OT: CPU heatsink "heat pipes"

Technology

It transports heat, and does so by fluid transport.

Hollow, like a pipe, hence the name.

It doesn't conduct heat the way a solid does, it transports it via fluid flow, and is orders of magnitude better than copper.

Well, there are some similarities but fundamental differences. A heatpipe uses the heat itself, combined with either gravity or capillary action, as the transport mechanism and, so, doesn't require a mechanical pump. It also operates on phase change whereas water boiling in a car cooling system is bad news.

Only half right. The last couple of steps in manufacturing a heat-pipe involve compeletely evacuating it, ten distlling in the desired amount of theworking fluid - usually water - then sealing off the tube.

The only vapour in a good heat pipe has come from the evaporation of the working fluid, so that the "boiling point" of the working fluid is always pretty much the temperature of the coolest point along the heat pipe. You might get some pressure drop along the heat pipe as the vapour flows from the hot spot from which it is boiling off to the coolest point where it condenses, but this doesn't usually amount to much.

A bad heat pipe - which hasn't been properly evacuated - doesn't work too well at temperatures below those required to make the vapour presure of the working fluid a good bit higher than the partical pressure of the residual (non-condensing) gases.

Is it hype, or is it technology? I'm wondering what is the usefulness of "heat pipes" in CPU heatsinks?

Seems to me that a "heat pipe" is not a pipe in the ordinary sense of the word, so it isn't going to transport anything the way a pipe does. I guess it is a solid rod of copper that conducts heat about the same as any other shaped copper conducts heat. Would its use have to do with the shape of the heatsink, perhaps allowing the heatsink to fit into a narrower space or allowing an easier/cheaper connection between the CPU contact area and the heatsink fins or whatever?

Thank you.

Heat pipes are not solid. They have a liquid in them that is constantly changing from a liquid to a gas and back. As it does so it transfers the heat from the end connected to the cpu to the end at the cooling fins. They are most commonly used in laptops so the heatsink and fan can be thin and located in a convenient location.

Technology.

I'm wondering what is the

Works on the same principle as a radiator on a car.

Radiator fluid passes through a car engine. In order for that to be the same principle, there would have to be fluid or at least air passing through the CPU. As far as I know, there are not even any channels through the rest of the heatsink.

Sounds like a troll to me. For that to make any sense, you would have to live in the United Kingdom? Tony Blair is at least as gung ho as George Bush Jr. Before Iraq, Tony Blair teamed up with Bill Clinton for the last invasion of a sovereign nation.

newssvr11.news.prodigy.com!newsdbm04.news.prodigy.com!newsdst01.news.prodigy.com!newsmst01b.news.prodigy.com!prodigy.com!newscon02.news.prodigy.com!prodigy.net!nx01.iad01.newshosting.com!newshosting.com!newsfeed.icl.net!newsfeed.fjserv.net!news.tele.dk!news.tele.dk!small.news.tele.dk!fu-berlin.de!uni-berlin.de!individual.net!not-for-mail

sci.electronics.design:614444

Plot the heat drop thoughout a heatsink, and you get a few C loss just next to the heat source. If you can spread this into a bar of effectively infinite conductivity, as a heat pipe can do, then you lose that several C.

Only for the "heatsink" part of the system; the heat pipe element is something completely different.

I'd stick to driving trucks if I were you.

...

It's been very informative, thanks to all of the replies. The most important information seems to be that the heat pipe is exceptionally conductive.

Cheers! Rich

No. In a car engine/radiator, the working fluid is the liquid. The heat transfer medium is the liquid. The mechanism is conduction; the "radiator" actually cools the fluid mostly by conduction. (to the air, of course.)

In a heat pipe, the working fluid is vapor. The working fluid boils at the hot end and condenses at the cold end, then wicks back to the hot end as liquid. The mechanism of heat transfer is phase change.

Hope This Helps! Rich

Fundamentally an enclosed tube with a bit of fluid in it . Fluid boils at one end, condenses at the other, then travels back through gravity or capillary action along the tube wall.

The vapour can carry heat much faster than solid copper can conduct it.

The tube often has a partial vacuum so the fluid boils at around the temperature the cooled device is intended to run at.

Do you know what convection is? The first car engines didn't have waterpumps...

Really? Care to explain how? It's using a moving fluid to transmit heat away then passing the liquid via pipes through fins to dissipate the heat. I wonder how a car cooling system works...oh yeah, the same.

I'd stick to working in Mcdonalds if I were you.

Wait-- water? Vacuum? These are *atrocious* refrigerants! Wouldn't it have to be a common refrigerating gas, such as perhaps a medium freon, or butane? Under sufficient pressure of course. Pressure then varies with temperature, as with any compressed liquified gas.

Tim

-- Deep Fryer: a very philosophical monk. Website:

A heatpipe is filled with a wick to transport fluid to the hot end, from the cold end,at the hot end the fluid evaporates, goes to the cold end ,condenses ,is sucked up by the wick and fed back to the hot side. Evaporation takes a huge amount of heat,at a very small change of temperature,(as in boiling water),so you can get a lot of heat out of a small place, and bring it to a place were there is room to get rid of it. In one type of laptop I had open,they used the whole area below the keyboard to get rid of the heat. So the pipe really ships gas in one direction , and fluid in the other(with the help of the wick).

That entry was wrong in a least one respect - open space inside the heat pipe is not filled with air, which would make the heat pump much less effective, but is filled with working fluid vapour, and is free from any other vapour. The vapour flow from the hot end to cold end goes much faster if there aren't any non-codensing gas molecules to slow thing up.

I've corrected it - I hope the correction sticks.

When my ex-employers went over to heat pipes to cool their Peltier junction, and needed low thermal resistance at close to room temperature, they had to build a test rig to check out every heat pipe that came in, and were rejecting one in three - sent back to be opened, pumped down to vacuum and reloaded with water - until the manufacturer got their act together.