Fill them with deoxit
Another thing I do is wiggle the RAM a load before pushing it the last little bit to latch it, to try and clean the points on the contacts.
-
Hello MLAers! We've re-enabled auto-approval for accounts. If you are still waiting on account approval, please check this thread for more information.
restoring a battery-bombed IIcx
- Thread starter bigmessowires
- Start date
Another thing I do is wiggle the RAM a load before pushing it the last little bit to latch it, to try and clean the points on the contacts.
I'd check the power switch contacts and deox if required.
And does the cold-start power-on happen with a Mac PSU? It may be a quirk of your ATX PSU plus adapter.
I poked at the startup circuit some more but still couldn't solve the cold-start power-on problem. I'll live with it as-is. I also can't explain the ROM-inator problems unless it's an SE/30-style issue with the PCB thickness. Bracing or pushing on the ROM SIMM didn't help. I can live without it a ROM SIMM. I can also live without an explanation for the 4MB RAM SIMM mystery, since it's working with 8x1MB SIMMs now and that's good enough for me.
So all that's left is a bit of cleanup: a few more caps on the motherboard, install the battery holder and the blue power LED, and recap the PSU.
Has anyone ever tried replacing the fan in a stock IIcx/IIci PSU? Mine is pretty loud.
So all that's left is a bit of cleanup: a few more caps on the motherboard, install the battery holder and the blue power LED, and recap the PSU.
Has anyone ever tried replacing the fan in a stock IIcx/IIci PSU? Mine is pretty loud.
Cleaning and lubricating generally gets them quiet again.
I tend to encourage maintenance over replacing fans because it can be tricky to replace a Mac fan. Mac fans are selected to push quite a volume of air at pressure through a tortuous path, common "high volume" modern fans are generally designed for a lower pressure drop in modern, open case designs.
Try putting a couple or three drops of silicone sewing machine oil in the bearing.
This may be a dumb question, but how does one get access to the fan bearing to lubricate it? Do you drip oil into the gap surrounding the hub, or is there a way to disassemble the fan and remove the blades?
As for the cold-start power-on, I noticed something about the startup circuit schematic:
Look at C9 and R30. When the Mac is first plugged in, C9 will be at 0 volts and then will begin to charge up to 5V through R30 and the 5V trickle supply. But it's a big capacitor and an even bigger resistor, 47 uF times 100K ohms is a time constant of 4.7 seconds. During the time that C9 is charging up, its input to the NAND gate will be low, forcing the NAND output to be high, which keeps Q2 turned off. What this means: for a period of a few seconds after the Mac is first plugged in, C9 prevents the Mac from turning on even if you press the power switch. I assume this is intentional, to prevent exactly the sort of spurious start-up that I'm seeing.
But my IIcx turns on within about 0.5 seconds of plugging it in or flipping the switch on the ATX supply. That should be impossible. So this points to a problem with C9, R30, or the connections between them and the NAND gate. But I've checked all those and they're OK.
As for the cold-start power-on, I noticed something about the startup circuit schematic:
Look at C9 and R30. When the Mac is first plugged in, C9 will be at 0 volts and then will begin to charge up to 5V through R30 and the 5V trickle supply. But it's a big capacitor and an even bigger resistor, 47 uF times 100K ohms is a time constant of 4.7 seconds. During the time that C9 is charging up, its input to the NAND gate will be low, forcing the NAND output to be high, which keeps Q2 turned off. What this means: for a period of a few seconds after the Mac is first plugged in, C9 prevents the Mac from turning on even if you press the power switch. I assume this is intentional, to prevent exactly the sort of spurious start-up that I'm seeing.
But my IIcx turns on within about 0.5 seconds of plugging it in or flipping the switch on the ATX supply. That should be impossible. So this points to a problem with C9, R30, or the connections between them and the NAND gate. But I've checked all those and they're OK.
You peel back that sticker and its underneath there that you want to put a few drops of oil in.
There is often access under the sticker. Try not to get oil on the sticker or it won't stick.
Has the cap lost capacity. Or has the resistor lost resistance? Did you check for shorts as well as breaks?
Last edited:
I've spent too too too much time looking at this cold-start power-on. Measuring all the voltages in the turn-on part of the startup circuit, UM2 appears to be working normally except its "high" outputs at pin 6 and pin 8 only reach about 4.4-4.5V even when its supply voltage is 5.0V. As a result I think pin 8 never gets quite high enough to fully shut off Q2 and the circuit is always biased towards almost turning on. I'll ignore pin 8 for a minute since it's connected to the transistor and other stuff, and focus on pin 6 which is just a simple interconnection to pin 9 on the same chip. There is nothing drawing current here, so pin 6 should definitely reach 5.0V highs or something very close to it.
My guess here is that UM2 is bad. Maybe its internal supply voltage connection is broken, and it's being powered by its other input signals through its ESD protection diodes. This would result in a "supply voltage" that's a diode drop lower than the input signal voltage, and output signals could never go above that. I'm measuring 5.0V right at UM2 pin 14, so I know the 5V trickle traces are OK and 5V is reaching the chip's supply pin. So if it's broken, it's broken internally. That seems unlikely and this is all speculation, but it sort-of makes sense and it's the only theory I've got.
My guess here is that UM2 is bad. Maybe its internal supply voltage connection is broken, and it's being powered by its other input signals through its ESD protection diodes. This would result in a "supply voltage" that's a diode drop lower than the input signal voltage, and output signals could never go above that. I'm measuring 5.0V right at UM2 pin 14, so I know the 5V trickle traces are OK and 5V is reaching the chip's supply pin. So if it's broken, it's broken internally. That seems unlikely and this is all speculation, but it sort-of makes sense and it's the only theory I've got.
First thing is to desolder, clean and resolder it. The cap juice can create high resistance or dry joints on chips.
I've "fixed" the cold-start power-on problem by adding a 1.2K resistor between the base of Q2 and the 5V trickle, and the IIcx now behaves normally with all three PSUs I've tested (one original and two ATX). This isn't a very good fix, but it does seem to address the observed problem. I'll try replacing UM2 and removing the new resistor when I can get the right part to replace UM2. I'd rather avoid doing too much hot air rework on the motherboard so I'm not going to try removing and resoldering it, although that might work.
I did test the relevant resistor vales and checked for shorts between pins of UM2 and power/ground, but didn't find any.
The 1.2K resistor biases the base of Q2 more towards 5V, so that the switching threshold at UM2's output is effectively lower than before. Before, any voltage from UM2 below about 4.4V would turn on Q2, which was a problem since UM2 didn't seem able to output more than 4.4-4.5V. After this change, the UM2 output voltage only needs to be above about 2.0V in order to keep the base of Q2 high enough to prevent it turning it on. If replacing UM2 doesn't help then I'll make this resistor addition permanent.
I did test the relevant resistor vales and checked for shorts between pins of UM2 and power/ground, but didn't find any.
The 1.2K resistor biases the base of Q2 more towards 5V, so that the switching threshold at UM2's output is effectively lower than before. Before, any voltage from UM2 below about 4.4V would turn on Q2, which was a problem since UM2 didn't seem able to output more than 4.4-4.5V. After this change, the UM2 output voltage only needs to be above about 2.0V in order to keep the base of Q2 high enough to prevent it turning it on. If replacing UM2 doesn't help then I'll make this resistor addition permanent.
Highly recommend picking some of these up for those pesky through hole pins that seem to be able to suck up stupid amounts of heat and still not clear. Game changer for me with them… very cheap too. I find a bit of twisting motion on them while you push and heat with the iron works wonders. HiLetgo 3 Lot Mixed Stainless Steel Non-stick Tin Hollow Core Needle Kits for Soldering Assist Accessories https://a.co/d/8ZqOLc3
When I was recapping my Mac II I got super frustrated until I found those.
When I was recapping my Mac II I got super frustrated until I found those.
Interesting... thanks! How do you use them? Do you apply your soldering iron to the needle, or to the rim of the pad?
Doesn’t surprise me that UM2 might be bad. I think a few folks (including me) had to replace those chips in that startup circuit. My battery bombed IIci needed all three of those chips replaced (one 74HC74 and two 74HC132), because the caps in that area eat those chips up and two fell off the board, one was dangling. I got Toshiba replacements (757-74HC132D for the 74HC132) from Mouser which seem to work but I didn‘t have the exact original part number so couldn’t compare specs. I have two extras still as I picked up four of them in case I needed some in the future.
It’s a little tricky to pull off but I’ve found for me that using the iron on one side of the board and the little pusher on the other side pushing through to the iron works best. Worth a try if you just can’t get that battery holder cleared.
An alternative I've used dozens of times is a dental probe. A fine and rigid stainless steel tip chases solder out of through holes easily. Heat should be applied to both hole and probe.
Thanks! I've ordered some of those needles and will give it a try. I'm embarrassed to admit how much time I've already spent trying to clear ONE HOLE in the PCB. Several hours at least. ChipQuik helped a little, but the hole remains plugged.
The issue with this IIcx's ROM SIMM slot was indeed related to SIMM thickness. I was testing it with a couple of older 1.2 mm ROM-inators, and could never get the computer to boot when they were installed. 1.2 mm has always worked fine for me in every Mac II model except the SE/30, and also works in the SE/30 if you brace the SIMM or use a retaining clip, but I had no luck in this IIcx. My clue was that a stock SE/30 ROM SIMM did work in this IIcx, so I tried a newer 1.27 mm ROM-inator and it booted straight up.
With that resolved, the cold-start power-on issue is the only remaining mystery before I can declare this battery-bombed machine 100 percent working.
With that resolved, the cold-start power-on issue is the only remaining mystery before I can declare this battery-bombed machine 100 percent working.
I added a few drops of oil to the bearing in the IIcx power supply fan. I don't think it made any difference though. Check out the before and after and tell me if this sounds normal.
before:
after:
before:
after:
