Announcement

Collapse
No announcement yet.

Fitting HT subwoofers in a pre-defined space

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

  • Fitting HT subwoofers in a pre-defined space

    It is time to upgrade two heavily Mini-DSP Equalized cheap Monoprice subwoofers with something better. The Monoprice units are currently behind speaker cloth covered openings under a countertop that fits into the bay window-shaped area in my basement home-theater room. The main part of the room is 11.5 by 19 by 7.5 feet with openings at the back to a stairwell, kitchenette and den. This probably put the room in a medium size category from a bass perspective.

    The audiovisual stage is against the shorter wall where the bottom of the projector screen drops to just below the countertop height. All primary reflection points from the three main front speakers are treated with acoustic panels and there are a couple of Rockwool bass traps in the front corners.

    First measurements indicate that it will be possible to fit a couple of custom made enclosures of up to 6.1 ft^3 each (internal volume excluding ports, braces and driver) in the available spaces under the countertop (and still maintain the WAF). The positions and shape of the enclosures will allow for the drivers to be positioned on the recommended 1/4 wall length points.

    Filling as much as possible of the available space with the new enclosure will (hopefully) also minimize the effect of the countertop and cabinet. A minimum gap of 1/2 inch will be maintained to prevent any potential mechanical interference with adjacent surfaces.

    I really liked what the parametric equalization the Mini-DSP unit could do for the Monoprice units so the Behringer NX3000D amp is on the shopping list. (The notoriously noisy fan will be replaced).

    A 60/40 music/movie ratio applies. Sound Quality is at the top of the list but if there is sound in the 20 Hz band I would like to hear it. I have absolutely no intention to break any SPL records or any interest in making the neighbour’s dentures rattle on his nightstand.

    I am in two minds about ported or sealed. I like the reported “tightness” of sealed but not the roll off. Not sure if the room gain in my setup will compensate for the roll-off. The REW Room Sim does not show that and I don’t recall seeing that effect when I did the DSP calibration for the Monoprice units. So I started with ported.

    In order for the box to fit around a plate on the back wall, there will be an off center 1.5 inch deep recess at the back of the enclosure. Since the back would be prime space for a conventional slotted design, this pretty much rule out the use of slotted vents.

    The Parts Express recommended enclosure size (5.58 ft^3) for the Dayton Audio UM12-22 drivers appears to be the closest fit for a ported enclosure. Applying the T&S parameters for the UM12-22 to WinISD indicates that the expected 5.1 ft^3 (net) box will require two 4” ports.

    The ports will be constructed from white PVC pipe and each port will require a combination of 90 degree and 45 degree elbows to shoehorn a port length of up to a 38 inch in the enclosure and still maintaining the clearances around the internal openings of the two ports. Four 3D printed flares will hopefully eliminate/minimize any chuffing.

    The drawing for the enclosure layout is attached. It is not clear from the drawing but the inside openings of the two ports are swiveled inwards to move them away from the inside surfaces. Sound absorbing material against the back of the baffle across from the inside port openings will hopefully reduce any higher order frequencies potentially being reflected into, and exciting the higher order resonances of the port.

    Simulating a sealed enclosure with the same net volume and UM-12-22 driver indicates that this will also keep the possibility open to plug the ports and get a transfer function very similar to that of the WinISD recommended sized sealed enclosure. Probably easy to reduce the internal volume after the fact if so required.

    Should I go for sealed from the word GO I would probably select the UM15-22 mainly due to the slightly wider bandwidth (compared to a sealed UM12-22) that WinISD predicts.

    Any other options that I should consider given the space available? Your opinions will be highly appreciated.







    Ported UM12-22 in 5.1 ft^3

  • #2
    If you have room for the ported boxes by all means do it. If it turns out that your room gain makes the additional extension of ported unnecessary you can always seal the cabs. On the subject of the ports, a single 4" is probably sufficient. It's unlikely that you'd ever run with more than 200 watts actual program power, and that's only potentially problematic below 20Hz, where it's even more unlikely you'll put more than 50 watts actual program power into it.
    www.billfitzmaurice.com
    www.billfitzmaurice.info/forum

    Comment


    • #3
      Originally posted by billfitzmaurice View Post
      If you have room for the ported boxes by all means do it. If it turns out that your room gain makes the additional extension of ported unnecessary you can always seal the cabs. On the subject of the ports, a single 4" is probably sufficient. It's unlikely that you'd ever run with more than 200 watts actual program power, and that's only potentially problematic below 20Hz, where it's even more unlikely you'll put more than 50 watts actual program power into it.
      Thanks, that will certainly free up some net volume and reduce the required port length with the associated increase in the first port resonance frequency. You are correct, feeding the driver with 100W results in an SPL between 102 and 104 dB and a max airport velocity of 26m/s at the 16.5 Hz (port is tuned to 18 Hz). The SPL is almost at maximum THX levels considering that there will be some room gain and there are 2 drivers in the room. I was wondering about under what real life (music or movies) any potential chuffing (in this kind of setup) will actually be heard over the other sound content.

      Comment


      • #4
        At the levels required for chuffing to occur I can't imagine being able to distinguish it.
        www.billfitzmaurice.com
        www.billfitzmaurice.info/forum

        Comment


        • #5
          1. Take advantage of the DSP and use a high pass filter to limit cone excursion
          2. If you replace the fan don't change the air flow direction of it. It blows across the power transistors and won't work as effectively if it's turned around to exhaust the air from inside

          Comment


          • #6
            Originally posted by devnull View Post
            1. Take advantage of the DSP and use a high pass filter to limit cone excursion
            Do you know if the high pass filter in the NX3000 can be tuned? Ideally I would prefer something around 16Hz

            2. If you replace the fan don't change the air flow direction of it. It blows across the power transistors and won't work as effectively if it's turned around to exhaust the air from inside
            Yes, I am also going to investigate if I could increase the surface area of heatsink.

            Comment


            • #7
              Thanks to billfitsmaurice's welcomed suggestion there are now more options for fitting a significant shorter single 16" port. Unfortunately, in order to keep a 4" clear area around the inside opening it may still require an elbow, either 45 or 90 degrees.

              Theoretically the port can be placed on any location on the front baffle as indicated by the green area in the attached drawing. What are the guidelines for placing the port in this case? I have read about coupling between the driver and having the port too far from the driver not being good for the sound at the listening position. So close may be good but not to close? 1 port diameter? Closer to the floor or higher up (we are talking about a port tuned to 18 Hz). Probably away from ear level (i.e. closer to the floor)

              Also, I assume it is preferable that the inside opening should not face the back of the driver. My thoughts are that the inside port should face some sound absorbing material to reduce the chance of any higher order component exiting the port resonance. Any other suggestions?

              Edit: This has kind of been discussed before:

              http://techtalk.parts-express.com/fo...-tube-location

              Last edited by BertusS; 08-14-2019, 09:02 PM.

              Comment


              • #8
                The port exit needs to be less than 1/4 wavelength from the cone. Since the shortest wavelength in the port pass band is roughly 28 feet you don't have a problem.
                www.billfitzmaurice.com
                www.billfitzmaurice.info/forum

                Comment


                • #9
                  Originally posted by BertusS View Post
                  Do you know if the high pass filter in the NX3000 can be tuned? Ideally I would prefer something around 16Hz

                  Yes, I am also going to investigate if I could increase the surface area of heatsink.
                  Lowest frequency in the Behringer software is 20Hz. There is a way around this. Set a steep high pass filter, something like a 24 or 48dB Butterworth at 20Hz, Then set a narrow Q 3dB peq filter at 20Hz to bring back the rolloff of the high pass filter down to 16Hz. I doubt you'll hear or feel anything adding it back in.

                  You won't be able to tell until you measure and don't forget you can EQ out peaks but you can't EQ up nulls.

                  Comment


                  • #10
                    Originally posted by devnull View Post

                    Lowest frequency in the Behringer software is 20Hz. There is a way around this. Set a steep high pass filter, something like a 24 or 48dB Butterworth at 20Hz, Then set a narrow Q 3dB peq filter at 20Hz to bring back the rolloff of the high pass filter down to 16Hz. I doubt you'll hear or feel anything adding it back in.

                    You won't be able to tell until you measure and don't forget you can EQ out peaks but you can't EQ up nulls.
                    Thank you devnull - I'm looking forward to play with the future toy!

                    Comment


                    • #11
                      The required length of the port and the distance between the baffle and enclosure backwall was a concern. So I entered the enclosure dimensions in "BoxNotes" https://www.subwoofer-builder.com/freesoft.htm and it calculated numbers that are a bit too close for comfort to my newbie mind. The side to side resonances in the calculation are incorrect because of the angle of the one side, but that is not what I was after.


                      Click image for larger version  Name:	SubWooferDrawing3.jpg Views:	1 Size:	87.2 KB ID:	1419524

                      The question: How far should the 1st port resonance be from any internal enclosure resonance? The graphic at the bottom of the screenshot suggest that I may be okay: - No?

                      Comment


                      • #12
                        May have found a way to increase the net box size to 5.8ft^3 resulting in a shorter port. 5.8 ft^3 is coincidentally also parts express' recommended net volume. I will also certainly follow billfitsmaurice's recommendation from another post to line the inside with sound absorbing material - especially the risky surfaces. Thanks again to billfitsmaurice and devnull for taking the time to respond to my posts.

                        Comment


                        • #13
                          Those resonances would be of concern if they were within, or at least close to, the speaker pass band, but they aren't. A potential resonance only resonates if it's excited.
                          www.billfitzmaurice.com
                          www.billfitzmaurice.info/forum

                          Comment


                          • #14
                            I wouldn't worry about it. This is a subwoofer so your AVR is going to low pass the signal going to it. I'm going to guess that it's using at least a 12dB an octave filter, if not an 18 or 24. Let's guess again that you're using 100 Hz as the crossover freq to the subwoofer. That means that at 400 Hz you are down at least 24dB which shouldn't be audible. If you have problems you can always set a low pass filter on the subwoofer amp.

                            Comment


                            • #15
                              Thank you billfitzmaurtice and devnull for taking the time to respond. Much appreciated. My main concern is if these resonances are excited by higher order harmonics caused by driver distortion, especially if these resonances, being so close to each other, have a combined effect. I don't have a gut feel for what to expect. I looked at my enclosure dimensions where I made a 2" provision for feet. Initially had cones on felt pads in mind to rest on the laminate floor. I have now decided to rather go for lower profile rubber feet and that would allow for some additional volume, resulting in requiring a shorter port length and a greater separation of the resonance frequencies. I am still planning on lining at least the back wall and the wall opposite the port opening with something like a mattress cover or cotton batting, (still reading about this part) which would hopefully further reduce the risk.

                              Comment

                              Working...
                              X