air flow

[Robert]

It's 0300 hrs and I'm sitting with my computer again. So I don't have a
life...

I was thinking about adding another exhaust fan to my computer case. I have
(of course) the one in the pwr supply, and one case fan at the lower rear.
If I add another fan to the lower rear, could I set up a condition, in which
so much air is being drawn across the lower half of my case, thus creating a
negative air pressure in the upper half of the case(were the pwr-supply
lives), and since there is no air to drawn through the pwr-supply, and out
of the case. Will this cause my pwr-supply to over heat, and fail
pre-maturely?

This is kind of long winded yes, but I didn't want to design in a possible
failure.

The more I thought about the more it bugged me...
R. Brooks ASEET A+ MCP 136/140 BSCS

 

[Guest]

It's 0300 hrs and I'm sitting with my computer again. So I don't have a
life...

I was thinking about adding another exhaust fan to my computer case. I have
(of course) the one in the pwr supply, and one case fan at the lower rear.
If I add another fan to the lower rear, could I set up a condition, in which
so much air is being drawn across the lower half of my case, thus creating a
negative air pressure in the upper half of the case(were the pwr-supply
lives), and since there is no air to drawn through the pwr-supply, and out
of the case. Will this cause my pwr-supply to over heat, and fail
pre-maturely?

This is kind of long winded yes, but I didn't want to design in a possible
failure.

The more I thought about the more it bugged me...

Slight or moderate negative pressure is good in a computer
case and should not cause overheating problems in a PSU.

 

[John Doe]

It's 0300 hrs and I'm sitting with my computer again. So I don't
have a life...

I was thinking about adding another exhaust fan to my computer case.
I have (of course) the one in the pwr supply, and one case fan at
the lower rear. If I add another fan to the lower rear, could I set
up a condition, in which so much air is being drawn across the lower
half of my case, thus creating a negative air pressure in the upper
half of the case(were the pwr-supply lives), and since there is no
air to drawn through the pwr-supply, and out of the case. Will this
cause my pwr-supply to over heat, and fail pre-maturely?

You definitely want air flowing through your power supply. Some power
supplies probably will compensate by heating up and automatically
increasing fan speed.

In addition to avoiding air flowing in through your CD drive and stuff
(some people think that's OK), putting an airflow strain on the
power supply sounds like another reason to avoid having negative air
flow pressure.

Good luck.

 

[John Doe]

....

Slight or moderate negative pressure is good in a computer case

Please explain.

and should not cause overheating problems in a PSU.

It definitely shouldn't, because that would be a problem.

 

[Guest]

...


Please explain.

It's basic thermal dynamics. Air follows the path of least
resistance. In a negative pressure flow, heat generated by
components inside a computer case is exhausted efficiently.
In a positive pressure flow, heat stays inside the case (at
least for a while), and tends to build up.

Positive flow does have a few benefits (e.g. less accumulation
of dust inside a case), but the OP is concerned with temps.

 

[John Doe]

It's basic thermal dynamics. Air follows the path of least
resistance. In a negative pressure flow, heat generated by
components inside a computer case is exhausted efficiently.
In a positive pressure flow, heat stays inside the case (at
least for a while), and tends to build up.

Seems to me that basic thermal dynamics is more concerned with
quantity of air flow than it is with negative or positive pressure.

 

[Guest]

Seems to me that basic thermal dynamics is more concerned with
quantity of air flow than it is with negative or positive pressure.

Nope. Quantity isn't the issue. If you're blowing the same
hot air around the inside of a case, that's not good --
regardless of how fast the air is moving.

 

[Skybuck Flying]

I have a nice airflow story:

My case looks like this from the top:

two gfx cards can blow little bit of air out.
/|\ vertical rear fan blowing air out
| /|\
| | power air exit
| | /|\
| | |
+-----------+
| power | <- horizontal power fan blowing air up, half to top.
| sndblast|
Ogfx cards| <- graphics cards can suck air in left/bottom.
O <--cpu | <- cpu fan blows directly out on left/top.
| memory |
| space |
|<--hd's-->|
O<-holes->O<--- two side fans, bottom, top.
+------------+
/|\ /|\ air gates and filter in door as well.
|
|
I patched/closed these two holes on the left side near the HD.

Because the air was being blown in from the right and across the hd's and
immediatly out left.

I believe that was bad.

The cool air should then go to the back of the case.

So first I tried plastic stickers on the left side of the case.

But they would always get loose because of the air blowing them loose.

So finally I got smart... and I placed the stickers on the inside of the
case...

They weren't as sticky anymore... but they are still in place !

The air is simply helping keeping them in place since the air now blows
against them !

Ofcourse the stickyness helps too.

And it looks much better now too.. I can't even see the stickers anymore...
they see through plastic... only if I look real carefully do I see a bit of
shine from light

I am happy with this solution.

Now the cool air goes from the front to the back of the case as it should.

The motherboard, cpu and graphics cards are now a few degrees cooler.

I also underclocked and undervoltaged my CPU that reduces about 20 degrees
of celcius temperature.

I also tried patching the other holes in the back near gpu/gfx card's... the
results were inconclusive... could help or could make it worse.

However patching the cpu hole would be bad.

It's good if the CPU can blow out it's hot air immediatly it seems

Bye,
Skybuck.

 

[SteveH]

So finally I got smart...

Could never happen..

 

[Skybuck Flying]

Could never happen..

Like you'd know.

Bye,
Skybuck.

 

[DaveW]

Adding another rear exhaust fan will work just fine. Your fears are
unfounded.

 

[peter]

Could you post the make and model of your Case??? as well as theType of
Heatsink on your CPU
It is a bit unusual to have a low rear exhaust fan as hot air rises
the CPU is usually up high and if the Fan from the PSU is directly above the
CPU that fan is sucking the heat away from the CPU and its heatsink.The fear
is that by adding a 2nd low rear exhaust fan will it suck that hot air down
and heat up other components..like the NB..the Video Card...resulting in a
higher case temp. If for example that PSU fan is sucking 45CFM and your
present rear fan is sucking 45cfm...that pretty well creates the situation
where each fan draws air out.............now you add a 2nd 45CFM fan and all
of a sudden you're over powering the PSU exhaust and hot air from the CPU
heatsink is now being pulled down across the other parts in the machine.
peter

 

[John Doe]

Nope. Quantity isn't the issue.

You have to be kidding.

If you're blowing the same hot air around the inside of a case,
that's not good -- regardless of how fast the air is moving.

By "quantity of airflow", I mean the amount of air flowing through the
case, not just airflow within the case.

Path: newssvr12.news.prodigy.net!newsdbm05.news.prodigy.net! newsdst01.news.prodigy.net!prodigy.com!postmaster.news.prodigy.com!
newssvr19.news.prodigy.net.POSTED!af76947f!not-for-mail
From: <[email protected]>
Newsgroups: alt.comp.hardware.pc-homebuilt
References: <[email protected]> <Yqlli.8280

[email protected]> <Jimli.2505$eY.810
@newssvr13.news.prodigy.net> <PNmli.8284$bz7.29
@newssvr22.news.prodigy.net> <Wknli.9877$Rw1.7159
@newssvr25.news.prodigy.net>

Subject: Re: air flow
Lines: 27
X-Priority: 3
X-MSMail-Priority: Normal
X-Newsreader: Microsoft Outlook Express 6.00.2800.1807
X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2800.1896
Message-ID: <[email protected]>
NNTP-Posting-Host: 75.15.156.195
X-Complaints-To: (e-mail address removed)
X-Trace: newssvr19.news.prodigy.net 1184241109 ST000 75.15.156.195

(Thu, 12 Jul 2007 07:51:49 EDT)

NNTP-Posting-Date: Thu, 12 Jul 2007 07:51:49 EDT
Organization: AT&T http://yahoo.sbc.com
X-UserInfo1: Q[RGW_SGQBWKRT@Y[JHD]_\@VR]^@B@MCPWZKB]MPXHTEPIB_NVUAH_[BL [\IRKIANGGJBFNJF_DOLSCENSY^U@FRFUEXR@KFXYDBPWBCDQJA@X_DCBHXR[C@
\EOKCJLED_SZ@RMWYXYWE_P@\\GOIW^@SYFFSWHFIXMADO@^[ADPRPETLBJ]RDGENSKQQZN
Date: Thu, 12 Jul 2007 04:52:21 -0700
Xref: prodigy.net alt.comp.hardware.pc-homebuilt:493173

 

[jameshanley39]

You have to be kidding.


By "quantity of airflow", I mean the amount of air flowing through the
case, not just airflow within the case.

He's covering the possibility that you're living in an oven. It was a
theoretical question about 'basic thermaldynamics' !!! ;-)

He makes a good point to add to yours though.
Not just the factor of Quantity of air.
But, the factor that outside air should be cool.

If your computer is in a poorly ventilated cupboard.. his example
would be relevant. Or perhaps, in a really hot country, with no air
conditioning in the room, the computers would suffer for it in a
similar way(hot air in and out, though not the same hot air!) . It's
not that outlandish or theoretical.

And I don't think he misunderstood you, as you suggest.

 

[jameshanley39]

I have a nice airflow story:

My case looks like this from the top:

two gfx cards can blow little bit of air out.
/|\ vertical rear fan blowing air out
| /|\
| | power air exit
| | /|\
| | |
+-----------+
| power | <- horizontal power fan blowing air up, half to top.
| sndblast|
Ogfx cards| <- graphics cards can suck air in left/bottom.
O <--cpu | <- cpu fan blows directly out on left/top.
| memory |
| space |
|<--hd's-->|
O<-holes->O<--- two side fans, bottom, top.
+------------+
/|\ /|\ air gates and filter in door as well.
|
|
I patched/closed these two holes on the left side near the HD.

Because the air was being blown in from the right and across the hd's and
immediatly out left.

I believe that was bad.

The cool air should then go to the back of the case.

So first I tried plastic stickers on the left side of the case.

But they would always get loose because of the air blowing them loose.

So finally I got smart... and I placed the stickers on the inside of the
case...

They weren't as sticky anymore... but they are still in place !

The air is simply helping keeping them in place since the air now blows
against them !

Ofcourse the stickyness helps too.

And it looks much better now too.. I can't even see the stickers anymore...
they see through plastic... only if I look real carefully do I see a bit of
shine from light

I am happy with this solution.

Now the cool air goes from the front to the back of the case as it should.

The motherboard, cpu and graphics cards are now a few degrees cooler.

I also underclocked and undervoltaged my CPU that reduces about 20 degrees
of celcius temperature.

I also tried patching the other holes in the back near gpu/gfx card's... the
results were inconclusive... could help or could make it worse.

However patching the cpu hole would be bad.

It's good if the CPU can blow out it's hot air immediatly it seems

Bye,
Skybuck.

I don't get the picture.. Or all of what you said, but you make some
interesting points.. i'm suprised underclocking made such a
difference. I'm sure i've done it and not found such a difference.
What processor, what speeds?

Regarding the CPU heatsink. It doesn't really blow its hot air out.
The fan sucks cool air in, and that air gets sent through the fins of
the heatsink. Test that it's an intake fan, by holding a tissue over
the fan.

I noticed that in a scott mueller video i found online or in his UGRP
book. Coincidentally, he also mentioned a spec designed at the time of
the Prescott(Hot processor), that utilised more air being blown out
than in. It had holes on the case which sucked cool air in , due to
the negative air pressure. The spec was meant to take over, he took
it seriously. The case had a funnel like thing on the side of the case
which went to the CPU for cooling it.
Another weird design of case i've seen pictures of, is with the PSU
on the bottom. That may not have been a standard though, could've just
been one adventurous designer.

 

[John Doe]

He makes a good point to add to yours though.
Not just the factor of Quantity of air.
But, the factor that outside air should be cool.

That's not what he was talking about.

If your computer is in a poorly ventilated cupboard.. his example
would be relevant. Or perhaps, in a really hot country, with no air
conditioning in the room, the computers would suffer for it in a
similar way(hot air in and out, though not the same hot air!) .
It's not that outlandish or theoretical.

And I don't think he misunderstood you, as you suggest.

He was talking about air circulating inside of the case, as if that is
what positive air pressure causes. Even if that were true, it would be
true only for the time it takes for the pressure to form, probably
instantaneously.

You probably do not want negative air pressure inside of your case.
The point about straining your power supply outflow fan, tending to
oppose the flow of air through your power supply, is a clincher IMO.
That combined with no benefit of negative air pressure makes the
endeavor pointless at best.

 

[Skybuck Flying]

I don't get the picture.. Or all of what you said, but you make some
interesting points.. i'm suprised underclocking made such a
difference. I'm sure i've done it and not found such a difference.
What processor, what speeds?

Undervoltage from 1.4 to 1.1 volts or something like that.
X2 3800+

Top=bottom = 3D in picture.

Take a look at pictures of the case then ll you understand

Regarding the CPU heatsink. It doesn't really blow its hot air out.
The fan sucks cool air in, and that air gets sent through the fins of
the heatsink. Test that it's an intake fan, by holding a tissue over
the fan.

Well to me it seemed as if hot air was been blowing out.

Maybe sucking cold air in then..

I noticed that in a scott mueller video i found online or in his UGRP
book. Coincidentally, he also mentioned a spec designed at the time of
the Prescott(Hot processor), that utilised more air being blown out
than in. It had holes on the case which sucked cool air in , due to
the negative air pressure. The spec was meant to take over, he took
it seriously. The case had a funnel like thing on the side of the case
which went to the CPU for cooling it.
Another weird design of case i've seen pictures of, is with the PSU
on the bottom. That may not have been a standard though, could've just
been one adventurous designer.

I have seen HP computers with special cpu ducts... which guide air to/from
cpu.

Bye,
Skybuck.

 

[JAD]

I don't get the picture.. Or all of what you said, but you make some
interesting points.. i'm suprised underclocking made such a
difference. I'm sure i've done it and not found such a difference.
What processor, what speeds?

your kidding right? laptops do this very same thing via a single push of a button.(for
awhile now)

 

[Crackles McFarly]

It's 0300 hrs and I'm sitting with my computer again. So I don't have a
life...

I was thinking about adding another exhaust fan to my computer case. I have
(of course) the one in the pwr supply, and one case fan at the lower rear.
If I add another fan to the lower rear, could I set up a condition, in which
so much air is being drawn across the lower half of my case, thus creating a
negative air pressure in the upper half of the case(were the pwr-supply
lives), and since there is no air to drawn through the pwr-supply, and out
of the case. Will this cause my pwr-supply to over heat, and fail
pre-maturely?

i dont want to anger anyone but i had a super overheated system and
gave up on tinkering with # of fans, placement of fans, balance of
fans, rpm's of fans, etc etc etc. My solution was to simply open the
whole side of the case. i figured it wouldn't help but why not try it?

it worked well [for me]
cpu dropped 2C
motherboard dropped 5C
hard drive only 1C but with $5 drive fan it dropped 4C with case off
and only 2C with it on.

It did increase the noise by about 25% which in my setup is VERY
noticable however the temp drops overall seem to have worked for me,
yes me. I am NOT suggesting someone try this but just mentioning how
it worked for [me, just me]

one more thing, dust! The amount of and type of dust increased 300% or
more! it's insane to clean it and then it be as dirty again in less
than 72 hrs. With the case on I could easily go 30-45 days without
this level of dust building up.

Also:I think it's kinda neat to not only SEE the components at all
times but get that heat loss benefit with it.


Again:This is just something that worked for ME and am not suggesting
you try it for more than 1 hour to see if you like it.

my 33 cents [inflation]

yours truly,
crackles

 

[Frank McCoy]

In alt.comp.hardware.pc-homebuilt Crackles McFarly

It's 0300 hrs and I'm sitting with my computer again. So I don't have a
life...

I was thinking about adding another exhaust fan to my computer case. I have
(of course) the one in the pwr supply, and one case fan at the lower rear.
If I add another fan to the lower rear, could I set up a condition, in which
so much air is being drawn across the lower half of my case, thus creating a
negative air pressure in the upper half of the case(were the pwr-supply
lives), and since there is no air to drawn through the pwr-supply, and out
of the case. Will this cause my pwr-supply to over heat, and fail
pre-maturely?

i dont want to anger anyone but i had a super overheated system and
gave up on tinkering with # of fans, placement of fans, balance of
fans, rpm's of fans, etc etc etc. My solution was to simply open the
whole side of the case. i figured it wouldn't help but why not try it?

it worked well [for me]
cpu dropped 2C
motherboard dropped 5C
hard drive only 1C but with $5 drive fan it dropped 4C with case off
and only 2C with it on.

It did increase the noise by about 25% which in my setup is VERY
noticable however the temp drops overall seem to have worked for me,
yes me. I am NOT suggesting someone try this but just mentioning how
it worked for [me, just me]

one more thing, dust! The amount of and type of dust increased 300% or
more! it's insane to clean it and then it be as dirty again in less
than 72 hrs. With the case on I could easily go 30-45 days without
this level of dust building up.

Also:I think it's kinda neat to not only SEE the components at all
times but get that heat loss benefit with it.


Again:This is just something that worked for ME and am not suggesting
you try it for more than 1 hour to see if you like it.

my 33 cents [inflation]

yours truly,
crackles

*I* had that problem.
Only I solved it with a case-change.

Something like the case in this:
http://www.windowsdevcenter.com/pub/a/windows/2005/02/23/Build_Perfect_BudgetPC.html
Only with a blue squared-off duct to the CPU-cooler instead of the
rounded one shown in the picture of the side-panel.

It makes an *enormous* difference to have the CPU-cooler sucking in cool
outside air directly from the side, while having an exhaust fan nearby
as well. The large main chassis-fan in front helps too.

The only thing is: You have to remember to leave room for air to get in
the side, and wipe the dust and cat-hair off the vent in the side-panel
every month or two. (Actually, much easier than getting into the case
and cleaning out the CPU-cooler about as often, which seems to not be
needed nearly so often with this case; the small grid of the air-entry
stopping most such crud.)