• Updated 2023-07-12: Hello, Guest! Welcome back, and be sure to check out this follow-up post about our outage a week or so ago.

Macintosh IIsi Restoration

PotatoFi

Well-known member
In May of 2020, a friend messaged me saying that he had some Mac stuff for me that he had picked up at his parents house. Needless to say, I was thrilled! Here's what everything looked like when I brought it home (but there are more details in this thread).

IMG_0310.jpg

Among the collection of stuff was a Macintosh IIsi, which is what we'll be focusing on in this restoration log. Let's take a closer look!

IMG_2672.jpg

I think there's a pretty nice machine hiding under all of that dirt and grime! There was some weird orange stuff on the front, which looks a bit like sticker or tape residue. It also looks like water got splashed on the front at some point.

IMG_2673.jpg

The top reveals something a bit hilarious. At some point, someone cut a hole in the top to pass a SCSI cable through! If I had to guess, I'd say that this was someone's way of achieving a CD-ROM drive upgrade. In fact, there was a CD-ROM drive in the pile that I'll bet went with this IIsi!

IMG_2674.jpg

Here's the back of the machine. There's one thing that I noticed right away: a twisted-pair Ethernet adapter! That alone makes this product a huge success, as that Ethernet card should also work an SE/30. More on that in a bit!

IMG_2675.jpg

Let's pop the cover off and see what is going on inside, and get this thing ready for cleaning and Retrobrite. But first, note the hilariously long SCSI cable!

IMG_2676.jpg

The first thing I decided to do was pop out that PDS card to see what we have. Sure enough, it's a Mac Con for IIsi and SE/30! To me, this is the equivalent of buying a Nintendo GameCube at a garage sale for $20 (yeah right, those days are long gone) and finding a $100 bill shoved under the expansion cover underneath. Whether this card stays in the IIsi or not... it will be super, super useful.

IMG_2678.jpg

With the Ethernet card removed, I pulled the hard drive and floppy. Next, I decided to remove the power supply. There are two tabs along the back to press...

IMG_2680.jpg

And then a tab on the front to pull back slightly, which allows the power supply to lift out.


IMG_2681.jpg

On the IIsi, the power connector is underneath the PSU, and connects directly to the logic board. Note the yellow/green tint on anything: cap juice. Cap juice everywhere. I'll bet that there are a bunch of leaky capacitors in this power supply that we'll have to deal with.

IMG_2682.jpg

With the power supply out of the way, we can get a good look at the logic board!

Wait a second. Where are all of the electrolytic caps?! Confession time: back in the middle of the winter, I took a few minutes to "decap" this machine. I twisted all of the caps off using the "twist and push" method, and scrubbed the whole board down in the kitchen sink with Dawn dishwasher detergent and water. Then, I took it out to the garage and blasted all of the water off of it with the air compressor. That way, it could safely wait for me to recap it without risking any more damage to the logic board.

IMG_2683.jpg

Next, the fan. This is by far the most stubborn part to remove on a IIsi. I used a screwdriver to work it free from the clips that hold it in.

IMG_2684.jpg

After much wiggling and prying, I was finally able to break it free without breaking anything.

IMG_2685.jpg

Now to remove the logic board. The way the IIsi retains it is brilliant, but I was a bit worried about breaking the plastic clips, so I asked my wife for some help. She held the tabs out on each side, while I wiggled the board forward. When it was free from the tabs, I lifted it out.

IMG_2686.jpg

Finally, I popped the speaker out, for which the contacts were previously trapped under the logic board.

IMG_2687.jpg

At that point, the case was empty, but there were a couple of quick things to do before cleaning.

One thing I considered was removing the metal shield from the back. Retrobrite can cause rust and corrosion on metal parts, but I decided that this was going to be too fidgety to remove and reinstall. There are a bunch of melted standoffs holding it into place, and I didn't feel like cutting them all off. I decided to go for a "fast" Retrobrite - in other words, get it done quick before corrosion can occur.

IMG_2688.jpg

One last thing to do before Retrobrite: pop out the Apple logo! The heat and sun can bleach the red and/or yellow out of the logo, which is very sad. I don't have any spare badges on hand, so I don't want to take any chances.

IMG_2690.jpg

I hope somebody makes reproduction badges, someday. These things are neat.

IMG_2691.jpg
 

PotatoFi

Well-known member
With everything out of the case, it was time for cleaning and retrobrite! I always like to do this out in the grass on a sunny day (I need sun for retrobrite anyway), and I mostly just use a toothbrush and some Dawn. Getting the parts super clean is not only good for their look - I think it's important for an even retrobrite job too.

IMG_2692.jpg

Sadly, there are some scratches and scuffs on the front of the machine. Hilariously, they always seem to be on the badging for the machine, where they are the most noticeable. I'm not sure why it always happens like that.

IMG_2693.jpg

The toothbrush and soap helped the weird sticker or tape residue a bit, but they didn't quite get all of it. For anything more stubborn like this, I always use baking soda and a wet rag or wet paper towel. It's amazing what the baking soda will remove! You've just got to be careful, as it works as a mild abrasive and removes some of the texture. Just work slowly, and don't scrub more aggressively than you need to.

IMG_2694.jpg

Looking good! Cleaning always makes a bigger difference than retrobrite.

IMG_2695.jpg

Speaking of retrobrite... time to get at it! I dumped a bottle of 30 proof Hydrogen Peroxide from Sally Beauty into a plastic container, and filled it with water until the top half of the IIsi case was covered. If you want to kick-start your retrobrite process, dump some how water in there! It's all about heat. In this case, I think the garden hose was full of nice hot water from the sun.

Full submersion retrobrite is very safe - there's almost not risk of "marbling" like you get with creme and plastic wrap. There's one big danger though... if a part floats and dries out on top, it will irreversibly bleach the top of the part! Because of this, it is super critical that the parts do NOT float.

My solution to this is to fill a glass pitcher with water, and place it on top of the part. It works great, and lets plenty of UV in to keep working on the parts.

IMG_2696.jpg

It was nice and hot, so the water hit 104°F (40°C) pretty fast! I find that retrobrite doesn't really work at all below 97°F / 36°C, but at 40°C, it really starts working!

IMG_2698.jpg

After a few hours, it was looking pretty good, but there was still work to do.

IMG_2699.jpg

I decided to switch to the bottom piece for awhile to get it moving. It was nice and hot in the afternoon, and because of the metal sheilding, I wanted the bottom to go pretty fast.

IMG_2700.jpg

The next day, things were looking pretty good, except for one side of the top half of the case. It was still pretty yellow, and the liquid Hydrogen Peroxide just wasn't making much progress on it anymore. I think that with retrobrite, the longer the parts go, the more you get diminishing results. I decided to try something a bit risky to finish up the retrobrite part of the project.

Normally, I don't recommend using cream and plastic wrap, as it can cause a horrible marbling effect, but it's also much more concentrated. In direct sunlight, plastic wrap and cream moves really, really fast. You just have to be super careful that it doesn't dry out! If it does, your part is ruined.

IMG_2711.jpg

Using a paintbrush, I applied 40 proof Hydrogen Peroxide developer cream to the one side of the case that was still yellow. Then, I applied plastic wrap to it, put it in the sun, and watched it VERY closely.

IMG_2712.jpg

After only 10 or 15 minutes, I removed the plastic and rinsed the peroxide away. Sure enough... streaks, but there are two kinds of streaks!

1. Streaks due to the peroxide drying out, which bleaches the plastic. This is unrecoverable.
2. Streaks from underexposure. If the hydrogen peroxide is too thin, it won't do anything to the plastic and you get brown or yellow spots. To fix this, all you have to do is add more peroxide cream and expose the part for a but longer.

These were definetely the latter, so it just needed a fresh coat of peroxide and some more time.

IMG_2713.jpg

One more exposure is all it took! After another 10 or 15 minutes, I took of the plastic, rinsed the peroxide cream off, and towel dried it. The retrobrite job looks perfect now!

IMG_2760.jpg
 

PotatoFi

Well-known member
Fast forward a few months, and I've finally got some time and energy to move this project forward. The first thing on the agenda is to recap the logic board, and to do that, we need to get all of the pads cleaned up.

Remember that I used the "twist and push" method some months earlier. I used to be horrified of this method, but after lifting pads and over and over again, trying to desolder the old caps the "right way", I finally gave this a go. To date, I've only lifted one pad like this.

IMG_3219.jpg

To get the leftover capacitor legs off of the logic board, I like to dab some flux paste on them, and then heat the pad with the iron. The flux burns all of the junk away, and then the leftover capacitor leg slides off and sticks to my iron. I jab the iron into my soldering station's copper ball cleanie thingie to dispose to the leg.

IMG_3221.jpg

With the capacitor legs removed, I usually add a bit of fresh solder to the pad, and then suck it all back off with desoldering braid. Sometimes I slide the braid around a bit while it's hot to polish the pads up, but you don't want to go crazy.

IMG_3222.jpg

Side note: I know that earlier, I said I scrubbed the board... but did I? Just looking at these photos makes me think that no, I did not. I did at least spot-clean all of the capacitor pads with alcohol and q-tips, but this whole thing still looks pretty gross. I wish I'd made absolutely sure that the board was clean before recapping it.

IMG_3240.jpg
 

PotatoFi

Well-known member
Time for caps! When I first started recapping SMD boards, I thought they were super intimidating, but these days, I find SMD work like this to be far easier than through-hole stuff. I struggled through recapping my Fall 2007 Xbox 360 a few weeks ago - the most difficult recap job I've ever done. It was all through-hole caps, but the difficulty was due to the massive ground planes everywhere, which stole all of the heat from the solder. Clearing the solder from the holes to install the new caps was a huge, huge pain.

But wait: there's something that we need to talk about, which is voltage derating. On the Macintosh IIsi, all of electrolytic capacitors are 47uF, rated for 16 volts.

A little while ago, @JDW posted a thread about the dangers of using 16 V Tantalum caps on the SE/30. I'm not going to try to explain it here, but the gist is this: for caps that see 12 volts, a solid tantalum capacitor rated at 16 volts isn't quite safe. If the cap sees 12 volts, you need to give some more margin and go with a cap with a higher voltage rating. JDW's recommendation is to bump to a 47uF capacitor rated for 25 volts. All of that said, I don't know of any documented case where a 16 V Solid Tantalum capacitor failed on an SE/30, Classic II, or IIsi. I'm personally not replacing any 16 V caps on my machines, but I am trying to use 25 V caps where needed, going forward.

And of course, just before JDW posted that thread, I made a huge investment in gobs and gobs of 47uF, 16 V caps, so I have a ton of them on hand. 25 V caps are a bit more expensive, so I've decided that I'd like to use up the 16 V caps on my upcoming projects, if I can.

There's an additional component to these components, which is that the 16 V capacitors fit better on the pads, and just look nicer, to me (and you can see this in the photo above). Here's a 16 V cap, and a 24 V cap, side-by-side.

IMG_3252.jpg

So how did a determine which capacitors could be replaced with 16 V caps, and which ones needed 25 V caps? I ran around the board with a multimeter, and looked for continuity on the 12 V rail. If the cap was connected to 12 V, I marked it for a 25 V Tantalum capacitor. If it wasn't connected to 12 V, I gave it a 16 V capacitor. I also looked at trace widths too - fat traces got 25 V caps.

Unfortunately, I've already discovered that my method wasn't foolproof - C22 in the photo above does indeed see 12 V. I won't be able to tell you exactly which ones see 12 V until I can check with the machine running, so please abstain from specifying capacitors based on my photos - they're not all correct! I'll give a parts list and parts numbers later - probably towards the end of this thread.

Ok, so with all of that out of the way, let's focus on how to install new caps.

First, start with nice, clean pads. Add some liquid solder if you like, but I don't always use it.

IMG_3243.jpg

Tin the pads on one side. I'm a leftie, so the left pads get tinned.

IMG_3244.jpg

I didn't have enough hands to show everything here at once, but what I do is bring my part in with tweezers. Then, I heat the solder on the left pad, and slide the part into place. You can easily reheat the part and carefully adjust with the tweezers until it's nice and straight. If the part isn't sitting flat on the pad, just push down on it with the tweezers while you heat the pad.

IMG_3245.jpg

Then, add some solder to the other side! Easy. And don't the caps at C9, C10, and C11 look great alongside the factory caps at C12 and C13!? I sure hope they only see 5 V here!

IMG_3249.jpg

As I go through these photos from several weeks ago, I'm very happy to report that I must have cleaned up the board! Check out the power connector - night and day difference from earlier. For extreme cases of cap juice, I always just use tap water, Dawn, and a toothbrush. Then, I always go straight out to the air compressor in the garage, and blast water out of everything, especially from under the IC's. Sometimes, I chase the water with alcohol, and then blast everything out with compressed air.

Also visible are two 24 V capacitors. You can see that they don't fit on the pads quite as nicely, but they do work.

IMG_3253.jpg

Next, I installed the board back in the machine. And looking back over these photos... there's a flaw. Don't bother looking for it - I'll reveal it soon.

IMG_3254.jpg

For now, I reinstalled the floppy drive, and I also cleaned up that massive ribbon cable.

IMG_3255.jpg

IMG_3256.jpg

And to wrap up this post, let's do one of the most satisfying parts of a vintage Mac restoration!

IMG_3257.jpg

IMG_3258.jpg
 

Attachments

  • IMG_3252.jpg
    IMG_3252.jpg
    807.6 KB · Views: 11

3lectr1cPPC

Well-known member
I've had a great time reading through all your past restorations - I always enjoy these! If you haven't already finished, I'm sure you'll be able to get this machine working like new again.
 

PotatoFi

Well-known member
I've had a great time reading through all your past restorations - I always enjoy these! If you haven't already finished, I'm sure you'll be able to get this machine working like new again.
Thanks @3lectr1cPPC! I'm super glad to hear that you enjoy them. As of this writing, it's not done, but I am getting caught up on documenting the process. Hopefully, these threads help other people with their own restorations.

---

And now, my least favorite part of any vintage computer restoration: the power supply. Ugh, I hate these things, but it's clearly oozing cap juice from all over the place. Let's get this over with.

IMG_3259.jpg

Ugh, nasty. I've never attempted to power on this machine - this is part of why. I can't imagine things going well with it looking like this.

IMG_3260.jpg

Just to further sell how gross it is.

IMG_3261.jpg

First, I use my Leatherman pliers to pinch these little standoff thins, and push the connector into the PSU case.

IMG_3262.jpg

After removing the small Phillips screw towards the mains power connector, I carefully pried the top cover off.

IMG_3263.jpg

Okay, we're in. To get the board out, we'll need to remove the screws holding a couple of components to the inside of the case, which are in the lower left corner of the photo. The large Phillips screw in the top right does NOT need to be removed (although you'll see that I did, because I didn't know yet).

IMG_3264.jpg

Underneath the mains power connectors, there are two more screws that hold the board to the enclosure. They need to come out.

IMG_3266.jpg

Ugh, this thing is disgusting. I can't wait to clean it up, and install shiny new capacitors in here!

IMG_3267.jpg

Yikes. Quite a bit of damage down here, but I think everything will still work once it's cleaned up and recapped. These big fat traces seem to be able to sustain a lot more damage than a logic board trace can.

The shiny stuff is all cap juice.

IMG_3268.jpg

I took a photo of the board, and used my iPad and Apple Pencil to label all of the caps. I also use digital calipers to determine the physical size of the capacitors. I always record:

uF Value, Voltage, Height, Diameter, Lead Spacing

The recording might look like this:

6.3 V, 2200uF, 30 mm H, 10 mm D, 5 mm LS

Next, I always make a list of parts on Digikey. Digikey doesn't allow sharing lists directly, but they do allow sharing of a cart. You can add all of the capacitors that I used to your cart with this link. At the time of writing, this cost me about $16, plus $5 for shipping. Not bad!

I always try to use Nichicon or Panasonic capacitors. I also typically go for at least 5,000 hours at 105°C. My hope is that they'll last longer than the original caps. I'm okay with paying a bit more to make sure I'll never have to mess with them again.

IMG_0690.jpg

Also included in the cart link was these two SMD caps. I decided to replace them with Aluminum Poly capacitors. I have no idea if the color means that they're Aluminum Poly caps already... but the new ones look identical.

IMG_0691.jpg

Next, I got to work desoldering all of the old capacitors, and cleaning up the board.

IMG_3270.jpg

They desoldered pretty easily, but yikes.

By the way, some of these caps were all glued together with hot glue. Someone told me to dribble Isopropyl Alcohol all over the hot glue once - it works wonders. You can easily remove the glue with some pliers after the Alcohol has soaked in for a moment.

IMG_3271.jpg

Here's the board with all of the caps desoldered.

IMG_3272.jpg

Good riddance! I do wonder how these capacitors compare to the new ones I'm installing from a quality perspective. I do see "105°C" on there in a few places, but I don't know how long they're rated for. I'm also not sure how Elna compares to Nichicon or Panasonic.

Also, it's worth noting that not all of these caps were leaking. All of the stuff on the high-voltage side of the board was in good shape, but I'm willing to spent a bit extra to just replace everything.

IMG_3274.jpg

Finally, it was time to clean up the board and get rid of that horrible cap juice. I went straight to the kitchen sink, and I used a toothbrush with Dawn to scrub everything up. I don't usually submerge complicated parts like this... but it was so gross. I didn't know how else to clean it up.

It's squeaky clean now!

IMG_3355.jpg
 

PotatoFi

Well-known member
At this point, the capacitor order from Digikey had arrived, so I got to work! I always bend the legs like this to capture the cap.

IMG_3356.jpg

Then solder, and clip the legs...

IMG_3357.jpg

And done!

IMG_3358.jpg

My wife even jumped in to solder one of them! Her normal thing is upholstery and sewing, but she was curious about soldering one day, and did a couple of practice kits for fun. Check out her skills!

IMG_3360.jpg

There's a small daughterboard on the PSU with these SMT caps. I used the solder sucker to desolder the board, so I could replace the caps.

IMG_3362.jpg

Caps desoldered!

IMG_3363.jpg

Pads cleaned.

IMG_3364.jpg

Pads tinned.

IMG_3425.jpg

Caps installed!

IMG_3426.jpg

Now to resintall the board. I soldered in one pin to make sure it was where I wanted it.

IMG_3427.jpg

And done! This PSU is ready to reassemble.

Oh, how I wish I'd given this photo a really close look before testing the PSU. Foreshadowing? Yes. But at least it looks nice.

IMG_3428.jpg

I took the enclosure down to the sink, and scrubbed the cap juice out. It was pretty bad, but it's clean now!

And oh look, the horrible foreshadowing thing is visible in this photo too! GREAT.

IMG_3429.jpg

After wiggling the PSU back into the case, I put a dab of thermal paste on this part to help it conduct heat to the enclosure.

IMG_3430.jpg

A small tab clamps it into place.

IMG_3431.jpg

Then, I bolted this thingie in. No idea what it is.

IMG_3432.jpg

I reinstalled the ground screw that I didn't need to remove in the first place, then I installed the two screws on this end that hold the board into the enclosure.

IMG_3434.jpg

Finally, I installed the top cover, and installed the two screws on the top and bottom of the enclosure.

IMG_3435.jpg

I also added a quick note for the next person (or me, depending on how long I keep this machine for).

IMG_3437.jpg

PSU done! Or is it?
 

PotatoFi

Well-known member
Floppy drive time! I'm not going to go through all of the details here, as I've covered them in one of my previous threads. I'll hit some high points though!

IMG_3438.jpg

Basically, I like to partially disassemble these drives, clean them up, and make sure everything is lubricated. I find that the dust + lubrication combo makes for some pretty nasty drives, and I always clean them before using them to make sure there's no damage to the drive or disks. These gears in particular are prone to breaking if the lubrication in the drive has dried up. Fortunately, they were in good shape - I just needed to clean all of the dust and hair out of everything. For lubrication on these, I use Dupont Teflon Silicon Lubricant. Just a couple drops is all they need.

IMG_3440.jpg

Here's the drive back together! I always take apart three things:

1. I remove the upper mechanical part of the drive, scrub them with Dawn and a toothbrush, rinse, and dry with compressed air.
2. I remove the lower mechanical part, which slides forwards and backwards, same.
3. I remove the eject motor and gear unit, and clean things up with alcohol and cotton swabs
4. I clean up the lower part of the drive, removing all grease and dust with alcohol and swabs. I clean up the R/W heads, and the rail that the heads slide on, and the screw drive for the heads.

Then, I use this lubricant that @erichelgeson sent me for all of the mechanical parts, except for the gears inside the gearbox, which gets the Dupont treatment.

When I reassemble the drive and manually eject a disk, they practically fly out like toast out of John Arbuckle's toaster.

IMG_3441.jpg

Drive reassembled and ready to go back in the machine. I like this plastic sleeve - a bit of a pain to get the drive out of, but I would imagine that this helps with the dust problem quite a bit.

IMG_3443.jpg
 

PotatoFi

Well-known member
At this point, it was time to test the machine!

IMG_3445.jpg

I reached around to the back and powered on the machine, and immediately heard the horrible grinding sound of a dying SCSI hard drive. I hardly had time to think about that though, there was a super loud POP!

I totally panicked, and started unplugging things, but it took me a moment - why I didn't just unplug the power cord from the back of the machine, I don't know, but after 5 or 8 seconds, I managed to get the thing unplugged.

Rollback plans... we use them for a reason. Next time something goes wrong, I'll try to remember to just unplug the cord from the back of the machine.

I immediately took it back apart, when I discovered this. It seems that I'd missed a capacitor entirely, but I knew this likely had nothing to do with the loud pop.

IMG_3446.jpg

Fortunately, the logic board passed the rest of the visual inspection. I was hoping for a blown Tantalum capacitor on the hard drive, but sadly, the problem wasn't there.

Next, I took the power supply back apart.

Well, there it is. I installed a capacitor backwards. When I went back and looked at the diagram I'd drawn on my iPad, I'd gotten the polarity wrong in the drawing. I also managed to ignore the markings on the board. And yes, this is the cap that my wife soldered in, but I am the one who installed the cap on the board. 100 percent my fault.

IMG_3447.jpg

Fortunately, I have a spare kit for my other IIsi Sony PSU, so I stole a capacitor from it.

IMG_3451.jpg

I double checked the polarity on a bunch of other caps, just to be sure, then I installed the missing cap on the logic board. So, so much fail here... and right at the finish line!

IMG_3452.jpg

I put the machine and tested it again, and here's what I get:

1. When I press the power button, the speaker crackles and the fans come on
2. There is no chime
3. The machine shows a brief white blank screen, and then flips to back
4. Nothing else happens

I am getting +12 V and +5 V at the hard drive power connector and floppy connector. Next, I'll check for -12 V (I'm not sure where to get that from, other than the PSU connector itself). The capacitor that blew up does sit between the -12 V rail and the +5 V rail inside the PSU. The next challenge is to find out how to tell the PSU to power on, so I can take measurements directly from it.
 
Last edited:

joshc

Well-known member
@PotatoFi

First off, well done so far on this - that IIsi PSU was cap goo hell and it's looking tons better!

If I'm being super critical, there is still capacitor leakage on the soft power board but as it is starting it's probably not causing any noticeable problem.

Are you using the bomarc IIsi schematics? Those will show where you can test -12V. You can get the schematics here: https://archive.org/details/Macintosh68kSchematics

-12V is on page 1, near Q8. Looks like you should see -12V on C39/C43/C36 (if I'm reading the schematic correctly)...

I didn't get sound from my IIsi until I replaced the speaker with a different one. However, you are getting a pop from the speaker on start so the sound circuit should be OK in that case...
 

LaPorta

Well-known member
This is great. I have a IIsi waiting for restoration. Though I don’t retrobrite, the rest of this is a huge help for when I go through mine. It’s funny: you and I have the exact same technique for removal and installation of SMD caps. Only difference is I like the solid polymer caps and don’t bother with tantalus. Partly for the reasons you stated, partly for OEM-look purposes.

I really wonder what the heck they were thinking with the long SCSI cable and hacking up the case. There’s a SCSI port for a reason…so odd.
 

olePigeon

Well-known member
This is a top notch restore thread. Excellent job. (y) I wish I had the patience and drive to document my restorations.

The thorough cleaning of the case before retrobriting is so important. I do wonder if that's the principal cause for marbling.
 

PotatoFi

Well-known member
I really wonder what the heck they were thinking with the long SCSI cable and hacking up the case. There’s a SCSI port for a reason…so odd.

I was able to ask the previous owner about this, and I got a very satisfying answer:

I think the story is that we discovered a way to add a CD-ROM to that particular machine, and my brother cut a hole in the case so that the cable could run down into the motherboard from the external drive.

I believe they were kids at the time (I certainly was), and they probably just didn't know about or have access to the "right" way to do this. My guess is that this happened in the mid 90's to early 2000's, at which time this machine would have been considered totally worthless. I remember being 14 or so, and only having a small pile of very old machines to work with. Good times.

At this point, I think the hole adds character, and I think I'm just going to leave the hole as-is.

This is a top notch restore thread. Excellent job. (y) I wish I had the patience and drive to document my restorations.

The thorough cleaning of the case before retrobriting is so important. I do wonder if that's the principal cause for marbling.

Thank you! My hope is that these notes help others with their own restorations. They don't take nearly as much energy as doing YouTube videos does... those are the people I have tons of respect for. Documenting everything on video is exponentially harder than snapping photos and doing some quick writing.

Agree on the thorough cleaning. I think it's pretty important, but I do also think that drying peroxide is the primary cause of marbling. Dirt and grime could definetely be a factor, though.

@PotatoFi

First off, well done so far on this - that IIsi PSU was cap goo hell and it's looking tons better!

If I'm being super critical, there is still capacitor leakage on the soft power board but as it is starting it's probably not causing any noticeable problem.

Thanks! It was rewarding to see the PSU all cleaned and recapped. It looks so, so much better now.

As for the soft power board (thanks for letting me know what it is, I was unaware), I agree that the caps were leaking. I scrubbed it clean with Isopropyl Alcohol, recapped it, and cleaned up the flux from the recap before reinstalling it. It should be very clean now.

Are you using the bomarc IIsi schematics? Those will show where you can test -12V. You can get the schematics here: https://archive.org/details/Macintosh68kSchematics

-12V is on page 1, near Q8. Looks like you should see -12V on C39/C43/C36 (if I'm reading the schematic correctly)...

Thanks for pointing me to these schematics - these are hugely helpful! In fact, this helped me move the project forward a little bit.

With @joshc's information about the -12 V rail, I identified these pads on the board that I could use for testing. With the machine powered on, they showed -12.35 V. With that, I feel like we can probably rule out an issue with the -12 V rail on the power supply.

IMG_3477-1.jpg

I turned back to the logic board, and notices some corrosion about this chip, which is a bus transceiver.

IMG_3478-1.jpg

I decided to give the board another cleaning, and this time, to be extremely thorough. I started with 91 percent Isopropyl Alcohol (the best I have on-hand right now), and scrubbed at the corrosion above. Then, I switched to Dawn and tap water, and went crazy.

IMG_3479-1.jpg

After rinsing it with very hot water, I went directly out to the air compressor and blasted all of the water off of the board. I am always especially careful under the chips - I don't want any mineral deposits in there! Then, it went in front of the fireplace for a couple of hours to be sure it was dry.

IMG_3480-1.jpg

After I was confident that the board was totally dry, I reinstalled it in the machine. Strangely, the board powered on automatically, but I think there's a mechanical state that the power button can be in that causes this to happen. Perhaps someone can confirm for me. Rotating the power button a but caused it to "pop out" and this behavior stopped.

Sadly, it's the same thing. Small crackle in the speaker, fans come on, screen flashes white, and turns black.

On a whim, I decided to try the ROM from me SE/30.

IMG_3482-1.jpg

The only change this made is that the screen stays white indefinitely.

IMG_3481-1.jpg

At this point, the only next step I can think of is to use the Bomarc schematics to find stuff connected to the -12 V rail, and try to test components. If anyone has ideas, I'd love to hear them.
 

PotatoFi

Well-known member
This is a bit of a side note, but using the Bomarc schematic, I was able make some guesses at which caps only see 5 V, and thus don't need a 24 V solid tantalum cap. If someone wants to double-check my work, I would appreciate it.

CapacitorSolid Tantalum VoltageVoltagePurpose
C516 VUnknownCouldn't find on schematic, + connected to pins 3 and 10 (inputs 1 and 3) and LM324M opamp. The trace between the + and opamp is teeny tiny, suggesting that this cap doesn't see 12 V.
C916 V5 V (assumed)Between audio IC's
C1016 V5 V (assumed)Left Audio
C1116 V5 V (assumed)Right Audio
C2024 V12 VMeasured with multimeter
C2224 V12 VMeasured with multimeter
C3116 V-5 VBehind -5 V voltage regulator
C3424 V12 V
C3624 V- 12 V
C3516 V5 V
C3824 V12 V
 
Last edited:

cruff

Well-known member
Don't assume that because a trace is tiny it can't carry a higher voltage. It is the current the trace carries that is the important factor in trace area (and thus width).
Did you mean to say that C5+ pin is connected to the LM324M pins 3 and 10? What supply voltages are supplied to the op amp (the datasheet says 32V max)? That will tell you the maximum voltage C5 could be exposed to as you can't drive the inputs much beyond the supply voltages without damage.
 

PotatoFi

Well-known member
Don't assume that because a trace is tiny it can't carry a higher voltage. It is the current the trace carries that is the important factor in trace area (and thus width).
Did you mean to say that C5+ pin is connected to the LM324M pins 3 and 10? What supply voltages are supplied to the op amp (the datasheet says 32V max)? That will tell you the maximum voltage C5 could be exposed to as you can't drive the inputs much beyond the supply voltages without damage.
No assumptions made! The tiny trace suggests (based on the trace widths I can see and measure elsewhere on the board) that it's not 12 V, but it does not confirm anything. Unfortunately, I can't find that circuit on the schematic, and since it is underneath the power supply, I can't take measurements with it running.
 

techknight

Well-known member
looking at the corrosion pictures, especially around the 245, I bet there are broken traces. you will need to check/inspect all of them. Potentially remove that IC.
 

zackl

Well-known member
These threads are absolutely golden! As noted in past threads (1) You should youtube this! (2) I would love to pay you to restore one of my machines w/a quality thread like this - have an 840av (that's been recapped but otherwise untouched) :)
 
Top