CFM calculator to get rid of HEAT (TDP or WATTS)

S

Skybuck Flying

Hello,

I'm curious if it's possible to build a powerfull PC/iCore/GPU/Cuda PC
without getting into heat problems ! ;)

For example CPU has about 30 TDP (which is close to watts, otherwise find
watts listing) and GPU could have as much as 200 to 300 TDP (watts) (top
model for maximum bandwidth).

Additional watts for memory( 20 watt ? just guessing ;)) /motherboard (10
watt ?)/4 harddisk (60 watt*4=240)

Let's say 800 watts.

How much CFM would be needed to get rid of all that heat ?!?

A CFM heat (PC) calculator would be most handy... maybe somebody can
make/program it.

What would be good formula's to use ?!? ;)

Bye,
Skybuck =D
 
S

Skybuck Flying

Hmmm... could be really simple...

Wikipedia shows conversion chart:


http://en.wikipedia.org/wiki/Conversion_of_units#Power_or_heat_flow_rate

1 atmosphere-cubic foot per minute = 47.820 074 682 24 W

Which is probably close to 48 Watts.

But this is atm-cfm and not cfm hmm... not sure if that is same thing... but
for now I'll go with it ;)

So (800 * 60) / 48 = 1000 CFM needed ! ;) =D


Let's see if antec case is sufficient lol:

http://www.antec.com/pdf/manuals/1200_EN manual.pdf

It's about 240 CFM...

So there is no way that the antec 1200 case would be able to cool such a
system sufficiently.

Yes perhaps the graphics card would output a little bit of air as well...
but it's waaayyyy too little.

And this manual probably assumes at maximum operating fan speed... so let's
divide 240 by 2 is 120 CFM... and let's divide it one more time because of
dust is only 60 CFM.

Yeah...

Well it's pretty clear where this is going:

OVERHEAT DAMAGE ! ;) =D

Bye,
Skybuck ;)
 
S

Skybuck Flying

This one also seems kinda cool but a bit old, but that could be an
adventage, since I am on a PIII 450 mhz ;) and best of all it's free, alas
only for MAC OS and Linux:

http://tetruss.larc.nasa.gov/

Could still be interesting though.

Bye,
Skybuck.
 
T

TheJoker

Really interesting stuff:

http://en.wikipedia.org/wiki/Heat_transfer

Getting down to basics now lol... getting down to the core of it ! ;) =D

Bye,
Skybuck.
You are still so dumb that you will lay your fat fingers all over the
circuits and ESD FAIL each and every one of them.

Somehow, I also think you have this erroneous habit of 'tweaking' your
motherboard BIOS settings on every machine you ever owned.

Essentially, you qualify as an idiot, not for what you think you are
taking in (learning), but for all the things you have decided to ignore.

Yet another proof is that you cross-posted this stupidity to an HVAC
group. An HVAC engineering group, no less.
 
S

Skybuck Flying

TheJoker said:
You are still so dumb that you will lay your fat fingers all over the
circuits and ESD FAIL each and every one of them.

Good thing you started it, I am glad you did.

I fokked my Pentium III 450 mhz so hard ! and it's still working hahaha !

So shove that ESD FAIL theory straight up your ass.

I disassembled it so many times, even placed it on table with fluffy stuff
on it.

I literally disassembled every screw of the whole computer system, I fokked
the motherboard.

I fokked the graphics card with water... It probably even spilled drops on
the motherboard...

And it still works ! LOL.

It never died.

I am computering on it right now !

Check the headers and shit ! ;) =D

Goodbye,
Skybuck.
 
R

Robert Myers

Yeah,

I heared about these motherfokkers too:

http://www.ansys.com/Products/Simulation+Technology/Fluid+Dynamics/AN...

They've been around for a long time... maybe old maybe not old maybe good
maybe not good...

So far it's a bit vague...

Butt the pirates have it too ! ;) =D

Bye,
  Skybuttpirate ! LOL.

It's only fair, Skybuck, that you probably don't read most of my
posts, because I don't read most of yours.

If you are trying to advance the state of the art in heat transfer as
applied to microelectronics, my suggestion for any of the codes you
have mentioned is that you find a dumpster for them so as not to waste
your time. Mitch Alsup might not agree; I'm not sure what he's up to.

The fundamental difficulty is that heat transfer such as would occur
in a microprocessor heat sink depends intimately on details of
microscopic turbulent flow that are poorly understood for reasons I
have been ranting about for years. The theoretical problems get worse
as you push to higher Reynolds number, which is where you are headed
if you are trying to push air faster.

If you are ever in a window seat on a modern airliner, look out the
window at the row of vortex generators on the upper surface of the
wing. Those tiny little pieces of metal have a dramatic influence on
the aerodynamics of that big old wing, and that should give you an
idea of how hard this problem is. Predicting heat transfer is harder
than predicting pressure forces, so the problem is even harder than
those teensy little pieces of metal would indicate. The problem is
absolutely fundamental, and there is no way out of it. For an
airliner, the largest scale is the wingspan. The smallest scale of
relevance may not be visible to the naked eye.

There probably are significant opportunities for improving heat
transfer in microelectronics. My suspicion is that most of the
advances are made by bending tin and making measurements, rather than
by running CFD codes. I'm not a tin bender, so I could be wrong.
Conceivably, someone with a deep understanding of fluid mechanics,
heat transfer in solids, and computation could make progress that
would make life a lot less random for the tin benders. If you don't
have those skills, you are probably better off putting your money into
a high-stakes lottery, because that's how much chance you have of
making progress by fumbling around in the dark.

Robert.
 
T

TheJoker

So shove that ESD FAIL theory straight up your ass.

You're a goddamned idiot.

The ESD FAIL is YOU fat fingering all your gear. Your grasp of
electronics is nil.

I have a computer that has lasted for over 8 years.

That one, and all of my previous computers all still work, and NONE ever
failed.

The fact that ALL of your pieces of shit have failed is absolute,
undeniable proof that YOU are the dumbfuck breaking your gear.
 
R

Robert Myers

        So you think 1,000 CFM, requiring a 2.5 - 3 HP motor, might be
excessive for a desktop computer ? :)

Heat transfer is not my thing, but you know that if someone is talking
about using out of the box CFD to decide if something is even
plausible, without seeing if anyone has done anything even remotely
similar, and without knowing anything about fluid mechanics, heat
transfer, or CFD, they're off on a wild goose chase.

Yes, I know, I could have figured that out from the poster. Somehow,
I just could not stop myself.

Robert.
 
S

Skybuck Flying

I have one computer from 1992.
I have one computer from 1995 or so.
I have one computer from 1999.

All three still working.

And now I have a dead computer from 2006.

So I have one computer which is now working for 19 years ! ;) :p*

Bye,
Skybuck.
 
S

Skybuck Flying

I would use it to test the airflow of the case itself, not necessarily the
tiny little processor ;)

Just to see if the case can transport the heat out of it.

I'm a bit worried that it might not run on my PIII 450 mhz to be usuable...
like too slow GUI... or maybe it will run... anyway that's why I posted
multiple CFD softwares, maybe one of them will work on 450 mhz ;)

Would be nice to be able to verify if my case can actually transport enough
heat away for any new computer design.

However if I would go with a very powerfull graphics card then the PC would
probably require a new supply which is a bit too expensive for my taste at
this point.

I have not tracked down the source of death and nothing has been done so
far... perhaps my "land lore" or whatever it's called... they guy/company
who collects the rent might contact me to see if something can be done about
it... but perhaps not...

Perhaps the appartment and everything else is hard-wire for 2 wires only...
of which one is nul-wire...

The weird thing is the kitchen has 3 wire plugs but perhaps those are
fake... the bathroom seems to have 2 wires to washing machine which is
weird...

Perhaps I just live in a shitty appartment if that's the case well then to
bad... but that will have consequences for my future PC's... I would then be
a bit too scared to buy expensive equipment because it's out-dated so
fast...

So then I would have to go with cheap equipment which kinda sucks as well.

So currently I see no way out of this, thus I am lingering waiting for a
solution ;)

I'm also investigating some software technologies to see if I can program
that... which will influence my decision how I shall repair or replace my
computer ;)

Perhaps I will even construct two computers, one for desktop, and perhaps
one for server, and then a router.

The problem is the old-cpu and old-graphics cards are so hot I don't dare to
re-use them for a server, so buying a super computer was probably a big
mistake... but I wanted to try it ;)

Bye,
Skybuck.
 
T

TheJoker

So I have one computer which is now working for 19 years ! ;) :p*

Bye,
Skybuck.

Still plugged in? Then you are a waste of electrical energy as well.
I'll bet that your carbon footprint is three times as big as it should
be. That too is due to your utter stupidity.

Now, idiot child, list all the ones you have broken.

It would not surprise me if the number wasn't the same as the number of
times you answered your own post AGAIN in this thread.

You AND your pathetic posts are about as stupid as it gets.
 

Ask a Question

Want to reply to this thread or ask your own question?

You'll need to choose a username for the site, which only take a couple of moments. After that, you can post your question and our members will help you out.

Ask a Question

Top