Is this a decent measurement?

Discussion in 'Headphone Measurements' started by Murderisfun, Mar 24, 2020.

  1. Murderisfun

    Murderisfun New

    Banned
    Joined:
    Mar 24, 2020
    Likes Received:
    0
    Trophy Points:
    1
    Location:
    earth
    I'm trying to measure a DIY headphone driver I made and I'm a total noob at this so I'm looking for guidance and outside interpretation.
    Not sure if my settings are correct but here it is.
    [​IMG]
    [​IMG]
    [​IMG]
    [​IMG]
    linear waterfall plot
    [​IMG]

    For some reason the log waterfall plot didn't display under 200hz unless I expanded the window to 60ms
    [​IMG]

    Do my settings look alright?
    Also what's the deal with all the valleys in the waterfall response? Is that really the driver decay?
    And is there a method for smoothing out all those wrinkles in the FR?
     
    Last edited: Mar 24, 2020
  2. Lyander

    Lyander Official SBAF Equitable Empathizer

    Pyrate Contributor
    Joined:
    Feb 25, 2017
    Likes Received:
    10,961
    Trophy Points:
    113
    Location:
    Philippines, The
    Important bit: this is the section for specific known quantities to post stuff. The one for noobs (like me, hi) is: https://superbestaudiofriends.org/index.php?forums/audio-science.24/

    Other things: what microphone are you using and how are you coupling the driver to the measurement rig? Also, if it's just a driver being measured how are you getting extension that far down low? Not just a driver but also a baffle with pads etc? How noisy was it in your room when you took these measurements? Any compensation applied? Also, without a known reference to directly compare to it's really kinda hard to say if this is any decent or not since the specifications and performance of your measurement system are basically unknown.

    Just guessing it's a planar/ortho from those measurements, but I could always be wrong. Rough treble behaviour would corroborate that, as well as what could be "ortho walls" in your CSD (https://superbestaudiofriends.org/index.php?threads/ortho-wall-for-lack-of-a-better-term.2119/)

    I don't bother with IR on my rig because the EARS rig is pretty poorly suited to them, but based on other measurements like @cskippy's here displaying 500m is a bit too far off; there's basically a 2m window here: https://superbestaudiofriends.org/i...io-borealis-australis.7650/page-3#post-258552

    Going up to 9th harmonic in distortion may or may not be of very much use in headphone measurements, since at least from what I can tell behaviour after 4th isn't greatly influential to perceived performance. It also makes the graph look a lot cleaner, which helps with digesting info. I also recommend hiding stuff under the measured noise floor because external sounds can EASILY bollix those figures.

    60ms window is a tad overmuch, I have a 7ms window with only 4s visible, 0.1s rise time and 241 slices (holy poop 1000 slices would kill my laptop). Range is 115dB to 60dB, but only because I use a top-down perspective; actual range from peak to floor is only about 35dB; a lot of this will depend on how quiet you can get your room during measurements, also how much dynamic range you believe you need before environmental noise becomes a problem ;)

    A lot of the above will depend on what gear you're using, mind you. I "just" use a MinDSP EARS so have to account for the weirdness that comes with the silicone pinnae and metal plates. I also don't bother with CSDs for stuff down below about 300Hz because of the aforementioned window problems— personally use spectrograms to check how "taut" bass is. Not sure how much use stuff before peak is on spectrogram BTW, I just set that to 1ms.

    Uploaded single channel measurements for a headphone just to show how I present data, may or may not work for you: https://imgur.com/a/w8wuuIK

    P.S.
    Dafuq is with your handle, hah.
     
  3. purr1n

    purr1n Desire for betterer is endless.

    Staff Member Pyrate BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    89,771
    Trophy Points:
    113
    Location:
    Padre Island CC TX
    This is a well disguised troll post. Advise to ignore.
     
  4. Murderisfun

    Murderisfun New

    Banned
    Joined:
    Mar 24, 2020
    Likes Received:
    0
    Trophy Points:
    1
    Location:
    earth
    The MiniDSP UMIK-1

    I'm just holding where I expect my ear to be located next to the driver.



    It's inherent to the driver technology. I've been developing it for some time based on a theory I had a couple years ago.
    Audble down to 1hz but it barely registers as sound until 5hz. 20hz and up can get pretty loud.
    The sub bass on this thing is insane, it makes the air feel solid.
    Beyond that, you wouldn't believe me if I told you how good it sounds so I'm going to keep my mouth shut.
    I'm working on making it marketable after I finish development.

    Quite noisy actually. Between the fan of my lab supply and a couple of heaters. But when I run the REW measurement and mute the sound the noise floor only comes up with some spikes in the lower frequencies, not all that crap up to 20khz like the graph is showing.


    I don't know what compensation you are referring to.

    What do you need to know?
    Close, it's a type of planar but in an unusual configuration.

    Is all that squirliness audible? Is it something I should try to fix? Do you know what the cause is?

    But the 60ms is required as I stated, otherwise I get this
    [​IMG]
    For some reason I don't know, the frequencies below 200hz are cut off unless I go to 60ms window for the full range which necessitates a 60ms rise time.
    Thanks I'll take a look
    I have a dark sense of humor :p
    You can't be serious...
     
  5. purr1n

    purr1n Desire for betterer is endless.

    Staff Member Pyrate BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    89,771
    Trophy Points:
    113
    Location:
    Padre Island CC TX
    I'm dead serious. Your handle does not meet acceptable social norms for SBAF. It's also a strong indicator to me that this DIY headphone effort is too big for your britches. Please try again.
     
    Last edited: Mar 24, 2020
  6. purr1n

    purr1n Desire for betterer is endless.

    Staff Member Pyrate BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    89,771
    Trophy Points:
    113
    Location:
    Padre Island CC TX
    The driver looks good. Proceed to make a headphone. Then come back. Focus on FR first. Don't worry about the other stuff until you get FR right.
     

Share This Page