IBM 5150

Another rusty pile of mess I dug out of a tent in a friend’s back yard… No, honestly, this machine sat in a tarpaulin tent for about twenty years before I discovered it.

It was a complete IBM 5150 with 64kB RAM and had been upgraded to act like a 5160 (20MB type 13 hard drive and Xebec controller). It had an original IBM CGA adapter as well.

The original case was unfortunately so rusty I wound up scrapping it… It was a serious shame. I kept all the guts, including the original power supply! I held onto the bits for a number of years before finally getting around to doing something with them. I knew it was going to be some work to get it going, but it was very fun and well worth it.

Starting off, I had to verify that the power supply was alright. It was the least damaged component in the entire system; even the original electrolytic capacitors passed my tests, although I still intend to replace them soon. The power supply works properly, so off to the next part… the motherboard. A good number of the bypass capacitors (see below) were shorted, so some were replaced with modern low-impedance electrolytics (I should do all of them). The CGA card also had shorted bypasses and I haven’t yet repaired it; instead I installed a spare MDA card. The inaugural power-up showed activity, including a POST error telling me there was bad RAM… what a surprise. I dug around in my things and found some 4116 RAM ICs and got it going.

I quickly decided that I was going to want more RAM (I wanted to run SpinRite on the hard drive for one), so I added sockets to try out 256kB, then threw in the towel and found a box of 41256 chips in my things and set about designing the circuit to decode the RAM out to 512kB even though the motherboard is made to go to 256kB! You can see in the photos below that it took a little bit of Kynar wire and an additional 74LS157 multiplexer in order to double the maximum RAM possible on the board. It feels kind of funny having only two banks populated and yet having double the max RAM, but that’s how it goes… It works charmingly!

Next step: Hard Drive. Would an ancient MFM hard drive that sat outdoors for twenty years be willing to work? Sure it would! It spun up smoothly with no trouble at all and started working right away. Sadly, there was definitely bitrot. SpinRite was able to bring a great deal of it back, and also reported that the media was okay for future use. I immediately imaged the drive to ensure I could save whatever time capsule goodies were on it! (I did later discover things that are, shall we say, interesting… and things which will not be mentioned here except to say they were definitely enjoyed by some business employee on coffee break.. nudge nudge, wink wink).

I bought a nice AT computer case from a friend and stuffed all of this in. Note that an AT case has closer ISA slot positioning than an XT one, so you have to be a bit creative to fit some of the card in some positions. I hope I find a real XT, or better yet, IBM case! You’ll see in the photos the keyboard plug lines up, but the cassette plug doesn’t… Shame, cause I want to use tapes on this thing!

I wiped the drive clean and installed DOS (can’t recall which version) and then mTCP and hosted a web server out on the internet for my friends to try for a day! It was SLOW, but it worked! A real website hosted straight up from a real IBM 5150!

I am sure I will add some more goodies as I explore this machine; but that’s where we’re at for now!

I recorded a Sound File (1.17MB) of the hard drive being tested by SpinRite for your enjoyment!

And onto the photos!!

The Insides!

ISA Slots

Bad Capacitor

More Capacitors

Added Sockets

Address Decoding

RAM Modification

RAM Wiring

First Try

Second Try

Trying SpinRite

Not Bad