convection

[AilSnail]

Will different materials have different convection abilities?

Specifically alu, copper, silver, and various coatings and anodizing.

Or do I only need to consider the conductive properties of the material and the geometry?

 

[Doug S]

[ QUOTE ]
AilSnail said:
Will different materials have different convection abilities?



[/ QUOTE ]

I believe that the answer is no. Surface *texture* however can affect the convection rates. I cannot tell you much more beyond the fact that the thermal resistance at the surface to air interface is referred to as the "film coefficient".

 

[Kurtos]

My heat transfer is a little hazy but I believe convection is dependant on the fluid properties and not the material. Geometry plays a part in convection as was alluded to above in texture.

 

[kakster]

So did i waste my money buying a silver plated copper heatsink for my CPU? I was thinking a heatsink for a 5 watter might be made with these materials for better thermal management.

 

[Kurtos]

Copper and silver have excellent thermal conductivities. So for channeling the heat out of the emitter they are pretty good. However you must remove that energy from the heat sink.

Assuming air is the working fluid you won't loose much due to conduction due to the low k(thermal conductivity) value of air. Energy must be lost then by convection or radiation.

Couldn't say if your heatsink is a lemon or a deal. Depends on fluid interaction, geometry and a few other things. Generally greater surface area is better (to a limit) and forced vs free convection is also better.

 

[Unicorn]

Copper will draw heat from the CPU better than aluminum. You will still need a good fan and case ventilation, possibly even more so. Also the design of the heatsink plays an important part in it's ability to transfer that heat to the air. Copper draws heat off better, but I think it also holds it more, so a good HSF is a must.

 

[Lux Luthor]

If I'm not mistaken, the term "convection" refers to both transfer of thermal energy due to mass transfer (via motion of the fluid), as well as to transfer of thermal energy due to conduction at the solid/fluid interface.

It could be that different microscopic surface properties of different materials could affect the rate of conduction at the solid/fluid interface, but I'm not sure of this.

Regardless, though, it seems reasonable that if you have a slab of solid, and if the solid has a higher thermal conductivity, then the temperature gradient across the slab would be smaller, and this would give you a higher temperature at the solid/fluid interface. This should help to promote convection, although the effect might be small.

 

[xrayzebra]

Any consensus on the copper vs. aluminum question? I thought that aluminum was actually a better heatsink in terms of drawing heat away and dissipating it. Could be I just swallowed some propaganda from someone trying to sell cheaper materials as beter materials back in the day when I was in the HVAC business.

Even in copper is better, aluminum may be a better choice when weight is a factor. In either case, the thermal connection between heat sink and the part you are trying to cool is certainly key. Sheer mass of the heat sink is okay for short bursts of heat, but dissipation is important for sustained heat. Fins, air movement, etc.

 

[NightStorm]

Copper has better thermal transmission (conductivity) capabilities and aluminum has better thermal volume. In other words copper has the ability to move the phonons quickly from the heat source to another point but aluminum has a better ability to absorb that heat.

Thermal Characteristics

Dan

 

[AilSnail]

"aluminum has a better ability to absorb that heat."

yes, per weight unit.
no, per volumetric unit.

 

[Chief_Wiggum]

^^^that means for the same size piece of material, copper conducts heat from the source faster and "absorbs" heat better...

In flashlight use, I don't think the weight matters so much. Anyway, it's probably all a wash, since none of us can really achieve perfect thermal conductivity from the emitter to the heat sink anyway.

 

[kakster]

But we can dream /ubbthreads/images/graemlins/smile.gif

 

[mattheww50]

Cooling is a double edged sword. The specific heat (energy required to raise the temperature of a fixed mass of the material)
aluminium is .212 (BTU/pound degree)
Copper is .094
brass is .090
Gold is .031
silver .056


by contrast water is 1, ice is .48.

Electrical conductivity is a good surrogate for thermal conductivity in metals,in fact in some cases the mechanism is the same (Peltier effect devices
carry heat on electrons).

Having a material with low thermal miss is not necessarily bad, since the energy you radiate goes up as temperature to the 4th power..

As result at least for radiating the heat away, a low thermal mass has some attractions. Yes it gets hot, it also cools off rapidly, and reaches
steady state condition more rapidly.

And there is no such animal as perfect thermal conductivity unless you are talking about temperatures approaching absolute zero and weird substances like superfluid liquid helium.

 

[AilSnail]

[ QUOTE ]
Having a material with low thermal miss is not necessarily bad, since the energy you radiate goes up as temperature to the 4th power..

As result at least for radiating the heat away, a low thermal mass has some attractions. Yes it gets hot, it also cools off rapidly, and reaches
steady state condition more rapidly.


[/ QUOTE ]

I don't understand this reasoning. Why is this good?

 

[Lux Luthor]

He's saying that if the thermal mass is low, the outer surface of the heat sink would be hotter, which would help promote radiational cooling.

The rate of radiational cooling is given by the Stefan-Boltzmann law and increases as temperaure to the 4th power (in absolute units). So a higher temperature at the outer surface corresponds to a higher rate of radiational cooling.

 

[AilSnail]

If two heat sinks are the same geometry, the same conduction (w/m.K) but different thermal capacity (joule/K).
Will they not have the same surface temperature when they have reached steady state?
Will not the one with greater capacity be colder where the heat source is until they have reached steady state hot?

An analogy: Two vertical cylinders of different circumference has splits from top to bottom where the water can seep out. The water is pumped in from the bottom, symbolising a LED generating heat. The pressure at the bottom symbolises temperature at the led.
As the water level quickly rise in the small cylinder, the pressure rise and more water is let through the split. the thicker cylinder reaches steady state later, and it is not until the water is turned off that the water level is lower in the small cylinder.