Announcement

Collapse
No announcement yet.

The Super Bees - Garage two way build

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

  • The Super Bees - Garage two way build

    Hi Guys,

    This build is my first original speaker design, and is intended for me to really learn and apply some crossover knowledge. I've got a head start since the boxes were pre-assembled door prizes with blank baffles from InDIYana 2017. I decided to use the HiVi L6-4R woofer and Dayton ND25FW-4 waveguide soft dome tweeter after doing some crossover simulations in PCD and seeing that I could achieve what I think will be good results from relatively modest XO parts in 2nd order LR filters. This project has also been a convenient way to start learning Autodesk Fusion 360 CAD software, which I'm finding is more full featured than google's sketchup.

    I've been working on these here and there, and wasn't totally sure when to start this thread, so I figured now is as good a time as any. I'll try to post meaningful pictures over the next few entries to catch everyone up. At a high level, here's where we're at:

    1) Naming them the Super Bees after the Dodge muscle car of the 70's. I like the name, and the super bee logo... and these are garage speakers, so naming them after cars only made sense!
    1a) This leaves me open to re-use an Ultimax-15 driver for a garage sub, name it the Rumble Bee, and then have the whole project called the Scat Pack!

    2) Bassbox predicts Fb=51.9 Hz / F3 = 56.5 Hz with a 4.75" long x 1.728" ID round port with internal and external flares.

    3) DATS measurement shows Fb is actually 45 Hz, so I'm clearly off in BassBox somewhere.

    4) Woofer LP target is 2nd order LR @ 2200 Hz, Tweeter HP target is 2nd order LR @ 2500 Hz

    My next steps are to take frequency response measurements of the woofer and tweeter in the cabinet with my Omnimic and follow Jeff Bagby's approach to combining nearfield and farfield responses to get quasi-anechoic frequency responses for crossover design. I currently have a 10 conductor - 18AWG wire divided up between the woofer and tweeter (3 wires + and - for woofer, 2 wires + and - for tweeter) so I can hook up only the drivers I want at any given moment.

    OK, enough of my jabbering for a while. Here's some pictures of where we are, and a rendering of what I hope it looks like when I'm done!

    Click image for larger version

Name:	Super Bee F360 render 01Sept2017.png
Views:	1
Size:	57.5 KB
ID:	1343819 Click image for larger version

Name:	HiVi L6-4R Dayton ND25FW-4 Summed Response 10dB scale.gif
Views:	1
Size:	20.7 KB
ID:	1343820 Click image for larger version

Name:	20170828_211224.jpg
Views:	1
Size:	539.1 KB
ID:	1343818 Click image for larger version

Name:	20170901_161114.jpg
Views:	1
Size:	586.8 KB
ID:	1343817 Click image for larger version

Name:	20170901_221949.jpg
Views:	1
Size:	600.1 KB
ID:	1343816
    Voxel Down Firing with Dayton SA70
    Translam Subwoofers - The Jedi Mind Tricks
    The Super Bees - Garage 2 way

  • #2
    Here's a few more details around what BassBox thinks is going on vs DATS impedance sweeps. Not sure why my real measured Fb in DATS vs BassBox prediction is much lower, but I think I'm OK with an FB of 45Hz.

    Click image for larger version

Name:	Super Bee BassBoxPro Estimates.PNG
Views:	1
Size:	111.6 KB
ID:	1343829 Click image for larger version

Name:	Super Bee impedance sweep 4.75 inch port length.PNG
Views:	1
Size:	69.7 KB
ID:	1343828
    Voxel Down Firing with Dayton SA70
    Translam Subwoofers - The Jedi Mind Tricks
    The Super Bees - Garage 2 way

    Comment


    • #3
      Last post for the night, more complete simulated XO details. Once I get the real FRD and ZMA files for these drivers in this cabinet, I'll repeat this exercise and see where I land. If anyone spots any "watch-outs" in these plots, please let me know!

      Click image for larger version

Name:	HiVi L6-4R Dayton ND25FW-4 Rev 1Summed Response 70-100 10dB scale.gif
Views:	2
Size:	57.9 KB
ID:	1343833 Click image for larger version

Name:	HiVi L6-4R Dayton ND25FW-4 Rev 1 Filter Transfer Functions.gif
Views:	2
Size:	17.8 KB
ID:	1343832 Click image for larger version

Name:	HiVi L6-4R Dayton ND25FW-4 Rev 1Summed Impedance.gif
Views:	2
Size:	15.6 KB
ID:	1343831 Click image for larger version

Name:	HiVi L6-4R and Dayton ND25FW-4 Rev 1 High Pass Filter.PNG
Views:	2
Size:	117.5 KB
ID:	1343834 Click image for larger version

Name:	HiVi L6-4R and Dayton ND25FW-4 Rev 1 Low Pass Filter.PNG
Views:	2
Size:	109.1 KB
ID:	1343835
      Voxel Down Firing with Dayton SA70
      Translam Subwoofers - The Jedi Mind Tricks
      The Super Bees - Garage 2 way

      Comment


      • #4
        I like the Super Bee scheme. Have always been a huge Mopar muscle car fanatic myself, favorite being the legendary Hemi Cuda.

        What port are you using? Are you sure it's 1.728" ID? Is it a tapered port? What setting did you use in BassBox for the damping? Based on the picture of your lining, you should be using "typical" for damping. That's a couple of things that might be impacting your tuning. Could just be BassBox itself, I've had trouble with it's length suggestions in the past and had to adjust to reach desired Fb.
        My "No-Name" CC Speaker
        Kerry's "Silverbacks"
        Ben's Synchaeta's for Mom
        The Archers
        Rick's "db" Desktop CBT Arrays
        The Gandalf's

        Comment


        • #5
          As along as you are practicing you might see if you can flatten the FR above above 15K with a smaller series resistor while keeping the impedance above 4 ohms. If these were real FR's I'd built it
          John H

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

          Comment


          • #6
            Great theme and nice rendering. The sim looks pretty good! You mentioned Jeff's paper on quasi anechoic but also take a look for his method on setting up the z-offset if you haven't already.

            Keep the posts coming!

            Comment


            • #7
              OK, even coming from a Chevy guy..... this is ALL KINDS of cool!!!

              David

              Comment


              • #8
                Nice job! Looking forward to see how the real measurements compare.
                Electronics engineer, woofer enthusiast, and musician.
                Wogg Music

                Comment


                • #9
                  Originally posted by Kevin K. View Post
                  I like the Super Bee scheme. Have always been a huge Mopar muscle car fanatic myself, favorite being the legendary Hemi Cuda.

                  What port are you using? Are you sure it's 1.728" ID? Is it a tapered port? What setting did you use in BassBox for the damping? Based on the picture of your lining, you should be using "typical" for damping. That's a couple of things that might be impacting your tuning. Could just be BassBox itself, I've had trouble with it's length suggestions in the past and had to adjust to reach desired Fb.

                  Hey Kevin, I agree the 'Cuda is fantastic. Maybe some day I'll own one (If I ever quit spending money on speakers...hah!) The port I'm using is actually a couple of modified Goldwood 1.5" x 4" ports that I had left over from the Voxel project I did last winter. Since these boxes were premade, there really wasnt an easy option for flush trimming any PVC ports unless I stuck them forward firing, and I used all my real estate on the front for drivers. If you look at those ports, they are actually a bit tapered. Might be hard to tell by internet photos alone, but the vernier calipers don't lie. I measured the ports with calipers and modeled them in Fusion 360 as well, then made a combined version to determine what my minimum ID would be after cutting two ports down equally and butting them together. I made some drawings of each to help detail the idea a bit more. That's where my oddly specific 1.728" ID came from.

                  As for BassBox, I'm using "typical" as the damping setting, and 1.728 inch as my vent diameter (Dv) and 4.75 inch as my length (Lv). I have the vent cross section set to "round", so I wonder if the port having a slight taper may be messing with the calculation. Additionally, I never did fully grasp port length when including flares. Was this supposed to subtract from your assumed port length? or add to it? May need to study the bassbox help files to understand the assumptions better.

                  As always, thanks for following along!
                  Attached Files
                  Voxel Down Firing with Dayton SA70
                  Translam Subwoofers - The Jedi Mind Tricks
                  The Super Bees - Garage 2 way

                  Comment


                  • #10
                    Originally posted by jhollander View Post
                    As along as you are practicing you might see if you can flatten the FR above above 15K with a smaller series resistor while keeping the impedance above 4 ohms. If these were real FR's I'd built it
                    Thanks John! I'll give that a go later today after I finish the honey-do list. Means a lot hearing you say you'd build it!

                    Originally posted by scottvalentin View Post
                    Great theme and nice rendering. The sim looks pretty good! You mentioned Jeff's paper on quasi anechoic but also take a look for his method on setting up the z-offset if you haven't already.

                    Keep the posts coming!
                    Hey Scott! Thanks for the compliments I am definitely considering z-offset here. Omnimic has some built-in functions to measure z-offset that I believe are based on Jeff's methods. I'll be using that information along with the quasi-anechoic measurements to give myself the total picture going into the crossover design with real frequency responses.

                    Originally posted by blue934 View Post
                    OK, even coming from a Chevy guy..... this is ALL KINDS of cool!!!
                    Thanks David! I consider myself an appreciator of all historic muscle. Although I currently drive a RAM 1500, so I guess that puts me in Team MOPAR whether intentionally or not

                    Originally posted by wogg View Post
                    Nice job! Looking forward to see how the real measurements compare.
                    Thanks wogg, I'm looking forward to that as well. Might take me a few days to fumble through Jeff's methods, so we'll see when I can file the full report! I always take my time when reading through a new method. Gotta get it right so I don't waste time and money later!

                    Overall, thanks to everyone who's commented so far. I'll do my best to keep updated progress listed here. Maybe I'll send the wife and dog out on some errands soon so I can keep the house quiet for some measurements!

                    Oh... and one question on that topic. My "speaker stand" is currently my living room left channel tower. Should I worry about that when doing frequency response measurements, or is that too much to worry about? Eventually these speakers will be sitting on a workbench or shelf anyhow, so having a ground plane of sorts near the speakers is probably more realistic.

                    Again, Thanks all!
                    Keith
                    Voxel Down Firing with Dayton SA70
                    Translam Subwoofers - The Jedi Mind Tricks
                    The Super Bees - Garage 2 way

                    Comment


                    • #11
                      Looking good! These are almost too nice to be garage speakers. I would try to pull the 6-8K range down a few dB, but that's just my personal aversion to sibilance. How's your electrical phase looking? I haven't quite figured out how to manipulate that aspect of a design.
                      Co-conspirator in the development of the "CR Gnarly Fidelity Reduction Unit" - Registered Trademark, Patent Pending.

                      Comment


                      • #12
                        Originally posted by tom_s View Post
                        Looking good! These are almost too nice to be garage speakers.
                        +1

                        Comment


                        • #13
                          Originally posted by tom_s View Post
                          Looking good! These are almost too nice to be garage speakers. I would try to pull the 6-8K range down a few dB, but that's just my personal aversion to sibilance. How's your electrical phase looking? I haven't quite figured out how to manipulate that aspect of a design.
                          Thanks Tom and Mike!

                          will probably try to relax the frequency response through the mid-high registers a bit more in the final version. I tinkered with these values a lot before arriving at the current XO model. I figured this was good enough for a simulation and proof positive that good things could be done with this combination.

                          As for the electrical phase, the simulation shows me being in excellent phase alignment at 1826 Hz, where the actual Fc looks to be ~2400 Hz. As far as I understand things, this is a perfect phase alignment, but I'm not sure about how far the point of phase alignment should be from the Fc frequency. Do you guys have rules of thumb on that point? I know everything is a compromise somewhere, just trying to understand what's the least troublesome compromise to make. As far as manipulating the electrical phase during designs, I don't really know what I'm doing, I just made sure to turn on the phase lines in the graphs and observe what happened as I pushed values of components up or down. I know the design goal is to make phase agree as best as possible, so that's what I tried to do.

                          Click image for larger version

Name:	Super Bee simulated XO - Black plot to highlight phase better.gif
Views:	2
Size:	61.9 KB
ID:	1343889
                          Attached Files
                          Voxel Down Firing with Dayton SA70
                          Translam Subwoofers - The Jedi Mind Tricks
                          The Super Bees - Garage 2 way

                          Comment


                          • #14
                            Those are going to look bada$$. Glad you decided to move forward with that tweeter. I think you will find its FR is even smoother when measured on the actual baffle. I always use 1/24th octave smoothing with my OmniMic.

                            Tom was asking to see the electrical (impedance) phase plot. On that you don't want to have wild phase swings. Rule of thumb is keeping the electrical phase between + and - 30 degrees, but I've broken that rule a bit ;)
                            Craig

                            Comment


                            • #15
                              And to expand a bit on what John was saying about lifting the FR above 15kHz... you could try placing a small value cap in parallel with your 5.1 ohm padding resistor. Try 1 uF and go up and down to see the effects.
                              Last edited by PWR RYD; 09-03-2017, 06:40 AM. Reason: Typo
                              Craig

                              Comment

                              Working...
                              X