1.     Don't be a dotard!
    Read these "rules" AND introduce
    yourself     before your first post
  2. A healthy approach to understanding measurements
    Being true to what the artists intended
  3. After being asked for the 308th time, Merv lists top 5 headphones
    (opinion / entertainment piece)
    Dismiss Notice
  4. $1000 to the first to design a headamp using discrete components with better measurements than Magni 3+
    Dismiss Notice
IMPORTANT: New Members Read This Don't Be a Measurement Fascist Merv's Top 5 Headphones - 2023 Edition $1000 Challenge

Burst Response! HD800, SR-207, HD650

Discussion in 'Measurement Techniques Discussion' started by purr1n, Jan 8, 2017.

Page 2 of 14
  1. zonto

    zonto Friend

    Pyrate Contributor
    Joined:
    Sep 30, 2015
    Likes Received:
    4,979
    Trophy Points:
    113
    Location:
    Boston, MA
    @Marvey were these HD650 results with your HD650M, HD650K, or stock?
     
    zonto, Jan 8, 2017
    #21
  2. Cspirou

    Cspirou They call me Sparky

    Pyrate
    Joined:
    Sep 27, 2015
    Likes Received:
    8,221
    Trophy Points:
    113
    Location:
    Northwest France
    I always appreciate mic measurements! Curious to see what this looks like with a tube amp.
     
    Cspirou, Jan 9, 2017
    #22
  3. Soliloqueen

    Soliloqueen Friend

    Pyrate
    Joined:
    Sep 27, 2015
    Likes Received:
    225
    Trophy Points:
    43
    Isn't this just sort of a slice of a waterfall plot? Isn't a waterfall plot just this but with a frequency axis?
     
    Soliloqueen, Jan 9, 2017
    #23
  4. Soliloqueen

    Soliloqueen Friend

    Pyrate
    Joined:
    Sep 27, 2015
    Likes Received:
    225
    Trophy Points:
    43
    i see. yes but the difference is this is real instead of deconvolved

    what happens if you band pass a frequency out the sweep for a csd?
     
    Soliloqueen, Jan 9, 2017
    #24
  5. purr1n

    purr1n Desire for betterer is endless.

    Staff Member Pyrate BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    89,909
    Trophy Points:
    113
    Location:
    Padre Island CC TX
    Not the same thing. CSDs are slices of frequency domain information over time. Each slice is an average of spectral content over an interval of time. Time domain information is lot.
     
    purr1n, Jan 9, 2017
    #25
  6. purr1n

    purr1n Desire for betterer is endless.

    Staff Member Pyrate BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    89,909
    Trophy Points:
    113
    Location:
    Padre Island CC TX
    Probably low level ambient noise. Environment was pretty noisy when I took the measurements. Planes, trucks, yard blowers, etc.
     
    purr1n, Jan 9, 2017
    #26
  7. purr1n

    purr1n Desire for betterer is endless.

    Staff Member Pyrate BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    89,909
    Trophy Points:
    113
    Location:
    Padre Island CC TX
    HD650K and HD800V
     
    purr1n, Jan 9, 2017
    #27
  8. OJneg

    OJneg The Most Insufferable

    Pyrate BWC
    Joined:
    Sep 25, 2015
    Likes Received:
    3,923
    Trophy Points:
    113
    Location:
    Grand Rapids, MI
    Any idea of the mechanism behind the poorer decay? Thin plastic diaphragm storing energy and still vibrating?
     
    OJneg, Jan 9, 2017
    #28
  9. ultrabike

    ultrabike Measurbator - Admin

    Staff Member Pyrate MZR
    Joined:
    Sep 25, 2015
    Likes Received:
    8,960
    Trophy Points:
    113
    Location:
    Irvine CA
    A CSD plot frequency domain magnitude slice corresponds to a windowed impulse response, where I think the window is of variable width and is reduced by shifting the left hand side of it. I believe a waterfall is different in that a fixed width window is sliding in time across the impulse response for each one of the waterfall slices.

    This is a windowed single tone (probably rectangular) and the window is not moving. It might correspond to the first slice of the CSD if instead of windowing the impulse response, we were windowing a tone, and not looking at the frequency magnitude (or phase) but the raw time domain response itself.

    I think the preferred approach for something like this in some of the stuff I've done in the past is a step response (looking at phase lock loop responses and control), but if one wants to visualize the response to a particular frequency (or set of frequencies), this looks pretty useful.
     
    ultrabike, Jan 9, 2017
    #29
  10. Kattefjaes

    Kattefjaes Mostly Harmless

    Pyrate
    Joined:
    Sep 5, 2016
    Likes Received:
    4,521
    Trophy Points:
    113
    Location:
    London, UK
    Congratulations, @Marvey - you have devised a way to visualise "limp dick" without recourse to the murkier corners of google \/
     
    Kattefjaes, Jan 9, 2017
    #30
  11. purr1n

    purr1n Desire for betterer is endless.

    Staff Member Pyrate BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    89,909
    Trophy Points:
    113
    Location:
    Padre Island CC TX
    I would hesitate to use a step response visualization, at least one ultimately derived from sweeps or noise. But hard to say.

    Would like to get more data from a variety of headphones. LCD2r2 (pre-fazor). LCD2 post fazor, HE560, HE6, HE60, etc. I don't think it's far fetched to think that a plastic diaphragm stretched out across a surface would continue to vibrate like a drum. There is certainly interesting stuff going on with the STAXen with the both the start and stop of the higher frequency signal. The diaphragm seems to shoot an initial load, followed up a period of lower energy, and then with a second less powerful load over a a period of time before settling. SR-007 and SR-207 are a bit different exhibiting these behaviors. The SR-007s tendencies toward this behavior are not as strong as the SR-207's and there is no second load at the signal start / attack. The SR-207 also has a second rest period for the signal start.

    SR-007 settling.png
    SR-207 Settling.png
     
    Last edited: Jan 9, 2017
    purr1n, Jan 9, 2017
    #31
  12. bigfatpaulie

    bigfatpaulie Tried to screw other friends while playing victim

    Pyrate Banned
    Joined:
    Nov 9, 2016
    Likes Received:
    194
    Trophy Points:
    43
    Location:
    Toronto, Ontario, CANADA
    From these charts it looks like the HD650 is actually a decent headphone... Who knew.

    Seriously though, I am surprised at these. I would not have matched any of the results to the headphones used. Wow. Thanks for doing this - very interesting!

    It is possible that there is a seal issue with the 207's? It just seems so odd that they would measure the way they did.
     
    bigfatpaulie, Jan 9, 2017
    #32
  13. aufmerksam

    aufmerksam Friend

    Pyrate
    Joined:
    Sep 28, 2015
    Likes Received:
    1,337
    Trophy Points:
    93
    Location:
    E. Lansing, MI
    something something Peter North something something...
     
    aufmerksam, Jan 9, 2017
    #33
  14. 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 actually think most of it is the enclosure, especially with the creatology foam, but the STAX Lambda drivers do tend to have resonances in the mid-treble region. In that case I think it's a driver geomety thing. I wonder how much of a role diaphagm thickness or the stator material/shape plays here. Thicker diaphragms are generally less ringy/more damped, but less sensitive and lower bandwidth.

    I did notice the same small ridges in my CSDs with the SR-202 that you can see in a lot of the changstar STAX CSDs. I think the HE90 had less of these tiny resonances.
     
    Serious, Jan 9, 2017
    #34
  15. gixxerwimp

    gixxerwimp Professional tricycle rider

    Pyrate
    Joined:
    Sep 30, 2015
    Likes Received:
    5,776
    Trophy Points:
    113
    Location:
    small island claimed by China
    Just curious, was the polarity of the HD800s and HD650s reversed? Or is the leading edge of the burst random?

    Ok. Now I'm certain you aren't actually 18 years old.
     
    gixxerwimp, Jan 15, 2017
    #35
  16. Kattefjaes

    Kattefjaes Mostly Harmless

    Pyrate
    Joined:
    Sep 5, 2016
    Likes Received:
    4,521
    Trophy Points:
    113
    Location:
    London, UK
    ..if it is random , would some kind of root mean square plot be easier to read?
     
    Kattefjaes, Jan 15, 2017
    #36
  17. purr1n

    purr1n Desire for betterer is endless.

    Staff Member Pyrate BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    89,909
    Trophy Points:
    113
    Location:
    Padre Island CC TX
    Random in the sense that I used alligator clips on the XLR pins and may have swapped them here and there between measurements. I couldn't find my balanced to SE phono adapter.

    Just flip in Photoshop.
     
    purr1n, Jan 15, 2017
    #37
  18. Serious

    Serious Inquisitive Frequency Response Plot

    Pyrate BWC MZR
    Joined:
    Sep 28, 2015
    Likes Received:
    2,594
    Trophy Points:
    113
    Location:
    near Munich, Germany
    Hahaha, I meant some old audio magazines that I found on the internet.
     
    Serious, Jan 15, 2017
    #38
  19. qetadg

    qetadg New

    Joined:
    Aug 11, 2016
    Likes Received:
    11
    Trophy Points:
    3
    No earphones? would be interesting to see how BA drivers compare
     
    qetadg, Jan 19, 2017
    #39
  20. JohnM

    JohnM Author of REW - Rando

    Joined:
    Dec 13, 2015
    Likes Received:
    134
    Trophy Points:
    33
    Using a rectangular gated toneburst means the stimulus has a lot of high frequency content even for a low frequency carrier, which may make the results a little harder to interpret - particularly when directly viewing the output, might want to use a Hilbert transform to generate the envelope of the responses if doing it that way. ARTA's burst decay plot derives results equivalent to using tonebursts shaped with a Gaussian envelope so that the burst has a consistent bandwidth, either 1/3 or 1/6 octave according to user selection. A wavelet approach is used to directly produce the envelope of the burst responses across the frequency range. Quite a bit faster :)
     
    JohnM, Mar 21, 2017
    #40
Page 2 of 14

Share This Page