[comp.sys.mac.hardware] SE/30 HD problem solution!

stanger@otago.ac.nz (Nigel Stanger) (03/08/91)

Whew! With all the replies I received, I know a lot more about
hard drives than I used to! I am posting the solution to the
problem (if you can call it that) in the hope that it will help
others with the same thing. Thanks to all the people who replied
to my question, especially Dave Platt (ntg!dplatt@apple.com),
whose reply I have appended below:

The original question:
---------------------
In article <1991Mar7.144304.130@otago.ac.nz> you write:
>
>Now here's an interesting problem for all you hardware experts
>out there. I have this SE/30 sitting on my desk. Every now and
>then, the hard drive (standard Apple 40 meg) throws a fit and
>makes a rather distressing continuous rattling noise. It sounds
>very much as if the heads on the drive are moving back and forth
>continuously, but the activity light flashes not. 
>
>We have had it looked at twice now, the first time we were told
>that there had been a bug in the ROM (which part I don't know),
>and so we told them to replace it. This done, the problem went
>away for quite a while, but has resurfaced again recently This
>time, when looked at, nothing wrong could be found. The problem
>is intermittent, and there seems to be no real pattern to
>occurrences, except that it always hapens after a disk access.
>Forcing another disk access usually stops it. It even happens
>sometimes *after* shutting down, i.e. after it says "You may now
>switch off you Macintosh safely."
>
>Has anyone else had this problem, or know what might be causing
>it? The data in the drive is intact, and always has been. What we
>are really worried about is that the drive might wear itself out
>with all this incessant activity. Is that possible? Help!

The answer!
-----------
Quantum (the original disk manufacturer whose 40-meg disks are/were used
by Apple) had a bit of a problem some time ago.  An unfortunate
interaction between a lubricant they used on the drive-head actuator,
and excessively high humidity in the fabrication clean-room, could
result in the lubricant absorbing moisture from the air inside the
sealed disk drive, and becoming "gummy".  This would result in the drive
having problems when moving the head from one track to another... it
would mistrack, and be forced to retry the seek.  Eventually, the
lubricant could get gummy enough that the arm would stick in place, and
the drive controller would not spin up the drive.

Quantum now uses a new lubricant.  They "fixed" the older drives by
installing a new firmware PROM, which [a] uses more current at startup,
if necessary, to unstick the arm, and [b] watches for signs the
lubricant is getting gummy ("seek errors") and institutes an "arm
exercise" sequence to loosen up the lubricant.

What you're observing in your SE/30 drive is very probably the
arm-exercise sequence taking place.  The controller is seeking the heads
back and forth across the drive as fast as it can.  As you've noticed,
this is rather noisy.

There's not really a permanent fix for the problem... if you don't use
your disk heavily enough to keep the lubricant loose, the drive
controller will activate the exercise program on your behalf ;-).

One workaround is to periodically run a heavy disk-exercise sequence
yourself... enough to keep the lubricant loose.  For example, if you
were to install DiskExpress II, and permit it to run a full optimization
of your disk every night at midnight (as it usually does), then you'd
probably never see the exercise program activated during your normal
working hours.  DiskExpress II does a lot of disk I/O, and this would
keep the lubricant nice and fluid.

The long-term solution is to replace your drive with another brand.

----------

Not much of a solution really, but at least I know what's wrong
now... ;-)

See ya

                                Nigel.

/******************************************************************************\
* "If I had a quote, I'd be wearing it."  *  Internet: stanger@otago.ac.nz     *
*       -- Dylan Thomas (I think)         *  SnailMail: Information Science,   *
*                                         *             University of Otago,   *
* "Say no MORE!!!"                        *             P.O. Box 56,           *
*       -- Eric Idle                      *             Dunedin, NEW ZEALAND.  *
\******************************************************************************/