Damping factor, x-over and speaker distortion beahaviour?

[Electron]

Have you been thinking speaker element behaviour with higher driving impedanse ?

It is easier for amplifier to drive higher load impedance. Especially crossover distorition smaller with higher load impedanse when amp works in A-class with higher amplitude.
Cross over distortion is easier to hear on higher frequencies because amplifier cannot compensate that effeciently on higher frequencies due to lower feedback and faster crossover region transition time compared to amplifier speed, not mentioning the distortion increase due to higher load to VAS stage.

Often Tweeter and midrange speakers need to be attenuated in x-over. If this done only with series resistance (or impedance) the amplifier sees higher impedance. This however makes speaker cone/dome control weaker, doesn't it?

Speaker element is mechanical unit with hysteresis and memory plus of course lot of differend kind of nonlinear distortion components.

Now my question is, which of the nonlinear components/phenomenons gets worse with higher source impedance and which possibly better?
Is the benefit in amplifier bahaviour bigger benefit than possibly incrreased some distortion phenomenons in speaker unit?

 

[Speedskater]

Before doing too much thinking on the subject, it might be good to re-read the old (1994-2002) Dick Pierce paper on the subject:

Damping Factor: Effects On System Response

Dick Pierce - Professional Audio Development

http://www.cartchunk.org/audiotopics/DampingFactor.pdf

 

[Electron]

Speedskater, the point is not in damping factor from T/S parameter point of view but more like distortion. The paper you linked (by quick look) talks only about T/S effect and amplitude changes not about phenomenom in speaker unit. Note that spekaer unit sees easily tens of ohms driving impedance due to x-over. That compared to voice coil resistance can be 10... times.
Now when moving voice coil directly from amp the system impedance level is equal to voice coil impedance but with x-over it can be tens time higher. Therefor it is possible that nonlinerities in spider or surround has tens of times higher effect to movement linearity of voice coil.

I see here similarity to non liner input impedance of transistor and too high source impedance causing distortion even though in lab measurement (when used couple of ohms source impedance) the stage is perfectly linear.