Announcement

Collapse
No announcement yet.

First swing @ a big passive 3-way. Input needed.

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

  • First swing @ a big passive 3-way. Input needed.

    This is my first swing at a big passive 3-way. Not sure about anything at this point so feel free to dive in if you have the inclination. I need the help. I have some experience but this is a pretty big bite for me. I already have some concerns.

    First the drivers:

    Woofer- Dayton RSS390HF-4
    Mid - Dayton RS150P-8A
    Tweet - Wavecor TW030WA12

    I want this project to be sealed.The 390 looks pretty good in 60-70 l sealed with an f3 of around 35hz and an f6 around.28hz It will take plenty of power and the curve looks ok to me up to about 400 hz where I will cross to the mid.

    Click image for larger version

Name:	390 TF.PNG
Views:	1
Size:	293.7 KB
ID:	1384183


    The 150 mid looks ok for power and excursion when crossed @ 400hz, however I have never heard this driver so there may be better choices. This is in a 2l sealed chamber.with a 2nd order high pass @400 hz

    Click image for larger version

Name:	RS150P8A EXC.PNG
Views:	1
Size:	319.3 KB
ID:	1384184


    Here is where I am getting out of my element. My first time doing a complex 3-way passive crossover and my first time using Xsim. There is a lot I don't know. My main concern with this is the low impedance of about 3.75 ohm in the middle. I was using a 4 ohm mid and tweet. I worked on those for about a week and I couldn't get the load up above 2.5 ohms in the pass band so I went to 8 ohms on them and I got it up to about 3.75. I thought of going to a different 8 ohm woofer but to be honest the 390 is pretty impressive in a smallish sealed box for a 15 and I didn't want to give up on it. But then I my be out to lunch on the crossover.

    Click image for larger version

Name:	X-OVER 6.PNG
Views:	1
Size:	480.7 KB
ID:	1384185

    I have concerns with the wavecor on the power. I wouldn't think there would be an issue with this tweeter with an LR 4 @ 2500 hz but maybe I am overlooking something. Like I said. I am not sure about anything at this point.

    Click image for larger version

Name:	POWER.PNG
Views:	1
Size:	474.3 KB
ID:	1384186

    Take a look at what I have so far. If you see anything I have out of wack please let me know. I have a box modeled in CAD and I can add that to this thread as I go.

    Matthew

    BSME
    Kannapolis NC

  • #2
    Other than the impedance looks pretty good. Have you loaded in the baffle response? Might want to guess at the offsets and look at the phase at the x-o, the slopes look asymmetrical. Hard to read but might consider bumping up L-3 and lowering C4. Then again the measured mid roll off might not be nearly that smooth where your sim work might not match reality.
    John H

    Synergy Horn, SLS-85, BMR-3L, Mini-TL, BR-2, Titan OB, B452, Udique, Vultus, Latus1, Seriatim, Aperivox,Pencil Tower

    Comment


    • #3
      Originally posted by jhollander View Post
      Other than the impedance looks pretty good. Have you loaded in the baffle response? Might want to guess at the offsets and look at the phase at the x-o, the slopes look asymmetrical. Hard to read but might consider bumping up L-3 and lowering C4. Then again the measured mid roll off might not be nearly that smooth where your sim work might not match reality.
      I have and it puts a pretty nasty dip roughly @ the 2500 crossover point. I did this with the new simX 3D (which has a baffle simulator kind of like EDGE) so I wasn't sure I should believe it or not. I wanted to have you guys check what I had before I move forward. I will post a few CAD drawings tomorrow. At this point everything is in rough state so I can make changes before building the test box.
      BSME
      Kannapolis NC

      Comment


      • #4
        When you get ready to order parts it might be a good idea to order caps/resistors a few sizes bigger/smaller than needed. If a coil is too large you can unwind it.
        http://techtalk.parts-express.com/fo...khanspires-but
        http://techtalk.parts-express.com/fo...pico-neo-build
        http://techtalk.parts-express.com/fo...ensation-build

        Comment


        • #5
          The only thing I see that looks "off" is how the summed response is below the woofer response around 200Hz. That's a phase issue, and with a 2nd order LP filter on the woofer, you might be better off targeting a 2nd order slope on the mid HP filter, which requires inverting the midrange phase relative to the woofer. Of course, you'd invert the tweeter too to follow the mid.
          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


          • #6
            I'm surprised by the limited padding that's going on with the tweeter. 0.5 series and 24 parallel. The TW030WA12 also has a rising response hump from 3-5k, and then droops back down with valley around 10k. I don't see these "anomolies" in the modeled response. I'm sure I'm missing something obvious.
            SEC DIY 2014 Speaker Show is on! *November 8th*
            Wanna win a set of Newform Research R30 ribbons?!?!
            SEC DIY 2014 Thread

            SEC DIY 2012 Speaker Show
            SEC DIY 2011 Speaker Show

            Comment


            • #7
              With your presumed power usage, I'm guessing you're going to be playing these pretty loud. Make sure you put adequate wattage on R6 pad in front of the mid. That's going to take some serious abuse. I would go with 3 or 4 10W film resistors in parallel to avoid stressing them long term.
              SEC DIY 2014 Speaker Show is on! *November 8th*
              Wanna win a set of Newform Research R30 ribbons?!?!
              SEC DIY 2014 Thread

              SEC DIY 2012 Speaker Show
              SEC DIY 2011 Speaker Show

              Comment


              • #8
                i love big speakers ! i'm not of much help but following for moral support .
                Paper Towers
                RS180P/28F surrounds
                Boombox

                Comment


                • #9
                  Originally posted by scottq View Post
                  I'm surprised by the limited padding that's going on with the tweeter. 0.5 series and 24 parallel. The TW030WA12 also has a rising response hump from 3-5k, and then droops back down with valley around 10k. I don't see these "anomolies" in the modeled response. I'm sure I'm missing something obvious.
                  Not sure why this is. Here is an image of the driver response included. I generated the frd and zma files for the Wavecor from measurement file plots. Not from thre published data of course but from measurement plots I got here: http://www.audioxpress.com/article/Test-Bench-Wavecor-TW030WA12-30-mm-Cloth-Waveguide-Loaded-Tweeter
                  I am not completely confident in the generated files. I would prefer to have the real files but I am having trouble finding them on line.

                  Click image for larger version

Name:	wavecore.PNG
Views:	1
Size:	419.2 KB
ID:	1384247
                  BSME
                  Kannapolis NC

                  Comment


                  • #10
                    Originally posted by Pete Schumacher View Post
                    The only thing I see that looks "off" is how the summed response is below the woofer response around 200Hz. That's a phase issue, and with a 2nd order LP filter on the woofer, you might be better off targeting a 2nd order slope on the mid HP filter, which requires inverting the midrange phase relative to the woofer. Of course, you'd invert the tweeter too to follow the mid.
                    Thanks for the input Pete. I spent the morning working on this. I reversed the polarity on the woofer and left the mid and tweet connected the same. Not sure if this is correct but the plot looks the same either way when I flip the phase as long as the mid and tweet are opposite the woofer.

                    There are less crossover components this way and the impedance curve looks a tick better. The phase looks a bit better with the woofer ( I think) and the crossover point on the woofer looks better @ 350 than where i had I around 400hz.

                    It looks like I might have a bit of a phase issue with the tweeter around 5000hz. I am not sure how to deal with this.

                    I also lost the padding on the tweet and mid but I will likely have to put them back for tuning. the system curve seems to have a bit more efficiency too. Let me know I I am on the right path.

                    Click image for larger version

Name:	XOVER 7.PNG
Views:	1
Size:	521.3 KB
ID:	1384250
                    BSME
                    Kannapolis NC

                    Comment


                    • #11
                      Just thinking out loud ... there are peeps on here that can sim everything from start to finish and wind up being dead nuts on when the build is finished.

                      That is so not me.

                      What I like to do is sim everything just to prove the concept should work, then build the speaker, take measurements with everything in the box, then start building/simn' the crossover for real. And of course order extra crossover parts for fine tuning.

                      Just saying, don't be so worried about everything being perfect now while all you really need is proof it's going to work like you want.


                      IMSHO, YMMV
                      http://techtalk.parts-express.com/fo...khanspires-but
                      http://techtalk.parts-express.com/fo...pico-neo-build
                      http://techtalk.parts-express.com/fo...ensation-build

                      Comment


                      • #12
                        Originally posted by Kornbread View Post
                        Just thinking out loud ... there are peeps on here that can sim everything from start to finish and wind up being dead nuts on when the build is finished.

                        That is so not me.

                        What I like to do is sim everything just to prove the concept should work, then build the speaker, take measurements with everything in the box, then start building/simn' the crossover for real. And of course order extra crossover parts for fine tuning.

                        Just saying, don't be so worried about everything being perfect now while all you really need is proof it's going to work like you want.

                        IMSHO, YMMV
                        I am right there with you.

                        I design front suspensions for a living (among other automotive systems). The model bump curve never really matches the car on the plate. But if you are good you can get it close. I average about a .05in in the model target to the built product.

                        Speakers are different for me. I am out of my element.

                        I have never had a good understanding of how a passive crossover works fundamentally. What each component does and its effect on the other components. I have built speakers yes, many. But because of my lack of knowledge in this area I have always taken short cuts such as using active crossovers and cutting sections out of existing speaker models and trying to match them to others.

                        For me this exercise is about learning to build a speaker from the ground up. I have a woodworking shop, I have the measuring equipment and I have the software. Now I need to put it all together and gain some experience.

                        I have so many more questions than answers. But then, I love being hungry.
                        BSME
                        Kannapolis NC

                        Comment


                        • #13
                          This thread's gettin' kinda long (and I can't read it all every time, so...) but I THINK someone offered up a 2nd and 3rd order HP for the tweeter.
                          TRY those (along w/maybe flipping the tweeter - if you have to) to see if you can change that phase issue.
                          As filters go up (or down) in order, they "rotate" (change the phase of) the signal that they pass.
                          IF you had a pair of drivers (let's say IDENTICAL drivers - like full rangers) that were perfectly in phase, then made a 2-way out of them by putting a 1st order LP filter on one (just a series coil - for woofer duty), and a 1st order HP filter on the other (just a series cap - for tweeter duty), they would no longer be "in phase".
                          Increase both filter orders to 2nd, and odds are you'll have to "flip" the tweeter (connect it "backwards") to get them back in phase.

                          Rule of thumb (for a 3-way) is to TRY to get the mid to cover 3 octaves (so, a bandpass whose Fcs are 8x apart - like 400Hz and 3.2kHz).
                          This isn't always doable, but the greater spread helps to keep the impedance propped up.
                          Also (IF your sim software allows it), try putting the (mid's) bandpass series components "up front", followed by the shunt legs (to ground).
                          THAT can also help hold the impedance up, for some reason.

                          Comment


                          • #14
                            For the phase issue at 5000 it might be the breakup bump from the rs150 . Try putting an 8uF cap in series with a 0.05mH inductor and putting those in parallel with the rs150 right next to the driver to see where the notch strikes.you can alter those values plus add a series resistor to lessen the impact of the notch

                            Comment


                            • #15
                              +1 on the fact that the sim only get gets you so far. I find the SPL match and phase are not that close to the sim
                              John H

                              Synergy Horn, SLS-85, BMR-3L, Mini-TL, BR-2, Titan OB, B452, Udique, Vultus, Latus1, Seriatim, Aperivox,Pencil Tower

                              Comment

                              Working...
                              X