xdaniel

Well, something's whirred back to life for the first time in ages yesterday... Alright, so what exactly do we have here? It's an IBM PS/2 Model 55 SX, from the looks of it in its stock configuration, or very close to it. The exact specs -- as far as I can tell anyway, more on that later -- are as follows: 16 MHz Intel 80386SX No FPU (does have a socket for one) 4 MB RAM (2x 2 MB) 60 MB ESDI hard disk 1.44 MB floppy drive Micro Channel architecture VGA, serial, parallel & PS/2 ports on-board It turns on, counts the memory (up to 3968 KB, because 128 KB video RAM, I assume?) and... gets stuck with errors 161 and 163. "Well, no big deal! Let's just reset the BIOS settings to their defaults and see what we get!" ... is probably what I would say, if I wasn't already at least slightly familiar with these machines. Errors 161 and 163 respectively mean "CMOS configuration lost" and "CMOS error, clock not updating", which indicates that the on-board Dallas clock chip, which can be seen on the last photo above with the cable tie around it, has failed. The big problem with these is that these Dallas chips don't just contain the circuitry for the clock, but also its crystal and, worst of all, the battery -- i.e. unlike most, if not all modern PCs out there, you can't fix CMOS issues by just replacing a small coin cell battery on the board. And this battery is what usually fails with these chips. Now, what's even worse with these PS/2s is that they won't boot when faced with errors 161 and 163. In that case, you can only boot it off of its Reference Disk, which is basically the BIOS setup and diagnostics in one. So I looked online for an image of the disk, which was much less painful than I had imagined, wrote that to a floppy disk using my ThinkPad 380ED, then put the disk into the PS/2 on boot, and... the drive started spinning, reading the disk, and eventually I was greeted by a white-on-blue IBM logo on the screen. The software detected the broken CMOS configuration, proceeded to automatically configure the system, then asked me to press Enter to restart. ...which still didn't fix the errors I was getting, so I assume that, at the very least, the battery inside that Dallas chip is as dead as it can be. For this battery issue, there's two courses of action: First, get a replacement chip off of eBay, most likely a fake from China, as apparently neither this nor other compatible models of the chip are still being manufactured officially. Or second, literally hack up the existing chip to expose the battery contacts, break them, solder on an external battery holder and attach that to the chip's package. I opted for the former, for now at least, hoping that the chip from China is at least a new, good quality fake, and not just a used and possibly already dead original. And that's... pretty much where I am with this machine right now. I'm waiting for the replacement RTC to arrive before I look for and purchase any additional parts for the system, although there's not much you can really get for it anyway -- Micro Channel cards are rather expensive, and the only one I might want is an ethernet card... which would have BNC connectors, which would mean I need some kind of adapter or hub with both connection types to connect it to my modern RJ45 network, which is a pain. So for data exchange, I'll probably just try to use the null modem cable I've used with my Amiga 500 before; MS-DOS has some tools for that, I believe (Interlnk/Intersvr). MCA cards aside, I suppose an FPU would be nice to have, although it probably wouldn't make much of a difference with the software I'm planning to use -- which, well, is really just MS-DOS and maybe Windows 3.11, anyway. As for what I'll be doing with this... I have no idea. It's a low-end 386SX after all, so applications and games alike won't be much fun to use on it, and if I do want or need to use some DOS or 16-bit Windows software, I still have my Pentium 100 tower, as described in my original post. I guess I'll simply be keeping it for its historic significance in IBM's bid to try and recapture the PC market after losing it to the XT/AT clones, as well as introducing technology that ended up becoming standards on many PCs for two or so decades afterwards (VGA, PS/2 mouse and keyboard ports, etc).