• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

MEGA cooling, MEGA potential, Cheap

R

RAGEdemon

n00b
Joined
Jun 27, 2004
Messages
38
Alright, I was thinking about this and i see no reason why one of you elites can't try it out on a Graphics card or CPU:

It is simple:
Link the main heatsink of either or both to the Main Case.
Case is made of Metal, the thin sheets ensure excellent head dissipation, it has massive surface area, it has fans all over the place, and its grounded. The only downside I can think of is the main contact will have to be big and solid so as to transfer the heat to the case adequately from the card/CPU. The case effectively becomes a massive heatsink.

I cant believe noone has thought of doing this before... or maybe someone has?

Can somone try this and let the rest of us know how it performs? Obviously the actual performance will be case dependent, but still, a rougn idea is good.

--RAGEdemon
 
First, there is a large distance between the GPU and the case that you'd have to bridge. This alone would introduce much more thermal resistance than you would ever want. The only way to do that would be with heatpipes instead of a solid metal path for heat to move through.
But then again, the case would make a horrible heatsink. Take a look at CPU heatsinks - they have many fins whereas the case has... well none. It just doesn't have that much effective surface area. Not to mention most cases are made of steel, which is a terrible conductor of heat compared to copper or aluminum.
So in theory this could work with a heatpipe moving heat to a passively cooled case, but it wouldn't really work very well...
 
Hi,

Thanks for the input.
This is the hardcore overclocking foum so i figured people would have invested in well cooled cases, Aluminium etc. The heat dissipation through the fins is good, but the case would effectively be a large fin, hich would have the effect f lots of small fins? Albeit not the solid contact for all bases for good heat transfer, but still, better than just a heatsink on its own, right?

As mentioned in the earlier post, i do see the problem of transferring the heat from the card/spu to the case itself... perhaps some welding is in order... again, something for the hardcores to try :)

It is an open idea, anything tha comes to mind. It merely seems to offer better cooling than a heatsink would able to manage on its own. :)

Perhaps someone with some time on their hands can give it a go.

Again, Thanks for the input.
--RAGEdemon
 
Since heatsinks have all those fins and ridges and stuff, I bet if you added up the surface area of a heatsink it would be pretty close to the side of a case. Maybe not, but it has a lot more surface area than you think. You would need a very good heat conductor to get the heat from the processor to the side of the case. You also need very good, very tight connections between the processor and the conduit and the conduit and the case. Also, heat dissipation from the side of the case would rely on passive radiation of the heat, whereas heat dissipation from a heatsink relies on active air exchange. I think this would be very suitable in some sort of embedded environment or something, but I don't think it would be very suitable for a power user's machine. Though I have thought about this before and I was considering it. I'm just lazy so the idea never left my head.
 
The case may have lots of surface area, but it is not geometrically well-designed to be a heatsink. It's basically a 2-dimensional plate - heat would have to transfer over a loooong distance to utilize the whole case, and that's not going to happen. With a heatsink, heat moves outward from the center only a few cm before transferring up into the fins.
With your reasoning, you'd think a flat plate would make a good heatsink to just set on top of your CPU or GPU.
 
Yeah, just reiterated what I said about the solid base to many fins vs no base to single fin in my previous post :)

Thanks anyway :)

--RAGEdemon
 
Zalaman already makes a case like that. It has heatpipes to move all the heat from key components to a case covered with fins.

It's perfectly quiet, but don't expect to overclock with it.
 
Perhaps a case is a terrible heatsink, but it could make an excellent radiator. Just make solder halves of copper piping to the inside, or outside perimiter of the case, and include it in the water loop. Should work well.
 
M4d-K10wN said:
Perhaps a case is a terrible heatsink, but it could make an excellent radiator. Just make solder halves of copper piping to the inside, or outside perimiter of the case, and include it in the water loop. Should work well.
Click to expand...
Uh, can you care to explain how radiators and heatsinks dissipate heat differently in a way that would allow this to work for a radiator but not a heatsink? If one has a sufficiently long piece of tubing then the case would not add much to the cooling capabilities of the loop.
 
You could also solder fins to inside of the case. Either way, it would have a greater surface area than tubing itself; and conduct heat to the parts of the case improving the surface area. Either way, this shouldn't be any better than a radiator; if the tubing is long enough it should be just as good.. however this eliminates the need for a radiator.

It's better as a radiator than a heatsink, because you would somehow have to transport the heat to different parts of the case in order to turn it into a heatsink; with this - water has to go somewhere anyway, no?
 
The heat transfer through the wall follows a simple equation:
Q=k/L(T1-T2)


We can draw some interesting conclusions from this equation. First, heat transfer is proportional to the temperature difference on the object. If the temperature differential doubles, the heat transferred doubles. Second, the conduction coefficient "k" is proportional to heat transfer. If the conduction coefficient doubles, the heat transfer doubles Alternatively, for the same differential temperature, twice as much heat may be transferred. The final observation is "L". As thickness (or distance) increases, heat transfer decreases. Alternatively, to maintain the same heat transfer through a material twice as thick requires twice the temperature differential.
Click to expand...

General Heat Transfer Guide
 
M4d-K10wN said:
That is so off topic. We're not moving heat; we're moving the material.
Click to expand...

if you really think that perhaps you should repeat thermodynamics :p
 
M4d-K10wN said:
That is so off topic. We're not moving heat; we're moving the material.
Click to expand...
In a thread about a different cooling method we are not discussing moving heat? ;) And if you are moving the material then there has to be some way for the heat to get to that moved material.

Ice Czar brings up a good point and I had suspected it all along. The more you try to move the heat around, the more it is going to hurt heat transfer.

I am really starting to like the idea of turning part of the case into a watercooling radiator. I'm tossing some ideas around in my head right now.
 
Fine I'll be like Ice Czar. Multiwalled, XX sided carbon nanotubes have proved to be a superconductor of heat. Now, go figure why that information is useful.
 
if you where designing an airplane, knowledge of carbon fibre would be useful
but not as useful as understanding Bernoulli's Principle ;)

the above equation governs conductance within any heatsink
and the guide linked gives you a relatively good overview of convection as well

for fiuture reference, the point of a link it to refer someone to information

the post in and of itself, highlights the difficulty of transfering the heat from a point source to a very large interface area by whatever mechanism, the larger the interface the more difficult that becomes, while that addresses conductance, there is a corrallary for convection (transfer from the heatsink to a fluid (air\water\ect)
its known as Newton's Law of Cooling


which basically states the same relationship regarding the temperature differential
since you dont seem to be grasping those points you might consider reading that link
or advanced studies A Heat Transfer Textbook Lienhard & Lienhard (MIT & U of Houston) PDF

regarding other thoughts, is the case as heatsink going to be passive or forced convection?

If you where actually going to be like me youd be both more polite, and have a sense of humor :p
 
Ice Czar said:
if you where designing an airplane, knowledge of carbon fibre would be useful
but not as useful as understanding Bernoulli's Principle ;)
...
Click to expand...
Carbon nanotubes != carbon fibre. Carbon nanotubes have as many interesting properties as configurations. You could have a varying internal diameter, and a variying wall thinkness. The only way to create a carbon nanotube right now is using an electron microscope, from what I understand. The carbon-to-carbon bond in the nanotubes is the strongest bond in nature; a material that would consist of possibly fabirc-like woven together nanotubes would be almost indestructible. Some configuration of nanotubes is a super resistor. Some othe one is a semiconductor; while yet another hints ar room-temperature internal electrical superconductivity, while yet another, multiwalled configuration is a proven superconductor of heat.. heat conductivity at 100% efficiency. Carbon nanotubes are only a few nanometers long, and will never get any longer, well, not with today's technology. There is a way to fuse nanotubes togeter; using an electron microscope, forgot how, but it involves high voltage ;). Another way of making several nanotubes at a time, makes them all attached to each other.. looking like a hairy ball... thing.

The topic of nanotubes is an interesting one. Seeing how you confuse them with carbon fibre, I suggest you do more research. You're a nice guy, and everything, but you tend to misinform people; just like in that A64 memory thread, you for some reason decided to say that regular A64s don't have an on-die controller.
 
its called an analogy :p

"drawing a comparison in order to show a similarity in some respect"

out of couriosity how old are you?
 
Riiiiiiiiiiiiiiiiiight. That was remarkably fast how you read my post. Me thinks you didn't read it at all. Either way, the analogy you're trying to turn it into doesn't work; I wasn't talking about carbon fibre in the first place.
 
the analogy was simple

advanced materials < aerodynamics > Bernoulli's Principle

advanced materials < thermodynamics > Newtons Law of Cooling & Q=k/L(T1-T2

would you have been happier if Id used Boron Composites?
 
Ice Czar said:
the analogy was simple

advanced materials < aerodynamics > Bernoulli's Principle

advanced materials < thermodynamics > Newtons Law of Cooling & Q=k/L(T1-T2

would you have been happier if Id used Boron Composites?
Click to expand...
Yes, but considering I was referring to something different entirely, and it seems like you were trying to continue with that, made you look stupid.
 
M4d-K10wN said:
Yes, but considering I was referring to something different entirely, and it seems like you were trying to continue with that, made you look stupid.
Click to expand...
No. It was an analogy. The fact that you don't understand what that means makes you look stupid. Just drop it. Calling a moderator stupid is also pretty stupid.
 
jpmkm said:
No. It was an analogy. The fact that you don't understand what that means makes you look stupid. Just drop it. Calling a moderator stupid is also pretty stupid.
Click to expand...
Talk to me when you're done kissing... ass.
 
This thread is skating on thin ice... Get it back on topic and end the senseless banter before it sinks.
 
Either way, I stand by my case > radiator idea. You could find a ribbed case/make your own case panels with ribs to increace surface area. You could also solder fins to tubing itself, and turn all tubing you use into radiators. They sell some frexible metal tubin in Home Depot, that metal is really soft..

Edit:

Someone is threatening to kill me if I don't stop arguing and inevitably insulting Ice Czar. I propose we put each other on ignore lists so we'll never see each others' posts again?
 
I think your main problem will be surface area. Your typical case has maybe 400 sqin of surface. That's a fraction of what a good heater core or similar radiator provides. Also, your panel will radiate heat just as well to the inside as the outside, so you will either have to insulate the inside (loosing you half your efficient cooling area) of set up a very clever laminar flow sytem that blows air along the panel inside, but does not allow that heat to mix back into your case. I'm sure it can be done, but it might not be as efficient as it looks from the outside.
 
I knew that idea sucked. On to my other idea, build the entire case out of fans! Now that would push a lot of air.
 
M4d-K10wN said:
Edit:

Someone is threatening to kill me if I don't stop arguing and inevitably insulting Ice Czar. I propose we put each other on ignore lists so we'll never see each others' posts again?
Click to expand...

Because Im a mod, I really cant do that, however our "issues" can be addressed privately
Im constrained as a Moderator from seriously argueing, which is really a pity, because I was considered quite good at it :p (strictly within the rules) or derailing a thread.

Im also loath to use my position as a means to gain advantage.
However respecting the moderators is a key principle here,
so let me extend to you the opportunity to privately insult me as much as you like :p

I think we can manage to both post in the same thread and even disagree civilly
so I'll start :p

http://www.amdmb.com/article-display.php?ArticleID=105&PageID=8
One also needs to be aware that there is such a thing as too many fans. Each fan draws power to create airflow. The fan converts this electrical power to heat through motor inefficiency and direct compression heat in the air. If a fan draws 5 watts to generate 42 CFM, it will raise the temperature of that air by about 0.22°C. This is a small number to be sure, but certainly not zero. So how many CFM are enough?
Click to expand...

which is a corrallary of what in a water cooled system is considered "pump heat"
the energy directly transfer to the fluid, as momentum. Water and Air both being fluids
however the actual impact in an aircooled system isnt nearly as great since the molecular density (and thus heat transfer potential) is so different, there is however a point of diminishing returns with fans

also, I thought I might point out that OC&C was my [H]ome forum here for the first 2 years
and nearly my first post was a Thermodynamic Reference Linkfarm that was unfortunately lost in the great thread dieoff a couple of years back.

It had a whole section on Carbon Nanotubes, Diamond Substrates and other advanced materials. ;)

Im sure we can "play nice"
Im not perfect and Ive made mistakes before and Im guranteed to make them again
however they are pretty rare.
 
regarding the heat transfer of the case both inward and outward
while that is true, consider that the temperature differential also enters into it
(per the first post) and that even without insulation on one side heat transfer will not be proportional in both directions
unless the ambient temperature and the airflow match on both sides

V0ltage's link demonstrates how heatpipes could be employed
and M4d-K10wN's idea is still valid as well,

ever seen a waterwall?
imagine a waterwall with the whole interior side of the case being the wall
thus a huge contact patch but also a very "thin" wall ideally out of copper, and a second insulated wall inside of that
then the outside of the case with fins for additional interface

consider the difference in the bottom of a typical waterblock
the Whitewater has a very thin bottom, to assist heat transfer which is quite good directly below the impinging jets but is actually a very small area\interface, now consider a peltier setup, that adds a big heat spreader of thick material, that by comparision to the thin wall is an insulator
but its job to to increase the total interface area, to match the size of the pelt which is much larger than the Whitewaters "interface" area and
because it is able to maintain such a high temperature differential, its able to remove the heat faster throughout the whole chain of conduction

tradeoffs, so you match the materials and mechanisms to match the goal

my orignal post wasnt meant as a dig
rather it was meant to point the thread in the right direction regarding the physics involved

Heat transfer is all about temperature differentials. Conduction through materials and convection away from surfaces is proportional to the temperature differential that exists.
Click to expand...

which points up the real question of aiding the heat transfer off the exterior of the case
forced convection being so much more productive than passive convection

if it is sitting in a drafty room in Siberia vs a closed room in Singapore
strangely I dont see a °C/W rating for that Zalman TNN

whatever thermal solution you have is typically rated with a °C/W rating (Thermal Resistance (Rht) in °C/W actually)

Delta = CPU temp* - Ambient Temp
C/W = Delta / CPU Watts*

Interpreting C/W: For every watt (CPUw) that the CPU consumes, the HSF will limit the CPU's temperature rise to (C/W x CPUw) plus the temperature at the HSF's fan inlet. For example, at an ambient temp of 25 C, a C/W of 0.25 with a CPU radiating 50 watts means that CPU temp will increase 50 x 0.25 = 12.5 C over ambient temp, or 37.5 C. The lower the C/W, the better.

* or in this case the whole case\CPU array and total heat from all the components
Click to expand...

basically the thermal resistance of a given solution changes with the temperature differential, which is why thermal solutions are rated in °C/W which leaves that variable ambient open but gives you a formula
 
V0ltage said:
http://www.xoxide.com/zalman6.html

just put a boxfan or 2 up against it and you have some uber cooling.
Click to expand...

There you go, that's the case I was talking about.

It cools alright, but don't expect to overclock. They advise that you only use it for high-end parts if you have air-conditioning and keep it out of the sun.

VoodooPC makes machines with that case.
 
You must log in or register to reply here.
Back
Top