The Cyber-Spy.Com Usenet Archive Feeds Directly
From The Open And Publicly Available Newsgroup
This Group And Thousands Of Others Are Available
On Most IS NNTP News Servers On Port 119.
Cyber-Spy.Com Is NOT Responsible For Any Topic,
Opinions Or Content Posted To This Or Any Other
Newsgroup. This Web Archive Of The Newsgroup And
Posts Are For Informational Purposes Only.
References: <email@example.com> <3DECDEF1.283802B2@webaccess.net>
Subject: Re: Hard drive spins with variable speed ... phase damage????
X-Newsreader: Microsoft Outlook Express 5.50.4920.2300
Date: Wed, 04 Dec 2002 18:03:36 GMT
NNTP-Posting-Date: Wed, 04 Dec 2002 13:03:36 EST
Organization: Cox Communications
"Chuck Simmons" wrote in message
> Mike wrote:
> > "nicolas" wrote in message
> > news:firstname.lastname@example.org...
> > > I need to rescue the data of a SCSI drive (Quantum Viking II 9.1GB
> > > 7200 rpm). It seems the disk has a problem with the motor because I
> > > can hear it spinning at one speed and then it slows down to another
> > > speed, and then again it goes up, and then down again. From what I
> > > have read it could be that one of the motor phases have damage, and
> > > thats why the motor canīt reach itīs 7200 rpm.
> > >
> > > To open the drive I can use a laminar clean bench at my university. My
> > > doubt is the following: Can I speed the motor myself with another tool
> > > and make the disk reach the 7200 rpm and then make an image of the
> > > drive to another disk? Is that possible? Should the disk spin at 7200
> > > rpm exactly ?? or will it work at 7000rpm or 7500rpm for example??
> > >
> > > Thanks for your time, I appreciate any help.
> > > Sorry for my english but it is not my native language. (I am from
> > > Chile).
> > I wish everyone had your command of English, Nicolas.
> > The read channel contains a PLL that needs to lock to the data being
> > from the drive. The pull-in range of the PLL used to be spec'd at 2%,
> > it's often less than 1% these days. This isn't generally a problem in a
> > working drive, since the motor speed is accurate to better than 0.1%.
> > the data rate is proportional to the motor speed, you'll need to keep
> > motor speed within roughly 1% or the PLL won't lock to the data, and you
> > won't be able to read.
> Pull-in range is a minor problem since all of the read channels I have
> seen in recent years use zero phase restart. This was mainly due to the
> wider tracking range required by zoned bit recording. That is, data rate
> depends on disk speed and radial head position. The disk is divided into
> radial zones each having a different data rate (near constant linear
I think we might be talking about different things (I should have qualified
my statement: the pull-in range *when locking to data* is often less than
1%). During idle mode, the PLL is locked with a phase-frequency detector to
a reference frequency that's very close to the data frequency on that
particular track. The range of frequencies in a zoned-bit recording drive is
over 2:1, so the PLL needs to be able to acquire over the entire range.
During idle mode, with the phase-frequency detector, that's no problem.
During read mode, the PLL input is switched from the reference frequency to
the data. The frequency difference between the reference and the data is
well below 1%, so the PLL doesn't need a wide pull-in range at that point.
In peak-detect drives, the PLL may have been able to pull in more than 3%
(as I recall, it wasn't much more), but once a cycle slip occurs, the PLL
won't be able to lock before the end of the preamble. So, we considered that
the onset of cycle-slipping was the limit of the pull-in range.
Zero phase restart was used in desktop drives long before zoned bit
recording was (that may not have been the case in your drives at DEC). For
us, the idea was that acquisition is faster with zero phase restart,
limiting the length of the preamble.
Even with zero phase restart, the pull-in range (when locking to data) on
peak-detect drives was only around 3%, and on sampled data channels it's
significantly less - often less than 1%.
-- Mike --
Go Back To The Cyber-Spy.Com
Usenet Web Archive Index Of
The sci.electronics.design Newsgroup