Sign in with
Sign up | Sign in
Your question

Prolonging Life of Hard Drive

Last response: in Laptops & Notebooks
Share
November 1, 2004 12:45:33 AM

Archived from groups: comp.sys.laptops (More info?)

The IBM hard drive in my Sony VAIO VX71P has recently broken after slightly
less than 3 years. I have replaced it with a Samsung MP0402H, which is far
quieter and runs far cooler.

I would like to use my laptop in such a way that will prolong the life of
this new drive. I use my laptop as my main computer and so it is generally
powered on all day every day. In the power settings, should I set the drive
to turn off after a period of time, or is it better for the drive to stay
constantly spinning whilst the laptop is on?

Many thanks in advance.
Anonymous
a b D Laptop
November 1, 2004 1:36:18 AM

Archived from groups: comp.sys.laptops (More info?)

> powered on all day every day. In the power settings, should I set the drive
> to turn off after a period of time, or is it better for the drive to stay
> constantly spinning whilst the laptop is on?

Anyone who gives you a simple answer to this is, to put it bluntly,
foolish. Until a few months ago, I worked as lead engineer at a company
that made appliances containing 2.5" IDE hard drives. We were
experiencing a large number of field failures, so I went through all the
information I could find trying to work out "the" answer (mainly so that
we could work out the right warranty period to offer). From researching
all the datasheets, it's clear that there is no simple heuristic you can
use to answer this question, because there are several different
"lifetimes" built into the drive, viz (typically):

1. An overall powerup lifetime, assuming you power it up once and run it
until it dies. This reflects the lifespan of various electronic
components - mostly electrolytic caps, probably - at anticipated drive
operating temperatures. You can't affect this with Windows power
management settings.

2. A spindle powerup lifetime, assuming you spin it up once and run it
until it seizes. This reflects the lifespan of the motor, bearings and
the drive transistors powering the spindle motor. This is the lifespan
you eat LESS rapidly when you set aggressive power management options in
Windows.

3. A spinup/spindown cycle lifetime. This indicates how many times the
drive is expected to survive going into "low power" mode (turning off
the spindle motor while leaving the rest of the electronics powered).
This is the lifespan you eat MORE rapidly when you set aggressive power
management options in Windows.

4. A head load/unload lifetime. This reflects how many times you can
expect the unload mechanism to work properly, and the probability of a
head crash during load.

5. A power-cycle lifetime. This indicates how many times the drive is
expected to survive the stress of a cold powerup operation. You eat this
number every time you switch the laptop off and on.


The problem is, it is EXTREMELY difficult to look at all these numbers
(and there are usually others as well) and decide what exactly are the
best settings to extend the overall drive life - partly because the
"right" choice depends on how you intend to use your laptop *today*. If
you're going to spend 3 hours watching a DVD, then you'll be best off to
set a fairly aggressive spindown timeout, so that you avoid eating into
lifespan #2 above. Tomorrow, you might be working on a word processing
document, and saving every five minutes. In that case, you'd be better
off leaving the drive spinning, because 8 hours of one spinup/down per 5
minutes will eat a huge chunk of lifespan #3.

Moreover, the details are specific to the manufacturer and series (if
not individual model) of drive.

The manufacturer will combine all of these lifetimes to obtain the MTBF
(this is the number you usually see quoted as the "lifetime" of the
drive). The datasheet usually says what the assumptions were built into
that number. Typically, it is an artificial test usage pattern involving
a couple of spinup cycles every 24 hours, a certain number of sector
read/write ops per hour, and a certain number of powered-down hours.

The bottom line is that consumer-grade 2.5" IDE drives are not designed
for 100% duty cycle operation, and, like all hard drives, you shouldn't
rely on them for long-term storage.

Avoid shocks. Avoid using the drive in conditions of reduced air
pressure or increased humidity. Strenuously avoid tobacco smoke. Use the
drive in a manner that gives you acceptable performance and battery
life, in accordance with your typical usage patterns. Install additional
RAM to reduce the frequency with which the OS pages out data to disk.
Back up all important files regularly.


PS: Three years is good going. Although drives DO frequently last
longer, I would say 15-24 months is a good lifespan for a 2.5" drive
that sees heavy usage.
Anonymous
a b D Laptop
November 1, 2004 1:36:19 AM

Archived from groups: comp.sys.laptops (More info?)

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi Lewin,

What does tobacco smoke have to do with the life of a hard disk. If
cigarette smoke is getting inside of your hard disk, then you have a
damaged hard disk.

30 years ago, when I worked on mainframes the prohibition on smoking
made sense because the smoke could effect the tape. These days hard
disk are *sealed*. Any kind of contaminate could damage a hard disk.
That is why they are sealed so that no contaminates can get to the disk.

Ciao . . . C.Joseph

That which a man buys too cheaply . . .
~ He esteems too lightly



Lewin A.R.W. Edwards wrote:

| Avoid shocks. Avoid using the drive in conditions of reduced air
| pressure or increased humidity. Strenuously avoid tobacco smoke.
| Use the drive in a manner that gives you acceptable performance and
| battery life, in accordance with your typical usage patterns.
| Install additional RAM to reduce the frequency with which the OS
| pages out data to disk. Back up all important files regularly.
|
|
| PS: Three years is good going. Although drives DO frequently last
| longer, I would say 15-24 months is a good lifespan for a 2.5"
| drive that sees heavy usage.
|
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.2 (MingW32)

iD8DBQFBhXZC6bFq6mlbLOwRAhn3AJ9jA2o3oAbNE9joU04mJPdRJ1UUiQCfaRaC
iycrJeybakB6T8aatIgTIPY=
=ozM+
-----END PGP SIGNATURE-----
Related resources
Anonymous
a b D Laptop
November 1, 2004 1:36:20 AM

Archived from groups: comp.sys.laptops (More info?)

"C.Joseph Drayton" <kalek1@mindspring.com> writes:
> 30 years ago, when I worked on mainframes the prohibition on smoking
> made sense because the smoke could effect the tape. These days hard
> disk are *sealed*. Any kind of contaminate could damage a hard disk.
> That is why they are sealed so that no contaminates can get to the disk.

Hard discs are not sealed. The air going in and out of them is filtered,
but smoke particles are very small and can get in through the filter pores.
Anonymous
a b D Laptop
November 1, 2004 3:40:10 AM

Archived from groups: comp.sys.laptops (More info?)

C.Joseph Drayton <kalek1@mindspring.com> wrote:
> What does tobacco smoke have to do with the life of a hard disk. If

It's full of small particles. About the size of the heads or domains ...

> cigarette smoke is getting inside of your hard disk, then you have a
> damaged hard disk.

No he doesn't. They're open to the air.

Peter
Anonymous
a b D Laptop
November 1, 2004 3:40:11 AM

Archived from groups: comp.sys.laptops (More info?)

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi Peter,

I just looked at one of my backup 2.5" hard disk, and if there is a
way for contaminates to get into it, I can't see it.

Ciao . . . C.Joseph

That which a man buys too cheaply . . .
~ He esteems too lightly



Peter T. Breuer wrote:

|C.Joseph Drayton <kalek1@mindspring.com> wrote:
|
|>What does tobacco smoke have to do with the life of a hard disk. If
|
|
|It's full of small particles. About the size of the heads or domains ...
|
|>cigarette smoke is getting inside of your hard disk, then you have a
|>damaged hard disk.
|
|
|No he doesn't. They're open to the air.
|
|Peter
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.2 (MingW32)

iD8DBQFBhXxS6bFq6mlbLOwRAqvFAKCvuF876tMR1zrATUgPdwUQmZ8JfgCg0fLy
oCX/u2nnQ6iXTLLpedet0+c=
=6e/i
-----END PGP SIGNATURE-----
Anonymous
a b D Laptop
November 1, 2004 3:40:12 AM

Archived from groups: comp.sys.laptops (More info?)

"C.Joseph Drayton" <kalek1@mindspring.com> writes:
> I just looked at one of my backup 2.5" hard disk, and if there is a
> way for contaminates to get into it, I can't see it.

There should be a little hole on the top plate, maybe adjacent to a
sticker warning you not to cover the hole. That is the air intake.
There is a little filter behind it, but air does have to circulate in
the drive for the drive to work.
Anonymous
a b D Laptop
November 1, 2004 4:16:43 AM

Archived from groups: comp.sys.laptops (More info?)

Lewin A.R.W. Edwards wrote:

>
> Anyone who gives you a simple answer to this is, to put it bluntly,
> foolish.

Certainly and all the explanations that follow are very interesting.
Still the OP question remains unanswered. I would just suggest a common
sense approach. Obviously, powering up and down a hard disk often is
increasing the chances of trouble. My not very scientific approach is
this: if for extended period of times, I am away from my machine, there
is no point in keeping the disk spinning. On the other hand, if I am
working on my machine, I do not want the hard drive to spin down and
force me to wait until it "wakes" up to be productive again. So the
short answer is, set up a value in Windows of for instance 5mn before
the hard disk spins down. Then, observe what happens during a typical
work session. If you notice that the disk needs to wake up more than
occasionally, this value is too short. Ideally, the hard disk should
only spin down in instances when it will remain stopped for an extended
(whatever that means !) period of time. So this is basically a trial and
error process.

But statistically, chances are you will replace your machine before the
HD dies on you, whatever you do. Just use common sense and settle for an
intermediate value.
--
John Doue
Anonymous
a b D Laptop
November 1, 2004 4:16:44 AM

Archived from groups: comp.sys.laptops (More info?)

In article <%7ghd.413$nC.118@read3.inet.fi>,
John Doue <notwobe@yahoo.com> wrote:
>Lewin A.R.W. Edwards wrote:
>
>>
>> Anyone who gives you a simple answer to this is, to put it bluntly,
>> foolish.
>
>Certainly and all the explanations that follow are very interesting.
>Still the OP question remains unanswered. I would just suggest a common
>sense approach. Obviously, powering up and down a hard disk often is
>increasing the chances of trouble. My not very scientific approach is
>this: if for extended period of times, I am away from my machine, there
>is no point in keeping the disk spinning. On the other hand, if I am
>working on my machine, I do not want the hard drive to spin down and
>force me to wait until it "wakes" up to be productive again. So the
>short answer is, set up a value in Windows of for instance 5mn before
>the hard disk spins down. Then, observe what happens during a typical
>work session. If you notice that the disk needs to wake up more than
>occasionally, this value is too short. Ideally, the hard disk should
>only spin down in instances when it will remain stopped for an extended
>(whatever that means !) period of time. So this is basically a trial and
>error process.
>
>But statistically, chances are you will replace your machine before the
>HD dies on you, whatever you do. Just use common sense and settle for an
>intermediate value.
>--
>John Doue


The #1 thing that shortens disk life is temperature. With a
desktop PC you can make sure that airflow keeps it cool. With laptops
you don't really have any way to modify the machine.

If you know the disk drive model, look up the detailed specs for it on
the manufacturers web site and see the recommended max operating
temperature, then install some SMART monitering software on your
laptop and make sure the temperature is below spec. If it's at or
above spec the lifetime will be shortened. If the laptop design mimits
airflow and the disk runs hot I can't think of anything to tell you.

There is really no guarantee. You can do everything right and
the disk may fail next week.

Backup Backup Backup.




--
a d y k e s @ p a n i x . c o m
----
Anonymous
a b D Laptop
November 1, 2004 5:03:32 AM

Archived from groups: comp.sys.laptops (More info?)

> What does tobacco smoke have to do with the life of a hard disk. If

General advice, mainly, to keep your computer out of tobacco smoke. A
machine that has been in a smoker's environment is filled with sticky
dust. It fills heatsinks, causes premature failure of fans, and blankets
all the electronics, keeping them warmer than they ought to be.

Reputedly it can help arcs cross boundaries they otherwise would not,
though I have never seen this. It also grabs any passing atmospheric
contaminants and holds them long-term against contact surfaces.

> made sense because the smoke could effect the tape. These days hard
> disk are *sealed*.

I suggest you take a close look at a hard disk. All hard drives have a
pressure equalization vent, which is led through a simple fine-grain
filter. There is a second filter in the airflow inside the drive to
catch internally generated debris.
November 1, 2004 5:16:14 AM

Archived from groups: comp.sys.laptops (More info?)

Thank you for the replies everyone.

Not only is it very useful, but also a very interesting discussion.

Thanks again.
Anonymous
a b D Laptop
November 1, 2004 9:37:17 AM

Archived from groups: comp.sys.laptops (More info?)

In article <7x3bzutpwz.fsf@ruckus.brouhaha.com>,
Paul Rubin <http://phr.cx@NOSPAM.invalid&gt; wrote:
>"C.Joseph Drayton" <kalek1@mindspring.com> writes:
>> 30 years ago, when I worked on mainframes the prohibition on smoking
>> made sense because the smoke could effect the tape. These days hard
>> disk are *sealed*. Any kind of contaminate could damage a hard disk.
>> That is why they are sealed so that no contaminates can get to the disk.
>
>Hard discs are not sealed. The air going in and out of them is filtered,
>but smoke particles are very small and can get in through the filter pores.


The hole, and filter, is necsesary to keep the pressure inside the
case the same as the atmosphere. A disk drive can get hot, and while
I've never heard a design engineer say it, I guess internal pressure
could distort the dimensions of the case and the hole is a simple way
to eliminate this variable.

The filter doesn't let anything thru that is larger that would be
found in the cleanrooms the disk was assembled in, in the first place.
The amount of air that flows thru the filter is microscopic.

--
a d y k e s @ p a n i x . c o m
----
Anonymous
a b D Laptop
November 1, 2004 4:59:09 PM

Archived from groups: comp.sys.laptops (More info?)

Al Dykes <adykes@panix.com> wrote:
> In article <7x3bzutpwz.fsf@ruckus.brouhaha.com>,
> Paul Rubin <http://phr.cx@NOSPAM.invalid&gt; wrote:
> >"C.Joseph Drayton" <kalek1@mindspring.com> writes:
> >> 30 years ago, when I worked on mainframes the prohibition on smoking
> >> made sense because the smoke could effect the tape. These days hard
> >> disk are *sealed*. Any kind of contaminate could damage a hard disk.
> >> That is why they are sealed so that no contaminates can get to the disk.
> >
> >Hard discs are not sealed. The air going in and out of them is filtered,
> >but smoke particles are very small and can get in through the filter pores.
>
>
> The filter doesn't let anything thru that is larger that would be
> found in the cleanrooms the disk was assembled in, in the first place.
> The amount of air that flows thru the filter is microscopic.

No it's not. If the temperature difference from ambient is 30C inside
the case, it's 10% of the air inside the case, because volume is
proportional to absolute temperature (and we are at 300K) in a gas at
constant pressure.

Run the cycle up and down a few times, and you easily exchange all the
air inside the casing.

As to what gets through the filter - well, either something gets
through or the filter clogs up. I know no more than that deduction
tells me.

Peter
Anonymous
a b D Laptop
November 2, 2004 7:00:34 AM

Archived from groups: comp.sys.laptops (More info?)

"C.Joseph Drayton" <kalek1@mindspring.com> wrote in message
news:2ulb23F2c05duU1@uni-berlin.de...
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
> Hi Lewin,
>
> What does tobacco smoke have to do with the life of a hard disk. If
> cigarette smoke is getting inside of your hard disk, then you have a
> damaged hard disk.
>
> 30 years ago, when I worked on mainframes the prohibition on smoking
> made sense because the smoke could effect the tape. These days hard
> disk are *sealed*. Any kind of contaminate could damage a hard disk.
> That is why they are sealed so that no contaminates can get to the disk.
>


You're wrong.

The drives are mechanically sealed,
but they're not airtight. They do
in fact have a ventilation hole.

Next time you have a chance to
hold a disk drive in your hands,
take a good, close look.



dk
!