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.
Anyone up for some IIcx troubleshooting assistance? We have clocks but no activity...
Posted by: Sideburn on 2023-07-03 23:47:40
Some isopropyl cleaned it up some image.jpg
Posted by: GRudolf94 on 2023-07-03 23:47:43
What @Phipli said. I just scrape the dark parts with the meter probe, and tin over - using an x-acto feels a bit aggro. Wire only goes over confirmed breaks. UV mask is a thing I usually skip as I find it usually just makes things uglier (it's only really needed in certain things).
Posted by: Sideburn on 2023-07-03 23:49:55
If there isn't actually a break, all you need to do is tin them to add some material and protect them.
There’s enough gap to need some wire
Posted by: Sideburn on 2023-07-03 23:50:53
There’s enough gap to need some wire
Oh yes you mean the iffy ones. Yes I’ll scrape then clean and coat with solder and UV mask them.
Posted by: Sideburn on 2023-07-03 23:52:35
What @Phipli said. I just scrape the dark parts with the meter probe, and tin over - using an x-acto feels a bit aggro. Wire only goes over confirmed breaks. UV mask is a thing I usually skip as I find it usually just makes things uglier (it's only really needed in certain things).
I use the probes too like that. But some are so bad they need careful scraping.
Posted by: GRudolf94 on 2023-07-03 23:52:53
Btw, I'll take the pin measurements and figure out what they mean in a sec.
Posted by: Sideburn on 2023-07-03 23:53:27
I use the probes too like that. But some are so bad they need careful scraping.
I also have a fiberglass pen. Those work great too.
Posted by: Sideburn on 2023-07-03 23:58:42
Uf13 is warm
Posted by: Sideburn on 2023-07-04 00:02:28
It also has a wonky square wave on pin 32 @ 4mhz
image.jpg
Posted by: Sideburn on 2023-07-04 00:05:23
Guess that’s not important since it’s I/o ports
Posted by: GRudolf94 on 2023-07-04 00:18:21
Pin 32 of what, exactly?
Posted by: GRudolf94 on 2023-07-04 00:33:11
Check pin 3 on those same ICs. However something's odd. All address lines generated by NuChip are quiet, but you're telling me those coming from the system side are asserted, or at least something is forcing them on.

Check UDs as well, feel free to check both A and B sides. I have a suspicion this might be close to the issue, especially considering how bad traces around that area look.

1688455806604.png
Posted by: Sideburn on 2023-07-04 00:34:37
Pin 32 of what, exactly?
Uf13
Posted by: GRudolf94 on 2023-07-04 00:36:12
Uf13
Ah that's far off from where we're looking. It's SCC B transmit clock. 4MHz-ish sounds right.
Posted by: Sideburn on 2023-07-04 00:36:54
On it. So when I hit reset, the address lines on the roms go high and then back to low. Does this mean the CPU is not “dead” ?
Posted by: Sideburn on 2023-07-04 00:37:24
Ah that's far off from where we're looking. It's SCC B transmit clock. 4MHz-ish sounds right.
Yeah I figured after checking schematics
Posted by: Sideburn on 2023-07-04 00:41:00
Check pin 3 on those same ICs. However something's odd. All address lines generated by NuChip are quiet, but you're telling me those coming from the system side are asserted, or at least something is forcing them on.

Check UDs as well, feel free to check both A and B sides. I have a suspicion this might be close to the issue, especially considering how bad traces around that area look.

View attachment 58924
Uc9 pin 3 - high
Uc13 pin 3 - low
Posted by: GRudolf94 on 2023-07-04 00:42:50
On it. So when I hit reset, the address lines on the roms go high and then back to low. Does this mean the CPU is not “dead” ?
No, it doesn't mean anything - somewhere earlier I mentioned broken silicon may still act in ways that seem sensible. While reset is asserted, the system is in undefined state. What reset does is clear the slate and get it all into a known default state the machine can start running from. That would include throwing the CPU at its reset vector, what on the m68kis called an IISR, which makes it go to the bottom of the address space (0x00000000) and then start executing from 0x04. As you can imagine, that location is physically pointed at by an all-low address lines combined.

A broken CPU in an otherwise functioning machine way well jumble the lines upon having reset asserted, and then beautifully clear them once reset deasserts, but never go from there. A broken CPU may have broken address line drivers. Something could short one of those address lines and force them to do weird things. It goes on.
Posted by: joshc on 2023-07-04 00:44:49
I seemed to recall those being socketed.
Earlier production IIcx boards have socketed CPUs, later ones do not. Same for the SE/30.
Posted by: Sideburn on 2023-07-04 00:46:26
Now that the boards been on for a while the uc and UD chips are getting pretty warm/ hot

That red one in the bottom center is uc12. 45 degrees Celsius
< 6 >