[iggybaseball]

I have read some old cooling articles and ran across using a heat pipe. how does this work>

[mx]

Originally posted by Diggrr
Forgive the exagerated scale, and the quickie rendering, but here goes.
It works much like air conditioning without moving parts.
The phase change liquid is evaporated at the hot side. The vapor collects on the cold side and returns to liquid (also releasing it's heat to the cold side). The liquid is then carried back to the hot side with either cappilary tubing or a 'wicking' material to start the process over again.

Heat pipes are usually too expensive, and perform too low for cpu use. They also tend to need a gravity assistance to keep the liquid material on the hot side. There are hundreds of variations and configurations, the one shown is an oversimplified version.
I found many by just using a Google search. Most companies have illustrations or pictures to give you an idea of what's possible.
The technology hasn't quite come far enough for our use.

Is the phase change liquid cooler than ambient temps when it reaches it's boiling point......and if not I can't see how this would be much more efective than air cooling.

[mx]

Originally posted by Diggrr
The two that I've actually seen would boil below room temp but were under internal pressure to stop that from happening. Sort of like the butane in a lighter. Balancing the pressure allows you to pick a boiling point.

Their main use (from two that I've seen) is cooling where you can't put a fan. Electromagnetic field measuring, and Seizmic measuring. On one the fan gives off way too much EMF for the equipment to see through, and the other vibrates far too much.
In both cases even a stock coolermaster would work better at cooling, but the fan interferes with what the equipment is doing, so they found a fan-free alternative.

They would be the ultimate cooler if they could handle 150+ watts of heat fast enough to keep our 'bird' from frying. For now, they cannot. (unless of course someone made a huge and expensive one. I'd hate to see the size of the cold side radiator that could cool one without any fans.)

I just visited a few sites that sell "flush mount solutions" that mount externally on the rear of the case, drawing air in to cool entire system, hot air is expeled to the outside"a.c. solution"......the air is cooled below ambient temps, they look like very simple devices........I think this requires further investigation!

[nihili]

The main advantage a heat pipe of this sort would have is that it uses convection to carry the heat away from the source. You still have to dissipate the heat somehow, but convection allows you to distribute the heat over a larger area.

In application to computers one problem is that this sort of heat pipe only operates vertically. Thus it's useless for tower cases.


nihili

[RycheX]

"The technology hasn't quite come far enough for our use"

NOT true. Heat pipe technology far surpasses any cooling methods used in PCs today. The issue is more like "the technology has not been properly applied for our use." Heat pipes are amazing devices and the technology shows alot of promise for our sector, someone just needs the $$ and the time to port it over to a form that will work better with PCs. There are heatpipes that have been designed and that are in use that move alot more heat than any processor currently puts out. I spent more than a year in engineering school studying heat pipes and worked with others on their implementation for cooling satellites. I believe the biggest issues with using them for modern computing is going to be overcoming the high cost for use on relatively cheap PC systems.

"The main advantage a heat pipe of this sort would have is that it uses convection to carry the heat away from the source"

Actually the main mode of heat transfer in a heat pipe is accomplished through the phase change of the working fluid.

[mx]

Originally posted by nihili
The main advantage a heat pipe of this sort would have is that it uses convection to carry the heat away from the source. You still have to dissipate the heat somehow, but convection allows you to distribute the heat over a larger area.

In application to computers one problem is that this sort of heat pipe only operates vertically. Thus it's useless for tower cases.


nihili

Not true I have just viewed heat pipes designed for vertical processor aplications, with heatsink and fan on the condensor side!.

The cost seems to be the main factor here!

[nihili]

Originally posted by RycheX
"The main advantage a heat pipe of this sort would have is that it uses convection to carry the heat away from the source"

Actually the main mode of heat transfer in a heat pipe is accomplished through the phase change of the working fluid.

Phase change is a method of heat absorption, not of heat transfer. The phase change allows more heat to be absorbed into the pipe, but convection is the method that transfers the heat from one end of the pipe to the other. Thus it is convection that is responsible for carrying the heat away from the source for the most part.

nihili

[nihili]

Originally posted by mx


Not true I have just viewed heat pipes designed for vertical processor aplications, with heatsink and fan on the condensor side!.

The cost seems to be the main factor here!

That's why I said "of this sort". There are heat pipes that will work in horizontal positions, but they are different than the one drawn. I'm guessing you looked at the pulsating heat pipe in the ZEN cooler. The pulsating heat pipe uses different principles to move the heat from the hot to the cold side.

Though as I recall, on the Zen the heat sink and fan run all along the pipe rather than just on the condenser side. Do you happen to have a link to what you saw?

nihili

[mx]

Originally posted by nihili


That's why I said "of this sort". There are heat pipes that will work in horizontal positions, but they are different than the one drawn. I'm guessing you looked at the pulsating heat pipe in the ZEN cooler. The pulsating heat pipe uses different principles to move the heat from the hot to the cold side.

Though as I recall, on the Zen the heat sink and fan run all along the pipe rather than just on the condenser side. Do you happen to have a link to what you saw?

nihili

I could'nt find both sites that I viewed, but here are a couple of links.
http://www.norenproducts.com/Heat_Pi...ion_Notes.html
http://www.norenproducts.com/Informa...eat_Pipes.html

[RycheX]

"Phase change is a method of heat absorption, not of heat transfer. The phase change allows more heat to be absorbed into the pipe, but convection is the method that transfers the heat from one end of the pipe to the other. Thus it is convection that is responsible for carrying the heat away from the source for the most part. "

Convection is the transfer of energy (heat) from a solid to a moving fluid when they are at different temperatures. This is NOT what is happening in a heat pipe.

In a heat pipe, heat enters into the pipe itself by conduction. From the pipe material, the heat once again conducts into the fluid. The additional heat that was just conducted into the fluid, adds to the total energy in the fluid and when that energy is sufficiently high, the fluid vaporizes. So at the 'hot' end of the heat pipe there is a large concentration of vapor. This causes a pressure differential between the 'hot' and 'cold' ends of the heat pipe. It is because of this pressure differential that the vapor then moves to the 'cold' end of the heat pipe. The vapor then comes in contact with the heat pipe material at the cold end of the pipe, and conducts heat into pipe material, thus lowering its total energy to the point that it condenses on the pipe wall.

Now 'cold' end of the heat pipe may be cooled itself on the outside by convection, but convection is not the manner that heat is being moved within the pipe itself.

[nihili]

Originally posted by RycheX
"Phase change is a method of heat absorption, not of heat transfer. The phase change allows more heat to be absorbed into the pipe, but convection is the method that transfers the heat from one end of the pipe to the other. Thus it is convection that is responsible for carrying the heat away from the source for the most part. "

Convection is the transfer of energy (heat) from a solid to a moving fluid when they are at different temperatures. This is NOT what is happening in a heat pipe.

In a heat pipe, heat enters into the pipe itself by conduction. From the pipe material, the heat once again conducts into the fluid. The additional heat that was just conducted into the fluid, adds to the total energy in the fluid and when that energy is sufficiently high, the fluid vaporizes. So at the 'hot' end of the heat pipe there is a large concentration of vapor. This causes a pressure differential between the 'hot' and 'cold' ends of the heat pipe. It is because of this pressure differential that the vapor then moves to the 'cold' end of the heat pipe. The vapor then comes in contact with the heat pipe material at the cold end of the pipe, and conducts heat into pipe material, thus lowering its total energy to the point that it condenses on the pipe wall.

Now 'cold' end of the heat pipe may be cooled itself on the outside by convection, but convection is not the manner that heat is being moved within the pipe itself.

I believe we are in substantial agreement over what's going on in the pipe and are at the point of quibbling about the proper definition of "convection". If you're in to quibbling I'd be happy to oblige, as I'm an old quibbler from way back. A brief survey of relevant lexical and scientific/engineering sites seems to support both uses however, so I don't know as we'll come to a definitive answer.

nihili

[nihili]

Originally posted by mx


I could'nt find both sites that I viewed, but here are a couple of links.
http://www.norenproducts.com/Heat_Pi...ion_Notes.html
http://www.norenproducts.com/Informa...eat_Pipes.html

Thanks, those were very informative. I hadn't been aware of the internal wicking systems. That's a good idea that would certainly allow for horizontal application. I guess now I know what that little copper pipe in my laptop is.

nihili

[RycheX]

nihili, while I may admire your great ability to 'quibble', there is only one 'proper' definition of convection, and it is exactly what I stated. I'm not trying to take credit for any great knowledge I have, I'm simply stating what is and has been known for a very long time. If you'd like to further research the true meaning of convection, I'd highly suggest getting your hands on an engineering heat transfer text such as "Fundamentals of Heat and Mass Transfer" by Incropera and DeWitt. Relying upon websites for in-depth, and accurate heat transfer engineering understanding is simply not possible because they are full of the random musings of many people that have not thoroughly studied heat transfer. Nor do websites have the space to properly address the theory behind heat transfer or it's practical applications.

While it can be argued that there are small elements of convection within a heat pipe, the fact is simple that it is not the means of the vast majority of the heat transfer.

[mx]

Originally posted by nihili


Thanks, those were very informative. I hadn't been aware of the internal wicking systems. That's a good idea that would certainly allow for horizontal application. I guess now I know what that little copper pipe in my laptop is.

nihili

I don't think we should easily dismiss this tehnology for possible use in overclocking.
The device itself is very simple, but we need to take into account other variables such as the proper degree of evacuation for various refrigerants with a boiling point below..... lets say 30C, the amount of refrigerant in relation to volume to achieve desired pressure.
The steep price tags really does'nt make it an option for the individual, like any new technology(not used successfully in processor cooling very long) this is probably due to tooling costs.

The words "american ingenuity" come to mind.......I think we can manufacture our own heat pipes after doing our homework.

I am not an A.C. or refregeration specialist, but I have freinds that are.
I for one am ready to meet this challeng head on!

[nihili]

Originally posted by RycheX
nihili, while I may admire your great ability to 'quibble', there is only one 'proper' definition of convection, and it is exactly what I stated. I'm not trying to take credit for any great knowledge I have, I'm simply stating what is and has been known for a very long time. If you'd like to further research the true meaning of convection, I'd highly suggest getting your hands on an engineering heat transfer text such as "Fundamentals of Heat and Mass Transfer" by Incropera and DeWitt. Relying upon websites for in-depth, and accurate heat transfer engineering understanding is simply not possible because they are full of the random musings of many people that have not thoroughly studied heat transfer. Nor do websites have the space to properly address the theory behind heat transfer or it's practical applications.

While it can be argued that there are small elements of convection within a heat pipe, the fact is simple that it is not the means of the vast majority of the heat transfer.

Since you seem to prefer hard copy, may I suggest having a look at Bejan's Convection Heat Transfer 2nd Ed. He clearly uses "convection" both in the sense of heat transfer within a fluid and also in the sense of heat transfer between a solid and a fluid. If you don't have access to a copy readily, then have a look through the abstracts of the Twelfth Internation Symposium on Transport Phenomena at http://www.geocities.com/Athens/Colu...stp12abook.PDF . Again both senses are clearly used throughout. I hardly think you can brush these people's usage off as the random musings of many people who have not thoroughly studied heat transfer. You may disagree with their usage, but their usage is neither casual nor untutored.

In fact, I think you'll find that the sense of convection as motion within a fluid is primary and that the use of "convection" to describe heat transfer from a solid to a fluid is derivative. After all it is only because of the convective (in the first sense) properties of fluids that free convection (in the second sense) differs from conduction. The former sense of "convection" is still overwhelmingly more common in meteorology and geophysics. So if you're going to insist that there is only one proper definition of "convection", I'm afraid it's not yours.

nihili

[Diggrr]

There now. I've deleted my posts...Happy?

I answered his question in the context it was asked.
I told him about the two instances I've seen it used.

Technology is useless unless those that need it can afford it.
That's where heatpipe technology falls short.

I had an oppinion of engineers before you posted, and still do, but that's not for debate here.

If you have an answer to the post, answer it already.

Bye.

[RycheX]

Diggrr is 100% correct in that my posts have long since strayed from the original intention of this thread and for this I apologize.

nihili, if you want to have a conversation with me about why i believe the definitions of convection presented throughout this thread do not apply to how heat pipes work, feel free to pm me and we can continue this elsewhere.

[BillA]

it's ONLY economics, not technology

all SPARC IIIs use a Thermacore heat pipe
(I've been trying to source one for some time, anyone got an "in" ?)

a year ago Thermacore gave me a ballpark price of $10 to $12 ea. for a flat plate heat pipe to fit Athlon and Pentium CPUs.

Problem: a minimum run is 50,000 units

but they recently had a pr blurb about a coutract for a mainstream application

RycheX and nihili

technical discussions are MORE than OK, they are the vehicle by which we (all) are exposed to the why of it

threads evolve, so be it
don't let the Luddites run you off

be cool

[neo86]

I am a bit stupid so excuse my ignorance. I am under the impression that a heat pipe works by evaporative cooling and then through convection the heat it carried to a cooler part of the pipe (usually the top). The question is:

How long will it take for the liquid to start evaporating and then comeback down? I don't want my processor to start catching on fire before any cooling has actually begun.

[Holst]

In laymans terms thats how they work.

With the correct type of heat pipe the process would happen very quickly.

This technology is very interesting and could work very well for cooling small CPU dies.

not only would it allow heat to be moved from the Die very efficently but you would als be able to increase the surface area that a cooler can be atached too. This would increase the efficency of the heatsink used.

Didnt the original thermoengine have a heat pipe in it ?
Or am I thinking of the cooler Billa memtioned

As for the quibbling I though that a heat pipe would utilise both convection and evaporation to move the heat, but im not a expert on those specific definitions. I did badly in my degree thermodynamics.

[nihili]

I'll take a picture of the heat pipe in my laptop (i'm pretty sure that's what it is) and post it. Unfortunately I don't have a digital camera, so it will take some time for film processing and the like.

nihili

[BillA]

I think almost all newer "big cpu" laptops use a heat pipe

for tinkerers, there's a surplus one on eBay with plates on either end

now if one cut the pipe out, and ran 4 in parallel swedged into a new copper plate . . . .

be cool

[mx]

Originally posted by BillA
I think almost all newer "big cpu" laptops use a heat pipe

for tinkerers, there's a surplus one on eBay with plates on either end

now if one cut the pipe out, and ran 4 in parallel swedged into a new copper plate . . . .

be cool

Could you provide a link BillA ?
I don't want to spend an hour searching for it!

[Ridenow]

I am a Dell certified tech (number 315282)

Dell Latitude laptops have a small heatpipe connecting the processor cover and the heatsink and fan. The processor is under the keyboard between about where the 0 key and the H key. The heatsink is along the right side of the laptop in the back.
All CP, CPi, CPt and CPx are like this. I think some of the other models are similar, but I do not work on them. They have been like this for years, at least before '97

[nihili]

Hmmmm, so if I could come across some old dead laptops....

Anyone have any ideas where the secret Dead Laptop burial ground is?

nihili

[mx]

Originally posted by Ridenow
I am a Dell certified tech (number 315282)

Dell Latitude laptops have a small heatpipe connecting the processor cover and the heatsink and fan. The processor is under the keyboard between about where the 0 key and the H key. The heatsink is along the right side of the laptop in the back.
All CP, CPi, CPt and CPx are like this. I think some of the other models are similar, but I do not work on them. They have been like this for years, at least before '97

Can't you get us a few from the scrap pile Ridenow?

[Ridenow]

Originally posted by mx


Can't you get us a few from the scrap pile Ridenow?

Sorry. When I work on them I am doing warentee work. I am required to send the used parts back to Dell or they charge me for the new parts.

As to the "dead laptop pile" I am good and do not let them die.

[Ridenow]

I thought I would give you guys a pic from the DCSE manuals. The heat pipe is the copper "L" shaped thing.

[Tiger]

Just wanted to join the debate about convection and latent heat of vapourisation. You guys are arguing about samantics.
All heat pipes incorporate capillary action in them. The best of them use a scintered metal as the tranfer interface(like a metal sponge). When the capillary is heated heat is conducted into the coolant causing the molecules to energise and therefore expand. This expansion results in the specific gravity of the coolant decreasing and then net result is that it rises since it is heavier than the molecules around it i.e. convection is established. However heat continues to be absorbed and will continue to be absorbed until it reaches its boiling point. Once the latent heat of vapourisation has been achieved it will vapourise. The key in heat pipe design is the size of the capillary in order for heat to be transferred in this way and the low pressure within the vessel.
So for a heat pipe to work effectively there must be convection in order to move the coolant in the capillary and phase change because phase change is the biggest consumer of energy.

[iggybaseball]

wow this whole thread is mind boggling, i only wanted a simple answer but i think i may have the jist of it now.