Oof. Where to start with this one. What an anus of a project. It should have been an amazing build. Its should have been an absolute breeze, once I had it all figured out. It should have been a well planned, well executed joy of a thing. It wasn’t. Best laid plans and all that…
Right from the start of my speaker building odyssey I’d been fascinated by ‘transmission line’ or ‘t-line’ speakers. They seemed to be somewhat of a unicorn in the speaker design world. Too sensitive to room placement and tricky to tune the internal absorption and damping materials to make them a viable commercial proposition. But when done right, offering amazing performance – especially in the B.A.S.S.
The subs I built in my sound system build were rear loaded tapped horns – similar to a transmission line, but not the same. Horn speakers work on exactly the same principle as brass instruments, such as a trumpet, trombone or french horn. You have a specific ‘pipe’ length which resonates at a certain frequency – its pitch, or tuning. The tapped horn subs were tuned to 25hz. Very low B.A.S.S.
This gets very technical, very quickly. Go here and fist your mind with quarter wave theory. Fill your boots. The rabbit hole is deep. And full of science. To summarise: Sound waves – like any wave – have a wavelength. That length is measured in physical terms (cm, inches, feet, metres). The longer the wave, the lower the frequency. The shorter the wave, the higher the frequency. The greater the amplitude of the wave, the louder the sound. Crafting clever enclosures can enhance certain frequencies. Like B.A.S.S.
Stay with me… speaker drivers generate sound waves by turning electrical energy from an audio source in to physical motion. By rapidly oscillating the speaker cone, air molecules are moved in waves. Our ears register these pressure waves (relative changes in air density) as sound. All speaker drivers have a specific resonant frequency – a wavelength which the driver resonates at, which causes all sorts of problems (think opera singers shattering wine glasses with their voice). Transmission line speakers take the principle of matching a speaker driver to the correct amount of air pressure behind the cone (in this case the volumetric capacity of a shaped chamber or pipe), to help sort out resonance problems AND enhance certain frequencies. Phew. Enough of that. I made some plans…
Because my big system was too bloody loud to actually use, I figured I’d build some nice hifi speakers that I could actually listen to instead. I’d learnt from my builds to date that setting constraints was the way forward. So the mandate for this build was thus:
- Tall
- Thin
- Good looking
- Front ported
- Serviceable
- Full range
Taking those constraints – I searched for some appropriate drivers, and settled on these Pluvia 7 full range drivers.
Pretty little things. I thought if I’m going to do this right, I should properly embrace the nerd life and plan the enclosures using some nifty free software I found online. Unfortunately I cannot link to it as it is no longer available, as the author took it down for whatever reasons.
It basically helps you make the trade off between enclosure size and what frequencies you want to enhance. After much trial and error I ended up with the above frequency curve from a decently small, thin enclosure. I thought I had compensated enough to give decent bass response without needing a subwoofer – because t-lines are meant to produce phat, low bass. Turns out I hadn’t compensated anywhere near close enough. Its plain as day to me now looking at that graph… but I suppose that’s why learning curves are a thing.
Anyways, as I was embracing the nerd life, I thought I’d model the enclosures in some free 3D software (Sketchup) and really geek out. It was a great idea. If I hadn’t done this I would have messed up a crucial measurement that would have rendered the entire build completely hatstand (oh the irony..). Only problem was I had no idea how to use the software, so I had to learn it as I went. Hello learning curve…
Unfortunately I have a mostly hate/hate relationship with 3D software, so the process wasn’t much fun. But it was VERY satisfying when I finally got there.
Because I wanted this to be a super accurate, ideally replicable build I’d decided from the start I was going to get the wood CNC cut from a local supplier. There was no way I was going to get the design accurate enough with a jigsaw, a wing and a prayer.
The main reason this thing ended up so complicated and frustrating was that I was trying to be too clever. I knew that t-lines needed to be really solid to help with negating unwanted frequencies, so I’d designed everything to be double thickness. Because I thought I was smarter than I was, I made everything join together like a jigsaw puzzle to help gluing. Then I got wood (Gnurgle).
O frabjous day! callooh! callay! He chortled in his joy. What a wondrous sight it was to behold. I’d designed my very own millimetrically perfect speaker kit. And there it was, all ready to build myself. Marvel at its wonder! Bask in its glory! I’ll test out fitting it all together…
Clever me! You lucky sausage – being so smart. S.M.A.R.T. Well bloody done lad!
Hold on a minute… When I put them all together properly, everything seems ever so slightly… off. A bit… rattly. A bit… gappy. Wait a minute, there are tiny gaps, just less than a millimetre on all the joins. That’s not good. What gives?
It turns out that 12mm birch ply is not exactly 12mm birch ply. It’s 12.5mm. And someone hadn’t realised this. That someone was me. Everything was off by around 1mm. Because I’d been so ‘clever’ making all these super strong, super sturdy stepped joins it was all ever so slightly on the piss. All speakers need to be 100% sealed to prevent air leakage ruining the enclosure effects. This was 0% sealed. I’d royally banjaxed the whole thing. Kill me.
But I don’t give up easily. I knew it would be an absolute nightmare, and even with all the good intentions in the world I might not be able to save it, but I went ahead and got jiggy with my little dremel and some sanding bits. I had to shave 0.6mm off every single joining piece. What. An. Anus. HUGE one. Massive. But I puckered up and sorted it.
All those dreams of a perfectly finished, exquisite build – done for. But I still reckoned I could do it. So I ploughed on. Even if I couldn’t make it look super pretty, I could still make something that sounded great. Or so I thought. Initial tests were positive. They sounded crisp, clear and lovely… but very little B.A.S.S.
Its alright I thought. Once they’re all damped and sealed properly, they’ll come in to their own. Plough on Davey, lad, plough on…
So, 4 months after getting the wood eventually they were finished. Halle-bleedin-lujah
Now to be fair, its not all doom and gloom. They look fab – so much so my wife let me put them downstairs. Unheard of. They also sound amazing – apart from the fact there is very little low end. Which after everything I went through on this build was a massive let down. You can get bass out of them, but it overwhelms the little drivers at low to medium volumes, so its just not worth it. I ended up pairing them with a cheapy old active sub that I had kicking around and they sound peachy.
As they are downstairs, they also have to deal with two barely contained monkeys messing with them. So I had to build speaker covers to protect the drivers – which kind of kills the point of having pretty exposed drivers. But it doesn’t seem to be to the detriment of the sound
So, what have I learnt from this build?
Keep it super simple (K.I.S.S)
I can scarcely believe I’ve had to learn that lesson uh-gain… but there you have it. At least I’ve learnt a stack of things and ended up with some properly decent sounding speakers. It’s just about killed my desire to continue building speakers though, which is… well, it is what it is. That’s designing and building speakers for you ¯\_(ツ)_/¯
Dave Cropley 