68kMLA Classic Interface

This is a version of the 68kMLA forums for viewing on your favorite old mac. Visitors on modern platforms may prefer the main site.

Click here to select a new forum.
Portable 5126 Logic Board Debugging
Posted by: jmacz on 2026-09-14 19:06:17 I am working on a Macintosh Portable model 5126 (backlit version) which I received. It came with two power adapters both producing the proper voltage and look to be working. It came with a battery that was dead. After removing the logic board, the caps were original and still leaking. I removed all the caps, gave it a good IPA bath, and replaced the caps. In terms of my setup, I am testing the logic board by giving it 6.5V @ 1.5amps coming from a bench power supply and feeding into the 4 pin connector on the logic board (near the SCSI connector). I am giving both power pins the 6.5V and the other two are ground. I am powering up the logic board by temporarily shorting a couple pins on the IO connector (where the keyboard connects). And for reference, I'm using the schematics for the 5120 model since I could not find one specific to the 5126 model but I heard things were close enough for the major components. I have already found a difference with the RAM pin for AS from the schematics (5126 vs 5120). Problem #1: Fuse I used my bench supply to give the logic board power and heard no chime. Tried shorting the keyboard pins and nothing. No current draw. I then checked power delivery to the 4 pin connector on the logic board and power was getting there. I then tested the voltage regulators, nothing. I then noticed the F1 fuse on the logic board near the connector and checked it, no continuity. It's blown. 7A fuse blown. Uh oh. That's a lot of current. I didn't have any other 7A fuses but I did have a 2A fuse. Since my bench supply was set to limit at 1.5amps, I decided to move forward with using the 2A fuse for now. With it replaced... I saw proper voltage at the voltage regulators including around 5.22V at the regulator providing 5V to the logic board components. Problem #2: Chime Initially I was getting no chime but could see that the board was drawing current from my bench supply. I plugged in headphones and heard a good chime. So why wasn't the speaker working. Tried multiple 32 ohm speakers (like stock) and nothing worked. But the speaker pins were connected properly to the circuit near the headphone jack. I then used deoxit on the speaker pins and that did the trick. Got good chime coming out of the speaker. Problem #3: No Progression During Boot I then connected the LCD display and found the display was working. Doesn't look like any dead pixels and it's a nice image. After power on, a cursor eventually shows up and with an external ADB mouse connected, the cursor moves and tracks my mouse movements. But no question mark floppy disk icon. Even with an external SCSI device plugged in (ZuluSCSI -- with external power provided since there's no terminal power on the Portable), nothing happens. I figured this must be something with the SCSI chip or data lines or something. I beeped out the SCSI, SWIM, VIA chips and found all of the pins had continuity with each other and other components they were supposed to connect to. But as I was ensuring each pin connected to ALL of the other places on the board, I found the AS connection was there between SWIM, VIA, CPU, and CPU GLU chips, but not with the MISC GLU chip. The schematics (5120 version) says the MISC GLU chip pin 7 is AS. But there's no continuity between it and those other chips. I could not see any traces coming out of pad 7 for MISC GLU and no vias under that pad on the bottom side of the board. I tried a bodge wire between MISC GLU pin 7 and VIA pin 25. Then powered on and finally got a question mark floppy icon! I then attempted to boot from the ZuluSCSI with System 6.0.8. Success! Got to the finder. Saw 1MB of memory. Ok great. I then did a restart via the Finder menu, the system reset but I got two chimes back to back and then the system was hung with no question mark floppy icon again and no boot.... argh. Problem #4: Restart Doesn't Work A cold power on will get to the question mark icon or will boot. A warm restart (via Finder or via the reset button) gets stuck with no question mark icon or boot. But the cursor still moves and is responsive.
Posted by: jmacz on 2026-09-14 19:16:38 Current Investigation There was nothing obvious to me regarding what's not working. Voltages seemed fine, even under an oscilloscope. Something about that AS line was bothering me because I saw no obvious breaks. I decided to remove the chips on the board where the leaking capacitors were. I found a good amount of cap juice still under the chips and cleaned it all up. The chips removed were the ASC, VIA, and PowerManager chips. After clean up, I didn't see any broken traces under the chips. Vias look ok. I scratched away the solder mask in a few areas just to be sure of continuity and things look good. The red arrows indicate where the AS connection points are (according to the 5120 schematics). All of them checked out EXCEPT the pin 7 on the MSIC GLU chip (bottom center of the picture above). The yellow lines indicate where the top of board and bottom of board traces are for AS. Note I'm missing one yellow line in the diagram between the CPU and the SWIM chip. The point is I get AS connectivity between the CPU, CPU GLU, SWIM, VIA, ROM, and PDS slots. I do not get anything on the RAM slot but I think this might be because the 5126 is different possibly as the AS pin on the RAM slot on my logic board is connected to an IC on another part of the board and I think I read that the 5126 handles memory slightly differently? Not sure. So I removed the MISC GLU chip as well. The above is a clean picture and the below is with annotations. Pin 7 on MISC GLU which is the AS pin is marked above with the red arrow (left one). And pin 25 on the VIA chip is the other red arrow. Note that pin 26 on the VIA chip looks a bit funky but I confirmed it's just some solder I didn't fully clean off. Same goes on the other pads for the VIA Chip. Problem is, that pin 7 pad is not connected anywhere on the top side of the board. There's no via from under the pad coming out the bottom side of the board either. I think pin 7 must be connected to AS, but given I see nothing on the top and nothing on the bottom layers, it must be embedded in a middle layer (which I think are power and ground so it must be isolated on one of those two planes?). But no obvious breaks here that I can see which would cause that pad 7 to be disconnected from the other AS pads/vias. Not sure if anyone has a 5126 logic board sitting out that can test if their board has pad 7 / pin 7 on the MISC GLU connected to either pin 25 on the VIA chip or pin 6 on the CPU? cc two folks who have extensive experience with these boards: @techknight @David Cook
Posted by: obsolete on 2026-09-14 20:43:34 I don't have any specific experience with the Portable, but I don't think anyone was doing buried vias or via-in-pad back then, so that pin 7 pad you highlighted looks like a true "no connect" to me.
Posted by: jmacz on 2026-09-14 23:04:05 I don't have any specific experience with the Portable, but I don't think anyone was doing buried vias or via-in-pad back then, so that pin 7 pad you highlighted looks like a true "no connect" to me. That’s what it looks like but I am scratching my head over why it’s connected on the 5120 model according to Apple’s schematics but would not be on the 5126 backlit version. And why adding a bodge corrected the no boot issue unless that was coincidence and something else changed.
Posted by: SuperSVGA on 2026-09-15 01:03:37 Problem is, that pin 7 pad is not connected anywhere on the top side of the board. Pin 7 is normally not connected because the MISC GLU is not used for RAM decode anymore.
Posted by: jmacz on 2026-09-15 06:53:53 Pin 7 is normally not connected because the MISC GLU is not used for RAM decode anymore. Ok, same reason the RAM slot isn’t directly connected to AS on the 5126. Thanks for the confirmation!
Posted by: jmacz on 2026-09-20 22:40:50 Put the chips back on. Haven’t had a chance to debug further… but one win today. The chime was a bit garbled at times. Found out when it happens, the sound chips aren’t getting a full 12 volts and flexing the board changes the behavior. Found a hairline break in a trace coming off an inductor near the sound chips. Sound is steady now.
Posted by: jmacz on 2026-09-28 21:43:04 Did a quick continuity check on the major signals and those are fine. Checked traces around caps and they check out. Current Observations: Long delay before boot Initial power on seems ok except it takes about 2 to 2.5 minutes to detect the SCSI device and start booting from it Floppy boot attempt has the same issue Tried with and without scsi and floppy device in combinations but it's the same Unstable restarts either from Finder or via reset button Either restarting from the Finder or via the reset button gives me unstable results.. sometimes it hangs without showing the cursor. Sometimes I get two consecutive chimes followed by a hang. Things "seem" fine after boot Booted into System 7.1 and ran some programs and they seem to work fine? But it's not fine ... I noticed the zoom rectangle when opening folders in the Finder are not smooth.. the zoom rect animation is jerky and sometimes slow I launched the Tetris game... upon load, there's an animation and then the game plays a song. This song is a little garbled and it slows down, speeds up, it's not fine. What's interesting is if I move the mouse while the song is playing, I notice it does make the song play faster (back to normal speed) with no garbled sound.. it's not perfect but I can get the song to sound more normal for periods while I'm moving the mouse but if I leave the mouse alone, the song is in bad shape. Feels like something might be wrong with a clock signal? But I put a scope on the cpu clock (16MHz), the main oscillator at 32MHz, the 60Hz signal, all of those seem fairly crisp and fine. I don't see any change in the wave form while moving the mouse for any of these. If anyone has any ideas, let me know. Ran out of time playing with it this weekend so need to find some more time to play with it.
Posted by: jmacz on 2026-09-29 19:31:21 Wow.. fixed it. My M5126 Portable seems to be fully working now. I spent time using a logic analyzer to watch the AS, DTACK, and a few other signals. All of those looked good. I watched the interrupt/irq signals and those also looked ok. I then decided try something more old school.. started tapping on various locations on the logic board to see if I could get a clearer sound like when I was moving the mouse. I failed for the most part until I got around the CPU and noticed I got a clear sound every so often. But it wasn't the tapping. It turned out it was touching, and in one corner of the CPU. Given my finger was changing the behavior and all the pins in that area were properly connected, I figured it was a floating signal. I used a disconnected probe and found the pin... IPL0. It was connected but touching it with the disconnected probe cleared the sound. I traced it to the two spots it connects on the logic board. One was on the PDS connector. Touching the probe there also cleared the sound. The other place was the CPU GLU chip. No change when I touched the IPL0 pin on the CPU GLU chip. Must be a break. There's a trace running from the pin on the CPU GLU chip to a nearby via. The via has continuity to the CPU and PDS slot, but not the chip. I used a microscope but I couldn't see any breaks in that 1 centimeter segment of trace from the via to the chip. I scratched off the solder mask in the middle of the trace. That part had continuity with the via but not the chip. So looked like it was around the pad. I used a bodge wire. That was it! Sound was crystal clear and all my restart issues and crashes went away. Everything seems to work. IPL0 is used for the interrupt priority level so the fact that it was floating was messing things up. I don't know why there's a broken trace here as it's no where near any caps. Now hopefully that was the only other issue.
1