What is the Reason For This Driver Design (Large Dust Cap)

[Arindal]

Please get yourself an R type KEF speaker and investigate how the surround is mounted.

I have seen an exploded sketch. Looked like an inverted surround to me, glued (?) to the inner and outer cone alike with the latter showing a little fold (you mean this one?). Seemingly a similar design like Peerless SBS-160F35AL01-04 which is another prominent example of a ´inverted dome replaces dustcap´ unit and one of the xmax record holders.

while high fs is beneficial for bass reflex, as it mitigates the problem with excursion explosion below tuning

I doubt that high fs is always advantageous for vented designs.

 

[AnalogSteph]

I doubt that high fs is always advantageous for vented designs.

Which seems justified, given that drivers have a habit of starting to do funny things around fs at high levels, like response peaking. (If anyone has a good explanation for this phenomenon, I'm all ears.)

Anyway, excursion basically is a function of only two things, Sd and SPL output at a given frequency. So a driver that natively has less bass output (that being the higher fs unit) is also going to have less excursion, but you could achieve the exact same effect with EQ.

I quite like the additional degrees of freedom provided by these modern times. TSPs still dictate your maximum SPL output but frequency response can be bent a fair bit, so you get a lot more leeway in what drivers you can actually use. This has enabled things like actually good budget subs (the $99 Monoprice model comes to mind, probably not as inexpensive as that for much longer) or 3" class desktop monitors that can get to 50 Hz flat at modest levels.

 

[Salt]

Which seems justified, given that drivers have a habit of starting to do funny things around fs at high levels, like response peaking. (If anyone has a good explanation for this phenomenon, I'm all ears.)

Fs is the resonance frequency of the speaker, so theoretically there is only low energy needed to make the system oscillate.
https://engineeryoursound.com/what-is-speaker-fs-fundamentals-explained/

 

[kyuu]

Some more insights

Your URL is broken, appears to be copy/paste error.

 

[Salt]

Your URL is broken, appears to be copy/paste error.

Corrected and link updated.

 

[DSJR]

Edit : He had the Kef 104aB. The passive radiator is the B139.

The passive KEF radiator in the Cadenza and 104ab (I don't think they were hugely different was the BD 139... In the 104, it flopped about all over the place with then-not idea vinyl sources, but the 104ab as a 'system' came to life in a wonderful way with master-grade and digital I remember.

As for large centre caps, I suspect the solid ones as per Dynaudio and others, act as radiators at higher frequencies and in ATCs case with the 19 and 20 driver family, the centre domed dust-cap acts as a mid driver, 'crossing over' (a bit too obviously by today's standards) mechanically at around 600Hz, just the frequency where the upper mid starts to rise in these models.....

 

[Keith_W]

So is it a rule that bigger dust caps mean bigger voice coil? And are bigger voice coils always advantageous?

A bit like Henry VIII's codpiece, you mean.

 

[Chr1]

The passive KEF radiator in the Cadenza and 104ab (I don't think they were hugely different was the BD 139... In the 104, it flopped about all over the place with then-not idea vinyl sources, but the 104ab as a 'system' came to life in a wonderful way with master-grade and digital I remember.

As for large centre caps, I suspect the solid ones as per Dynaudio and others, act as radiators at higher frequencies and in ATCs case with the 19 and 20 driver family, the centre domed dust-cap acts as a mid driver, 'crossing over' (a bit too obviously by today's standards) mechanically at around 600Hz, just the frequency where the upper mid starts to rise in these models.....

At risk of sounding stupid, the bit I don't get is how both concave and convex dustcaps can both work.
They are kind of the opposite, when it comes to the dissipation/radiation of sound waves, and a major part of the diaphragm when large, so how come they both work?
And what are the differences, pros/cons, if any etc?

 

[Arindal]

drivers have a habit of starting to do funny things around fs at high levels, like response peaking. (If anyone has a good explanation for this phenomenon, I'm all ears.)

Can confirm this from listening to many examples, but cannot really give a valid explanation. Seemingly drivers which are incarcerated into an enclosure way too small pushing the resulting resonance frequency to kick band (around 100Hz is worst) tend to show strange and unpredictable dynamic behavior at different SPL.

So, yes you can boost the level between driver resonance and port tuning to get a perfect FR, but it might sound different at different levels. One explanation might be the stiffness of air enclosed which combined with compression also tends to drift the port tuning depending on the level. Whenever I see minimum active driver excursion frequency drifting at different SPL, it is a big red flag.

So a driver that natively has less bass output (that being the higher fs unit) is also going to have less excursion, but you could achieve the exact same effect with EQ.

Certainly true for a sealed enclosure. For a reflex design which is non-standard, i.e. high resonance of the active driver and lower port tuning than calculated, the equation is changing as the port also shows decreased excursion, but distributed over a much broader frequency range. If you apply boost on such a construction you have active driver and port playing in parallel over a broad band. Not ideal in my understanding.

Fs is the resonance frequency of the speaker, so theoretically there is only low energy needed to make the system oscillate.

That is true, but usually higher efficiency, i.e. less energy needed to oscillate, is expressed in a sharp peak in impedance exactly at Fs, while the output level stays relatively proportional to input voltage. At least if the total quality factor is not overly deviating from the desires area 0.5...0.75

So, to apply all this on the OP´s initial question:

bigger dustcaps coming with a bigger voicecoil are just indicative of a different driver design. Same is true to high-excursion woofers as a measure of giving stability to the moving system. I am reserved when it is coming with shallow woofers or high-fs/Qts designs.

Regarding AnalogStephs comment on micro- and high-excursion drivers: It is indeed astonishing what modern driver designs can achieve. But I noticed that most of really capable units to produce bass from a small enclosure and little diaphragm area are actually showing a pretty deep and long motor design (may it be with a double spider or extended VC former or alike.

I don't get is how both concave and convex dustcaps can both work.
They are kind of the opposite, when it comes to the dissipation/radiation of sound waves, and a major part of the diaphragm when large, so how come they both work?

For lower frequencies, hence longer wavelengths, the diaphragm geometry is not really relevant, so they both produce a very similar wavefront. The shape of the dustcap or second cone is in woofers and midwoofers rather a question of rigidity and geometry of the whole design.

It is very relevant for upper midrange and treble, though, particularly with designs having no equally rigid diaphragm (like the aforementioned ATC, or BMRs), or driven by a large voicecoil with any diaphragm that is not completely stiff (as they show a tendency towards ring-radiator behavior).

 

[Chr1]

A bit like Henry VIII's codpiece, you mean.

View attachment 450939

Indeed. He was most likely a bit "rapey."

 

[Salt]

At risk of sounding stupid, the bit I don't get is how both concave and convex dustcaps can both work.
They are kind of the opposite, when it comes to the dissipation/radiation of sound waves, and a major part of the diaphragm when large, so how come they both work?
And what are the differences, pros/cons, if any etc?

Imagine the air in front of a transducer lying on it's surface like a carpet, and then move the membrane: would a convex make any difference to a concave membrane?

 

[Chr1]

Yes, but the effective shape of the diaphragm making the soundwave is fundamentally different, so how can they make the same soundwaves? (Basic physics.)

Any benefits of a concave/convex dustcap?

 

[Salt]

The soundwave will be cylindric from the surface of the mebrane, and because we're talking about bass, and 1 meter wavelength corresponds to 340 Hz, 5 cm more or less are irrelevant.

 

[Jaxjax]

One of the characteristics of a Dynaudio driver is the tabs between the dust cap and the cone, where the voice coil is attached, in a way that the cone and dust cap can be formed in one piece of polypropylene. The design of these tabs solved a couple of problems (how to attach the coil + how to prevent too soft attachment of the dust cap part hence preventing phase issues) is the brainchild of Mark Allan Thorup. C/P

 

[somebodyelse]

Yes, but the effective shape of the diaphragm making the soundwave is fundamentally different, so how can they make the same soundwaves? (Basic physics.)

When the diaphragm is small relative to the wavelength its shape doesn't make much difference. When the size of the features gets similar to or larger than the wavelength it becomes important, but by then you've often got diaphragm resonance modes to consider too. At school we did experiments with a vibrating dipper making waves in a tray of water to see similar effects from the size of the dipper, the size of a gap in a barrier etc. Someone might have published some animated models online somewhere - there's probably too much learning involved in modelling yourself in something like akabak.

Any benefits of a concave/convex dustcap?

Maybe, but it would depend on the specifics of the rest of the cone, probably involving its resonance modes. Wideband / 'full range' drivers often make use of controlled diaphragm resonance modes to extend the usable upper frequency range.

 

[Heinrich]

Indeed. He was most likely a bit "rapey."


View attachment 450940

Yeah, this discussion, again, gets a bit astray. There is so much half-informed belief in the DIY scene, that you cannot expect from experts dealing with data all day long. I don't want to pick on somebody in particular, but air layers, really? I leave it at that.

 

[Arindal]

The soundwave will be cylindric from the surface of the mebrane

Would rather call it spherical, if the diaphragm is a pointsource, small in all dimensions compared to the wavelength. Cylindric wavefront would be expected from a source which is small in two dimensions, but large in the third, compared to the wavelength, i.e. an ideal line source.

the effective shape of the diaphragm making the soundwave is fundamentally different, so how can they make the same soundwaves?

Sound pressure waves move with the speed of sound to all directions where there is not obstacle. If the wavelength is big enough compared to the diaphragm size, i.e. low frequency, during a fraction of the period the air has enough time to get evenly distributed in terms of pressure and follow any pattern of the diaphragm, and even ´evase´ to all directions without an obstacle outside the diaphragm area (the latter phenomenon is diffraction).

Any benefits of a concave/convex dustcap?

As mentioned for bass and midbass drivers, there is no difference to be expected in terms of dispersion. The shape of the dustcap is mainly a question of rigidity of the whole moving part, which area it has to cover (in case of larger voicecoil diameters for example) and of course how it can be practically produced.

 

[Chr1]

Thanks. OK, it seems my basic knowledge of soundwaves was lacking. Interesting.

 

[teashea]

No.
DSP can correct and refine, but never will substitute the basics of TSP.

... but DSP can work with good TSP to produce superlative results, beyond what TSP can do alone ...... eg Neumann KH150's.

 

[Heinrich]

... but DSP can work with good TSP to produce superlative results, beyond what TSP can do alone ...... eg Neumann KH150's.

The Thiele/Small parameters describe the speaker and enclosure combined as a filter. In this theory the whole doesn't have mechanical or other limitations. Instead, it just follows the math, theoretically.

Filters of whatever design can be concatenated with other filters. The sequence and mutual association doesn't matter. That is what contemporary drivers (in p/a and studio) are made for since the 1990s. The need to be supported by electronic/digital, additional filters. E/g a subsonic filter to prevent overexcursion below the bassreflex tuning, and to combine this with a bit of a peaking just a above the cut-off, which is ironed-out by the subsonic, in total having the very exact, best extension down low.

One might argue that extreme designs hit the limitations earlier, e/g too much power needed to drive the enclosure, but that is decidedly not (no, no) something originating in the T/S parameters. The latter only allow to calculate, theoretically, the filter's properties, any filter! What people do is to mess up the means to describe something, with what is described.