Why not underhung voice coil?


Most of the speakers have overhung voice coil, meaning that magnetic gap is very narrow and most of the coil sticks out.  In underhung design magnetic gap is (horizontally) wide and whole coil movement is contained within the gap.  It requires much larger magnet, but supposed to be more linear (lower distortions), especially for big excursions.  It applies mostly to woofers, but there are even tweeters with underhung coil.  Very few speaker companies use underhung design.  One of them is Acoustic Zen.  As I understand it the only disadvantage is increased cost because of much larger magnet, but it should be irrelevant, at least for high end/cost speakers.  Why overhung coil design dominates.  Please help me understand.
128x128kijanki
Okay so slow work day here thanks to the CCP virus combined with misinformation keeping everyone afraid to go out (when they could just cover their face and be fine) so I had time to look this one up. Which I had to do seeing as its such an arcane and normally insignificant bit of audio trivia it had long ago been forgotten. 

It seems you know the answer but dismissed it: cost. Since you know about Acoustic Zen then you probably know this too, it was easy enough to find:
To compliment the speed and low distortion characteristics of his tweeter, Mr. Lee decided to develop his own woofers. The custom designed woofer uses an underhung voice coil that is lower in distortion compared to the overhung voice coil found in the majority of loudspeaker designs.

An overhung driver has a long voice coil in a short magnetic gap and the under hung type uses a short voice coil in a long magnetic gap. Short coils dramatically reduce harmonic and transient distortion because they have complete control within the magnetic field.

In an underhung setup the voice coil operates inside the boundaries of the edges of the magnet. On the other hand, the voice coil in an overhung driver will not only vibrate within the gap between the magnets but also extend beyond the end of the magnets. When the voice coil functions inside and outside the strength of the magnetic field it produces more distortion. Although underhung drivers are more expensive to produce, because of larger magnets, they are lower in distortion then their aforementioned counterparts.
So there you have it. Its possible to get really good results both ways. Underhung is lower distortion but as always that is only when looking at it with blinders on, as if nothing else matters. When in fact everything else matters. Distortion also comes from the cone material, dust cap material, surround, basket, baffle material and shape, cabinet material and shape, on and on and on. Every single little bit of it either contributes or detracts from performance and distortion.

In other words your question boils down to help me understand ad copy. That's easy! Ad copy is written to tell a story that will make guys go out and buy stuff. Is it working?
"...supposed to be more linear (lower distortions), especially for big excursions."

I do not agree that underhung motors are more linear for large excursions as a blanket statement; imo the situation is more complicated than that. Yes they are somewhat more linear as x-max is approached, but typically underhung motors have less linear excursion (x-max) than overhung motors, so for a given SPL it depends on the specifics. And when x-max is exceeded with an underhung motor, it doesn’t take much overshoot for a large portion of that short coil to exit the gap, which causes distortion to rise rapidly.

An underhung motor can achieve higher efficiency because the moving mass is usually reduced by the voice coil being shorter and therefore lighter.

On the other hand that short voice coil typically has less thermal mass than the much longer voice coil of an overhung motor, so in general underhung motors are more susceptible to thermal compression. (The short voice coil being typically surrounded by thermally conductive metal tends to lessen but not reverse this disparity.)

The inductance of that short underhung voice coil will typically be less than the inductance of a longer overhung voice coil, unless the overhung motor includes Faraday rings.

I have used woofers with both types of motors and if we’re comparing high quality, roughly equal-cost woofers, it has not been obvious to me that underhung motors inherently outperform overhung ones. Perhaps one reason why the theoretically improved linearity of underhung motors wasn’t obvious lies in human hearing: The low-order distortion that woofer motors produce is not highly audible to begin with.

If there were more underhung motor woofers out there today I might be using one, assuming it did what I wanted better than its overhung competition.

Duke
@audiokinesis  Thank you.  I can understand it better now.  I will try to find out more on Faraday rings.  AFAIK they are used to lower influence of magnetic field produced by the coil, but I have no idea what they look like.  I cannot find any picture showing their location.
They’re typically copper caps or rings(sometimes aluminum), situated on(or around) the magnet’s pole piece/gap.      Variously: called flux stabilization, demodulation or shorting rings.     ie: http://www.audioheritage.org/images/projectmay/technology/1500AL_draw.gif     and: https://celestion.com/speakerworld/patech/4/117/Demodulation_Rings/     This has a page, with a nice diagram: https://www.focal.com/sites/www.focal.fr/files/shared/catalog/document/Sopra_WhitePaper.pdf
@rodman99999  Thank you.  As shorting ring they must work as short turn, reducing leakage flux.