Thermal conductivity air to metal

[themaea]

Let's say you have a small metal case that is 3 inches by 3 inches versus 2 inches by 2 inches. Inside at center is a heat generating source (let's say a small processor chip). How efficient is the transfer of heat from the processor to the air... and from air to metal between the 2 different size metal box? How do you compute for this? I know directly gluing the metal to the processor would be more efficient.. but I want to know the thermal transfer if air is in the middle between them where the heat sink is not directly connected to the source but has air medium between them like that occurring in many enclosure housing electronic devices with processor chip inside (like in cctv camera). Thank you.

 

[Dr. Courtney]

Convection and radiation are playing important roles, and one cannot consider the mechanism to be exclusively conduction.

 

[themaea]

Convection and radiation are playing important roles, and one cannot consider the mechanism to be exclusively conduction.

Here's an actual setup I tried. The square metal case is 1 inch high and 1.6 inch in the 4 of its sides. There is a small camera processor (13mm x 13mm) at bottom with temperature of 70 degree Celsius.. I opened the top metal cover to test if the temperature just rise straight up to outside from convection but what really happened is the sides of the metal case are still hot (even if the top cover is open). So what really goes on microscopically? When the processor gets hot.. it conducts heat to the air.. but since the top cover is opened.. why is its least resistance is not the open air above (and outside) but the metal at the sides? Before doing the experiment. I was guessing the metal sides would remain cold since the heat from the hot cpu at bottom would just rise up. You also mentioned about radiation. In this experiment, how many percentage (approximately) of the heat from the cpu goes to the metal at the sides in form of radiation and in form of conduction and convection? The cpu and metal case are all colored black so let's assume an emissivity of 0.95. Thank you.

 

[Dr. Courtney]

The split between radiative, convective, and conductive cooling are notoriously hard to calculate with confidence.

Most designers use several prototypes and approach the problem experimentally, adding heat sinks, vents, and fans if needed.

 

[themaea]

or let's just use a very simple setup... just two metals let's say one inch apart.. when one is heated 70 degree celsius.. is there significant radiative heat transfer from one metal to the other? i thought IR radiative heat transfer is only important for the sun-earth heat transfer bec the sun is great concentrated source of ir

 

[themaea]

http://imageshack.com/a/img913/9240/RH8zTD.jpg
http://imageshack.com/a/img537/8577/NIdMHw.jpg
http://imageshack.com/a/img537/563/m7ER9p.jpg

The above picture is a small pinhole camera that you can buy for $30. The processor inside is so tiny no heatsink can't be put inside. The black aluminum metal case is sized 1.6 inches at the sides and 1 inch height. Here's the mystery. When put on table such that the top cover is open. The thermograph shows the processor to be 71.8 Celsius.. but the metal case is measuring about 44 degrees Celsius. Convection should make the heat go straight up without touching the sides. Conduction through air is said to be poor. Is it really possible the IR radiation from the processor is what heat up the metal case? This is just for theoretical understanding and not about putting heatsink to the cheap camera. I thought only the sun IR radiation can bestow heat to earth. I'd like to understand the role of IR radiation in terrestrial object.. usually how much does it contribute to heat versus conduction and convection in daily objects.. can you give another example? Thank you.

 

[themaea]

notice the black metal case is like a blackbody (because it's painted black) and the chip is black too.. so the question is.. would there be significant transfer of heat from the chip to the metal case by plain ir radiation? I think it should, isn't it. because I can't see how conduction by air can increase the temperature of the metal case by 10 degrees Celsius, could it?

 

[gmax137]

The methods for calculating the things you're looking for are explained in any number of textbooks on heat transfer. You should read one of these books and then ask questions on the parts you don't understand. I like this one,
https://www.amazon.com/dp/0070479801/?tag=pfamazon01-20

 

[themaea]

The methods for calculating the things you're looking for are explained in any number of textbooks on heat transfer. You should read one of these books and then ask questions on the parts you don't understand. I like this one,
https://www.amazon.com/dp/0070479801/?tag=pfamazon01-20

I just want to know whether the metal case heating up is caused by convection or ir radiation.. what is usually the case.. that's all.

 

[CWatters]

Convection tends to dominate many similar configurations. You didn't mention if the plates are vertical or horizontal. The aspect ratio of the problem may also matter. If the plates are very large in relation to the gap between them (like a double glazed window) then conduction might dominate over convection. Exactly where that cross over is I couldn't say.