Temps that shorten drive life

[Rod Speed]

Arno Wagner wrote

There's a chemist's rule of thumb

Irrelevant to hard drives.

- if you raise the temperature of a reacting system by 10 degrees
(obviously not Farenheit, or Badley, or other such archaic tosh),
then you double the speed of a reaction.

Complete drivel, even with chemical reactions.

Not strictly relevant to mechanical failures of bearings,

It aint the bearings which fail, stupid.

but of considerable relevance to things like the oxidation of lubricating oils,

It aint the bearings which fail, stupid.

the re-crystallisation of solders

Thanks for that completely superfluous proof that you dont
actually have a ****ing clue about anything at all, ever.

and the diffusion of dopant atoms.

Pathetic, really.

 

[Aidan Karley]

Well, that was an easy decision.

 

[JohnH]

Well, that was an easy decision.

Fat lot of good it will do you, ****wit child.

 

[Folkert Rienstra]

I agree on this.

What a surprise.

And this should be at reasonable load.

For some drives it may need to be even lower. For example
I am the person with the about 40 Maxtors running reliably
24/7 for 3 years now. But they are in a server room and with
outside air on each, i.e. they don't go over 30C under heavy load.

WD is lying to you.

Of course they are, babblebot.

Some background: Semiconductors have
about 20-30 years lifetime at 25C.

This degrades by a factor of about 2 every 10C or so.

15 years at 35C
7 years at 45C
3 years at 55C

Still within the warranty period even with 24/7 use

 

[Folkert Rienstra]

Been debating which one to shitcan myself.

Which is worse,

hooker or hookee?

Or who's who between them?

 

[Guest]

Well, depends. Actually it is difficult to get current information.
The figures are from Intel's memory components Handbook from 10-15
years back.

That would mean semiconductors are expected to have shorter average
lifespans than electrolytic capacitors, which are usually considered
the least reliable electronic components in any electronic device.

 

[Arno Wagner]

That would mean semiconductors are expected to have shorter average
lifespans than electrolytic capacitors, which are usually considered
the least reliable electronic components in any electronic device.

As I said, the figures are from a memory handbook from way back.
Of course different semiconductors have different characteristics.
But don't expect 30 years at 200C. You will not get it. Today you
often get something like 5 years at 80C or so, if you get any
information at all. The 30 year point would then be around 50C.

Electrolythes age differently than semiconductors. Their "lifetime"
is the time when capacity has dropped to 50%. For semiconducors
it is when the failure rate rises again. Also electrolytes
age all the time, just a bit faster when used. Semiconductors
age almost not at all when not used.

In addition, if you buy quality capacitors, they are not the least
reliable component. It is just that you can get very cheap,
short-lived ones, rated at, e.g. 1000h at 85C.

In short, no, it is not as simple as that.

Arno

 

[Guest]

WD is lying to you. Some background: Semiconductors have
about 20-30 years lifetime at 25C.


Well, depends. Actually it is difficult to get current information.
The figures are from Intel's memory components Handbook from 10-15
years back.


As I said, the figures are from a memory handbook from way back.
Of course different semiconductors have different characteristics.
But don't expect 30 years at 200C. You will not get it.

I said 10C below maximum rating. Lifespan is expected to be several
times longer at 30C below maximum.

Today you often get something like 5 years at 80C or so, if you get any
information at all. The 30 year point would then be around 50C.

50C is cooler than even maximum allowed even for ultra reliable
equipment.

Electrolythes age differently than semiconductors. Their "lifetime"
is the time when capacity has dropped to 50%. For semiconducors
it is when the failure rate rises again. Also electrolytes
age all the time, just a bit faster when used. Semiconductors
age almost not at all when not used.

That's irrelevant to your claim of 20-30 years expected lifespan for
semiconductors at 25C.

 

[Arno Wagner]

I said 10C below maximum rating. Lifespan is expected to be several
times longer at 30C below maximum.

Does not matter. This is not a hard phase change. 10C below maximum
you get neraly the same behaviour as at maximum.

50C is cooler than even maximum allowed even for ultra reliable
equipment.

What pleae is "ultra reliable equipment"?

That's irrelevant to your claim of 20-30 years expected lifespan for
semiconductors at 25C.

Juts saying you capacitor comparison is bogus.

And would you please post a source for your 30 years @1 90C claim?

Arno

 

[Folkert Rienstra]

That's a rather broad range, isn't it? Still, if you take the lower,
what would the package outside temperature be?

So much for your "WD is lying to you", babblebot.

I said 10C below maximum rating. Lifespan is expected to be several
times longer

at 30C below maximum.

Which would be what, in practice?

The 30 year point would then be around 50C.

Too bad of your "55C is way to hot", babblebot.

 

[willbill]

Aidan Karley wrote:

Well, that was an easy decision.

i see that i now need to do
a global kill on johnh (in my
Agent n/g program)

re Curious George, despite being
the hookee he's said some sensible
things and seems to have a brain.
i expect he'll adjust ok to
rod speed's frequent
nonsense posting

bill

 

[willbill]

Ed Light wrote:

55C is *way* too hot for a hard drive.

You should be aiming for max 40C - no more.

Bear in mind that these temp readings tend to relate to the actual
casing - at which stage the components on the logic board (which are far
more susceptible to heat damage) are probably twice the temperature.
(I.e. dangerously warm.)

i tend to agree with you on a max of 40C
for hard drives

OTOH, 55C is "only" 131F, a nasty summer
day in the shade in Death Valley

my warmer PC reports a typical case temp
of 39C at idle, and up to 43C under load.
since my fans for my hard drive stack (4 of
them, 1 boot (Raptor) and 3 for a raid 1E
array) blow air into the case, directly
onto the hard drives, i'm confident that
they are running at 40C (or less), and they
also pass the finger test where i take the
side door off the case while it's running
and put my finger on top of each drive

as a side note on processor temps, my
slower/cooler PC has an nVidia GeForce
6600GT which shows a normal GPU temp of
54/55C (i haven't yet played games on this
machine), with a stated slowdown point
of 145C!!

my CPU (Opty 940) shuts down if it gets
to something like 75 or 80C

bill

 

[Guest]

Previously (e-mail address removed) wrote:

Semiconductors have about 20-30 years lifetime at 25C.

Well, depends. Actually it is difficult to get current information.
The figures are from Intel's memory components Handbook from 10-15
years back.


As I said, the figures are from a memory handbook from way back.
Of course different semiconductors have different characteristics.
But don't expect 30 years at 200C. You will not get it.


Does not matter. This is not a hard phase change. 10C below maximum
you get nearly the same behaviour as at maximum.

My information came from an engineer who worked at a Motorola bipolar
fab.

Today you often get something like 5 years at 80C or so, if you get any
information at all. The 30 year point would then be around 50C.


What please is "ultra reliable equipment"?

Space-rated equipment will fit the bill (at least it's supposed to),
although its components usually have higher maximum temperature
ratings.

Electrolythes age differently than semiconductors. Their "lifetime"
is the time when capacity has dropped to 50%. For semiconducors
it is when the failure rate rises again. Also electrolytes
age all the time, just a bit faster when used. Semiconductors
age almost not at all when not used.


Just saying your capacitor comparison is bogus.

Nonsense, and a large sample of any electronic equipment will almost
always show more failures among the electrolytic capacitors than the
semiconductors. For equipment at least 10 years old, it's common for
most of its electrolytics to have slightly high ESR and none of the P-N
junctions of the power semiconductors to have excessive leakage.

And would you please post a source for your 30 years @1 90C claim?

I don't have any, and I haven't knowingly bought any semiconductors
rated for 200C max. since the 1970s. Maybe www.powerelectronics.com
has an article.

 

[Arno Wagner]

My information came from an engineer who worked at a Motorola bipolar
fab.

Ok, power semiconductors are different. My numbers are for CMOS.
Power semiconductors have larger structures and are therefore far
less succeptible to aging. They usually die by melting. For small
scale cmos, this is different.

Space-rated equipment will fit the bill (at least it's supposed to),
although its components usually have higher maximum temperature
ratings.

An lower power allowances. The trade-off done here for non-cooled
chips is that e.g. military TTL is slower, and can tolerate
higher ambient temperatures.

Nonsense, and a large sample of any electronic equipment will almost
always show more failures among the electrolytic capacitors than the
semiconductors. For equipment at least 10 years old, it's common for
most of its electrolytics to have slightly high ESR and none of the P-N
junctions of the power semiconductors to have excessive leakage.

I don't dospute that. The problem is low-quality capacitors combined
with a different ageing mode.

Arno

 

[Arno Wagner]

Odie Ferrous wrote:



i tend to agree with you on a max of 40C
for hard drives

OTOH, 55C is "only" 131F, a nasty summer
day in the shade in Death Valley

my warmer PC reports a typical case temp
of 39C at idle, and up to 43C under load.
since my fans for my hard drive stack (4 of
them, 1 boot (Raptor) and 3 for a raid 1E
array) blow air into the case, directly
onto the hard drives, i'm confident that
they are running at 40C (or less), and they
also pass the finger test where i take the
side door off the case while it's running
and put my finger on top of each drive

Should be fine then.

as a side note on processor temps, my
slower/cooler PC has an nVidia GeForce
6600GT which shows a normal GPU temp of
54/55C (i haven't yet played games on this
machine), with a stated slowdown point
of 145C!!

Nvidia is incompetent or stupid here. My 7600GT has
a slowdown-point of 105C but it becomes flaky and crashes
around 70...80C. I think the slowdown point is way to
high on these things.

my CPU (Opty 940) shuts down if it gets
to something like 75 or 80C

That is the right point.

Arno

 

[willbill]

Nvidia is incompetent or stupid here. My 7600GT has
a slowdown-point of 105C but it becomes flaky and crashes
around 70...80C. I think the slowdown point is way to
high on these things.

i have a 7600GT on my faster machine ((Opty
248/940) presently working on it and therefore
down for the next day or two), and my memory
is that you might want to check that "105C"
slowdown number coz i'm pretty sure that it's
actually the same (or higher) than that of
my 6600GT

another possible factor is that the 6600GT is
120nm (110?), whereas the 76000GT is 90nm.
they both seem to run the same idle temps tho.
even at idle the 7600GT gives off noticeably
more heat, but it's when under load (which
i've yet to do with my 6600GT) when it is
especially noticable

fwiw, i've actually seen my 7600GT briefly
peak at 69/70C while playing Oblivion with
both a 20 minute continuous graph of CPU/case
temps running, as well as the new nView
20 minute continuous graph of GPU temps.
the CPU peaked at 59/60C

my 7600GT has been decently stable, but i've
not tried to do any overclocking of it.
my 6600GT has been flakier, but i do not
think temperature is a factor in the odd
glitches that seem to happen with somewhat
more frequency

if you're not running the latest nVidia
drivers (91.47), you might want to do that
coz the new Control Panel inclues that
useful continuous 20 minute temp graph

bill

 

[Folkert Rienstra]

Aidan Karley wrote:






i see that i now need to do a global kill on johnh (in my Agent n/g program)

You one of those ostriches?

re Curious George, despite being the hookee

and hooker.

he's said some sensible things and seems to have a brain.

Often so perceived by those that lack one.
The type that sticks it's head in the sand.

 

[Ed Light]

That is the right point.

Note that at least some A64's are rated for only 65C.

--
Ed Light

Smiley :-/
MS Smiley :-\

Send spam to the FTC at
(e-mail address removed)
Thanks, robots.

Bring the Troops Home:
http://bringthemhomenow.org

 

[willbill]

... My 7600GT has a slowdown-point of 105C but it
becomes flaky and crashes around 70...80C.

^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

one thought for you: i suspect you have
a case hot spot where your video card is
coz it shouldn't be running at 80C
(unless you are overclocking it)

put your hand on the case and if it is
noticeably warm then you have a hot spot
(with poor air flow at that point);
assuming you've got positive (or negative)
air pressure within the case, remove
the slot plate immediately below the
video card and you'll remove the hotspot

I think the slowdown point is way to
high on these things.

i simply don't know

what amazes me is that nVidia seems to
be ok with running their GPUs at *much*
higher temps (at least 20+C higher) than
anything i've ever seen for any modern CPU

i have to think that even if they
don't run reliable at higher temps,
that they somehow don't fail coz if
they did fail it would all show up
on nVidia's doorstep

bill

 

[Arno Wagner]

Arno Wagner wrote:


i have a 7600GT on my faster machine ((Opty
248/940) presently working on it and therefore
down for the next day or two), and my memory
is that you might want to check that "105C"
slowdown number coz i'm pretty sure that it's
actually the same (or higher) than that of
my 6600GT

another possible factor is that the 6600GT is
120nm (110?), whereas the 76000GT is 90nm.
they both seem to run the same idle temps tho.
even at idle the 7600GT gives off noticeably
more heat, but it's when under load (which
i've yet to do with my 6600GT) when it is
especially noticable

fwiw, i've actually seen my 7600GT briefly
peak at 69/70C while playing Oblivion with
both a 20 minute continuous graph of CPU/case
temps running, as well as the new nView
20 minute continuous graph of GPU temps.
the CPU peaked at 59/60C

my 7600GT has been decently stable, but i've

Same here. The cash around 78C or so was with reduced cooling.
I was experimenting because I replaced the very noisy original
cooler. I now have a Zahlman heatpipe cooler, highest temp
I have seen is maybe 60C.

not tried to do any overclocking of it.
my 6600GT has been flakier, but i do not
think temperature is a factor in the odd
glitches that seem to happen with somewhat
more frequency

if you're not running the latest nVidia
drivers (91.47), you might want to do that
coz the new Control Panel inclues that
useful continuous 20 minute temp graph

I have a tool for that.

Arno