Squeal on Older Hard Drives

Billy Pettit Billy.Pettit at wdc.com
Fri Feb 24 12:40:39 CST 2006


Jules Richardson wrote:

> ... although the spring seems to be the norm on 5.25" hard drives,
presumably 
> for cost reasons?

I can't answer that.  None of the 5.25' drives I worked on had a spring.
You are probably right; the spring is way under a dollar, while the ferro
seals were several dollars each.

>> I would never ever bend a spring away from the spindle.  Too many bad
>> experiences.  The noise was a necessary pain.  You can ease it by moving
the
>> spring slightly, so a different part of the button is touching. All the
>> designs I know have oblong holes on the spring, so you can tweak the
>> adjustment.  It's unfathomable to me why an OEM would tell its customers
to
>> bend the spring away from the spindle.

>Hmm, unless on 5.25" drives they have both the spring and the magic seals
for 
>some reason - although I can't think why. If so though, maybe the advice
about 
>bending the spring only applies to 5.25" drives?

Can't answer 'cause again, I have never seen such a drive. But I find it
very hard to believe that a drive manufacturer would design in the static
discharge spring and then recommend it be disabled.  Margins on disk drives
are minuscule.  Every item that can be cut, is.  I know of entire teams
working on 2 and 3 cent cost reductions.  To add an unnecessary part costing
a dollar is madness. 

I have changed dozens of heads and preamps blown by worn springs.  So I'm a
believer.  I admit it.  I've too many bad memories of loose or worn static
springs destroying electronics.  And, by the way, the discharge through the
head creates an intense magnetic field and usually erases a chunk of data on
the disk.  It's a real nasty gotcha.

I've worked daily with hard disk design since 1976, and static is one of the
biggest nightmares in this world.  It gets worse with time.  Moving to the
first MR heads, involved revamping the entire factory, eliminating any
potential charge over 100 volts.  GMR took that down to sub-10 volt.  Now,
with TunnelingMR and Perpendicular heads, the level goes down again. We are
working with flying heights smaller than the wave length of visible light.
So it doesn't take much of a charge to bridge the gap.  And static also has
some very exotic effects on magnetic domains in a modern head; a whole new
industry has grown around suppressing static electricity in disk drive
production.

The older drives you are talking about don't share this new type of
weirdness.  But static was always a major problem because of the disk
ripping through the air.  Static is a natural consequence of the technology.


>I wonder why the squeal happens? Whether it's wear on the spring side or
the 
>spindle side? If the former, then a donor spring from a scrap unit could
fix 
>things.

The springs are designed to wear, albeit slowly.  You do not want any wear
on the spindle.  And it is normally a very hard stainless steel. The button
on the spring is usually an ablative type of material so it will wear before
the spindle.  The squeal comes from uneven wear creating vibration.  The
same as the squeal from worn brakes on your car.  The frequency is usually
determined by the length of the spring and its tension.

>I also wonder if using graphite grease is a viable way of curing the
problem - 
>  or if the grease just wouldn't last long enough or wouldn't provide a
useful 
>ground path...

Graphite was tried.  And various type of grease.  But you are dealing with a
constantly spinning piece of metal.  It usually sloughs off any lubricant
quickly.  And that creates a new problem of introducing contaminants all
over the base of a drive.  Eventually, some of them make it back through the
breather filters or the bearings.  And yes, the static spring had the same
problem, though  to a lesser degree.  It was not the ideal solution by any
means and was eliminated as quickly as possible.

Billy


>cheers

>Jules




More information about the cctech mailing list