Pistonic cone materials

[noah katz]

In the TCSounds new driver thread, AJ said

"Pro units and pistonic cones = oil + water."

Does AJ or anyone have any evidence that metal is more pistonic than paper (cones are actually a cellulose "alloy")?

Pistonic behavior/relative stiffness is indicated by the first breakup freq, except where it's damped out of existence.

If you make a beams or plates (which structurally is just a 2-D beam) of wood and metal that are the same width and weight, the wood one will be far stiffer because of it's greater height; stiffness increases with the cube of the height.

As an aside, it's always been a mystery to me why, if you're set on a metal cone (which I believe is marketing driven as a result of people's "intuitive" experience of metal), why more aren't made of magnesium.

It would be the best of all metals (except beryllium) because it's the lowest density (thus thicker/stiffer for same the weight, per above), and has internal damping an order or magnitude higher than other metals.

[cotdt]

metal cones are not pistonic either! except metal-coned subwoofers in their passband. if metal cones were pistonic, the FR would be flat. but 7" metal-coned woofers are only flat to around 600Hz or so.

[Dennis H]

Noah, I agree that you can't generalize about cone materials. You need to measure each driver individually. But the measurements show (okay, here I go generalizing ) that many metal cones are more pistonic in their passband and many paper cones have more breakup in the passband but usually have a wider useable passband than the metals.

Useful measurements include:

1. Distortion sweeps.

2. Jon's favorite of measuring frequency response with the mic almost touching the cone and then moving the mic across the cone. Where the curves diverge indicates the frequency where the cone starts being non-pistonic.

3. Impedance sweeps. Bobbles may indicate a breakup.

If you go to the Seas Vintage page, you can download spec sheets for identical Excel drivers with different cone materials and see some of those differences.

But, all that said, the Peerless Exclusive (Nomex) and the Seas Excel (mag) drivers both perform quite well with very different cones.

[Dennis H]

if metal cones were pistonic, the FR would be flat.

What proof do you have of that?

but 7" metal-coned woofers are only flat to around 600Hz or so.

What proof do you have of that?

[cjd]

I think metal cones happen to hit certain levels of pistonic behavior within a given size, weight, shape, etc. relationship better than many other materials - probably also importantly, it does so within a certain ease of manufacture and cost.

If we look at Adire's Parthenon demo, IIRC they were using a honeycomb core carbon fiber structure or something. High cost, but capable of withstanding the crazy motion and outputting at low frequency. how well would this do at 200Hz, 2kHz, 20kHz?

It's a balancing act, as with almost everything else in sound reproduction. Life for that matter.

C

[noah katz]

"metal cones are not pistonic either!"

Yes, I should have said pistonic (implyies behaving as a rigid body) up to it's first mode.

Which is also overgeneralizing; looking at the generalized AR (amplitude ratio) vs. freq graph in my vibratiions text, amplitude ratio (output/input) is 1 at 0 Hz and rises to Q at f(1). For Q =2, AR is about 1.5 at .5 of f(1).

This means that strictly speaking there is piston behavior only at 0 Hz. So the higher f(1) is, the more pistonic it is.

[AJINFLA]

Does AJ or anyone have any evidence that metal is more pistonic than paper (cones are actually a cellulose "alloy")?

http://www.infinitysystems.com/homeaudio/technology/whitepapers/cmmd.pdf

Pistonic behavior relative stiffness is indicated by the first breakup freq, except where it's damped out of existence.

Rubbish. It is very much still in existence, just spread out and not as easily discernible in the FR.

Pro units and pistonic cones = oil + water.
Almost every pro woofer uses (very) low mass paper cones in order to a) achieve high sensitivity b) have good dispersion vs size.
Which means...cone flex.

Very cool....for me at least

I'm trying to be clear that metal cone pro units aren't for everyone. Hence them being practically non-existent..till now.

cheers,

AJ

[noah katz]

http://www.infinitysystems.com/home...papers/cmmd.pdf

Thanks for the intersting read.

I agree that the CMMD is a super material, as sandwich structures tend to be.

I don't get their connection of 1st mode with sound velocity within the material; it's a function of mass and stiffness.

Maybe sound velocity ends up with the same relation, but why be obtuse?

Using their numbers, I calculate that if the same weight cones were made from paper and aluminum, the paper would have 3.7 times the bending stiffness, and thus a 1st mode 1.9X higher in freq.

So the advantages they show for the stiffer denser materials are not apples to apples because they haven't been made with the same cone mass to give equal efficiency.

> Pistonic behavior relative stiffness is indicated by the first breakup freq, except where it's damped out of existence.

"Rubbish. It is very much still in existence, just spread out and not as easily discernible in the FR."

I in no way meant the mode isn't there, just that it can hidden by damping.

I seemed to have dropped a word or two in my response; I didn't mean that that the mode is damped out of exisitence, but that its effect in FR is.

[David G]

For a given moving mass, aren't metal cones thinner? Does that mean then that more internal energy in the box escapes through the cone? B&W state in one of their white papers that they found that paper/epoxy cones have more impact than metal cones.

[noah katz]

Yes, the same flexibility that lets the modal vibrations happen, lets sound pass through, with the additional factor that a heavier cone is excited at reduced amplitude.

[AJINFLA]

Does that mean then that more internal energy in the box escapes through the cone?

What box? :W

B&W state in one of their white papers that they found that paper/epoxy cones have more impact than metal cones.

"more impact" is a bit hazy to define in terms of accuracy. Heck, "accuracy" is tough enough :roll: .
I have no problems with using heavy paper composite cones & boxes for very deep bass duties. Since that is exactly what I use.

cheers,

AJ

[Jim Holtz]

http://www.infinitysystems.com/homea...apers/cmmd.pdf

Rubbish. It is very much still in existence, just spread out and not as easily discernible in the FR.

Pro units and pistonic cones = oil + water.
Almost every pro woofer uses (very) low mass paper cones in order to a) achieve high sensitivity b) have good dispersion vs size.
Which means...cone flex.


I'm trying to be clear that metal cone pro units aren't for everyone. Hence them being practically non-existent..till now.

cheers,

AJ

Hi AJ,

Interesting link. Thank you for providing the reference. I, like you, am a big fan of metal cone drivers. I also have really liked Kevlar and carbon fiber based cones too. Audax used to have a chart on their website (long gone) that ranked cone material which showed Kevlar and carbon fiber coned drivers to be very close in stiffness to metal. The chart in the Infinity link does not.

Do you have any other links to information of this type?

Thanks!

Best regards,

Jim

[David G]

What box? :W

AJ

Fair call

[noah katz]

"Audax used to have a chart on their website (long gone) that ranked cone material which showed Kevlar and carbon fiber coned drivers to be very close in stiffness to metal."

From a design handbook:

Material modulus (Mpsi) (for composites, 0/90 deg fabric+resin)

E glass 2.7
S glass 4.1
Kevlar-49 4.9
magnesium 6.5
aluminum 10.9
carbon fiber 7 to 15 (4 types)
titanium 16.4
steel 30

[Jim Holtz]

"Audax used to have a chart on their website (long gone) that ranked cone material which showed Kevlar and carbon fiber coned drivers to be very close in stiffness to metal."

From a design handbook:

Material modulus (Mpsi) (for composites, 0/90 deg fabric+resin)

E glass 4.1
S glass 4.1
Kevlar-49 4.9
magnesium 6.5
aluminum 10.9
carbon fiber 7 to 15 (4 types)
titanium 16.4
steel 30

Thanks Noah. That looks similar to what I remember. Since most cones are a composition of different materials, I'm not sure if any real decisions can be reached by comparing data, but it's fun to look at. Paper isn't listed. Do you have any idea where coated paper drivers fall in this line up?

Thanks again!

Jim

[cotdt]

diamond or cabon nanotubes would be nice

[noah katz]

You're welocome, Jim.

BTW, I corrected the E glass number, it's 2.7 Mpsi.

I don't know what paper is, but woods are from .5 to 1 Mpsi, I'd guess papers aren't too different.

[Dennis H]

Well, nobody responeded to my comments on comparing Seas mag and paper cones so here are some small pics. To see larger ones get the pdfs from the Seas page. These are both Excel 8" drivers, same motors, one magnesium cone and the other paper.

Note the breakup and/or cone edge resonance on the paper cone that shows up as an impedance bobble and a dip in frequency response. It also has a 2nd harmonic above 1% between 400 and 700 Hz. I'd attribute the second order midrange stuff of the paper cone to a lack of stiffness leading to different behavior on the push and pull strokes -- like the difference between pushing and pulling on a string.

By contrast, the harmonics are below 0.3% below 1000 Hz for the mag cone but there's a 3rd harmonic spike up to 1% at around 1500 corresponding to the 4500 breakup.

Clearly, the mag cone is superior below 1 kHz although it could be argued that the paper cone might be better if you need a high crossover point. For the purposes of the discussion here, the mag cone behaves more "pistonically" below 1 kHz, whatever the reason.

Paper cone:



Mag cone:

[AJINFLA]

Do you have any other links to information of this type?

Hmm, I seem to recall seeing some over the years, but finding them is another story Jim. I'll search again one of these days.

Clearly, the mag cone is superior below 1 kHz although it could be argued that the paper cone might be better if you need a high crossover point.

Sure Dennis. But so would CMMD. Or at least according to Dr. Toole .
IIRC, some inexpensive Infinity's got a decent review in S-pile and I believe the XO's are paperish high.
Too bad Harman can't hang a DD motor on a CMMD cone. Shame.
But back to TC. Again, hooray for them taking the lead here with these aluminum cones. I can only hope that we start to see JBL, B&C, 18sound, BMS, etc. take the hint and come up with some of their own (at mass production prices).
I'm not holding my breath, but hey, I'm finally starting to see more coaxials, so maybe progress is slowly on the way.... :W

cheers,

AJ

[David G]

Well, nobody responeded to my comments on comparing Seas mag and paper cones so here are some small pics. To see larger ones get the pdfs from the Seas page. These are both Excel 8" drivers, same motors, one magnesium cone and the other paper.

Note the breakup and/or cone edge resonance on the paper cone that shows up as an impedance bobble and a dip in frequency response. It also has a 2nd harmonic above 1% between 400 and 700 Hz. I'd attribute the second order midrange stuff of the paper cone to a lack of stiffness leading to different behavior on the push and pull strokes -- like the difference between pushing and pulling on a string.

By contrast, the harmonics are below 0.3% below 1000 Hz for the mag cone but there's a 3rd harmonic spike up to 1% at around 1500 corresponding to the 4500 breakup.

Clearly, the mag cone is superior below 1 kHz although it could be argued that the paper cone might be better if you need a high crossover point. For the purposes of the discussion here, the mag cone behaves more "pistonically" below 1 kHz, whatever the reason.

Generally audibility of 2nd harmonic is about 1% - see http://www.audioholics.com/techtips/...Distortion.php

so maybe this is academic. I'd be more interested in a CSD comparison in the 500Hz- 2kHz range.

[Andrew M Ward]

Almost every pro woofer uses (very) low mass paper cones in order to a) achieve high sensitivity b) have good dispersion vs size.
Which means...cone flex.
AJ

What if low sensitvity is the goal (of course, not pro-gear) because of more neutral tonal balance and truer flat responce?

any thoughts?

[Dennis H]

Generally audibility of 2nd harmonic is about 1% - see http://www.audioholics.com/techtips/...Distortion.php

so maybe this is academic. I'd be more interested in a CSD comparison in the 500Hz- 2kHz range.

Well, that's one opinion. I wouldn't necessarily quote audioholics as a definitive reference. CSD is just another way of looking at frequency response. No additional information, just another view of it. See the old MAD thread for all the ugly math.

Edit: I was in a hurry last night. Now I've had a chance to look at the audioholics article and it's okay as far as it goes. (I wonder if he got permission to use SL's graphs?) So, let's say 1% (-40dB) is the threshold of audibility. Keep in mind that the Seas tests are at 96dB/1m. That's 84dB/4m which isn't all that loud. Cranking the volume up more will increase distortion.

[jkrutke]

Well, nobody responeded to my comments on comparing Seas mag and paper cones so here are some small pics. To see larger ones get the pdfs from the Seas page. These are both Excel 8" drivers, same motors, one magnesium cone and the other paper.

Good post Dennis, I was going to respond with the same sort of thing from my own measurements. Unfortunately, the bad thing about reading these forums at work is that I rarely have time to respond, so you beat me to it.

One recent example in my 5.5" driver comparison is the Seas Nextel vs the Mag Excel. The mag is far smoother. (and of course usable higher up) I could come up with a lot more examples of this behaviour. I have to agree that in general metal cones are "pistonic" higher in frequency.

I have to wonder if a metal version of the Vifa XG/XT18 would have the 400hz dip that they currently have. A metal version would probably have lower bass distortion too.

[noah katz]

Thanks for the mag/paper comparison, Dennis.

What are the Mms of the two drivers?

[Dennis H]

Mms is 26g for the paper and 29g for the magnesium. Aside from material properties and cone thickness we also need to consider overall cone shape. I'm pretty sure the mag cones are straight tapers but the paper ones may be a curvilinear shape (many paper cones are).

http://www.seas.no/excel_line/excel/E004.PDF

http://www.seas.no/excel_line/excel/E013.PDF

There are too many variables to draw conclusions based just on cone material. You need to look at the individual driver and see how it performs. Jon's whole "pistonic" philosophy involves finding drivers that perform exceptionally well within a certain limited frequency range and using steep crossovers and/or notch filters to strongly suppress the frequencies where the drivers misbehave badly. Other designers eschew steep filters and will accept a certain amount of misbehavior in the passband in order to avoid really bad behavior in the stopband.

You pay your money and you make your choices.

[Piotr]

Generally audibility of 2nd harmonic is about 1% - see http://www.audioholics.com/techtips/...Distortion.php

so maybe this is academic. I'd be more interested in a CSD comparison in the 500Hz- 2kHz range.

Such a test is about the threshoold of pure 2nd order on a sine wave. In a system playing wide BW material things change a whole lot with all kinds of modulation products. IOW a midrange driver that peaks up to 1% or more HD will have a "very" colored sound.

Even though it´s tempting to use the term "pistonic" it´s of course not really correct. Stiff drivers flex as softer paper and plastic ones does. They generally flex less though and that means less distortion.

/Peter

[capslock]

Such a test is about the threshoold of pure 2nd order on a sine wave. In a system playing wide BW material things change a whole lot with all kinds of modulation products. IOW a midrange driver that peaks up to 1% or more HD will have a "very" colored sound.

Even though it´s tempting to use the term "pistonic" it´s of course not really correct. Stiff drivers flex as softer paper and plastic ones does. They generally flex less though and that means less distortion.

/Peter

Another aspect is how much energy gets dissipated (converted into heat) in the flexing. Paper and highly dampled polys tend to have significant bending hysteresis, so the losses will vary within the cycle and of course with the amplitude of the driving signal.

Metal has much lower internal damping, so even when it flexes, the losses are still proportional to the amplitude of the signal (or the flexing going on), so they may show up as time smear (linear distortion), but not as nonlinear distortion.

The venerable Peerless HDS sandwich cones also seem to be an example of a floppy, reasonably well damped material with low hysteresis. All the old HDS and CSX drivers have low second harmonic in the 500 - 1200 Hz flexing region, and the phase plugged versions, which also sport a better motor, have exceptionally low distortion for 2nd to 5th.

[AJINFLA]

Eric, where have you been? Did you see Paul Hilgemans W22 coax? https://www.htguide.com/forum/showthread.php?t=20478
Did you ever get around to using/making a neo dome suitable for coax mounting?

cheers,

AJ

p.s. I wonder if Bjorn saw it too? :W