The Double Hard Floppy
This blog entry is going to read like a Control Exception Report - or a 'you fscked up' report. I had a brain wave earlier - when I first got the PPC640 I used Topspeed Modula-2. My thoughts had turned back to my original plan for retrochallenge, implementing code in Modula-3 (which I had been unsuccessful getting working) and I thought, OK, Modula-2 is close enough.
The disk box containing my Topspeed disks has lived in the basement since we moved to this house. This is the first control exception - the basement is not a good place for floppy disks. I tried to read a couple of the disks in the box. This was the second control exception. This was a bad idea, generating a quite alarming grating noise. When I tried to read my known-good DOS 3.3 disk afterwards, the drive generated errors. I tried rebooting, but to no avai. The double hard floppy had struck, and broken a floppy drive. A drive that has lasted in working condition, up till tonight, for approximately 22 years!
So, I take the PPC640 apart - initially going through the bottom of the box, and then realising that the way to the drives lies through a separate door at the top of the main unit. Having found the screw holding the drive enclosure under the cables, I got the drive out and had a play. So, fellow retrochallengers, how could a bad floppy cause a drive to fail? I cleaned the heads with Isopropyl to no avail. I decided that maybe the heads had been knocked out of line, so to enable me to play with screws, I plugged the drive in resting on the metal plate. This was the third control exception - had I checked before turning on the power I'd have realised that two lugs were resting on the bottom of the drive. The result however was not entirely predicable. I lost power. The power supply was broken - no fuses blown, just no power. I found a couple of alternatives and the first powered the computer up to when the floppies kicked in which then overloaded it, the second one I tried with more headroom worked fine. I swapped the first and second drives (it was the first that had broken) but the machine wouldn't boot. This was the the fourth control exception - I hadn't checked for jumpers on the back of the drives, and found when I did that each drive had a jumper in a different place. I swapped the jumper positions. Still no boot. The fifth control exception was created when I put the bottom of the unit together - which must have knocked the DIP switches out of their original positions. Once I'd had a play I got it to boot again.
I then later played with all sorts of screws (probably deeming the drive useless guaranteed for ever more!) but with no joy. I tried swapping the drive for a 1.44MB version (these are 720KB drives) and thought it was working until I realised that I was getting a directory listing back for the A: drive. My guess is that newer floppy drives rely on the cable to determine whether the drive is A: or B: and these drives rely on the jumper.
So, I am one floppy drive and one power supply down. Everything else is working fine. Mind, I was going to give the machine away to anyone who wanted it (for the cost of postage), but I guess that's not the point - it's never nice when a piece of retrokit dies a little.
In control exception reports you need to have preventative actions. Mine would be:
1. Rely on your gut feeling. I should have thrown those floppies in the bin - I knew there was no chance they'd sill be readable.
2. Don't power on a computer with the case open unless you are really sure that there is no chance of shorting out anything.
3. Use the internet - don't press on without giving yourself the best chance of knowing what's what.
4. Don't do electronics late at night.
So, floppy drive, power supply, I salute your long service. RIP.
Below are a few images of my exploits.