cooling computer with refrigerator

[Ken Maltby]

gutless, even someone as stupid as you should be able to grasp that
when a particular chunk of air is cooled, the humidity will increase until
eventually its 100% humid and will start condensing at the dew point.


Wrong with the HUMIDITY.


What was being discussed is the HUMIDITY.


Wrong again, its only lost SOME of the water, because
the cold side can be below the dew point of the air.


Nope, that wasnt what was being discussed.

Humidity is the concentration of water vapor in the air.

The air being talked about loses a great deal ( in some cases more than
95%) of its water vapor, through its contact with the cold surfaces of the
air conditioner. (This is not by accident, one of the factors that air
conditioning addresses, in its attempt to create a more comfortable
environment, is to lower the humidity along with the temp.) This is
designed into the air conditioner. If you take the water out of the air,
as you cool it, then the cold air is dry. If it could have given up any
more water vapor to the still very cold surfaces of the air conditioner, it
would have.

Cold Dry air being pushed from an air conditioner by a fan will displace
a considerable amount of the warm humid air, and reduce the humidity (by
taking up some of the water vapor) of the air it hasn't completely
displaced.
If this were to take place into a leaky box, there would quickly be only
cold
DRY air within the box. No matter how cold the cold dry air can make any
surface in the leaky box, that surface won't be able to pull any water out
of the
dry air.

Luck;
Ken

 

[Ken Maltby]

Which "he" are you refering to?
If you had a duct straight from a compressor type air-conditioner leading
straight into the case of your computer, the cold dry air from it would
both cool and dry the inside of your computer and the cold computer case
would have some moisture condensing on the outside of it from the
surrounding (slightly warmer) air in the room.
I have tried this and that is exactly what happens.

As I posted earlier, it can work very well. It helps if you can use the
extra cooling, like here in central Texas. The system I setup for the kid's
multiplayer games, uses the $85 5000BTU window air conditioner and
he isn't monitoring the setup to any extent. Its been working that way for
over a year now.

Luck;
Ken

 

[Rod Speed]

Humidity is the concentration of water vapor in the air.

You quite sure you aint one of those rocket scientist fellas ?

The air being talked about loses a great deal ( in some cases more than 95%) of its
water vapor, through its contact with the cold surfaces of the air conditioner.

Not unless those are below the DEW POINT, it doesnt.

(This is not by accident, one of the factors that air conditioning addresses, in its
attempt to create a more comfortable environment, is to lower the humidity along with
the temp.) This is designed into the air conditioner.

Its more complicated than that when its heating.

If you take the water out of the air, as you cool it, then the cold air is dry.

Depends entirely on how much of the water is removed.

Most dont in fact find that very dry air is very
comfortable, which is why humidifiers are used.

If it could have given up any more water vapor to the still very cold surfaces of the
air conditioner, it would have.

Wrong again, see above.

Cold Dry air being pushed from an air conditioner by a fan will displace a considerable
amount of the warm humid air,

You aint established that the warm air is necessarily humid.

and reduce the humidity (by taking up some of the water vapor) of the air it hasn't
completely displaced.

If this were to take place into a leaky box, there would quickly be only cold DRY air
within the box.

Not necessarily, see above.

No matter how cold the cold dry air can make any surface in the leaky box, that surface
won't be able to pull any water out of the dry air.

Have fun explaining how come you still get condensation
on cold glasses even in an air conditioned room.

 

[Rod Speed]

Humidity is the concentration of water vapor in the air.

You quite sure you aint one of those rocket scientist fellas ?

The air being talked about loses a great deal ( in some cases more than 95%) of its
water vapor, through its contact with the cold surfaces of the air conditioner.

Not unless those are below the DEW POINT, it doesnt.

(This is not by accident, one of the factors that air conditioning addresses, in its
attempt to create a more comfortable environment, is to lower the humidity along with
the temp.) This is designed into the air conditioner.

Wrong again, most obviously with split systems.

If you take the water out of the air, as you cool it, then the cold air is dry.

Depends entirely on how much of the water is removed.

Most dont in fact find that very dry air is very
comfortable, which is why humidifiers are used.

If it could have given up any more water vapor to the still very cold surfaces of the
air conditioner, it would have.

Wrong again, see above.

Cold Dry air being pushed from an air conditioner by a fan will displace a considerable
amount of the warm humid air,

You aint established that the warm air is necessarily humid.

and reduce the humidity (by taking up some of the water vapor) of the air it hasn't
completely displaced.

If this were to take place into a leaky box, there would quickly be only cold DRY air
within the box.

Not necessarily, see above.

No matter how cold the cold dry air can make any surface in the leaky box, that surface
won't be able to pull any water out of the dry air.

Have fun explaining how come you still get condensation
on cold glasses even in an air conditioned room.

 

[Ken Maltby]

Changing the temp. of a gas (air included) doesn't change the amount
of water vapor it contains, until the dew point is reached and the water
vapor can condense (and then only if there is some dust or something
to condense around). You could also say that a gas can't take up any
more water vapor unless the temp. is above the dew point. You can't
increase humidity by lowering the temp., only by adding water vapor,
which usually means raising the temp. (Ever wonder why you get
thunderstorms after a hot day?)

You quite sure you aint one of those rocket scientist fellas ?


Not unless those are below the DEW POINT, it doesnt.

See my paragraph below. Air conditioners are designed to have surfaces
that
are below the dew point of the warm air they are conditioning.

Its more complicated than that when its heating.

What's that supposed to mean? All my air conditioners are dedicated to
cooling. None of them do any heating. (Of the conditioned air that is.)

Depends entirely on how much of the water is removed.

As I said, almost all of it, more than 95%.

Most dont in fact find that very dry air is very
comfortable, which is why humidifiers are used.

Which is why the air in a room is not normally made up of only
the direct output of an air conditioner, but a mixture of conditioned
air and the ambient air. Also one of the reasons they operate in
cycles.

Wrong again, see above.

See what above? Where is there anything above that would explain how
air
with water vapor could make contact with the cold surfaces of an air
conditioner
and NOT give up its water vapor. Don't tell me you think the cold coils of
an air
conditioner would be above the dew point of the air?

You aint established that the warm air is necessarily humid.

If it is not humid then it is dry and not able to condense on any cold
surface, that may be created within the computer. But most air that is
perceived as warm, is well above the dew point, and will take up water
vapor, if any is available.

Not necessarily, see above.

Again, what above? Where do you say anything that refutes what I said in
that sentence?

Have fun explaining how come you still get condensation
on cold glasses even in an air conditioned room.

I have taken a perfectly cold, and perfectly dry, can of soda from
a computer room into a less conditioned office and seen it get wet.

The conditioning for a room is moderated to match the needs of people,
providing a lower humidity but not so low as to dry out peoples sinus.
( Check with anyone responsible for operation an HVAC system, one
of the most common required adjustments is to fix overly dry air, causing
health issues.) What we were discussing here was dumping the output of
a window air conditioner, directly into a computer without mixing it with a
whole room full of moister air.

/Ken

 

[Rod Speed]

Changing the temp. of a gas (air included) doesn't change the amount of water vapor it
contains,

Yes, but it will clearly change the HUMIDITY that is being discussed.

until the dew point is reached and the water vapor can condense (and then only if there
is some dust or something to condense around).

That last is just plain wrong. There is
always something for it to condense on.

You could also say that a gas can't take up any more water vapor unless the temp. is
above the dew point.

Irrelevant to the HUMIDITY which was being discussed.

You can't increase humidity by lowering the temp.,

Wrong. Thats what happens to the relative humidity.

only by adding water vapor,

Wrong again with the relative humidity.

which usually means raising the temp.

Have fun explaining the condensation on cold objects.

(Ever wonder why you get thunderstorms after a hot day?)

Utterly mangled all over again.

See my paragraph below.

Completely useless, you mangled the story completely, most
obviously because air conditioners DONT attempt to produce
the lowest humidity possible, because that doesnt produce
the best comfort. In spades with split systems.

Air conditioners are designed to have surfaces that are below the dew point of the warm
air they are conditioning.

Some are, some aint. Most obviously with split systems.

What's that supposed to mean?

That they clearly dont attempt to reduced the humidity when they are heating.

All my air conditioners are dedicated to cooling. None of them do any heating.

Irrelevant to what reverse cycle air conditioners do.

(Of the conditioned air that is.)

As I said, almost all of it, more than 95%.

You can keep repeating that claim till the cows come home
if you like, its just plain wrong. Essentially because the lowest
humidity doesnt necessarily produce the best comfort.

Which is why the air in a room is not normally made up of only the direct output of an
air conditioner, but a mixture of conditioned air and the ambient air.

Wrong again with air conditioners used for heating in winter.

Also one of the reasons they operate in cycles.

Wrong again, thats for the set point.

See what above?

The silly claim that air conditioners even
attempt to produce the most dry air possible.

Where is there anything above that would explain how air with water vapor could make
contact with the cold surfaces of an air conditioner and NOT give up its water vapor.

Most obviously when the cold side isnt below the dew point.

Don't tell me you think the cold coils of an air
conditioner would be above the dew point of the air?

Corse they can be with quite dry warm air.

If it is not humid then it is dry and not able to condense on any cold surface, that may
be created within the computer.

Its no news that you dont always get condensation.

But most air that is perceived as warm, is well above the dew point, and will take up
water vapor, if any is available.

Irrelevant to whether it will necessarily
be below the dew point after its cooled.

Again, what above?

Again, they dont necessarily produce DRY air, because
that doesnt necessarily produce the best comfort.

Where do you say anything that refutes what I said in that sentence?

Where I said that DRY air doesnt necessarily produce the best comfort.

Most obviously with split systems where there isnt any source
of external air available at the inside part to mix with the cooled air.

I have taken a perfectly cold, and perfectly dry, can of soda from
a computer room into a less conditioned office and seen it get wet.

All that shows is that the humidity levels are different.

The conditioning for a room is moderated to match the needs of people, providing a lower
humidity but not so low as to dry out peoples sinus.

So your more than 95% claim is clearly just plain wrong.

( Check with anyone responsible for operation an HVAC system,

Dont need to, I know that your more than 95% claim is just plain wrong.

one of the most common required adjustments is to fix overly dry air, causing health
issues.)

So your more than 95% claim is just plain wrong.

What we were discussing here was dumping the output of a window air conditioner,

Wrong again, it wasnt restricted to WINDOW air conditioners.

A split system would be a hell of a lot better way to do it.

directly into a computer without mixing it with a whole room full of moister air.

And you mangled the story completely with your more than 95% claim.

In spades with split systems which dont even have
external air that can be mixed with the cooled air.

 

[dj_nme]

As I posted earlier, it can work very well. It helps if you can use the
extra cooling, like here in central Texas. The system I setup for the kid's
multiplayer games, uses the $85 5000BTU window air conditioner and
he isn't monitoring the setup to any extent. Its been working that way for
over a year now.

Luck;
Ken

Fair enough, central Texas in summer must be a lot less humid that
Sydney (Australia) in summer.
There is enough water vapour in the air (AKA humidity, for Mr Speed)
during Summer here that there was water beading on the surface of my old
computer's case.
I found that it didn't improved the operating of the computer to any
noticable extent, while adding to the corrosion of the screws and other
exposed metal on the outside of the case.
Perhaps it never got hot enough here to warrant such a drastic cooling
solution, forced ventilation by ordinary 12V 80mm case-fans seems
perfectly adequate.

 

[GEO Me]

Try listing the populated bits of Canada that are 'terribly dry'

So now is only selected parts of Canada??


Geo

 

[subc]

wrote:

So now is only selected parts of Canada??


Geo

Geo, he is a troll.

 

[Rod Speed]

So now is only selected parts of Canada??

The populated bits are the only relevant ones given
that what was being discussed was PC cooling.

 

[Ken Maltby]

wrote:


Geo, he is a troll.

I am coming to that conclusion myself, he seems to just want to
argue totally ridiculous and unsupported ideas. I guess I should
have read more of his posts before I replied to his drivel. Well,
we live and learn.

Luck;
Ken

 

[Rod Speed]

I am coming to that conclusion myself, he seems to just want to argue totally ridiculous
and unsupported ideas.

Or you hate having your nose rubbed in your stupiditys.

Most obviously with your stupid pig ignorant more than 95% claim.

I guess I should have read more of his posts before I replied to his drivel. Well, we
live and learn.

Yep, that you couldnt bullshit your way out of a wet paper
bag even if your pathetic excuse for a 'life' depended on that.

In spades when a split system is used to cool the PC.
There is no external air to mix with, and they dont produce
air which has had more than 95% of the water removed either.

 

[Ken Maltby]

Cold air from an AC does have less water in it, BUT, the
parts that get cooled by this air may come in contact with
OTHER air besides that from the AC vent, air which is warmer
and has more water in it, that water settling out onto the
cooler parts if the temp is low enough.

True, but if the cold dry air displaces the warm moist air, so
that there is only dry air in the vicinity of the cooler parts, it
is no problem. By forcing the denser cold dry air into what
is essentially a leaky box, with a fan, all the warm moist air is
forced out, until there is only dry air in the box. There is also
the fact that there is a heatsource in the box, which will turn
some of that cold dry air into air warm enough to accept a
little water vapor, if any should creep in, this in effect, makes
for an actively drying environment.

Now if the dry air were to be stopped, there is a short period
of time where the cold dry dense air dominates, but it will slowly
be infiltrated by the ambient air. Normally you would also have
shut down the computer during this time.

So you can have a period where you have shut down both the
computer and the air conditioner and the inside of the leaky box
is cold enough that, too the extent that warm most air infiltrates,
condensation on the computer surfaces will occur. If the
computer were then turned back on, there can be problems.

Over time the computer will return to the ambient temp of the
room and no longer condense water vapor out of the air, (Here
in central Texas, that can be a relatively short period of time, and
what was wet, will soon dry.)

So aside from a brief predictable vulnerable period, this form
of cooling is quite workable. As I said, I've had my less than
totally responsible son operating such a setup for over a year
now without incident. I know he hasn't paid a lot of attention
to my instructions on how he should operate the computer.

Luck;
Ken

 

[Rod Speed]

True, but if the cold dry air displaces the warm moist air, so
that there is only dry air in the vicinity of the cooler parts, it
is no problem. By forcing the denser cold dry air into what
is essentially a leaky box, with a fan, all the warm moist air is
forced out, until there is only dry air in the box. There is also
the fact that there is a heatsource in the box, which will turn
some of that cold dry air into air warm enough to accept a
little water vapor, if any should creep in, this in effect, makes
for an actively drying environment.

Now if the dry air were to be stopped, there is a short period
of time where the cold dry dense air dominates, but it will slowly
be infiltrated by the ambient air. Normally you would also have
shut down the computer during this time.

So you can have a period where you have shut down both the
computer and the air conditioner and the inside of the leaky box
is cold enough that, too the extent that warm most air infiltrates,
condensation on the computer surfaces will occur. If the
computer were then turned back on, there can be problems.

Over time the computer will return to the ambient temp of the
room and no longer condense water vapor out of the air, (Here
in central Texas, that can be a relatively short period of time, and
what was wet, will soon dry.)

So aside from a brief predictable vulnerable period, this form
of cooling is quite workable. As I said, I've had my less than
totally responsible son operating such a setup for over a year
now without incident. I know he hasn't paid a lot of attention
to my instructions on how he should operate the computer.

Thats a separate issue to the original question tho,
why you dont see too many PCs cooled that way.

With the systems like
http://www.newegg.com/Product/Product.asp?Item=N82E16811221001
you will get higher humidity cold air into the case, because
that system cant be running the room air over very cold
surfaces and having the moisture condense out before
it enters the case, so you can have a problem with the
higher than normal humidity inside the case.

You'll get the same potential problem if you use a split system too
and many would prefer one of those to a window airconditioner,
just because its much quieter than the window air conditioner.

 

[shegeek72]

2 words: peltier devices

I couldn't find any CPU / GPU coolers that use Peltier devices. The
heat from the hot side of the Peltier device has to be removed. So
there's the same problem of how to get rid of the heat produced by the
Peltier.

The reason I want to cool the entire inside of the case is water
coolers only cool the CPU / GPU. The HDs and other components aren't
cooled, except by fans.

For example, my Maxtor (bought by Seagate) SATA HD has a low operating
temp of 5C (41F). I contacted AMD about the lowest operating temp for
my 3000+ and got this nebulous answer:

"We can't really define 'lowest operating temperature' since there are
so many cooling solutions (some of them use liquid nitrogen for cooling
which can cool your processor to the temperature below 0 C)."

I'd be interested in seeing the performance curve of a CPU as it
reaches 0C and below and if there's really any increase in performance
below a certain temp.

 

[kony]

I couldn't find any CPU / GPU coolers that use Peltier devices. The
heat from the hot side of the Peltier device has to be removed. So
there's the same problem of how to get rid of the heat produced by the
Peltier.

The reason I want to cool the entire inside of the case is water
coolers only cool the CPU / GPU. The HDs and other components aren't
cooled, except by fans.

That's fine, fan cooling works.

For example, my Maxtor (bought by Seagate) SATA HD has a low operating
temp of 5C (41F).

There are no hard drives that need to be at 5C, 41F. Why
would you try to operate it this low? Do you have any
evidence that it is expected to do better?

I contacted AMD about the lowest operating temp for
my 3000+ and got this nebulous answer:

"We can't really define 'lowest operating temperature' since there are
so many cooling solutions (some of them use liquid nitrogen for cooling
which can cool your processor to the temperature below 0 C)."

What did you expect? There's no need to test to an
extremely low temp because there is no reason to do it when
they sell parts spec'd to run stability at more moderate
temps, and do not spec nor guarantee extreme overclocking.

I'd be interested in seeing the performance curve of a CPU as it
reaches 0C and below and if there's really any increase in performance
below a certain temp.

No, there is not. CPUs are based upon clock rate. If they
were able to operate faster, they won't until that clock
rate changes.

 

[shegeek72]

That's fine, fan cooling works.

To a certain extent. Heat is going to fatigue electronic components.
Indeed, adequate cooling is a real concern in server buildings.

There are no hard drives that need to be at 5C, 41F. Why
would you try to operate it this low? Do you have any
evidence that it is expected to do better?

No, but I'm going to do some research.

What did you expect? There's no need to test to an
extremely low temp because there is no reason to do it when
they sell parts spec'd to run stability at more moderate
temps, and do not spec nor guarantee extreme overclocking.

I frequently read about excessive heat making CPUs / GPUs unstable and
the cooler they are they better they perform. It's a hot (sorry about
the pun) topic in the gamer community.

I've noticed better performance in demanding video games when my CPU /
GPU were running cool (relatively) v. hot, i.e. stuttering and
jerkiness.

I'm not an expert in this area, but from my rudimentary electronic
training cool electronic components are going to perform better and
last longer than those running hot.

 

[kony]

To a certain extent. Heat is going to fatigue electronic components.
Indeed, adequate cooling is a real concern in server buildings.

Yes, emphasis on "adequate", not run-it-in-a-freezer, cold.
The goal is not to try to get temp as low as possible,
rather to use the time tested and proven method(s) to keep
the temp from getting too high.

No, but I'm going to do some research.

Parts are designed to be used in a typical, human hospitable
environment. Even in extreme (hot) environments, the goal
is still to return system to a hospitable temp range, not to
go as cold as possible.

Further, you degrade the function of the electrolytic
capacitors severely by going under roughly 10C, even
significantly above that in the 10-25C range.

I frequently read about excessive heat making CPUs / GPUs unstable and
the cooler they are they better they perform.

No, you read that overheating effects stability but not that
"the cooler they are the better they perform".

Their performance only needs to be well enough to work at
the set speed. If you are overclocking, you may find heat
again rises, or reduction in temp allows lower voltage per
same frequency of operation, but these are specific
situations not a generic "better performance" idea, because
they cannot perform better so long as they're stable, still,
at the target temp unless their frequency is changed too.

It's a hot (sorry about
the pun) topic in the gamer community.

Then they're wasting their time.

I've noticed better performance in demanding video games when my CPU /
GPU were running cool (relatively) v. hot, i.e. stuttering and
jerkiness.

No, you mean when it overheated it was a problem.
The issue is not "how cool" it is, as if cooler is better.
The issue is not letting it overheat. It is like a line in
the sand, pass which it cannot operate properly. Under that
threshold, you gain nothing performancewise by going any
colder.

I'm not an expert in this area, but from my rudimentary electronic
training cool electronic components are going to perform better and
last longer than those running hot.

No, they are not going to perform better.

You are trying to overthink the obvious- don't let parts
overheat.

Yes parts will last longer if cooler, but again it only
matters if they were hot enough that it degraded their
functional life below their needed life. If you managed to
get your CPU or GPU to run for 30 years by lowering temp
instead of 15 years, did you really gain anything of value?
Since the 15 year old CPU/GPU is practically worthless, the
maintenance alone on the cooling design was not worth the
gain even if it had no cost, size, power, or noise
detriments.

There is no mystical secret that the entire industry has
overlooked, particularly not in basic concepts such as
operating temp. If you want to overclock to extreme levels
you have the same criteria to consider, to merely keep it
cool enough to be stable and have acceptible lifespan. Even
highly overclocked gaming systems kept cool enough (room
temp range not fridge or freezer range) will last until
they're long obsolete, too slow for contemporary games.
That is unless you have some unique flaw in the system, in
which case the flaw should be corrected.

 

[GEO Me]

[more nonsense]

Geo, he is a troll.

Thank you, I had previously read comments about Rod Speed's
stupidity, and I guess they were right on the mark.

Thanks for the warning.

Geo

 

[SubKool_Ntrance]

On 23 Aug 2006 15:56:30 -0700, "shegeek72"
<[email protected]> wrote:

<snip long winded text>

Like i stated in an earlier post, i use an old heater core (copper tubes
brass fins)from a car, put it over one of the vents in your box, and use
circulated water to lower the temp, a fishpond pump works just fine,grab
two fans, one behind the heater core, one on the other vent of your box,
make sure the one on the heater core pulls air through the core, and
into the box, and the other one forces air out of the box.

My CPU operates at a constant 11.3 C, and the ambient air temp in the
box is about equal to taht, never stuck a thermometer in there, dont
even own one, lol.

Should you need a diagram or a custom cooling core, please feel free to
chase me up, the only reason i know these things is that i spent a year
working for a car radiator place, and did a shite load of custom builds
for hotrods, and miniature model cars (yes it is possible, copper
cooling coil around the head)


David.