SEOS12 and AE TD12M design

Discussion in 'DIY Speakers and Subwoofers' started by BillWaslo, Sep 22, 2012.

  1. Here is a link to measured response and impedance files for the TD12M and the SEOS12 drivers, these are for measurement with the FRONT edge of the SEOS12 being on the same plane with the TD12M.. if the TD12M is modeled using the Z offset shown in the Notes.txt file.

    The curves were taken at 22degrees off the horizontal axis, at the tweeter's vertical center. The 22deg offset is why the very high treble drops like that. I figure that since these will always be used toed-in (RIGHT???) that 22deg is a better angle to adjust for.

    This was with a driver kindly loaned by pgwalsh over at AVS, and a phase plug supplied by HomeTheaterDoc. I also modeled (but didn't build) a crossover design for these drivers, here are the simulated freq response and impedance curves:
    (ignore the glitch around 800Hz, that's a PCD thing)


    This is the schematic:


    The schematic file is at

    Very nice driver, that too. Heavy suckers, though!
  2. Thank you for this B. This is the speaker I went with. I don't have it yet, but when I do I look forward to starting my build
  3. Thanks, Bill, looks good.

    From my rudimentary grasp of electronics, it looks like a 2nd-order LP on the woofer and 4th-order HP on the CD; is that correct?

    What's the XO freq?

    The notes say "Measured 15" from baffle, on tweeter axis, 22.5deg Horizonal, 0deg Vertical."

    Was the combined response measured from that close or just the individual drivers?

    I'd think it takes a much greater distance for the drivers' responses to integrate.
  4. wow..finally a TD12M model.

    Mr. B, why not build the x-over and have a listen?
  5. There is no set minimum distance to where drivers will integrate. Besides, the closer you're able to measure, the less of a chance of room artifacts interfering.
  6. But you don't want to be so close that it will make you way off axis from one or both drivers while measuring.

    I think I would call that 2nd order on both woofer and CD with some HF shaping.

    Looks like another fine design from Mr. Waslo
  7. Wow. Thank you so much Bill. 8)
    I'm at the stage of gluing up my test enclosure tonight, so I started adding parts to my cart on PE, but there are backorders on the 10 & 12 Ohm resistors! They've got these at Meniscus:, and ,but they are 10% tolerance. Doable?
    For L3, I'm not seeing PE#266-259. Would this work, it's .19 Ohm (<0.4Ohm):{11}184
    Also, I don't have the DNA-360, but I've got the an order with Erich for CD's in my future? ;)
  8. Mr. Waslo, how about this inductor for L3 at .36 Ohm?
    Found the 10 Ohm resistors at mouser electronics, but the 12 Ohm are backordered until 10/16. And there I thought I would be making sound this weekend... :(
  9. You can always parallel resistors to get what you need.
  10. Bunny, what box size are you using? Sealed or ported?

    I am thinking of a 2cf box sealed for my build.
  11. Bill,

    I'm on agin/off again on using active XO; if you've still got the drivers might you be able to come up with the filter parameters?


  12. Those BMS drivers seem to measure really well in the SEOS-12 from what I've seen. I'd definitely want to use them if I were you! But yes you would need a different crossover.
  13. A bunch of questions have been asked. I'm always reading this on a smart phone, on which answering isn 't so easy. So, time to catch up, if I miss anything, let me know.

    The actual measurements for the frd files were made at about 4' distance. The 15 in measurements were just used to find the effective z-offset to enter into PCD. But, the frd for the modeling had to be further because, yeah, that would be way off-axis for the woofer.

    The resistor values can be obtained in a number of ways -- for 12 ohms, for instance, you can put a 2 ohm in series with a 10 ohm to get there, or an 8 in series with a 4, or two 24 ohm resistors in parallel.... lots of ways.

    Similarly capacitors can be paralleled to get off-values (paralleling capacitors ADDS their values; series connecting resistors (or inductors) adds their values -- you seldom would use capacitors in series or inductors in parallel, since they are expensive at larger values. Sure, 10% is fine (except for resonators... none in this circuit).

    Sure, for the big inductor on the woofer, you can always go for one with lower DC resistance (the "ohms").

    BMS driver would almost certainly need a different crossover to be optimized (though of course it would still play, just not as well as it might with a better suited crossover).

    The crossover is electrically 2nd order both ways, with CD compensation from both C2 and L2 (which also do some other shaping). If the sound seems to bright (it might in a hard room, though using a subwoofer helps that oddly enough), you can increase the value of R2 to 15 or even 18 ohms to bring the treble down. The crossover as-is is for approximately flat at 22deg off axis.

    Building? That's a lot more work (and parts shopping) than just designing with PCD! As far as measured results, PCD doesn't lie (assuming data input to it is correct!). To "voice" the speaker to taste, play with R2 if needed, but one size may not please all.

    BTW, it was a lot easier to get the TD12M to combine smoothly with a SEOS12 than it was for a TD15M with a SEOS15. The TD15M gets a little squirrely a little below 1kHz, the TD12's were smoother there. But the 15's will get louder, I bet.
  14. Thanks tux, hadn't considered that. For R2, I'd use two 24 ohm and for R3, two 20 ohm? Would I use the same tolerance and power handling as the 10 and 12 ohm resistors?

    smokarz, the woofer is in 1 cu.ft., slot port tuned to 50Hz. F3@~80hz. I had to get this thing as small as possible to fit the space. Can't wait to see your build, it was your sub build that inspired me to build my sealed mfw-15's. Still don't have them painted either! Lol.

    Thanks GranteedEV. That's exactly why I got the BMS and that Autotech used them in measurements with the fiberglass seos12. They have a bit more extension, but I'd probably never hear it anyway! lol. Perhaps I'll keep them for use in the surrounds I'm working on, seos10/td10m :)

    Thanks Bill. My brain was going for the values to achieve the same in parallel, it's just a matter of what's in stock, and trying not to deviate too far from your schematic.
  15. XO freq?

  16. About 1200Hz for this one.
  17. Great, thanks
  18. All xo parts ordered, and yes, 3 DNA-360's :) You guys in OH do not mess around...I ordered this stuff yesterday (late morning) and everything will be here in KS this Friday. So, big thanks to Erich for shipping the cd's so quickly and to PE as well. See, I knew I'd be making sound this weekend. ;)
    For the common, I was going to strip some copper from some 12/2 Romex. Good there?
    Also, will the drivers' magnet have any effect on the xo components?
  19. You don't necessarily have to use double the value. If you're good at math: 1/R1 + 1/R2 = 1/R3. So for instance a 15ohm and a 30ohm in parallel would net 10ohm. 1/15 + 1/30 = 0.06666 + 0.03333 = 0.1 = 1/R3 ===> R3 = 10.
  20. Thanks again! I found the resistors I needed and the L3 inductor at Meniscus. Can't wait to lay it all out and prep the board...

  21. For what it's worth, that's what I did for my build. I found that it was easier to hold off hot gluing it down until after you wrapped/soldered the wires.


  22. Speaking of building these, I wonder how Erich is doing on XO boards.
  23. That would be cool if he was doing the xo. I would go for that
  24. Amazing. <---Period. I am beside myself, Bill. Like you, I'm doing all this from my phone, otherwise I would post some pics. I'm just floored by the clarity, the result of your hard work no less.

    Building the crossover was incredibly simple thanks to your schematic. Hooked up the drivers and they measured at 7.6 ohms.
  25. Glad it worked out robotbunny. The clarity is from the SEOS and the drivers more than anything else. Electrostats don't have anything over those in the clarity department. Did you have to bring the treble down or was it ok as in the schematic?

Share This Page