Broken Macintosh II Repair

[davidg5678]

I am working on a Macintosh II motherboard which will not boot up. It has been recapped, cleaned, and has had new batteries installed. Using this excellent article, I checked for continuity throughout the entire power circuit, and I have patched all of the broken traces I could detect. I have confirmed that there is a connection between R3, R18, and C6, as I read that this often goes bad. For good measure, I also added a patch wire from C14's positive side to UI14's pin 10. I desoldered UB1 and UB2 and cleaned their pads before reattaching them to the board.

When I press either power button, sometimes the PSU turns on, but sometimes nothing happens at all. The power button on the back of the computer turns the machine off, but it then immediately turns itself back on. There is never a boot chime and nothing ever appears on the screen. I verified that the PSU voltages are normal. Sometimes I can make the PSU turn on with a jump-start, but the computer still does not boot, and even the ability to jump-start is intermittent.

When working on this same board a few years ago, (with significantly worse equipment and skills) @AwkwardPotato recommended that I check pin 8 of UB2's voltage while pressing the keyboard power button. Apparently, the voltage is supposed to drop to 0v from 5v when the button is pressed; however, I have never measured a voltage at this point that is higher than 0v. The pin always seems to be pulled low. (assuming this is the correct terminology) I also measured voltages from the NAND gates 3 & 4, but they seem to be behaving abnormally. Instead of changing from 5v to 0v when a power switch is pressed, their voltages remain at 5v. I think this is part of why the computer is not turning on, but I am not sure how to proceed. (Maybe a component is defective, but I don't really know how to tell.) At least Q3's emitter is measuring within the right range.

The computer doesn't chime -even on the rare occasion I can manage to get the PSU to turn on, nor is there is not a video signal. I am unsure whether or not this is a related or separate issue from the power circuitry problems.

If anyone has any ideas or advice about how I can fix this computer, I would love to hear it! Thanks!

I've placed AwkwardPotato's quote below along with a few links to some other, older posts for context.

Measure the voltage on pin 8 of UB2 too, this should be 0V when the keyboard switch is pressed. Measure the voltage at the emitter of Q3, it should be around 5-7V at all times. 

In a working power-on circuit, gates 3 and 4 (rightmost ones in the power-on circuit) should output 5-7V when neither switch is pressed, until one of the switches is pressed, at which point it goes to 0V. The output of those gates then goes to the base of the PNP transistor at Q3. To turn on that transistor (let current flow from the emitter to the collector), the voltage at the base should be lower than that at the emitter (which is always at 5-7V). Since the collector is connected to the pin that turns on the power supply, the base needs to be pulled to 0V to turn on the computer.

 

[davidg5678]

I've decided to try replacing UB1 and UB2, along with the Z53C80 SCSI controller chip. Surprisingly, all of these parts are still for sale on Mouser.com, even the ancient SCSI controller. I have suitable replacement SMD diodes and transistors on hand, and I may also swap these out too. Hopefully, these replacements are all I need to make things work again! I'll keep this thread updated with further developments once I swap the chips.

If you have any ideas or experience in fixing Macintosh II power circuits, I'd love to hear from you!

 

[buserror]

I'm currently doing the same job, and I decided outright to change UB1 and UB2 -- make sure ALL the traces connect while the chips are off the board, as on my board, some of the traces between UB1&2 were corroded enough to remove connection between pads.
Also check on the power button side, as there was quite a bit of damage there, I decided to change the SCC ICs there as well as they were badly corroded.

I've been fixing another problem with another piece of kit so I ahven't had time to power mine up yet, but I hope I can do that in the next few days!

 

[davidg5678]

I'm currently doing the same job, and I decided outright to change UB1 and UB2 -- make sure ALL the traces connect while the chips are off the board, as on my board, some of the traces between UB1&2 were corroded enough to remove connection between pads.
Also check on the power button side, as there was quite a bit of damage there, I decided to change the SCC ICs there as well as they were badly corroded.

I've been fixing another problem with another piece of kit so I ahven't had time to power mine up yet, but I hope I can do that in the next few days!

Thanks for the advice!

I removed UB1 and UB2 and checked for continuity throughout the entire power circuit. I found no broken traces in the circuit (apart from a few that I was already aware of -these were fixed before testing, of course). The traces and functionality of the rear power switch are normal.  I replaced the UB1 and UB2 chips with brand new parts; however, the computer now seems worse than before. When plugged in, it turns itself on and off and seemingly random intervals, each time without chiming or displaying a picture on the screen. The power buttons no longer have any effect on the computer.

This one has me stumped. Any ideas @Bolle@techknight?

Thanks!

 

[techknight]

You still have an open trace somewhere. 

This is hallmark of a floating digital/control line. Almost as if one of the pullup or pulldown resistors has gone open circuit either the resistor itself, or, the trace/via connecting to it. 

 

[davidg5678]

You still have an open trace somewhere. 

This is hallmark of a floating digital/control line. Almost as if one of the pullup or pulldown resistors has gone open circuit either the resistor itself, or, the trace/via connecting to it.  

I took a closer look at UB1 underneath my microscope and found that a few legs of the IC were not making contact. I soldered these legs down, and the erratic power on and off problem has been resolved. The computer is now in the same state it was in before I replaced the ICs. :sadmac: (It will not turn on -even with a jump start, and it never chimes or shows video)

I decided to check every resistor on the back of the board with my multimeter and I got a number of strange readings. (I removed the batteries before taking measurements, but I didn't desolder each resistor before measuring it.)

• R16: Reading jumps around and multimeter cannot lock onto anything
  
• R20: Reads 50.6 Ohms instead of labeled 120 Ohms
  
• R21: Reads 46 Ohms instead of labeled 68 Ohms
  
• R28: Reads 5.5 Ohms instead of labeled 10 Ohms
  
• R31: Reads 5.4 Ohms instead of labeled 10 Ohms
  

Nothing was open, but almost every reading I took was off by at least a small fraction of what it should have been. Are resistors just like this, or do I have defective parts? (Maybe I should have removed each component before measuring it? :huh: )

Does my problem lie with the power circuitry, or could something else be broken elsewhere in addition? As I am unable to jump-start the computer much of the time, I know something is seriously wrong with the power circuit. On the rare occasion where I can jump the computer, I don't hear a boot chime, but instead a tiny pop from the speaker. I wonder if the ROM (or one of its connections) is broken, or if ESD mysteriously fried something?

 

[Phipli]

Reading values in circuit isn't reliable. So...

Weird non stable : there might be a cap in parallel and your meter is charging it.

Way off value : If you're reading a 1k resistor, and there is a second 1k resistor in parallel in the circuit, you'll read 500 ohm. Components that aren't resistors can cause this. You shouldn't expect to see marked values in circuit, but it is a way of comparing with another machine.

Slightly off readings : in addition to the same as above... not all parts are created equal. Sometimes you don't need an accurate value componet - for example, a resistor in an ADC might need to be 0.1% accurate... but a pull-up resistor might only need to be within 20%! To save cost you use low accuracy components where you don't need precision.

My advice - concentrate on checking continuity. Also, take care that you're not fixing a broken connection by pushing down on it while measuring! I'd put fresh solder on pins that look excessively crusty, although I'd only remove a chip as a last resort as putting too much heat into something might kill it. Check every identified pin is connected to each destination on this schematic  :

https://www.downtowndougbrown.com/2015/03/explanation-of-the-macintosh-iiiix-power-onoff-circuit/

If you find a difference, check with another owner of a II, just in case you have a different board revision - you don't want to fix things that aren't broken! I'll help if I can as I've just gone through the same thing... but I have a IIx which as a different layout and some component labels are different... even before you get to the real differences!

Note you say about fixing a trace to C6 - on my IIx... C6 isn't the same capacitor! If I'd done that repair I think it would be wrong! Have a check with continuity + following traces and make sure C6 is the part that the schematic says it is and that you haven't made an unintended change. Good luck!

 

[davidg5678]

Thanks for your reply @Phipli!

The information you have provided about resistors is very useful. I'll be sure to remove resistors from the circuit the next time that I have to test them. It sounds like my measurements were probably slightly inaccurate, to say the least.

I have already checked the entire power circuit for continuity using the same website you linked. I did find a couple of broken traces, but these were repaired. On subsequent checks, I could not find any breaks in the circuits. The schematics on Downtown Doug Brown's website are really helpful when working on these computers!

Since there is continuity to all of the resistors in the power circuit and none of them are completely shorted open, I think that I can partially rule them out for now. I think what @techknight was referring to as a floating line was when the pins on UB1 were not soldered correctly and were allowed to randomly jump from high to low. In theory, the floating line should now be fixed, as the randomness of everything has gone away -it just stays broken. I suppose if there were to be another floating line somewhere else, it could be causing issues, but I have made multiple passes of successful continuity checks and microscope inspections that have not turned anything up so far. Maybe I need a logic probe to see what the various gates are doing? (If this is wrong, please correct me, @techknight!)

I hope I haven't been pressing too hard on the pins of components when testing continuity, but it is hard to say if this has created inaccurate results. I'll probably have to check the circuitry for continuity again, and when I do, I experiment with pressing different amounts. The computer's issues are consistent which probably means that whatever is broken remains broken -consistently.

In regards to C6, I don't recall running a wire to it. I think I just confirmed that the traces connected to it are good. The component label deviations between the different models of Mac II, IIx, etc. certainly make working with Mac II computers slightly more confusing!

Please let me know your thoughts -I'd love to hear!

In the meantime, I think I'll try pulling the five resistors with strange readings from the board and test them out just in case they are to blame.