The thermal conductivity λ (lambda, GER) or κ (kappa, INT) of copper is ~ 380 W/(m·K), the one of oil 0,15 W/(m·K) and of water 0,5562 W/(m·K).
Liquids don't have great conductivity, but they make up for that with convection (i.e. they move). This is why water cooling is the most efficient way to cool your CPU (in that case, forced convection with a pump). The catch with liquid cooling would be managing the refractive index of the liquid (and avoiding bubbles)
The convection and heat effects on refractive index would be just like those on using a long telephoto lens on a hot summer's day with the sun baking the ground!
Liquids are used to cool CPUs because those are better than air (!!!). But worse than solid metals.
Yes, front side liquid cooling would be a challenge. If a proper seal could be made, backside liquid cooling is a possibility. Just think, a micro-miniature CPU cooler, pump and all
. It could be very lightweight and would not be subject to dust like a fan. The coolant hoses would likely be more flexible than the graphene strips in the patent.Unless you have a river handy, the heat has to be eventually dissipated to air. Some metals have great conductivity, but typically, the closer you can get a moving liquid to the heat source (i.e. the less metal in the way), the better the result will be. Your "passive" cooled CPU is still air convection cooled. The massive blob of metal is just an extended heat spreader. CPU "air coolers" actually use vapor phase cooling, which can be fully as efficient as liquid cooling, but for CPUs, water cooling has more capacity, simply because you can get more surface area on a radiator (for that final connection to the air) than will fit on top of the CPU "air" cooler. I, too, have cooled many power Semis over the years along with some engines and other things that make heat and noise.
I don’t understand why you‘re trying to explain air cooling to me, when I already explained it.
I was responding to your statement "Liquids are used to cool CPUs because those are better than air (!!!). But worse than solid metals". I was simply pointing out that moving (or boiling as in vapor phase) liquids can be more efficient at heat transfer than the best of metals. You used your "passively" cooled CPU as an example, but you would have a hard time "passively" (i.e. conductively) cooling an RTX 5090, no matter how big the block of copper. Vapor phase or liquid are the only practical means that will move 600 watts from an area that small while keeping the temp under 100 deg. So, yes I was talking about heat transfer (and disagreeing with your statement).
The better the longer the heat pipe is, I know.
Graphene is useful in this context because of its anisotropic thermal behavior (over 100 times more lateral heat flow than vertical). It sort of resembles a solid-state heat pipe. Flexible heat pipes are possible but not common, but liquid lines can be very flexible and that was where I started this discussion.
If you know so much better than Canon developers (read the patent) why not apply at their R&D?
I never suggested that I know better than Canon developers. I just suggested that liquid cooling could a viable possibility (with some obvious hurdles to be overcome). Graphene is a unique material and clearly an interesting choice for the application. I suspect Canon's engineers would avoid the complexity of liquid cooling if at all possible and that makes tons of sense. You do seem a bit cranky about your rather absolutist statement being corrected. Sorry if I hurt your feelings.
But the camera sealing must be very good then. Not to avoid dust/moisture getting in, but
That would be HeBut the camera sealing must be very good then. Not to avoid dust/moisture getting in, but H2 from getting out.
. It is even sneakier than H2.
I didn‘t get hurt.
Your‘re partly right: just He.That would be He2
