Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!uwm.edu!gem.mps.ohio-state.edu!uakari.primate.wisc.edu!ames!pacbell!att!ll1a!cej
From: cej@ll1a.att.com (Jones)
Newsgroups: comp.sys.mac
Subject: Re: Apple to Fix Quantum Drive Failures
Message-ID: <7118@ll1a.att.com>
Date: 13 Oct 89 01:43:30 GMT
References: <5327@shlump.nac.dec.com>
Reply-To: cej@ll1a.att.com (Jones)
Organization: AT&T, Lisle, Il.
Lines: 50

In article <5327@shlump.nac.dec.com> long@rainbo.enet.dec.com
(Richard Long) writes:
>In article <957@bridge2.ESD.3Com.COM>, ngg@bridge2.ESD.3Com.COM
>(Norman Goodger) writes... 
>>
>>Problem: [HD death due to "sticktion"]
>>Solution: An extended repair program that will continue to October 1991. 
>>The fix will replace a ROM on the HDA controller PWA. Users that have
>
>umm...err...maybe I'm missing something, but I don't see how the problem
>cited above could be solved by a software fix; sounds like a hardware
>problem to me...

	Actually I can imagine how a ROM swap might help (but *not*
fix) the problem.

	It takes much more power to get a hard disk up to speed than it
does to keep it spinning, even without the spindle sticking.  (Ever
notice the similarity between the platter/spindle assembly in a hard
drive and a flywheel?)

	Old, large, hard drives used pretty good sized power-amplifiers
and discrete timing circuits to feed several times more power to the
spindle motor during spin-up.  (Forget the price per Meg, do you realize
the reduction there's been in weight in the last decade?  Years past, I
dragged around 2.5M drives that weighed 60 lbs.)

	(Since I haven't worked on disk drives for a few years the
following is a bit of speculation.)

	In new drives I would imagine that rather than discrete
circuits, the start-up current and it's duration is determined by the
ROM.  So, if this is true, a new ROM might either increase the start-up
current (if possible), or increase it's duration.

	Or, if the the logic detects the fact that the drive has not
started to spin in some maximum time, the ROM could "pulse" the current
to the motor, attempting to "rock" or "jar" the spindle loose.

	[Again, this is just educated speculation, but if it's true, I
just hope the spindle motor is rated to handle the additional power.
And I'm only speculating since no one who has the inside scoop seems to
be talking.]

...att!ll1a!cej  Llewellyn Jones  [Just me, not AT&T]  cej@ll1a.att.com
>> This article may NOT be forwarded by the 'In Moderation Network' <<

How many would vote for putting Pete Rose in the Hall of Fame,
	*and* taking his picture *off* of Wheaties?  -  Mark Russel