Noctua prototype passive cooler can handle 120W in fanless case

[HEXUS]

1.5kg finned and heatpipe threaded block can thus cope with an Intel Core i9-9900K.

Read more.

[spacein_vader]

Coolers that can work in a fan free case come up every now and then but they still have the same flaw: they weigh almost as much as said case.

Might be ok for a desktop case but j wouldn't want one hanging off a mobo in a tower no matter how robust the bracket was.

If you really need that kind of silence those cases that act as the heatsink themselves and connect to the CPU via heat pipe seem a much better bet.

[Tunnah]

Bit misleading/disingenuous to use the 9900k as an example considering intel's dodgy use of TDP. Doesn't the 9900k use 200w or something under load ?

[tinbin]

good coolers not cheap

[Corky34]

No announcement of a 140mm Sterox fan.

[Xlucine]

1.5mm fins? That seems unusually thick, normally a low convective heat transfer coefficient means you can get away with thinner thins - there's no way they'd have a problem with fin efficiency at a normal fin thickness if their forced air coolers work fine

[aidanjt]

Bit misleading/disingenuous to use the 9900k as an example considering intel's dodgy use of TDP. Doesn't the 9900k use 200w or something under load ?

120W still leaves comfortable above-base clock TDP margin, even if it doesn't stay pinned at 5GHz under load. If your application is more latency/bursty in nature than sustained, it'd be perfectly fine to run an i9-9900K under this.

[aidanjt]

1.5mm fins? That seems unusually thick, normally a low convective heat transfer coefficient means you can get away with thinner thins - there's no way they'd have a problem with fin efficiency at a normal fin thickness if their forced air coolers work fine

Yeah, when you're actively cooling thinner, tightly packed fins is more optimal. In the case of passive cooling, more thermal mass and more spaced out fins is better.

[Xlucine]

Yeah, when you're actively cooling thinner, tightly packed fins is more optimal. In the case of passive cooling, more thermal mass and more spaced out fins is better.

I agree on the spacing, but thicker fins are desirable under active cooling - the higher convective heat transfer means more heat flow, so you need the reduction in thermal resistance across the fin to maintain good efficiency (or to look at it another way, a thicker fin boosts effectiveness). With the lower convective transfer of free convection, a thinner fin will be fine (if the efficiency is already high, then a thicker fin won't shed more heat). I can see your point about thermal mass, racing-to-idle makes a lot of sense with passive cooling

[RonanH]

1.5Kg hanging off your CPU socket seems like an awful lot. If you install this puppy on a vertical motherboard, be sure not to bump the case too much!

[aidanjt]

but thicker fins are desirable under active cooling - the higher convective heat transfer means more heat flow, so you need the reduction in thermal resistance across the fin to maintain good efficiency

That's what the fan is for. Thinner fins means you can pack more of them together, and it takes less air pushing across each fin to cool them off, ensuring that there's a wider temperature gradient to keep heat moving around. That's why active air coolers are designed the way they are.

[aidanjt]

1.5Kg hanging off your CPU socket seems like an awful lot. If you install this puppy on a vertical motherboard, be sure not to bump the case too much!

That's not nearly as much of a problem with modern cooler mounting mechanisms (like Noctua's SecuFirm2) that clamps the motherboard from both ends of the socket and distributes the load. I still wouldn't go dropping it down a flight of stairs, of course.

[Xlucine]

That's what the fan is for. Thinner fins means you can pack more of them together, and it takes less air pushing across each fin to cool them off, ensuring that there's a wider temperature gradient to keep heat moving around. That's why active air coolers are designed the way they are.

The fan gives airflow, which leads to the higher convective transfer coefficient. This leads to more heat flowing from the heatpipe to the extremities of the fins, and so a greater temperature gradient across the fin - which means it's not working as efficiently as it could be, as the heat loss from the fin extremities is not as high as it could be. If you up the convective transfer coefficient then you need thicker fins (decreasing thermal resistance across the fin, or you could see it as reducing heat flux) to keep the fin efficiency, except in this cooler noctua have decided to do the opposite.

In the steady state (i.e. where TDP limits for the cooler are calculated, as everyone uses prime95 or similar to test heatsinks these days) then the thermal inertia of the fins doesn't matter, by definition, so the amount of air movement required to carry the heat away from a fin scales with surface area (as that's what the fin effectiveness can be approximated by, and surface area of a given fin is effectively constant for any sensible thickness).