Originally Posted by tuxedocivic
A word of warning. I built a couple downfiring 12" end tables. ... But they love to wander around. ... If I did it over again, I'd have to think hard about doing it differently. They are adequately braced!
My build is not quite done, and owe the group a write-up, but this applies here.
I vibration isolated the subwoofer in my end table sub. Single, down-firing Tempest X and it's rock solid at any power level. Here's a link to the theory
And a cross-section of the design, showing only the rails, subwoofer capsule, and support system. Legs are not visible in the diagram; it's just floating in thin air, like any good cross-section.
The idea is to create a spring-mass system with a resonant frequency below the lowest expected driving frequency. As you can see from the link, transmissability falls like a rock above resonance. They suggest a 70% margin to driving frequencies; I aimed for an octave... and assumed a 10Hz driving frequency. I targetted a 5Hz resonance empirically, by adding small springs until I got 10 cycles in 2 sec on the stopwatch.
The only tricky part is finding springs with the right spring rate and rest length. I wanted to support 85lb. with only 1/3 deflection (concerned with creap over time), but I had a lot to choose from (thanks to my employer). I varied placement to get equal deflection from all the springs, so it sits level. At this point, the only thing that happens when I send it a 20Hz sine wave is a whistle where the leads enter the box; nothing moves at all.
I expect a glass of water on the table would jiggle much less than one on the floor next to the sub, but that's what's missing - the table top has a BIG hole in it.