Announcement

Collapse
No announcement yet.

Crossover simulation

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

  • #16
    Here's my "stab" (Fc = 2k, 83dB +5dBSC)
    2nd order HP - 10uF series cap, 0.60mH shunt coil, L-pad: SR=2 / PR=3
    2nd order LP - 2.0mH low DCR series coil, 8uF shunt cap, Zobel = 10ohms+6uF
    or just skip the Zobel and bump up the coil to 2.5mH

    Comment


    • #17
      Thanks Chris. THAT looks more like I would have expected. The original crossover I found was third order and you just don't see any 3rd order crossover in a two way configuration.

      Comment


      • #18
        Here you go. Ignore the "peak" in the 850122 response - it's an artefact of the splice I did to combine port and nearfield bass response. Not relevant to phase tracking here.

        The only issue is that is as low as I'd take the XT25. Depending on your volume needs (the 850122 can take a pounding - I'm running it against a 3 way and it can almost hold its own), you may need to go steeper order on the XT25. I tried 3rd and 4th order electrical but couldn't get good phase tracking with the limited time I had

        Comment


        • #19
          PS: The bass circuit is what I am listening too with the Seas tweeter (i.e. 2.5mH inductor) and has the right amount of BSC to my ears, in my room.

          Comment


          • #20
            Thanks Dave for all your help. I feel confident going forward with your advice as well as Chris Roemer's, and rpb. This is where the heavyweights hang out.

            Comment


            • #21
              Originally posted by B King View Post
              Thanks Dave for all your help. I feel confident going forward with your advice as well as Chris Roemer's, and rpb. This is where the heavyweights hang out.
              No problem. I'm no heavyweight but have a method that gets me very close to measured results so reasonably confident. Whether or not those results are are preferred is really down to the owner / listener. A tapered FR curve, BBC dip, driver distortion issues all change things. As I said my only reservation above is how low the XT25 is crossing and energy sub 1KHz given the tweeters distortion profile.

              Comment


              • #22
                Originally posted by Dave Bullet View Post
                Here you go. Ignore the "peak" in the 850122 response - it's an artefact of the splice I did to combine port and nearfield bass response. Not relevant to phase tracking here.

                The only issue is that is as low as I'd take the XT25. Depending on your volume needs (the 850122 can take a pounding - I'm running it against a 3 way and it can almost hold its own), you may need to go steeper order on the XT25. I tried 3rd and 4th order electrical but couldn't get good phase tracking with the limited time I had
                If you have the sim still, can you try the filter I posted in post #9, and tweak it for a 2.5k x-over point? Maybe 2.5mH instead of 3mH.

                Or, maybe post the frd, and zma files.

                Comment


                • #23
                  Here's the post #9 sim. Note: without a resistor in the series notch - you get a high Q - hence the dip in response. A resistor would lessen the Q - but I don't think the 850122 benefits from a upper midrange notch anyway. A cap/resistor combo provides the required shape to the lowpass

                  Comment


                  • #24
                    Here's my sim for XO in post #16

                    Comment


                    • #25
                      Originally posted by Dave Bullet View Post
                      Here's the post #9 sim. Note: without a resistor in the series notch - you get a high Q - hence the dip in response. A resistor would lessen the Q - but I don't think the 850122 benefits from a upper midrange notch anyway. A cap/resistor combo provides the required shape to the lowpass
                      Try adding a couple ohms to the notch, and making the big coil smaller. See if you can push the x-over up to 2.5k, or 3k.

                      The filter was suggested in another thread by Daryl. I used it here because it was the same, or similar software that you guys are using. It showed a deep notch. I assembled the filter, and measured the response. The notch for some reason was not as deep as predicted. Playing with the resistance in the notch can be useful to shape the roll-off between 2k, and 4k..

                      Comment


                      • #26
                        Originally posted by rpb View Post

                        Try adding a couple ohms to the notch, and making the big coil smaller. See if you can push the x-over up to 2.5k, or 3k.

                        The filter was suggested in another thread by Daryl. I used it here because it was the same, or similar software that you guys are using. It showed a deep notch. I assembled the filter, and measured the response. The notch for some reason was not as deep as predicted. Playing with the resistance in the notch can be useful to shape the roll-off between 2k, and 4k..
                        I can push the xo point up as you suggest, but this doesn't help the XT25 if that is a concern. A notch isn't really required to get a reasonable LR4 acoustic target given the reasonable top end response of the 850122

                        Comment


                        • #27
                          Originally posted by Dave Bullet View Post

                          I can push the xo point up as you suggest, but this doesn't help the XT25 if that is a concern. A notch isn't really required to get a reasonable LR4 acoustic target given the reasonable top end response of the 850122
                          I'd like to play with this using PCD. Can I borrow your woofer files?

                          Comment


                          • #28
                            Certainly!

                            The first file - blended.... is an FRD

                            the 2nd - impedance... is a ZIP file
                            Attached Files

                            Comment


                            • #29
                              Here's a woofer filter for ya. Might need more BSC. I was going strictly by the measurements. Looks like Dave likes more BSC than this in his room. I thought there was some concern about how low the tweeter could cross. Maybe I misunderstood. Anyway, this filter is what I was trying to get you to try. I couldn't open your zma file so I used mine. I shaved off a couple dB around 800hz on your frd, because it made my graph look ugly. This is a simple filter, and it matches the target very well.

                              Click image for larger version  Name:	850122 roughed out filter.jpg Views:	1 Size:	525.4 KB ID:	1406659

                              Comment


                              • #30
                                Here's one more, with more BSC, and a 2.6k x-over. I smoothed up your frd a little bit more.
                                850122rev.frd

                                Click image for larger version

Name:	850122 better filter.jpg
Views:	1
Size:	523.9 KB
ID:	1406662
                                Attached Files

                                Comment

                                Working...
                                X