No announcement yet.

2-way crossover design issues

  • Filter
  • Time
  • Show
Clear All
new posts

  • 2-way crossover design issues

    I recently designed a 2-way crossover for a Dayton Audio DC300-8 and Dayton Audio RST28F-4. I used VituixCAD to design the crossover and all seemed good within the software. However upon building the crossover and attaching the drivers, low frequencies are still getting through to the tweeter side. I know that I should probably have this in a 3-way but, this is what I'm working with. Any help would be appreciated.
    Click image for larger version

Name:	DC300+T_crossover.PNG
Views:	48
Size:	186.7 KB
ID:	1420915

  • #2
    I haven't used either of those drivers, but here's my 2 cents.
    If the .61 ohm resistor is the resistance of the wire or coil in your model, that's OK. If it is a physical resistor, ditch it.
    I'd go 2nd order crossover on the tweeter at minimum. Just a guess, try a .30-.35 mH coil to ground after the 10 uF cap.
    That breakup node on the woofer is going to give you trouble. Might take a few iterations to tame it.

    Wait! Is that really a .22uF cap to ground for the woofer?
    Probably want to use a 8-10 uF cap there.

    I think I hear a difference - wow, it's amazing!" Ethan Winer: audio myths
    "As God is my witness I'll never be without a good pair of speakers!" Scarlett O'Hara

    High value, high quality RS150/TB28-537SH bookshelf - TARGAS NLA!
    SB13/Vifa BC25SC06 MTM DCR Galeons-SB13-MTM
    My Voxel min sub Yet-another-Voxel-build

    Tangband W6-sub


    • #3
      Thank you for your suggestions.


      • #4
        Been working on some revisions and have a large peak at 1800Hz but, I believe it is greatly improved.
        Click image for larger version

Name:	DC300+T_crossover_1.PNG
Views:	20
Size:	169.6 KB
ID:	1420938


        • #5
          Try inserting a resistor in series with the woofer's shunt capacitor. It will soften (damp) the knee of the filter and reduce that large peak.

          The lowest possible F3 box alignment is not always the best alignment.

          Designing and building speaker projects are like playing with adult Lego Blocks for me.


          • #6
            What Z offset are you using for the DC300? Are you using the raw Dayton supplied FRD/ZMA files or have you applied your own baffle diffraction and step to these?


            • #7
              I am using the Dayton FRD/ZMA files.


              • #8
                Here's an attempt. I've applied 3dB BSC to the woofer to being net system sensitivity to 87dB. There's some phase mis-alignment just above the crossover for the tweeter, but it might help you get closer. Impedance is a healthy 6 ohm minimum. Note the deliberate reverse polarity of the tweeter. Good luck!

                You'll probably need another series network (in parallel) with the woofer to tame that 2K bump. But this should make for a nice relaxing sounding speaker.
                Attached Files


                • #9
                  Then your simulation will not come out as it will in reality. They need to be adapted to the cabinets you are using, and the offsets are important for phase alignment.

                  One more thing- a 12" dome-assisted 2-way is not really ever a good idea. Prosound, maybe, but not a conventional home style. The blending and lobing and limited bandwidth will never really align the best with a woofer that is mainly bandwidth limited to about 800 Hz. This is the reason why 12" 2-ways are few and far between of this type.

                  "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:

                  My blog/writeups/thoughts here at PE:


                  • #10
                    As Wolf says - this isn't really doable, off-axis is going to be an issue. The low end response does not include the box - so no indication of frequency extension. I made an arbitrary call on baffle size / placement which will affect step and ripple response.

                    Essentially - the Z offset is so large (4.5" by my rough estimation) for this 12 inch woofer on the tweeter listening axis, when you move off-axis the phase relationship is going to change drastically, meaning all sorts of funky peaks and dips in the response. A good on-axis response doesn't mean a good power response - especially when working with such divergent driver offsets.

                    If you have already bought the drivers... then I suppose you have nothing to lose! Apart from XO parts money... still you can add them to your spare parts bin for the next project!


                    • #11
                      Try this. First order filter. (ie, a coil), combined with a notch centered on the peak. my WAG would be to start with a 1mH coil for the first order filter. For the notch (lcr), start with .5mH, 10uf, and 3 ohms. These are in series with each other, and across the woofer terminals. Your sim will show a pronounced dip somewhere from the notch. Adjust the cap value to move the notch where it is centered with the peak. Then increase the resistor value, until the peak is level with 1k. If you need a wider notch, make the coil smaller, and repeat the process. When it starts looking flat out to about 2k, add an additional cap to make the filter a second order with a notch. You don't want the roll-off very steep on the woofer, so you might need to add a resistor along with the cap. Like maybe 10uf, and 8 ohms. In fact, you might want to start out with this "impedance compensation" early on. Shoot for a 1.5k x-over point, if the tweeter can handle it. My computer is not available, or I'd sim this for you.

                      Let me revise the suggested values. I connected some xo parts to my woofer, and did a little trial and error.

                      2mH low-pass with a notch made from a 2mH coil, and a 4uf cap. Might need a resistor as well, but I didn't use one here.

                      Click image for larger version

Name:	dc300 wag.JPG
Views:	28
Size:	435.2 KB
ID:	1420962
                      Last edited by rpb; 09-05-2019, 08:26 PM.


                      • #12
                        Dude, your tweeter is SCREAMin' ! (needs about -10dB of attenuation)
                        (circuit drawn L>R = amp>driver)
                        2nd order HighPass (on tweeter) w/an "L-pad": use a 16uF series cap, then a 0.50mH shunt coil (to ground), then an "L" next to the tweeter: SeriesResistor = 4n(ohms) / PR = 2n, reverse polarity is indicated on the tweeter
                        woofer needs 6 parts: 1st a 2.0mH series coil (use an iron core, about 0.2ohms DCR), ACROSS THIS coil (in parallel w/it) run a 4n resistor inline w/a (small) 1.5uF cap, after that run a 10uF cap to ground (so this will be in parallel w/the woofer), also in parallel w/the woofer will be a Zobel, an 8ohm resistor in series w/a 50uF cap (use a cheap npe here)

                        Your woofer peak is gone. You have +6dB of baffle-step compensation (so pull these out a couple feet from the wall, on stands to get the tweeters at "ear height". These run about 82dB and cross near 1.4kHz.

                        Hope you ported those (for 30Hz bass), else 60 is all they'll do (w/out room gain).


                        • #13
                          Well you've got lots of options to try.
                          not sure I agree a 2.0mH could gives full 6dB bafflestep comp but maybe the northern hemisphere is different


                          • #14
                            at least +6dB
                            Click image for larger version

Name:	12in2way.jpg
Views:	22
Size:	190.3 KB
ID:	1421032


                            • #15
                              Odd - Is your XO causing a mid-bass hump or assuming IEC baffle? Looks like it is too sensitive at 100Hz