My case is hot enough to cook a turkey

[jack]

Hello,

I recently built a new system around an Asus P5E (bios 402), C2D
E6850, ATI 2600 XT, and a CoolerMaster 534 case. I ran Prime95 last to
load test the system and about 25 minutes into the test, Asus PC Probe
was reporting a CPU temp of 91 deg Celsius and MB temp of 192 deg
Celsius ( yes, those numbers are Celsius not Fahrenheit... I'm not
making this up). Needless to say I immediately shutdown Prime95. The
CPU temp slowly came down to the mid 80's C and stayed there for about
3 or 4 minutes until I rebooted. I entered the Bios and went to
Hardware Monitor and the CPU temp was 40-42 deg C and MB temp was 35
deg C. I reran Prime 95 this morning for about 20 minutes and the CPU
temp never rose above 53 deg C and MB temp never went higher than 36
deg C.

Any opinions on what might have been going on last night. I should
also add that I am not overclocking the CPU nor the memory. Everything
is running at stock frequencies and voltages. Could this have damaged
the CPU in anyway?

 

[Eric Parker]

Hello,

I recently built a new system around an Asus P5E (bios 402), C2D
E6850, ATI 2600 XT, and a CoolerMaster 534 case. I ran Prime95 last to
load test the system and about 25 minutes into the test, Asus PC Probe
was reporting a CPU temp of 91 deg Celsius and MB temp of 192 deg
Celsius ( yes, those numbers are Celsius not Fahrenheit... I'm not
making this up). Needless to say I immediately shutdown Prime95. The
CPU temp slowly came down to the mid 80's C and stayed there for about
3 or 4 minutes until I rebooted. I entered the Bios and went to
Hardware Monitor and the CPU temp was 40-42 deg C and MB temp was 35
deg C. I reran Prime 95 this morning for about 20 minutes and the CPU
temp never rose above 53 deg C and MB temp never went higher than 36
deg C.

Any opinions on what might have been going on last night. I should
also add that I am not overclocking the CPU nor the memory. Everything
is running at stock frequencies and voltages. Could this have damaged
the CPU in anyway?

Well that is fortunate about this time of year ;-)

But seriously -

The thermal paste may have now settled.
I have noticed that it takes 2 or 3 hot/cold cycles before cooling
efficiency
reaches optimal, but nothing on that scale. Perhaps the heatsink was not
settled but did afterwards. The CPU may have been damaged. I'd give it
a 12 hour Prime95 * 2 test (both CPUs at 100%). Initially keeping a
close
eye on temps. The CPU may give up somewhere down the line but if it
was mine I'd just run it.

Eric

 

[UCLAN]

Hello,

I recently built a new system around an Asus P5E (bios 402), C2D
E6850, ATI 2600 XT, and a CoolerMaster 534 case. I ran Prime95 last to
load test the system and about 25 minutes into the test, Asus PC Probe
was reporting a CPU temp of 91 deg Celsius and MB temp of 192 deg
Celsius ( yes, those numbers are Celsius not Fahrenheit... I'm not
making this up). Needless to say I immediately shutdown Prime95. The
CPU temp slowly came down to the mid 80's C and stayed there for about
3 or 4 minutes until I rebooted. I entered the Bios and went to
Hardware Monitor and the CPU temp was 40-42 deg C and MB temp was 35
deg C. I reran Prime 95 this morning for about 20 minutes and the CPU
temp never rose above 53 deg C and MB temp never went higher than 36
deg C.

Any opinions on what might have been going on last night. I should
also add that I am not overclocking the CPU nor the memory. Everything
is running at stock frequencies and voltages. Could this have damaged
the CPU in anyway?

192 Celsius? That's close to 380 degrees F.

I think you were getting erroneous readings.

 

[Emerald Saint]

You can put an ordinary thermometer in the case just for a test. Maybe a
good way to calibrate the case temp monitor.

I have used ASUS Probe a lot. I notice CPU and Mainboard temps have shifted
10 deg C - maybe more - for no reason. This on two different ASUS P4P800SE
machines at different times. I don't think there is any actual temp
difference - it is the sensor or Probe utility. The shifts occurred after
more than1 year of use. Temp also shifts back. These changes are all
long-term. But it's 10 or 15 deg C changes - not hot like you say.

Bill S.

 

[~misfit~]

Somewhere on teh interweb jack typed:

Hello,

I recently built a new system around an Asus P5E (bios 402), C2D
E6850, ATI 2600 XT, and a CoolerMaster 534 case. I ran Prime95 last to
load test the system and about 25 minutes into the test, Asus PC Probe
was reporting a CPU temp of 91 deg Celsius and MB temp of 192 deg
Celsius ( yes, those numbers are Celsius not Fahrenheit... I'm not
making this up). Needless to say I immediately shutdown Prime95. The
CPU temp slowly came down to the mid 80's C and stayed there for about
3 or 4 minutes until I rebooted. I entered the Bios and went to
Hardware Monitor and the CPU temp was 40-42 deg C and MB temp was 35
deg C. I reran Prime 95 this morning for about 20 minutes and the CPU
temp never rose above 53 deg C and MB temp never went higher than 36
deg C.

Any opinions on what might have been going on last night. I should
also add that I am not overclocking the CPU nor the memory. Everything
is running at stock frequencies and voltages. Could this have damaged
the CPU in anyway?

Dump PC Probe as your monitoring software, download the latest version of
CoreTemp to check on the die temps of your CPU and SpeedFan for anything
else.

You give no particulars about the case cooling. If the case were minimally
cooled then you running Prime may have created a lot of heat that was then
stored in the air in the case, as well as all the other components. However,
that wouldn't explain the temps the second time around.

Also, you don't mention ambient temps. Was it 40°C cooler this morning?

 

[~misfit~]

Somewhere on teh interweb Emerald Saint typed:

You can put an ordinary thermometer in the case just for a test. Maybe a
good way to calibrate the case temp monitor.

I have used ASUS Probe a lot. I notice CPU and Mainboard temps have
shifted 10 deg C - maybe more - for no reason. This on two different
ASUS P4P800SE machines at different times. I don't think there is
any actual temp difference - it is the sensor or Probe utility. The
shifts occurred after more than1 year of use. Temp also shifts back.
These changes are all long-term. But it's 10 or 15 deg C changes -
not hot like you say.

I find that, often after updating the motherboard BIOS, the temps in ASUS
Probe change. They periodically change the software correction factor (which
is part of the BIOS) (hysteresis?) of the thermistor output in an attempt to
more accurately display the actual temp. For instance, a BIOS update on my
P5K-E last week raised the reading in Asus PC Probe II for the CPU by about
20°C from what it reading before the update. (Niether is correct, earlier it
was ~15°C low, now it's ~5°C high).

 

[kony]

Somewhere on teh interweb jack typed:

You give no particulars about the case cooling. If the case were minimally
cooled then you running Prime may have created a lot of heat that was then
stored in the air in the case, as well as all the other components. However,
that wouldn't explain the temps the second time around.

Also, you don't mention ambient temps. Was it 40°C cooler this morning?

If that Coolermaster case didn't come with a rear exhaust
fan (I vaguely recall reading some review where it might've
had a fan in front of the HDD bay but not the rear exhaust
fan), it would be good to either move the front fan to the
rear (assuming it is the same, 120mm size) or buy a fan to
place in the rear.

If it has the two vertical mesh panels with the solid front
as intake, and those mesh panels have foam filter inserts
behind them, this foam will also significantly reduce
airflow... even moreso after it begins to get clogged with
dust. I like filtered cases but so many of them implement
the filter poorly.

 

[Kent_Diego]

".....

If that Coolermaster case didn't come with a rear exhaust
fan (I vaguely recall reading some review where it might've
had a fan in front of the HDD bay but not the rear exhaust
fan), it would be good to either move the front fan to the
rear (assuming it is the same, 120mm size) or buy a fan to
place in the rear.
...

I really like having the fan in the front. It blows air over the hard drives
keeping them cool. The cooler your hard drive, the longer it lasts. I just
got a new Cooler Master case and it came with one rear exhaust fan. I
removed the front cover and re-located fan there. The Cooler Master cases
have ample venting in back for exhaust. The OP's problem is related to his
temperature sensors/monitoring software and not actual overheating.

 

[kony]

".....

I really like having the fan in the front. It blows air over the hard drives
keeping them cool. The cooler your hard drive, the longer it lasts. I just
got a new Cooler Master case and it came with one rear exhaust fan. I
removed the front cover and re-located fan there. The Cooler Master cases
have ample venting in back for exhaust. The OP's problem is related to his
temperature sensors/monitoring software and not actual overheating.

A HDD in front can be useful, particularly with their
ill-conceived HDD bay that is turned sideways for easier
access instead of better HDD cooling, BUT that fan should
not be added until _after_ a rear chassis fan is used, it
should always be the 3rd fan added after PSU and rear
chassis exhaust.

 

[~misfit~]

Somewhere on teh interweb kony typed:

A HDD in front can be useful, particularly with their
ill-conceived HDD bay that is turned sideways for easier
access instead of better HDD cooling, BUT that fan should
not be added until _after_ a rear chassis fan is used, it
should always be the 3rd fan added after PSU and rear
chassis exhaust.

I'm pleased to see someone else who has problems with the current trend of
side-entry HDD bays. I find it *significantly* impacts cooling, causes the
case to be wider (and/or necesitates the use of 90° SATA connectors) and for
what? How often does the average user, hell make it the average PC
enthuaiast, fit and remove HDDs? I tend to shuffle drives around from PC to
PC a bit more than most I guess and there's still no way I'd trade off good
cooling for 2 minutes faster HDD removal/fitting.

Whew! Thanks for letting me get that off my chest. I've built a few
enthusiast systems of late in cases with side entry HDD bays (parts selected
by the owners, (usually with my input, although I've been leaving the choice
of case to the end-user 'til now) sent to my address, I build, overclock and
then test stability before delivery). These otherwise well desgined cases
are ruined by the turbulence and lack of directed airflow caused by this
sideways bay. In traditionally mounted HDD systems I often consider the HDDs
themselves as part of the cooling solution, they act as baffles to direct
airflow towards the bottom rear of the case and the graphics card. In cases
with side mouted drives this doesn't happen, the graphics card area cooling
suffers, all for the "covenience" of saving a few minutes a year maybe.

It's a gimmicky thing that needs to stop. IMO.

 

[milleron]

Somewhere on teh interweb kony typed:

I'm pleased to see someone else who has problems with the current trend of
side-entry HDD bays. I find it *significantly* impacts cooling, causes the
case to be wider (and/or necesitates the use of 90° SATA connectors) and for
what? How often does the average user, hell make it the average PC
enthuaiast, fit and remove HDDs? I tend to shuffle drives around from PC to
PC a bit more than most I guess and there's still no way I'd trade off good
cooling for 2 minutes faster HDD removal/fitting.

Whew! Thanks for letting me get that off my chest. I've built a few
enthusiast systems of late in cases with side entry HDD bays (parts selected
by the owners, (usually with my input, although I've been leaving the choice
of case to the end-user 'til now) sent to my address, I build, overclock and
then test stability before delivery). These otherwise well desgined cases
are ruined by the turbulence and lack of directed airflow caused by this
sideways bay. In traditionally mounted HDD systems I often consider the HDDs
themselves as part of the cooling solution, they act as baffles to direct
airflow towards the bottom rear of the case and the graphics card. In cases
with side mouted drives this doesn't happen, the graphics card area cooling
suffers, all for the "covenience" of saving a few minutes a year maybe.

It's a gimmicky thing that needs to stop. IMO.

I hear ya, but how do you know that the difference is significant? I
have an Antec P180 with the side-mounted SATA drives, and they both sit
in a tunnel with air pumped from front intake to rear exhaust by a 120mm
case fan and the 120mm PSU fan. I deduce that the HDs are actually
running cooler than they would in a conventional case, but I've no way
to to tell.
Have you built essentially identical systems in the two different kinds
of cases, so that you can actually make a direct comparison? I'd be
very interested in the numbers. You may well be on to something here.

Ron

 

[Noozer]

I'm pleased to see someone else who has problems with the current trend of

side-entry HDD bays. I find it *significantly* impacts cooling, causes the
case to be wider (and/or necesitates the use of 90° SATA connectors) and
for what? How often does the average user, hell make it the average PC
enthuaiast, fit and remove HDDs? I tend to shuffle drives around from PC
to PC a bit more than most I guess and there's still no way I'd trade off
good cooling for 2 minutes faster HDD removal/fitting.

I REAAAAALLY love the sideways bays. There is nothing worse than having to
unloom cabling and pull a video card just to take a hard drive out of a
system.

....and, although case makers don't do this, it does give a straight path
through the system to cool the drives without introducing the heat into the
system. Simply seal the drive bays from the rest of the system and vent the
drive bay area on both sides. Add a fan to one side and there you go!

 

[Paul]

I hear ya, but how do you know that the difference is significant? I
have an Antec P180 with the side-mounted SATA drives, and they both sit
in a tunnel with air pumped from front intake to rear exhaust by a 120mm
case fan and the 120mm PSU fan. I deduce that the HDs are actually
running cooler than they would in a conventional case, but I've no way
to to tell.
Have you built essentially identical systems in the two different kinds
of cases, so that you can actually make a direct comparison? I'd be
very interested in the numbers. You may well be on to something here.

Ron

I have an Antec Sonata. The lower bays are sideways. But they
are mounted next to the intake vents. The two drives currently
installed, run at 31C and 32C respectively. (Room temp about 25C.)
There are also holes in the rack, to allow airflow. That is why
it all works.

If I used the regular bays in the upper part of the computer,
there are no intake vents up there, so the drive temp would be
hotter in that case. Only an optical drive sits up there currently.

Paul

 

[kony]

I REAAAAALLY love the sideways bays. There is nothing worse than having to
unloom cabling and pull a video card just to take a hard drive out of a
system.

True, but look into case that have a removable front bezel,
then you can pull them out the front instead of the rear.

...and, although case makers don't do this, it does give a straight path
through the system to cool the drives without introducing the heat into the
system. Simply seal the drive bays from the rest of the system and vent the
drive bay area on both sides. Add a fan to one side and there you go!

Unfortunately, fans mounted on external panels leak noise
into the room. It is especially true with side panels
because they are usually straight sheet metal instead of
made more rigid with small flat areas before there are folds
in the metal. If I had to choose between cool drives and
quiet, cool would win every time but given a case where all
the passive intake flows through the drive bay, you can have
both w/o a front fan.

 

[kony]

I hear ya, but how do you know that the difference is significant?

Common sense?

On the one hand there are properly engineered HDD bays where
100% of the intake fan is through the HDD bay, and on the
other hand, there are sideways bays where maybe 20% goes
trough them.

I'm not saying a sideways bay can't work, certainly it can,
but when a fan is mounted on the front of a case it usually
results in the most noise emission of any (otherwise low
RPM/quiet) fans in the system.

If a fan is meant for HDD rack cooling, the key to keeping
good temps at low noise levels is maximizing efficiency.

With a good case design, drawing all air intake through the
HDD bay, a front fan isn't even needed. A resonable PSU and
rear 120mm chassis fan can adequately cool a few HDD if the
intake is all flowing through the bay.

Unfortunately, too many cases are poorly designed such that
all the intake is not through the HDD bay. They often have
short-loop intake holes across from the motherboard slots,
passive intake holes in the side panel, and low tolerance
parts that result in intake between external bays and
elsewhere. It is a shame they take the lazy way out but
given some prep-work (and a case worth the work) these
shortcommings can be overcome.

I
have an Antec P180 with the side-mounted SATA drives, and they both sit
in a tunnel with air pumped from front intake to rear exhaust by a 120mm
case fan and the 120mm PSU fan. I deduce that the HDs are actually
running cooler than they would in a conventional case, but I've no way
to to tell.

It may depend on how you define conventional.
Certainly there are worse cases. I always strive for *best*
(within reason).

Have you built essentially identical systems in the two different kinds
of cases, so that you can actually make a direct comparison? I'd be
very interested in the numbers. You may well be on to something here.

It's not (temp) numbers, since there are several ways one
can build what is cool enough, it's noise and dust buildup
from any given design, and/or, maintenance interval since it
is easy enough to just mount a half dozen low RPM fans on a
case, but stray airflow does more to build up dust than hit
problem cooling areas, unless everything is filtered but
doing that again increases maintenance interval to change
filtes, and fan noise to overcome the impedance to airflow
from the filters.

Getting maximal cooling depends on routing airflow across
the parts that need cooling and minimizing stray airflow.
Key areas that need more airflow (than the rest) include
HDDs, video card, CPU, motherboard VRM subcircuit
(especially if overclocking or marginal motherboard
components) and PSU.

PSU is a catch-22, you want good intake with minimal
obstruction but still a 2nd case exhaust so the PSU is not
drawing in all the heated air.

 

[~misfit~]

Somewhere on teh interweb Paul typed:

I have an Antec Sonata. The lower bays are sideways. But they
are mounted next to the intake vents. The two drives currently
installed, run at 31C and 32C respectively. (Room temp about 25C.)
There are also holes in the rack, to allow airflow. That is why
it all works.

There are also holes up by the fans where the word "Antec" is punched out of
the side of the case. I'd think that would provide a short-cut for at least
some air.

If I used the regular bays in the upper part of the computer,
there are no intake vents up there, so the drive temp would be
hotter in that case. Only an optical drive sits up there currently.

milleron you have a non-conventional case (I believe?) where the only area
of entry for fresh air is directly in front of the bays. For the purposes of
airflow patterns you essentially have two cases, one long thin one (think
direct front-to-back airflow) for the HDDs/PSU and another for the rest
sitting on top. As you say, a tunnel. Your experiences are only valid for
others with that case type.

I've built a few systems recently in conventional cases that have 120mm slow
running fans directly in front of the HDD bay pushing air into the case.
Some with traditional HDD bays and a couple with side entry bays. Just
looking at them you should be able to predict the result. However, I've
monitored them as well and see nearly 10°C higher temps on the side entry
bays HDDs. The traditional bays let almost all the air pass through between
or around the drives. The side-mounts rely on a small percentage of the air
going through some smallish holes between the drives. However, when most of
the air is actually bouncing off the steel sheet that the holes are in it
tends to create a flow pattern that makes it even more unlikely that
significant amounts of air will go through the holes. The air exiting the
area to the side creates an area of low pressure behind it that *draws* more
air after it.

Air takes the path of least resistance and, unless the fan is jammed up
against the holey panel with nowhere else for the air to go *but* the holes
then it'll mostly go sideways. If it *is* jammed up against the panel then
it's efficiency will be severely compromised.

This is obviously more important where you have 'stacked' drives, as in a
RAID array or several as JBOD. If you've got just one or two HDDs set well
apart then convection/diffusion will help keep them cool.

 

[milleron]

Common sense?

Common sense tells you that there's going to be a difference. However,
I asked about the SIGNIFICANCE of the difference. Only a thermometer is
going to answer that question.

 

[kony]

Common sense tells you that there's going to be a difference.
However,
I asked about the SIGNIFICANCE of the difference. Only a thermometer is
going to answer that question.

Only a thermometer will give an exact number, but an exact
number is not needed to see significance. For example I
can see the trees on the hill are a significant distance
away - I don't need an exact measurement to know it.

 

[Pecos]

A HDD in front can be useful, particularly with their
ill-conceived HDD bay that is turned sideways for easier
access instead of better HDD cooling, BUT that fan should
not be added until _after_ a rear chassis fan is used, it
should always be the 3rd fan added after PSU and rear
chassis exhaust.

You are so right Kony.

I have a Cooler Master Mystique and a sideways mounted drive bay. It was
only a nuisance until this week when I wanted to do some renovations.

I have two RAID SATA drives and one PATA drive that was sitting in the 3
1/2 inch bay above my flash card reader. I wanted to move the PATA drive
to the drive bay where it would get more airflow.

The problem with that idea became apparent rather quickly. I tried *two*
18 inch IDE cables and neither had enough cabling between the two
Master/Slave connectors to span the distance from the 5.25 inch DVD burner
mounted in the lowest bay slot and the PATA drive in the sideways mounted
top drive bay slot.

I had to use two IDE cables with each drive as Master to two different IDE
connectors on the motherboard. I looked for a motherboard that had two IDE
channels so I was fortunate in that sense, but most motherboards these days
only have one IDE channel.

And then there are the motherboard designs where DIMMs are snug up against
one another if the DIMM has a heat spreader. These heat spreaders do a
great job spreading the heat to its twin sister DIMM in the adjacent slot.

And while we are ranting about motherboard and case design, which I think
is fair at best when it comes to cooling, why is it that my passively
cooled video card has to sit in the case with the heatsink on the
*underside* where it gets little to no airflow at all?

And someone needs to take a look at a way to cool DVD burners. I had one
that stopped working this summer while I was watching a movie. The DVD
drive was very hot - so hot that I thought I had fried it. I thought the
cable was bad because the DVD drive worked after a cable swap, but looking
back on it the drive might have simply cooled down enough to work again.

Most cases have 3.5 and 5.25 inch bays that are enclosed on three sides
allowing for no airflow at all. Not a good place for a hard drive or a
$500 Blu-ray burner!

Honestly, I think this group could design a better case than 90% of what
you can buy for under $100 U.S.

 

[milleron]

Only a thermometer will give an exact number, but an exact
number is not needed to see significance. For example I
can see the trees on the hill are a significant distance
away - I don't need an exact measurement to know it.

Uhhh . . . that's not going to win the prize for Best Analogy of Western
Man. It's going to take measurements to tell if this difference is
significant. That measurement might even be a finger-touch test if the
difference is large enough, but intuition tells me that's not going to
be the case.