Re: disk drive spin direction

From: AEF (spamsink2001_at_yahoo.com)
Date: 01/21/05 

Date: 21 Jan 2005 10:58:09 -0800

David Mathog wrote:
> Keith Cayemberg wrote:
>
> > I have now found a much better description of the "flying magnetic
head"
> > technology...
> >
> >
http://www.storagereview.com/guide2000/ref/hdd/op/heads/opHeight.html
> >
> > Modern drive heads float over the surface of the disk and do all of

> > their work without ever physically touching the platters they are
> > magnetizing. The amount of space between the heads and the platters
is
> > called the floating height or flying height. It is also sometimes
called
> > the head gap, and some hard disk manufacturers refer to the heads
as
> > riding on an "air bearing".
>
> Special disk drives or pressured rooms are required at high altitudes
or
> said "bearings" will fail resulting in a head crash.
>
> Back to the original question at the start of this thread (clipped
out
> somewhere along the line), I think the original poster was asking,
for
> a drive like this:
>
> http://en.wikipedia.org/wiki/Image:Hard_disk_dismantled.jpg
>
> if you imagine the arm to be a vector, with the arrow pointing
> towards the head, then is the disk spinning under the head moving
> into the vector or away from it?

YES! Thank you for clarifying my question. I did try to be clear.

> I don't know the answer to that. However, that air bearing is going
to
> exert two force vectors on the head.
>
> 1. The one perpendicular to the disk. This is the air bearing
effect.
> 2. The one in the plane of the disk, normal to the track where the
head
> is currently positioned, in the direction the disk is turning.
This
> one is due to drag from the air layer attached to the suface of
> the disk.
>
> Naively I'd expect it would be easier to maintain arm
> stability if that (unit) second vector was in the same direction as
the
> (unit) arm vector (dot product ~= 1). On the other hand, it might
well
> be easier to reduce the seek time if that vector was in the
> exact opposite direction (dot product ~= -1). There's precedent
> for the latter type of design: some modern fighter planes are
> intrinsically unstable because it improves maneuverability,
> they are only flyable because of constant computer control.
>
> If I had to bet it would be that the dot product is ~= 1.
>
> Now I'm curious too. Is it 1 or -1?
>
> Regards,
>
> David Mathog

Well, I got a broken IDE drive from our help desk. I plugged the SCSI
disk power supply from a MicroVAX 3100 Model 80 into the drive and
powered it up (I had my safety goggles on, of course ;-) At first the
platter just swung slighty back and forth. Then there was a brief
pause. Then it sped up CCW, with the direction of the head assembly. So
the dot product is very close to 1. I hope there is no difference in
the power supply for IDE and SCSI so that my test is valid. For now I'm
assuming it is valid. (Correction welcome!)

If it were -1, I assume it would be easier to find that out on the Web
as it would be contrary to most people's expectations. (And perhaps
this is why some responders didn't correctly interpret my question!) So
I'm going with +1.

I read some of the links provided by Keith and I find that disk drives
are truly wonders of technology. Why can't tape drives be even one
tenth as good?! (And by "good" here I mean quality-wise, not
capacity-wise.)

Thanks to the responders and my company's help desk.

Relevant Pages

  • Re: aligning the heads on a 5.25" floppy disk drive
    ... To properly align a drive you need an alignment disk and a scope. ... head cleaning disk. ... On some drives you can pull the heads ...
    (comp.os.msdos.misc)
  • Re: off topic - disk crash
    ... > IDE disk crashed. ... This specific line of drives is infamous for a failure rate that's at ... to recover data for our babysitter's computer, ... Yes, this might actually help, provided the read head is not yet too ...
    (freebsd-hackers)
  • Re: Bad Clusters
    ... On one of the external drives, ... indicating the clusters were not really fixed by Windows, ... > head positioning wobble that reduces accuracy of pickup, ... i.e. where the head touches the disk and thus damages it. ...
    (microsoft.public.windowsxp.general)
  • Re: Bad Clusters
    ... >drives I use for backup. ... head positioning wobble that reduces accuracy of pickup, ... Why would a previously good HD develop a bad sector? ... i.e. where the head touches the disk and thus damages it. ...
    (microsoft.public.windowsxp.general)
  • Re: Perhaps SCO has a legitimate gripe
    ... >>you, when they realize its their head on the block, they usually ... One department orders a new Model 80 IBM and he was heard ... > That machine was to be their first AIX on Intel. ... So some of these disk sets were inserted ...
    (comp.unix.sco.misc)