May 3, 2026
This story began in 2021 when I entered my computer science classroom for the first time that year and found an impressive collection of old Macintoshes:
From left to right: some eMac, the fastest iMac G3 ever made, a Performa 5260, a Mac mini G4, a mirrored drive doors Power Mac G4 (that wouldn't boot), some early Intel iMac, the Plus, a Bondi Blue G3 (I don't remember what specs), another early Intel iMac, and a later Intel iMac. The Performa was how I made floppy disks for the Plus, but only after swapping its dead hard drive with the blue G3's drive (and cracking a bunch of brittle plastic along the way).
I've always liked old technology but at that point I had never had the chance to use an old computer. I was of course most interested in the oldest one up there, an original 1986 Macintosh Plus, upgraded to 4MB of RAM, perched atop an Apple Hard Disk 20 (the version that plugs into the floppy port, not the SCSI version). I pulled it down and it powered on. I wish I had pictures of that first boot-up, or lack thereof, since I didn't put in any disks.
I took this computer home and had fun with it
on and off for over a year. It was missing a keyboard and mouse, so I bought a PS/2-to-Atari-mouse
adapter on eBay and rewired it for the Plus, then I made my own keyboard emulator using an Arduino
and mostly somebody else's code, which only kind of worked. It's still on my GitHub as mac-plus-serkey.
My final project was getting the Bad Apple!! video running on it,
and while the video ran well, combining it with the audio didn't really work, so I gave up
and put the computer back. (I also had some fun with the other ones...and wiped too many of their
hard drives. Don't do that.)
The Plus wasn't mine, so once my project attempt ended, the computer had to go back.
In April 2026, my friends and I went to an electronics warehouse run by some auction company (who set prices instead of actually auctioning). This was, frankly, an awesome discovery:
There's a Sony Watchman for $10 next to the Osborne computer
Most of the items for sale at this place were overpriced, but not what I found: a PowerBook 145B in impressively good condition for $60. The auction guy let me take some 1.4MB floppies, an ADB keyboard, and an HDI-SCSI adapter with that price (which changed $60 from a slightly high price to an extremely good price).
I got this thing home and opened it up to find the insides and display in great condition. Even the plastic screw mounts, which usually break, were fully intact. I wasn't able to find a power supply at the warehouse, so I bought a modern reproduction on eBay.
It looks thick, and it is, but it doesn't feel quite as thick as it looks.
This took a few days to ship. When it arrived, I plugged in the PowerBook, hit the power button, and...
It works, and it even boots from the hard drive! Digging around on this drive revealed that this computer was used by a handyman in the 1990s to print out invoices on a portable printer, presumably for services rendered. The drive also has (password-protected) tax forms for that person's business and some drivers for an interesting portable printer called the GCC WriteMove. Neither I nor the original owner will probably ever need the data, but I took a backup of the hard drive with Apple Disk Copy and it's resting on my computer. I already had a BlueSCSI v1 from my days with the Mac Plus, and the HDI-SCSI adapter lets me boot whatever I want hassle-free.
The floppy drive didn't work, and the battery it came with was completely dead. The computer itself was missing the rear I/O door and the right lifting foot. The rest of the computer was in great shape, including the display. The adhesive in these displays is known to break down in a process called vinegar syndrome, which causes significant but repairable damage to the display. This computer was used and stored in the dry Southwest US, which I suspect is why the screen hasn't deteriorated.
These computers were made in the 1990s, so they have 1990s capacitors, which leak all over the place. If you're lucky, they haven't done any damage, but if you're unlucky, they've destroyed a working computer. You must replace them AND clean up after them. Even though the 145B worked, the display wouldn't hold contrast consistently, which is a sure sign that the electrolytic capacitors in the computer have gone bad.
When a lot of pixels are turned white, the contrast shifts dramatically.
The recommendation for the display capacitors is to use tantalum capacitors and the recommendation for the inverter board inside the main chassis is to use a normal radial electrolytic capacitor. The capacitor on the trackball module apparently doesn't leak, and there's a capacitor on the floppy drive but it also doesn't seem to leak, or at least not nearly as much as the others in the computer. (There seems to be some interest on the 68kMLA forum in cataloguing capacitors in the floppy drives, because there isn't a lot of info.)
I bought a big box of parts from DigiKey to use on the computer: tantalum caps and that inverter board cap, a bunch of capacitors to use on an official Apple power supply (which I'd like to get at some point), a replacement backup battery (Panasonic VL2330), and an anti-static strap for my toolbag.
Disassembling these computers is super easy. There's four Torx T10 screws in the bottom and one T8 screw on the back. I recommend taking out the T8 screw first, because it holds the bottom half of the computer against a small piece of plastic, which recieves a lot more strain when the T10 screws have been removed. Once the screws are out, gently pull the two halves apart at the back, and pull upward a little to disengage two plastic tabs at the front (palmrest side) of the computer. I like to do this with the lid closed and the computer held upright, with the I/O ports pointing at the sky.
Be gentle when you pull apart the halves because there's a ribbon cable that connects the screen and speaker to the CPU board, which itself is mounted on the mainboard. If the connector doesn't pop loose, you should be able to reach your hands into the computer from the back and pull it free. I like to do this before I detach the plastic tabs because once the ribbon cable is free, I can be fully confident that I'm not going to break anything electronic inside the computer.
Those plastic tabs are very sturdy, and the whole chassis is designed so that the halves push against metal leaf springs for electrical isolation. Don't be afraid to use a little force. Generally you want to slide the two halves apart parallel to each other to disengage the tabs, then you can fully separate the computer.
You'll now have two halves. From here on, the "bottom half" is the side with the mainboard and floppy/HDD mount, and the "top half" is the rest of the computer, with the lid, trackball module, and ribbon cable.
These pictures are from before I cleaned it up, which is why there's a spider carcass in there and a lot of dust. Notice how the smaller CPU board (with the big Motorola CPU on it) is mounted on top of the mainboard.
The inverter board is a little blue PCB underneath the metal shield in the top half of the laptop. It has two potentiometers on it that are mechanically coupled to the contrast and brightness controls on the keyboard side of the computer, and some circuitry to generate a high-voltage supply for the CCFL backlight on the display. This is evidently a fairly complex piece of engineering, since it was manufactured by TDK, not Apple (or maybe TDK had already done the work and Apple just borrowed it, I don't know).
The inverter board is mounted with two T8 screws. It's connected to the long green PCB in the middle of the top half, called the interconnect board, by a small rectangular connector. It's easy to take out, but pull it up gently, because the backlight cable is directly connected to the side of the interconnect board closest to the floppy drive side of the computer. This cable wraps around a little plastic peg, and I recommend putting it back there when you reinstall the inverter board.
The through-hole capacitors on the inverter board are weirdly stout. The capacitors I bought for it were the smallest I could easily find on DigiKey that matched the requirements. I used a Panasonic Industry ECE-A1CKA470, which is 5mm diameter by 7mm tall. Because the board butts up against the top of the case, there isn't much room for tall caps, so I folded mine over:
I don't actually know how much extra vertical space there is, but this works fine. I think it's safe to assume that any components shorter than the potentiometers are fine.
The interconnect board links the keyboard and mouse, display, inverter board, backup battery, and speaker to the CPU and mainboard. It's also mounted with two T8 screws, but you have to take out the inverter board first. Be careful when you remove the display and keyboard cables, although the plastics don't seem to have degraded, they're still 35-year-old connectors and mounted in a way that's difficult to access.
Desoldering the old PRAM battery is kind of difficult. I recommend just cutting the old terminals and using your favorite desoldering method to pull out the big pin and clean its hole, but the little pin has a very small hole that doesn't want to come clear. I ended up just leaving the small hole empty and mounting the big pin, then heating up the small hole and pushing the other pin through it. I took the plastic cover off of the old battery and stuck it on the new one. The adhesive has mostly failed but there was enough stick left to keep it in place.
These batteries are rechargeable lithium coin cells, which you don't see very much. It's very easy to get them concerningly warm when you're soldering, so be quick.
That was the easy part.
To get to the display, you have to take the lid apart by removing the bezel. You'll probably want to do this with the two halves of the computer at least on top of each other, if not screwed in place. It's a lot easier to take the display bezel off if the lid is supported by the hinges instead of lying flat on a table. The screen is fully accessible from just the top half, so you can leave the rest of the computer assembled.
Tilt the screen to about a 145-degree angle and pull out the rubber caps right beneath the display panel, if you have them, then unscrew the two T8 screws. Slide the bezel slightly down with your fingertips. It should only move maybe a few centimeters and will otherwise stay in place. The service manual says that at this point, you should "pull the display bezel down and away to release it from the mounting tabs at the top of the display". This is not helpful.
It's hard to see but the bezel is slightly shifted downward in the second image.
I didn't like this advice. Until you take it apart, you can't tell whether the plastic clips on the display bezel will survive any amount of force. My solution is to pull the bezel off from one side, and then simply pop it free. Hold one side of the bezel with your fingertips and flex it slightly toward the center of the screen while pulling outward, and the bezel should pop free. This is why you want the bezel shifted slightly downward: there's four plastic clips at the top of the display and pulling the bezel down clears them from their mounting tabs. (Shifting the bezel may or may not actually be necessary depending on how much flex your bezel has.)
The screen (or display, whatever you prefer) itself is mounted with four long T8 screws and connected to the computer by a fairly fragile flat ribbon cable and the thick backlight wires. The screen is also shrouded by a foil EMI shield that covers almost every part of it except for the actual visible panel area.
This display is not like modern displays. It's thick, it's heavy, it has a very thin flourescent tube at one side, it has disassemble-able layers, and it has 1990s-scale electronics along three sides. To pull the display module out, place a piece of clean paper on the keyboard, then unscrew the four mounting screws and gently lay the display on the paper. When the screen is fully assembled, it's pretty sturdy, but make sure you protect the viewing surface. I like to take out the top screws first, then the bottom ones, because it makes the screen slightly less likely to go sliding out of control.
With the screen on the keyboard, gently disconnect the pink and white backlight cable. You now have a choice: if your display cable is still glued to the back of the screen, you can unstick it, or you can leave it in place. I recapped my display with the ribbon cable still glued to the display module, and it was quite inconvenient, because I couldn't lay the pieces of the display anywhere. The risk of pulling the cable off is that if the glue is still in good condition, the cable is held down and needs a significant amount of force to remove. This can cause you to rip the cable, which makes it basically trash. Either way, make sure you disconnect the display cable from the connector on the display itself.
The display is held together with screws and metal flaps. Gently unfold the metal flaps with a pair of pliers, then flip the display over and unscrew the four small screws. The three screws on the left hold the backlight tube and the fourth is extra reinforcement for the metal frame that holds the LCD panel against the backlight. The LCD should disassemble into two halves and (technically) three pieces:
Be super careful with the CCFL tube! It floats free when the display is apart and it's very thin. I didn't break mine but I definitely put it in precarious situations multiple times. Your diffuser layer (the cloudy plastic) may be attached to the LCD panel or to the back half. It doesn't matter where it stays as long as you put it back in the right place. Note how it needs to cover the whole glass panel, and how the rubber strips are aligned on top of the brown pixel driver cables.
As someone who's fond of displays, I thought this was pretty neat. The LCD panel itself is a standalone unit with three PCBs mounted to a glass panel with fragile pixel driver ribbon cables. The backlight is an entirely separate device with a thick slab of plexiglas side-illuminated by the CCFL tube. I assume the dots in the plexiglas are there to help scatter light into the diffuser.
At this point you can simply pick up the LCD panel and place it somewhere to work on it. You can pick up the panel just from the PCBs, but you probably want to pick it up from two or more at a time to distribute the force. I placed it on a piece of paper (you'll notice this is a recurring theme) with the capacitor side up, then covered the back of the display with another piece of paper. I don't recommend keeping the display in the metal frame, because it's hard to get your soldering ion to all the pads.
The capacitors on my display had barely leaked, which is probably why it worked so well before I even opened it. I pulled them off with pliers---there was enough corrosion to make them come off the solder without even heating it---then cleaned up the pads. I then installed the tantalums:
Reassembly is pretty easy. Start with the display by remounting the actual LCD panel into the steel frame. The diffuser goes first, then the tube mount holder on the left side, then the display panel on top of that. Fold the metal tabs back into place and reinsert the four small screws, and the display is back together. Reconnect its cables and reattach it to the back of the lid. I like to do this with the lid flat on a table so the screen doesn't go sliding around. Make sure the backlight cable is routed properly in the tube mount:
and in the lid itself (note the little grooves for the wires):
You can simply push the bezel against the back of the lid to reinsert it, then slide it upward to engage the clips at the top. Screw the screws back in and reinsert the rubber covers if you have them.
Everything else should be basically obvious. I like to get the ribbon cables on the interconnect board inserted before I mount the board to the top half. It can be a little difficult to slot the interconnect board into its space because of the grounding tabs on the top half of the computer.
Before you put back the inverter board, make sure the plastic dust sheet beneath it is properly aligned. It actually isn't in the pictures below, and it takes a little wrestling to get the backlight cable to coil properly and keep the sheet flat.
The shield on my inverter board was still fully attached to the transformer, until I ripped it off. It was originally mounted with some kind of foam pad and adhesive, but the foam has completely hardened and the shield won't go back on. I initially tried to stick the shield on with a loop of electrical tape in place of the remnants of the foam, but there's not enough contact area. I later found that two pieces of electrical tape on either end of the shield, looped around the board, is the best approach. Once the shield is mounted, you can put the inverter board back into the computer.
The inverter board goes back similarly, but you have to make sure the potentiometers are coupled to the plastic sliders. The Apple-recommended way to do this is to push both sliders and potentiometers as far apart from each other as possible, then to insert the inverter board. I find that this works well. The backlight cable has to already be connected to the inverter board, and you should wrap it around the little peg at the corner of the case:
Maintain this coil as you align the inverter board on its connector, and press down to align it with the screw holes. The 145B has little white discs that act like the surrounding columns of the screw posts, and they have to be properly aligned with the surrounding case to fit the inverter board. Install the screws and you should be done.
The last step is to reconnect the interconnect cable. I like to do this with the computer held vertically like the initial opening procedure, and reach my hands in to push the connector onto the CPU board. The two halves should now be floating on top of each other, so you can set the computer flat on a table. Angle the top half so that the two clips at the front of the computer engage, then push the back half down. It will still float some, that's caused by the grounding clips and it's okay. Screw down the five case screws and the computer's back together!
At this point I was pretty exhausted from several hours of PowerBook repair, but confident that I had a fully restored machine. I plugged in the computer, pressed the power button, and the computer started spewing magic smoke. I unplugged the computer right away and investigated, to find that the display (!) was the source of the smoke. I opened the display to find the charred remains of my freshly soldered tantalum capacitor on the C9 footprint.
This picture was taken after I removed the capacitor, but imagine a small charred black chunk on those pads.
In a display of uncharacteristic lack of thought, I replaced the capacitor with a second one and powered the system again. I guess I figured that I had just gotten a bad capacitor, or something. Naturally the capacitor smoked again, and now I was out of replacements for that capacitor. I tried one last test with an electrolytic of about the same size (82uF vs 100uF) soldered into C9, and the screen just didn't work at all.
The first capacitor let out a lot of smoke because it took me 15+ seconds to realize it was burning. The computer smelled like magic smoke for a week after this. It has since fully dissipated and now it just smells like a computer.
I assumed that I'd broken the display, so I gave up for the day and prepared to replace it. I unstuck the ribbon cable from its glue point close to the display connector, then tried to take it off the glue point near the hinge, but used too much force and ripped it. I put the machine back together and set it on a shelf.
Naturally, I was pretty mad at this point. I'd followed all the instructions I could find to a T and I broke the display and its ribbon cable. The ribbon cable can be replaced with a modern reproduction, but the displays are one of a kind. (Actually, there are some shady websites that sell them, and they're even on eBay for like $200 a piece, but they're basically one of a kind). I went on eBay and found the cheapest parts PowerBook I could find, a PowerBook 160 listed as "falling apart" and missing the trackball module, with the intent of harvesting its display and display cable for the 145B. I could see that the display looked somewhat grimy in the eBay photos but for $30, I decided I'd take the chance. The 160 is the second-generation passive matrix PowerBook 1xx model and actually has the same computing specs as the 145B, a 25MHz 68030 and 4MB of base RAM, and its display is compatible with the 145B. I ordered another set of capacitors, this time with more extras, to go with the new display.
For some reason USPS decided that the 160 needed to circle the continental US twice before arriving in my city, so I had a lot of time to think about the mistakes I'd made. Maybe a week after ordering the 160, I realized that the capacitor that exploded was rated for 6.3V. The computer runs at 5V, and when I had the display connected, the computer entered some kind of boot loop or something where the speaker made a periodic buzzing sound. I realized I had some kind of association between that behavior and overcurrent conditions: the computer is supposed to make the characteristic Apple chime when it turns on, but it was looping with a buzzing sound, like a short-circuit protection circuit was continually engaging, and like a speaker amplifier wasn't getting power. I then had a dim memory of learning that tantalum capacitors (which I hadn't really used before this point) have weird polarity markings. I Googled it and found that yes, in fact, they do:
I installed every one of the 11 display capacitors backwards because I assumed that the black stripe on an electrolytic capacitor serves the same purpose as the stripe on a tantalum capacitor. You can scroll up and check the picture yourself. It's honestly a bit of a wonder that just one capacitor exploded, and it seems that I was rather fortunate to get just magic smoke, not an actual fire. I then thought some more and realized that if the capacitor that exploded was sinking enough current to explode, and it's rated for 6.3V, there's a decent chance that it's actually a filter or decoupling capacitor for the display. This would mean that it's routed between the power supply and ground, and the excess current the display drew because of the backwards capacitor might have bypassed the display electronics. This doesn't explain why nothing happened when I tried powering the display with the axial electrolytic, but I was confident enough in my theory to hope that the display was actually still functional.
Around this point I also realized that I would probably have more fun if I made the 160 my main machine instead of the 145B. The 160 has the same computing specs, but it can drive the internal display at 4-bit grayscale and it has an external video output. The floppy drive from the 145B was also broken before I even got it, so I figured I could transfer the trackball module over to the 160, recap the 160, and maybe have a slightly cooler working PowerBook than I originally planned.
The 160 came in and I cautiously powered it up. It booted, but something wasn't right with the display:
I forgot to take a picture with the bezel on
This combined with the computer not chiming and the backlight control not working are basically sure signs that the interconnect cable needs to be replaced. The 160 is electrically redesigned from the 14x models, so the interconnect board and cable aren't compatible, but routing a new cable isn't especially hard. I was confident that was the only electronic issue with the computer.
I took apart the computer and opened the display. It smelled like vinegar but the display itself hadn't actually undergone any deterioration except for one tiny sliver on one side of the display:
You're looking at the back of the display, the side that faces the diffuser and the plexiglas. I think the little bite mark near C1, C4, and C5 is vinegar syndrome damage. Some of those weird dots on the glass are dust, some are...not. They kind of look like mold, with little spidery legs, but they might be dirt or more vinegar syndrome symptoms.
There was also corrosion on the metal frame, presumably from the acetic acid produced by the decomposing adhesives:
This is actually after I scraped away what corrosion I could. Look at the weird dirt on the viewing surface of the display, and the flux (I think) on the capacitors under the frame.
Fortunately the corrosion was minor, the vinegar smell dissipated quickly, and the PCB hadn't been severely damaged. The grime on the screen that I noticed in the eBay listing might (?) be caused by the vinegar syndrome, but it might also just be dirt. I can't tell and I didn't try cleaning it, for reasons that will soon become clear. The capacitors on the 160's display had leaked more than the 145B's display, unfortunately.
However, my first mission was not to work on the 160's display, it was to test the 145B's display. Armed with a new set of tantalum capacitors and isopropyl alcohol, I took off all the capacitors from the 145B's screen and installed a new set. The capacitors that didn't explode were probably fine but I didn't want to take any chances. I then put the screen in the 160 (whose display ribbon cable was already conveniently unglued from the display panel) and it worked!
Words cannot convey how happy I was to see this
I then cleaned up the 160's screen, scratched off the corrosion as best I could, removed all the capacitors, cleaned the pads and PCB, and installed new tantalum ones. It also worked. These caps had leaked more than the 145B's display and I was able to take off the solder mask with as little as a Q-tip soaked in alcohol. One capacitor pad even lifted off the board.
I recapped the interconnect board, transferred the trackball module over, and now I had a kind-of-working PowerBook 160. I noticed that the interconnect cable had gotten absolutely crushed between the halves of the computer, probably by the eBay seller who didn't know any better. You can replace the interconnect cable in a 160 using a standard 80-wire Ultra ATA cable, so I ordered one on Amazon and replaced it a few days later. The guide linked is all you need to do that, but my advice for the process is that a) you can use more force than you think on the metal ribbon cable top bracket thing, and b) use the thinnest, softest tool possible (not your thumbnail) to get the wires into their little teeth without damaging the connector.
With that, the screen showed a full image!
I had my doubts but the grayscale was worth it, it looks so good
On the lower half of the display, you can see that the contrast is slightly different and that there's something going on with alterating columns of pixels. The alternating columns thing went away after a little while, but the contrast difference appears to be dependent on how many pixels are on or off in each half of the display. I think the two halves of the display might be driven independently by the display drivers, or something like that. It's a mild effect and apparently within the range of normal for these displays.
This display is a passive-matrix LCD, so it shows characteristic passive-matrix LCD behavior: vertical and horizontal pixel bleed, slow pixel response time, and not-great viewing angles. But it's enough for me! Apple made PowerBooks in this series with much nicer active-matrix displays, but those all show a behavior called tunnel vision that makes them more interesting to work on than to actually use.
Although these computers have strong plastic, the mounts that hold the metal screw posts shrink and break as they age. The 160 in particular had a number of screw posts that broke when I took apart the bottom half, and I fixed these using Loctite Plastics Bonding System. This product appears to be some kind of alcohol-based activator and superglue, and it worked fine, but I wouldn't buy it again. Just use epoxy if you're repairing screw posts, it's stronger and you can shape it however you need. I also broke one of the screw domes on the bottom half that holds the two halves together, so my advice is tighten the screws just enough to make the case rigid, and no more.
The speaker still didn't work after replacing the interconnect cable, so I harvested the 145B's speaker and the 160 started boot chiming as it should. I also took my new PRAM battery from the 145B and installed it in the 160. The 160 added a microphone and with it some electronics on the interconnect board, so the PRAM battery has two pieces of plastic protecting it instead of just one. These had mostly lost their adhesive so I used a few drops of superglue to stick them back on. You could also just use electrical tape instead of the plastic.
I mixed and matched plastic parts between the two machines to get the 160 looking as nice as I could. It was falling apart because it had no screws, so I put the 145B's screws in it. I have a very nice 160 and an about-as-nice 145B for parts. The 145B will be my experimentation machine, for some things I want to explore:
As of writing this, I traded the 160's original display with a 68kMLA forum member for some parts that are still being shipped. I'll make more blog posts when I've done more with the computer, and I'm considering making a collection of my own guides for PowerBook procedures with less documentation: floppy drive restoration, battery rebuilding, etc.
I made some mistakes when I worked on the 145B:
If you get a PowerBook, don't be like me! These are wonderful devices to work on and if you have the tools, the skills, and the aptitude, you will have a great time.
Artist: shapoco