I recently tried to revive an old ThinkPad 600 (with a 300 MHz mobile Pentium II processor). The system wouldn’t boot up and reported errors (173, 163) which are usually a good indication of a dead CMOS battery. In a 15-year old system, that’s not particularly surprising.
So I opened the memory cover on the bottom of the laptop, removed the CMOS battery from its housing, and pulled on the cable… only to break off the connector from the board instead of just unplugging the wires.
Unfortunately the CMOS battery connector is located right under the shell and can’t be re-attached without more or less completely dismantling the entire laptop, which I proceeded to do.
While removing the processor board, the CPU fan connector needed to be unplugged. After applying only a very moderate amount of force, it broke off too!
It should be added that while the CPU fan worked before I started disassembling the laptop, the system intermittently reported error 192 (fan error), which may or may not have been related to the barely-attached fan connector.
With the CMOS battery connector, I assumed some previous user was probably too rough. But the CPU fan had almost certainly not been touched before, and it’s about three times the size, so it shouldn’t break off easily… at all yet it did.
This must be a manufacturing defect, although it’s not clear to me how it happened. On the broken-off connectors, it is apparent that both the PCB and the SMD (Surface-Mounted Device) were slightly oxidized, as though there were a tiny amount of space between the board and the connector contacts. The connectors must have worked, but were just barely attached to the board.
The laptop was manufactured in IBM UK’s Scottish facility in Greenock in early 1999. I don’t know if the surface-mounting would have been done there or if the boards were shipped to Greenock pre-assembled.
I successfully repaired the fan connector, but there’s serious doubt that I managed to re-solder the battery connector correctly. The project will continue once a replacement CMOS battery arrives.
Pentium Chameleon
The original ThinkPad 600 is interesting in that there were both Pentium MMX and Pentium II models which shared the same system board. Since the processors required different chipsets, how is that possible?
The answer is that IBM placed the CPU on a separate processor card—together with the north bridge. This was either the 430TX in the Pentium MMX case, or the famous 440BX in the Pentium II case. The PIIX4 south bridge was the same in both cases and lived on the system board, together with the graphics controller, CardBus controller, audio chip, and the rest of the system.
The follow-on 600E models were exclusively Pentium II powered, and the final 600X series used Pentium III processors. All the while of course using the 440BX chipset, from a 233MHz Pentium II (66MHz FSB) up to a 650MHz Pentium III (100MHz FSB).
I’m not sure what’s up with the Scottish-built ThinkPads, but I had a similar experience with a Scottish-built ThinkPad 600.
The Mexican-built ThinkPad 600s are more reliable–in fact, I have one that has has been on for months and controls a CNC.
The 600 feels like the “last” of the old ThinkPads: the variety where PS2.EXE is still a real thing.
By “similar experience” you mean stuff falling off the board? I should say that I have several Scottish-made ThinkPads here, both older and newer, and the 600 is the only one with such obvious manufacturing defect. For the most part it seems to me that the unit’s condition depends only on how the previous owner(s) treated it.
What I noticed is that the 600X is the first ThinkPad (at least of the higher-end models) without diagnostics in the BIOS. Starting with the T20 I believe IBM switched to Phoenix BIOS and the whole system feels different, even though hardware-wise there’s not that much of a difference between a 600X and a T20.
Actually how do you distinguish between PS2.EXE that’s a “real thing” vs. one that isn’t?