IM and Doppler distortion in multi-way vs one-way systems.

[Nevod]

Reading about ESL systems, i've noticed a point on IM/Doppler distortion - that when one driver is used for all the audio spectrum, these distortions are quite high, and multi-way systems are used to reduce such kind of distortion.

However, I wondered, does the separation really help? May it be so that sound radiation from one driver affects another driver in such a way that it produces IM and/or Doppler distortion? And how much is distortion really reduced that way?

[breez]

Originally posted by Nevod
However, I wondered, does the separation really help? May it be so that sound radiation from one driver affects another driver in such a way that it produces IM and/or Doppler distortion? And how much is distortion really reduced that way?

Yes, it does help. Doppler distortion results from the membrane being displaced by significant amounts while at the same time emitting high frequencies.

Sound radiation from nearby sources couldn't possibly cause it unless there were absurd amounts of pressure change to affect the high frequency emitter.

[Nevod]

breez
I understand that, but, I am talking about air affecting air, not the emitter itself. The emitter isn't of course likely to be affected by air pressure change(but in case of ESL, it is more probable), but the air it emits to, is affected, causing probable changes in level of air loading.

[SY]

Remember, though, the displacement of large panel transducers is still quite small. As a practical matter, ESLs have rather low IM.

[breez]

Originally posted by Nevod
breez
I understand that, but, I am talking about air affecting air, not the emitter itself. The emitter isn't of course likely to be affected by air pressure change(but in case of ESL, it is more probable), but the air it emits to, is affected, causing probable changes in level of air loading.

Linear superposition should hold true at practical SPL levels and I think the only effect could be compression.

Originally posted by SY
Remember, though, the displacement of large panel transducers is still quite small. As a practical matter, ESLs have rather low IM.

Yes, I wouldn't worry about doppler effects in such large membranes.

For example, wavelength at 10KHz is 34.4 mm. For any appreciable effects to occur we need at least several millimeters of displacement. For a large panel this means SPL in the range 110dB+ @ 20Hz.