Path: utzoo!attcan!uunet!lll-winken!ncis.llnl.gov!helios.ee.lbl.gov!pasteur!ucbvax!decwrl!purdue!mailrus!cwjcc!neoucom!wtm
From: wtm@neoucom.UUCP (Bill Mayhew)
Newsgroups: comp.sys.amiga
Subject: Re: IEEE-488 Question.
Summary: I think it is not the spindle motor
Message-ID: <1475@neoucom.UUCP>
Date: 23 Jan 89 14:11:00 GMT
References: <2030146@hpcilzb.HP.COM> <410@madnix.UUCP> <5145@super.ORG>
Organization: Northeastern Ohio Universities College of Medicine
Lines: 33


The spindle motor on the current generations of Seagate (and almost
all other) drives is a DC servo motor that is about the same as
used on direct drive phono turntables.

On drives such as the ST-251R, etc., the problem is apparently the
stepper motor that controls positioning of the heads.  When the
drive powers up, the stepper is commanded to release the head
assembly from park.  There is a fail-safe that does not spin up the
platter if the heads don't release.  At least, that's how it was
described to me.  Sounds backwards to me.  I'd think you'd want the
platter moving before releasing the heads, so that there is
sufficient air motion to fly the heads.  Hmmm.  The repair person
said the poking the head stepper motor shaft would cure the sticky
Seagates.

I have a 20 meg drive here that says "Epson" on it, but appears
that it might have been OEM'ed by Quantum.  This drive suffers from
so-called stiction (sticky + friction ??).  Apparently, the
platters were supposed to be lubricated with some sort of silicone
stuff,... but it didn't work.  When the drive stops, the heads
adhere to the platter in much the same way that machinists guage
blocks stick together due to the ultra flat surfaces.  The only way
to get the drive moving is to poke the flywheel with a screwdriver
because the spindle motor doesn't have enough torque to break the
heads loose.  Miraculously, the drive hasn't gotten any damage from
the sticking heads, eventhough it is not auto-parking.  The drive
electronics for the spindle motor are working correctly.  There was
an article in one of the August 1988 issues of PC Week that said
stiction was relatively frequent failure mode seen by a drive
repair company in the midwest US.

--Bill