Announcement

Collapse
No announcement yet.

Is it possible to implement an elliptic filter using op-amps?

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • Is it possible to implement an elliptic filter using op-amps?

    Elliptic filters seem to be a very nice option for avoiding resonances in metal-coned drivers. However, while implementing them passively is fairly simple, I'm a bit fuzzy on how one might go about implementing an active elliptic filter.

  • #2
    Re: Is it possible to implement an elliptic filter using op-amps?

    Use science

    But seriously folks, just use a 6th or 8th order slope if you're going to go that route. That's basically what elliptic filters are trying to emulate.

    Although, it's not really needed. 4th order implemented correctly will work just fine in almost any case.
    I am trolling you.

    Comment


    • #3
      Re: Is it possible to implement an elliptic filter using op-amps?

      Originally posted by MSaturn View Post
      Use science

      But seriously folks, just use a 6th or 8th order slope if you're going to go that route. That's basically what elliptic filters are trying to emulate.

      Although, it's not really needed. 4th order implemented correctly will work just fine in almost any case.
      From what I understand, the main value of the Cauer filter is that while the output is somewhat irregular, it's a very efficient way to put an extremely steep secondary filter to iron out cone resonances.

      Comment


      • #4
        Re: Is it possible to implement an elliptic filter using op-amps?

        A Cauer lowpass is basically a standard lowpass plus a notch filter.
        Dennis

        Comment


        • #5
          Re: Is it possible to implement an elliptic filter using op-amps?

          He's correct, that's why output rises again after the Fc of the tank filter.

          Like I said, just use a high order filter if you're using op-amps. There's no reason to use a cauer filter at all.
          I am trolling you.

          Comment


          • #6
            Re: Is it possible to implement an elliptic filter using op-amps?

            Originally posted by MSaturn View Post
            But seriously folks, just use a 6th or 8th order slope
            The biggest disadvantage of that is the aditional phase rotation across the filter. There is also greater LF energy imposed on the tweeter, as the knee of the HP filter is raised.

            Originally posted by Dennis H View Post
            A Cauer lowpass is basically a standard lowpass plus a notch filter.
            As a practical matter that is correct, although the stop-band ripple is a bit different, and it points to the easiest implementation. One gets the low phase rotation of the underlying "standard" filter, plus the steep stopband rejection contributed by the notch . . . particularly effective if the notch can be centered on a (the first) cone breakup. It's really easy to model (using, for example, the "active filter" side of PCD), and to build. It also makes it easy to properly match a quasi-eliptic low-pass to a corresponding "standard" high-pass.
            "It suggests that there is something that is happening in the real system that is not quite captured in the models."

            Comment


            • #7
              Re: Is it possible to implement an elliptic filter using op-amps?

              If it's an second-order filter with the tank on the coil.

              If it's any higher order, and you have an inductor also in the series line, it won't rise above the Fc of the notch, or at least not as strongly.

              If you are using active filters, you may still be required to use a passive notch at the driver for material resonances, cauer-elliptic or not. Yes the slope will rolloff the driver, but it might not kill the resonances strongly enough depending on the location of the xover.

              Later,
              Wolf
              "Wolf, you shall now be known as "King of the Zip ties." -Pete00t
              "Wolf and speakers equivalent to Picasso and 'Blue'" -dantheman
              "He is a true ambassador for this forum and speaker DIY in general." -Ed Froste
              "We're all in this together, so keep your stick on the ice!" - Red Green aka Steve Smith

              *InDIYana event website*

              Photobucket pages:
              https://app.photobucket.com/u/wolf_teeth_speaker

              My blog/writeups/thoughts here at PE:
              http://techtalk.parts-express.com/blog.php?u=4102

              Comment


              • #8
                Re: Is it possible to implement an elliptic filter using op-amps?

                Lancaster's Active Filter Cookbook has a section of a chapter that gives some detail on how to build one using op-amps. So, yes, it can be done.

                Comment


                • #9
                  Re: Is it possible to implement an elliptic filter using op-amps?

                  YES. You have to implement the circuit by simulating the inductors using a generalized impedance converter (GIC). Then you can just scale the component values of the active version like you would a passive filter.

                  Here is a web page about efforts in this direction:
                  http://ldsg.snippets.org/FILTERS/Cuadra/elliptic.php (but didn't use a GIC)

                  If you can open these pages, it gives details about GICs and their use for inductor simulation:
                  http://www.ieee.li/pdf/viewgraphs/fi...igurations.pdf
                  http://www.springerlink.com/content/...1/fulltext.pdf

                  This book:
                  Electronic Filter Design Handbook, Fourth Edition (McGraw-Hill Handbooks),
                  Arthur Williams (Author). See page 128 for an excellent section on this topic. I got it out of my local University library. It used to be available online for free, but it looks like it was pulled and now only the contents are posted.

                  -Charlie
                  Charlie's Audio Pages: http://audio.claub.net

                  Comment


                  • #10
                    Re: Is it possible to implement an elliptic filter using op-amps?

                    Here you go, found a book preview covering the GIC simulation of active inductors by Walt Jung.

                    http://books.google.com/books?id=dun...page&q&f=false

                    -Charlie
                    Charlie's Audio Pages: http://audio.claub.net

                    Comment


                    • #11
                      Re: Is it possible to implement an elliptic filter using op-amps?

                      Yes,

                      Jens Rasmussen & Bob Ellis did a write-up on such a thing over on diyAudio.

                      It is included in the Active Filter 4 Manual.

                      Shoot me message with your e-mail and I'll copy you.
                      Mongo only pawn in game of life
                      ____
                      Ed

                      Comment


                      • #12
                        Re: Is it possible to implement an elliptic filter using op-amps?

                        Originally posted by Deward Hastings View Post
                        The biggest disadvantage of that is the aditional phase rotation across the filter. There is also greater LF energy imposed on the tweeter, as the knee of the HP filter is raised.
                        Not necessarily Deward. Nothing says you have to have a knee region that is any more challenging than LR4. The shape of the transfer function near the XO region is not set in stone with elliptic filters of any order.
                        R = h/(2*pi*m*c) and don't you forget it! || Periodic Table as redrawn by Marshall Freerks and Ignatius Schumacher || King Crimson Radio
                        Byzantium Project & Build Thread || MiniByzy Build Thread || 3 x Peerless 850439 HDS 3-way || 8" 2-way - RS28A/B&C8BG51

                        95% of Climate Models Agree: The Observations Must be Wrong
                        "Gravitational systems are the ashes of prior electrical systems.". - Hannes Alfven, Nobel Laureate, Plasma physicist.

                        Comment


                        • #13
                          Re: Is it possible to implement an elliptic filter using op-amps?

                          Not necessarily Deward. Nothing says you have to have a knee region that is any more challenging than LR4.
                          You missread either my comment or the one I was referring to (which suggests using 6th or 8th order instead of eliptical). 6th or 8th order typically raise the knee . . . a potential disadvantage of that approach in addition to the greater phase rotation.
                          "It suggests that there is something that is happening in the real system that is not quite captured in the models."

                          Comment


                          • #14
                            Re: Is it possible to implement an elliptic filter using op-amps?

                            As far as I'm concerned (and aware) extra phase rotation is inaudible so long as the drivers still mesh with one another as far as that goes. Remember that off-axis response is audible, and any normal LR4 filter (or even LR2) is going to start rotating like crazy 20 degrees off axis or so.

                            Besides that, the knee of the filter could be modified easily with a small bit of EQ. say -1 or 2 dB near the apex of the curve, at a corresponding Q to the filter's.
                            I am trolling you.

                            Comment


                            • #15
                              Re: Is it possible to implement an elliptic filter using op-amps?

                              I'm afraid that I'm somewhat lost in all of this - my comprehension of filters is basically limited to "woofer turns off, tweeter turns on, keep everything in phase with minimal time delay."

                              I guess my thoughts are as follows:

                              1. Higher-order crossovers produce group delay. This is undesirable, as you get weird time smearing effects between the HP and LP sections.
                              2. Cone resonance on metal drivers causes two things: First, a a harmonic of a given frequency (approx. 2.2khz according to Wolf) which sounds horrible, and second, a huge impedance peak which make conventional crossovers go berserk.
                              3. A secondary filter (such as a Cauer filter) above the crossover point can reduce driver output at the resonance point to the point where you can't hear it. By placing the filter above the crossover point, any time smearing would be at frequencies with amplitudes low enough you couldn't hear it.
                              4. An active crossover has an effectively zero output impedance, meaning that any increase in impedance due to resonance is effectively moot.

                              That Jason Caudra design looks pretty darn close to what I want, though I'm a bit fuzzy on some of the math. If I have this right, adjusting the capacitor values for a 1.5khz crossover point will have the woofer amplitude down 30dB at 2.1 khz, which is effectively a 12th-order crossover - hopefully enough to keep the resonance from being an issue. However, the amplitude of the output is attenuated no further. Would adding a second -12dB/octave filter at about 2.1khz eliminate this problem?

                              Comment

                              Working...
                              X