Who makes the lowest distortion speaker drivers

Addendum

Most audible distortions (departures from a realistic performance) are beyond your control. They relate to the acoustics of the recording venue, miking techniques employed, the technology used to record it, and mix-down details used to produce the master recording, a copy of which is the source you will be listening to. Beyond this, the acoustics of your listening space and the number and placement of loudspeaker units within it is next. This is followed by signal processing facilities available in your system. Next is the directivity of the loudspeaker drivers used and the crossover points selected, the drive signal filtering, and enclosure design details. Last and least important, are the loudspeaker distortion products that measure, say less than 10%. At the low levels used in a domestic listing space, most are inaudible or masked by the characteristics of the original music signal that has already been recorded. Dr. Bose figured this one out a long time ago.

Regards,
WHG

So did Dr. Geddes as well!
 
IMHO instead of loosing your minds in the theories (even if you are really good at theroisation) you perhaps should perform your own recordings in order to evaluate these theories especially if these ones are out of the standard test procedures.
Just an exemple, my unmastered ORTF raw recordings never lie and never fail when i have to evaluate the sound imaging.
 
Linkwitz did some driver testing that may interest you:

Midrange distortion test


Having read lots of what Linkwitz’s work, I took an opportunity to listen to his systems at the seaside ranch, during my last trip to the Bay Area. The larger system was much closer in fidelity performance to what I listen to on a small system. He has attributed this to the cheaper electronics on the smaller system. The dynamics low frequency dynamics is better than what I usually listen to for both systems, but the focus and image depth is not as good.

I will be at the Loudspeaker Sourcing Show in Guangzhou next week to see if I can find quality source component suppliers. First time there, don’t know what to expect. But definitely going to take a day or two for some touring.
 
a moving cone does have a 'geometric nonlinearity' - its not a Linear (superposition principle) air velocity source in a unmoving plane


This is why I get confused because I understood that the air velocity source is relative to the magnet, which is unmoving I hope.

I assume that the reference to a piccolo note riding on the motion of a single driver cone refers to cone surface oscillations, break up if you will, that are not referenced to the magnet. They are parasitic in nature. If the piccolo is a note playing without cone break up then these motions of the cone are referenced to the magnet - how can that cause a Doppler effect? To get Doppler don't you need the 'flabby' cone (I like that expression Silversprout!) that no longer moves with reference to the magnet ? i.e. parasitic motion. I've never heard anything resembling FM, as Earl says it's probably inaudible.
 
Last edited:
no, a "perfect" piston driver still has Doppler from the displacement necessary to generate air velocity

Rod went round about, reasoning about the difference in air path delay when the high frequency is radiated from the top vs bottom of the excursion of the (much larger displacement amplitude) low frequency

it turns out the displacement description gave the same equations as the Doppler explaination


breakup can make the measurement of pure Doppler difficult, but is is there AND likely inaudible
 
no, a "perfect" piston driver still has Doppler from the displacement necessary to generate air velocity

I understand why folk eroneously think this way. They see the cone as the source of the sound and since the cone moves relative to the observer, there should be a Doppler shift. But there is a subtle issue here that is missed.

The Doppler shift occurs when there is movement of the source relative to the observer because the peaks of the sound waves are no longer separated in time by the same amount as when the source is at rest. When the source moves closer to the observer the pressure waves bunch up, when the source recedes the pressure waves stretch.

In the case of a pistonic speaker cone the pressure waves are generated when the cone moves back and forth relative to the fixed magnet in lock-step with the applied electrical signal. Multiple frequencies can be present in the signal, each one creates movement of the cone relative to the magnet. These frequencies do not mix in this linear system. It doesn't matter if the cone is slowly oscillating to a low frequency electrical signal at the same time as it wiggles to a higher frequency electrical signal. The pressure waves generated at the higher frequency will be regularly spaced in time and not modulated by the lower frequency.

The reference point for the speaker as a sound source is the permanent magnet. If the whole assembly were to move relative to the observer, then you would have a Doppler effect.

Put the speaker inside a black box, nobody knows how it generates it's sound except that it is a linear device. If the black box is stationary, there is no Doppler effect. If the whole box moves, then the frequencies it emits will all be subject to the Doppler effect, depending on the relative motion to the observer.


Rod went round about, reasoning about the difference in air path delay when the high frequency is radiated from the top vs bottom of the excursion of the (much larger displacement amplitude) low frequency

I'm afraid Rod didn't see clearly that the sound pressure waves are generated from the motion of a cone relative to the fixed magnet, regardless of the excursion of the cone caused by other frequency components providing that it operates in a linear range (Xmax anyone...). When the cone has a large positive excursion it wiggles at high frequency in lock-step with the applied high frequency signal just as well as at large negative excursions - the frequency of these wiggles dictated by the applied signal acting against a stationary fixed magnet. There is no modulation or change in the frequency of these high frequency signals. No Doppler effect.
 
Last edited:
Part of why small full range drivers never really worked well for me has been because, at most systems I have heard/had personally, sound like a garbled mess when you up the volume a bit. Perfectly fine at lower levels, but for instance the TB W5-1880 sounded horrible when the cone started flapping about a little bit (well within xmax, maybe +/-2,5-3mm at the very largest swings).

I can not speak for anyone else regarding subjective opinion, but I much prefer a cone that does not move as much when playing a little bit loud, so that means you need a bit more SD to get the same SPL, then you get more issues with breakup up top, but that does not bother me to the same extent.

Silverprout:
I agree that a half decent 15" would be preferable for the same price, pending on the application, but quality of the drivers was the point of this thread, not price?

Would you rather have a wall filled with cheap 15"s or a wall filled with the SD equivalent in Seas driver build quality?

Edit:
I guess the B&C 15PS100 is more comparable to the Seas 10", but they are not equal, FS is much higher, it does not have the same performance goals. You can certainly not compare the Seas 10" to something like 15LB075-UW4.
 
Last edited:
Silverprout:
I agree that a half decent 15" would be preferable for the same price, pending on the application, but quality of the drivers was the point of this thread, not price?

Edit:
I guess the B&C 15PS100 is more comparable to the Seas 10", but they are not equal, FS is much higher, it does not have the same performance goals. You can certainly not compare the Seas 10" to something like 15LB075-UW4.

As i’ve said in my post above with the linear motion mathematicaly theorized as a curvilign function, things are not as simple.
But under 100Hz Sd is god.
The problem is that large loudpeakers needs space to breathe, because the bigger the soundwaves are the more they trigger the room modes.
Even an entry level well designed 15’ PA woofer in the encolsure recomanded by the manufacturer can soud extremely detailed and refined comparatively to a very expensive hifi woofer.

Entry level low QTS well designed 15’ PA under 100€ :

Fane Sovereign 15-400
Eminence Gamma 15A2
P audio 15BM-500B/E15-300S/15BM-300B
Beyma SM-115/N
Oberton...
Sica...
... there are a lot of other manufacturers.
 
Last edited:
When using a full range driver, the BL curve and Km curves are important. Additionally, Xmax is important. For a small 3” full range driver that I use, I have to filter the frequencies below 40Hz to prevent that part from driving it beyond Xmax. This filtering was tuned through listening to lots of various music.
 
But they do not compare at all to the Seas drivers build quality.

My seas E0022-08S W22EX001 are not perfect, the build quality is not so high, but i don't care since they are hidden by cloth grills.

When using a full range driver, the BL curve and Km curves are important. Additionally, Xmax is important.

Absolute values are meaningless, the explanation of the optimisation of each parameters and the technical solutions that link them to their environment are more interesting IMO.
 
If you are talking about the Magnesium drivers I will agree that they are a bit overrated, I think they are impressive because Magnesium is a difficult material to "get right" for loudspeaker cones, but it requires too much work/adaption of the crossover to tame the sonic character of these drivers, and in many cases you end up with "dull" sound. It can be done, it's just very difficult.

What I do think has a real influence on the quality and level of harmonic distortion on the Seas drivers are the other technical implementations:
Heavy copper rings mounted above and below the T-shaped pole piece reduce non linear and modulation distortion and increase overload margin.

Extremely stiff and stable injection moulded metal basket keeps the critical components in perfect alignment.

Large windows in the basket both above and below the spider reduce sound reflexion, air flow noise and cavity resonance to a minimum.

A solid copper phase plug enhances the performance of the copper rings and improves heat conduction away from the pole piece.

There is very little that disturbs the movement of air around the moving pieces, steps have been made to reduce cavity resonance and influence of variations in the magnetic field. The basket is very stiff and with good internal damping, but at the same time very thin, both to reduce reflections and also "break up" the little reflection they have.
 
If you are talking about the Magnesium drivers I will agree that they are a bit overrated

Well implemented they one of the most transparent drivers available (less colorated by cone resonances).

There is very little that disturbs the movement of air around the moving pieces, steps have been made to reduce cavity resonance and influence of variations in the magnetic field. The basket is very stiff and with good internal damping, but at the same time very thin, both to reduce reflections and also "break up" the little reflection they have.

Since it is impossible to grip the air by acuating directly the molecules we must use an achaic mecanical mecanism for the energy conversion.
Perhaps in the future a lab guy will fin a way to create a cloud of nanoparticles of magnetic material directly into the air in order to perform a perfect energy transfer... :santa: