New speakers inspired by karma-8

Discussion in 'General Topics' started by josephjcole, Mar 15, 2013.

  1. I just wanted to share these speakers that I'm building for a friend. EOS-8 waveguide, Denovo 150, and Faital Pro 8FE200. I've got them hooked up to a mini-DSP right now, tweaking the crossover. After I get it dialed in I'll build an analog crossover and send them off to their new home.

    I was inspired by the karma-8's, but my friend doesn't have a subwoofer so we wanted something with a little more bass. I only hooked them up last night so they still need some work in the crossover, but they show a lot of promise! Suprisingly large and dynamic sound for such a small speaker.


    I can most measurements, and a crossover later after I get them figured out if people are interested.


    Thanks to Erich for all the work and great products...
     

    Attached Files:

  2. Great looking speakers. What kind of wood is that and how was it built? Looks like all solid wood?


    Thanks for posting!
     
  3. It's solid walnut. The top, front, back and bottom are 'tongue and dado'. The sides are frame and panel construction to deal with expansion and contraction. There is some internal bracing, walnut, screwed in with pocket screws.


    I'm glad you like them.... I'm pretty proud really. I should get better pictures. They sound great, so far I've got a 4th order filter around 2khz with a little compensation at the top end for the Denovo-150.
     
  4. They look really nice, though to be honest the first thing I thought of when I looked at it was that the side resembled a cabinet door ;D

    The DNA-150 does really well in the EOS-8, I cooked up modified version of the Karma-8 design using the EOS-8 waveguide a few months ago and liked it even more then original. The passive highpass I came up with for the DNA-150 on the EOS-8 used only three parts. I should get that design posted up here.
     

  5. Yeah, I think it's hard not to look like a cabinet door when you use frame and panel construction. I've made small solid wood speakers before with out frame and panel construction but it seems like the bigger the speaker is the more likely you are to run into expansion issues if you don't deal with the expansion/contraction of wood.


    I'ld certainly love to see the high pass you've come up with for the [SIZE=78%]DNA-150 on the EOS-8. I'm on a mac, but I've set up a virtual machine, and am starting to wade into 'Passive Crossover Designer'. I've built crossover's before, but they were all pretty simple, and I haven't messed with them at all since I got myself a miniDSP. The Faital Pro has a pretty heavy rising response which will have to be dealt with.... lots to learn![/SIZE]
     
  6. MTG90


    I would still love to see the three component crossover you came up with, for your denvo150/EOS-8!
    I'm starting to get a handle on Passive Crossover Designer... took a while though as I'm measuring everything with Fuzzmeasure2 which outputs frd files, which don't load up properly in PCD. I've ended up running everything through SPL-trace.

    Anyways, I'm getting somewhere, but my crossover for for the Denvo is a second order with 2 parts, and then a LCR notch filter to deal with the impedence peak around 2k... I'ld love to see/hear how to do it with less parts.

    thanks so much
     
  7. He posted it elsewhere on the forum.

    Funny the frd won't load. Does fuzz measure output txt files? How are you getting phase with the SPL trace?
     
  8. Sorry Joseph I must have missed that last post of yours, here is the thread with the EOS-8 crossover if you have not already found it yourself: http://www.diysoundgroup.com/forum/index.php?topic=190.0

    I'll try to get my PCD transfer functions up tomorrow as well so you can see those too.

    Since your woofer is rear mounted it may not need as steep a filter to get the phase aligned as the 4th order I used on mine.
     
  9. Thanks mtg90! I had completely missed it...


    tuxedocivic- fuzzmeasure does put out a text frd file. How ever the in the second column (amplitude) it doesn't give the numbers PCD wants. Fuzzmeasure doesn't put it out in "real" db values. It's centered around 0 so there are some negative number where the amplitude is low. If there was a way to go through the file and say, add 50 to each number then I would be in business.


    Either way though it doesn't export zma files. I believe both of these issues have been cleared up in version three, unfortunately it would cost me $75 to upgrade from version 2 to version 3.... so I'm trying to find a work around.


    You're right though, both the frd files and the zma files just have zeros through the phase column. Any advice you might have would be greatly appreciated!
     

    Attached Files:

  10. You can use excel or response modeler to increase the SPL. I have this issue with HolmImpulse. It's a pain but once you do it a couple times its a snap. For phase, Jeff B has a phase extractor, but you'll have to use techniques to find the z off set.

    Alternatively you could try REW which should do impedance and fix a lot of your problems. Available for Mac and windows.
     
  11. Great! thanks so much, I had always heard about REW, but never looked into it figuring it was for windows only.... I'll give it a go. Thanks so much!
     

Share This Page