It's been a busy year for me and so I haven't got as much retro computing done this year. That being said, I have been able to sell four computers and give away a couple more this year so I am still having success with thinning out the collection. For quite a few years I've been building a backlog of interesting broken machines. I keep this hardware out in the garage and have a document going for each of these potential projects. When I'm feeling inspired to take on a new project I can go fine one any time by looking at that folder and see what I'm interested in taking on next. I've had a bit more time as of late and I've repaired a series of PCs that I'll describe in this article.
The first repair is an interesting Zenith 386 PC that came from Free Geek Twin Cities in the spring of last year. This machine was badly rusted from water damage in and around the power supply. The Free Geek team handled this machine well, they knew that it should be powered up, but the machine included a nice matching keyboard and a full box of documentation and software so it was worth some effort to fix it up.
I started right in on this machine last year. I took the machine all apart and cleaned it up. It's model number ZBF-3340-EK, an early 386 machine built on a passive backplane. The backplane is based on ISA but extended to 32 bits. It was a very good thing that Free Geek didn't power the machine up because there were pieces of rust floating loose inside the power supply. I took all the metal pieces apart and cleaned them up. I spray painted to the rusty pieces to stop further rust, but this wasn't a full restoration, just trying to halt the deterioration enough to get some basic testing done and decide if the PC would be salvageable or not. To my great surprise after cleaning up the machine and some basic repairs, the PC was able to Post and I got video working with an old VGA card which had been paired with the machine later in life (It originally had a proprietary monitor and video card). It took quite a few hours of effort to reach this point so I set the project aside. I thought I'd come back to it soon but it ended up being a year till I came back to it.
After moving this PC around in the garage for the 3rd or 4th time I decided it was time to finish this project. I took the machine apart enough to get a look at the four drives. It had two 1.2MB 5 1/4" Floppy Disk Drives, and two Full Height MFM Hard Drives. I cleaned and tested both of the Floppy Drives. One of them was working and the other was not. I replaced the non-working drive with a 3 1/2" drive I had in my spares, and set the 5 1/4" drive aside for future repairs. Once I figured out the BIOS settings which were needed for the two Hard Drives, I was happy to see that both of them worked (after a few seek errors, which makes alot of sense if the heads had not moved for a decade or more). It's very cool to have a machine with Dual MFM drives, although I may just leave one in there and keep the second one for another machine.
Adding a clock battery was the final step. After adding the clock battery the PC still wasn't holding time and BIOS settings. It turns out there was a bit more corrosion on the board around the battery area so I replaced a few corroded components, some diodes and resistors, and I added a bodge wire for a damaged trace. After that the clock battery worked and it is holding the BIOS settings. The machine looks nice...
Due to the solid construction, large chassis, and dual full heigh MFM drives, this computer is very heavy, and it's great to have the documentation and a matching keyboard with lovely mechanical switches.
The next PC up on the bench is an XT class machine. The seller let me know that it had bad tantalum capacitors which had blown. They didn't want to do a recap on the machine so they sold it on to me. This is machine sat on my project intake queue for even longer than the Zenith. I have quite a few XT PCs so I knew this would be a PC that I wouldn't hold onto for my collection. It turns out that it had one bad tantalum (which failed short on the +12 volt rail), when I removed that, the computer came up to the BIOS Post screen. After a few minutes of operation, another tantalum failed (this time on the -12 volt rail), but again that was easily removed after some basic troubleshooting probing around with the multimeter on the board. I didn't have 4.7 uF Tantalums in stock so I ordered some of those and did the recap when those arrived.
Other than that, my main tasks were to take the machine all apart and clean it. It was quite dusty. I cleaned the mainboard with my ultrasonic cleaner. After it dried I cleaned and tested the 360K 5 1/4" Floppy Disk Drive. It was working good. Sadly, the MFM hard drive inside the machine was not working. I set that aside for further troubleshooting. The faceplates for this machine were incomplete. I was able to find a second half-height black Floppy Disk Drive to cover the gap in the front panel of the machine, and I also 3D printed a Full-Height Hard Drive blanking plate for the other drive bay, which completes the cosmetics of the front panel. The machine is fully functional, and I'm happy with how it looks...
The last PC in this group was an original PC, the IBM 5150. As readers of the Blog may know, I already have one IBM 5150 that came from a pickup of Facebook Marketplace some years back. But over the last couple of years I've been making an effort to find a second machine if I can find one at a good price. I'd been collecting bits and pieces for a second IBM PC for well over a year now and I finally came back to this project since I was already working on the XT Clone from that era.The IBM PC also had bad tantalums on the main board and that was the reason why the seller sold it to me. I was able to locate the shorted capacitor and replace it. I also tested the power supply with a load but disconnected from the PC, the power supply voltages were all in range. I was able to find a VGA card that would work in an 8-bit slot. This enabled me to do testing with a more modern monitor and I also left the original Monochrome MDA card in the machine as well, since the IBM PC does support two different types of video cards at the same time. The machine has 256K of memory (the maximum capacity on that mainboard) and it Posts and comes up as expected. The next order of business was to clean the two floppy disk drives. After cleaning they were both working. I had two cases and so I was able to put together the best cosmetic version of the case. It's got some rust spots on the top, but as a spare machine it will be just fine and I'm happy to have it in the collection...
Of these three machines, I'll be moving the first two on too new owners, either here locally or at VCF Midwest in the fall. The IBM PC 5150 will go back into storage to be used as a spare. Given how long it took me to get to these repairs I probably won't be seeking out more machines of this type to repair, but it is always gratifying to get a broken computer fixed and to have more projects off my to-do list.
If you have any memories of using a Zenith 386, an Ameridata XT PC, or an IBM 5150, please post in the comments below or drop me an email (my email address is in my profile).
No comments:
Post a Comment