It's been a while...
I've had a lot of SGI systems in the last 16 years, but I got rid of everything I had left about 7 years ago when I moved from the Midwest to the PNW. I briefly became active again shortly before Nekochan kicked the bucket, in that brief time, I rounded up a couple of o2's (one care of everyone's favorite curmudgeon, Hamei), and a really rad Tandem re-badged Challenge L deskside (that's a different topic, though...). Well, life got in the way for four or five years and my machines sat untouched and unloved. I can't say why I decided to pull my O2s down off the shelf a few days ago, but I did. After kicking around a bit, I decided the best course of action would be to take all the best parts I had and put them into the chassis with the best plastics and leave the rest as spares...
The best CPU I've got on hand is a 350 Mhz RM7000, so I installed it into the best R5k motherboard I've got, filled it up with RAM, stuck it into the chassis and...nada. Red LED. The O2 boot diagnostic flowchart says it's a bad CPU. But O2s are finicky, right? I spent a while cleaning contacts, re-seating things, swapping motherboards, CPU risers, you name it. I sometimes got it to boot, but it was usually very unstable. Eventually the "sometimes" became "never", and I assumed the worst. Swapping other CPUs into the same board worked. Maybe I'd just settle for a lousy 180 Mhz R5k.
But why quit there? The odd thing was, I was usually able to get it to boot once after cleaning the contacts with alcohol. Not every time, but probably half the time, if I cleaned the CPU connector off real good, waited a bit, then gave it a go, she'd boot. She might even run for a bit, then lock up. Odd indeed. Then I thought, mayhaps...just a little bit of alcohol was hanging out and bridging a marginal contact or broken solder joint between the CPU connector and the daughterboard? See figure A for a visual aide if you've never seen the bottom side of an O2 CPU daughterboard:
(Figure A)
What the hell, I had time during a boring conference call. I installed the CPU riser blocks (see figure B if you have no idea what I'm talking about) into the CPU and did a continuity test on every last pin.
(Figure B)
All seemed continuous (ominous foreshadowing!!!). If it wasn't the connection between the CPU daughterboard and these riser blocks, then surely there was a faulty solder joint somewhere. With no shortage of conference calls at $DAYJOB that I can completely tune out of, I heated up the soldering iron with my trusty SMT tip installed and re-melted every last one of those ~200 contacts. Did it fix the problem? No. Of course not.
So what now? Defeat? Nay! I did what I should have done in the first place. I closely visually inspected the CPU connector on the daughterboard. The picture above isn't the R7k module (it's one of my R5k modules), but the connectors are identical. Except for the fact that maybe 10 or 12 of the little metal fingers inside one of the connectors had lost their spring. They were sitting flush up against the plastic connector housing rather than standing proud like their 190ish other comrades. So I did what anybody in my position would do and raided my wife's sewing kit for a needle and got to work prying them away from the connector housing and back into shape. Once again standing proud and ready to make a solid connection, I re-installed the cpu and...success! I booted into Irix (much more quickly than with the R5k) and away I went. Here's what she looks like now:
I'm currently riding high. Now I just need to dig a PS/2 keyboard and mouse out of storage so I can setenv console g and have the full experience...
I've had a lot of SGI systems in the last 16 years, but I got rid of everything I had left about 7 years ago when I moved from the Midwest to the PNW. I briefly became active again shortly before Nekochan kicked the bucket, in that brief time, I rounded up a couple of o2's (one care of everyone's favorite curmudgeon, Hamei), and a really rad Tandem re-badged Challenge L deskside (that's a different topic, though...). Well, life got in the way for four or five years and my machines sat untouched and unloved. I can't say why I decided to pull my O2s down off the shelf a few days ago, but I did. After kicking around a bit, I decided the best course of action would be to take all the best parts I had and put them into the chassis with the best plastics and leave the rest as spares...
The best CPU I've got on hand is a 350 Mhz RM7000, so I installed it into the best R5k motherboard I've got, filled it up with RAM, stuck it into the chassis and...nada. Red LED. The O2 boot diagnostic flowchart says it's a bad CPU. But O2s are finicky, right? I spent a while cleaning contacts, re-seating things, swapping motherboards, CPU risers, you name it. I sometimes got it to boot, but it was usually very unstable. Eventually the "sometimes" became "never", and I assumed the worst. Swapping other CPUs into the same board worked. Maybe I'd just settle for a lousy 180 Mhz R5k.
But why quit there? The odd thing was, I was usually able to get it to boot once after cleaning the contacts with alcohol. Not every time, but probably half the time, if I cleaned the CPU connector off real good, waited a bit, then gave it a go, she'd boot. She might even run for a bit, then lock up. Odd indeed. Then I thought, mayhaps...just a little bit of alcohol was hanging out and bridging a marginal contact or broken solder joint between the CPU connector and the daughterboard? See figure A for a visual aide if you've never seen the bottom side of an O2 CPU daughterboard:
(Figure A)
What the hell, I had time during a boring conference call. I installed the CPU riser blocks (see figure B if you have no idea what I'm talking about) into the CPU and did a continuity test on every last pin.
(Figure B)
All seemed continuous (ominous foreshadowing!!!). If it wasn't the connection between the CPU daughterboard and these riser blocks, then surely there was a faulty solder joint somewhere. With no shortage of conference calls at $DAYJOB that I can completely tune out of, I heated up the soldering iron with my trusty SMT tip installed and re-melted every last one of those ~200 contacts. Did it fix the problem? No. Of course not.
So what now? Defeat? Nay! I did what I should have done in the first place. I closely visually inspected the CPU connector on the daughterboard. The picture above isn't the R7k module (it's one of my R5k modules), but the connectors are identical. Except for the fact that maybe 10 or 12 of the little metal fingers inside one of the connectors had lost their spring. They were sitting flush up against the plastic connector housing rather than standing proud like their 190ish other comrades. So I did what anybody in my position would do and raided my wife's sewing kit for a needle and got to work prying them away from the connector housing and back into shape. Once again standing proud and ready to make a solid connection, I re-installed the cpu and...success! I booted into Irix (much more quickly than with the R5k) and away I went. Here's what she looks like now:
Code:
bash-4.2# hinv
CPU: QED RM7000 Processor Chip Revision: 3.3
FPU: QED RM7000 Floating Point Coprocessor Revision: 2.0
1 350 MHZ IP32 Processor
Main memory size: 1024 Mbytes
Secondary unified instruction/data cache size: 256 Kbytes on Processor 0
Ternary unified instruction/data cache size: 1 Mbyte on Processor 0
Instruction cache size: 16 Kbytes
Data cache size: 16 Kbytes
FLASH PROM version 4.18
Integral SCSI controller 0: Version ADAPTEC 7880
Disk drive: unit 1 on SCSI controller 0
CDROM: unit 4 on SCSI controller 0
Integral SCSI controller 1: Version ADAPTEC 7880
On-board serial ports: tty1
On-board serial ports: tty2
On-board EPP/ECP parallel port
CRM graphics installed
Integral Ethernet: ec0, version 1
Iris Audio Processor: version A3 revision 0
Video: MVP unit 0 version 1.4
AV: AV1 Card version 1, Camera not connected.
Vice: TRE