Andromeda measurements + discussion

Discussion in 'Measurement Techniques Discussion' started by Serious, May 19, 2017.

  1. Serious

    Serious Inquisitive Frequency Response Plot

    Pyrate BWC MZR
    Joined:
    Sep 28, 2015
    Likes Received:
    2,594
    Trophy Points:
    113
    Location:
    near Munich, Germany
    @Muse Wanderer came over for a listen today after High End and brought his Utopia and Andromeda with him, for me to give a 2nd listen and also take measurements. This is also the thread where I wanted to continue the discussion that @k.e. started in the Andromeda thread.

    First up, here are the standardized measurements of UERM and Andromeda on an 8mm inner diameter tube that was calibrated to match the IEC type couplers. Referenced to about 2.83Vrms. Keep in mind that insertion depth and ear canal size play an important role in the resultant FR and sensitivity.
    Andro UERM standardized measurements.jpg
    The channel imbalance with the UERM is the effect of a slightly different insertion depth and I mainly did that to show the effect of the insertion depth. Even with customs there's a bit of room for different insertion depths. It's harder to get a good seal with customs, so bass extension might even be better than this. Subjectively I prefer the UERM bass.

    The Andromeda seems to be about 11db more sensitive at 100Hz, but considering a 4 Ohm impedance for the Andromeda and a 50 Ohm impedance for the UERM at that frequency, they are both almost exactly the same efficiency.

    The answer is it depends. With my own measurements I look at the raw, uncompensated measurements and I use a compensation curve to get the measurements to look like the IEC measurements. With the IEC measurements I basically use the compensation curve as a target.
    There is a difference between ear-canal opening measurements and measurements with the microphone at the eardrum level and it's not just the ear canal. So no, the HD600 shouldn't show a 10db bump at 3.5kHz, at least not with that type of measurement.
    True, but overall I like to think that IEMs don't have random parts to decrease efficiency in them for no reason. If the parts serve to better achieve a target FR then that's fair, but should definitely be factored into the efficiency calculation. The correlation with sound quality doesn't have to be perfect.


    The Andromeda sounded far too bassy from a near 0 OI source for my tastes. It was almost perfect from the Ragnarok SE out, but you have to use low gain to avoid hiss.
    Here's the Andromeda FR from the Ragnarok SE output (3.3 Ohm):
    UERM 0 OI vs Andro Rag SE.jpg
    Three key differences:
    1. Andromeda has less energy in the fundamentals region around 500Hz. This adds more thickness to the lower mids.
    2. Andromeda has more energy in the 1.5kHz range. This might not be a bad thing.
    3. Andromeda has less energy in the lower treble around 6kHz. This can be good or bad depending on your preferences.

    I also attached a comparison between the 3.3 Ohm and 0 Ohm measurement.

    Subjectively the UERM bass still sounds clearer (but not as clean as HD800 or my speakers) and the UERM is also less closed-in.
     

    Attached Files:

    Last edited: May 19, 2017
  2. bartzky

    bartzky Acquaintance

    Joined:
    Dec 22, 2016
    Likes Received:
    62
    Trophy Points:
    18
    Location:
    Germany
    Home Page:
    I have to say your use of the word compensation is kind of confusing. Compensation for me is DF-compensation and so on, while making measurements look like 60318 is something like calibration to me.

    Of course! I should have read your post more carefully...
    So you tried to EQ your HD600 towards one of the speaker measurements but didn't like it that way?

    Of course they aren't just radom parts, but serve a function in the crossover network.
    There might as well be parts to make the IEM an easier load or to make it less prone to hiss.


    To be more precise: When I said "As the frequency rises the sound will become more direct so the free field will become more important." I was talking about around a growing amount to around 20% FF @ 10kHz.

    In fact there isn't such a thing like the DF, but multiple DFs from different sources and for different purposes. As we are interested in IEC 60318-4 measurements we will have to use the DF from ISO 11904-2 as this was measured with a ITU-T P.58-spec manikin who's ears are essentially 60318 couplers.
    As you already stated this DF was never intended to be an IEM target and will sound like shit. But however I think it will serve as a good basis for developing a real target for IEMs and headphones.
     
  3. 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 forgot to mention: The two Andromeda measurements aren't perfectly comparable. The 0 OI one was measured with Comply tips and the other one with silicone tips. The insertion depth wasn't exactly the same and I also had to use a slightly wider tube. This mainly affects the resonances in the treble. I feel both are close to representative to how I hear them.

    Yes. Basically a +10db boost centered at 4kHz. Not surprising that it doesn't sound good. The dummy head measurements are more accurate in this regard-

    You're right. You could say it's calibrated to the IEC 60318 standard. I tried to match my measurements to the raw Rin Choi measurements, but I think it's not accurate above 10kHz. The area above 10kHz is hard to measure with IEMs anyway.

    Here's what the raw, uncompensated measurements for the UERM look like:
    UERM raw measurement.jpg

    Now the measurements look quite a bit too warm. Here's what happens when I apply the B&K target. LR average for UERM and Andro (3.3 Ohm):
    UERM Andro BK target.jpg
    Not bad I'd say. The AudioQuest "Ideal room curve" gets even closer in the upper midrange in my opinion, but they're close enough really.
     

    Attached Files:

    Last edited: May 23, 2017
  4. stratocaster

    stratocaster Friend

    Pyrate
    Joined:
    Sep 29, 2015
    Likes Received:
    495
    Trophy Points:
    63
    Location:
    Austria
    FWIW, these are my UERM measurements using the exact same UERM @james444 had sent for Rin to measure. I am using a 25mm tube with a diameter of 8mm as a coupler and a calibrated Beyerdynamic MM1 mic and the Jot for measurements. These are RAW. The 7kHz peak goes down quite a bit if you reduce the lenght of the tube by about 1-2mm or insert the IEM more deeply.

    uerm.jpg

    Blown-up Y-axis
    uerm2.jpg
     
    Last edited: May 23, 2017
  5. bartzky

    bartzky Acquaintance

    Joined:
    Dec 22, 2016
    Likes Received:
    62
    Trophy Points:
    18
    Location:
    Germany
    Home Page:
    Additionally the area above 10kHz really isn't very interesting. Knowing a rough tendency how the FR may look like is sufficient IMO.

    May you plot and post the AudioQuest curve? It sadly is not shown on the website anymore.

    Thanks for sharing the link to the AudioQuest article. A few aspects are pure bullshit though...

    That's simply not true and a misquotation. The ISO does show how the A-weighted values can be calculated but the values given in ISO 11904-2 are not A-weighted as well as the plotted curves on the AudioQuest website are not A-weighted.

    DF and FF are referenced to the eardrum so they of course must contain the pinnae and canal effects. The same is true for a HATS measurement. As AudioQuest refers to HATS measurements they are actually comparing a with-pinnae measurement to a with-pinnae target - so everything is correct and there is no emphasis of "the pinnae and ear canal transfer functions" as stated by AudioQuest. It would of course be wrong to compare a with-pinnae target like DF or FF to a non-pinnae measurement, but this is obviously not the case here.
     
  6. Crinacle

    Crinacle Friend

    Staff Member Pyrate IEMW MZR
    Joined:
    Dec 12, 2016
    Likes Received:
    812
    Trophy Points:
    93
    Location:
    SG
    Home Page:
    Here's my measurements for an additional perspective. Compensated to IEC711 standards.

    UERM
    [​IMG]

    Andromeda (deep, medium, shallow inserts respectively)
    [​IMG]
    [​IMG]
    [​IMG]

    Officials for reference
    [​IMG]
     
  7. bartzky

    bartzky Acquaintance

    Joined:
    Dec 22, 2016
    Likes Received:
    62
    Trophy Points:
    18
    Location:
    Germany
    Home Page:

Share This Page