DIY Diode housings

[keeperx]

well buffo i think we are talking about heat a little differently.. Im reffering to heat sink saturation, and i do not believe in it..

http://www.ocforums.com/showthread.php?t=29041

neither does he..

[mixedgas]

heat sinks are rated in degrees C per watt, if your heatsink in free air has less capacity then say twice the device heat output, it will start to heat up and head towards the attached device melting if there is enough energy supplied. Critical cooling, ie the heat sink has exactly the amount of cooling needed, is a bad thing.

If there is even more energy above the rated heatsink capacity, the heatsink will head towards its melting point. The saturation does occur, Saturation occurs at the heatsink melting as the molten aluminum will not rise above its melting point just like boiling water.




Now if you have a longer path to the last heat removal medium in the chain, which is the air, the device will stay hotter then a finned heat sink that has short path.

That is solid state Physics, and the last time I checked God doesn't allow those rules to be bent.

Steve

[buffo]

Wow - until I checked that over-clocker forum you linked to, I had no idea that there were so many otherwise intelligent people that were utterly clueless when it comes to basic thermodynamics. Heat sink saturation? You've got to be kidding me!

As Steve pointed out, a heat sink never truly saturates. It might *MELT*, but at that point whatever you were trying to cool has long since stopped working.

A heat sink can loose heat 3 ways: Conduction (to the baseplate), convection (to the air around it), and radiation. And all other things being equal, a higher temperature will give you a greater rate of heat transfer for all three processes. Thus there is no such thing as "saturation", because as the temperature rises, so does the rate of heat transfer, though in practice the temperature rise can be quite dramatic.

I suspect that people started using the term saturation to refer to a condition where you are adding more heat to the heat sink than it can remove *at normal operating temperatures*. Once you reach this point, the temperature of the entire system begins to rise, and while it's not "saturated", it does lead to a condition where the temperature can quickly rise beyond the rated maximums for semiconductor devices. So yeah, it's still a problem, but using the term "saturation" is misleading at best.

Adam

PS: Steve, don't forget that even at the melting point you can still have an increasing rate of heat removal . You're forgetting the latent heat of fusion that is being absorbed as the heat sink melts! Of course, once it melts, the surface area will be lower (because it's now in a puddle), so once you are all liquid your heat transfer rate will be lower. But at that point the liquid temperature will begin to rise, raising the transfer rate once again...

[Doc]

How many megawatts is this diode? Will it burst a balloon with a black dot on it?

[danielbriggs]

Wow... (snip)

4 words:
Newton's Law of Cooling

[allthatwhichis]

That's funny!

[keeperx]

that was kinda my point..

the previous poster said that the entire block has to heat up before it reaches its cooling point.. that did not make any sense to me..

the block is constantly radiating heat, as soon at it is above room temp, no matter how thick the block is.

as long as you can radiate/convect/conduct enough heat away from the block your good. Fins help with convection and radiation. but with a good baseplate, like in my setup, everything works quite well. The point is that the aixiz module is more than good enough for most setups as is.

[Laserman532]

i dont know what diodes you are running but I am running LOC 660 diodes at 450ma and they dont even get hot, heck...i would add they dont even get warm (to the touch) i use my clamshell heatsink and no puddles of molten aluminum................yet

see two sizes....no molten aluminum...............

[rfourt]

didnt any one see these

http://www.z-bolt.com/order/Product240

[drlava]

didnt any one see these

http://www.z-bolt.com/order/Product240

"you are unable to view this portion of the site because your browser will not accept cookies from nexternal"

hmm?