How to pick a headphone as an EQ base.

Discussion in 'Modifications and Tweaks' started by JewBear, Nov 24, 2017.

  1. TomHP

    TomHP Facebook Friend

    Joined:
    Nov 24, 2016
    Likes Received:
    153
    Trophy Points:
    43
    Location:
    Taiwan
    Case in point: quick simulation on a headphone driver with Le changing from 0 to 0.3 to 0.6 mH.

    Impedance
    [​IMG]

    Phase
    [​IMG]

    Sorry, I haven't figured out how to unwrap the phase yet in this new software tool, but you can see the shift induced by the inductance.
     
  2. Serious

    Serious Inquisitive Frequency Response Plot

    Pyrate BWC MZR
    Joined:
    Sep 28, 2015
    Likes Received:
    2,594
    Trophy Points:
    113
    Location:
    near Munich, Germany
    Wait, phase shift due to inductance is non-minimum? Shouldn't that mean my woofer measurements should have a ton of excess phase? I'll have to look at them again
    Edit : I see you meant regular phase shift, not excess phase. In which case I don't see the problem.
     
  3. TomHP

    TomHP Facebook Friend

    Joined:
    Nov 24, 2016
    Likes Received:
    153
    Trophy Points:
    43
    Location:
    Taiwan
    But it is excess phase, no? It's only for parts of the frequency range where the inductance becomes important (function of jw).

    For your woofer, it probably won't matter much since the x-over point will be below where the inductance starts to act up.
     
  4. Serious

    Serious Inquisitive Frequency Response Plot

    Pyrate BWC MZR
    Joined:
    Sep 28, 2015
    Likes Received:
    2,594
    Trophy Points:
    113
    Location:
    near Munich, Germany
    I definitely don't see anything like what your graphs show or what you seem to suggest. Measuring just the Beta 15A woofer (Le 1.1mH, but that's even with the crossover attached*) I get an excess phase graph that's absolutely flat to 2kHz (less than 1° of phase shift) and then it depends more on the measurement than measuring an intrinsic woofer parameter it seems. The woofer just rolls off so much >2kHz that the SNR is too low to really analyze >5kHz. Since my widebander has a whizzer cone there are broad regions of excess phase shift at the crossover regions between big cone, whizzer cone and voice coil. Every other driver I've measured was virtually perfectly minimum phase. And headphones are essentially just a driver in an enclosure, so not really any different there.

    *I also don't see how crossover components should make the driver response deviate from minimum phase. Of course they affect frequency response magnitude and phase, but it's still minimum phase. Only when you add the outputs of two drivers you usually start to get excess phase issues, time alignement goes out of the window, etc.

    EDIT: Attached the graph**. I don't think I ever posted my impedance measurements of just the woofer and I don't think I ever measured the woofer + crossover impedance, but even without the crossover the inductance is very high. Without the crossover it's about 7 Ohm at 100Hz and 24 Ohm at 3kHz for one woofer.

    We pretty much already had the same discussion here: http://www.superbestaudiofriends.org/index.php?threads/linear-phase-discussion.4594/
    I still stand by my opinion that headphones and speaker drivers essentially measure like minimum-phase systems. Maybe when there are very severe enclosure resonance issues we might see some excess-phase, but I don't think I've ever seen that.

    **The excess phase in the bass region I think is from ADC/DAC filters or possibly sampling rate issues. Always seems to be there with the UMIK-1. When I use the same device for ADC and DAC I get none of this.
     

    Attached Files:

    Last edited: Nov 28, 2017

Share This Page