Welcome to PCForumz.com!
FAQFAQ      ProfileProfile    Private MessagesPrivate Messages   Log inLog in

Computer cooling issues

 
Goto page 1, 2, 3
   Hardware Problem Solving Community! (Home) -> Home Built RSS
Next:  opening dropped drive  
Author Message
TomS

External


Since: Dec 07, 2005
Posts: 3



(Msg. 1) Posted: Wed Dec 07, 2005 5:55 pm
Post subject: Computer cooling issues
Archived from groups: alt>comp>hardware>homebuilt (more info?)

Have a P4 2400, 845GRG Intel motherboard with 1GB RAM and 2 x 80GB HDD's.
This machine has been running quite cool a per Intel Active Monitor, around
30-35 deg C. Recently I bought FS2004 and a MSI 6600GT video card. Now the
PC is running at around 55 deg C in the processor zone and 45 deg C in the
system zone when I do flightsim. There are currently 3 fans in the machine,
1 on top of the CPU, 1 on the graphics card and the powersupply fan. There
is still provision to install 2 more 8cm (?) chassis fans, 1 on the front
and 1 on the back of the case, which I think of installing to cool the PC.

First, what is the proper fan arrangement? Should 1 be extracting and 1
blowing air into the case, or should both be extracting (or both blowing air
in)? Bought an extra fan some time ago (12V, 4.3W) which made such a noise
that I uninstalled it again. So preferably I look for something that cools
and is not too noisy as well.

Second, what are proper upper threshold temperature settings for Intel PC's?
Currently the monitor is set for 70 deg C in the processor zone and 50 deg C
in the system zone. Keeping in mind I want the PC to last, also I live in
Australia where it is summer now and temperatures in the house can be over
30 deg C in daytime.

 >> Stay informed about: Computer cooling issues 
Back to top
Login to vote
kony

External


Since: Jan 03, 2004
Posts: 6148



(Msg. 2) Posted: Thu Dec 08, 2005 9:55 am
Post subject: Re: Computer cooling issues [Login to view extended thread Info.]
Archived from groups: per prev. post (more info?)

On Thu, 08 Dec 2005 01:15:43 GMT, "TomS"
<someone RemoveThis @somewhere.biz> wrote:


>Unfortunately can't change the fan grills as they are one piece with the
>chassis, short of disassembling everything and cutting a hole into the case,
>will leave that until I have run out of other options.

It illustrates a point that there are far too many cases on
the market today, let alone past years that are not designed
very well and need modified for best results, prior to any
system building.




>> You might see if it'll run on 5V, there are fan adapters
>> that you can convert to supply 5V, or proper fan controllers
>> to allow broad RPM adjustments between roughly 5V and 11.5V.
>
>Actually the fan I have is of Taiwanese origin with one of those mini 3 pin
>connectors that get the power from the motherboard, not the powersupply.

Doesn't necessarily matter, there are also adapters that can
power those from a power supply plug and also, still be
changed to power from 5V rather than 12V.

>Fan
>label says DC12V - 4.3W. The motherboard has 3 of those power connectors,
>and with the right fan the speed should be controlled by the motherboard (I
>think, have to read the manual once more).

That's a nice feature but it depends upon both the
motherboard having a complimentary fan RPM to temp threshod
adjustment AND the fan itself to be spec'd accordingly...
not seeing your case it's a bit hard to be certain but in
general a 4.3W fan is overkill even for a motherboard header
with this thermal control... optimal would be a fan that
starts at lower default RPM @ 12V, _and_ having that (or
another method of) further RPM reduction control from the
motherboard.


>
>Thanks again - after reading your reply I am less worried. It seems to me
>all I need in the first instance is a 2nd fan to work properly and all
>should be fine.

Probably, the primary issue with the CPU around that level
is still the video or the hard drive(s), since a case rear
exhaust with stamped grill reduces exhaust rate, and if the
PSU also has stamped grill, the total airflow is going to be
a lot lower, or quite a bit noiser per same flow. CPUs
now have thermal throttling or shutdown protection but
discrete parts of a hard drive, video, motherboard will
continue running even when temps get higher, to the
detriment of their lifespan. Even so, given the CPU temp,
the temps are likely not severe as the CPU temp is an
indication of chassis rise over ambient, whether there's
some flow through it too.

 >> Stay informed about: Computer cooling issues 
Back to top
Login to vote
kony

External


Since: Jan 03, 2004
Posts: 6148



(Msg. 3) Posted: Thu Dec 08, 2005 9:55 am
Post subject: Re: Computer cooling issues [Login to view extended thread Info.]
Archived from groups: per prev. post (more info?)

On Thu, 08 Dec 2005 11:17:55 GMT, "SteveH"
<steve.hough.DeleteThis@REMOVEMEblueyonder.co.uk> wrote:


>As kony said, your CPU temps are fine for a P4. Personally, I wouldn't add
>to the noise with more fans unless the heat becomes a problem.
>As you are obviously a fan of Flight Sim (well you did buy it a video card!)
>I would now concentrate on tweaking the Flight Sim settings, of which there
>are many to get optimum performace with maximum frame rates.

One of the issues in modern PCs is that among all parts that
heat a significant amount, the CPU is among those most
resistance to temp-induced lifespan reduction below levels
considered acceptible per lifespan of system.

In other words, better to have a CPU at 60C and hard drive
at 40C, than both at 50C.
 >> Stay informed about: Computer cooling issues 
Back to top
Login to vote
TomS

External


Since: Dec 07, 2005
Posts: 3



(Msg. 4) Posted: Sat Dec 10, 2005 10:55 pm
Post subject: Re: Computer cooling issues [Login to view extended thread Info.]
Archived from groups: per prev. post (more info?)

"kony" <spam.RemoveThis@spam.com> wrote in message
news:vidgp1lmb2uvg5lp9k806jd80ua55387s4@4ax.com...
> On Thu, 08 Dec 2005 11:17:55 GMT, "SteveH"
> <steve.hough.RemoveThis@REMOVEMEblueyonder.co.uk> wrote:
>
>
>>As kony said, your CPU temps are fine for a P4. Personally, I wouldn't add
>>to the noise with more fans unless the heat becomes a problem.
>>As you are obviously a fan of Flight Sim (well you did buy it a video
>>card!)
>>I would now concentrate on tweaking the Flight Sim settings, of which
>>there
>>are many to get optimum performace with maximum frame rates.
>
> One of the issues in modern PCs is that among all parts that
> heat a significant amount, the CPU is among those most
> resistance to temp-induced lifespan reduction below levels
> considered acceptible per lifespan of system.
>
> In other words, better to have a CPU at 60C and hard drive
> at 40C, than both at 50C.

Have to agree with you here, I want to keep the two HDD's cool and
alive. --- Bought 2 fans today, they are 12V 1.5W type with 3-pin
motherboard connectors and installed one of them as a rear extracting fan.
Intel Active Monitor software recognized the fan right away and seems to
control its RPM depending on thermal load. The machine still runs very quiet
and the temperatures dropped both in the system and processor zones, so will
not install a front fan at this stage. So far - so good, but its a cool day
today maybe 20-25C, and I have to wait until it's hot again to see what is
really happening.

Thanks again for all the valuable input.
 >> Stay informed about: Computer cooling issues 
Back to top
Login to vote
Allen Weiner1

External


Since: Aug 05, 2004
Posts: 8



(Msg. 5) Posted: Tue Dec 13, 2005 2:55 pm
Post subject: Re: Computer cooling issues [Login to view extended thread Info.]
Archived from groups: per prev. post (more info?)

"kony" <spam.TakeThisOut@spam.com> wrote in message
news:41dgp1l6f3drbtjj70alk9el1fc2cau5sl@4ax.com...
> On Thu, 08 Dec 2005 01:15:43 GMT, "TomS"
> <someone.TakeThisOut@somewhere.biz> wrote:
>
>
> >Unfortunately can't change the fan grills as they are one piece with the
> >chassis, short of disassembling everything and cutting a hole into the
case,
> >will leave that until I have run out of other options.
>
> It illustrates a point that there are far too many cases on
> the market today, let alone past years that are not designed
> very well and need modified for best results, prior to any
> system building.
>

<snip>

Are there cases on the market that, in your opinion, are well designed (with
respect to air flow) and would not require owner mods in order to provide
good results? If there are, would you please list a few.

If you were searching for such a case, what information source would you
use?

The cases forum on Silent PC Review seems pretty informative (even though
I'm not concerned about system noise.)
 >> Stay informed about: Computer cooling issues 
Back to top
Login to vote
Gringo

External


Since: Dec 10, 2005
Posts: 30



(Msg. 6) Posted: Wed Dec 14, 2005 9:21 am
Post subject: Re: Computer cooling issues [Login to view extended thread Info.]
Imported from groups: per prev. post (more info?)

Back to top
Login to vote
kony

External


Since: Jan 03, 2004
Posts: 6148



(Msg. 7) Posted: Wed Dec 14, 2005 1:55 pm
Post subject: Re: Computer cooling issues [Login to view extended thread Info.]
Archived from groups: per prev. post (more info?)

On Wed, 14 Dec 2005 09:21:56 -0800, Gringo
<spammers.TakeThisOut@piss.off> wrote:

>On Wed, 14 Dec 2005 05:33:29 GMT, kony <spam.TakeThisOut@spam.com> wrote:
>
>>Only
>>when cases have inherant design flaws are side-ducts or
>>side-fans, top fans, necessary except in the worst of
>>ambient room temps or highly overclocked parts.
>
>Sorry, but that is not quite correct,

It is correct. Even one single system running fine like
that would be evidence of it. Now consider that the vast
majority of systems are OEM boxes that do not have these
side ducts or fans.

>if you look up Intel's Thermally
>Advantaged v1.1 design spec it calls for a side-duct for the cpu, a
>side grill area at the bottom of the case and a rear case fan of at
>least 92mm.

Why would I be interested in that? It's intel's attempt to
sway case manufacturers towards accomodating their
lower-cost-performance-by-higher-clock-speed design.
Propaganda designed to shift costs from Intel to others.
Even so, as I mentioned above it is not necessary and proven
so by most systems sold.

>If you have a case that meets this design spec then you
>don't need an open front area or a fan in the front at all. You make
>it sound like a side-duct is a bad thing when in fact it is a good
>thing to have.

no it is definitely a very bad thing to have a side duct and
not one in front, it unquestionably decreases flow through
the hard drive rack, and the lowest front area of the
motherboard which may include southbridge or long cards in
lower slots.

A case can have a side duct and still manage to cool these
areas but it requires further re-engineering efforts, it
cannot just be added without changing the rest of the
chassis and/or fans too. In other words, you'll have to
redesign whole thing when none of it was necessary.

The whole tailor-cooling-to-CPU is largely a false benefit.
The CPU always has the far largest heatsink and itself is
not subject to mid-term heat induced failure at temps
commonly seen below the instability or throttling threshold.
 >> Stay informed about: Computer cooling issues 
Back to top
Login to vote
Gringo

External


Since: Dec 10, 2005
Posts: 30



(Msg. 8) Posted: Thu Dec 15, 2005 12:39 pm
Post subject: Re: Computer cooling issues [Login to view extended thread Info.]
Imported from groups: per prev. post (more info?)

Back to top
Login to vote
kony

External


Since: Jan 03, 2004
Posts: 6148



(Msg. 9) Posted: Thu Dec 15, 2005 5:55 pm
Post subject: Re: Computer cooling issues [Login to view extended thread Info.]
Archived from groups: per prev. post (more info?)

On Thu, 15 Dec 2005 12:39:00 -0800, Gringo
<spammers DeleteThis @piss.off> wrote:

>On Wed, 14 Dec 2005 17:52:17 GMT, kony <spam DeleteThis @spam.com> wrote:
>
>
>>The whole tailor-cooling-to-CPU is largely a false benefit.
>>The CPU always has the far largest heatsink and itself is
>>not subject to mid-term heat induced failure at temps
>>commonly seen below the instability or throttling threshold.
>
>Have you any experience using both case designs? I have.
>Both designs are efficient at cooling, only the Intel TA v1.1 case is
>quieter.

That is a ridiculous judgement. Noise is based on several
variables, the least of which is the one you're considering.
I could easily make either seem louder or quieter than the
other.


>It has small vents in the front too but not wide open vents
>like my Antec case has that made for a noisy computing environment.

You have arbitrarily guessed that this design is the cause
for noise reduction but it isn't. Specific cases must be
weighed on all parameters.


>There is a plce for a low front fan in the TA v1.1 case but because
>the HDD cage sits sideways the fan would'nt be doing much good except
>causing turbulence and noise.

Then how did you plan on cooling the drives?
If it makes you happy to have the CPU a degree or two cooler
for no good reason so you can have hotter running hard
drives- be my guest. Even after I pointed it out you still
miss the important point there- it is not a contest to get
the CPU to the lowest temp possible, rather necessary to
keep the REST of the parts cool too. By altering the
airflow it's not just a free benefit, you change airflow in
other areas too.

What Intel did is product a design that helps their hot
running CPUs stay cooler, their interest is in their bottom
line on their product, not on your other parts or system as
a whole. That doesn't mean that solution can't work, but as
I already wrote that's a lot more to consider and in the end
a reworking of whole case with no need to do so.

In other words, if the only (more than moderatly warm) part
in your system is CPU, it might be optimal, and warranted
design if the CPU is overheating. If it's not overheating
there's no reason to choose it.
 >> Stay informed about: Computer cooling issues 
Back to top
Login to vote
Gringo

External


Since: Dec 10, 2005
Posts: 30



(Msg. 10) Posted: Thu Dec 15, 2005 6:25 pm
Post subject: Re: Computer cooling issues [Login to view extended thread Info.]
Imported from groups: per prev. post (more info?)

Back to top
Login to vote
kony

External


Since: Jan 03, 2004
Posts: 6148



(Msg. 11) Posted: Fri Dec 16, 2005 12:55 am
Post subject: Re: Computer cooling issues [Login to view extended thread Info.]
Archived from groups: per prev. post (more info?)

On Thu, 15 Dec 2005 18:25:17 -0800, Gringo
<spammers DeleteThis @piss.off> wrote:

>On Thu, 15 Dec 2005 22:19:17 GMT, kony <spam DeleteThis @spam.com> wrote:
>
>
>>You have arbitrarily guessed that this design is the cause
>>for noise reduction but it isn't. Specific cases must be
>>weighed on all parameters.
>
>The only thing ridiculous here is your inability to accept the facts.

Actually this is all a rehash of what reviewers found back
when the spec was released.


>All you seem to be concerned about is HDD temps. My HDD temps are the
>same in this TA v1.1 case as they were in the Antec case.

Again, only comparing two specific cases is not evidence of
a design's benefits, only of those two specific cases.

>Even AMD's
>white paper on efficient cooling doesn't recommend a front fan

I never insisted a front fan was necessary.
Rather, it often is NOT necessary unless the design gets
altered to interfere with proper flow- for example, putting
in a side fan or duct.

Can a front fan help some cases? Yes. It helps even more
on the type you prefer than standard ATX.

>so why
>you thinik a case necessarily needs one is beyond me. It's obvious to
>me why the Intel TA v1.1 design is quieter - with no open front the
>sound is baffled from the user's ears. Simple case of deduction,

Again, you have no evidence of this from comparing only two
cases, and further, even those two cases you are considering
will vary in noise depending on specific fans used.

The necessary cooling of a computer system involves
examination of all heat generating parts and tailoring the
airflow to accomplish this. If one has enough drives in
front (or very high RPM, or high ambient temp) it may then
be warranted to have a fan. These is no case design that
allows one to ignore individual parts temperatures when an
individual takes on the role of system designer.

Again I mention obvious evidence you continually ignore-
that there are millions of OEM cases out there that do not
have this side ducts or fans and still run quietly.
 >> Stay informed about: Computer cooling issues 
Back to top
Login to vote
Gringo

External


Since: Dec 10, 2005
Posts: 30



(Msg. 12) Posted: Fri Dec 16, 2005 3:11 pm
Post subject: Re: Computer cooling issues [Login to view extended thread Info.]
Imported from groups: per prev. post (more info?)

Back to top
Login to vote
Gringo

External


Since: Dec 10, 2005
Posts: 30



(Msg. 13) Posted: Fri Dec 16, 2005 10:16 pm
Post subject: Re: Computer cooling issues [Login to view extended thread Info.]
Imported from groups: per prev. post (more info?)

Back to top
Login to vote
kony

External


Since: Jan 03, 2004
Posts: 6148



(Msg. 14) Posted: Sat Dec 17, 2005 3:55 am
Post subject: Re: Computer cooling issues [Login to view extended thread Info.]
Archived from groups: per prev. post (more info?)

On Fri, 16 Dec 2005 22:16:32 -0800, Gringo
<spammers DeleteThis @piss.off> wrote:

>On Sat, 17 Dec 2005 00:44:19 GMT, kony <spam DeleteThis @spam.com> wrote:
>
>
>>You apparently own TWO WHOLE CASES.
>>
>>WOW.
>
>Three, actually, one is not being used currently.
>
>>Wanna guess how many I have in my basement alone just
>>waiting to have parts put into 'em?
>
>Good for you, but I bet you don't have one that meets the Intel TA
>v1.1 spec so you still have no experience to go by.

You mean with a side-duct? Yes, is that supposed to be some
kind of rare thing?

I"ve modded cases and added side-duct as well as fan(s)
too... so what? It's laughable that with 3 cases you feel
you have a comprehensive understanding of the differences in
noise between different case design specifications.


>>If you want to find an alternate way to cool the lower
>>quadrant of a case, fine, choose one that has a lower side
>>fan and a rotated rack. Either way, one cannot just
>>"decide" to use a case without an open front to reduce
>>noise, unless they ALSO find an alternative cooling to make
>>up for this flow change. Such is the choice facing any
>>design/designer, what tradeoffs to make. The tradeoff Intel
>>made favors the CPU, which is more important if the CPU
>>overheats but nothing else runs hot.
>
>I also said there is a lower square grill and not just a side duct.

Nevertheless, the side duct reduces flow through the bottom
half and drive rack. Result is either hotter running parts
or addition of a fan to offset this flow reduction, which is
not complimentary to lowering noise, unless it also means
you manually reduce the rear exhaust fan speeds.


>
>
>>I further suggest that you have very little experience with
>>fans if you feel either design necessarily makes much
>>difference contrasted with fan selection.
>
>
>Suggest what you will. You don't know jack about what experience I
>have with fans. I'll tell you this though, less is more.
><snippity snip>
>You're turning this into a crusade now. Adios.

I do know you have insufficient experience with fans based
on your statements. If only you knew a little more about
fans, you'd then know why the statements were revealing- oh
the irony of it all.

Anyway, no, less isn't more. The quietest configuration is
running fans at the lowest RPM possible before they get into
a pre-stall threshold where there is a pulsing sound as each
coil energization causes a sudden current increase and
torque. So it's not a matter of less # of fans, it's a
matter how many fans are needed with each at this lowest
noise RPM. It can mean only one fan, in a fairly low end,
low heat system. It always means more than one on any
well-endowed system, often 3 or more. However, most systems
are not so endowed, will be fine with one fairly sizeable
rear chassis fan and the one in the PSU. If you add a
side-duct to the case, you may easily need add another fan
just to counter-act this reduction in flow to other areas.

Is it a crusade? YES! I modify cases regularly, and right
this minute have one about 18" away that has been modified,
and three more behind it, and (hmm, I can't see behind those
3 at the moment!). In fact I"ve probably modified 10X more
cases than you've ever owned. I pay careful attention to
noise, and airflow, as these are the whole point of
modifications... only someone inexperienced, judges noise
based on an arbitrary idea about # of fans, not considering
the RPM or additive effects of multiple fans being lower
noise than fewer fans moving same amount of air.

I do it because doing same thing different day is boring,
and because I like to overclock too (non-critical systems)
but not at the expense of noise.

Again I suggest you rethink your whole argument, you do not
have evidence with only two particular cases. Each case,
and each fan selection choice makes a difference regardless
of which spec is used.

Even then, it's incredible that you don't see why the
side-duct is working against low noise. Which SPECIFIC area
of system cooling do you attribute to the noise reduction?
Since the front intake is not excluded in the Thermally
Advantaged spec, we can ignore it completely and focus on
this side-duct. Do you think the CPU fan will run quieter?

Can you hear your CPU fan in either system? I can't in any
of the ATX cases you claim would be louder. So with CPU fan
out of the way, what is the noise source that's going to
make regular ATX louder because it lacks the side-duct?

It's possible the PSU exhaust will be a little quieter with
a side-duct, but only if the case has too much restriction
to intake in other areas. With any case, either design, one
should avoid cases with poor intake so we can rule this out
as well, it's not a constant but varies per case.

Yet again I"ll mention it just for the heck of it (since you
seem to have totally ignored it the last couple times)-
OEMs build quiet systems without side-ducts. The vast
majority of systems do not have them. This case you have
that is ATX, not Thermally Advantaged, is almost certainly
not as quiet as a semi-standard ATX make by (for example)
Dell, HP or Gateway. This in itself is yet again evidence
that you'e basing your conclusion on the wrong variable, it
is not whether the case is "Thermally Advantaged" or not.

In summary, you would do well to start modifying cases if
you really have an interest in case layout and cooling
designs. Only after you have a case and measure temps,
THEN modify it and compare, will you begin to see the
significance of any alterations. Then get some fans, some
quiet ones since you apparently do not have quiet fans, else
you would not be reporting "noisier" in either config, you'd
be reporting "hotter". This alone shows you are comparing
individual fans and fan mounts more than cases.
 >> Stay informed about: Computer cooling issues 
Back to top
Login to vote
Gringo

External


Since: Dec 10, 2005
Posts: 30



(Msg. 15) Posted: Sat Dec 17, 2005 3:55 am
Post subject: Re: Computer cooling issues [Login to view extended thread Info.]
Imported from groups: per prev. post (more info?)

Back to top
Login to vote
Display posts from previous:   
Related Topics:
Cooling issues - getting a new system 2600+ sempron 512 DDR400 9600 pro 128mb gigabyte mobo 20GB 5200 15GB 5200 dvd cdrw raidmax case I know my systems not the hottest thing hot there, but i like to be over cautious. My cpu temp on my old system(1ghz celeron, 256mb)is....

Help with Memory issues - Folks, I seem to have developed a problem with the system's memory and was hoping someone may be able to offer some advice. The system was: Abit AT7-Max-2 Athlon XP2000 1Gb of Ram consisting of 1x 256Mb Kingtson ValueRam (266), 1x256Mb Aria Generic..

Brand New Dell 8400 with issues - I just purchased a DEll 8400 with a Radeon X300 SE video card. I plugged in my two year old Trinitron Monitor which was working great up until that moment and it turned green. All attempts to reset or calibrate have failed. If I calibrate it fades in..

More cooling woes... - Folks, Some of you may remember my posting a few months ago regarding cooling my machine... I have an XP1900 sat under a Zalman Flower (all Copper, not Copper / Alum), plus two rear exhaust 80mm fans, 1 side intake 80mm, and 2 front intake 80mm fans..

Cooling And Power - Ok, so I have an AMS G-mono http://www.amselectronics.com/Products/PC_Servers/CF-2029.html case, that supports three extra fans. I have two 80mm in the back, and an 80mm (in the 120mm or 80mm mount) in the front. If you look at the bottom of the front of...
   Hardware Problem Solving Community! (Home) -> Home Built All times are: Pacific Time (US & Canada) (change)
Goto page 1, 2, 3
Page 1 of 3

 
You can post new topics in this forum
You can reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum



[ Contact us | Terms of Service/Privacy Policy ]