Length of warranty length versus expected lifetime of disks

Re length of warranty versus expected failure rate and disk
life:

One or more of Scott Moulton mentions that some lubricant
evaporates and is supposed to be recycled inside of the drive.

He suspects that not all of the lubricant will be recycled
and that therefore the lifetime of the drive is limited.

Not specifically stated, but implicit in the comments, are
that there are other factors that may lead to end of life
of the disk, such as bearing life for the motor and
other moving parts. (What we [not working for the disk manufacturers]
don't know is if the life of the motors is 5 years and the
life of the lubricant is 50 years or the other way around.)

My additional take on things is that things eventually get
balanced. Thus, with a 3 years from date of manufacture
warranty the life of the lubricant and the motors will both tend
to 3 years from date of manufacture rather than the old 5 years
spinning life.

Mark F wrote:
Re length of warranty versus expected failure rate and disk
life:

One or more of Scott Moulton mentions that some lubricant
evaporates and is supposed to be recycled inside of the drive.

He suspects that not all of the lubricant will be recycled
and that therefore the lifetime of the drive is limited.

Not specifically stated, but implicit in the comments, are
that there are other factors that may lead to end of life
of the disk, such as bearing life for the motor and
other moving parts. (What we [not working for the disk manufacturers]
don't know is if the life of the motors is 5 years and the
life of the lubricant is 50 years or the other way around.)

My additional take on things is that things eventually get
balanced. Thus, with a 3 years from date of manufacture
warranty the life of the lubricant and the motors will both tend
to 3 years from date of manufacture rather than the old 5 years
spinning life.

An FDB motor is frictionless, once it starts to work. The
constantly pumped fluid, which travels over the bearing
surface, is what helps to make it frictionless. But during
startup (just like your car), there could be a little bit
of friction.

There is an attestation, to how good a fluid bearing can be, here.

http://en.wikipedia.org/wiki/Fluid_bearing

"According to the ASME (see reference link), the first
Michell/Kingsbury fluid bearing in the USA was installed in
the Holtwood Hydroelectric Power Plant (on the Susquehanna River,
near Lancaster, Pennsylvania, USA) in 1912. The 2.25-tonne bearing
supports a water turbine and electric generator with a rotating mass
of about 165 tonnes and water turbine pressure adding another 40 tonnes.
The bearing has been in nearly continuous service since 1912, with
no parts replaced. The ASME reported it was still in service as of
2000. As of 2002, the manufacturer estimated the bearings at Holtwood
should have a maintenance-free life of about 1,300 years."

You can see a sample of a test report for FDB motors here. See figure 3.

http://www.seagate.com/docs/pdf/whit...otor_tp574.pdf

They measure how well things are going, by doing a "weight change" test.
If the lubricant is leaving the sealed FDB motor, that is an indication
that eventually, it will fail due to no lubricant being present.
Presumably, operating temperature is an element to lubricant
loss.

The lubricant is inside the motor itself. It isn't floating around
within the HDA. The quantity of lubricant is pretty small.

What the test results don't show, is what percentage of FDB motors
leaving the factory, are defect free. Obviously, motors and
bearing are still failing, but that may not be due solely to
the nature of the design. When a disk drive sells for $40 retail,
how carefully are all the parts going to be made ?

I would hope, that a real reliability calculation, is based on field return
data, and takes into account, what percentage of "failure to function"
is caused by the motor, head assembly, landing ramp, controller board,
and so on. One of those Mickey Mouse theoretical calculations would
not make sense, when the manufacturer is determining how much money
to set aside for warranty claims. By using real field data (collected
as drives come back under warranty), the manufacturer gets a real
good idea, what part of their design sucks. They're not going to
put that number in print. Instead, the spec section of the web
site, will contain the Mickey Mouse calculation (as someone who
has played that game, both supplier and customer use the Mickey
Mouse method, each with a smirk on their face - the truth may
be arrived at by negotiation between the parties, in cases
involving large contracts).

In the study done here, it was noted that disk drives do not follow
the normal "bathtub curve" that people assume. In fact, replacement
rate steadily increases with time. So the statistic is not what
you'd expect.

http://www.siliconsystems.com/techno...re_Studies.pdf

One of my expectations would be, that the manufacturer relies on
some percentage of customers *not* exercising the warranty, in the
same way as a rebate coupon program relies on only a small percentage
of customers bothering to correctly fill out the paperwork.

Paul

Mark F wrote:
Re length of warranty versus expected failure rate and disk
life:

I have no idea about your post but FWIW you can often double the warranty
period simply by paying with a credit card.

Many (most?) credit card issuers toss that in as an added feature and it can
be handy...I've collected at least twice on hard drives that died shortly
after the stated warranty period expired.

dadiOH

On Mon, 25 May 2009 17:14:43 -0400, Paul wrote:

Mark F wrote:
Re length of warranty versus expected failure rate and disk
life:

One or more of Scott Moulton mentions that some lubricant
evaporates and is supposed to be recycled inside of the drive.

He suspects that not all of the lubricant will be recycled
and that therefore the lifetime of the drive is limited.

Not specifically stated, but implicit in the comments, are
that there are other factors that may lead to end of life
of the disk, such as bearing life for the motor and
other moving parts. (What we [not working for the disk manufacturers]
don't know is if the life of the motors is 5 years and the
life of the lubricant is 50 years or the other way around.)

My additional take on things is that things eventually get
balanced. Thus, with a 3 years from date of manufacture
warranty the life of the lubricant and the motors will both tend
to 3 years from date of manufacture rather than the old 5 years
spinning life.

An FDB motor is frictionless, once it starts to work. The
constantly pumped fluid, which travels over the bearing
surface, is what helps to make it frictionless. But during
startup (just like your car), there could be a little bit
of friction.
The lubricant that I was referring to was for the surface of the
disks, not for the motors. I just mentioned the motors as one
of the other items in disk drives that were subject to wearing
and that I expected would over time be changed to have a lifetime
reduced to that of the least long-lived component.
(rest of post valid, but deleted)

Mark F wrote:
On Mon, 25 May 2009 17:14:43 -0400, Paul wrote:

Mark F wrote:
Re length of warranty versus expected failure rate and disk
life:

One or more of Scott Moulton mentions that some lubricant
evaporates and is supposed to be recycled inside of the drive.

He suspects that not all of the lubricant will be recycled
and that therefore the lifetime of the drive is limited.

Not specifically stated, but implicit in the comments, are
that there are other factors that may lead to end of life
of the disk, such as bearing life for the motor and
other moving parts. (What we [not working for the disk manufacturers]
don't know is if the life of the motors is 5 years and the
life of the lubricant is 50 years or the other way around.)

My additional take on things is that things eventually get
balanced. Thus, with a 3 years from date of manufacture
warranty the life of the lubricant and the motors will both tend
to 3 years from date of manufacture rather than the old 5 years
spinning life.
An FDB motor is frictionless, once it starts to work. The
constantly pumped fluid, which travels over the bearing
surface, is what helps to make it frictionless. But during
startup (just like your car), there could be a little bit
of friction.
The lubricant that I was referring to was for the surface of the
disks, not for the motors. I just mentioned the motors as one
of the other items in disk drives that were subject to wearing
and that I expected would over time be changed to have a lifetime
reduced to that of the least long-lived component.
(rest of post valid, but deleted)

Info on platter lubrication, here.

Picture of platter stackup.
http://www.hitachigst.com/hdd/resear.../im/index.html

Z-DPA mentioned here.
http://www.hitachigst.com/hdd/resear...gradation.html

More details on Z-DPA. Lubricant is very thin (order of one nanometer),
and consists of a bonded part and a mobile part.

http://www.freepatentsonline.com/y2005/0282045.html

Paul

Mark, It seems that you're in the market for HD's. If this is the case,
you need look no farther that the reliability records of the various
brands. Quantum, the best of the bunch for both SCSI and IDE has, sadly,
been absorbed by a less quality-minded concern. I've had two Q drives
for 12-13 years and they run flawlessly. Luck. Please report your
findings here.


Mark F wrote:
Re length of warranty versus expected failure rate and disk
life:

One or more of Scott Moulton mentions that some lubricant
evaporates and is supposed to be recycled inside of the drive.

He suspects that not all of the lubricant will be recycled
and that therefore the lifetime of the drive is limited.

Not specifically stated, but implicit in the comments, are
that there are other factors that may lead to end of life
of the disk, such as bearing life for the motor and
other moving parts. (What we [not working for the disk manufacturers]
don't know is if the life of the motors is 5 years and the
life of the lubricant is 50 years or the other way around.)

My additional take on things is that things eventually get
balanced. Thus, with a 3 years from date of manufacture
warranty the life of the lubricant and the motors will both tend
to 3 years from date of manufacture rather than the old 5 years
spinning life.