Saturday, October 29th, 2005
11:53 pm - Rebuilding Speedy - Episode 2 - If You Build It, He Will Come...  
More rebuild notes:

8:30 a.m. - The case was already in a bit of disarray fron my attempted repair, so finishing the dissembly of Speedy wasn't all that difficult. I disconnected all of the power cables, then removed the old (well, used exactly once) power supply. Detached all the IDE ribbon cables, and then removed all the drives (six of them). Next, I unscrewed the mainboard mounting tray, and slid it out the back of the case. Removed the the CPU fan and heat sink assembly from the motherboard, which surprisingly brought the old CPU with it, despite the lock down lever still being in, er, locked position. Huh. I guess the thermal pad gets sticky after awhile. I removed all the motherboard mounting screws from their brass tray standoffs.

That left one lone nylon standoff/clip in the corner down near the PCI slots. One thing that has always confused me about the design of this case is that there's no location for a solid brass standoff in the part of the case where it is most likely that the motherboard will be flexed while some is is adding or removing a card. Instead, the only thing that works there is a nylon compression fitting that looks like a a little white Christmas tree. You stick the thing up through the tray, where the bottommost branchs spread out to prevent it from falling back through, and then jam down the hole on the motherboard over the top of it until the uppermost branchs do the same. The thing is that once it's attached, it's almost impossible to get it back off, sort of like trying to take a Christmas tree through a narrow entrance top end first. Usually, you have to cut off from the bottom. Problem is, when I checked my spare parts box, I didn't have any more of them, so I had to try to save this one. I got lucky, and was able to squeeze the upper branches together with a needle-nose pliers until they could slip through as I pulled up on the board. Go me! Finally, I vacuumed up the piles of dust on the tray, and got ready to start adding stuff.

10:30 a.m. - But first, I went to the post office, and then upon my return got a message from veejane that there were rumors that Paul DePodesta was about to be fired by the Dodgers. That kept me distracted for a bit, plus there was lunch to think about. One weird thing I discovered was that the web mail interface for my domain won't display the first paragraph of Vee's message (the stuff about DePo). I didn't get to read what she said until I went to reply to something else she'd said, and noticed her first paragraph down in the quoted text. Now I'm wondering what else the interface hasn't displayed. It does provide another reason for me to hurry up and get the new machine up and running

1:30 p.m. - Back to work. I had to move one of the brass standoffs to fit the configuration of the new board, but other than that it's the same pattern as the old board. I fitted the new back panel onto the back of the tray, then attached the motherboard to the nylon standoff, and pivoted the board towards the new back panel, lining everything up, and the snugging the various onboard ports into their slots. Everything fit fine, so I started screwing the board down onto the brass standoffs.

I unsealed the Intel package, and fished out the CPU. The CPU is actually a tiny thing, about an inch square and an eighth of an inch thick, with 478 pins arranged in a square on its underside, except that at one corner it's missing two pins. I lined up the missing pins with the missing holes in the socket on the motherboard, dropped the CPU into place, and closed the locking lever. I swabbed the top of the CPU with a little isopropyl alcohol, just in case there were any fingerprints on it, let it dry, then fitted the Intel supplied heat sink/fan assembly on top of the CPU. The heat sink comes with a thermal pad already attached, so I didn't need to use any thermal transfer gunk on the interface. Fitting this assembly down into its cradle is probably the trickiest part of the entire project. It has to sit just so, and the plastic latches and attachment arms are a bit fragile. Finally, I fitted the four RAM modules into the DIMM slots. They were a tight fit, which is good.

3:00 p.m. - I went out to mow the lawn. It was the first chance I've had since my vacation, and we've had a lot of rain since. The grass was very high. It was fairly nice out when I started, albeit chilly, but then the sun went behind the clouds. I also finally got around to picking onions out of the garden. I only picked about half of them, and left the rest to see how they'll do over the winter. Some had already sprouted again. None of them are particularly big, which is why I want to see if the ones I left there will survive to grow larger next year.

5:30 p.m. - Vacuumed out the case, including the fans. The case is an enormous InWin server case I originally bought when I was building Flash back in 1998. It has five 5¼" drive bays, plus several 3½" bays, and I needed most of them. The most recent configuration included a CD-ROM, DVD burner, and a 5¼" floppy drive, along with a 3½" floppy and two hard drives. At one time I had a SparQ drive in the thing, too. I probably won't install the 5¼" floppy drive this time around, although I might just throw it into Flash. One of these days I need to go through all my old 5¼" disks to see if there's anything I want. If they're still any good, that is. Anyway, the main disadvantage to the case is that it's so tall. I've occasionally had to hunt to find cables long enough to reach from the motherboard in the bottom of the case to the drives up near the top of it. More on that later.

I slid the tray back into the case, screwed it in place, and then connected the leads for the power switch and speaker. Next I installed the power supply, and connected the power leads to the motherboard. I threw in the old video card so I could do a quick systems test.

Meanwhile, the Dodgers held a press conference to announce DePo's firing, or rather that Frank McCourt was making a change. That sounds so much nicer. Apparently the fact that McCourt just cleaned out his PR department and replaced them with an outside firm of political spin-meisters last week was no coincidence. Time for dinner. Also, it's started raining outside. Good thing I mowed the lawn.

8:30 p.m. - The system test worked. The BIOS automatically detected the CPU, so I didn't really have to do anything other than go into setup and confirm its choice. Cool. Now to install some drives.

11:30 p.m. - I've got the drives partially installed. The problem at this point is cabling. The mother board came with two very nice non-ribbony cables, one for the floppy drive, and one IDE cable. The floppy cable is long enough to reach the floppy drive which sits at the very top of the machine, so that's good. However, the IDE connectors are now at the very bottom of the machine, so the supplied cable just reaches the hard drives, less than half way up. I need to get another, much longer, IDE cable to connect the two optical drives. The old ribbon cable is just barely long enough, and ribbons tend to block airflow. This is a bad thing in newer machines, because newer CPUs and video chips generate a lot more heat than they used to. I'll make a quick run up to CompUSA tomorrow to see what they've got. I also need to stop by work and pick up a Win98 startup disk so I can setup the operating system. Unless I decide to go with XP. I just don't know yet.

What I do know is that I have to go set the clocks back an hour, then hit the sack. Busy day tomorrow.
Current Mood: sleepy
( Post a new comment )
vw bug: Boovwbug on October 30th, 2005 - 01:35 pm
Wow. Busy day today too!
(Reply) (Thread) (Link)