Copper vs. Aluminium

[shadowjagans]

whats better for cooling, copper or aluminium?
i herd that copper absorbs heat faster but the aluminium dissipates
heats faster, and new vga cards are comin out with aluminium heatsinks
so.... comments?

 

[AlaskaFox]

copper=better heat transfer
aluminum=cheaper and lighter.

when you are talking about mounting a heatsink on an add on card, going heavy is asking for trouble, especialy in the OEM market. I am sure that is the last thing you want is to buy a brand new brand X puter only to have yer vid card snapped in half from shipping...

 

[chuckshissle]

Copper is better than aluminum when it comes to heat transfer, that is why mostly all heatpipes on cpu coolers are made of copper. Aluminum is not as good but it is lighter and they made up of mostly all radiators and fins on cpu HSF or water-cooling rads where weight is a concern. Some of the best cpu coolers and water cooling rad are made up of both copper and aluminum.

 

[wun911]

Copper is better at heat transfer, however it is heavier than aluminum. Also copper will tarnish after a while but aluminum wont.

 

[Jackalope73]

The thermal resistance of copper is lower than aluminum as is its electrical resistance. This is the case with all precious metals. From lowest to best conductivity of heat and electricity; aluminum, copper, silver, gold, platinum. That is the scale of the precious industrial metals excluding high noble metal processes and alloys. The bottom line on the video heatsink are as previously stated light weight against brittle silicon pcb and aluminum is way cheaper and easier to work with. It's all economics mostly but some heatsink manufacture use a copper heat spreader with a massive aluminum fin structure to get the most surface area per unit of weight.

 

[PhyberOptik]

The thermal resistance of copper is lower than aluminum as is its electrical resistance. This is the case with all precious metals. From lowest to best conductivity of heat and electricity; aluminum, copper, silver, gold, platinum. That is the scale of the precious industrial metals excluding high noble metal processes and alloys. The bottom line on the video heatsink are as previously stated light weight against brittle silicon pcb and aluminum is way cheaper and easier to work with. It's all economics mostly but some heatsink manufacture use a copper heat spreader with a massive aluminum fin structure to get the most surface area per unit of weight.

Sir, your knowledge of the relevant chemistry is inadequate. I shall attempt to remedy the situation.

I refer you now to WebElements.com, and the image below which has been reproduced from there in terms of the WebElements copyright.

Nature's best performers in terms of thermal conductivity for pure metals are the group 11 metals. Copper, Silver, and Gold. The relative value of precious and semi precious metals have nothing to do with their thermodynamic properties. The value of a metal is completely arbitrary in humans terms, and is a result of its rarity, and/or usefulness for a particular purpose. For example, if copper was more scarce than gold, it would traditionally (in pre-industrial times) have been hardly be more valuable than gold, simply because it tarnishes, discolours skin, and as such its desirabity as a cosmetic metal would have been low when compared to gold.

But I digress.

As you can see from the graph, Platinum's (group 10, next to gold) TC is woefully inadequate for heatsink use. Even a heatsink made from pure silicon, would offer better thermal conductivity. And yet, ultra-pure silicon is in fact used as a thermal insulator on the Space Shuttle's exterior tiles....

Copper's thermal properties seem too good to be true - it closely follows that of Silver, the metal with the highest thermal conductivity, yet it is cheap and abundant. However, aluminium's popularity is due to a number of factors, notably, Aluminium's high TC:density ratio, BUT PRIMARILY, its low melting point. The low melting point allows extrusions to be made very cheaply. Molten Aluminium is "squeezed" through a shaped orifice, which are cooled to form long extrusions (complete with fins, etc). These are then cut up into individual items and machined for detail)

What is interesting, is that while pure carbon (a non-metal) has a low thermal conductivity, nothing beats carbon in one of its allotropic forms, which of course is pure diamond.

Another thing, the core of a cpu is not called a "PCB". PCB is notionally "Printed circuit board", but actually it stands for what the early circuit boards were traditionally made of - Polychorinated Biphenyls. Today, the substrates used for circuit boards are much less hazardous.

 

[shadowjagans]

wth r u ppl chemistry teachers or somthin, even i didnt learn this mch in Chemistry or IT classes

 

[PhyberOptik]

No, but if you're passionate about computers and overclocking, you aquire this sort of knowledge over time. Wait 'till you start messing around with phase-change cooling, then you'll learn all about the difference between mineral and synthetic compressor oils, the different refrigerants, and calculating heat load and capillary tube lengths, and so on.

It's a very satisfying hobby, if you can afford all the equipment :?

 

[shadowjagans]

dude this is jus interacting with ppl in the forums and a way to "make friends", yes i admit i probably did overdid it on the forums, but theres totally nofin wrong with replying to ppl, its a friendly thing to do, theres nothing wrong with doing so

 

[PhyberOptik]

shadowjagans, stop flooding the forums with your worthless questions. and then you say stuff like "what are you chemistry teachers?". dude, get lost. why do you people even respond to his stufff? i did just now cause i'm pretty tired of his stuff and one other person i know for sure is too... don't you have your avatar or whatever it is already? geez...

This tirade is utterly uncalled for. The "chemistry teacher" bit sounded somewhat like a compliment, if you ask me.

Habitat87, you may not like his questions, but they have their place. If everybody knew everything, then these forums would not exist. Neither you nor myself are entitled to draw an arbitrary line as to what sort of question is "worthless" or not.

What we are entitled to, is to establish whether answering a question is worth our time, and you are not excercising that right properly. If the questions bother you, then ignore them. You are able to see who created the thread, yet you still chose to enter and reply.

Considering the above, the fact that you entered the thread only to end up flaming, your behaviour can arguably be described as trolling, not so?

 

[shadowjagans]

i have to thank PhyberOptik for backing me up =]

 

[Datman]

I'm curious as to all the questions so quickly together, I don't have a problem with it.
They are good and general questions, just jump around from topic to topic.

 

[PhyberOptik]

Shadowjagans, no sweat. As you learn more and more, you'll ask less questions and start answering more of them for others.

Of course, one must never ever stop asking questions.

 

[shadowjagans]

geez, alrite, i apologize for making ppl unhappy in this forum, so no need to argue with each other

 

[PhyberOptik]

habitat87 is right, forums has one purpose and that's to be productive. 90% of your threads aren't and have already been discussed in past threads.

Well, I don't agree. How does one measure "productivity"? "Productivity" is not the goal, because you cannot quantify it in real terms. If the goal of a forum is technical discussion, then the asking and answering of technical questions, no matter how mundane, fullfills that goal admirably.

The aim is not "to further the aim of the hardware/overclocking/hacking art", because not everybody here is at the cutting edge. If that's the sort of community you wish to have exclusively, then it exists elsewhere such as at xtremesystems.org.

I agree that many questions may be answered if the asker chose to search first, but in my opinion, life shouldn't be a series of googles. The internet, such a great bringer-together of people, fails dismally if its citizens were to choose not to interact for silly reasons such as the meritocracy in question here.

 

[PhyberOptik]

yeah, but i didn't realize this was a "make a friend" forum... and as for the trolling, yes i agree, but that has it's place when something seems completely stupid and wasteful am i not correct? if you can make a comment that mines wasn't appropriate i can express my opinion that his was a completely waste of time too. except your was all opinion, his pasts posts evidently a waste of time... why would you want to start of with a complete waste of a forum? i stand with with what i say.

I'd like to point out that by using the original post's "worthlessness" as a rationale for your own "waste of time" post, you are logically validating the original post existence, and therefore you're not allowed to comment on it in the first place. Can you see the circular logical paradox that you are creating?

The correct way to avoid the paradox, is to sidestep it and not comment on posts that you feel are worthless. They may be worthless to you, and you might be 100% correct, but you're obviously not the only person who frequents this forum, not so?

On the topic of perceived "productivity", I'd like to point out the irony of what this particular thread has devolved into - from a technical discussion about cooling and heatsinks, into one about the desired ettiquette of the forum. Yet the devolution was created exactly by those who purport to reject worthless, "unproductive" posts.

 

[Bluefinger]

I studied A-Level Physics, which I learnt quite a bit about thermal dynamics, including what makes a better coolant. In terms of an HSF, I will just reiterate what everyone has said before. Aluminium is popular purely for the reason it has a high heat capacity (takes more energy to heat it up than copper) and its very easy to work with in order to produce thin fins for heatsinks. However, Aluminium does suffer slightly that its thermal conductivity isn't as good as copper's, so the heat doesn't spread across a aluminium fin as quickly as a fin made out of copper. Also, aluminium doesn't surrender the heat as easily to air because of its high heat capacity, but this is usually overcome by making sure the heatsink has a very high dissapation area, so the more surface area exposed to the air, the easier it is to dump heat into it.

 

[PhyberOptik]

Also, aluminium doesn't surrender the heat as easily to air because of its high heat capacity

This is certainly interesting! If you ask most people in the hardware and overclocking game, we've been led to believe that it's the other way round. I certainly thought that it was the other way round (which answered the question of why many heatsinks have a copper base with aluminium fins - issues of mass aside).

I'll definitely have to read up on it more.

 

[cubicleslave]

One glance at this table, and you'll know which material performs better at heat transfer.

http://www.monachos.gr/eng/resources/thermo/conductivity.htm

 

[chunkymonster]

whats better for cooling, copper or aluminium?
i herd that copper absorbs heat faster but the aluminium dissipates
heats faster, and new vga cards are comin out with aluminium heatsinks
so.... comments?

Alrighty there pal...this has to be the 7th or 8th thread that you've created asking these BS questions about "Which is better, this or that?"...wtf dude...seriously, you got nothing else to do with your time...c'mon, I'm all for people asking questions and participating in the forumz to find answers, but for effin' sake, this is just friggin' ridiculous!

Not for nothing, but a simple google search will return scads of info on the thermal conductivity of copper and aluminum. Better yet, go to school, take some engineering or physics classes, and learn this $hit for yourself and answer your own damned questions!

Wtf...another bull$hit thread!!!!

 

[cubicleslave]

Forgot to address your "absorb" vs "dissipate" conjecture. I think that's related to how much heat a given material can hold. Here's another link:

http://www.ami.ac.uk/courses/ami4817_dti/u01/supplementary/sup_02.html

aluminum = 0.896 kJ per kg per Kelvin
copper = 0.383 kJ per kg per Kelvin

It appears that aluminum is the clear winner at storing heat, however the table shows specific heat in terms of mass. We know that copper is a lot denser than aluminum. We are probably interested in comparing a given volume of aluminum vs the same volume of copper, so gotta convert the table to volume instead of mass.

Here is a table of densities for various materials:

http://www.mcelwee.net/html/densities_of_various_materials.html

aluminum = 0.002643 kg/cm^3
copper = 0.0089 kg/cm^3

multiplying by these factors,

specific heat of aluminum = 0.00237 kJ per cm^3 per Kelvin
specific heat of copper = 0.00341 kJ per cm^3 per Kelvin

so by mass, aluminum is better at absorbing heat. But by volume, copper is the clear winner. Given that we are usually space-constrained inside a computer case, copper would be both better at absorbing heat and dissipating it (see my previous post). Sorry I had to run you through this calculator exercise, but i couldn't google up any webpage showing specific heat in terms of volume.

 

[Bluefinger]

Also, aluminium doesn't surrender the heat as easily to air because of its high heat capacity

This is certainly interesting! If you ask most people in the hardware and overclocking game, we've been led to believe that it's the other way round. I certainly thought that it was the other way round (which answered the question of why many heatsinks have a copper base with aluminium fins - issues of mass aside).

I'll definitely have to read up on it more.

Aluminium is used because it stores a lot of energy without increasing in temperature much, so its easier to keep cool despite it being rather deficient in transferring the heat. Thats why copper is used for bases and heatpipes, and why heat pipes are used so much in high quality heatsinks, to overcome the deficiency in heat transfer in aluminium fins. The copper transfers the heat, and the aluminium stores the heat away from the CPU, thus keeping the CPU cool.

 

[shadowjagans]

that must be one of the best explanations ever... except im too dumb to understand most of it... O_O

 

[rodney_ws]

shadowjagans, stop flooding the forums with your worthless questions. and then you say stuff like "what are you chemistry teachers?". dude, get lost. why do you people even respond to his stufff? i did just now cause i'm pretty tired of his stuff and one other person i know for sure is too... don't you have your avatar or whatever it is already? geez...

Amen. Enough BS questions from this guy... he's definitely exceeded his quota.

 

[Jackalope73]

"Sir, your knowledge of the relevant chemistry is inadequate. I shall attempt to remedy the situation. " PhyberOptik

Slur, your knowledge of being a jerk off is irrelevant, but certainly adequate.
Congratulations you are a fool with a search engine! You don't even know what a PCB is in a computer forum. It is commonly used as an abbreviation for PRINTED CIRCUIT BOARD a substrate for the video card. But since you are concerned with chemistry it's root is in older substrates that were made from Polychlorinated Biphenyl. Also Mr. Search engine, copper, was more precious than gold through out history. In the Americas the Native Americans used to sell gold for copper and much of the territories and trading was done in precious copper. This is something you would learn if you actually went to college and had to take anthropology. But chemistry man lets stick to the question at hand which is METALURGY not chemistry. The process of manufacture and rare earth elemental formations and heavy isotopes of a material alter its properties. This was covered in my disclaimer of high nobles and alloys (iridium), but maybe you can't read. As to the silicon conductors on the shuttle that is industrially produced silicon that sub-atomically is quite different than what occurs naturally on earth anyway. Maybe we could argue the thermo dynamic properties of Jupiter’s Hydrogen metal core? Listen up here, synthetic and bort diamonds are not special they are assembling a molecular structure conducive to heat and electrical transfer using a delocalized plane of free electrons to "glide" energy across excited electrons. Hey you forgot nano tubes! You should spend more time actually reading and trying to comprehend, instead of crawling the internet in an attempt at trying to look cool in front of people who have better things to do. Hey I got something for you it's your AXX. Now if you'll excuse me I have better things to do than hand it to you all day. P.S. Nice graph Search Engine, Maybe you were Lord Kelvin in your last life .
P.P.S No Search Engines were harmed in the production of this response.

Yeah that's right you call me sir, and next time, I'll send you a bill.