TH900 mod extravaganza (stream of consciousness)

Discussion in 'Modifications and Tweaks' started by E_Schaaf, Jun 16, 2020.

  1. Gadril

    Gadril New

    Joined:
    Sep 30, 2020
    Likes Received:
    1
    Trophy Points:
    3
    Location:
    Slovenia
    Very impressive! Now that is some really in depth modding here.

    This thread gives me a new energy to pick up the Aliexpress Foster 50mm drivers and go back to messing with em.
    While I mostly figured out how to dampen the baffle with foam and filtering (especially higher density ones), but I was very much loath to touch the filtering on the back of the driver (even tho I knew that it needed mods) as I feared having to find a replacement that wont ruin the thing. The medical tape you mentioned here very much helps me solve this problem and honestly I would never have even thought about it.

    I also have a questions about the Sushi mod, since it seem to very nicely reduce the severity of the dip at 3k-3.6k. Is this mod basically a kind of flow guide on the back of the driver? If so I am guessing your profile picture is roughly how it looks :) (but not encompassing the baffle vents, instead just being a extension on the driver back?).

    Anyways keep up the great work and keep posting! This is really interesting to follow :punk:
     
  2. E_Schaaf

    E_Schaaf MOT: E.T.A Headphones

    Pyrate BWC
    Joined:
    Mar 28, 2017
    Likes Received:
    9,589
    Trophy Points:
    113
    Location:
    Staten Island, NY
    Home Page:
    [​IMG]

    Playing with some new venting schemes and baffle foams. Addicted to this unabashed basshead mod.

    edit - smoothed out just a bit. I don't mind the 1.2k push, the preceding dip lines up with the upper mid area so it doesn't feel overly forward or hollowed out.

    [​IMG]
     
    Last edited: Nov 12, 2020
  3. E_Schaaf

    E_Schaaf MOT: E.T.A Headphones

    Pyrate BWC
    Joined:
    Mar 28, 2017
    Likes Received:
    9,589
    Trophy Points:
    113
    Location:
    Staten Island, NY
    Home Page:
    A story in 5 parts

    [​IMG]
    [​IMG]
    [​IMG]
    [​IMG]
    [​IMG]
     
  4. yotacowboy

    yotacowboy McRibs Kind of Guy

    Pyrate Contributor
    Joined:
    Feb 23, 2016
    Likes Received:
    10,697
    Trophy Points:
    113
    Location:
    NOVA
    Home Page:
    @E_Schaaf - I just blitzed through this thread after picking up AH-D5200s. First, thanks for doing all this!! Second, what would be your recommendation to get a slight boost/lift below 50hz, without changing pads? Does opening up the holes in the baffle do that? So far I've removed all paper screens behind the frame of the drivers and over the pole piece, and dynamatted the magnet and driver frame, and cups. Will be picking up some open cell foam or wool felt to line the cups. Also, interestingly, the foam around the driver (front side) is noticeably more porous than either my old D2000s, or my old TH-X00s.

    [​IMG]
     
  5. E_Schaaf

    E_Schaaf MOT: E.T.A Headphones

    Pyrate BWC
    Joined:
    Mar 28, 2017
    Likes Received:
    9,589
    Trophy Points:
    113
    Location:
    Staten Island, NY
    Home Page:
    The 5200 does have a different baffle build and cup dimensions vs the Fostexes so the same methods might not work as consistently, but for more sub bass you need both more airflow AND a better seal on the front. So a front foam ring that's denser than the porous ring, but also with some holes around the edges of the ring to allow pressure exit / avoid flex is what you'd need. Angling of the front foam holes does tune the sound (is it a solid front foam ring or does it have little vents like the Fostex rings?) - you want to have open airflow, but not direct airflow.

    This is a balancing act against airflow on the baffle itself. You can puncture through some of the holes (only outer circle, not the inner circle) to effectively tune the venting (imagine it's a subwoofer), but then you're also allowing more opportunities for weird phase-related dips if the airflow is too direct from the cup's rear to the ear. The ideal compromise for linearity and extension is like halfway between a port and a labyrinth if you can imagine that. Or you can seal the baffle completely and use longer cup screws so that it kind of floats behind the driver. Then you can have a LOT of airflow with the dense front rings and get a nice bassy tone without mid dips given there's no rear-to-ear pathway through the baffle.

    edit - Also I don't suggest running ALL the rear vents wide-open until you've figured out a cup damping scheme that can control the backwave. Use micropore tape to test - different patterns of vent covered vs uncovered will direct treble energy with more or less focus to the ear. Again, a balancing act vs whatever stuffing you have going on in the cup. In general, of the 11 vents on the driver, 5-7 open seems to be the sweet spot for most mods.

    edit 2 - For cup stuffing (this is going to be weird)... imagine the cup is a full bowl of water, with many laminar streams of water pouring directly into the center of the cup (that's your backwave energy coming through whatever driver vents you have open). Think about the fluid dynamics here, how the energy is going to flow according to the shape of the cup and travel back out through your vents (wherever they may be and however large / symmetrical or asymmetrical they may be). Think about where pressure is escaping and where that's positioned relative to your ear. The best damping schemes have a gradient filtering of that energy from the center out through the sides.

    The best stuffing often looks like concentric circles, with a bit of airflow in a few key places, more or less density in a few places, and some directly-coupled mechanical (as opposed to acoustic) damping in a few other key places. Key places meaning relative to each other - again, there's no absolute solution. Once you find the right general balance, positioning can shift things like transient and image focus. The degrees of subtlety here definitely hit diminishing returns quickly vs a brute force KISS stuffing solution, but it's the key difference between a mod that's pretty good but not necessarily free from trade-offs and one that feels like it has a refined sonic vision. Sometimes asymmetry is the solution too, if you're trying to address certain narrow bands in the tone. It's like sculpting water. More airflow and more symmetry on both sides of the driver generally will push these Foster bio variants towards a more contrasted W-tone. Try symmetry on the back and asymmetry on the front for a more linear slope, or vice versa.

    edit 3 - @yotacowboy if you still have a pair of Fostex front foam rings and take a side-by-side pic with the Denon rings, I'll be happy to cut and mail you a few different ring types for you to play with on the 5200. Cutting with scissors is impossible with such thick foam, so I use leather dies and a hammer. If you want to tip me a bit for materials cost and future R&D for my own builds, send me a DM with your address and I'll get a few different pairs (different firmness, different venting) out to you in a few days.

    It's fickle and complicated, this is a warning! You'll find there's no singular mod that fixes these headphones. Most mods come with trade offs that require further mods to control and dial things in. It's about the synergy between each choice and change you make. I suggest working back to front. Good luck, happy to help with further tweaks once you've found a good first step.
     
    Last edited: Dec 3, 2020
  6. yotacowboy

    yotacowboy McRibs Kind of Guy

    Pyrate Contributor
    Joined:
    Feb 23, 2016
    Likes Received:
    10,697
    Trophy Points:
    113
    Location:
    NOVA
    Home Page:
    Thanks so much for all of this! From your description, it seems to me that one possible takeaway is that these closed Fostex/Denon cans seem to have a most even response using an aperiodic damping/venting scheme. I'll need to do some more reading on this. But your work spacing the cup from the baffle is very similar to something I've seen in DIY box speakers years ago where you create a lossy baffle, sort of slot-loading the driver by spacing off the baffle by a few, or several, millimeters, and by placing a ring of acoustuff between the back of the driver frame and the baffle you resistively tune how lossy it is.

    I will say, I don't have any HP measurement gear, so I'm tuning by ear, for better or worse. So far the mods I've done in the pic above have increased both macro and micro dynamics (macro, considerably), smoothed out the 40hz bass hump, while lowering the broad sub-100Hz "bass boost tilt", and added a ca. 2kHz peak. If I can either shift that 2k peak up into the 5k range (better for "imaging") or tame it altogether, and then add back in a 2-3dB rise in the sub 50Hz range, these are going to make me sell both Aeolus and Australis. I'll be futzing with PEQ to see (generally) where things can be improved, and then try to make physical tweaks to try to get the desired changes. I think I might order a few different sheets of this stuff to play around with in front of the driver:

    https://www.mcmaster.com/wire-cloth/super-small-particle-filtering-stainless-steel-wire-cloth/

    My Aeolus have what looks to be a ca. 100-200 micron filtering cloth in front of the drivers, so perhaps worth experimenting with.
     
  7. E_Schaaf

    E_Schaaf MOT: E.T.A Headphones

    Pyrate BWC
    Joined:
    Mar 28, 2017
    Likes Received:
    9,589
    Trophy Points:
    113
    Location:
    Staten Island, NY
    Home Page:
    Pad swap could probably yield those changes. These headphones like shallow pads with oval earholes and medium to high density internal pad foam. The older Denons, EMU, and THX00 cheat a bit by using pads that partially obstruct the driver's front face with an earhole that's narrow and offset from center to help attenuate treble and bolster lower midrange. The Audio Technica AP2000 Ti pad is the only pad I've found so far that doesn't obstruct the driver's front face to any degree and also doesn't exacerbate the lower mid dip and mid treble spike. I've tried basically every leather pad from 85mm-110mm, circular and ovular, available on aliexpress. Then again the D7200 quite evenly on Marv's thread here, and I haven't seen D5200 measurements, so they might not need this.

    And then there's these pads ( https://www.amazon.com/Replacement-Maestro-GMP400-435-450PRO/dp/B08F9CJ3GQ/ ), which give the smoothest treble, but can make the middle mids (800-1200hz) a bit hot depending on the mod. And they don't seal quite as well. T60rp stock pads could be a good choice but haven't measured those yet.

    Haven't played with metal meshes yet, didn't know where to source them. Thanks for this pointer! Hope it's easy to cut. My fancy Gingher shears are starting to dull.
     
    Last edited: Dec 3, 2020
  8. E_Schaaf

    E_Schaaf MOT: E.T.A Headphones

    Pyrate BWC
    Joined:
    Mar 28, 2017
    Likes Received:
    9,589
    Trophy Points:
    113
    Location:
    Staten Island, NY
    Home Page:
    @dncnexus sent me a stock TH900 for mods. I realized I never posted a stock measurement on this thread, so here we go:

    [​IMG]

    And a fresh set of measurements for one of mine (always tweaking this pair, #7 rebuild 4) -

    [​IMG]

    (#8 rebuild 2) -

    [​IMG]

    Note something extremely interesting I hadn't noticed before: the ESX900 mods make the headphone about 6-8db more efficient through most of the audible band. I didn't change the volume between these measurements. If the stock TH900 is 100db/mw, these are 108db/mw. Exceptionally easy to drive. Truly an anti-HE6.

    In the impressions thread I mentioned that these headphones make me think about what 'BWC' means in the headphone world. Effiecency and airflow certainly play some part - this headphone is a true lens for a source and amp. @DEATHxMACHINE mentioned this in his review too. I love how this headphone scales but also retains its tonal character as you swap to lesser chains.

    I wish I could capture the improvement in dynamic clarity with the improved airflow. Something I have no way to measure...
     
    Last edited: Dec 7, 2020
  9. E_Schaaf

    E_Schaaf MOT: E.T.A Headphones

    Pyrate BWC
    Joined:
    Mar 28, 2017
    Likes Received:
    9,589
    Trophy Points:
    113
    Location:
    Staten Island, NY
    Home Page:
    Just a teaser, still have quite a bit of work to do with the voicing -

    [​IMG]
     
  10. yotacowboy

    yotacowboy McRibs Kind of Guy

    Pyrate Contributor
    Joined:
    Feb 23, 2016
    Likes Received:
    10,697
    Trophy Points:
    113
    Location:
    NOVA
    Home Page:
    Holy shit! is that a passive radiator?
     
  11. E_Schaaf

    E_Schaaf MOT: E.T.A Headphones

    Pyrate BWC
    Joined:
    Mar 28, 2017
    Likes Received:
    9,589
    Trophy Points:
    113
    Location:
    Staten Island, NY
    Home Page:
    Yup!
     
  12. yotacowboy

    yotacowboy McRibs Kind of Guy

    Pyrate Contributor
    Joined:
    Feb 23, 2016
    Likes Received:
    10,697
    Trophy Points:
    113
    Location:
    NOVA
    Home Page:
    Also, found some perhaps easier to deal with micron-sized SS filtering screens to play with: Aeropress aftermarket filtering screens! Like I mentioned earlier, my old Aeolus had what looked like a 100 or so micron filter screwed down over the drivers, so trying out some different porosities might be an option for tuning. I've ordered a few different porosities to try. I've been digging around trying to find a way to calculate the acoustic impedance of a screen based on the screen aperture without much luck. Only thing I did find was that it looks like the manufacturer for the screen material used in Sennheiser headphones is probably these guys:

    https://www.bopp.com/
     
  13. E_Schaaf

    E_Schaaf MOT: E.T.A Headphones

    Pyrate BWC
    Joined:
    Mar 28, 2017
    Likes Received:
    9,589
    Trophy Points:
    113
    Location:
    Staten Island, NY
    Home Page:
    Never crossed my mind to try the Aeropress filter, that's honestly genius! I've had 'good enough' results with the materials I have so I might stick to my now-standard mods as far as TH900s go, but that'll definitely be useful for my upcoming scratch build with high impedance N52 drivers.
     
  14. yotacowboy

    yotacowboy McRibs Kind of Guy

    Pyrate Contributor
    Joined:
    Feb 23, 2016
    Likes Received:
    10,697
    Trophy Points:
    113
    Location:
    NOVA
    Home Page:
    This should probably be in the 5200 thread, but if I can find a way to knock down the response from about 1k to 2k by 2db without affecting anything else, the 5200's are really close to HD600 with a subwoofer... I've got a couple more tricks to pull back the 1-2k hump down but waiting on materials.
     
  15. Armaegis

    Armaegis Friend

    Pyrate BWC
    Joined:
    Sep 27, 2015
    Likes Received:
    7,468
    Trophy Points:
    113
    Location:
    Winnipeg
     
  16. bobboxbody

    bobboxbody Friend

    Pyrate BWC Contributor
    Joined:
    Sep 6, 2016
    Likes Received:
    1,465
    Trophy Points:
    93
    Location:
    GR, MI
    Does the passive radiator replace some/all of the venting and make for a more closed system, and how does that effect isolation from outside noise? I have my donor 610's in hand and am 6 weeks in to the 6-8 week full compass window, but might have to wait and see how this passive radiator idea pans out, the cool factor is very high for me.
     
  17. E_Schaaf

    E_Schaaf MOT: E.T.A Headphones

    Pyrate BWC
    Joined:
    Mar 28, 2017
    Likes Received:
    9,589
    Trophy Points:
    113
    Location:
    Staten Island, NY
    Home Page:
    This particular pair of cups defeats the air gap system that vents out through the sides of the front mount ring like we see on the stock cups. This was a minor miscalculation during the build process, but I can work around it and hope to use it to my advantage after some experimentation. Tommytakis pair of XL cups also does this.

    I am still tuning airflow to see what scheme of baffle vents gives best performance. I have some 3D printed front mount rings that have additional ports built into the sides to get around the blockage if I find it to be problematic. The funnel shape on the passive radiator cups is meant to direct and concentrate backwave pressure and reduce cup volume to really get the passive moving. The passive element is slightly heavier than I would prefer, so might have to investigate other radiators if I'm not pleased with the end result. Or just really crank things up to get it moving. Or replace the baffle paper with something else to trap more energy. Too soon to tell.

    Also, this particular design may be a one-off if I can't get around some of these little quirks. The further I go, the more I'm realizing I'll stick to my guns with commissioned TH900 mods now that I have a recipe that consistently works. In many ways, the stock cup shape does work most consistently. I'm not trying to 'best' what I have vs the standard mod, this is just something I'm playing with out of curiosity. In the future I'll keep all the wacky experimentation reserved for the scratch build which is coming along nicely. I do see passive radiators making it into that design scheme as well.
     
    Last edited: Dec 22, 2020
  18. Armaegis

    Armaegis Friend

    Pyrate BWC
    Joined:
    Sep 27, 2015
    Likes Received:
    7,468
    Trophy Points:
    113
    Location:
    Winnipeg
    No one has done isobaric headphones yet. Maybe you should try :D

    (I mean, if you're doing passive radiators you're halfway there anyways...)
     
  19. E_Schaaf

    E_Schaaf MOT: E.T.A Headphones

    Pyrate BWC
    Joined:
    Mar 28, 2017
    Likes Received:
    9,589
    Trophy Points:
    113
    Location:
    Staten Island, NY
    Home Page:
    It's certainly crossed my mind, give me a year and I'll let you know how it goes :p
     
  20. memoryerror

    memoryerror New

    Joined:
    Apr 20, 2019
    Likes Received:
    8
    Trophy Points:
    3
    Location:
    California
    [​IMG] [​IMG] I have been modding my Emu Teaks and I really like Dan Clark Audio's Ether 2 Perforated pads https://www.danclarkaudio.com/perforated-ether-2-ear-pads.html
    But the adhesive does not work the greatest, so I used some carpet tape on the pad ring.
    I added the acoustic paper like ring (from a Sennheiser HD540 or HD250 pad kit) to try and bring down a peak somewhere between 5k an 8k.
    I am interested in where you sourced the foam for your "Sushi Mod".
     
    Last edited: Jan 10, 2021

Share This Page