Woofer enclosure bracing

The woofer enclosures badly need bracing, because they (like the Asawari Mark I) have huge side panels which will sing like a drum if they are not braced. Since this enclosure will generate only low frequencies, any bracing which moves the resonant frequency of the side panels to the KHz range will effectively block the vibrations. As per Seigfried Linkwitz, if bracing is applied in a matrix or grid manner where adjacent braces have gaps less than about 4-5 inches, then the resonant frequencies are pushed up to the KHz range. So I intended to do that.

I built the bracing frame as a separate sub-assembly out of 20mm ply (my material of choice for bracing). The pieces were assembled and then joined together with screws and Araldite, one per enclosure. They are almost three feet tall each, and are amazingly strong, rigid, and light for their size. You can lean on them and lift your feet off the ground, and they will take your weight but will not bend or tilt at all. I tried it. Quite satisfactory.

Woofer enclosure front baffle

Now for the front baffle. This, as the plans show, is made out of three sheets of MDF, each one inch thick. This makes for an inert baffle -- it feels like a solid block of wood against a rap with the knuckles.

The three sheets have the round hole cut in them for the woofer. The top-most or outermost sheet has this hole with straight sides, without chamfering. The second sheet has its hole chamfered, but the places where the screws will go in are not chamfered. This allows me more material for the screws to grip. The innermost sheet is fully chamfered, and therefore has a wider hole than the middle sheet.

The chamfering trickery I am using here allows me to get a full two inches of MDF in which my wood screws will get place to grip, to mount the driver. I have not yet begun to use T-nuts, finding them to be more of a hassle than I need. Six screws penetrating into two inches of MDF should be good enough to give me a lifelong grip on the 10" woofer.

A special router bit

This chamfering of edges of the hole is done by a specific router bit:

This is sold on eBay US by the seller ghsu2ia3 who has been selling on eBay since 2000, and operates an eBay shop called Super Carbide Tools. He has a score of more than 22,000 with 100% positive feedback. He's clearly a serious businessman. His router bits ship from McHenry, Illinois, USA. I imported a bunch of bits from him through Borderlinx.

What's special about this specific router bit? Well, it has these properties:

• It has no bearing at top or bottom
• it has a 45-degree slope and a 25mm depth (which means a 50mm top diameter)

Just try finding another router bit from somewhere else with these features. I couldn't. I will explain why this is special:

• The absence of any bearing at top and bottom means that I can plunge this bit into thick MDF when it is rotating, and I can start my routing without drilling any leader holes. All other chamfering bits I find have a bearing at the bottom, which means they can cut from their sides all right, but they cannot cut downwards.
• The combination of 25mm depth and 45-degree slope angle is pretty rare. Even if I were to accept bearings, I found just one bit with 25mm depth and 45-degree slope, and it was selling for USD 57. The 25mm depth allows me to do chamfer cuts in 25mm-thick MDF sheets (and thinner sheets if I need to someday).

Therefore, I realise that this bit is quite special, and I will need it for all midrange and woofer mounting holes in all speakers I build in future. I have located and procured a Chinese brand router which can accept 1/2"-shank router bits, so these bits work very well for me.

And I am using the same trick of recessing which I have used in the past -- the 4mm of veneer becomes one of the components to help me get the recessed fit of my drivers. If a driver (some tweeters) has a front plate thinner than 4mm, then I use a gasket behind its front plate to raise it and also give me the air-tight seal I need. This is how the Asawari tweeters were mounted, for instance.

Bracing, bracing... yet again

After I had gotten the woofer enclosure bracing framework and posted pictures to some online fora, some friends congratulated me, and one friend actually used the word "over-built", all in a very appreciative tone. My chest swelled and there was a clear swagger in my step.

After the bracing was fitted inside the enclosure with Araldite and a few screws, I tested the walls with the good old knuckle test and lost all my swagger. The bracing was not helping as much as I had hoped. It was quite unbelievable -- I had never built such a carefully designed bracing before, and I was still getting hollow sounds from the knuckle knocks.

After some thought, I realised that the side walls were not sticking to the bracing as tightly as I had wanted them to. Just Araldite on the edges of the bracing to fix it to the side walls is not of any use -- one must press the pieces together hard while the Araldite solidifies. I had not done that. In fact, I had wanted to do that but my know-it-all carpenter had ignored it. I had asked him to drill holes every six inches or so, on the side walls and put screws through them to grip the bracing tightly after he fitted the bracing and before the Araldite had time to solidify. This would have meant at least about 25 screws on each sidewall. Instead, my carpenter had fixed half a dozen screws, pretty close to the edges of the sidewalls, just to hold the bracing in place. Now there was no way to remove the bracing and re-do it -- the Araldite locked the bracing in place, making removal impossible.

I then made him drill holes every six inches as I had wanted initially, fill the holes with Araldite paste and then put screws through them and tighten. He thought it was rubbish and total overkill, but I can be as obstinate as he when the need arises. This was a hunch -- I felt that the only way to make the side walls really stick to the bracing was to do this post facto hack. These holes had to be drilled from the outside precisely along the lines where the edge of the plywood sheets of the bracing lay on the inner surface. The carpenter carefully did it. He poured Araldite, tightened the screws.

And ... it worked! After a day of the Araldite drying, I tried the knuckle test again and the side walls gave a satisfying "tup-tup" sound instead of the earlier "bonk bonk" hollow sound. I had managed to save my investment and deaden the sidewalls and work around my obstinate carpenter after all. That was one of the happiest days of my Darbari building story.

To those friends who had thought I was over-building, I say... "Think again!" :)

The upper enclosure

The upper enclosure is easier to build well, simply because it has smaller panels, therefore they turn out acoustically dead without much struggle. I am using bracing similar to what I have used for the woofer enclosure, with slightly better results.

In these photos, a vertical brace is visible with a squarish hole in its centre. This will eventually be directly behind the midrange driver. On hindsight, this is probably going to cause unwanted levels of reflection of the back wave of the midrange driver, though it is doing its job as bracing of the side walls splendidly.

One point very visible in the photos is the tweeter chamber. It is a separate sealed chamber running the full width of the top of the enclosure. Such a large tweeter chamber is not required, but I built it that way because it's easier to build than having a half-width chamber. But a sealed chamber is useful; it reduces vibration of the tweeter due to the back wave of the midrange driver. In the case of the Darbari, this issue may not be as serious as in the case of 2-way designs, because the deep bass which causes the hardest-to-manage vibrations are not going to be present in this chamber in the case of the Darbari.

After the front baffle is shaped and fitted to the upper enclosure, and the veneer sheet is glued on, this is what it looks like:

These photos clearly show the bevelled corners and the overall shape. The chopped-off corners and bevelled edges "soften" the boxy shape. One thing not very clear from these photos is that the front veneer is a much darker pattern than the ones on the rest of the surfaces (including the bevelled edges and corners). As a result, the visual impression of the boxes head-on will be communicated by the shape of the front pieces, which have chopped-off corners and therefore look a bit "rounded" if one may use the word.

Enclosures are built

These are shots before the polishing, but after the enclosures are completely built.

Drivers are fitted

The glossy PU polish on the enclosures is done. Speakers have been shifted to the living room. Drivers are fitted.

I am using Dayton Audio insulated binding posts mounted on matching mounting plates also from Dayton Audio. Both items are very well made, and are easy to assemble and join wire leads to. I am using female crimp-type QC interconnects to connect leads to the binding posts from the inside. I never depend on pure QC for a long-term connection, specially since the pollution in the Indian air makes surfaces acquire oxide and sulphate layers. So, I did soldering here too, but soldering QC to QC is a much faster and neater job than soldering bare wire to a binding post.

The speaker drivers have been mounted using absolutely stunning hex-head screws from The Audio Crafts, a Delhi-based supplier. I have no idea where else these screws are available from. The screws I am using are these copper-finish ones for the tweeter and midrange, and these larger black ones for the woofer. I would have loved it if TheAudioCrafts carried large ones in copper finish, but he doesn't. I also feel that 45mm is a better length for larger woofers, but he stops at 30mm. Whatever it may be, but the copper-finish screws make my speakers look like a million dollars.

The speakers are standing on three spikes and a metal plate. This metal plate is an 8mm-thick sheet of 304-grade stainless steel. It has been cut, drilled, tapped (for the spikes) and finished in brushed finish by Sound Foundations, an amazing boutique manufacturer of accessories for very high-end audio equipment. Ali of Sound Foundations does not make any money from his venture, I am quite certain. The spikes are manufactured by Sound Foundations, and he also supplies floor-saver discs for them. Each speaker stands on three spikes.

After listening to the speakers for a couple weeks, I decided to get some wooden wedges made to place between the upper and lower enclosures. This wedge tilts the upper enclosure forward so that the upper drivers point more directly towards the listener's head, instead of firing above it. I placed a few dozen small vibration-absorbing pads on the bottom of each upper enclosure and on the bottom surface on each wedge. This absorbs vibration and also prevents slippage of the upper enclosure.

The speakers were a real dog to work upon, because of their size and weight. Fitting the metal plates to the bottom meant up-ending the enclosure, standing it on its head, placing the plate, drilling carefully, tightening four two-inch-long steel screws, then fitting the spikes, and finally, gingerly, carefully, laying the box on its side and straightening it up. Each steel plate weighs 14kg. I suspect each channel's speakers (upper and lower combined) will exceed 100kg now.