Is anything known about the layout of the PowerBook 170/140 motherboard?

[PB170]

Does anyone here have any information about the layout of the PowerBook 170/140 motherboard?

Since a couple of months back, I've had problems with my PowerBook 170, somehow related to the AC connector (I've been using the computer more or less daily since I got it back in 2001, averaging about 1000 hours/year, so I'm quite impressed that it lasted this long).

Power to the computer is never interrupted but the voltage varies between 6.5 and 7 V (as opposed to the normal ≈7.6 V). Moving the power cord causes the backlight to flicker, the speaker to pop and sometimes also disturbs the video signal, although this behavior is not constant; sometimes it's fairly insensitive and at other times even the slightest movement is enough. The voltage, however, is constantly at between 6.5 and 7 V. Curiously, when the voltage is lower, the brightness of the backlight gets brighter at all but the highest settings. When running off the battery the computer behaves normally.

At first I assumed the connector had started to get loose from the board, but upon inspection all solder points looked good and the connector is firmly attached to the board. As a first attempt I tried to simply remelt the solder. This helped slightly at first, but the problems continued.

Following that, I decided to desolder the connector and change it to a known working one from a spare PowerBook, in hopes that this would solve the issue permanently. At first everything seemed to be back to normal, but after a couple of days the problems returned.

Since problems with the connector itself are ruled out, I assume that the traces or vias surrounding it have been damaged in some way. My idea at this point is to solder wires from the pins on the connector to the end of the closest direct trace on the board to make an extra connection (preferably on a free pad somewhere). The problem is, I'm unable to see exactly where the traces go. There appears to be at least one additional layer inside the board, which makes it hard to inspect. From what I can tell there seems to be a positive power layer inside the board and ground traces on the top and bottom, but I'm not sure.

So, my question is: does anyone here have any information about the layout of the board? Also, any ideas are greatly appreciated.

Attached are two pictures of the top and bottom side of the board where the connector is located.

 

[Macdrone]

have you recapped the motherboard? All the 100 series except the 150 should need to be done at this point.  There are not many that I remember exactly, also the screen has some inside also.  Also the power supply brink has some to replace so definatly do a search on that as many posts before doing much more and chasing your tail.

 

[PB145B]

have you recapped the motherboard? All the 100 series except the 150 should need to be done at this point.  There are not many that I remember exactly, also the screen has some inside also.  Also the power supply brink has some to replace so definatly do a search on that as many posts before doing much more and chasing your tail.

Only the 100 LB should need to be done. 140 and up are all tantalum on the LB. The passive matrix models do have electrolytics on the screen and inverter that should be replaced. The 170 is also all-tantalum on the LCD as well. There is 1 electrolytic on the 170’s inverter that may need replacing.

 

[bibilit]

I agree with Macdrone, i will check the power brick first, there is a lot of nasty capacitors inside those, all the ones i have seen so far were gone.

The PSU is probably too weak when load is required and this can be the source of your issue, logic boards are bullet proof.

 

[PB170]

I feel I have to intervene a bit here

Like PB145B said, there are no electrolytic capacitors on the PB170's motherboard. And the power source is a new switched power supply (inside the original case), so the problem is not there. I've also tried different power supplies with the same symptoms.

To me, everything points to it being a problem with conductivity somewhere since everything works fine if the plug is positioned carefully and the cord is not moved (or when running off the battery).

 

[Macdrone]

i did not say electrolytic caps, but caps or cold solder joints can go bad after this much time.  And if that isnt the issue, many traces can be suspect as green wire fixes are pretty common on these I had found over time, not as much as in the 100 itself, but they still had them.

Its great if the power supply isnt the issue.  I have plenty of them but they do go bad quickly.

 

[PB145B]

I would just replace the mainboard at this point. The 140, 145, 145B and 170 all share the same mainboard, so finding one should be very easy and cheap.

Cold solder joints are very possible though, but it’ll be a needle in a haystack to find.

 

[PB170]

Thank you for your input!

I've been so focused on problems with the pins from the power connector and the surrounding vias and traces that I haven't considered problems with the solder joints elsewhere. I will look into that.

This past weekend, I had the idea to take voltage measurements on the board while the computer is turned off but plugged in, to more easily determine possible direct paths to the power connector. The unoccupied solder point next to pin 20 on the modem connector shows exactly the same voltage as on the positive input pin so I made the assumption that this is connected directly to the connector and went on and soldered an extra lead between the two points (photo below). I also added a lead between the negative pin and the ground traces.

Unfortunately, while I guess my assumption was correct since the computer still runs fine with the extra leads, this didn't solve the issues I'm having. Considering that the issues point at there being a problem with connectivity somewhere, I guess the next logical step would be to remelt the solder around the other components on that end of the board. Hopefully that will help.

i did not say electrolytic caps, but caps or cold solder joints can go bad after this much time.  And if that isnt the issue, many traces can be suspect as green wire fixes are pretty common on these I had found over time, not as much as in the 100 itself, but they still had them.

Its great if the power supply isnt the issue.  I have plenty of them but they do go bad quickly.

Would you say non-electrolytic capacitors are more likely to fail than other components, such as power transistors etc.?

Also, the traces themselves wouldn't likely go bad, unless damaged by leaking caps/batteries or other external damage, would they?

I would just replace the mainboard at this point. The 140, 145, 145B and 170 all share the same mainboard, so finding one should be very easy and cheap.

I have several spare PowerBooks so I could definitely do that, but lately I've rather been trying to fix the problems that occur, since I realize I will eventually run out of spares

 

[PB170]

Success, at last!

A couple of weeks ago I decided to go ahead and re-solder all the solder points on both sides the protruding part of the motherboard to see if that would solve the issues I've had. After reassembling the computer, the operation seemed to be successful and everything was back to normal again. However; four days later the problems returned…

At this point I was wondering whether just taking the board out and putting it back again was enough to temporarily fix the problems, or if my work actually improved the situation. (Actually, each time I've worked with board the problems have temporarily gone away).

In lack of other options, I decided to re-solder the solder points on all of the remaining components on the top of the board this past weekend. Just as I had finished the work and was inspecting the board, I noticed what looked like flux residue around two solder points not connected to any component (and untouched by me), just where the protruding part of the board meets the main part (see the first photo). Upon closer inspection, the trace between the two solder points goes to two vias, one of which was only partly soldered, while the other had too much solder (photos 2 and 3). Indeed, after comparing it with another board (photos 4 and 5), it clearly seems to be a production fault. I re-soldered the vias and put the computer back together, and voilà, problems gone and everything back to normal again!

Who would have guessed the cause was an error during production just waiting to cause problems, rather than broken solder joints or failing components.

So if you ever have similar problems, that's another thing to look for!

 

[PB145B]

Awesome! Glad you found the issue!