On Apr 15, 2024, at 1:15 PM, Tom Uban via cctalk
<cctalk(a)classiccmp.org> wrote:
I recall around 1980, the "A" machine at Purdue University Electrical
Engineering, a PDP-11/70 running Version 7 Unix had a RS04 drum drive used for swap. It
was getting long in the tooth and when a power failure occurred, someone would have to get
a wrench to help spin it up as the head lubricant was no longer as good as it once was...
RS04 is a fixed head disk, similar properties as a drum but shape-wise it's a flat
platter with heads on one side.
The "lubricant" bit reminds me of our college RF11 swap disk (for RSTS/E). When
we got the RP04 it was no longer so interesting so it was not configured in, but it was
still powered up.
One day I noticed that the "timing track error" light was lit, and since the
system was under contract we called DEC field service to repair it. The local tech
checked things out and realized the drive was not spinning even though it had spindle
power. Oops. He took the drive apart and discovered that the heads had landed, and
melted so they were essentially hot-melt-glued to the platter. Those drives use rather
low powered motors, intentionally, so with the heads stuck it would not spin up.
He ended up replacing all the heads and the platter, not sure about the motor. Got an
alignment disk and timing track writer from Maynard, and had to figure out how to use
these -- the documentation was sparse, to put it politely. In the end, everything worked.
Oh by the way, I don't think that field repair of a drive like that was a standard
procedure, but for Jim Newport this wasn't a big deal.
paul