I completed this subwoofer a few months ago and figured I should share the results.

The goal for this design was to have a subwoofer capable of going subsonic, with excellent sound quality (ie, no boominess), and do it all on a minimal budget. After doing extensive research on the net I found what I was looking for; a transmission line. I could easily tune it to 20 Hz, it would have a slow rolloff, and it would be acoustically pleasing as well.

After running the calculations I figured I needed a tapered line that was 14 feet long. This resulted in a box that measured 10"D x 20"H x 84"L, which just so happened to fit nicely behind my couch.

The woofer is a Diesel Audio 12" dual voice coil I picked up at Fry's, but the specs are good for the $25 I paid for it. Powering this thing is an old Kenwood stereo amplifier, 150+watts RMS per channel (into 8 ohms). I have each channel going to its own voice coil, which is only 4 ohms each. I'll let you do the math... but I will tell you this, this is the second 12" woofer in this enclosure. The first one was a different brand and I managed to let the smoke out of that one.

The performance of this sub is phenominal. It doesn't come into play until 45 Hz (when my mains start to trail off), and will extend all the way down to 15 Hz. My DB meter can't measure anything below 25 Hz or so, but judging by the resonant frequency of the closet doors and the windows in the room, I get decent output to 15 Hz. It works wonders for those super low effects in movies, and at the same time, adds just enough bass for music when it requires it. If I can manage it, I will video the shaking window to show you, since, afterall, you can't hear a 15 Hz tone!

Here is another model shot with the single 12" woofer and a transparent front panel.

And here is the unfinished yet highly functional product...

And another shot...

Last pic with the couch back in its proper place...

Nice work. Solidworks?

Hi. Im interested at building something similar.
How the calculation were performed? Did you use Martin's datasheet with Mathcad? Or just thumb rules?

thanks

Yep, Solidworks and MathCad! It is difficult finding information about these things on the internet. I did all the research I could, ran tons of simulations using MathCad, and just jumped right in and built one. These things are physically HUGE, but they sound really good too. They have the response similar to a sealed enclosure, but can extend to subsonic levels if designed to do so. I'm hooked.

DAve have you tried the 4th order and 6th order BANDPASS DESIGNS.The calculators are readily available on www.diysubwoofers.org and the similar sites.I made a 8" SUBWOOFER (4th order bandpass and it sounds nice .I made it in the wall so no vibration of the panels.They occupy much less space.I have found one more thing that the set ups wherein the subwoofer cannot be seen by naked eyes sound much better and deeper than the ones with the speakers on the front , back or facing downwards.

The best commercial speakers i have heard yet are from B&O.Great sound .Greater depth and the greatest realism.

I can’t say that I’ve ever built a T/L speaker and so I set out to find Martin’s MathCAD data sheet but couldn’t find it. Apparently he took it down after commercial manufacturers started using it to build. Do you know where to find a copy?
And no I’m not going to be selling speakers

DJ

I have never tried a bandpass design. I have never been a big fan of ported enclosures in general because of there unnatural boosting effect. I guess I could easily build one so it wouldn't boost a lot, but I guess I just wanted to try something different. I have heard a lot of subwoofers, and the sad part is, the majority of them overboost the 50-60Hz range and then have terrible harmonics below that. I think bass is very subjective, different people like different amounts and types of bass. The goal of this design was to have a subwoofer that would be relatively flat, response wise, and the rolloff would be very gradual.

I would love to hear an in-wall sub setup. As it stands right now, being in a rental, I don't think the land lords would appreciate me tearing into the wall to install a badass subwoofer!

Ironic, this, since a properly tuned ported enclosure is supposed to be the cleanest bass that you can get. The port is supposed to boost bass to match the speaker's natural rolloff at lower frequencies down to the tuning frequency. Response does drop off rapidly below that point, more rapidly than a sealed enclosure, so for deep bass extension, a sealed enclosure is better.

Though inefficient, I've seen one other idea for low bass extension that was interesting.

The idea is to use a sealed enclosure that's actually too small. The small enclosure forces a first offer rolloff of bass response (6dB/octave) at the high end that you want the subwoofer to play. You then pair this with a first order low pass filter from your lower end frequency. When you superimpose the two response curves over each other, you get a response graph that rolls off gently at both the top and bottom parts of your frequency range. The downside that it requires between 2 and 4 times the amplifier power that you would otherwise need for the same SPL levels.

I must agree that a "properly tuned" ported enclosure will sound good if designed properly. I guess I haven't heard many properly tuned store bought subwoofers, even some costing a few hundred dollars (I'm frugal ). My judgments about ported subwoofers stem from a lot of "blaaaaa" sounding subwoofers, whether it be poor material choice, poor design, poor tuning, or "all of the above". I have even designed quite a few ported enclosures, but never put one into use. I have constructed a few sealed enclosures that seem to work quite well.

I think a big contributing factor to subwoofer performance is room size, or room gain. I have not looked into finding ways to simulate this variable, but I haven't looked that hard either.

Cabin gain: I use a rule of thumb..

Take the longest dimension of the room, and divide by two. Determine frequency has that wavelength. (Example, 100Hz has a wavelength of 11.3ft at sea level.) Assume a gain of 3dB/octave slope working down from that point, and it's close.

This is what I used for my car stereo system, and it seems to pan out about right.

It's not exactly right, it's a more complex function than that, and depends a lot on the room itself, however it works well enough for audible frequencies, and is better than assuming zero room gain.

Ok, I'm hooked as well. Been wanting to replace this pathetic sub, Jensen, on my 5.1 for a while.

Daveoxide, think you can post up some exact measurements which includes the interior and holes to be cut/spaced, plus any ideas on some affordable amps/wattages you suggest would be good for this build? Would appreciate it.

Wait, so is that 3dB/octave gain from 100Hz (lets just assume that is my room gain frequency)? So does that mean I will have 9dB of gain at 25Hz? That seems like a lot to me... but then again, maybe not.

I don't think me giving out dimensions to build the exact same box is the best plan. Things like driver specs, room sizes, and WAF's can vary greatly, so I would much rather give you the "tools" to build one to your own specs and WAF . And I am more than willing to help you with any problems you encounter. I made a lot of sacrifices for my build, and you may or may not have to do the same thing, so I think me helping you design one that fits your needs is better than trying to make my design fit your needs

The first "test enclosure" I built was made of that cardboard tube used for concrete pillar molds. It surprisingly worked pretty well, but was a bit too short ultimately.

To calculate the length of your transmission line just take the speed of sound in air (1130 ft/s) and divide that by your desired "tuned" frequency (usually Fs of the speaker, but can vary above or below that value slightly) and then divide that total by 4 (since we are dealing with 1/4 wave lengths). The resultant value is the length, in feet, of the transmission line needed for that frequency.

Mine is:
L = 1130/(4*Fs) = 1130/(4*20) = 14.125 feet

As you can see, a VERY long enclosure is needed. That is why I folded mine, and even then, the thing is huge!

I will upload the information/programs that I used to design my enclosure, to my server, and pm you the link.

Oh, and as far as amps go (the most expensive part of this project), I used a dual voice coil woofer, so I picked up a used stereo amplifier (not an integrated amp or receiver, just a plain old power amp) and wired one channel to each voice coil. PLENTY of power setup this way. Another option would be to purchase a typical sub plate amp if you wanted to use a single voice coil sub and get the most bang for your buck.