Port diameter trade-offs: midrange SPL leakage

Collapse
This topic is closed.
X
X
 
  • Time
  • Show
Clear All
new posts
  • MarcE
    Member
    • Jun 2007
    • 41

    Port diameter trade-offs: midrange SPL leakage

    Hi everyone!

    I have been playing around with Unibox and have a few questions about some port length trade-offs. First, I have actually heard this driver in the 3" port configuration, and there was not any problems with the port configuration, i.e. no port chuffing or obvious midrange SPL peak. That being said, I was looking at a lot of the general recommendations for port diameter for 10" drivers and it seems that people basically recommend the larger the better, so I decided to do some simulations with a 4" port. I will be using dual-flared ports.

    Most of my simulations are at half-x_max, x_max, and x_sus, although the last simulations are really not a guiding factor in the decisions. While I understand that a lot of these comparisons should be made with a crossover slope and frequency in mind, perhaps we can leave that out of the discussion for now (if possible).

    First, the resulting air speed graphs look very reasonable, so I will not post them. At x_max, the port speed is just about tangent with the 5%, .05 vent mach line with the 3" port, but this occurs under 20Hz, so I'm not so concerned. Also, the 4" port is just about tangent to .05 vent mach at x_sus. The 4" port looks even better at x_max. However, the 3" port at x_sus crosses the .05 vent mach line at 24 Hz-ish and peaks way over the .08 vent mach line at 15 Hz or so. Again, this is so far down in the noise, and plus at levels that could damage the drivers, so I'm not concerned. Does this analysis look good to you as well?

    [ note: the first attachment is for the 3" port, and the other two are for the 4" port ]

    So, looking at at the midrange SPL leakage in the first attached graphs, we see a a mild peak of 70 dB at 729 Hz due to port resonance and a fairly flat SPLtotal. To me, this looks great and the 3" port seems to be a good choice, with very few tradeoffs. The above doesn't change (according to Unibox) at x_sus for the 3" port.

    Now, the 4" port has different results, quite possibly for the worse, which is surprising. There are two large resonance peeks at 412 Hz and 820 Hz at around 80dB and 70dB respectively. These measurements are taken at x_max (see the second graph). Also, there is some disturbance in the SPLtotal graph, as you can see. Even if we look at half-x_max (the third graph), we see the the twin resonances and the disturbance starting to form.

    I guess my question is, it seems that the 3" port measures better... I find it curious that I cannot see an advantage with the 4" port, other than the air speed graphs, which seem to be so low in the frequency range (at at such high excursion levels) as they will have little effect.

    Also, before someone suggests a rear or down firing port, I'd rather keep it on the front baffle to keep everything in phase and have the ability to run my speakers closer to the rear wall, if setup dictates.

    Thanks for your help!
    Attached Files
  • cjd
    Ultra Senior Member
    • Dec 2004
    • 5568

    #2
    Are these drivers going to be actually playing that high? For a sub, there's absolutely no problem with a 400Hz port resonance when it's crossed 80Hz... it's never excited.

    I run a 3" port on my MTM (pair of RS150's) - yeah, it's large. Probably could have gone 2". But, you know, I think there may be things we don't think about normally lending advantage to the larger ports. Perhaps one reason things like a TL have a "preferred" sound.

    Regardless, the resonance artifacts aren't actually that significant in the grand scheme of things according to your graphs. Well within the range most of your diffraction artifacts will exist as far as +/- deviation.

    Also, I'd tune it a little bit lower by 1 to 2Hz.
    diVine Sound - my DIY speaker designs at diVine Audio

    Comment

    • MarcE
      Member
      • Jun 2007
      • 41

      #3
      Originally posted by cjd
      Are these drivers going to be actually playing that high? For a sub, there's absolutely no problem with a 400Hz port resonance when it's crossed 80Hz... it's never excited.
      well, I am going to try my hardest to use a first-order crossover, so that 400 Hz comes up pretty quickly. I am trying to have a good two flat octaves in the pass band on both this woofer and my midrange (a Audiotechnology 15H520613SDK). I realize I'm skimping on the details here a bit and I suppose your observations are right on the money... okay, where did I put those Speaker Workshop files, time to revisit them I've been meaning to make a post with my thoughts, but I maybe suffering a bit of design paralysis ops:

      Originally posted by cjd
      I run a 3" port on my MTM (pair of RS150's) - yeah, it's large. Probably could have gone 2". But, you know, I think there may be things we don't think about normally lending advantage to the larger ports. Perhaps one reason things like a TL have a "preferred" sound.
      So, are you implying that you'd go with the 4" port in my case? Anything "wrong" with the 3" port? The reason I ask is that it may be a moot point, as my rudimentary CAD drawings are coming along and it's a bit harder to stuff a 29 cm port low on a sloped baffle, compared to a 8 cm length in the case of the 3" diameter.

      Originally posted by cjd
      Also, I'd tune it a little bit lower by 1 to 2Hz.
      ok, I'll play around with that. Just curious, any specific reason for the suggestion from my graphs, or just a seasoned veteran's knee jerk response 8)

      Comment

      • cjd
        Ultra Senior Member
        • Dec 2004
        • 5568

        #4
        If the larger port doesn't fit, go with the smaller port! Or, you know, slot port it. Nice thing about a slot port is you can in-between size it far more easily.

        Lower tuning is a preference. You'll get boundary/room gain down low almost surely, which will push bass response above nominal. Perhaps desired, in which case ignore my comment. But I have found systems tuned flat end up going a little bloat. Keep the Q low and tune to a light diminishing response and things open up sound-wise.

        First order slopes? You'll pick up worse driver distortion artifact noise than you're going to notice from that port resonance in this case. So definitely don't worry about the port noise.

        FWIW on first order slopes, I'd be shooting for 3 or 4 clean octaves in the passband. I like 1+ octave on a 24db/octave slope. Again, preference... but -12dB on grunge isn't anywhere near out of audibile range.
        diVine Sound - my DIY speaker designs at diVine Audio

        Comment

        • MarcE
          Member
          • Jun 2007
          • 41

          #5
          Originally posted by cjd
          First order slopes? You'll pick up worse driver distortion artifact noise than you're going to notice from that port resonance in this case. So definitely don't worry about the port noise.

          FWIW on first order slopes, I'd be shooting for 3 or 4 clean octaves in the passband. I like 1+ octave on a 24db/octave slope. Again, preference... but -12dB on grunge isn't anywhere near out of audibile range.
          well, now that I've let the cat out of the bag ... I'm looking at using multiple first-order slopes, or multislope design. If anyone has more information on this, I'd appreciate it. All I can find is some information by a single poster on DIYaudio, and I am already in dialog with this person.

          Surely there is another term for this type of crossover that speaker builders use? I've looked at the THIEL site and they seem to call this "gradual-transition "first-order" slopes"... I'd love to find some AES articles or something, but I haven't had much luck... heck, a analog filter design textbook would probably suffice anyone?

          The concept is quite simple. Use two (or three) first-order crossovers such that you have a second pole about an octave away from the crossover point, and a third another octave away. The extra poles should have minimal impact on the phase of the resulting crossover as well as allow a narrower pass band. Well, this is all in theory of course and I have played around a lot with this in LTSpice only, not using one of the more specialized speaker building programs.

          cjd: Thank you so much for your comments!

          Comment

          • Dennis H
            Ultra Senior Member
            • Aug 2002
            • 3791

            #6
            Bob Cordell uses 1st order slopes just for an octave or so and lets them roll off faster beyond that. I remember him discussing it on the old Mad board but those archives have gone to digit heaven.

            Comment

            • MarcE
              Member
              • Jun 2007
              • 41

              #7
              Dennis H:

              Thanks for the link! I contacted Bob and he didn't know of many other resources either. Fortunately I have someone in my family that I can hit up about this :P

              If anyone else has some more information, please contact me!

              Comment

              • noah katz
                Senior Member
                • Dec 2005
                • 188

                #8
                Does anyone know if Unibox accounts for the box's acoustical absorption?
                ------------------------------
                Noah

                Comment

                • MarcE
                  Member
                  • Jun 2007
                  • 41

                  #9
                  Does anyone by chance have "Introduction to Loudspeaker Design" By John Murphy? I noticed that this review http://www.trueaudio.com/ild_rev1.htm mentions the discussion of first-order parallel and quasi-first-order-series crossovers. Any idea on what a "quasi-first-order-series crossover" is in this context?

                  Comment

                  Working...
                  Searching...Please wait.
                  An unexpected error was returned: 'Your submission could not be processed because you have logged in since the previous page was loaded.

                  Please push the back button and reload the previous window.'
                  An unexpected error was returned: 'Your submission could not be processed because the token has expired.

                  Please push the back button and reload the previous window.'
                  An internal error has occurred and the module cannot be displayed.
                  There are no results that meet this criteria.
                  Search Result for "|||"