Ancient timepieces and the life expectancy of spinning hard drives

[Norm X]

Ancient timepieces and the life expectancy of spinning hard drives

In museums and private collections there are ancient timepieces that still
keep accurate time. A skilled craftsman would fabricate the bearings from
emerald, sapphire or some other extremely hard precious mineral. Silicon is
as durable as any mineral. Unless abused, silicon chips should stay alive
until the end of the earth. I was wondering about the life expectancy of
current spinning hard drives. Eight years ago I had one Seagate that was
warranted for 24 months that died at 25 months, when it stopped spinning. My
current Seagates only get better with time. I've become invested in SSD. My
first Intel was troublesome at six months. My new Sansdisk microSD has great
support. The file system did get corrupted by errant software. But using the
SDHC adapter I was able to reformat using Win7 and miraculously all files
were recovered.

I'm a feeble old man. It is of vital interest to me whether my hard drives
will outlive me or vice versa. I might live for another 30 years.

Comments?

 

[Paul]

Ancient timepieces and the life expectancy of spinning hard drives

In museums and private collections there are ancient timepieces that still
keep accurate time. A skilled craftsman would fabricate the bearings from
emerald, sapphire or some other extremely hard precious mineral. Silicon is
as durable as any mineral. Unless abused, silicon chips should stay alive
until the end of the earth. I was wondering about the life expectancy of
current spinning hard drives. Eight years ago I had one Seagate that was
warranted for 24 months that died at 25 months, when it stopped spinning. My
current Seagates only get better with time. I've become invested in SSD. My
first Intel was troublesome at six months. My new Sansdisk microSD has great
support. The file system did get corrupted by errant software. But using the
SDHC adapter I was able to reformat using Win7 and miraculously all files
were recovered.

I'm a feeble old man. It is of vital interest to me whether my hard drives
will outlive me or vice versa. I might live for another 30 years.

Comments?

Hard drives switched from ball bearings, to fluid (FDB) bearings. And
the motor would now be considered frictionless. At least, until the
fluid in the bearing housing escapes. Which it does.

Many aspects of hard drive design are "perfect", as far as what is
drawn in the engineering drafting tools. But the practical qualities
of the drive, just don't seem to match.

There was an announcement, that a new generation of drives, will be
filled with helium and sealed. (Previous generations were at atmospheric
pressure, and used a "breather hole".) Which is a pretty strange development,
considering how hard it is to keep helium inside a sealed container. I
won't be buying any of those, and will leave the testing to others.
My last two drives (purchased this year), still had breather holes.

The longest lasting drive I've experienced, was a hard drive at
work, which ran for eight years 24/7/365. It was ball bearing
based. And was thrown away, rather than dying. It was in sad
shape, as after a power failure, it needed to be rotated 90 degrees,
so that the motor could overcome the friction present in the assembly.
It would not start if "upright". But if you could get it spinning,
it worked just fine. Still, pretty impressive for a non-FDB design.
Since work is air conditioned, that drive would have been at 40%
humidity for its entire life.

*******

If you have an SSD, it's more likely to die from a firmware bug,
than from worn out Flash chips. The failures evident on SSDs now,
are happening before a significant part of the "wear life" of the
drive has been used. The SSD has a processor and firmware inside.
Hard drives also have this exposure - corrupted data structures
are still an issue with them. The SSD appears to be no different.

Paul

 

[Norm X]

[snippage]

There was an announcement, that a new generation of drives, will be
filled with helium and sealed. (Previous generations were at atmospheric
pressure, and used a "breather hole".) Which is a pretty strange
development,
considering how hard it is to keep helium inside a sealed container. I
won't be buying any of those, and will leave the testing to others.
My last two drives (purchased this year), still had breather holes.

[snippage]

Thanks Paul,

Helium has an extremely high thermal conductivity and it is chemically
inert. It diffuses through solids three times faster than air but for that
to be a measurable phenomenon, a pressure differential (generally one
atmosphere) is required. (Helium comes from gas wells and it has not
diffused to the surface in geological time.)

Helium has about the same viscosity as air. Air is 22% oxygen which is
chemically reactive especially in combination with water vapor. Contained
within the spinning hard drive disk enclosure, its lack of chemical
reactivity would promote longevity of the disk surface and the high thermal
conductivity would protect against thermal shock. Given a one or greater
terabyte HDD in a small package, an investment in helium inside a sealed
container is a good idea. There are fluoroplastics that contain helium well
against a zero pressure differential.

X

 

[Paul]

[snippage]

There was an announcement, that a new generation of drives, will be
filled with helium and sealed. (Previous generations were at atmospheric
pressure, and used a "breather hole".) Which is a pretty strange
development,
considering how hard it is to keep helium inside a sealed container. I
won't be buying any of those, and will leave the testing to others.
My last two drives (purchased this year), still had breather holes.

[snippage]

Thanks Paul,

Helium has an extremely high thermal conductivity and it is chemically
inert. It diffuses through solids three times faster than air but for that
to be a measurable phenomenon, a pressure differential (generally one
atmosphere) is required. (Helium comes from gas wells and it has not
diffused to the surface in geological time.)

Helium has about the same viscosity as air. Air is 22% oxygen which is
chemically reactive especially in combination with water vapor. Contained
within the spinning hard drive disk enclosure, its lack of chemical
reactivity would promote longevity of the disk surface and the high thermal
conductivity would protect against thermal shock. Given a one or greater
terabyte HDD in a small package, an investment in helium inside a sealed
container is a good idea. There are fluoroplastics that contain helium well
against a zero pressure differential.

X

They're also working on "zero clearance" hard drives, where the
head is in contact with the platter. In the lab test drive, it only
took a month of operation, to grind the heads off You won't
be seeing that development, for a while.

The Helium development seems more realistic, even if I don't
like the idea.

In the blurb I read, it also didn't state what the intended pressure
was. Whether it would be 1 ATM or 5 ATM of He, was not stated.

Paul

 

[~misfit~]

Somewhere on teh intarwebs Paul wrote:
[snip]

There was an announcement, that a new generation of drives, will be
filled with helium and sealed. (Previous generations were at
atmospheric pressure, and used a "breather hole".) Which is a pretty
strange development, considering how hard it is to keep helium inside
a sealed container. I won't be buying any of those, and will leave the
testing to others.

From what I have read HGST [WD subsidiary] intend to have helium-filled
drives ready for market before the end of this year. They will be at
atmospheric pressure so that, the majority of them in use will have roughly
equal pressure inside and out. Their biggest concern is helium leakage due
to being transported in bulk in unpressurised aircraft holds - they are
fairly confident that, in use they will have the helium contained.

As the transport issue is only likely to be for a short time it's not as big
an issue as it might be. Still, that is their biggest challenge as far as
they see it.

Cheers,
--
/Shaun.

"Humans will have advanced a long, long, way when religious belief has a
cozy little classification in the DSM."
David Melville (in r.a.s.f1)