My IBM RS/6000 Power PC AIX Unix Machine

As readers of my bLog will no doubt know, I'm a big fan of the Vintage Computer Federation Forums.  The community there has helped me with several projects that I would not have been able to complete on my own. I recently saw a post from a gentleman who had listed four different Unix workstations.  I reached out, and he generously agreed to pass on an IBM RS/6000 machine, a big thanks to Peter M. for this machine.

First things first, Peter let me know that there were a couple different versions of IBM AIX which were not yet archived including the first version of AIX that can run on this box, IBM AIX 4.3.3, and also IBM AIX 5L V5.1, so those are online now for anyone else who may need them.  There are earlier and later versions of AIX already archived, and there's still quite a bit of software avaliable online including a person who is building GNU tools for the OS.

This machine is a Power3 computer.  It's a single core PowerPC based processor from about year 2000.  Specifically, this is a 44P Model 170 at 333MHz, 1MB L2 Cache and with 512MB RAM.  This machine was assembled in the USA at the IBM manufacturing facility in the Rochester, MN factory.  I've been in that facility many years ago so this machine is a part of Minnesota Computer History.

It was reported that the machine had not been used for several years and so as usual I started in with a light dusting and cleaning.  I didn't have to take the machine far apart but I had been warned that the CPU daughter card would come loose and prevent the machine from starting, so I addressed that right away.  The RAM also needed to be re-seated, the battery reinstalled, and the GPU reinstalled.  Thankfully, all the hardware manuals for this machine are still online at the fantastic IBMFiles.com website, so getting through the error codes on the front panel was pretty straightforward.  If the machine stops booting it's quite easy to look up the error codes where it got hung up, much like the POST codes on a PC.

It was interesting to install the oldest version of the OS that would run on this machine, version 4.3.3, so I did that and looked around a bit at that but eventually I decided that the last version would be more useful, so I went looking for AIX version 5.3, and was able to find disk images for that online.  I burned the 8 CD set and started the install upgrading over the top of a 5.1 install.  That worked quite well and the machine is up and running with networking...

I'm happy to have the machine cleaned up reloaded and up and running.  This completes the first phase of work on this machine, but I look forward to spending more time getting software tools loaded onto the machine, doing some more networking with it, and learning more about this era of AIX.

If you have one of these machines, or any stories to share please feel free to comment below, or drop me an email, my email address is in my profile.

Compaq Portable III Restoration

 One of my contacts at Free Geek recently reached out, he had a computer that had been on my wish list for a while, a Compaq Portable III.  I already have a Compaq Portable I (which also was a project computer from Free Geek) and Compaq Portable II, and I really like these early Compaq machines because of the story of the company.  If you haven't read about it, I recommend Rod Canion's book "Open".

Anyway, this repair story starts with a water damaged Compaq Portable III.  It was quite dirty and since I'd been told it was water damaged I had quite low expectations, purchasing it as a parts machine for a future repair.  When I got it home, I immediately got started with cleaning since it was so dirty I didn't even want to touch it.  Here's a picture from "Before"...

Definitely too gross to work on.  After cleaning for a while I started to take the machine apart.  I was pleasantly surprised.  The machine was very dirty and dusty inside as well, but there were no signs of water inside.

The next step was a Power On test. I could hear some clicking from the Power Supply, but no fans, no backlight from the display or any other signs of life.  Then, there was a snapping sound and smoke!  I suspected a capacitor.  Testing with the machine off, there was a short on the 5 Volt power rail, which for this era of machines often indicates a bad Tantalum capacitor.  I was able to find the blow tantalum and replace it, and as soon as that was done the machine booted.  I was glad to see that the Plasma Display appears to be intact as this is the unique new feature that really sets this machine apart from the Compaq Portable II which still had a CRT, and the next generation from Compaq which had a more normal LCD display.

The system had the following errors, Keyboard Error, Disk Controller Error, and Time of Day / Clock Error.  The keyboard error and Disk Controller Errors were easy to resolve because these peripherals were simple disconnected.  I cleaned the 1.2MB 5 1/4" Floppy disk drive with a cleaning floppy and I was able to boot the machine from a DOS disk.

I also discovered that the Hard Disk in the unit had been disconnected.  When that was reconnected, I discovered that the disk was blank with only a few DOS files left on the machine.  I tried re-installing a newer DOS, but I was not able too do so, there are write errors when formating the drive, so I'm disconnecting it and replacing it with an SD to IDE adapter solution that I've used in some other vintage machines.  Based on this other helpful bLog article, I set the system up for a Type 42 hard drive which provides just over 500 MB of disk space for this machine on a 1 GB SD card.

Fixing the clock error proved to be the most difficult part of this repair.  This is important for these early Compaq machines because they do not have the BIOS setup program in ROM, you have to boot up from the Compaq diagnostics floppy every time the computer looses power and that make the machine alot less usable.  Apparently there were two different clock circuit options for the 286 version of this machine.  One option is the Dallas Clock chip, which are still being made today, but that is not the option this machine had.  This machine has a rechargeable battery which had been removed from this computer probably due to concerns about leakage.  I tried jury rigging a battery holder for a CR2032 battery into this space where the rechargeable battery was but I made an error in connecting that and my modifications were a failure.  After doing some more research online I found this fantastic article on MinusZeroDegrees.net.  This article contains enough detail that I thought I would try to switch my board over from the unobtanium battery pack too the Dallas clock chip circuit configuration.  I was able to make those modifications, and now I have a working battery backed up clock and drive settings!

I installed MS-DOS 6.22 on this machine, and copied many of my commonly used utilities over to the machine.  That wraps up the repairs on the Compaq Portable III for now.  It needs a bit more cleaning, and perhaps some day I'll find a RAM expansion for it, but for now I'm quite happy to have one of these in the collection.

Please reach out in the comments or via email (My email address is on my profile) if you have any thoughts about the Compaq Portables.

Trash to Treasure - Osbourne 1a Computer

 The Osbourne One is widely regarded as the first portable computer.  It was released in 1981, and although it didn't include a battery, it was advertized that it could fit under the seat of an airplane.  I have a soft spot for CP/M machines and so I've had the Osbourne 1 on my wish list for a couple of years.

Last year my friend Steve gave me a "fixer upper" Osbourne Executive.  That was an interesting machine, too, and this example was is really good cosmetic shape.  I spent quite a few hours restoring the machine, working on the CRT alignment, cleaning and lubricating the floppy disk drives, making boot media, and troubleshooting the main board.  But ultimately, I was not able to get it booting.  I isolated the fault to the main logic board, I think there is something wrong with the floppy disk controller because the machine can move the drive heads and step to tracks, but when it starts to boot, it reads in track zero, steps the head to the next track, then stops responding.  The machine behaves the same way with a Go-Tek drive emulator, so that's why I'm pretty sure the fault lies with the main logic board.  So I had put that repair on hold and was going to come back too it.

However, a more interesting option presented itself a couple of weeks ago.  I was shopping at Free Geek Twin Cities as I often do and they mentioned that an Osbourne had just come in.  It was in really rough shape, quite yellowed, and keys missing from the keyboard, but it appeared to be booting and asking for a boot floppy.  It wouldn't boot from the disk that had been left in the drive, but in my experience almost all of these drives from the early 80s need cleaning and lubrication before they can work correctly.  So, they offered me a good price on the Osbourne as a project machine and I purchased it...

The first order of business on a restoration like this was for me to get the machine booting.  I started by making some "known good" boot disks using my SuperCard Pro disk imaging setup.  I think I've covered this workflow before, but my current workflow for making boot disks for new to me machines is to find disk iamges on Dave Dunfield's site here.  I then use the fantastic tool HxE Floppy Emulator Software to convert the IMD disk image format to the SCP flux format used by my disk imaging hardware.  Then, I use the SuperCard Pro to write out the disks to hardware.

One hack that I've learned on the hardware side with these CP/M Machines is that since they must boot off Drive A it will have alot more wear and tear on it.  It makes sense to swap Drive A and Drive B as well as service both drives.  In this case, the drives mechanisms were from two different vendors but the control boards were the same.  The Osbourne 1 has power going to the floppy drive over the ribbon cables, this was not the case for the Osbourne Executive.  It turns out that all you need to do to tell which drive is the A drive is swap the termination resistor, there are no other jumpers on the drive controller board, after this the machine was booting!

Once I got the machine booting, it was time to swap over the external case parts from the Executive.  This article explains the differences between the very early Osbourne One computers and the 1a model that I have here.  I was quite fortunate to have purchased the 1a, not realizing at the time, but the case is exactly the same shape as the Osbourne Executive.  The front bezels are unique, so I cleaned up the front bezel of the Osbourne One with soapy water.  The Executive had a bulge under the handle in order to accommodate a fan, but many people feel the machine should have had a fan so I'm happy to have that as part of my Osbourne One even if it's a slight anachronism.

Another surprising item was that although the keyboards look identical and have the same connector they do not appear to be compatible with one another (I guess the analogy would be the XT and AT keyboards with the same connector but different scan codes). So, I removed all the yellowed keycaps from the Osbourne 1 computer and swapped the keyboard over into the Executive Housing.

The machine turned out lovely, I'm very happy with how it looks...

From the sides it is even more of a contrast as the old case was badly yellowed.

I'm glad to have the machine booting and running, but I would also like to add a Go-Tek with Flash-Floppy firmware.  I'm not quite sure how to do that with the non standard drive wiring on the Osbourne One, but that will be a project for another day.

Hackers and the PiPD-10

I've previously written about using simulation and buying replica kits as a more attainable way of getting hands-on time with retro computers that are rare.  This article is two years old (as of this writing), and since then I've built another PPD-8 like system and Oscar Vermeulen's PiPD-11 (in Mid 2023).  I still think these kits are a great way to learn about old systems.  The learning continues with Oscars newest kit, the PiPD-10.

As soon as I heard about the PiDP-10 project I was excited.  I've known something of the original Hacker culture since reading the Jargon file online in the early 2000s and I've had a copy of the "New Hackers Dictionary" since then.  My college, the Milwaukee School of Engineering had Digital Equipment Corporation VAX machines when I was there in the 1990s.  I'm sure that without being conscious of it I was exposed too and adopted a midwestern version of Hacker culture as it existed there in the late 1990s.  

I've been anticipating Oscar's new machine for quite a while.  During the wait I read Steven Levy's book "Hackers".  This is a book that describes longer periods of the history of computing, but it has chapters focused on the people and the culture of the MIT AI Lab in the early days of computing in the 1960s that I had not read much about before.  This is the time period of the PDP10.  Even though the ITS system did not become prominent and is not a direct predecessor to the Operating Systems of today, so many of the people who worked on it and the ideas behind it have been incorporated into today's software.

But enough background and back to modern times.  I got into the PiDP10 Google Group for conversation about PiDP-10 and I asked to be added to Oscar's "expression of interest" list as soon as the Beta test period ended.  My PiDP10 Kit arrived in mid April 2024.  I build the kit on the first weekend after it arrived.  The build went pretty well.  The biggest challenge that I had was damage around a trace that I caused by not being careful with a component, but once I repaired the broken trace all the LEDs worked as expected.  It was a long build but well documented as usual for one of Oscars kits.  The PiPD10 looks great and since it has some graphical elements I'm going to keep it in a place of honor on the shelf above my desk where I can also keep it plugged into a Monitor, Keyboard, and Mouse...

The next step for me after building the kit was a challenge.  Not knowing anything about ITS I really needed the manual to make progress.  At one point I'd thought the system wasn't coming up because I'd corrupted the file systems so I re-installed a fresh image on the Pi and went through the entire install procedure again,  But that was not the root issue, I just didn't know how to bring up ITS.

Oscar published a draft version of the manual in the PiDP10 Google Group shortly after my kit was built and with that I was able to bring up ITS on the simulated hardware and begin a bit of happy hacking!  I haven't had much time with the system yet, but having it convenient at my desk will make it easy to go back and explore what it was like to use a computer back in an era when having computer access was not a part of normal life for people.  I'm looking forward to that!

Let me know in the comments, or send an email if you have memories or a connection to Hacker culture.  Also let me know if you have suggested about the ITS or TOPS operating systems and things that I should try on the PiPD10.  Thanks!

A Tandy 4000 Computer Repair

Adrian Black recently posted a video about a Tandy 4000 computer, an early 386 machine.  I was excited to see this because (as he has mentioned in his video) you don't see alot about these machines online and I have one.  One of the commenters on his video said that these are rare because the machine was only sold through the specialized Tandy computer stores and not at regular Radio Shack stores like the Tandy 1000 range was.

My Tandy 4000 came from another collector here in town last year.  He sold me the machine (at a very good price) because he knew that it had some blown capacitors and that I enjoy doing repairs. I hadn't blogged about it at the time because this was a somewhat routine repair, but if there is interest from Adrian's video I thought I should share my repair experience.

When I got the machine home and began troubleshooting I was able to identify a couple of bad Tantalum Capacitors (which was also a problem on Adrian's machine).  After I replaced those, the motherboard worked!  However, when I plugged everything in again, it stopped working, it turns out the Seagate ST-251 drive in my machine did not work, it kept the machine from powering up.  It's possible this drive also has some blown capacitors.  Thankfully at this point I had a spare ST-251 that I'd repaired after another project ended.  I was able to install that ST-251 into this machine and set the failed one aside.

The next challenge was a cosmetic one.  The faceplate / bezel of the 3 1/2" drive was broken off from my machine and it was nowhere to be found.  I removed the drive that came with the Tandy and replaced it with a beige 1.44 MB drive and bezel kit from Free Geek Twin Cities.  It looks quite nice, if maybe not quite exactly right...

I was not able to find the correct settings for the EGA card in the machine and the display was showing odd color artifacts so I ended up installing a VGA card.  I was also able to upgrade the RAM to 8MB which would have been a very expensive upgrade in 1988.

As with many early systems the BIOS setup program for this machine is not in ROM.  I tried GSETUP as I already have that on a boot disk but that didn't work.  I was able to find the Tandy 4000 specific setup program online here.  I created a DOS 5 boot disk with it and I'll be keeping that with this machine.  I used the Tandy setup program to set the drive type to 39 for the ST-251.  Another quirk was that I had to manually enter the RAM size, as 7168, the setup program did not pick up the RAM amount automatically.  Odd, but it works and boots without any faults.

I'm happy with how this repair turned out and glad to have this machine as a part of my collection!

A Kaypro II from 1982

My latested repair story is about a Kaypro II from 1982.  The Kaypro machines have such an odd numbering scheme.  They used Arabic numerals and Roman numerals to refer to different computers, it is so confusing that collectors have taken to referring to them by the year of their release.  The Kaypro II 82 is a different computer from the Kaypro 2 84 released two years later.  There is so much confusion caused by this so many articles online contain inaccuracies in describing the features and functions of the machines.

I've had a Kaypro 2 84 for several years, and acquired a Kaypro 1 from 1986 last year, but I've long wanted a Kaypro II 82 both for historical reasons, because it's the first of an iconic line of CP/M computers, and also for nostalgic reasons because my uncle who bought me my first computer bought one of these machines for my Grandpa and Grandma.  I knew from day one that Grandpa and Grandma's machine was a much more serious tool than the one that I had, and I thought it has an amazing design, it looked like a "real computer".

This Kaypro II 82 came from Free Geek Twin Cities via my friend Chris Gallaty.  Special thanks to Chris for getting this machine when he did and for passing it on to me.  The machine came with a note "Powers On - Keyboard Cable Needs Simple Repair".  One of the basic rules of troubleshooting is that you should not take other people's troubleshooting work at face value, you always want to do that work yourself, but I was hopeful that the label was correct.

The plug on the keyboard port had been smashed in and that repair did need to be done so I started out with the obvious repair first.  Chris has also given me a 4 pin modular jack from a phone handset.  Once I got inside the Kaypro the jack inside had taken a pretty hard hit and was beyond repair, so I needed the new part.  I soldered the replacement in and hot glued it on the inside to improve the support.  It feels quite solid.

I was able to drag out another working Kaypro machine and test the keyboard.  It was working as expected!

Next, I powered up the Kaypro II machine.  The CRT came up, but the normal prompt that appears on power up was not appearing, normally the machine would have a prompt that says 

*Kaypro II*

Please Insert a Disk

This message appears before booting and so a working floppy drive is not needed to get this message, just a working power supply, main board, and CRT.  I checked the power supply voltages and they were OK.  On the CRT I did not see any raster pattern when the brightness was turned up, or after adjusting the knobs.  I did some basic troubleshooting on the mainboard, and ended up moving the main board over into a different Kaypro.  I could see that it was outputting video, so that isolated the issue to the CRT analog circuit.  This was discouraging because I'm not the best at CRT repair, but I started doing some reading and found this article on Vintage Computer Forums.  I normally don't like doing "Shotgun" repairs, but I replaced as many of the caps as I could from my stock of Capacitors and that helped, now the image looked like this...

There was raster and the CRT was displaying text!  Also, I cleaned the disk drives with a floppy cleaning disk, and I made SS/DD Boot Disks for the Kaypro II.  The machine booted from the floppy!

The non-centered but visible image was evidence that I needed to complete the recap so I captured the rest of the capacitor sizes and put in an order on DigiKey.  When the caps arrived and I had time to get back to repairs on this machine I identified a large capacitor that was totally dead, so once the recap was complete, the image looked great and I completed the CRT adjustments.  I now have a fully functional Kaypro II machine...

I feel quite fortunate to have one of these systems, it's been on my wish list for quite some time.  My next project for the Kaypro II 82 will be make some more floppy disks.  I'm also thinking about swapping one of the drives for a Go-Tek as it would be really cool to be able to boot and run from Flash Storage.

Bargain Retro Computer Fixer-Uppers

It is quite common for me and my fellow retro computer collectors to complain about the high prices that Retro computer hardware gets on eBay, but there are still good deals to be had in local markets from people who are doing cleanups.  Personally I use Craigslist and Facebook Marketplace.  I find many of these deals as referrals from my network of friends who know I'm looking for these machines.  It's good to have realistic expectations going in, as most of these lower price computers will require repairs of some kind to bring them back to use.  

Recently I've been able to acquire the following small lots:

$70 for an Apple IIe and a tote full of retro PC software and hardware (Apple IIe was advertised as untested, and it had broken keys)

$50 for an Apple Macintosh Plus (Tested at the store before purchase, boots to ? icon)

$80 for an original Mac and a Macintosh Plus (advertised as not working, and without a keyboard, mouse, or other hardware)

None of these computers were advertised as working.  When I see a local ad like this, I look at the other items that this seller has listed.  If they have alot of Retro computer stuff for sale that is working then I'm less likely to believe that items are "untested", tested and not working is more likely.  But sometimes the seller will indicate that they don't have a monitor to test and I find this explanation more believable.  

The Apple IIe was listed this way, stating that the seller didn't have a monitor to test with and so it was sold untested.  Apple IIes are know to be quite reliable (unlike Commodore 64s of the era), and so when I got it home and connected to an Apple II compatible monitor, it fired right up.  I did have to replace a RIFA cap. in the power supply, and I had to remove the keyboard and replace two key switches and key caps.  This was something I knew would be needed going into the purchase from the sellers pictures and description.  I have fixed Apple keyboards before with parts I purchased from an eBay seller that specializes in Apple keyboard parts.  After the keyboard repair the machine was complete and fully functional.

The first Macintosh Plus was purchased because I have several of the early compact Macs and I wanted to have another computer of this type to use as a Parts machine.  This Mac Plus was working at the store, but I ended up spending time reflowing many of the solder joints on the Analog board to get it starting more reliable.  I also took the floppy disk drive out and lubricated and tested that.  This machine is fully working and I stored it with my other compact macs.

The original Mac and Mac Plus lot was not a purchase I had intended to make, but a friend tipped me off and since original Macs don't come up very often at a good price I gave the listing a good look.  The case was clearly in rough condition but I could see from looking at the ports on the back that the logic board was still inside.  So I made an appointment to buy the systems,  They looked as advertised so I paid for them and brought them home.  

At home, I confirmed that both of the systems were not working (the same symptoms the seller had indicated) and I got to work on the original Mac.  I was hoping that there would be enough parts to make at least one working machine from the two.  It turns out that the original mac wasn't booting because it has a memory expansion board that was not properly seated.  I removed the logic board, re-seated the card, and the machine came to life!

The second Mac Plus was more difficult to troubleshoot.  I have narrowed the problem down to the Aanlog board, but there are no obvious burn marks or leaking caps.  I have put this repair on hold for now and will store the Mac Plus with my other parts machines.  Even without getting this final Mac Plus working, this purchase was still a bargain, but I am sure that I'll come back to the Mac Plus at some point.

So, if you enjoy repairing retro computers there are still many bargains to be had on these machines, it just requires patience and a willingness to accept machines that may have some damage.  For me, I very much enjoy the hands on experience with these machines and so I'd rather have a good solid functional machine with some blemishes than a cosmetically perfect machine that doesn't boot up and run.

Have you fixed any vintage tech. recently?  Let me know in the comments below, or feel free to send me an email and let me know what you are working on.

My First PC - Slimline Gateway 2000 486

My first PC was a Gateway 2000 Slimline 486 which I think I got in 1994.  I've been wanting to re-acquire one of these machines for quite a few years since I got back into Retro Computing, but they've become a bit rare and I've been unwilling to spend eBay prices for one.  At the end of last year I got lucky and one came up for sale at Free Geek Twin Cities as a "Project Machine".   Free Geek is transparent about the problems with these machines and they sell them at really low prices.  The notes on the machine said "Floppy Controller Might be Bad", but the price was low enough that I could afford to fail.

When I got the machine home, I did power up the machine and confirmed the symptom, but I also noticed that the entire machine was filthy.  It wasn't just normal dust either, it was a thick sticky layer so I took the machine all the way apart and cleaned the circuit board with alcohol, opened the power supply and blew it out, and the case with water.

When I got the machine back together I was able to find notes about the motherboard on TheRetroWeb.  I was even able to contribute a BIOS dump and an image to the site!  Neither the floppy drive nor the CD-ROM drive worked on this machine.  I took them apart and cleaned the heads, but I wasn't able to bring these drives back to function.  I had the exact correct floppy drive and CD-ROM replacement from another Gateway 2000 machine that I'd tried and failed to repair earlier.

The other upgrades I installed include 16MB RAM and a Compact Flash adapter that is attached to a slot cover at the back of the machine.  This is attached to the IDE on the Motherboard and configured in the BIOS as a 485 MB HDD, but of course it is faster than a spinning drive.  But the main advantage of the CF solution is that it makes it easy to move files back and forth to the PC from a modern computer.  I also had a Sound Blaster 16 Value card from another lot of retro computer parts and I installed that into this machine as well as a 3COM Ethernet NIC card.

This machine is a 486DX2-66, and my machine was only 25MHz with 4MB RAM, so this feels really fast compared to what I had back in the day.  This machine currently has Windows 95 on it, but I may go back and setup another CF card with DOS 6.22 and Windows 3.1 which would have been what I ran on this computer back in the day.

I'm quite happy with how this build turned out and enjoying the nostalgia of having a Gateway 2000 machine configured with the options I really wanted back in the day.

The 286s are Coming

I've been on a bit of a tear with 286 PCs over the last couple of months.  Actually, I'd planned one of these repairs over a year ago when I got an ALR Fast 286 at an estate sale in the summer of 2022. This was a 286 machine from 1988, and I had been able to get that machine pieced back together and working but it was experiencing lock ups.  So I had been on the lookout for a nice 286 board to use as a replacement for that machine.  I'd purchased a board that needed some work, but that was yet another project that had been sitting for months waiting for time to do board level work on that.  But in the meantime, Free Geek Twin Cites came through once again with a "Project PC" 286.  It was an incomplete machine from 1990, but it had the motherboard, RAM and a functioning Floppy Disk drive, so I was happy to get it.

Shortly after that there was another 286 Project PC at Free Geek that was also incomplete, but included a nice case, Power Supply, and most of the computer.  I purchased this machine as well, thinking that I'd have enough parts already at home to build out both PCs.  One of the oddities of this era is that sometimes a 286 just isn't quite powerful enough for higher resolution VGA graphics, so I had been torn about rebuilding the 286 with VGA or CGA/EGA graphics.  But since I got the second machine, I would no longer have to make that decision.  I could configure the faster machine from 1990 as a VGA machine and the slower machine from 1988 as a CGA/EGA machine.

I did the rebuild of my existing 286 machine first with a target date of 1990.  The "new" motherboard from Free Geek has a 16MHz 286 with 4MB of RAM on it.  I was able to re-use the external battery pack I had previous built for the machine and the ATI All-In-Wonder VGA card.  There is a nice complement of ISA cards in the machine, including a FDD/HDD IDE controller card, a Serial Parallel I/O card, a more modern 3Com 3C509 Network card in it, and I added an CF to IDE card adapter instead of a hard drive.  The CF card is 256 MB and I experimented with the BIOS settings to get 200MB usable capacity from it.  The sound card is a more modern Sound Blaster Vibra 16 CT4170 which is less expensive than a period correct Sound Blaster.

I'm very happy with how the machined turned out, and it runs fast and stable...

After getting my "best" 286 running, I started on the more basic machine targeting a build date around 1988.  This PC is a 286 at 12MHz. It has 640k RAM, and I opted to stick with a CGA / EGA card.  The HDD / FDD controller and Serial Parallel I/O cards in the machine are period correct although again I did opt to install a CF adapter in order to make it easier to load software from my modern computers.  Given the low RAM configuration I won't be installing a Sound Card or other peripherals at this time, but the machine is working good and it looks really nice...

The final project for this article is another 286 PC, a Compaq Portable II.  Thanks to fellow collector Steve H. who passed this machine to me in a lot of luggable computers.  This machine was starting up but it had an error on the screen and the floppy drives were in unknown condition.  I had heard that it would have Citizen Floppy Disk Drives and that these drives were notoriously difficult to service, but when I opened the PC, the floppy drives were made by Canon.  I cleaned and lubricate the drives and after this both drives boot up and run very well, so that ended up being a non-issue.  This generation of Compaq PCs does not have the BIOS setup program in ROM.  When I was restoring my Compaq Portable I machine back in 2020, I had already downloaded and created 360K Floppy Disks for the Compaq Diagnostic and BIOS setup programs.  One of the startup error codes was related to the clock.  Thankfully, Compaq used Tadiran batteries, which I have never seen leak.  Even cooler than that, you can still purchase a replacement cell today.  Once I had installed the new Tadiran cell I was able to go into the Compaq Diagnostics and set the time, date, and machine configuration so that there would not be any error messages at start.

Since this was the lower cost Compaq II without a hard drive I knew I was going to want to install some kind of permanent storage to make the machine more useful.  Since the machine had an IDE controller built in I was able to add another CF to IDE adapter, as before with the other 286 machines.  This one is configured in the Compaq BIOS to give about 120MB of storage, which is more than this machine ever would have shipped with.

The machine boots up and runs great, I'm very happy with how it has turned out...

I think this third machine will conclude my 286 adventures for a while.  Let me know if you have any 286s, or if you have memories from that era of computing.  What games should I load on these PCs?  What programs would make them more useful in 2023?  Leave a comment below, or feel free to reach out to me via the email address in my Bio.