Announcement

Collapse
No announcement yet.

Mltl, Nd91 Mtm

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

  • Mltl, Nd91 Mtm

    I am working on a design exercise to model an MLTL using the ND91 in an MTM configuration. Tweeter hasn't been decided yet...

    Anyway, what I have so far is a 50Hz tuning using a 40.8" x 4.5" x 7" internal volume. Port is 2.5" x 3.25", with 0.5lb/cu ft stuffing. The simulation looks good with 8W input giving .03 port velocity and Xmax at 5mm.

    My concerns are what I don't know to be looking at/for and the in room results.

    Plots: infinity baffle response, woofer excursion, port velocity and corrected response. I just don't know if the corrected response is as an issue....
    Attached Files
    If life were fair, Elvis would still be alive today and all the impersonators would be dead.
    ~ Johnny Carson

    Bungelow Ed's Photo Album http://techtalk.parts-express.com/album.php?u=8594

  • #2
    Re: Mltl, Nd91 Mtm

    get in touch with paul k. im sure he will advise you.
    " To me, the soundstage presentation is more about phase and distortion and less about size. However, when you talk about bass extension, there's no replacement for displacement". Tyger23. 4.2015

    Quote Originally Posted by hongrn. Oct 2014
    Do you realize that being an American is like winning the biggest jackpot ever??

    http://www.midwestaudioclub.com/spot...owell-simpson/
    http://s413.photobucket.com/albums/pp216/arlis/

    Comment


    • #3
      Re: Mltl, Nd91 Mtm

      Based on your first graph, which is essentially the anechoic system response, I'd tune the system a bit lower and/or possibly increase the stuffing density and/or length to smooth out the response more as well as minimize/eliminate that peak at the knee. That's what I would do but it is a personal preference.
      Paul

      Originally posted by bungelow_ed View Post
      I am working on a design exercise to model an MLTL using the ND91 in an MTM configuration. Tweeter hasn't been decided yet...

      Anyway, what I have so far is a 50Hz tuning using a 40.8" x 4.5" x 7" internal volume. Port is 2.5" x 3.25", with 0.5lb/cu ft stuffing. The simulation looks good with 8W input giving .03 port velocity and Xmax at 5mm.

      My concerns are what I don't know to be looking at/for and the in room results.

      Plots: infinity baffle response, woofer excursion, port velocity and corrected response. I just don't know if the corrected response is as an issue....

      Comment


      • #4
        Re: Mltl, Nd91 Mtm

        If this turns out anything like Jim Griffin's design using the Aura NS3, it will be a hit. The tweeter he chose sounds very good in this app. Wolf turned me on to this design - I've built 4 sets for family & friends & they never fail to impress. Good luck!
        Audiophiles listen to the equipment, not the music.

        Comment


        • #5
          Re: Mltl, Nd91 Mtm

          Originally posted by arlis_1957@yahoo.com View Post
          get in touch with paul k. im sure he will advise you.
          Like the Mayor of Gotham City, You shone the light and our hero arrives! Thanks!
          If life were fair, Elvis would still be alive today and all the impersonators would be dead.
          ~ Johnny Carson

          Bungelow Ed's Photo Album http://techtalk.parts-express.com/album.php?u=8594

          Comment


          • #6
            Re: Mltl, Nd91 Mtm

            Originally posted by Paul K. View Post
            Based on your first graph, which is essentially the anechoic system response, I'd tune the system a bit lower and/or possibly increase the stuffing density and/or length to smooth out the response more as well as minimize/eliminate that peak at the knee. That's what I would do but it is a personal preference.
            Paul
            Hi Paul, I've seen your posts helping other and was hoping you would weigh in. Thank you for sharing your insights.

            I have tuned the design as low as 30Hz and modeled both longer lengths and higher stuffing densities. Do you have any thoughts on what tuning might be best?

            My biggest concern is the ~10 dB dip at 150Hz in the corrected SPL. That seems to be an issue no matter the tuning, speaker or mic orientation.
            If life were fair, Elvis would still be alive today and all the impersonators would be dead.
            ~ Johnny Carson

            Bungelow Ed's Photo Album http://techtalk.parts-express.com/album.php?u=8594

            Comment


            • #7
              Re: Mltl, Nd91 Mtm

              Originally posted by Steve Henry View Post
              If this turns out anything like Jim Griffin's design using the Aura NS3, it will be a hit. The tweeter he chose sounds very good in this app. Wolf turned me on to this design - I've built 4 sets for family & friends & they never fail to impress. Good luck!
              Jim Griffin's design is a favorite with me and my benchmark with this design exercise. I am trying very hard not to build a clone of his design. My goals are to gain experience with Martin King's worksheet and the PCD suite.

              Having a pair of Aura T-lines for subjective comparison is my next best thing to having and knowing how to use measurement equipment. If my results compare favorably to Jim's I will consider myself successful.
              If life were fair, Elvis would still be alive today and all the impersonators would be dead.
              ~ Johnny Carson

              Bungelow Ed's Photo Album http://techtalk.parts-express.com/album.php?u=8594

              Comment


              • #8
                Re: Mltl, Nd91 Mtm

                I determine the "best" tuning frequency simply based on how flat and non-peaky at the knee the response is. I've attached a comparison graph I created quite some time ago for an ML-TL to show the differences between tuning too high and tuning too low as compared to what I consider to be optimum. As to that dip at 150 Hz for the in-room response, there's probably not a whole lot you can do to minimize it since it's a natural outcome of the drivers' and port's locations. Frankly I almost never go beyond the first part of Martin's worksheets, particularly when modeling for others, because there's such a variety of room effects, and I suspect most speaker designs, whether from DIYers or commercial sources, don't really try to do anything corrective, just live with the results. The graphs in the attachment came out rather small; the first one is for a too-high of a tuning frequency, the second is for a too-low, and the last is what I call optimum. We each have our preferences, of course, so YMMV.
                Paul

                Originally posted by bungelow_ed View Post
                Hi Paul, I've seen your posts helping other and was hoping you would weigh in. Thank you for sharing your insights.

                I have tuned the design as low as 30Hz and modeled both longer lengths and higher stuffing densities. Do you have any thoughts on what tuning might be best?

                My biggest concern is the ~10 dB dip at 150Hz in the corrected SPL. That seems to be an issue no matter the tuning, speaker or mic orientation.
                Attached Files

                Comment


                • #9
                  Re: Mltl, Nd91 Mtm

                  Originally posted by bungelow_ed View Post
                  I am working on a design exercise to model an MLTL using the ND91 in an MTM configuration. Tweeter hasn't been decided yet...

                  Anyway, what I have so far is a 50Hz tuning using a 40.8" x 4.5" x 7" internal volume. Port is 2.5" x 3.25", with 0.5lb/cu ft stuffing. The simulation looks good with 8W input giving .03 port velocity and Xmax at 5mm.

                  My concerns are what I don't know to be looking at/for and the in room results.

                  Plots: infinity baffle response, woofer excursion, port velocity and corrected response. I just don't know if the corrected response is as an issue....
                  Ok, first off, Paul is correct, the first graph is the anechoic system response. The blue line in that graph is the driver's IB response, but it's only included for reference. I also agree with Paul that the shape of the curve in that first graph is not optimal. There's a huge amount of energy between tuning and all the way up to 150 hz or so. It would probably sound pretty good outside, but in room it will very likely be terribly boomy. I suggest taking that graph, making a .frd out of it, opening the .frd in Bagby's Diffraction and Boundary simulator to see how much boundary gain and closed vessel gain you are going to get on top of the already pronounced bottom end. Then, consider the fact that you will also have at least one room mode in that bass heavy area AND a bump down at tuning and (IMO) there will be way too much bass.
                  Don't even try
                  to sort out the lies
                  it's worse to try to understand.

                  Comment


                  • #10
                    Re: Mltl, Nd91 Mtm

                    Graph #2 is excursion. Generally, we consider the point in the graph around 60 hz to be the excursion limit, since below tuning we assume it will be protected by a high pass filter.

                    In that light, you've only used up just over 3 mm excursion as shown.
                    Don't even try
                    to sort out the lies
                    it's worse to try to understand.

                    Comment


                    • #11
                      Re: Mltl, Nd91 Mtm

                      Graph #3 is port velocity, which the sheets recommend to keep under 10 m/s. That's a good goal but sometimes not practical. Once you apply more power (see previous post) to get to xmax, you will be above 10 m/s. I've gone as high as 22 m/s (although I hate to go that high) without audible chuffing. Also, remember that some (most?) people will turn it up way past xmax on occasion, so if you intend to do that and expect to not hear chuffing, then you have to allow for it by allowing your simulation to go past xmax.
                      Don't even try
                      to sort out the lies
                      it's worse to try to understand.

                      Comment


                      • #12
                        Re: Mltl, Nd91 Mtm

                        Fourth graph is... what exactly? Need details. Like Paul, I hardly ever get down that far in the worksheets, I usually only go as far as the first section. So what are we looking at here? Is this the full model including diffraction and BSC? I prefer to do these steps in the PCD Suite tools. Just do a simulation using the first section of the worksheet, make a .frd and a .zma and continue with PCD. But of course you are free to do it any way you like.
                        Don't even try
                        to sort out the lies
                        it's worse to try to understand.

                        Comment


                        • #13
                          Re: Mltl, Nd91 Mtm

                          Originally posted by Paul K. View Post
                          I determine the "best" tuning frequency simply based on how flat and non-peaky at the knee the response is. I've attached a comparison graph I created quite some time ago for an ML-TL to show the differences between tuning too high and tuning too low as compared to what I consider to be optimum. As to that dip at 150 Hz for the in-room response, there's probably not a whole lot you can do to minimize it since it's a natural outcome of the drivers' and port's locations. Frankly I almost never go beyond the first part of Martin's worksheets, particularly when modeling for others, because there's such a variety of room effects, and I suspect most speaker designs, whether from DIYers or commercial sources, don't really try to do anything corrective, just live with the results. The graphs in the attachment came out rather small; the first one is for a too-high of a tuning frequency, the second is for a too-low, and the last is what I call optimum. We each have our preferences, of course, so YMMV.
                          Paul
                          I prefer the graph in the middle, the top and bottom graph would be too boomy in the rooms I have available to set up speakers. Of course this is purely up to personal preference and the room will make all the difference in the world.
                          Don't even try
                          to sort out the lies
                          it's worse to try to understand.

                          Comment


                          • #14
                            Re: Mltl, Nd91 Mtm

                            You're not alone in preferring a response as in the middle graph, but I use only ~3 dB of BSC because my speakers are placed with their backs within 12" of the wall behind. They never sound boomy and they always sound good, too, at DIY events, not lacking in bass when others often sound bass heavy. Now I listen to almost 100% instrumental symphonic music where usually the bass is limited to 40 Hz or so, and maybe that has something to do with it.
                            Paul

                            Originally posted by diy speaker guy View Post
                            I prefer the graph in the middle, the top and bottom graph would be too boomy in the rooms I have available to set up speakers. Of course this is purely up to personal preference and the room will make all the difference in the world.

                            Comment


                            • #15
                              Re: Mltl, Nd91 Mtm

                              Paul, Thanks for posting the Fr plots. Having examples of plots and their implications is a "pictures worth a 1000 words" moment.
                              If life were fair, Elvis would still be alive today and all the impersonators would be dead.
                              ~ Johnny Carson

                              Bungelow Ed's Photo Album http://techtalk.parts-express.com/album.php?u=8594

                              Comment

                              Working...
                              X