115L AV15-X sub – One of a pair.

  • Member
    russ.will on #553

    Posted elsewhere, but I promised to post my build on here, if only to help others learn from my mistakes.

    It should be noted that prior to these subs, I have never built or designed a speaker of any sort, never owned nor used a router and so I am figuring it out as I go along. On that basis, if I can do it anybody can, to which end, I offer three pieces of advice:

    1. If your local DIY store offers a wood cutting service, use it. Getting panels cut millimetre perfect and SQUARE takes a serious amount of grief out of the process.

    2. When it comes to routing/drilling/fitting panels, measure, adjust, measure, adjust and if that’s what it takes, measure and adjust again. It’s a lot less grief to get it exactly right in the first place than it is to sand, or worse, fill in mistakes afterwards.

    3. Do a bit at a time. If you’re doing the build in spare evenings, don’t continue beyond the point where you are tired and actively forcing yourself to think, because in the morning you’ll find you thought wrong!

    I had originally planned a pair of 75L boxes, but decided to go for a 65L box (as stuffing would allow it to behave like anything up to a 90L box) and also a 115L box and then compare the two in terms of listening preference. I would then build a second box of whichever version I preferred. Power comes from a Behringer EP4000 (2×1400 Watts(ish) which should be more than ample for my fairly typical 16x13x8ft UK room.

    If you have the patience for the full original thread, then it’s in AV Forums in the UK here : http://www.avforums.com/forums/diy-subwoofer-build/1087529-dual-ae-av15-x-sealed-subwoofers-diy-build.html

    So, here’s the progress so far…..

    russ.will on #6086

    The drivers have arrived.:eek:

    The pictures can’t convey the loveliness, or the fact that each one weighs as much as a bag of cement.

    Some teasers:

    Me? I’m unnaturally excited by metal in a way that hasn’t happened since I last owned an Italian Motorbike.;)


    russ.will on #6087

    They are quite large. About 7″ across, but heres some more [-]porn[/-] pictures to illustrate:

    For you Sir? Of course:


    russ.will on #6088
    Russell.Williams;10646101 wrote:

    Last night I glued the last panel on the rough 65L box:

    The rear ‘panel’ will look a touch spartan compared to an active sub:

    And as I had some decent cable for internal wiring lying around, I couldn’t resist soldering to the rear terminals and installing the driver:

    How does it sound? Dunno. I’m too knackered to dismantle the rack to pull out the 1/4″ jack to RCA adaptors and rewire it all without the BFD in the loop. It’ll have to wait until tomorrow or Saturday.

    Even in that rough arsed box, it does look rather mean though.:devil:


    russ.will on #6089

    Spent a very productive day with Adam (AngelEyes) finally shaping up his cabinet and refining up my 106L cabinet. The point of interest raised during the day that was inspite of having panels cut to exactly the dimensions calculated by Google Sketchup, that we both found that certain panels were 2-3mm larger than they needed to be.

    For instance, my sub is 550mm across which means the internal brace should be 550mm minus 18mm x2 for it’s internal dimension. The internal brace was thus cut to 514mm (confirmed by a steel ruler) and yet when offering up the panels, there was a clear overlap of 2mm.:confused:

    Luckily, the presence of a Festool track saw made it the work of moments to trim the extra couple of millimetres off, but I’m still a bit confused given that the panels were cut and latterly measured to be millimetre perfect.

    Anyhow, here’s the pictures from todays work. The awesome Festool Track Saw being operated by the extraordinarily hung-over Adam:

    I promise you you WANT this £1500 tool:

    It makes cutting straight lines so easy that even idiots like us can do it. I won’t explain why, just Google it. Here’s me pillar drilling the small cut outs in the internal baffle. Bear in mind that the camera is cruel and adds at least four stone to anybody pictured: :god:

    Okay, maybe I meant four pounds…:rolleyes:


    russ.will on #6090

    Originally posted by my partner in crime (he’s having 2 of the 6 drivers I ordered):

    These photos are a little out of sync as I was under the impression I had deleted them, anyway these are from a couple of weeks ago when we started on Russell’s ‘rough’ box. The infamous Festool track saw, Russell’s Router and in the background a biscuit jointer, yet to be used in anger.

    The woodmaster himself! We learnt a lot from this early build, mostly how not to do things but it was well worth the effort as the latest boxes really are spot on. Simple things like nailing down the baffle cut-out as well as securing the baffle itself during routing helps prevent any movement when you finally cut through the final piece.

    It was a real struggle lining the baffle and end piece up because we measured and cut them seperately. For the next ones we plan to centre drill through both pieces, so the circles will exactly line up. We even discussed gluing the two pieces before making the cuts but are not quite convinced it is necessay. You live you learn :rolleyes:

    The observant among you will note the proximity of a forklift truck, having moved up to 26mm MDF, this may become quite useful once my box is finished. I can’t help thinking Graham could do with it more than me though 😉

    It was surprisingly fiddly to glue and assemble the pieces and keep everything located correctly. I am probably going to use biscuit joints to make this process much easier. It is another job but will add to the strength and make the gluing together a bit easier. I also think we learnt some lessons about the amount of glue required!

    I quite enjoyed watching this part and trying to work out how Russell got so much glue on his face, worth another shot methinks 😀

    Kudos to Russell for going first, someone had to do it and we both learned loads. It might seem wasteful financially to build a ‘rough’ test box but the lessons were invaluable and have given us greater confidence to move forward with the other boxes.

    I think we may well build several versions each to test performance as well as aesthetics. It does help having a skilled professional a few yards away who is only too happy to offer advice.

    Hopefully we will get another crack on Wednesday?


    russ.will on #6091

    Posted with respect to the 65L box:

    Response stuff as promised. The first image is the predicted response curve as modeled in Unibox which I must say, is a much nicer program to use than WinISD. Yes the roll off looks steep and from quite a high frequency, but this will be EQd to flat along with any peaks introduced by the room’s contribution.

    To appreciate what this actually means as an in room response, you have to look at this graph like this. Say you want a solid 20Hz roll off (which means -3dB at 20Hz) then pick a point on the dark blue line 3dB above 20Hz and try to EQ everything to flat at that level. As the predicted output at 20Hz is a maximum of 103dB, then that means everything else is Eq’d flat to about 106dB. Effectively, even if you’re boosting at 20Hz to bring the bottom end up, or cutting like hell to bring the mid bass down you are giving yourself a maximum SPL of 106dB across the range as that is the limit imposed by the driver and it’s excursion at lower frequencies.

    By this graph alone this means that however you boost the bottom or cut the upper bass, you’re looking at a 16dB boost at the bottom, or a 16dB cut higher up to get to flat.

    Except you’re not, because that would be the response without the contribution of the room which, when EQing you can’t ignore and to a greater degree, you are relying on to help you out. It’s dependent on the size of room you are in, with bigger rooms helping less, but from here on I’m only relating things to my room. It also doesn’t take into account the processor imposing an 80Hz roll-off (-3dB at 80Hz) which actually means that the natural response peak will be flattened as the crossover starts kicking in from about 50Hz. This means less boosting/cutting already.

    In the graph below, the test tone (Pink Noise) has been calibrated to about 75dB with EQ and so the graph below shows the un-EQd response to see what the room is actually adding. Room gain only kicks in below the lowest room mode (27Hz in my room) and adds from this point about 12dB/octave, co there is no boost at 27Hz, but 12dB extra gain at 13.5Hz. This is particularly important with a sealed sub, because they naturally roll-off at 12dB, so from just below your lowest room mode, they perform more or less flat naturally.

    And there they are; Those room modes, loud and proud and exacerbated by the sub’s humped response. There are (if you look carefully) 27Hz, 32Hz and a huge second harmonic 64Hz of the 32Hz mode, the latter being particularly reinforced by the sub’s mid-bass hump. I prefer the waterfall graphs, because where you see a hump extending from the back of the graph to the front, you know you are dealing with a real room resonance; These are the peaks you need to hit accurately with EQ filters of exactly the right frequency, the right cut and the right filter width (Q) to deal with them effectively.

    These are the frequencies that ‘ring’ and so sustain for longer in the room and it’s as important to shorten the ringing, as it is to achieve a flat response. The ear is poor at discriminating a response peak from an extended ringing and so it’s as important to achieve a short response as a flat response, even if that means creating a bit of a dip in the frequency response to compensate.

    So when you look at the EQ’d graph below (look at the back) you will see dips in the response that when dealing with 2D graphs you might try to avoid, but in the 3D waterfall are required in order to ensure a short response which to the ear sounds more even and altogether tighter.

    I should point out that having meant to reproduce this graph at a relative 75dB like the one above. I forgot to save that graph and so you should mentally subtract 5dB from this graph which was a resulted after some tweaking whilst listening to music to confirm results. What can I say? It was nearly 11 o’clock and the brain was fuzzy; Sorry!

    The point is, even given a relative 5dB boost, not only is it significantly flatter, but look how much shorter it is. By the front of the plot, the energy still ringing (resonating) in the room is clearly massively reduced. According to REW the RT60 (the time taken for the original signal to fade to -60dB, the limit of audibility) has been reduced from 0.58sec to a shade over 0.26sec. This is very nice, but as EQ can only be truly optimized for one listening position, it should be noted that this result probably isn’t true for elsewhere in the room, although it’s likely to improve to a lesser degree. A more even result for all room positions would require room treatment (diffusers and bass traps) that most people wouldn’t consider visually acceptable in a living room.

    I finally had a chance to listen with music tonight and its clear that the 65L sub offers a very different type of bass to what I’m used to. With the sub’s Qtc of 0.8(ish) it is extremely tight and possibly a touch more damped than I would like. Lower frequencies are extremely well differentiated in tone, but perhaps lack a touch of impact, that the higher frequencies (50Hz+) deliver like the proverbial mule kick. The double bass of Harry Connick Jnr’s A Nightingale Sang On Berkerly Square was tonally very well differentiated, but lacked a little of the body/resonance and therefore weight to make it sound truly satisfying. By comparison Gwen Steffani’s If I Was A Rich Girl kicked so hard it hurt and had me grinning like schoolgirl, with the right balance been the basso drops and chest thumping rhythm. If I was a pure dance/club music listener, I’d probably stop here.

    I suspect that a more even balance between the two, from experience, is probably closer to what I would be looking for to suit my taste for not only music, but movies too. Adding some stuffing to the box should help drag the Qtc down toward the more theoretical ideal of 0.707 if not all the way. The 115L box being built should not only achieve this ideal more closely, but also flatten the natural response peak, plus add a couple of dB lower down too, so comparing this, to the stuffed smaller box should be interesting.

    Looks like I’ll be frequenting the haberdashers this weekend for some long fibre box stuffing.:god:

    russ.will on #6092

    Today’s progress.

    I decided to use the side panels to position the front sub-baffle, but experience from the 65L box showed me that it was very hard even with lots of pencil marks to reposition, glue and clamp the sub-baffle accurately. I decided to carefully remove the panels and drill a couple of 2mm pilot holes to help with positioning when clamping. Thing was, I noticed that the screws pulled up the baffles really nice and tight, so I decided to do away with clamps altogether and use a lot of screws to pull the panels together with an even pressure over the entire surface:

    Again, based on the previous box I used a very thin smear of the dreaded, but very effective Wurth saBesto to minimize excess being squeezed out all over the shop. It’s hard stuff to clean off, so best to use no more than is needed. The end result after removing the tiny excess:

    The side benefit of this method is that even whilst curing, the baffle assembly is absolutely rock solid and so I could get on with fixing the side panels too it. By now, I’ve noticed that even if the saBesto doesn’t need the reinforcement of screws, they make the job a lot easier than trying to keep panels in line as you tighten up multiple clamps. Offer the panel up, drill a 2mm pilot hole and a No6 x 1.25″ screw nips it up nice and tight exactly where you want it, without the panel sliding around on the glue. Lurvely. The clamps are still used just to make sure the corners don’t ‘spread’ whilst curing, but only two were needed rather than four:

    A lttle detail to show that the extra time spent measuring, adjusting, remaeasuring and adjusting again makes for a really tight fit when you do commit to glue and screws. Planing/sanding a little extra wood off isn’t too bad, but adding/filling a panel cut even a millimetre too short is a real sod:

    A couple of pictures of the internal brace, which Adam posted pictures of me drilling on the pillar drill earlier. Already having the round-over bits for the plunge router, it was too much to resist ‘airflowing’ the cutouts. If nothing else, it looks lovely even if in reality it will make little, if any difference to the performance. Here’s the internal brace presenting it’s best side: 😉

    Running out of screws stopped play for the evening, so that’s where we are for now. As I rebated the screw heads in, I realised that I will need to fill the rebate holes, plus any other dings and chunks where the edge of the MDF is less than totally clever. Does anybody have any suggestions for a top notch product for the job?


    russ.will on #6093

    Tonight, I decided to unscrew the top and bottom panels that had been effectively used as jigs whilst the side panels cured onto the baffle, so I could re-use the eight screws to glue and screw the centre brace.

    I wanted to fit the brace as it would add stability to the structure and allow me to offer up the driver and drill holes on the sub baffle to fit it. I decided this would be easier as I I could stand the box on end and still have two open sides through which I could reach to position and hold things in place whilst holding the t-nuts in place until they had fully bitten into the MDF.

    I may yet apply some adhesive to the the t-nuts, but they seem to have bitten tight exactly where they need to be. I used the driver (sans gasket) as a jig to drill 2mm pilot holes, then after the driver was removed I redrilled with a 5mm bit. This is a tight fit for the 5mm bolts, but I see no reason to remove more material from the baffle than is necassary. Once drilled, I replaced the driver, pushed the bolts though (with a bit of force) and finger tightened the t-nuts as far as I could. Then, whilst gripping the t-nuts with pliers, I torqued the bolts up until the t-nut was nearly flush with the rear of the sub baffle. The t-nuts seem remarkably secure, but paranoia means I may yet apply a dap of saBesto to each anyway.

    So, Pictures. First a couple of views of the secured driver and brace from behind:

    A view of a t-nut screwed up tight with the 35mm bolt. 35mm seems about right given the 18mm of MDF, thickness of the driver basket and gasket:

    Driver porn:

    View from the front:

    And finally, a close up showing the slight extra depth allowed to recess the the driver a little deeper than the gasket. The idea is that after applying an extra millimetre ot two of veneer to the baffle, I’ll use a round-over bit on the router to radius the edge to ‘exactly the front edge of the gasket for a really neat finish, even if it’ll make no difference what-so-ever to the sound of the sub:

    I must remind myself to let some magnets into the baffle before I veneer, so I can have invisible fixings for the grill which I’ve yet to figure out.


    russ.will on #6094

    So, that’s things up to date as of a few hours ago. I’m now sitting here at 12:40am (our time) at the thick end of two excellent bottles of wine (an ’07 Rioja followed by an equally deep ’06 Syrah) with no plans except to buy a lot more screws that I had originally though necassary.

    Any suggestions would be totally appreciated even if this box is a little far along the line to change right now. However, MDF is cheap and I’m enjoying this so much I’m pretty much open to anything should anybody have any better ideas for the next box. Just don’t suggest ported, because if you remeber post #1 in this thread, we Eurpoeans don’t have the wide open spaces to loose huge subs in that you, our colonial friends do! 😀


    simon5 on #6095

    It looks nice, you are great woodworkers it seems ! 😀

    I hope you will like the sound of that 115 liters box… I think you will ! 8)

    Maybe the next step will be to buy some sound treatment for your home theater room ? :mrgreen:

    russ.will on #6096

    Thanks Simon. I don’t think the pictures make it clear how much the second box’s build is improved compared to the first. I hope I prefer the sound of the 115L cabinet, because if I did prefer the 65L job, I’d have to build two – the origi nal is a real dog!

    Quick question; When I ordered the drivers, I thought I selected the 4 Ohm option and yet each driver has two sets of terminal like I might expect (with my limited experience) if I were looking at a dual voice coil. What am I looking at?



    simon5 on #6097

    Good question… do you have a multimeter handy ? That would be the quickest way to check…

    I will try to ask John in the mean time !

    simon5 on #6098

    I managed to reach John.

    John told me that he only does dual coil woofers now because that improved the reliability a bit and it’s also preferred by most customers.

    Your woofers are dual 2 ohms, so in series it is single 4 ohms… 😀

    russ.will on #6099

    Sorry; A bit sluggish on the updates. I did indeed multimeter the coils once I remembered I actually had a meter and of course quickly discovered what should have been obvious – thery’re dual 1.6Ω coils giving 3.2Ω in series.

    I finished the box above and about a week later started on the second one. It took six hours from beginning to end. Amazing how much quicker it is when you don’t have to think what to do next and just get on with it.

    I’ll get to the sound of them later as after about a week of sheer unmitigated joy, I had to move my new babies out so a leviathan of a THX Ultra2 sub could be moved in for review purposes. 🙁


Viewing 15 posts - 1 through 15 (of 26 total)

You must be logged in to reply to this topic.