A passing idea on vibration analysis.....

Discussion in 'Measurement Techniques Discussion' started by OJneg, Oct 30, 2015.

  1. OJneg

    OJneg The Most Insufferable

    Pyrate BWC
    Joined:
    Sep 25, 2015
    Likes Received:
    3,923
    Trophy Points:
    113
    Location:
    Grand Rapids, MI
    So it occurred to me back at the Source dealer show that headphone construction makes a huge difference. I'm talking sonically of course.

    Sitting down the Dharma next to the HD800 there's obviously a lot of similarities, and some striking differences. I think a lot of similarities can be attributed to the fact that each headphone has similar earcup dimensions and positioning. This alone might sound like silly audiophool logic but it starts to make sense when you think about the physics of these systems. The obvious differences that can be easily heard are the less capable dynamic driver (i.e. bass distortion) and the ES tweeter which has a distinct texture. But beside those issues I also think there's the question of the Dharma's enclosure. Now I'm not writing this because I think Dharma's a hack job or that the enclosure is a resonant POS. In fact, the Dharma's enclosure is probably more dead than a lot of other headphones. Comparing the Dharma's spatial and imaging qualities against the HD800 was also interesting to note. They're more similar than different in this sense (great wide and deep staging) but I still came away from the comparison with the feeling that the Dharma just doesn't quite have that pitch-black, micro-dynamic, high-contrast, out-of-nowhere transparency that can really fool you into thinking you're hearing the real deal. Again, you could try to pin this on the previously stated issues, but as we all know, the HD800 (even modded) isn't without its flaws either. The fact that the HD800 and Dharma were so similar tonally and spatially but still different in this last nagging regard got me thinking about what other technical difference might be the root of what I'm hearing.

    The vast majority of headphones I hear don't have this "fool-ya" quality. Most stuff makes the term "grey" come to mind. Off the top of my head I can only think of a few...the HD800, Merv's modded Abyss from a few years back, and the SR009. These headphones don't have too much in similar on the surface; all different driver technologies, different warts in different places, etc. They all pass the necessary prereqs for hi-fi of course: Acceptably neutral tonal balance and very low distortion and well separated staging. But more than anything else I think what binds these few together are the enclosures. The Abyss is basically a big heavy chunk of aluminum. The SR009 is all machined aluminum which seems fairly dead (although FWIW, I think the SR009 is behind others in terms of resolving power for whatever reason). And then there's the HD800, while not made of aluminum, is still very dead. I think the HD800's enclosure might be the most dead of any headphone. We should thank whatever space-age composite the evil German scientists at Sennheiser decided to use. I remember when idiots on Head-Fi used bag on the HD800 for being made of "plastic". The Beyer T1 of course would have been their weapon of choice over the HD800, because obviously being made of metal was the sign of quality. The look AND sound of metal. :Violin:

    So our predictions of "fool-ya" factor based on the materials and construction involved can only go so far. But it's obvious to me that cheap plastic enclosures can create a certain coloration. Anyone who's modded a TH02 or T50rp could tell you that. There's also the issue of certain 3D printed plastics, which quite honestly, sound like crap. Alpha Dog/Prime come to mind; that confusing click-clacky sound as Merv calls it. Then you can look at AT's woodie series which have a distinct (non-plastic) more romantic sound. More pleasant to hear but still fatiguing to me for long sessions. I'd be willing to bet that whatever material and damping scheme that Fostex uses on the TH900 is more dead than any of the AT woodies. You could list many other headphones (closed and open) that mimic the sound of their enclosures.

    Anyone can readily test the deadness of their headphone by just tapping on the outside of the enclosure while wearing it. Try it with different headphones and the same tap will produce a different sound, just like knocking on different speaker cabinets will produce a different sound. I am interested in devising a system and methodology with which to quantify the deadness of headphone enclosures. The process should be repeatable and comparable for different headphones.

    Similar to how CSD's were thrown around during the early CS days as a way to explain what folks were hearing, I think this idea lies in a similar vein. Or again, it might be a massive dead-end that's either too hard to realize practically or just not terribly revealing of what's going on. I accept that possibility but still think it's worth investigating. Any ideas, research, or critique are welcome; I'll add them to this first post.

    Links:

    http://www.pcb.com/Linked_Documents/AutomotiveSensors/AUTO_NVH_Lowres.pdf

    http://www.meas-spec.com/downloads/Piezo_Film_Product_Guide.pdf
    ....
    ....
     
    Last edited: Oct 30, 2015
  2. kapanak

    kapanak Canucklehead - Friend

    Pyrate
    Joined:
    Sep 27, 2015
    Likes Received:
    259
    Trophy Points:
    63
    Speaking of the HD800 enclosure, I doubt any other manufacturer has put as much thought into making the enclosure as 'dead' as you say. Everything from the cups, to the hinge and even the headbands. Certainly an interesting perspective to look at other headphones and their enclosure 'sound'.

    A possible way of quantifying this "deadness" is some variation of mechanic's stethoscope or a LDV (Laser Doppler Vibrometer), performing point to point vibration velocity measurements while playing a constant frequency tone. There are continuously scanning versions of LDVs as well, so you could potentially perform a frequency spectrum sweep and measure the enclosure throughout at various points simultaneously, or just use a single point LDV and choose multiple points to measure at different frequencies. Definitely not something as accessible as the typical measurement system with a mic and coupler lol. I'm also not sure it is DIY-able. You might have to actually buy or have access to a very precise LDV.
     
    Last edited: Oct 30, 2015
  3. OJneg

    OJneg The Most Insufferable

    Pyrate BWC
    Joined:
    Sep 25, 2015
    Likes Received:
    3,923
    Trophy Points:
    113
    Location:
    Grand Rapids, MI
    Interesting idea, but an accelerometer or contact piezo mic seems like the cheaper more doable solution
     
  4. thune

    thune Friend

    Pyrate
    Joined:
    Sep 26, 2015
    Likes Received:
    50
    Trophy Points:
    18
    Let me state first: I think enclosure resonances are only an issue if the resonance is bad. The importance of minor resonances may be exaggerated due to claims made by modders who are hoping for big changes from limited methods.

    Anyway:
    The trick in a standard is achieving a standard "knock". And arguably the only "knocking" that is important is what the driver produces. A resonance that never gets stimulated by the driver activity (but gets stimulated by a precisely located rapping) is no big deal.

    The way Atkinson of Stereophile achieves a consistent "knock" is to excite the loudspeaker with a Frequency Response measurement signal (in his case MLS) and extract the impulse response and CSD from the signal produced by the accelerometer fastened to the enclosure in various spots. (Atkinson limits the bandwidth to 2khz, a range that the accelerometer can cover.)

    So a reasonable method for such a measurement would be to use you favorite measuring software and impulse response measurement method, and to pick a fairly high but reasonable SPL.

    The other question is what points to measure with the accelerometer.

    I'm not sure all this is worth the effort, but it might be fun.
     
    Last edited: Oct 31, 2015
  5. Klasse

    Klasse Friend

    Pyrate
    Joined:
    Sep 28, 2015
    Likes Received:
    117
    Trophy Points:
    43
    Headphone = Driver + Enclosure
    Headphone - Driver = Enclosure

    Let's consider M as the measure operator

    M(Headphone - Driver) = M(Enclosure)

    Hopefully,

    M(Headphone) - M(Driver) = M(Enclosure)

    /////

    M(Headphone) : a piece of cake...

    M(Driver) :It would be interesting to discuss what would be the most useful way of measuring the Driver separately in order to reveal the Enclosure properties.
     
  6. drfindley

    drfindley Secretly lives in the Analog Room - Friend

    Pyrate
    Joined:
    Sep 25, 2015
    Likes Received:
    1,533
    Trophy Points:
    93
    Location:
    Austin
    I was just thinking about this. I really love the sound of my UERMs. There's something about the immediacy of them, the driver being near your eardrum that I find so pleasing. You can hear the sound of the plastic they resonate in. I feel like I can hear the different things in my HD650/HD800s that work to shape the sound.

    On the flip side, at Red Rocks amphitheater, on of my favorite sounds was the sound of the drums ricocheting off the to large rock walls. The attack was so delightful and I've not experienced much like it before and since.
     
  7. Thad E Ginathom

    Thad E Ginathom Friend

    Pyrate
    Joined:
    Sep 27, 2015
    Likes Received:
    14,134
    Trophy Points:
    113
    Location:
    India
    Is plastic being wronged here?

    I've got lots of metal things around the house that vibrate in a very extreme way. Apart from the ones actually called bells, there a a variety of pots, pans, steel plates, brass things etc etc that will give me a note if struck, which means they'll resonate at the note --- assuming something isn't damping them, of course: I used to have fun "playing" the pots and pans in the kitchen when I kept them hanging up, but they don't ring much when sitting on their bases.

    There are other materials, of course, that vibrate wonderfully. I can think of singing wine glasses and ceramic dishes (various sizes, varying amounts of water, makes a musical instrument). There is a Spanish percussion instrument that looks like not more than a wooden box with a hole in it (is it the wood vibrating? Or the air inside it?)

    I'm not an engineer, but the last thing I would think of making a half-decent bell out of would be plastics. But then, there are so many different kinds of plastic, that maybe I could be proved wrong about that. Maybe this post is all woffle. hey, it is! but I don't think the materials thing is that clear cut.

    Which makes the idea of the measurements all the more interesting.
     
  8. OJneg

    OJneg The Most Insufferable

    Pyrate BWC
    Joined:
    Sep 25, 2015
    Likes Received:
    3,923
    Trophy Points:
    113
    Location:
    Grand Rapids, MI
    I would like to test both methods of excitation. Run PN and test enclosure for vibration. And also give external "tap" excitation and look at spectral components.

    Bandwidth of 2k corresponds to most off-the-shelf accelerometers I see. I'd like to order a few different ones, pair them with a LNA, and see what sort of results I get.

    The HD800's enclosure is "plastic" and I suspect it is the most dead enclosures out there. Sorry if any plastic fanboi's out there feel insulted:p

    In terms of metals I suspect thick, machined, unibody aluminum enclosures will prove to be more rigid and inert than most other materials. You often see the best turntables use such design.
     
  9. purr1n

    purr1n Desire for betterer is endless.

    Staff Member Pyrate BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    89,777
    Trophy Points:
    113
    Location:
    Padre Island CC TX
    Grado uses resonant wood for RS-1 / RS-2. It does add a sonic flavor that the metal SR-325 and plastic SR-225 (and below) don't quite have. You can see these microvibrations the CSDs. The floor isn't quite as clean.
    RS1 L flatpads.txt.jpg
     
  10. Psalmanazar

    Psalmanazar Most improved member; A+

    Pyrate Slaytanic Cliff Clavin
    Joined:
    Sep 27, 2015
    Likes Received:
    5,345
    Trophy Points:
    113
    Aluminum dents.

    Could bass vibrations be why Audeze's crack?
     
  11. kapanak

    kapanak Canucklehead - Friend

    Pyrate
    Joined:
    Sep 27, 2015
    Likes Received:
    259
    Trophy Points:
    63
    I still think LDV is a better idea for measurements compared to an accelerometer contact based sensor. The cost may be higher, but for a small enclosure like that of headphones, I don't think we would get as accurate a reading with a physical contact based sensor. Even speaker designers use LDVs, among other acoustic device/instrument designers and engineers.

    Now on the matter of where to get a single point LDV, I have an idea, but need to check with my supervising professor first.
     
  12. keanex

    keanex Martian Bounty Hunter - Friend

    Pyrate
    Joined:
    Oct 1, 2015
    Likes Received:
    320
    Trophy Points:
    63
    Interesting topic. I think thait makes sense and looking at headphones that were modded to have wooden enclosures compared to before seem to point to this.
     
  13. Armaegis

    Armaegis Friend

    Pyrate BWC
    Joined:
    Sep 27, 2015
    Likes Received:
    7,464
    Trophy Points:
    113
    Location:
    Winnipeg
    My first thought on reading this: attach sensor to enclosure, run a frequency sweep, find natural resonances of the enclosure, compare to acoustic data and look for trends (particularly in distortion; and I'd be intrigued if it found its way into impedance).

    Of course, then a followup to this would be to minimally alter the enclosures in such a way as to alter the enclosure resonances (both frequency and amplitude) and again compare to changes (if any) in the acoustic profile.
     
  14. johnjen

    johnjen Doesn’t want to be here but keeps posting anyways

    Pyrate
    Joined:
    Sep 29, 2015
    Likes Received:
    505
    Trophy Points:
    93
    Location:
    Well Grounded
    I just stumbled upon this thread.
    It does bring up a few interesting paths of inquiry, at least for me.

    But (the ubiquitous But), given my circumstances, I'm taking an alternative approach.

    Performing structure born resonance measurements will undoubtedly provide further insights, especially if there is a demonstrable relationship between them and the HP's FR or other measurable metrics (THD, impulse response etc.)
    And these further insights, in some cases may point at significant DIY'r solutions or improvements.

    But (again), in my case with limited HP's to use, specifically my HD800's, which use a very carefully designed and crafted set of materials and the overall 'deadness' of the entire structure seems to have been a purposeful design goal.
    One that they achieved to a degree that only Sennheiser is aware of.

    But (last one) the 800 is what it is and modifications to the structure(al) will probably just further complicate an already complex situation.

    So, instead of focusing upon the structure and what its characteristic 'signature' is, I'm focusing upon vibrational management techniques.

    I'm hoping to have some measurements to demonstrate the relative effectiveness of my approach before too long.

    JJ
     
  15. purr1n

    purr1n Desire for betterer is endless.

    Staff Member Pyrate BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    89,777
    Trophy Points:
    113
    Location:
    Padre Island CC TX
    Dynamat or massloading on cups can reduce bass distortion and clean up the CSDs (and as an effect, make driver ringing even more apparent). I had some very old measurements of AT woodies which demonstrated this, but I can't find them.
     
  16. Armaegis

    Armaegis Friend

    Pyrate BWC
    Joined:
    Sep 27, 2015
    Likes Received:
    7,464
    Trophy Points:
    113
    Location:
    Winnipeg
    I can't remember who, but there's an audio guy on youtube (more PA stuff than hifi) who showed measurements of a few headphones and the differences produced by simply touching the cup. It's all related... mass loading, stiffness, vibration absorption, all contribute to a cleaner driver behaviour.

    Years ago in one of the Grado modding threads (before ortho madness took over), there was some experimentation with using small pegs to couple the driver to the grill with some mild compression to reduce vibration. Kinda pointless these days though since all the mods/transplants float the driver on a foam ring.
     
  17. MF_Kitten

    MF_Kitten Banned per own request

    Banned
    Joined:
    Jan 1, 2016
    Likes Received:
    181
    Trophy Points:
    43
    The Beyerdynamic Custom One Pro Plus and Custom Studio both have a round cylinder sticking out of the rear of the cup, which the back of the driver is pressed against with a rubbery material between them. Very interesting!
     
  18. Armaegis

    Armaegis Friend

    Pyrate BWC
    Joined:
    Sep 27, 2015
    Likes Received:
    7,464
    Trophy Points:
    113
    Location:
    Winnipeg
    There is indeed. Given how the large the baffle is, and how it simply "floats" (it is not screwed down; merely sandwiched into the cup with a small retaining ring and the foam acting as a gasket), that peg is needed to literally help hold it in place.
     
  19. MF_Kitten

    MF_Kitten Banned per own request

    Banned
    Joined:
    Jan 1, 2016
    Likes Received:
    181
    Trophy Points:
    43
    Not really though, consider that the DT880 and 990 don't have those, and the DT770 has it in some models but not in others. That cylinder is there for a different reason in those.
     
  20. Armaegis

    Armaegis Friend

    Pyrate BWC
    Joined:
    Sep 27, 2015
    Likes Received:
    7,464
    Trophy Points:
    113
    Location:
    Winnipeg
    It does help with the stability though. The COP baffle plate is even kinda wobbly without it
     

Share This Page