Why the rule on c2c spacing???

[OlogyAudio]

Hello I've had this little question burning at the back of my mind for awhile now...

So I know the near field line array rule is 1/2 of a wavelength c2c for cone/domes and from the edge of a ribbon element... I've heard the explanation that since the "coil" is the drive element we view the center of the cone as the point source, and a ribbon the entire element is excited so that’s why we view the entire element as the source...

I'm just a bit confused though since the voice coil is exciting the entire surface of the cone and this wave propagates through the cone at an extremely high speed, effectively making the entire surface of the cone the wavefront... Wouldn't that mean that the same rules apply to conical speakers as ribbons??? Am I missing something? Please fill me in.

Thanks,
Dylan

[Hartono]

Hi,

for flexible cone , the higher frequency tend to come only from the area near the voice coil.

for infinitely stiff cone, your model is correct.

stiffer matelrial flex less ( ceramic , diamond )

paper cone flex more.

Hartono

[OlogyAudio]

Even through paper though the wave will propagate between 1600-4000 m/s depending on the blend.

Compare that with the propagation in the air where the very first sound was generated from the cone near the voice coil at a whopping 343 m/s.

For a drastic oversimplification of the scenario, say the cone is 25mm deep and has an internal wave propagation speed of 1600 m/s:
in a 9" diameter cone, the sound that was emanated from the area near the voice coil will be at +25mm at the exact same time the rim of the cone is making its first sound waves... If I did my math right...

Now if theres breakup modes this waveform will have ripples in it that will sum out and lead to nulls and peaks at different frequencies, but nonetheless it is defiantly not an expanding point source in the air that was created only at the voice coil ...

Now if only I had a free FEA tool to play with that would do all these calculations for me all while simulating the breakup modes and summing everything together in terms of frequency response at different listening positions..... I wonder if you can use this mechanism to improve the off axis response a bit ...

But anyways... I think some measurements are in order??? Do you know if anyone did any measurements on line arrays and worked backwards to get to the 1/2 wavelength c2c thing??? Or even some measurements demonstrating comb filtering of an improper linearray?

[Javachip]

I am no expert, but I'll take a shot at this. For higher frequencies, most of the sound is produced near the voice coil, because that is the only part of the diaphragm that is actually vibrating and setting air in motion.

Imagine an 18 inch paper woofer trying to reproduce a 10 kHz tone. Obviously it can't. The big paper cone has so much inertia, and is so flexible, that it can't vibrate that fast. Only the innermost part of the cone, closest to the voice coil, will be able to vibrate that fast (with a lot of distortion).

Now imagine an 18 inch solid diamond (absolutely rigid) woofer trying to play 10 kHz. There is no breakup of the diaphragm, but its mass and inertia prevent any part of it from moving at all. The voice coil, which is rigidly attached to the cone, simply remains motionless and heats up in the magnetic field.

Direct propagation of sound waves through the diaphragm material is very small compared to sound produced by in-and-out motion of the diaphragm/voice coil unit.

I'll shut up now and let you teach me something -- you designed and built your own driver, which I can only daydream about at this point.

[Ron E]

Originally posted by OlogyAudio
Hello I've had this little question burning at the back of my mind for awhile now...

So I know the near field line array rule is 1/2 of a wavelength c2c for cone/domes and from the edge of a ribbon element...

Please fill me in.

The rule on c2c spacing is due to off axis concerns at high frequencies. A line array of drivers is not the same as a true line source because it is not continuous. When the wavelength is much longer than the spacing, the line acts as if it were continuous.

I think Jim Griffin has a white paper out that may explain it better.

[OlogyAudio]

Originally posted by Javachip
I am no expert, but I'll take a shot at this. For higher frequencies, most of the sound is produced near the voice coil, because that is the only part of the diaphragm that is actually vibrating and setting air in motion.

Imagine an 18 inch paper woofer trying to reproduce a 10 kHz tone. Obviously it can't. The big paper cone has so much inertia, and is so flexible, that it can't vibrate that fast. Only the innermost part of the cone, closest to the voice coil, will be able to vibrate that fast (with a lot of distortion).

Now imagine an 18 inch solid diamond (absolutely rigid) woofer trying to play 10 kHz. There is no breakup of the diaphragm, but its mass and inertia prevent any part of it from moving at all. The voice coil, which is rigidly attached to the cone, simply remains motionless and heats up in the magnetic field.

Direct propagation of sound waves through the diaphragm material is very small compared to sound produced by in-and-out motion of the diaphragm/voice coil unit.

I'll shut up now and let you teach me something -- you designed and built your own driver, which I can only daydream about at this point.

Nice post if we keep this up we can get to the bottom of this…
The speed of sound oversimplification was stemming from the idea that if force is applied at one point in the cone, the actual physical movement of the cone will be done through interaction of different sheer and compression waves that travel through the cone (cone geometry dependant). Those speeds of sound listed above are for compression only in solid bars, while sheer waves behave slightly differently and in thin sheets there are all kinds of other things that creep up and I only know how to scratch the surface on that one here...

I think it helps to try and visualize what is going on in the cone (I *really* wish I had this program... but it is about 4400 euros...)
This should help though:
http://www.loudsoft.com/my_documents...0Examples2.pdf
http://www.loudsoft.com/my_documents...per%202003.pdf

That should give you a good idea about how the cone is acting on the wavefront.

Off axis on a larger cone the high frequencies roll off, this is partially explained by one part of the cone being further away from another part of the cone, both parts of the cone are creating this high frequency wavefront at pretty much the same time and the phase interactions cause a drop in spl. This would not happen if the high frequency was only generated near the voice coil, we would see a much better off axis response, such as the manger or other bending wave or distributed mode transducers...

On another topic... I imagine that you could tailor an aerogel to the correct properties that it could more or less match the impedance of the air it is next to and make one heck of a air motion transfer unit...

[OlogyAudio]

Originally posted by Ron E


The rule on c2c spacing is due to off axis concerns at high frequencies. A line array of drivers is not the same as a true line source because it is not continuous. When the wavelength is much longer than the spacing, the line acts as if it were continuous.

I think Jim Griffin has a white paper out that may explain it better.

I agree with that fully, the question on my mind is why is the spacing considered center to center, instead of edge to edge (or well... somewhere slightly in of the edge of the cone since it is a circle after all...)

I wonder if we could make an acceptable quality square-ish midrange… HAH...

[bjorno]

The rule on c2c spacing is due to off axis concerns at high frequencies.

True, DIY’rs crossing over at c-c = lambda/2 measured using a simple folded ruler don’t suffer from insomnia and don’t have to be hypnotized in prior for a good night sleep.

Others nitpicking near c-c= lambda to close to the cone break up area using micrometers, OTOH, are constantly worried having nightmares that don’t prolong a long life expectance.

b

1(1)