Archived from groups: alt.comp.periphs.mainboard.asus (
More info?)
In article <SjQxc.56325$Np3.2505841@ursa-nb00s0.nbnet.nb.ca>, "Jody
Pellerin" <jodypell@NOSPAMnb.sympatico.ca> wrote:
> Thank you. That was very interesting. Today, the temperature outside was 34
> C and it wasn't any cooler in here. So the room was probably quite high as
> well. I have one 80mm case fan, on the back. It's a generic one which came
> with the case. I have two fans on the power supply, also built in. Antec
> 350watt. And then there's the stock CPU heatsink and fan. However, I'm using
> Artic Silver 5 compound.
> I can feel that cloud of heat around this computer. It's right by my legs
> and it's in a bit of an enclosed area because of the way this room is
> arranged, leaving no room for it elsewhere.
> I can see I'm probably going to have heat problems this summer. The CPU is
> now at 52. It doesn't go too much higher under load, for some reason. Today,
> it's ranging from what it is now to 57 C. It hovered at 55 C for the most
> part.
> So if I understand right, the highest temperature this can reach safely is
> 75 C?
Well, I don't know whether the CPU temp is a diode temp or a socket
temp, as that can make a small difference. So, maybe treat the 75C
as the limit for now.
The fact that there isn't much difference between idle and loaded
temperature, could mean that the CPU heatsink is working pretty
well, and the problem is getting the heat out of the case. One
simple test, is to take the side off the computer and repeat the
readings. If temps drop a lot, then an extra fan on the case would
help. Case air moves from lower front to upper rear, so place the
fan to enhance that movement.
As far as max temperature is concerned, I would be more worried
about the disk drive, and the power supply, than the processor
at those temperatures. The disk drive is sensitive to relative
humidity as well, so if the room isn't air conditioned, a RH
of 60% can occur in summer, and that isn't good for the drive.
For the disk drives, depending on how many you have in the case,
you might want to space them out, to allow as much air to move
between them as possible.
It is pretty hard to increase flow rate through the case, without
getting an increase in noise. One concept used in cases now, is to
place a door over the drive area on the front of the computer, to
muffle the sound. As well, some cases "breathe" via a vent right
at the bottom front of the case, so fan noise doesn't come right
out and hit the user. On the Antec Sonata I have here, the noise
is shifted "down spectrum", so fan noise is more of the low
frequency component, than is the case with some of the older cases.
It is still annoying though. The other enemy in case design, is
metal resonance, and this case is made from 1 "U" shaped piece of
sheet metal, with no rivets to join the pieces.
You might consider moving the computer case away from its
enclosure. I place mine on top of my computer desk.
HTH,
Paul
> "Paul" <nospam@needed.com> wrote in message
> news:nospam-0906042214150001@192.168.1.177...
> In article <E7Kxc.56161$Np3.2496894@ursa-nb00s0.nbnet.nb.ca>, "Jody
> Pellerin" <jodypell@NOSPAMnb.sympatico.ca> wrote:
>
> > Today has been really hot here and my CPU temperature is at 56 C and it's
> > not even under load. It's a P4 2.8C HT. P4PE-X motherboard.
> > What would you guys say the safe range for temperatures are for my CPU?
> What
> > is the danger zone?
> > Thanks in advance!
> > Jody
>
> The max case temperature listed here is 75C.
>
>
http://processorfinder.intel.com/scripts/details.asp?sSpec=SL6WT&ProcFam=483&PkgType=ALL&SysBusSpd=ALL&CorSpd=ALL
>
> Power dissipation listed is 70W.
>
> If the thermal resistance of the HSF is 0.25C/W, and the air
> inside the case is resting at 25C, the processor temp will be
> 25+(70*0.25)=42.5C at full processor load. If you cannot keep
> the case air temp down, by using enough case fans, then the
> first number in the equation will contribute to a warmer processor.
> A room temp of 32C plus a delta_T from room air to case air of
> 10C, would make the first number in the equation 42, and give
> a processor temp of 42+(70*0.25)=59.5C at full processor load.
>
> Since you are running idle and seeing 56, I would expect your
> load temp to be much worse. I've measured my 2.8C (but with
> HT disabled), and at idle in Win2K, it draws 13W. Using the
> last, pessimistic equation again 42+(13*0.25)=45.5C and you
> are seeing 56.
>
> Notice in the discussion above, that three numbers are needed
> to conduct a thermal analysis. The room temperature, the case
> air temperature, and the processor temperature. Without all
> three, it is hard to make a recommendation. My room to case
> delta is maybe 3 or 4C right now, and if yours is 10-15C from
> room temp to case temp, an extra fan might help. (I recommend
> a minimum of the PSU fan plus one 80mm case fan on the back
> of the computer, for any processor. The PSU fan thermal control
> is not set up to cool the computer case properly by itself, and
> you can effectively ignore the PSU fan as being of any use.)
>
> In the analysis above, I used a thermal resistance of 0.25C/W,
> a number you might see with a product like the Zalman 7000A.
> The Intel boxed processor HSF is somewhere in the area of
> 0.42C/W (according to Tomshardware).
>
>
http://www.zalman.co.kr/product/cooler/7000a-478_motherboard_list_eng.html
>
> You could fit one of these, if there is at least a centimeter
> of space between the top edge of the S478 bracket and your
> power supply. In a mid ATX case, there frequently isn't any room
> between the edge of the board and the PSU, but check your case to
> see if there would be enough room to fit it. The 7000a is the
> best compromise solution I can see right now for a S478. It isn't
> excessively loud, and not a lot of air blows through the fins.
> You can feel a warm cloud of air around it while it is running.
> You still need good case fans to move this cloud out of the
> computer case.
>
> HTH,
> Paul