Anyone ever built an active crossover based on Rod Elliot's designs?

Discussion in 'Home Theater Projects' started by RichardHOS, Apr 29, 2003.

  1. RichardHOS

    RichardHOS Second Unit

    Joined:
    Mar 11, 2003
    Messages:
    454
    Likes Received:
    0
    If so, what did you think of it? Easy to build? Opinion on noise, distortion specs., etc.? Sound good?

    Did you add output gain adjustments?

    Roughly what was the finished cost per channel?
     
  2. Ellen

    Ellen Stunt Coordinator

    Joined:
    Jan 8, 2000
    Messages:
    142
    Likes Received:
    0
    Real Name:
    Ellen
    I built Rod's Project 09 24 dB/oct LR crossover. Stuffing the board was easy as pie. It's been a long while since I built it so I don't remember the costs.
     
  3. RichardHOS

    RichardHOS Second Unit

    Joined:
    Mar 11, 2003
    Messages:
    454
    Likes Received:
    0
    What did you think about the sound? Did you notice any extra noise or distortion after inserting the crossover into the signal chain?
     
  4. Jonathan M

    Jonathan M Second Unit

    Joined:
    Jul 23, 2002
    Messages:
    267
    Likes Received:
    0
    Richard,

    I've built his design. It works well - no noticeable noise or distortion introduced, and I used just vanilla TL072's as the opamps (All except the output buffers are running unitary gain, so I don't think this is an issue).

    Really easy to put together, and it's a very flexible board. I used to use it to xover to my sub, so it was configured as 24dB/octave LR at 80Hz. I used the output buffers (there's room on the board) and altered them slightly for a larger gain range - only the lowpass had the gain, the highpass was buffered with no gain. Cost of parts here in New Zealand were around US$20 or so. I used bi-polar electros for the input and output caps - these can be poly based if you feel like it. As I said, I could not audibly tell the difference. No increase in noise that I could discern.

    Pics can be seen on my diy sub site:

    http://www.massey.ac.nz/~jcmarsha/hometheatre (Click through to the sub site)

    Hope this helps.
     
  5. Ellen

    Ellen Stunt Coordinator

    Joined:
    Jan 8, 2000
    Messages:
    142
    Likes Received:
    0
    Real Name:
    Ellen
    I should have said in my prior post that I could detect no noise introduced by the crossover. Mine is configured at around 85Hz. I stuffed the board with TL072 opamps as Jonathan did. I had done so for testing, planning to swap out with something better but never felt the need to do so.
     
  6. RichardHOS

    RichardHOS Second Unit

    Joined:
    Mar 11, 2003
    Messages:
    454
    Likes Received:
    0
    Cool! From what I can gather, there's something about the "state variable" designs used in the prosound gear I've looked at that introduces the self-noise. Rod E. mentioned on his site that making a L-R circuit 'variable' would be a pain in the ass, so "state variable" must be some trickery done (I'm guessing at the op-amp stage?) that allows great flexibility in frequency selection... at the cost of noise and (possibly) distortion.

    $20? That's about what I had guessed. With the $16/channel PCB (3-way... and I think the PCB would be worth it to keep me from screwing up the circuit [​IMG] ) that's around $40 per channel after shipping. Depending on how "good" I want it all to look, it might wind up at $50 or $60 per channel after power supply filtering and chassis.


    A couple more simple questions for you guys...

    - Does the PCB offered by Rod include traces for output gain adjustment section?
    - What provisions are on the PCB for inputs and outputs? Are the traces layed out for a standard board mount phono plug? Just solder points for you to wire in short sections of interconnect cabling?

    Hard to tell from the small pic on his site. Thanks.
     
  7. Jonathan M

    Jonathan M Second Unit

    Joined:
    Jul 23, 2002
    Messages:
    267
    Likes Received:
    0
    Yes, the PCB contains all the necessary traces etc. for output gain and output buffers.

    The PCB contains just solder points for connections to input and output sockets - you need to use small lengths of shielded cable (Or twisted pairs or whatever) to hook these up.

    There's a pic of my finished one showing these connectors and the hook up of the pot for gain adjustment on the Electronics page of the link above.
     
  8. RichardHOS

    RichardHOS Second Unit

    Joined:
    Mar 11, 2003
    Messages:
    454
    Likes Received:
    0
    Ah, yeah, a separate board for input/output connections would work even better and allow more flexibility for mounting than what I was thinking.

    Looks great! I think I'll build a twelve-pack. [​IMG]
     
  9. Ellen

    Ellen Stunt Coordinator

    Joined:
    Jan 8, 2000
    Messages:
    142
    Likes Received:
    0
    Real Name:
    Ellen
    I strongly encourage you to wait for the boards to arrive before buying parts. Rod's boards are designed to be as small as possible. I bought a bunch of caps for the project before my boards arrived and they were too big to fit. Wait for the boards and check things out and then read spec sheets very carefully when ordering your parts, paying particular attention to lead spacing and footprint.
     
  10. Dave Milne

    Dave Milne Supporting Actor

    Joined:
    Jul 2, 2001
    Messages:
    568
    Likes Received:
    0
    Richard,
    I've built quite a number of active crossovers of my own design, but essentially the same topology as Rod's circuits. With quality parts (don't use electrolytics -or even mylars -in any signal path) they sound very good indeed. One of my old subwoofer crossovers is currently supplementing my Lexicon MC-12 so that I can have bass management in bypass (SACD) mode.

    I've always put them in some sort of small chassis so that I had a place to mount connectors, level control, LED to show power supply active etc. For amplifiers, I have used 5534, TL-072, LM-833, some fast video op-amps, AD797, etc. I really like the AD797, but they're singles... so they won't work with Rod's board.

    A nice part about doing your own layout is that you can put the frequency-setting resistors on a header that plugs into an IC socket. That way, you can change them easily without risking lifted pads. You can even make up a few extra headers for different frequencies to swap around.

    I've been thinking about laying out a new "state-of-the-art" subwoofer crossover using AD797s, nice poly or teflon caps, super-quiet regulated supply, etc. The finished product would feature a sexy machined chassis, WBT connectors, Alps "black beauty" level control(s), etc. I'll probably include capability for subsonic filter and Linkwitz Transform. I could consider provisioning for some extra stages to make it usable for multi-ways.

    Any interest out there? I can post the design and board CAD files or consider having extra bare boards built to pass on at my cost...

    Unfortunately with my current workload, this could be a couple of months out.
     
  11. RichardHOS

    RichardHOS Second Unit

    Joined:
    Mar 11, 2003
    Messages:
    454
    Likes Received:
    0
    If the boards were layed out so that 2-way and 3-way were options, then I'd certainly be interested. I won't need the finished XO's for six months or so still. In the meantime, I'll be playing with an inexpensive prosound XO that has very flexible XO frequencies and output gain adjustments to find the best points to use with my particular speakers and amplifiers.

    I won't be ordering any boards from Rod or elsewhere for quite a while still (a few months or so), so keep me informed on what you decide to do. Having a transform option for the low-pass sounds very appealing.
     

Share This Page