Where there's smoke there's... a hard disk!?!

Recently, an accquaintaince was cleaning up his house and came upon some old hardware, and asked - do I want some old hard disks? Sure! He gave me a pair of old Avid external 9 gig hard drives. Basically just 9 gig 5 1/4" FH Micropolis 50 pin SCSI drives in little-ish external boxes. Perfect for some of these old boxen I run around here. Anyway, I decided to hook them up tonight. I cabled them up to an old PC, powered everything on, and the SCSI card saw both disks. I left the machine run while I went in the next room to burn a Debian netinst CD. Five minutes later... what's that smell? I went in the next room, now slightly hazy and full of that acrid 'burnt electronics' smell. I hit the switch on the power strip and started smelling things looking for the source. Old 14" DigiView monitor? - no... Old Compaq desktop PC? - no... Avid external 9 gig SCSI hard drive? - <cough>.. yes <gasp>. I took the drive apart, expecting to find a burnt up tantalum cap in the power supply or something - nothing. Then I took the drive out (not easy, the standoffs seem to like to stay with the screws rather than the drive..), and flipped it over. There was a burnt mark on the board the size of a nickel, and the board had started to delaminate. In the center, a charred husk of a something that resembled a surface mount capacitor of some sort, a small rectangular component about the size of a standard jumper laying flat. Hmm, I haven't got anything to lose at this point, so I picked the burnt thing off the board, it had already unsoldered one side of itself, so I wiggled the other side off, and removed the surrounding debris. Morbid curiosity made me hook the drive back up. Hmm, it spins... it comes ready... I hooked it back up to the computer, and turned the computer on, and lo and behold, the SCSI controller sees the drive! I am currently running the Adaptec SCSI controller's media test function on the drive now, as I ponder why the Linux installer insists on crashing on this particular box, but I really wonder a couple things - what caused that capacitor to burn, why does the disk still work, and how long will it continue to work? I imagine that the capacitor wasn't particularly important, seeming as though the drive still runs, but what baffles me is that even with sizeable damage to the board, (it's bubbled up and delaminated in that spot) the drive works. The capacitor was near three voltage regulator type transistors on the board, so it could have been a bypass type capacitor, and not crucial to the circuit, but that doesn't explain why it caught fire. Computers never cease to amaze me. -Ian