Graph Feppar
Active Member
- Joined
- Nov 19, 2018
- Messages
- 269
- Likes
- 258
This is another common confusion I see in headphone/speaker forums, if you know the harmonic distortion levels of headphone or speaker, you automatically know the IMD levels too. Both harmonic and inharmonic distortion comes from same reason, the driver as a piston is not in precise position its supposed to be.But we don't know the IMD of the 490 pro, do we ?
Among other articles, Harman themselves tried to evaluate the impact of various forms of non-linear distortion on listener preferences, the main article being this one (includes a Stax) : https://www.aes.org/e-lib/browse.cfm?elib=17441
In that test IM did not seem to have a particularly significant impact on preferences. Generally speaking for most of the headphones that were recorded and then played back non-linear distortions were low enough to have little impact on preferences.
I'm all in for more perceptually relevant non-linear distortion metrics, but I am not certain that focusing on IMD is the best bet to make.
As the driver goes further, the restoring force pulling it back is higher so the error in position is bigger. Now if you have headphone with 1% THD at 100hz and lets say 0.01% THD at 7200hz, becose the driver excursion is much higher at the 100hz, the poor 7200hz that would otherwise be clean will be riding on top of the 100hz driver motion and pushed into the highly non-linear region, thus making it highly distorted too. And since this is time variable, depending on at what phase of the 100hz cycle the driver is at, it will create time variable distortion of the 7400hz sine, thus creating inter modulation tones
Its like waterfall plot and freqency/phase response graphs. If you know the frequency and phase response of a headphone, you can predict exactly how will fast will it decay across the spectrum in waterfall plot.