Because I am never one to be content (a Head-fier's curse), I thought I would make some notes about my new SA5000 and the latest round of tinkering.
When the SA5k arrived I thought that the sound was how I remembered the SA3k's sounding before I stuffed them full of blu-tac - notably, lighting fast, with loose bass and a slightly plasticky timbre. I opened up the 5k's and stuffed the baffle ring, this time with a denser material: a sound damping material I had found at Jaycar, similar to paxmate / dynamat, a 1.5mm thick layer of butyl rubber with a 3.5mm layer of foam on top. The material seems to work reasonably well as a sound damper / vibration absorber and the butyl certainly seems denser than blu-tac. The additional mass damping helped tighten up the bass, though the effect was, as always subtle. (I verified that it did make a difference though by doing one side and then listening for any changes first.
I tried pad swapping with the M50 pads again, with similar effect. However after a few comparisons with the stock leather earpads I found that the cloth layer that the pads add over the driver do indeed attentuate the treble, it also affects soundstage placement quite negatively. I assume this is because besides just reducing the treble, the cloth layer scatters and diffuses the treble in different directions, causing small delays in the time required for the high frequencies to reach the ear and thus smearing the transients. So I decided to stick with the stock leather earpads, which are frankly a little more breathable anyway.
I still found the timbre a little off. Following on from the dimpled blu-tac I added to the side of the baffle where the driver was angled (inspired by the HD800 experiments and the idea that the driver cup may be acting as a reflective surface and causing resonances), I added a strip of the sound dampening material to that curved wall and again verified that there was a change to the sound - the treble became a little more subdued and cohesive. Again, I stress, it was a subtle change but it was there.
Then I got a little paranoid about the hard edge that the strip of damping material had, and also wondered if the curve / exposed top of the baffle also contributed any ringing. So I fashioned a strip that would curve around the entire section.

If you are wondering why the strip has a slightly complex shape, its both to allow the material to curve around the lip, but also to allow access to the crew holes, and also to avoid blocking the vents at the bottom of the driver. It looks a little odd, but once the cups are on, it actually looks subtle:

Again I compared one side with the larger foam strip to the other, and again heard a subtle difference. The total effect is that the SA5000 sounds more natural without reducing (and possibly improving) the level of actual detail. Now, I'm not going to put it down completely to the material eliminating any reflections from the plastic, since a strip this size is doing a few things based on my educated(?) guesses:
1. Reducing the potential for plastic reflections from the exposed plastic right next to the driver by absorbing / diffusing them.
2. Providing additional mass damping to the baffle.
3. Causing the earpads to rise slightly on one side, creating a natural angle that should improve soundstaging.
I think that most of these effects (besides the mass damping as the butyl rubber material is quite dense) could be replicated just by using a piece of felt as suggested in the HD800 experiments. Since this is an easy enough and reversible thing to do on the baffle side (though be careful as any adhesive that sticks to the existing felt ring around the driver may pull that felt ring off when removed) I hope my findings encourage someone else to try this and tell me I'm not just hearing things. :S
Edited by a_recording - 5/1/12 at 7:05am