Announcement

Collapse
No announcement yet.

Seas/Anarchy MTM Xover Critique Please

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

  • Seas/Anarchy MTM Xover Critique Please

    I bought some Anarchys back when they originally came out along with Seas 27TDFC. They've been in my basement for 10yrs now and finally deciding to get around to building something with them. This is my first time designing a passive xover. Any and all input on my design is appreciated.

    Building a floor standing MTM for mixed purpose HT/music duties. Not much "critical listening", general purpose enjoyment for myself and family but would like to build it right the first time.

    Images of the system response, reverse null, impedance and tweeter and woofer schematics are attached.

    Thank you in advance

  • #2
    Here is the driver response as well if that helps

    Comment


    • #3
      You might end up needing a small dip in the summed response right at the crossover to make them sound less "forward". I'm talking 1dB, maybe 2dB at most. Usually achieved by simply pulling the crossover points slightly apart. Not everybody hears it, or cares, especially if you're not concentrating and doing critical listening. But if it sounds a bit "in your face" that usually fixes things. Take a look at what Jeff did here:
      http://techtalk.parts-express.com/fo...180-mtm-design
      Francis

      Comment


      • #4
        Originally posted by fpitas View Post
        You might end up needing a small dip in the summed response right at the crossover to make them sound less "forward". I'm talking 1dB, maybe 2dB at most. Usually achieved by simply pulling the crossover points slightly apart. Not everybody hears it, or cares, especially if you're not concentrating and doing critical listening. But if it sounds a bit "in your face" that usually fixes things. Take a look at what Jeff did here:
        http://techtalk.parts-express.com/fo...180-mtm-design
        Thank you. I actually had to pull the xover points apart to get it that smooth, otherwise there was a 3db bump at the xover point. The tweeter is at 1500 and the woofers are 1300, going from memory.

        Comment


        • #5
          Originally posted by Impious View Post

          Thank you. I actually had to pull the xover points apart to get it that smooth, otherwise there was a 3db bump at the xover point. The tweeter is at 1500 and the woofers are 1300, going from memory.
          It's a track-by-track thing whether you'll hear it. On some tracks the forwardness is fine. On others, not so much...
          Francis

          Comment


          • #6
            Originally posted by fpitas View Post

            It's a track-by-track thing whether you'll hear it. On some tracks the forwardness is fine. On others, not so much...
            I notice Jeff says likes the -3db in the power response at the crossover point to keep them from sounding forward. That is approximately where mine end up. So I may be okay there

            Comment


            • #7
              Originally posted by Impious View Post

              I notice Jeff says likes the -3db in the power response at the crossover point to keep them from sounding forward. That is approximately where mine end up. So I may be okay there
              You just have to listen and decide. He made a comment elsewhere that he sometimes puts an actual response dip there, too. I've had the same experience with my MTMs. But nothing beats listening for yourself.
              Francis

              Comment


              • #8
                Originally posted by Impious View Post
                Here is the driver response as well if that helps
                Did you use frd files measured on a speaker baffle? If not, then expect everything to change substantially, unless you also used software to simulate the effect of the baffle.

                Will these speakers be close to the wall? Depending on location, you may need more, or less baffle step compensation.

                Comment


                • #9
                  Originally posted by rpb View Post

                  Did you use frd files measured on a speaker baffle? If not, then expect everything to change substantially, unless you also used software to simulate the effect of the baffle.

                  Will these speakers be close to the wall? Depending on location, you may need more, or less baffle step compensation.
                  I used SPL Trace to pull the FR and Z curves and WinPCD to design the passive. I would've imagined the baffle step could be simulated in the program but couldn't find a way to do it.

                  Comment


                  • #10
                    Care to post your F/Z curves for a look-see?

                    Comment


                    • #11
                      Originally posted by Chris Roemer View Post
                      Care to post your F/Z curves for a look-see?
                      Attached are the raw FR/Z curves for the drivers.
                      Attached Files

                      Comment


                      • #12
                        That explains some things, you're not using a box model or applying baffle effects to the files before simulating. That crossover will change drastically when you do that and without it, it won't match reality when you build it. Are you going to port the tower? Those Anarchy's will easily reach super low in a ported box, so I was suspicious looking at that slow roll off curve below 100Hz.

                        There are tons of resources for this, but I run through using Bagby's Response Modeler in This Post on my site to apply both a ported box adjustment and baffle effects during the sim process.
                        Electronics engineer, woofer enthusiast, and musician.
                        Wogg Music
                        Published projects: PPA100 Bass Guitar Amp, ISO El-Cheapo Sub, Indy 8 2.1 powered sub, MicroSat, SuperNova Minimus

                        Comment


                        • #13
                          Originally posted by Impious View Post

                          I used SPL Trace to pull the FR and Z curves and WinPCD to design the passive. I would've imagined the baffle step could be simulated in the program but couldn't find a way to do it.
                          Try this. Pad the tweeter another 5dB. Then work with the woofer filter to start sloping the response downward, starting at about 300hz, and meeting the tweeter at about 1k.. (Straight line with a 5dB slope from 300hz to 1k on the woofer.) That will put you closer than you are now. If the speaker will be real close to the wall, then target a 3dB slope.

                          Comment


                          • #14
                            Originally posted by wogg View Post
                            That explains some things, you're not using a box model or applying baffle effects to the files before simulating. That crossover will change drastically when you do that and without it, it won't match reality when you build it. Are you going to port the tower? Those Anarchy's will easily reach super low in a ported box, so I was suspicious looking at that slow roll off curve below 100Hz.

                            There are tons of resources for this, but I run through using Bagby's Response Modeler in This Post on my site to apply both a ported box adjustment and baffle effects during the sim process.
                            Thank you. I want sure how to account for baffle step without being able to measure it myself. I'll try to run the response modeler and see how it goes!

                            And yes, I did plan on porting them.

                            Comment


                            • #15
                              Originally posted by rpb View Post

                              Try this. Pad the tweeter another 5dB. Then work with the woofer filter to start sloping the response downward, starting at about 300hz, and meeting the tweeter at about 1k.. (Straight line with a 5dB slope from 300hz to 1k on the woofer.) That will put you closer than you are now. If the speaker will be real close to the wall, then target a 3dB slope.
                              Thanks for the info. If I cant get the response modeler to work (old PC, so for example Jeff B's PCD wouldn't work properly) I'll try to use this as a guide to shape the response instead

                              Comment

                              Working...
                              X