HD800 Measurements

Discussion in 'Headphone Measurements' started by Marvey, Sep 1, 2016.

  1. Marvey

    Marvey Loves sex and records

    Staff Member Friend
    Joined:
    Sep 24, 2015
    Likes Received:
    27,161
    Trophy Points:
    113
    Location:
    Calabasas, CA
    HD800 Measurements (stock, no mods)

    I realized I never posted these for the V2 coupler

    HD800 Frequency Response (0db = 100dbSPL)
    HD800 FR.png

    HD800 Distortion (at above levels)
    HD800 L HD.jpg

    HD800 CSD
    HD800 L.jpg
     
    eddypoon, Vanheim and rayfalkner like this.

  2. songmic

    songmic Friend

    Friend
    Joined:
    Sep 28, 2015
    Likes Received:
    815
    Trophy Points:
    93
    Location:
    Seoul, South Korea
    Why do I see a long schlong in the upper region?
     
    Priidik likes this.
  3. Marvey

    Marvey Loves sex and records

    Staff Member Friend
    Joined:
    Sep 24, 2015
    Likes Received:
    27,161
    Trophy Points:
    113
    Location:
    Calabasas, CA
    Haha. Because there is. It's quite penetrating, unless you use mods to protect yourself.
     
    Priidik, swamp and cskippy like this.
  4. Klasse

    Klasse Friend

    Friend
    Joined:
    Sep 28, 2015
    Likes Received:
    103
    Trophy Points:
    43
    What version of earpads?
     
  5. Marvey

    Marvey Loves sex and records

    Staff Member Friend
    Joined:
    Sep 24, 2015
    Likes Received:
    27,161
    Trophy Points:
    113
    Location:
    Calabasas, CA
    Original version.
     
    Klasse likes this.
  6. swamp

    swamp Acquaintance

    Joined:
    Jun 27, 2016
    Likes Received:
    48
    Trophy Points:
    18
    I... I can't stop looking at it.
     
  7. songmic

    songmic Friend

    Friend
    Joined:
    Sep 28, 2015
    Likes Received:
    815
    Trophy Points:
    93
    Location:
    Seoul, South Korea
    Listening to stock unmodded HD800 be like:

    [​IMG]

    For those of you who watched Scary Movie.
     
    rainer.skill and Priidik like this.
  8. Marvey

    Marvey Loves sex and records

    Staff Member Friend
    Joined:
    Sep 24, 2015
    Likes Received:
    27,161
    Trophy Points:
    113
    Location:
    Calabasas, CA
    More measurements: 40Hz, 500Hz, and 2kHz tones.

    HD800 40.png
    HD800 500.png
    HD800 2k.png
     
  9. Vanheim

    Vanheim Facebook Friend

    Joined:
    Mar 28, 2016
    Likes Received:
    81
    Trophy Points:
    28
    @stratocaster did quite an amazing job "neutralizing" his HD800. The stock HD800 schlong seems to have mostly disappeared. [​IMG]
     
  10. RonShaun

    RonShaun Rando

    Joined:
    Aug 19, 2016
    Likes Received:
    11
    Trophy Points:
    3
    So how did he achieve that FR?
     
  11. Vanheim

    Vanheim Facebook Friend

    Joined:
    Mar 28, 2016
    Likes Received:
    81
    Trophy Points:
    28
    physical mods and talent?
     
  12. RonShaun

    RonShaun Rando

    Joined:
    Aug 19, 2016
    Likes Received:
    11
    Trophy Points:
    3
    ...Physical mods and crazy good EQ, maybe?
    Don't judge me Marv.
     
  13. Marvey

    Marvey Loves sex and records

    Staff Member Friend
    Joined:
    Sep 24, 2015
    Likes Received:
    27,161
    Trophy Points:
    113
    Location:
    Calabasas, CA
    Vanheim likes this.
  14. Vanheim

    Vanheim Facebook Friend

    Joined:
    Mar 28, 2016
    Likes Received:
    81
    Trophy Points:
    28
    He didn't eq at all. This is the result of just physical modifications, go read the thread Marv linked. This mod does reduce the soundstage from what I understand, but that neutrality seems amazing coming from the hd800.
     
  15. RonShaun

    RonShaun Rando

    Joined:
    Aug 19, 2016
    Likes Received:
    11
    Trophy Points:
    3
    That's an impressive change from just changing earpads. I didn't even know this mod existed. I feel less knowledgeable than I originally thought I was; Oh well, more to learn.
     
  16. Serious

    Serious Immature child

    Friend
    Joined:
    Sep 28, 2015
    Likes Received:
    972
    Trophy Points:
    113
    Location:
    Munich
    I doubt that this mod is actually neutral once you remove the thing from a coupler and put it on your head.

    BTW @Marvey what's up with these bad distortion measurements in the bass? The ones on CS looked quite a bit better (but the newer ones seem to be less limited by the measurement rig as far as D2 goes):
    tmp_16663-hd800l1691381553.GIF

    ...Let me grab mine...
     
    Last edited: Sep 2, 2016
  17. Marvey

    Marvey Loves sex and records

    Staff Member Friend
    Joined:
    Sep 24, 2015
    Likes Received:
    27,161
    Trophy Points:
    113
    Location:
    Calabasas, CA
    Different coupler and that one above was probably run at slighty lower volume.
     
  18. HitmanFluffy

    HitmanFluffy Hoping to see real genitals someday! - Friend

    Friend Banned
    Joined:
    Sep 26, 2015
    Likes Received:
    1,352
    Trophy Points:
    93
    Location:
    Washington D.C.
    The stratocaster mod with HM5 pads works at getting an even FR, but even ignoring the effects on soundstage, the loss in clarity and transient response was too much for my liking.
     
    Bagged Milk, zonto and OJneg like this.
  19. Serious

    Serious Immature child

    Friend
    Joined:
    Sep 28, 2015
    Likes Received:
    972
    Trophy Points:
    113
    Location:
    Munich
    Wanted to get back on this...

    @Marvey: I get much lower distortion numbers than you, but my results are pretty noisy. My WM61A mic actually is less noisy than the UMIK-1, but has more 2nd order distortion. I'm using the UMIK microphone calibration, which should be accurate.
    Here are my results for 90db (at 500Hz) on the coupler. This is my modded HD800. The coupler results are almost exactly 5db lower than the in-ear measurements (about 4.8 to 5). This isn't the case with a "V2" type coupler like you're using. 90db on the coupler is 95db on the head, because the coupling is much more "lossy" on the coupler.

    HD800 90db coupler UMIK-1.png

    Also attached the WM-61A result. With the WM-61A the 5th and 4th order distortion in the bass is still slightly above the noise floor, but I get a lot of 50Hz + harmonics hum, which obviously makes the results hard to interpret. The UMIK also has some hum but not nearly as bad.

    I think the 2nd order distortion results for the midrange are still limited by the microphone + electronics. 2nd order at 90db should be more than 66db down. When I do a long FFT of a 440Hz tone the 2nd order distortion is at about -72db or 0.025% and the 3rd order distortion is entirely below the noise floor (more than 90db down), but I think that's still limited by the microphone setup. For some reason my UMIK seems to have much less distortion than @sorrodje's with pretty much exactly the same SNR, which means the levels should be the same.
    Like many others I do think the bass distortion is audible above 80db or so.

    I have to measure at 103db (a real 108db!) to get distortion measurements like yours and even then the 2nd order distortion doesn't hit 0.1% until 125Hz and the 3rd order distortion doesn't go above 0.5%*. The 2nd order distortion goes to about 2% at 20Hz in that case. From the overall shape of the THD curve I would say that these results are more consistent with Tyll's results.


    As for the 2nd order distortion spike at 3-4kHz I'm not sure if it's a driver problem or not. I've been trying to find this out haven't really come to a conclusion. To be sure I'd really like to measure the driver itself on an infinite baffle as a speaker driver. I think it's there but not as bad as your measurements seem to show. The CSDs look clean. At 100db or more it reaches about 0.2% with my measurements, but it could be lower. I would like to see your results using the harmonic frequency as reference to make more sense of it as I think the 4kHz dip on the coupler is mostly an artifact.

    I also wonder how much of the "6kHz peak" is due to baffle diffraction issues with the metal ring. A 66.8mm outer diameter would make for a 5150Hz bump and 10kHz dip, but I'm not sure how directional the driver is at that frequency (should become directional at about 4kHz or so). I think that might be why covering the hole in the middle rises 10kHz.


    EDIT:
    The main thing I was trying to get across here is that either my or your volume calibration isn't accurate.
    I like to use a real level of 95db for the distortion measurements as that feels like a more realistic "high listening level" to me. My mic is also much noisier than ears so using 100db or more might not be a bad idea. But I think my mic starts to distort at such high levels.

    *This is weird. D3 for that 103db measurement seemed to be lower from 20 to 40Hz than at 90db. Let's just say it stays below 1% much like the 90db measurement. However the 3rd order distortion is above the noise floor from 500Hz on (0.02%) and hits 0.1% at 300Hz. I wonder if it's proportional to driver excursion.
     

    Attached Files:

    Last edited: Sep 21, 2016
  20. Serious

    Serious Immature child

    Friend
    Joined:
    Sep 28, 2015
    Likes Received:
    972
    Trophy Points:
    113
    Location:
    Munich
    Here are my 95db (500Hz) in-ear mic distortion results for the HD800. Compare these to my SR-202 measurements.
    HD800L 95db.png
    Again, I had to cut them off at 30Hz because I get a lot of 50Hz + multiples hum. This means that there is a huge 25Hz D2 spike that shouldn't be there. You can see the distortion rising from 40Hz on in this plot. Realistically the distortion maybe reaches 0.5% at 20Hz or so and most likely less than that.
    Distortion is very low across the board. 3rd order distortion is slightly higher than with the STAX, but 2nd order is quite a bit lower. The 3rd order distortion is also more limited by the SNR with the HD800 because of its less screwy FR. Even in the mid-bass the distortion is lower than the SR-202. The SR-202 is a bit cleaner at 20Hz, but subjectively that doesn't matter much.

    I do get higher bass distortion on the coupler (without touching the volume dial). I think the coupler gets a different kind of seal that the driver doesn't like as much. I think these are more accurate in this case.
     

Share This Page