Announcement

Collapse
No announcement yet.

A good tweeter to match 5" SB Acoustics Magnesium Woofers

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

  • Originally posted by johnnyrichards View Post


    I do the same thing, although I sometimes refer to it as a "BBC dip". Flat on-axis on so many of my design attempts sounded poor, when I finally disciplined myself to take off-axis measurements, I figured out why.

    Here's a sim of diffraction response on axis and 30 degrees horizontal off axis for a small box/driver i was playing with. Sim is accurate, verified by my measures.

    Eq the difracttion on axis by getting rid of the dip (eg with a high Q high pass) and the 30 degrees response 9and others off axis) suffers significantly. I like tilted baffles because the "on axis" response is closer to the 30 degree cone average and eqing close to flat on axis creates a much more balanced tone in room than when eqing on axis flat for a box with a vertical baffle.

    Click image for larger version

Name:	Diffrraction on axis vs 30 deg.JPG
Views:	1
Size:	46.4 KB
ID:	1360100

    Comment


    • Power response is everything, if other factors are audible but not clearly measurable in regards to strain on the tweeter
      that seems like more than enough to me. But measurements will continue anyway.
      Guess xmax's age.

      My guess: 15. His grammar is passable. His trolling is good.

      Comment


      • Originally posted by DDF View Post


        Here's a sim of diffraction response on axis and 30 degrees horizontal off axis for a small box/driver i was playing with. Sim is accurate, verified by my measures.

        Eq the difracttion on axis by getting rid of the dip (eg with a high Q high pass) and the 30 degrees response 9and others off axis) suffers significantly. I like tilted baffles because the "on axis" response is closer to the 30 degree cone average and eqing close to flat on axis creates a much more balanced tone in room than when eqing on axis flat for a box with a vertical baffle.

        [ATTACH=CONFIG]n1360100[/ATTACH]
        Precisely.
        Don't listen to me - I have not sold any $150,000 speakers.

        Comment


        • Originally posted by xmax View Post
          Power response is everything,
          No.
          Don't listen to me - I have not sold any $150,000 speakers.

          Comment


          • Originally posted by xmax View Post
            The way I deal with some of these issues. This helps avoid a BBC dip on axis.

            Here's an example of one of my designs with much heavier absorption treatment and the effects off axis. Reduces first diffraction peak but made speaker sound lifeless in this case, so abandoned it and instead used a strip of felt between woofer and tweeter only. Reflection/diffraction off woofer was a bigger issue than off cabinet edges, but the "cabinet' is a tube and generously rounded over.

            Click image for larger version

Name:	Diffract1.JPG
Views:	1
Size:	114.7 KB
ID:	1360281

            Comment


            • Originally posted by dlr View Post
              Here's my experience with M/T crossovers on this pair which I would expect to work on similar small midwoofer 2-ways....
              You're in an enviable position with the UE. I think your comparison is a bit apples and oranges though
              - in your application, you can use a brick wall filter and push the tweeter down very low
              - you're willing to accept higher distortion
              - tweeter had a waveguide
              - midrange is smaller than usual

              Given all these, you were able to push the tweeter low to meet up with the woofer off axis. However remove any one of these, and when you have differing off axis dispersion to deal with, IME its still better to use a soft knee and wider overlap.

              johnk's web site has a good comparison of differing filter orders and impact of 'power response" but these comparisons (as was Jeffs) are too high level and really need to include some model of driver directivity. Of course any xover that is "in phase' at listening axis will be out of phase at other angles and not sum as strongly, so the greater the overlap, the greater the power response hole. But that's the point, it is 'filled in' by the excess tweeter dispersion near fc. i also agree with gfiandy over at diyaudio that a boost in the low pass on axis right at the point it starts to roll off is very helpful with providing fill in for the woofer's beaming. However, i think its best to use a woofer that has this feature in it's natural response and still use overdamped filters. I have an example design that does this posted over at DIYaudio for my JR149 re-imagining (its one of my favorite speakers, if limited in dynamic range).

              Comment


              • Guess xmax's age.

                My guess: 15. His grammar is passable. His trolling is good.

                Comment


                • Originally posted by djg View Post
                  Guess xmax's age.

                  My guess: 15. His grammar is passable. His trolling is good.
                  15, eh? After reading your " trolling" post, I then must assume that you are 12, 13 maybe?

                  Comment


                  • Close.

                    Comment


                    • Originally posted by xmax View Post
                      Power response is everything...
                      I will repeat what John said... No.

                      While I expect that you will not accept this sound advice, I am suggesting that you should buy and read with an open mind, the 3rd edition of Dr. Floyd E. Toole's book, "Sound Reproduction, The Acoustics and Psychoacoustics of Loudspeakers and Rooms". Turn off the confirmation bias filters and set aside what you think you already know, and learn.
                      "Our Nation’s interests are best served by fostering a peaceful global system comprised
                      of interdependent networks of trade, finance, information, law, people and governance."
                      - from the October 2007 U.S. Naval capstone doctrine
                      A Cooperative Strategy for 21st Century Seapower
                      (a lofty notion since removed in the March 2015 revision)

                      Comment


                      • Originally posted by DDF View Post
                        I like tilted baffles because the "on axis" response is closer to the 30 degree cone average and eqing close to flat on axis creates a much more balanced tone in room than when eqing on axis flat for a box with a vertical baffle.
                        Reminds me of something little different, but maybe of interest to you. Take a look at entry number 42 on D.B.Keele's webpage of AES papers. I will try to link that below, but this board software doesn't always play well with my old tablet.

                        edit: Well that did not work, so trying hand coding the link.
                        D.B.Keele's AES papers (see entry number 42):
                        http://www.xlrtechs.com/dbkeele.com/papers.htm

                        I noticed Horbach-Keele linear phase digital crossover filters for pair wise symmetric multi-way loudspeakers mentioned here in the article directly above a different unrelated article about DBK's CBTs.
                        http://www.linkwitzlab.com/Constant_...#Horbach-Keele










                        Click image for larger version

Name:	xo_filters.jpg
Views:	1
Size:	46.9 KB
ID:	1360339
                        .
                        Last edited by JRT; 01-13-2018, 04:16 PM.
                        "Our Nation’s interests are best served by fostering a peaceful global system comprised
                        of interdependent networks of trade, finance, information, law, people and governance."
                        - from the October 2007 U.S. Naval capstone doctrine
                        A Cooperative Strategy for 21st Century Seapower
                        (a lofty notion since removed in the March 2015 revision)

                        Comment


                        • Dave, this is a response to your post 247. Somehow that got lost in the edit.

                          True except for the low driver, at least for purposes of this thread discussion. The OP originally asked about a tweeter to match an SB 5" driver as was my midrange and the design by Jeff that was criticized without having been heard by xmax is a 2-way with an SB12 NRX, a 4" driver. Which makes Jeff's crossover even more reasonable. I wouldn't expect much tweeter flare in comparison to the off-axis of a 4" driver crossed as he did.

                          But I wouldn't call an LR8 to be a brick wall filter and the waveguide is really rather limited.

                          I had thought the DXT (95db, rated by Seas down to 2K) to be the equivalent motor as the 27TDFC (90db), but that one is rated down to 1500Hz and higher power. I may swap out one of my other standard tweeters for comparison, but I've been pleased with the system as it is. I auditioned it at my last DIY event and pushed it as hard as anyone wanted to hear it, higher than I usually do. No one complained about distortion or unpleasantness at all. I heard none. I'm of the opinion that distortion of a good tweeter is a lot less of an issue than many believe, at least with typical music content.

                          These graphs are for my Chameleons, the 2-way with 6.5" SB woofer and 1" SB tweeter. They can be generated in WinPCD. The calculations are for various polar responses as indicated for the crossover and driver variation with polar vectors. In WinPCD you examine a single off-axis point or generate these graphs. The off-axis uses a modified Bessel function for driver directionality. Simplistic, yes, but at least it provides a reasonable idea. It doesn't have any diffraction, so the step isn't part of it. Maybe one day. But it does show the crossover integration in the polar response reasonably well. I added in a Bessel rolloff as well to the result of the uncorrelated power response calculations using the 45 degree angle as a guesstimate for the directionality influence.

                          The horizontal is -90 to +90 as is the vertical. The system plot includes the calculated response for a system assuming the on-axis response to be non-correlated for power, similar to Jeff's PCD. The hemispherical graph shows the integrated response for each ring of a set of concentric circles at 5 degree intervals and the total front hemispherical response. Taken together they define a hemisphere at 1m radius.

                          The power response calculated by simply considering all driver responses to be non-correlated isn't all that different from a set of hemispherical calculations integrated over the front hemisphere.

                          I can't (yet) export any of the calculated power curves, but I set the graphs roughly equivalent screen sizes.

                          dlr Click image for larger version

Name:	Horizontal Polar.gif
Views:	1
Size:	30.0 KB
ID:	1360325
                          Click image for larger version

Name:	Vertical Polar.gif
Views:	1
Size:	33.3 KB
ID:	1360326
                          Click image for larger version

Name:	Calculated Power.gif
Views:	1
Size:	31.1 KB
ID:	1360327
                          Click image for larger version

Name:	Hemispherical.gif
Views:	1
Size:	46.3 KB
ID:	1360328
                          Click image for larger version

Name:	Calculated Hemishperical vs Uncorrelated Power.gif
Views:	1
Size:	37.3 KB
ID:	1360329
                          Last edited by dlr; 01-14-2018, 09:51 AM. Reason: Added detail of my 5" midrange
                          WinPCD - Windows .NET Passive Crossover Designer

                          Dave's Speaker Pages

                          Comment


                          • Is the tweeter about 50dB down at Fs? Just curious.
                            Guess xmax's age.

                            My guess: 15. His grammar is passable. His trolling is good.

                            Comment


                            • Originally posted by xmax View Post
                              Is the tweeter about 50dB down at Fs? Just curious.
                              No

                              dlr
                              WinPCD - Windows .NET Passive Crossover Designer

                              Dave's Speaker Pages

                              Comment


                              • About how many?
                                Guess xmax's age.

                                My guess: 15. His grammar is passable. His trolling is good.

                                Comment

                                Working...
                                X