than a lighter competitor, because it will inherently have greater damping properties - pushing those resonances down to a lower frequency, as well. ;)

Not much different than sheetmetal on a car - thin sheetmetal rings, buzzes when tapped or rubbed.
Then, you cover that sheetmetal with heavy Dynamat.
Now, it no longer rings and buzzes.
A heavy cone isn't much different. ;)