Announcement

Collapse
No announcement yet.

Damping/Decoupling material experiment

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

  • Damping/Decoupling material experiment

    I'm building a 3 way speaker with a pretty massive subwoofer in it and I'm planning to decouple it from the enclosure. I know, easier said than done.

    I was planning to do a traditional decouple, that is:

    Between the woofer rim and the enclosure, there will be some foam and a wooden ring (to bolt into). These bolts will go through the mounting screw holes on the front of the driver, through the wooden ring, through the foam, through the enclosure, and out the other side. Then, I plan to add some foam on the other side and use a washer and nut to keep the bolts in place with the foam preventing the nuts from touching the enclosure. Details are still being worked out.

    Onto the question: foam seemed like a really good decoupler because of its low energy transmission. If the foam is similar to stuff like memory foam I'd imagine it wouldn't transfer much energy. However, I have some green glue on hand and I was curious if that would be a better solution to the foam on the outside of the enclosure. Despite Green Glue company claiming its not an adhesive, it is VERY strong and VERY sticky. It would absolutely be able to hold a subwoofer of this caliber up. I have two pieces of wood in a CLD formation with green glue in the middle right now. Knuckle test tells me it's very dead, but I don't know if it is effective in sub frequencies.

    In short, do you guys think foam or green glue is a better mechanical decoupling agent?

  • #2
    I personally don't think you should decouple a subwoofer. It's operating too low in frequency to excite panel resonances where decoupling is most helpful. It is in the region where the internal pressure flexes the side walls, making it act like another source. Bracing is what stops this action most effectively.
    ~Brandon 8O
    Please donate to my Waveguides for CNC and 3D Printing Project!!
    Please donate to my Monster Box Construction Methods Project!!
    DriverVault
    Soma Sonus

    Comment


    • billfitzmaurice
      billfitzmaurice commented
      Editing a comment
      Agreed. Best results are had when the enclosure is structurally inert and the driver is rigidly coupled to it.

  • #3
    This subwoofer is playing up to 200-300hz. My other large speaker I built had substantial panel resonance at 70hz. This is obviously an entirely different box but from my experience, box resonance is somewhere in the 100hz to 200hz range, which will absolutely be excited by the sub. (This box is going to be 70L)

    I am also making a 3 way, so every frequency from 20hz to 20khz will be covered. Bracing pushes the resonant frequency upwards, no? Wouldn't a different driver just excite it instead of the subwoofer? The subwoofer box is the same as the mids box, which is the same as the tweeter box. Granted, the mids will be sectioned off so it doesn't share the same air as the sub and the tweeter is sealed, but anyways.

    If bracing is the most effective solution, what's the best way to brace? If tried cross bracing dowels but it has not been efficient. Maybe I needed more of them idk. I used one per 8 inches of wood that wasn't covered. It just seems like I can add dowels and the resonance doesn't do anything.

    Is there any use for CLD damping the inside of a speaker box then bracing the sandwiches? Or is that a bunch of work for no gain?

    Comment


    • billfitzmaurice
      billfitzmaurice commented
      Editing a comment
      The raising of the box resonant frequency is a by-product of bracing, it's not the goal of bracing, which is rigidity. You can get rigidity via bracing, you can also get it via mass. Six inches of concrete doesn't have a high resonant frequency, and yet it's very rigid. We don't usually build cabs from concrete because it's not practical. BTW, a three way that covers 20 to 20kHz isn't the best option, as ideal placement of low frequency and high frequency sources is almost never within the same foot print. The exception is when the speaker is used outdoors, where room modes and boundary reflections are not a concern.

  • #4
    You seem to be getting some of your physics confused and bracing isn't the answer for a single cabinet in the way it can be for a woofer cabinet isolated from a mid and tweeter cabinet.

    The cone assembly in a sub tends to be pretty heavy compared to the frame assembly which means that if you isolated the driver from the cabinet perfectly with an infinitely soft spring the frame would move noticeably in reaction to the cone motion adversely affecting it's intended position. A solid base for the frame is preferable which can be created by a stiffly attached heavy cabinet or better a force cancelling twin driver moving in the opposite direction. It is not uncommon for a single sub in a light cabinet to move around the floor which is another indication of the size of the issue.

    Compared to a woofer/sub a midrange tends to have a significantly lighter cone compared to the frame/magnet. This has lead some designers to mount mids on soft grommets in order to isolate the driver from the cabinet (e.g. Andrew Jones). A more fiddly but likely better performing approach is to isolate a separate midrange cabinet from the woofer cabinet.

    CLD (done properly) is an effective way to introduce damping. If your cabinet has resonances in the passband of the drivers then you need damping not stiffness to make a cabinet quiet. So bracing is just the job for a sub or woofer cabinet which can be made stiff enough for the lowest resonance to be above the passband. This isn't the case for a midrange cabinet which will include resonances in the passband. They need plenty of damping with some approaches even opting for floppy rather than stiff cabinets due to how we perceive resonances.

    Comment


    • #5
      Originally posted by andy19191 View Post
      You seem to be getting some of your physics confused and bracing isn't the answer for a single cabinet in the way it can be for a woofer cabinet isolated from a mid and tweeter cabinet.

      The cone assembly in a sub tends to be pretty heavy compared to the frame assembly which means that if you isolated the driver from the cabinet perfectly with an infinitely soft spring the frame would move noticeably in reaction to the cone motion adversely affecting it's intended position. A solid base for the frame is preferable which can be created by a stiffly attached heavy cabinet or better a force cancelling twin driver moving in the opposite direction. It is not uncommon for a single sub in a light cabinet to move around the floor which is another indication of the size of the issue.

      Compared to a woofer/sub a midrange tends to have a significantly lighter cone compared to the frame/magnet. This has lead some designers to mount mids on soft grommets in order to isolate the driver from the cabinet (e.g. Andrew Jones). A more fiddly but likely better performing approach is to isolate a separate midrange cabinet from the woofer cabinet.

      CLD (done properly) is an effective way to introduce damping. If your cabinet has resonances in the passband of the drivers then you need damping not stiffness to make a cabinet quiet. So bracing is just the job for a sub or woofer cabinet which can be made stiff enough for the lowest resonance to be above the passband. This isn't the case for a midrange cabinet which will include resonances in the passband. They need plenty of damping with some approaches even opting for floppy rather than stiff cabinets due to how we perceive resonances.
      Edited because blowhards just don't matter to me anymore
      Craig

      I drive way too fast to worry about cholesterol.

      Comment


      • Steve Lee
        Steve Lee commented
        Editing a comment
        I am trying hard to understand where the blowhard is wrong this time.

    • #6
      I'm still new to this site and im not super sure on how to use it yet so I will just say:

      In response to bill:
      Sounds good. Makes sense that mids and highs placement aren't ideal, but we run into a couple other issues then.
      1. I don't have amplifiers to run two separate subs. (Plate amps, that is).
      2. If I don't make a 3 way tower here, I will need to run my midrange much lower than I was intending to ensure the crossover frequency to the sub is omnidirectional (which I think is roughly 60hz). It is a 5-inch midrange that has a power handling of 50W (which I think is either full range or from 40hz-5khz) I will be running it from 200hz to 2khz, which is ~3 octaves less, which theoretically should "increase" power handling. Along with increased power handling, the subs can cross over higher.
      I would have to change the crossover slope if I were to do separate enclosures for my mids and subs, which affects my mid's power handling.

      In response to Andy:
      I am aware that bracing is not a "one size fits all" solution. Sometimes CLD is better and sometimes it's not, depending on scenario. I was trying to figure out a "best of both worlds" scenario which I'm sure plenty are trying to do. It seems what you are saying is that I should build a sub box along with midrange and tweeter boxes, because individually dealing with the resonances caused by the drivers is significantly easier than dealing with the whole thing. This is true. However, for the reasons listed in my response to Bill, this is particularly challenging to do.

      Comment


      • billfitzmaurice
        billfitzmaurice commented
        Editing a comment
        The simple cure is to use a midbass driver capable of crossing over between 80 and 100Hz. It would help to know what driver you intend to use. As for the amp situation, you don't have to use two separate amps for two subs. Below 100Hz the content is summed to mono. Where 'ensure the crossover frequency to the sub is omnidirectional', I don't understand what you mean.

    • #7
      I didn't catch that you wanted to house the subwoofer and woofer in the same box. That changes things somewhat. There is still no reason to decouple the subwoofer. I can guarantee the panel resonance at 70hz you thought you had wasn't a panel a resonance at all, but simply flexing of the panel due to internal pressure. So I standby bracing being most effective there. Along with rigid mounting of the sub. But since your woofer/mid will be in the same enclosure, you should do some sort of CLD construction, since this driver will operate in a passband that will likely excite panel resonances.
      ~Brandon 8O
      Please donate to my Waveguides for CNC and 3D Printing Project!!
      Please donate to my Monster Box Construction Methods Project!!
      DriverVault
      Soma Sonus

      Comment


      • #8
        Bill,
        Yes, the simple cure is to get a different driver that can play that low. I did look at the specs of the driver again and maybe 100hz is in the cards but I definitely couldn't run them at 60hz. The midrange/midbass I'm using is the SB15NBAC30-4 from SB acoustics. Unfortunately, I already have these drivers on hand so changing the driver isn't exactly an option I can pick. As for the info you gave about content summing to mono underneath 100hz, thank you I will keep that in mind.
        What I meant by "ensuring the crossover frequency is omnidirectional" is that I wanted to make sure that the direction my subwoofer is facing doesn't affect my sound. (I don't want it to sound thin if I face my sub into a wall) and to do this, I have to cross both the mids and the subwoofer low enough.

        Auger,
        You're probably correct about the panel resonance thing. I'm still learning audio and I'm not exactly sure what it was other than a roughly 15db peak at 70hz. I should note that this speaker was 9 cubic feet and was 4 feet tall with minimal bracing, so you can understand why I thought it was a panel resonance.
        As for the CLD/bracing mix you suggested, I'm very intrigued. Do you mean I should do CLD on the inside and 'brace the CLD' so to speak? (So just do CLD and then some vertical, horizontal, and depthwise braces?) Or was there an option you had in mind that I haven't considered yet?

        Comment


        • billfitzmaurice
          billfitzmaurice commented
          Editing a comment
          There's no need to go to 60Hz with the midbass, although with that driver you certainly could. It wouldn't be my choice for midbass, the low Fs/high Mms forces crossing it over to a tweeter no higher than 2.5kHz, but since you already have them you're locked in. Sub output is omni-directional up to where the baffle is a wavelength in dimension, which even at 200Hz is 5.6 feet, so that's not a concern. You do want to cross over where the subs aren't directionally locatable, which means no higher than 100Hz. As for CLD, I've tested it, didn't find it beneficial. I only use bracing, and never use thicker than 1/2" plywood construction. This includes horn loaded subs that will crack drywall before they'll suffer any panel vibration sourced coloration.

      • #9
        Originally posted by Thephantompsychic View Post
        I am aware that bracing is not a "one size fits all" solution. Sometimes CLD is better and sometimes it's not, depending on scenario. I was trying to figure out a "best of both worlds" scenario which I'm sure plenty are trying to do. It seems what you are saying is that I should build a sub box along with midrange and tweeter boxes, because individually dealing with the resonances caused by the drivers is significantly easier than dealing with the whole thing. This is true. However, for the reasons listed in my response to Bill, this is particularly challenging to do.
        Separating the woofer/sub box from the mid+tweeter box enables each box to built using an appropriate approach: stiff with no damping for the woofer/sub and significantly damped for mid+tweer. It is more fiddly but offers the highest performance in terms of minimising the radiated sound from the cabinets. A single cabinet for all drivers is unlikely to perform as well and getting a good performance isn't straightforward because you must provide effective damping which tends to conflict with providing stiffness (e.g. the CLD structural layer is only roughly half the wall thickness). Grommets for the midrange might be worth considering in this case. CLD for the baffle is normally unwise because stiffness here reduces the amount of energy put into the cabinet. This is not the case for the other 5 walls which need to get rid of the energy pumped in via the baffle using damping (or radiating away as sound which we are trying to minimise!). A good design is more challenging/fun but It is perhaps worth keeping some perspective given sound radiation from a reasonable cabinet isn't particularly intrusive.

        Comment


        • #10
          Originally posted by Thephantompsychic View Post
          I'm not exactly sure what it was other than a roughly 15db peak at 70hz. I should note that this speaker was 9 cubic feet and was 4 feet tall....
          That is quite likely a standing wave. Entirely different problem.
          Constructions: Dayton+SB 2-Way v1 | Dayton+SB 2-Way v2 | Fabios (SB Monitors)
          Refurbs: KLH 2 | Rega Ela Mk1

          Comment


          • #11
            Bill,
            I'll be honest, a big reason I went with my midrange has been the consistently good distortion measurements from this woofer and related woofers. Why must it be crossed no greater than 2.5k? Dispersion? I have in my CAD right now a 3.2k crossover 3rd order on each my mids and tweeter (SB26ADC). I suppose I could cross my tweeter lower and it wouldn't kill anyone. In fact, its 120W power rating is with a 12/octave @ 2600hz. I might do that tbh.
            As for crossing no greater than 100hz... yeah. My midrange has decent excursion but I didn't really pick it for that. I'd still need a plate amp to run the subwoofer because a passive circuit would be heavily reactive and would cause a massive dip in impedance if I had to cross that low.
            I'm trying to find excuses to build a mock box to test damping styles vs just bracing but it's kinda literally throwing money away because I can't use at least one of the boxes I make.

            Auger,
            Interesting ideas to be sure. I am aware that building one box to suit all 3 drivers is not ideal. (As it might be obvious at this point, I'm REALLY trying to keep it a 3 way. A lot of work has gone into designing it up until this point so redesigning wouldn't be particularly fun. On top of that, driver selection was made for a 3 way not a 2.1way) One interesting idea is to build a box, rigidly brace the heck out of it, then cut up some MDF squares and CLD them on the inside of the rigid box on the parts that don't have braces. Food for thought I guess. I'm not exactly sure what grommets accomplish for a midrange if I'm totally honest. I read an experiment someone did with them and it seems as if it didn't really make a difference. Iirc he did it for a subwoofer and not a midrange though. You still have to screw through the grommets and into the cabinet, no? Because if you do, that means the screws couple the driver to the cabinet.

            Comment


            • billfitzmaurice
              billfitzmaurice commented
              Editing a comment
              There are plenty of midbass drivers with appropriate Fs, 60 to 100Hz, that would work just as well where distortion is concerned, and they'd have better dispersion at the upper end, allowing a higher crossover. Higher Qts would also be beneficial, especially in a sealed enclosure. Lower Qts is something you want in a ported bass enclosure, not a sealed midbass enclosure. FWIW you keep referring to thermal power ratings. They're more or less meaningless. As for separately amplifying a sub that crosses over at 100Hz or less, that's SOP, be it in a three way or in separate cabs. I can't see building 3 ways that don't work anywhere near as well as separate subs and mains to save $100 or less for a sub amp.

          • #12
            Originally posted by Thephantompsychic View Post
            I'm not exactly sure what grommets accomplish for a midrange if I'm totally honest. I read an experiment someone did with them and it seems as if it didn't really make a difference.
            Some discussion and measurements here for a midrange. I cannot see how it would work for a sub for reasons mentioned above plus the size of the deflection, the internal air pressure moving the driver and the difficulty of getting a soft enough "grommet" to isolate. Do you have a link to a diagram of how the sub driver was mounted?

            Comment


            • #13
              Originally posted by Thephantompsychic;n1485500/
              I'm not exactly sure what grommets accomplish for a midrange if I'm totally honest. ...You still have to screw through the grommets and into the cabinet, no? Because if you do, that means the screws couple the driver to the cabinet.
              Grommets are also found as Well Nuts and Expansion Nuts. They do not screw into the cabinet but are a rubber insert that expands in the hole as the driver bolt is inserted. The bolt itself is not coupled to the cabinet and the rubber theoretically will dampen the vibrations to the Baffle. Augerpro did an experiment that showed a Neoprene gasket with screws didn't do much (due to your exact reasoning) but a Neoprene gasket with well nuts did measure differently.

              Comment


              • #14
                Originally posted by a4eaudio View Post

                Augerpro did an experiment that showed a Neoprene gasket with screws didn't do much (due to your exact reasoning) but a Neoprene gasket with well nuts did measure differently.
                I didn't use well nuts, but simple o-rings between the screw and driver flange. I do have some well nuts and would like to test them, but haven't gotten around to it. One issue with them is they require a thin baffle to work as intended, like 1/8" thick metal sheet sort of thing. Makes for a more complicated construction. More here: https://www.somasonus.net/box-construction-methods
                ~Brandon 8O
                Please donate to my Waveguides for CNC and 3D Printing Project!!
                Please donate to my Monster Box Construction Methods Project!!
                DriverVault
                Soma Sonus

                Comment


                • #15
                  Originally posted by augerpro View Post
                  ...One issue with them is they require a thin baffle to work as intended, like 1/8" thick metal sheet sort of thing...
                  Not saying you are wrong, but why is that? Here are #6-32 neoprene well nuts 0.5 inch and 1.0 inch. This weekend I'm going to use some to attach an RS100-8 full-range to a small speaker's baffle. BUT, I'm not doing any testing with NOT having the well nuts, so I won't know if it makes a difference or not.



                  Click image for larger version

Name:	6-32 1-inch.jpg
Views:	196
Size:	35.3 KB
ID:	1485515 Click image for larger version

Name:	6-32 half-inch.jpg
Views:	223
Size:	36.0 KB
ID:	1485514

                  Comment

                  Working...
                  X