• 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.

Fix0ring a Macintosh IIfx

Arisotura

Well-known member
I was looking around on eBay when I saw this one, so I thought, hey, why not give it a try. It was provided as 'not working' and having undergone battery leakage, but it did not look that bad...

maxellmac3.jpg

maxellmac2.jpg

Some of these components aren't looking too hot, but compared to the pics of Maxell-bombed Macs I have seen, it's certainly not as much of a disaster.

So, can we get this going again?

-

I started by cleaning up the motherboard and inspecting it to see what would have to be fixed. This is where I am tonight:

IMAG1686.jpg

As far as capacitors go, this motherboard is equipped entirely with tantalums and axials, so there's no cap gunk to deal with here. And that shows: outside of the battery leakage area, the board is in mint condition.

Anyway, I started removing the components that were affected by the leakage. I did some cleanup, but I ran out of solder wick, so I have to wait for some orders to arrive before I can go on.

The CPU got a tiny bit of corrosion on a couple pins, so I removed the CPU socket to check underneath. The socket and the board are actually fine, somehow. I cleaned up the CPU, I'm confident it's going to be fine.

For similar reasons, I removed one of the RAM sockets, and it did have battery gunk underneath, so that was a good thing. One of the pins had been corroded, I couldn't even desolder it because the solder had turned to some crust, I had to just push that pin through the board. The socket otherwise went out with no fuss, and it's going to be fine.

The two 74F573 chips between the RAM and the CPU did get a lot of the corrosion. One of them has a really weak leg, so I'm going to replace it. The board underneath is fine and the connections are all good, the pads just need a deep clean.

The RTC chip is fine as well, and all direct connections to it are good.

The worst is probably the start-up circuitry -- one of the diodes was missing, and two more came off as I scrubbed the board. The solder joints on the remaining ones aren't looking too hot, either, so I will have to rework these. There is also some trace damage to repair, in one case the corrosion even ate through a via and started eating the trace on the other side.

I'm confident I can get this running again :)

-

I also disassembled and inspected the power supply. All seems good inside, nothing's asploded, there's no trace of cap gunk or bulging caps. I wasn't able to verify that it's working, apparently these need an external power source to turn on. In the meantime, the motherboard will first be tested with a known-good ATX power supply, when it's finished, that is.

 

Byrd

Well-known member
If you're comfortable desoldering sockets and other ICs, you'll have that bad boy up and running in no time :)   Consider the rusty RF shielding to be "patina" unless significantly corroded.

 

LaPorta

Well-known member
There are a few more things going on here that I can visually see.

1. Your second battery holder is completely gone. The machine won't start without it's two PRAM batteries.

2. There ARE two SMD electrolytic caps which also seems to be gone on yours. They are capacitors C9 and C24, which are located in the upper-right hand corner by the startup circuit, and to the right of the RAM slots above the axial electrolytic, respectively.

Compare your board with the photos from mine which I resotred less than a year ago.







By the way, are you sure the thing wasnt partially submerged in water? That is an insane amount of corrosion on the RFI shielding. I guess it's possible?

 
Last edited by a moderator:

Arisotura

Well-known member
Second battery holder was gone already when I got it, and I planned for that in my parts orders. As a side note, the minus via for that one is completely and entirely gone, so I will have to get creative to reconnect that.

C24 was present, it was a tantalum but it was already half detached, so I removed it. I'm getting a new cap for that one, seeing what the battery gunk did to the axials.

C9 however is still present, it's also a tantalum on mine (all of them are).

Regarding the RF shield, I'll see what I can do there, I guess I'll atleast get rid of the 'big' part of the rust, if that makes any sense.

 
Last edited by a moderator:

LaPorta

Well-known member
Sounds great! Thankfully, the negative terminal isn't bad usually: drill the hole open, get the battery holder terminal through, and just solder wire from it to where the trace goes.

 

Arisotura

Well-known member
According to the schematics, the minus terminal for BT2 connects to ground. I'll find a good ground on the back on the board and run a bodge wire :)

 

Arisotura

Well-known member
Thinking about it, another possibility would be to try and enlarge the hole to expose the traces that connected the minus-terminal via to the ground plane, and solder to that. However that might be problematic, depending on the order of the layers inside the board (if there is a power plane before the ground plane, we might be getting too close to that).

 

Arisotura

Well-known member
By the way, are you sure the thing wasnt partially submerged in water? That is an insane amount of corrosion on the RFI shielding. I guess it's possible?
Seeing as I can't really do much until my orders arrive, I looked closer at the RF shield...

IMAG1688.jpg

The rust on the RF shield is typical of a Maxell bomb, from what I've seen. I hadn't noticed how bad it was near the speaker. I might need to get this shield out somehow, derust it and maybe polish it, but it's not going to be factory-fresh.

This also tends to confirm what I thought seeing the motherboard: that computer was stored vertically, front down. That may well have saved it from the Maxell bomb, as most of the battery acid dripped down and corroded the RF shield, instead of stagnating and ruining the motherboard.

That's one lucky Maxell victim, I guess.

-

Seeing as this is a weekend, I may begin working on the case. I wanted to see if the computer would work at all beforehand, but eh, I need to keep myself busy :p

 
Last edited by a moderator:

Arisotura

Well-known member
Lil' update!

IMAG1702.jpg

Picture might not be that good, but basically:

* all SMD parts are resoldered, except for the F573 for which I haven't received the replacement part yet

* all damaged traces should be fixed, everything had its connections checked thoroughly before being soldered back

* a few vias were bad so I had to get creative with my bodging... dealing with 0.2mm wires is a lot of fun

* my neck is screaming at me

I will do the through-hole parts tomorrow :)

Once I receive the F573, I'll be able to give this a test run. I'm a bit excited to see if it's gonna work, and what's inside :)

 

Arisotura

Well-known member
Thanks :p

Regardless, here's where we are this morning:

IMAG1706.jpg

The through-holes are installed. Except for the RAM socket, I will do that one after I have received and installed the F573 -- it would hinder my already limited ability to solder that SMD chip in place.

As usual, all connections have been tested, and all the traces that run under the battery holders were verified, they are fine. BT2 is looking lopsided because it's sitting on one of the bodges.

Speaking of, a close-up of the creative bodging:

IMAG1707.jpg

I still have to clean up that board, heh.

IMAG1708.jpg

The original reset/intr buttons. The right one is actually still good, but the left one is toast -- it won't even press at all, plus one of the legs broke off during removal.

I replaced both for the sake of consistency mostly. I wonder if there is any functional reason why the original ones have these holes in the button part.

-

I can't test this fully yet (missing one of the F573 latches that sit between the data bus and the RAM means we would only get a sad-mac at best), but I did a lil' test with batteries installed: pressing the power switch sends ~5.7V on pin 15 of the power connector, so atleast that part of the computer is working :p

 

moldy

Well-known member
so I removed the CPU socket to check underneath
@Arisotura I was wondering, what technique do you use to remove the sockets safely? A desoldering station? Solder wick? I also have a battery victim (a compact mac though), but I'm not sure what would be safest for the board itself.

 

Arisotura

Well-known member
I actually used a manual solder sucker, and it took forever. I since ordered an electric one. Key is to ensure that all the pins are loose before trying to pull out the socket.

Good luck with your fix0ring, btw :)

 
Last edited by a moderator:

Arisotura

Well-known member
Well... looks like I finally got that F573. Of course I immediately went to solder it in place, and...




Side notes since the video:

* sad-mac tone happened because, as I feared, the RAM is mismatched -- there are 5 1MB sticks and 3 4MB sticks. I'll have to see what I can do there, but for now the thing is stuck with 4MB of RAM.

* the HDD seems to work fine, the system on it won't boot tho, probably because there isn't enough RAM

* the floppy drive needs a good cleanup/lube, and currently doesn't read disks (even though I cleaned the read heads)

* sound, RTC, ADB, SCSI, floppy interface, are working

* the rusty video card (which I did get separately, and quickly recapped) is working

* reset switch isn't working, NMI is working

 

Arisotura

Well-known member
Lil' update for tonight:

Reset switch is fixed. There was a flaky via which I'd missed in the reset circuitry, and R11 was bad too -- the resistance was too high for the reset-button circuit to function. So I had to go hunting for a 100ohm SMD resistor... lots of fun.

The video card supports resolutions 512x384 and 640x480, but interlaced. My LCD doesn't like these modes, it treats it as a 720x240 resolution but the picture is real shitty and shaky. A bit odd considering my LC works like a charm with this monitor.

I put the IIfx's power supply back together, and as I'd expected, it works absolutely fine. There's something satisfying to pressing that power button and hearing the relay click on :)

There's still one part of this computer giving me trouble (besides the rusty RF shield and weird RAM): the floppy drive is not working. I was able to use my LC's floppy drive and boot into System 6 from that, so there's no problem with the motherboard.

I figured the floppy drive just needed a good clean/lube, seeing how slowly it latched down when inserting a floppy... so I did exactly that. Then it was clean and snappy and all was good... except it still didn't work.

So I looked deeper into this, and... it's an interesting failure mode to say the least. Basically, the drive doesn't even move its head back to track0 when power is applied, and when a floppy is inserted, it doesn't attempt to do anything other than immediately spit it out (and by the way the eject motor is weak).

Further inspection shows that the drive is actually trying to do the right thing -- when turned on, it _tries_ to move its head back, as I can see current applied to the head motor pins, and that stops if I manually move it to track0. So atleast the logic is working, but something is wrong with the power sent to the motor -- I measured about 1.8V on the motor pins, which isn't right. And I tried the motor itself in my good drive, it worked absolutely fine (and the good drive sends 5.8V to the motor).

The head motor is controlled by a Sony CXA1035 chip, which is getting 4.4V of power, presumably not enough for proper operation (seeing as the same chip in my good drive is getting 5V).

The power seems to come from the little circuit with Q3 and C16. I suspect Q3 might be faulty, but I'm going to learn more about this...

 

Arisotura

Well-known member
Thanks  :p

re: RAM

That's a nice option to consider, quite pricey tho (a full 128MB set would be more that what I got the Mac for, heh). I wonder if it would be possible to make compatible RAM sticks using more modern chips (but I doubt they make low-capacity chips these days, so, yeah).

re: floppy drive

Upon further inspection I found out that, on the drive's logic board, Q4 was shorted, causing the motor 12V line to be taken down to about 2V. Now I need to get a good replacement and we can see whether that drive works.

 

MrFahrenheit

Well-known member
Thanks  :p

re: RAM

That's a nice option to consider, quite pricey tho (a full 128MB set would be more that what I got the Mac for, heh). I wonder if it would be possible to make compatible RAM sticks using more modern chips (but I doubt they make low-capacity chips these days, so, yeah).

re: floppy drive

Upon further inspection I found out that, on the drive's logic board, Q4 was shorted, causing the motor 12V line to be taken down to about 2V. Now I need to get a good replacement and we can see whether that drive works.
The 128mb IIfx ram kits for ~$200 are made by a forum member @hyperneogeo here.  The pricing is absolutely reasonable. 
 

If you think ~$200 for 128mb in a IIfx is expensive, you should have seen the pricing last few years before he made them. I paid almost that for 32MB (8x4mb). Even $20 per 1mb simm wasn’t unreasonable until recently. 
 

Once the supply of 128mb kits is used up, I fear pricing will return to the stratosphere. 

 
Top