When Do You Need A HDD Cooler?

On Tue, 27 Sep 2005 22:49:30 GMT, kony wrote:

IMO, one should not expect good service from a drive if it's
operating in the upper end of it's allowed temp range.

Why do you believe that temp is important for hard disks but not for
CPUs? Same silicon, isn't it?


--

Greatest Movie Line Ever
http://home.houston.rr.com/rkba/MovieLine.wmv

"What is history but the story of how politicians have
squandered the blood and treasure of the human race?"
--Thomas Sowell

Bob wrote:

Why do you believe that temp is important for hard disks but not for
CPUs? Same silicon, isn't it?

What's the average lifepan of a CPU run at its rated maximum allowed
temperature verses an HD run at its rated maximum?

On 28/9/05 11:17 am, in article
, "Bob" wrote:

On Tue, 27 Sep 2005 22:49:30 GMT, kony wrote:

IMO, one should not expect good service from a drive if it's
operating in the upper end of it's allowed temp range.

Why do you believe that temp is important for hard disks but not for
CPUs? Same silicon, isn't it?

We have had many systems break because of heat from a hard drive, especially
when you are using 2 close to each other. I guess too much heat will warp
the platter or something. We have found the best way to prevent breaking of
hard drives in the manner is to install a 80mm fan at the foot of the drive
and have it suck air from outside the box to inside and use the exhaust
blowers at the back to expel the hot air.
--
Nik
http://www.pc-gear.co.uk

On Wed, 28 Sep 2005 10:17:50 GMT, (Bob) wrote:

On Tue, 27 Sep 2005 22:49:30 GMT, kony wrote:

IMO, one should not expect good service from a drive if it's
operating in the upper end of it's allowed temp range.

Why do you believe that temp is important for hard disks but not for
CPUs? Same silicon, isn't it?


That's just it, it's not "same silicon".
Drives are comprised of many other materials, mechanical
stress, wear, and dependencies in addition to power
regulation. Intel will be incorporating power regulation on
their CPUs in the future, and we may then see CPUs more
susceptible to heat too, but even so it's likely higher than
HDD limit mentioned previously of 55C. 55C is not a limit
seen on modern silicon, clearly the HDD limit is other
componentry. For the time being, the core logic chips on
CPUs are not generally seen to be the failure points, save
for some random manufacturing error like a bad batch of
epoxy in their casings.

this is in addition to the issue LM&C mentioned, that the
average lifespan of a CPU is far beyond that reasonable for
a system in general, that the CPU is among the parts
expected to be longest lasting while the drive is far from
it. Further, "many" people would suffer far less loss from
a CPU failure than a drive failure (even discounting a
presumption that people aren't making regular backups like
they should).

On 28/9/05 3:03 pm, in article ,
"kony" wrote:

On Wed, 28 Sep 2005 10:17:50 GMT, (Bob) wrote:

On Tue, 27 Sep 2005 22:49:30 GMT, kony wrote:

IMO, one should not expect good service from a drive if it's
operating in the upper end of it's allowed temp range.

Why do you believe that temp is important for hard disks but not for
CPUs? Same silicon, isn't it?


That's just it, it's not "same silicon".
Drives are comprised of many other materials, mechanical
stress, wear, and dependencies in addition to power
regulation. Intel will be incorporating power regulation on
their CPUs in the future, and we may then see CPUs more
susceptible to heat too, but even so it's likely higher than
HDD limit mentioned previously of 55C. 55C is not a limit
seen on modern silicon, clearly the HDD limit is other
componentry. For the time being, the core logic chips on
CPUs are not generally seen to be the failure points, save
for some random manufacturing error like a bad batch of
epoxy in their casings.

this is in addition to the issue LM&C mentioned, that the
average lifespan of a CPU is far beyond that reasonable for
a system in general, that the CPU is among the parts
expected to be longest lasting while the drive is far from
it. Further, "many" people would suffer far less loss from
a CPU failure than a drive failure (even discounting a
presumption that people aren't making regular backups like
they should).
I don't think it's fair to sell a product with a 5 year life that has a 1
year warranty and fails in 18 month.
--
Nik
http://www.pc-gear.co.uk

On Wed, 28 Sep 2005 15:31:16 +0100, "Nik Simms (Web Developer)"
wrote:

I don't think it's fair to sell a product with a 5 year life that has a 1
year warranty and fails in 18 month.

You haven't bought a new American car lately, have you.

It used to be that if you could get a new Chrysler out of the lot and
right back in without a tow truck, you were lucky.

I've seen cars with the window sticker still on being towed.



--

Greatest Movie Line Ever
http://home.houston.rr.com/rkba/MovieLine.wmv

"What is history but the story of how politicians have
squandered the blood and treasure of the human race?"
--Thomas Sowell

You cannot determine drive temp with only the software temp
reading, because it is only the temp of the chip taking it,
a chip isolated for each and every other chip. The PCB
insulates more than conducting heat (including traces) so
the temp report can only be used in conjunction with other
means.

I'll measure the drive casing with the temperature probe on my dmm then and see
if it reads differently to software..

If they're too hot to keep your hand on comfortably, yes I'd
try to increase cooling some. Those drives aren't even
relatively very hot running so there may be some airflow
issues in the whole chassis causing this?

I thot that myself... but moving from a fairly roomy case with no cooling, to a
brand new case with 120mm fans front and back, hasn't made much difference..
(maybe between 2 and 4C max). The front fan is directly infront of the drive
cage.

Regards,
Chris



----== Posted via Newsfeeds.Com - Unlimited-Uncensored-Secure Usenet News==----
http://www.newsfeeds.com The #1 Newsgroup Service in the World! 120,000+ Newsgroups
----= East and West-Coast Server Farms - Total Privacy via Encryption =----

On Thu, 29 Sep 2005 22:31:35 -0400, "Skeleton Man"
wrote:

You cannot determine drive temp with only the software temp
reading, because it is only the temp of the chip taking it,
a chip isolated for each and every other chip. The PCB
insulates more than conducting heat (including traces) so
the temp report can only be used in conjunction with other
means.

I'll measure the drive casing with the temperature probe on my dmm then and see
if it reads differently to software..

That will provide a useful, 2nd temp, then all you'd need
for a more comprehensive assessment is the temp of all the
other chips, too. The reported HDD temp is best seen as a
gauge of relative change.



If they're too hot to keep your hand on comfortably, yes I'd
try to increase cooling some. Those drives aren't even
relatively very hot running so there may be some airflow
issues in the whole chassis causing this?

I thot that myself... but moving from a fairly roomy case with no cooling, to a
brand new case with 120mm fans front and back, hasn't made much difference..
(maybe between 2 and 4C max). The front fan is directly infront of the drive
cage.


Does this new case have mostly unobstructed intake and
exhaust area, including low restriction from the case front
bezel? Fans are only so good as the case allows them to be.
A rack behind a 120mm fan, drives having fair amount of
space between them, shouldn't feel hot. However one
problem with the bays that stand back from fan some, AND
using 120mm fan is that far lower percentage of flow goes
between the drives, most flows around them instead. IMO,
the optimal arrangement for rack cooling is a 92mm fan with
an extended rack, such that the vast majority of fan exhaust
must travel between the drives to get out through the rack
into the rest of the chassis (save for a trivial amount
escaping through the rack mounting holes).

Ironically and unfortunately, few systems use such a setup
so you'd have to DIY.