View Single Post
  #7  
Old July 6th 20, 06:15 PM posted to microsoft.public.windowsxp.general
Paul[_32_]
external usenet poster
 
Posts: 11,873
Default High case temperatures

J. P. Gilliver (John) wrote:
On Mon, 6 Jul 2020 at 00:27:18, Paul wrote:
AK wrote:
My computer was doing an image save of my hard drive to a second hard
drive. (Gparted)
I have done that many times.
But I heard some unusual sounds coming from my computer case.
So I took off one of the side panels and could feel some heat that
seemed uncharacteristically high.
I put a thermometer in and it read around 100 degrees F.
After the process, the temp went to around 80 degrees.


80-100 *F* I don't think would worry me that much.

Do I need to be concerned?


Worth looking into the "unusual sounds".

Thanks,
Andy
https://www.goodnewsnetwork.org/


Find the fan that isn't spinning ?


And/or the vent/filter/whatever that's even partially blocked.

Does your power supply provide the only exhaust on
the back of the PC ? Is there a separate fan for venting
hot air ? Is there an intake vent near the front of the PC ?


Especially if it's low down, so likely to pick up more carpet fluff.

Paul

John


"So I took off one of the side panels and could feel
some heat that seemed uncharacteristically high."

When someone mentions a "hot cloud" effect, regardless of
temperature, that tells me there's no airflow in the computer
case. Take the side off and look. How is the ventilation
supposed to work ? What is missing from the setup ? Does
the setup make sense ? Is there a predominant front to back
cooling path, versus a confused mess with 7 fans pointing
in all directions ? While a BTX casing might have the
directions modified for front exhaust, you can still see
the "plan" on OEM machines, what the cooling is supposed to do.
There should be a predominant direction, not fans fighting
with one another for supremacy.

I have a CPU with a 156W heat output at max. Is there
a hot cloud behavior in the case ? Not that I can detect.
I have had cases with a hot cloud, even though the CPU
drew less power (95W max) and back in those days, the CPU power
used came in at the rated TDP or lower. On really modern
equipment, you'd be better off monitoring with a Kill-O-Watt,
to understand just how much the power consumption shoots up
under load. (The Kill-O-Watt won't lie.)

The hot air from the ATX PSU exhaust has largely disappeared,
with the advent of 80+ ATX power supplies (more efficient,
less waste heat in the PSU itself). The idle power on computers
has dropped a bit. An 8800GTX from idle to gaming, had only
a ratio of 2:1 on power (it didn't save a lot of power when
it dropped back to idle). Modern video cards can have a
ratio of 10:1 to 15:1 or so, with the clock dropping to 100MHz
if nothing is going on. And this behavior helps manage the heat
better. Less chance of contributing to a heat cloud.
But not zero chance.

It's a job for a calibrated eyeball. Can you justify the behavior ?
Is something amiss in there ? On some video cards, it can be
pretty hard to get a mirror in there so you can visually verify
the video card fan is still working. Some of the others are
easier to check.

If the case has filters (one of my cases has a "window screen"
in the front), make sure the filter isn't blocked with dust.

The worst plugging of a computer I've ever seen, involved several pounds
of human hair. (The machine operator had long hair, and seemed to
shed like a puppy :-) It's an unbelievable amount of hair for
a human to lose and not be bald.) I checked that computer even
though it "wasn't mine" and a funny noise was coming from it.
"Strangely muffled" fan noise was the giveaway. The computer
was fine, because it had a bipolar (ECL!) CPU and could take the heat.
Could run at 150C without losing its mind. The heatsink on it was
a rather large milled aluminum assembly with three fans for cooling.

If something you see doesn't add up, check.

You can use the case cooling equation, to estimate the CFM rating
of the exhaust fan needed. A 7C (10F) delta T is a good target
for a well cooled computer. Some computers have close to zero
air circulation near the hard drive, and that's something I would
try to improve if possible. One of the drive positions in my
other computer is in an airflow shadow and the drive ends up
warmer than the other drives.

But without a hand calculator, just your calibrated eyeball can
spot a problem where the behavior can be improved.

And don't be over-aggressive cleaning fan blades... You can
easily damage some brands of cooling fans, just by handling
them (bearing damage on spring-loaded style setups).

Paul
Ads