Last week I rescued an old Dell Latitude CPx H600GT - a 600 MHz PIII - from the 'to be recycled' pile. Not a bad machine, but a generation short and about 1.4 GHz shy of the department's minimum for supported equipment. So, it came home with me to see if I could renovate it and get it to one of my young nieces for school use.
After a couple of hours the battery showed no charge. The system had also exhibited some flakiness, but, you know, I thought that had more to do with trying to run the thing with only 128 MB of RAM (it had been stripped, and I only had a single 128 MB laptop DIMM on hand). I dug out a different battery -- a 4.5 Amp-Hour battery from on old Dell P4 Latitude. Now, the PIII Latitudes shipped with 3.6 Amp-Hour batteries. The P4s had a heavier power adapter -- 4.5 Amps vs 3.5 A. I had gotten away with this before with a different CPx, but I knew that it put a real stress (evidenced by a lot of heat) on the power adapter. So, I thought I'd play it safe and load the new battery in after I shut the laptop down -- thinking that, by only having to charge the battery and not run the laptop at the same time, I would cut down on the electrical/heat stress on the power adapter.
Well. That wasn't the component I should have been worrying about. That off-and-on flakiness is also a sign of a Latitude with a power controller that's teetering on the edge of failure. Presented with the overload of charging the 4.5 AH battery, one of the components on the power controller daughter board failed. And I do mean failed, as in let the magic smoke out failed. I autopsied the machine this morning while the Little Guy was napping. It looks like one of the chip capacitors cracked, flamed, it blew itself clean off the board.
This is the power controller daughter-board showing where the component blew itself off the circuit board
This is the component itself - a capacitor, I think - cracked in half and badly scorched.