No announcement yet.

My foolish journey into making a DIY center channel speaker

  • Filter
  • Time
  • Show
Clear All
new posts

  • #16
    Anyone want to look at my .dxo before I start ordering parts??


    • #17
      Originally posted by jvphotog View Post
      Anyone want to look at my .dxo before I start ordering parts??
      Do you expect the sim to be accurate, or are you willing to "sneak up" on a good xo using some sim info, and some trial and error? Do you have any xo parts now? Do you plan to have a stash for future projects?

      Where exactly will the center be used? Up high, partially blocking the screen, or down low inside the equipment cabinet?


      • #18
        I'll kind of add on to what the others have said and suggest using just one mid that can cross over lower, unless you need 2 to keep up with woofers. Maybe this:

        Furthering what rpb asked above, a stash of cheap electrolytic caps is one way of trying out/tweaking various crossovers, and you can replace them with poly's or whatever after you finalize things. A bunch or resistors too.


        • #19
          Well, I was hoping it would at least be in the ball park! But I expect to play around it it - I've come this far.
          I ordered some extra resistor values, but I should have ordered some more cheap electrolytics. I used to have a TON, but unfortunately sold off the lot when I stopped building guitar amps. Had I known...
          I know the side by side drivers are a bad idea, but I was mirroring the diysg 1099 speaker. I already had the custom baffle cut and the speaker is made.


          • #20
            Side by side mids are fine at your listening distance.

            What does your mid filter look like? (Forget the rest. Work on getting a good mid response.)


            • #21
              Ive been busy! XO parts came finally came in, so I wired it up and took it outside for some tests. It measured pretty close to the simulation, with only about a 1.5db variation in two spots. AND no fires! So Im calling this a success for my first ever XO attempt.

              This is on axis at one meter with 2.8vac applied. All other tests are quieter as to not bother neighbors. First reflection is at about 8ms, so I gated it there. 1/24th smoothing applied, but it doesn't change much.
              Click image for larger version

Name:	jvc_xo1_2.83vac_actual.png
Views:	195
Size:	289.6 KB
ID:	1478586

              This is the horizontal off axis. Seems decent until round 35 degrees. What do you guys think? I think it looks pretty good for a MTM...
              Click image for larger version

Name:	Screen Shot 2021-11-11 at 8.29.26 AM.png
Views:	182
Size:	324.5 KB
ID:	1478587

              Vertical off axis was difficult for me to measure. I had a hard time figuring out the angle, so these might be more than listed. Measurements aren't great where the horn and mids meet. But at this angle it will be pointing up at the ceiling/floor very quickly. At my listening distance Im well within 10 degrees.
              Click image for larger version

Name:	Screen Shot 2021-11-11 at 8.29.50 AM.png
Views:	186
Size:	367.6 KB
ID:	1478588


              • #22
                Here is the XSim, overlaid with the actual measurements. Red is actual. There is about 1db dip in the 1.5khz range and the top end is about 1.5-2db hotter than simulated... and I don't know why. I can pull the XO components off and measure them with the DATS if I get bored... I did enter manufactured listed resistance for the inductors.

                Click image for larger version

Name:	sim_actual_xo1.JPG
Views:	202
Size:	385.3 KB
ID:	1478590

                Here is an actual measurement with the mid inverted overlaid with the simulation with the mids inverted:
                Click image for larger version

Name:	sim_actual_xo1_inverted.JPG
Views:	183
Size:	388.9 KB
ID:	1478591
                My filters
                Click image for larger version

Name:	sim_xo1_.JPG
Views:	188
Size:	836.5 KB
ID:	1478592
                Impedance measurements were pretty spot on. 4ohm speaker.
                Click image for larger version

Name:	65821086988__62690DCD-F548-4494-88D9-346A5369DCE6.jpeg
Views:	205
Size:	694.8 KB
ID:	1478593


                • #23
                  So what do you all think? What should I do next, if anything? Should I try lowering XO points? Any suggestions would be appreciated. Thanks!
                  Also, lifting this 92lb speaker into the air hasn't been easy!

                  Click image for larger version  Name:	IMG_6955.jpeg Views:	0 Size:	730.9 KB ID:	1478595 Click image for larger version  Name:	IMG_6937.jpeg Views:	0 Size:	760.1 KB ID:	1478596 Click image for larger version  Name:	65828463290__E34A9D16-DCB1-4BFD-862C-69C77A71CBF8.jpeg Views:	0 Size:	532.2 KB ID:	1478597


                  • #24
                    Attached Files


                    • #25
                      Looks perfect! Don't change anything unless you feel a certain range needs adjustment.


                      • #26
                        Tweeter level varies and is usually tweaked as necessary. Measurements look good! I'd give it some time and listen in all the various spots where people might listen to see if anything needs adjustment. Pretty good results for a first attempt, especially a 3 way


                        • #27
                          Thanks guys! I'm actually surprised it worked on the first try. I almost gave up after spending days in XSim to try and make things work. I had a 2-way baffle cut, just in case I couldn't get this to work.
                          It never really occurred to me that the horn and mids would have phase issues on the vertical axis... I think I spent so much time worrying about the MTM horizontal response that I didn't pay attention to it.
                          I HAD a better XO designed, with better phase alignment and response, but when I was done with it I checked the wattage dissipation on the resistors and it was really high. I got freaked out and resigned it all. But now that I have this one built and running, Ive checked it with a Laser Infrared Thermometer and it never goes above room temperature. So I think I would have been fine. Oh well, now I know.
                          Im not sure what to do at this point, besides listen listen listen


                          • #28
                            What wattage resistors did you use? The ones on the tweeter are usually on the safer side, but if there are any on the mids they can get hot. Play something really loud and check them. I've had doubled up ones get very hot and now will triple or quadruple them up, or I'll double up on higher wattage resistors. For the vertical axis you can flip the cabinet over so the tweeter and mids are inverted and see if thats any better...or just simulate it since the cabinet is so heavy


                            • #29
                              One of them might be a 20 watt but the others are your typical 10 watters.


                              • #30
                                I've too have noticed xsim doesn't accurately simulate the high frequency level and always shows it slightly lower in the simulation compared to how it actually measures. So it's not your crossover parts out of spec. I now try to figure this in and leave a gentle taper to the top end in the software if I want it to measure more flat.