Hot Ass Hard Drive? When is hot too hot?

[kony]

Hogwash. 35C is neither arbitrary nor unrealistic for
adequately cooled hard drives. As for 30C I said it
requires a lot of airflow, although currently I have three
Seagate X15's running in a midtower case with just the
stock front fan, and none of them ever get over 30C.

You fail to mention if this is from the HDD temp sensor,
report.

You fail to mention what your ambient, room temp is.

In general, you are mistaken. The typical drive reports
more than 6C above ambient (if we arbitrarily presume 24C is
a typical room temp).

If you want to create a louder (to create), unnecessary
level of airflow, go right ahead - but plenty of cases and
users demonstrate there is a sane middle ground between
overheating drives and going overboard to hit an arbitrary
target temp like 30C.

 

[Guest]

You fail to mention if this is from the HDD temp sensor,
report.

It's from the drives' SMART.

You fail to mention what your ambient, room temp is.

Average, ~22C

In general, you are mistaken. The typical drive reports
more than 6C above ambient (if we arbitrarily presume 24C is
a typical room temp).

Close enough to 30C.

If you want to create a louder (to create), unnecessary
level of airflow, go right ahead - but plenty of cases and
users demonstrate there is a sane middle ground between
overheating drives and going overboard to hit an arbitrary
target temp like 30C.

I never claimed there wasn't "sane middle ground" between
the two. Look, I'm a hardware design engineer and qualify
hard drives for a living: on average, most drives will run much
longer at 35C than 40C, and a bit longer at 30C than 35C.
Whether people choose to do what's necessary to achieve
these temps is entirely up to them.

 

[kony]

I never claimed there wasn't "sane middle ground" between
the two. Look, I'm a hardware design engineer and qualify
hard drives for a living: on average, most drives will run much
longer at 35C than 40C, and a bit longer at 30C than 35C.
Whether people choose to do what's necessary to achieve
these temps is entirely up to them.

If you had a particularly problematic design such that there
was an individual component overheating, something that is
not reflected by the temp taken at another spot and reported
by SMART, it could certainly help to cool the drive more,
rather than less. That still does not bear directly on
whether the smart temp is lower and does not in itself
justify a random presumption that it is useful to get a
drive down to especially low temps.

A drive might run a bit longer at 30C than 35C, but how long
was really useful? We could already expect in excess of 5
years from 35C or more, some even closer to a decade, with
anything over that being a gamble, and that with a drive of
lower capacity and performance. Keeping something running
beyond it's reasonable lifespan isn't always so important.

.... or to put it another way, if data retention is important
to you, are you going to keep using a 6+ year old HDD no
matter how cool you manage to get it? The answer is no, and
the extreme airflow rates necessary for the expectation of
longer life is in itself more of a detraction than the last
years of a HDD's potential life. Even in mission critical
servers, drives do not usually run as low as 30-35C.

 

[Bob Brown]

Ok, is your drive in the very bottom tray (the tray adjacent
to the two front intake fans? Are the two front fans installed?

This case only came with one fan, blowing air out the back just below
the PSU fan outlet.

Have you tried running with the front panel open? Does it
lower your drive temps?

No, I have not tried that. So you think a range of 44C-50C for the
hard drive is bad, very bad?

 

[Bob Brown]

It is arbitrary and unrealistic to have a goal of 35C max as
a drive reported temp, and practically no drive in any
reasonable chassis would ever be under 30C.

There is a relevant benchmark about drive surface temp here,
http://www23.tomshardware.com/storage.html?modelx=33&model1=117&model2=676&chart=39

nice link. seems my flavor of drive is basically the worst, the WD
160GB drive seems very good. agreed?

I need more than my current 80gb drive but only about double+ that
size.

I wouldn't need 500gb right now.

importance to me seems to be

temp > speed > price > size

 

[Bob Brown]

Mount a heatsink on your hard drive, if there is space. In a
commercial application in a hot environment I have used a finned
heatsink of a similar base size to the base of the hard drive and with
7 x 1" fins on it. (it was taken from a junked 60w audio amp) screw
it firmly to the drive, making sure it doesn't short circuit on
anything in the process, and use heatsink compound on touching
surfaces.

I've never heard of a heatsink for a hard drive. I'm way behind the
times I guess. What would work for me and can you tell me if newegg is
a good place to buy these from?

thanks,

 

[John Doe]

Using one of those cute monitor programs I noticed my hard drive
is going up to 50C, idles around 44C.

Is this hot as hell like I think it is?

Is there a way to stop, limit, prevent my hard drive from getting
so GD Hot?

A computer case with an intake fan blowing on the hard drive rack.

Also, why are drive made this way?

They are expected to be installed in a proper environment.

My TV and MONITOR don't overheat

Some people would disagree with that.

 

[Rod Speed]

This case only came with one fan, blowing air out the back just below
the PSU fan outlet.


No, I have not tried that. So you think a range of 44C-50C for the
hard drive is bad, very bad?

Maxtors dont live that long at those sorts of temps.

 

[kony]

The case's rear exhaust fan suffers degraded exhaust rate
due to the stamped metal grill behind it. Cutting out that
grill would improve general system airflow rate some.

The drive rack design is flawed, putting the drive in the
upper of the two positions will work better overall (though
the spot-temp reported by the SMART monitor may not change
much either way). Further the front metal wall of the case
allows too great a percentage of the airflow to flow around
and above the drive rack instead of through it. If the rack
could be positioned higher (difficult to determine, "Maybe"
drilling a hole or two on the motherboard tray and a space
block on the closer side would allow this, but it's a fair
amount of work and would best be done after removing all
parts from the case to avoid bits of metal getting
everywhere) and a partition or duct installed that kept the
majority of the intake flowing under the drive that would
help.

In other words, the two best alternatives are to replace the
case or modify the existing case. If you can accept higher
noise levels, you could instead install fans in the two
positions available in front of that drive rack, and they
wouldn't need to be particularly high RPM. Even then,
having those stamped-metal grills cut out would improve the
airflow:noise ratio considerably.

Maxtors dont live that long at those sorts of temps.

Keep in mind that Rod has a bug about Maxtors, your drive is
going to run a little higher than the ambient temp due to
your case, and that case could be improved, but a SMART
report of 44-50C isn't exceptionally high for any brand of
drive.

 

[kony]

nice link. seems my flavor of drive is basically the worst, the WD
160GB drive seems very good. agreed?

You are drawing the wrong conclusion from the data. The
conclusion is your drive isn't running very hot.

I need more than my current 80gb drive but only about double+ that
size.

I wouldn't need 500gb right now.

importance to me seems to be

temp > speed > price > size

I would do the case modifications mentioned in my prior post
first, or replace the case. Even with those modifications
the case can't cool two drives very well because they're
stacked too closely together, but having two front fans and
a partition (piece of plastic screwed tot he case wall
should suffice) would probably be good enough, unless the
room temp gets significantly higher at other times of year.

 

[Rod Speed]

The case's rear exhaust fan suffers degraded exhaust rate
due to the stamped metal grill behind it. Cutting out that
grill would improve general system airflow rate some.

The drive rack design is flawed, putting the drive in the
upper of the two positions will work better overall (though
the spot-temp reported by the SMART monitor may not change
much either way). Further the front metal wall of the case
allows too great a percentage of the airflow to flow around
and above the drive rack instead of through it. If the rack
could be positioned higher (difficult to determine, "Maybe"
drilling a hole or two on the motherboard tray and a space
block on the closer side would allow this, but it's a fair
amount of work and would best be done after removing all
parts from the case to avoid bits of metal getting
everywhere) and a partition or duct installed that kept the
majority of the intake flowing under the drive that would
help.

In other words, the two best alternatives are to replace the
case or modify the existing case. If you can accept higher
noise levels, you could instead install fans in the two
positions available in front of that drive rack, and they
wouldn't need to be particularly high RPM. Even then,
having those stamped-metal grills cut out would improve the
airflow:noise ratio considerably.

Keep in mind that Rod has a bug about Maxtors,

No I dont, I just realise that maxtors dont live long at those sorts of temps.

your drive is going to run a little higher than
the ambient temp due to your case,

That is a hell of a lot more than a little higher.

and that case could be improved, but a SMART report of
44-50C isn't exceptionally high for any brand of drive.

Pig ignorant drivel with winter room temps.

 

[KR]

I've never heard of a heatsink for a hard drive. I'm way behind the
times I guess. What would work for me and can you tell me if newegg is
a good place to buy these from?

thanks,

It is something you will probably have to make yourself, you can buy
aluminium heatsink from
electronic component suppliers, (see below) or you can buy a dead hi-
fi amplifier (would have to be over 50wrms rating) from the rubbish
dump sales, flea markets etc for a couple of $ and use it's heatsink,
cut and drill it yourself to suit the drive.

I had a look at the newegg site for you, they seem to only have CPU
heatsinks.

This is the site of a local Australian component supplier, something
like this item

http://www.dse.com.au/cgi-bin/dse.storefront/45c7dfd305f7a2362743c0a87f9c075b/Product/View/H3426

is what you need for heavy duty hard drive cooling.

With some cutting and drilling, this item is sufficient size for 2
hard drives (cut it in half longways) and substantially brings down
the temperature
with ease ! . I have used these before with great results.

 

[kony]

That is a hell of a lot more than a little higher.

Previously I linked a chart at Tom's hardware, note there
are several drives in the mid to upper 40's (C), it is not
"a hell of a lot" of anything, that is not at all an
unexpected temp, especially considering the sub-optimal
case cooling.

Pig ignorant drivel with winter room temps.

If you were to ever venture out of your log cabin, consider
hiking down the mountain to civilization. We now have
marvelous contraptions that produce heat, do a reasonably
good job of keeping a room warm even in winter.

 

[Rod Speed]

Previously I linked a chart at Tom's hardware, note
there are several drives in the mid to upper 40's (C),

And that is more than a little higher than ambient.

it is not "a hell of a lot" of anything,

Corse its a hell of a lot higher than ambient.

that is not at all an unexpected temp, especially
considering the sub-optimal case cooling.

Irrelevant to whether its a little higher than ambient.

Clearly it isnt.

If you were to ever venture out of your log cabin, consider
hiking down the mountain to civilization. We now have
marvelous contraptions that produce heat, do a reasonably
good job of keeping a room warm even in winter.

Thats still an exceptional temperature for a hard drive in winter.

 

[Walter Shreeve]

On 5 Feb 2007 17:59:28 -0800, KR wrote

<snip> you will probably have to make yourself <snip>
http://www.dse.com.au/cgi-bin/dse.storefront/45c7dfd305f7a2362743c0a87f9c075b/Product/View/H3426
is what you need for heavy duty hard drive cooling.

Au contraire. There is a selection of hard drive coolers here
<http://www.frozencpu.com/cat/l3/g34/c147/s238/list/p1/Bay_Devices-Bay_Coolers-HDD_Coolers-Page1.html>.
Unfortunately, every item is more expensive than that massive
heatsink.

-- Walter Shreeve

 

[Ed Medlin]

Maxtors dont live that long at those sorts of temps.

About 1/3 of Maxtors seem to die no matter what temp they are.........and
much too quickly it seems.

Ed

 

[Ed Medlin]

The case's rear exhaust fan suffers degraded exhaust rate
due to the stamped metal grill behind it. Cutting out that
grill would improve general system airflow rate some.

The drive rack design is flawed, putting the drive in the
upper of the two positions will work better overall (though
the spot-temp reported by the SMART monitor may not change
much either way). Further the front metal wall of the case
allows too great a percentage of the airflow to flow around
and above the drive rack instead of through it. If the rack
could be positioned higher (difficult to determine, "Maybe"
drilling a hole or two on the motherboard tray and a space
block on the closer side would allow this, but it's a fair
amount of work and would best be done after removing all
parts from the case to avoid bits of metal getting
everywhere) and a partition or duct installed that kept the
majority of the intake flowing under the drive that would
help.

In other words, the two best alternatives are to replace the
case or modify the existing case. If you can accept higher
noise levels, you could instead install fans in the two
positions available in front of that drive rack, and they
wouldn't need to be particularly high RPM. Even then,
having those stamped-metal grills cut out would improve the
airflow:noise ratio considerably.



Keep in mind that Rod has a bug about Maxtors, your drive is
going to run a little higher than the ambient temp due to
your case, and that case could be improved, but a SMART
report of 44-50C isn't exceptionally high for any brand of
drive.

Not just Rod, Kony. I have had many problems in the last couple of years
with Maxtors myself. I just won't use them for anything critical anymore. I
certainly wouldn't use them in a customer's system. I use mostly WDs now
since I purchase from a local distributor. I rarely have problems with them
and the OEM drives I get have a full 3yr warranty instead of the 1yr with
retail drives.

Ed

 

[kony]

About 1/3 of Maxtors seem to die no matter what temp they are.........and
much too quickly it seems.

Perhaps 1/3rd of yours have but this is definitely not the
average seen across the industy, a failure rate that high
would be constant news everywhere... and I mean constant,
not scattered reports of a few failing, there'd be hundreds
of reports every day considering number of people using
them, not even considering the several Maxtors here that
have not failed at even 1/10 rate (within the first 4 years,
after which I too often lose track of them or they aren't
being ran often enough to draw any useful conclusion about
whether a more regular usage pattern would have a different
result).

So I have to wonder why my Maxtors aren't dying? I can't be
THAT lucky, out of dozens of drives I've had a couple from a
few brands fail, but the Maxtors no moreso than other
brands. The (now several generations old) Plus 8 series did
seem susceptible to overheat logic-lockup, but that was not
a permanent failure. I could put one out on a desk and
leave it sitting till it became inaccesible but powering
off, letting it cool or putting into reasonably ventilated
system case was enough.

 

[Rod Speed]

About 1/3 of Maxtors seem to die no matter what temp they
are.........and much too quickly it seems.

Yeah, there's a reason Maxtor went bust and got absorbed by Seagate.

 

[Charlie]

It's from the drives' SMART.


Average, ~22C


Close enough to 30C.


I never claimed there wasn't "sane middle ground" between
the two. Look, I'm a hardware design engineer and qualify
hard drives for a living: on average, most drives will run much
longer at 35C than 40C, and a bit longer at 30C than 35C.

How much longer will a drive run at 35c rather than 40c and how much longer
still if it is kept at 30c?

Charlie