"W" Sandwich Structure Speaker Cone/Speaker Cone Profile

Hello, I've read a lot about sandwich structures and some renowned companies have used them.

I'm currently working on a project of mine to make sandwich cones for my speakers.

The materials are:
Faces: Kevlar carbon fiber hybrid
Core: Low density expanded polystyrene (LDEP)

I want to know if there is another sandwich structure or better material than those I mentioned.

Currently I think the best sandwich structure for speaker cone.
I want to know how the speaker cone profile can affect the sound performance.

The speaker cone profiles I know of are:

Concave cone, widely used for Subwoofer, Woofer and Mid-Ranger.

Dome cone, often used for Tweeters and a few Mid-Rangers.

Flat cone, somewhat unknown, used for Woofers and Mid-Rangers.
The photo (image) below is of the Nautilus by Bowers & Wilkins.

Note that the mid-bass is a flat profile cone.


  • images (1).jpeg
    images (1).jpeg
    27.2 KB · Views: 67
I do not know what foam core is used by such companies as Focal or Eton in their sandwiched drivers, but the answers as usual should be : the ligther and stiffer and most damped foam structure. Then choose you two external driver sides devil : A german company made a nice aluminium with inside foam core line of drivers. I would be curious about non symetrical material surfaces : for instance : external thin aluminium sheet - foam core - internal side of the driver with something more damped though still stiff (glass fiber ?).

Anyway you should find some nice sandwich from Focal but alas in the car line.
Last edited:
Rohacell is a newer foam core - seems to be a technical update instead of Polystyrol. BW used Rohacell for their bass drivers.

However I was thinking about a Polystyrol core worked with CNC machines to be a honeycomb structure. Then layerd between alu or other material.

But you have no access to most high tech materials as a DIYer (do it yourself).

You normally have no access to manufacture parts of loudspeakers in order to assemble them yourself.
  • Like
Reactions: 1 user
It is my understanding that non-paper/non-conical drivers have very peaky resonances requiring steep filtering to tame them, so keeping them out of the intended pass band will be the key to success, or not! The SEAS magnesium cones spring instantly to mind. Given that paper cones can be made to work so well, it is probably a purely academic exercise to do otherwise, but a fascinating project nonetheless and I wish the OP well.
Eton has managed to tame the resonances in their Kevlar honeycomb sandwich drivers.

Maybe due to special glue I guess.

So it's not true that only paper can have good resonance behavior.

Also the HDA drivers from Audax were similar to paper.

They took high tech fibers and mixed them like paper pulp. And you got the good damping from paper cones but much stiffer cones.
In post 4 in this thread, follow the link and there is a citation from Mr. Barlow


So you do not need to seek, here the history and theory. I read in a compendium of articles from the Audio Engineering Society AES, about it:

From theory it is known that the combination of metals and paper can bring a lot more stiffness to the diaphragm (known since the 60ies, but history of loudspeaker building describes this in patents since the 30ies of the last century - it means right from the beginning of loudspeaker production of the dynamic type). For germany it was Dr. Emil Podzus (aluminium foil and hard foam), later it was english speaking D.A. Barlow in the 60ies with duraluminium sheets and expanded polystyrene of thick quality. In germany these aluminium foiled loudspeakers are known as Podzus-Görlich loudspeaker drivers. Mr. Görlich was helping out as a young man at the factory of Mr. Podzus. He later produced these diaphragms with aluminium foil and hard foam. He does it until today.
I measured the loudspeaker diaphragms paper + aluminium very often. One sided and both sided (last means it is a sandwich-cone). Coating of the driver on just one side brings resonances down - makes the speaker sometimes more linear - but always the loudspeaker driver becomes much more stiff and has therefore a significant better impulse reproduction.
Best is this checked by ear while covering one driver with aluminium foil and leaving the second stereo speaker undoped. You can then use your balance knob on your amplifier for an A / B listening test.
Here about the theory:
D.A. Barlow: The Development of a Sandwich-Construction Loudspeaker System. Page 159-171. From the AES anthology of articles on loudspeakers, article written 1970.
Citation begin (page 160):
"As paper cones are thin, these resonances are bending modes. The stiffness of a material in bending, for any given geometry and edge condition, is proportional to Young`s modulus and the cube of the thickness, ignoring variations in Poisson`s ratio. ... On this basis, paper is much stiffer than metals (except beryllium) in spite of its much lower modulus, and lower density materials such as expanded plastics are even stiffer, as may be seen in Table. A method of obtaining still greater stiffness is sandwich construction. ... In bending, the maximum stress and strain occur at the outer fibers, the material at the neutral axis being unstressed. Better use of material can thus be made by concentrating it at the outer fibers. A familiar example is the tube. In the case of large areas, the same effect is obtained by using a thin high-modulus material for the outer surfaces, and a light-weight material or form of construction for the core. ... This sandwich will of course be much stiffer than the same total weight of either material used separately. The skin material should have the maximum ratio of modulus/density. Beryllium, the best material, is impractical due to difficulty of rolling and possible toxicity, so aluminium is the obvious choice. The core should be as stiff as possible in the thickness direction and have minimum density. Honeycomb aluminium or impregnated paper are frequently used in aircraft construction and could be used for flat diaphragms."
Citation end.
Joined 2009
Paid Member
Thank you,
I read it and it made obvious points about trying to limit membrane breakup modes and dampen their effects.
All in all i'm seing this as a Constrained Layer Damping approach to membrane, useful where it become fractioned ( at breakup where modes on membrane happen, limiting their strength).

Have you played with different kind of glue? For his CLD approach ( on box but principle remains the same) Earl Geddes used melamine glue on which he incorporated micro balls to optimise even further loss occuring within the soft layer.
I think it would be difficult to implement with loudspeaker membrane but still it's interesting approach.

I agree with you about severity of breakup wrt materials, it can be mitigated.
Some carbon membrane i've heard had hard resonnance ( Davis drivers) it's true but my honeycomb aluminium drivers ( i own the Technics loudspeakers i linked previously) are not this much an issue for the woofers / mid drivers wrt breakup and far less an issue than pure aluminium membrane drivers..

It's apple to orange comparison but i think it comes down to construction principle and realisation and this kind of approach can have benefits.

By the way i like the sound of the Technics which is overall very neutral for loudspeakers of this area ( 1980). They have issue from the acoustic design choices made though ( a 15" played to 750hz and crossed to a 4" imply a mismatch in radiated power... even lowered by half an octave thanks to dsp it's still a mismatch in directivity behavior and limit max output level...).