• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

how effective are heatpipes??

S

Soup

n00b
Joined
Feb 17, 2004
Messages
40
compared to a very, very big hsf and ram sinks,how would a heatpipe do?
 
A heatpipe doesn't dissapate heat like a normal heatsink does. All a heatpipe does is transfer heat to where it can be dissapated more effectivly.

What do you mean when you say "compared to a very, very big hsf and ram sinks" A heatpipe is used with a big HSF. What do ramsinks have to do with anything?
 
Originally posted by OneMadPoptart
A heatpipe doesn't dissapate heat like a normal heatsink does. All a heatpipe does is transfer heat to where it can be dissapated more effectivly.
Click to expand...

Isn't that exactly what a heatsink does too? :D
They just do it in different ways. A heatsink transfers heat only by conduction through a solid metal. Heatpipes apply another principle in conjuction to simple heat flow- a liquid-vapor phase change, like when sweat evaporates from your body to remove heat. The vapor carries the heat further away to a "heatsink" where it is then cooled and condensed back to liquid form, returning to the original "hot" location.
 
imo, they are pretty damn good

lots o laptops are running heatpipes to cool the internals

and the whole line of shuttle (except for the old generation model) sff barebones are using the heatpipe technology

i got my 2500+ locked barton @2.5ghz w/ 1.75v

so i'd say its working
 
My 8100 has a heatpipe built in. Dell allows temps to get warm before kicking in the fans on the heatsinks the heatpipe leads to, but the heatpipe is very effective.

I'm looking to get an SP-94 for my desktop, which also uses heatpipes. I'm curious to see how it goes.
 
Its obvious that you all dont know exactly what a heatpipe does. Let me explain in a little more detail.

A heatpipe is not a heatsink. It does not cool the processor directly, nor does it dissapate heat. The way it functions is that it effectivly transfers heat from one 'end' of the heatpipe to the other 'end'. In a shuttle system, there is a finless block that is mounted to the processor which absorbs heat. This heat is channeled through the heatpipe to another heatsink block attached to the other end of the pipe, but this one has fins and a fan attached to it. All the heatpipes do is "move" the heat from the processor to the other heatsink where a fan can be inserted much easier, as space is limited in a SFF computer. The thermalright heatsinks that have heatpipes basically move as much heat as they can from the base to where they connect with the fins farther up the heatsink and closer to the fan. By brining as much heat as close to the fan as possible, better dissapation by the heatsink is possible.

Understand? A heatpipe is not a heatsink, but rather a space saver or more effecient way to get heat to where it is dissapated better.
 
Originally posted by OneMadPoptart
It seems that everyone from Rochester NY seems to know... How bout the rest of you?
Click to expand...

I know how they both work... but in the broader view of things, they both move heat from one place to another- the CPU to air. Like I said, a regular heatsink moves heat from the CPU to the surface of the fins by convection only. A heatpipe does it by conduction to the liquid inside, a phase change, natural convection of the gas, another phase change, then conduction to the surface of the fins. I guess if you want to be technical, you can say a heatpipe cooler is not a heatsink... but then a heatsink doesn't just sink heat either, so you'd better call it a heat-dissipator instead :D
 
Heatsinks are great for dissipating heat then, in your words. Heatpipes aren't so good, but they're a highway for transferring all that heat that would be sitting around in a heatsink to a location further from the heatsource where it can enter another heatsink and dissipate there.

Think of them as clearing space for the processor to put more heat in to the copper block sitting on it, they can move heat out of an area faster than the heatsink can. They just can't get rid of it as efficiently though and they need a heatsink to get rid of it at the other end, otherwise they'd get all jumbled up and wouldn't cool for sh*t.
 
here is the just of what heatpipes do...heat transfer wise

we all understand that the die of the processor gets really hot. However, since we can mount a heatsink 100% on it, we need a thermal compound to transfer heat from the surface of the die to the bottom of the heatsink.

Now we loose some heat transfer ability through the thermal paste itself in called "contact resistance" and "microvoids", these microvoilds are usually filled with air and since air has a lower heat transfer coefficient that say copper it adds resistance.

Now after the thermal paste we are at the heatsink. To be 100% efficient a heatsink must have the same fin temperature at the tip of the fins as the base, now we all know that that dosent happen. Here is why, their is inherent resistance in the copper itself in the form of grain boundaries, microvoids etc, all of these add up to creating a certain resistance in the material. And the father the heat has to travel through this material the more additive resistance is added.

Thermalright heatsinks are soldered to the base, their again is more contact resistance.

Finally, one way to overcome the inherent resistance in any material and in this case copper you can utilize heatpipes. Heatpipes are very special in that they can absorb HUGE amounts more than 20times than copper of heat and then it goes from a liquid to a gas and travels up the heatpipe to the top of the heatsink where the fan cools the heatpipe down and thus it condenses back to a liquid and flows down the the bottom of the heatpipe and then starts all over again. Now you are probally wondering why use heatpipes?

The heatpipe enables you to kinda skip over most of the heatsink material thus by=passing the inherent resistance and thus going to the top of the heatsink and redistributing heat their where it can be effectively dissipated.

Quickly think of shuttle computers how they have heatpipes on their heatsinks that travel to another heatsink with a fan, heatpipes are a conduate for heat to flow from one place to the next.

IF and that is a big IF you could make material with 100% efficency thus with no inherent resistance you wouldnt need heatpipes but do to material defects you need them.

Hope that helps, this is a materials engineering point of view, if you want more technical i can supply you with tons of formulas and you can have the fun of calcuting it out..ps it takes a few pages.
 
Hence we have designs like the Thermalright SP94/97 that use fins to dissipate the heat and heatpipes to try to get more heat out to the whole area of the fins (and away from the base of the heat sink and thus the CPU).
 
Originally posted by c00z
here is the just of what heatpipes do...heat transfer wise

we all understand that the die of the processor gets really hot. However, since we can mount a heatsink 100% on it, we need a thermal compound to transfer heat from the surface of the die to the bottom of the heatsink.

Now we loose some heat transfer ability through the thermal paste itself in called "contact resistance" and "microvoids", these microvoilds are usually filled with air and since air has a lower heat transfer coefficient that say copper it adds resistance.

Now after the thermal paste we are at the heatsink. To be 100% efficient a heatsink must have the same fin temperature at the tip of the fins as the base, now we all know that that dosent happen. Here is why, their is inherent resistance in the copper itself in the form of grain boundaries, microvoids etc, all of these add up to creating a certain resistance in the material. And the father the heat has to travel through this material the more additive resistance is added.

Thermalright heatsinks are soldered to the base, their again is more contact resistance.

Finally, one way to overcome the inherent resistance in any material and in this case copper you can utilize heatpipes. Heatpipes are very special in that they can absorb HUGE amounts more than 20times than copper of heat and then it goes from a liquid to a gas and travels up the heatpipe to the top of the heatsink where the fan cools the heatpipe down and thus it condenses back to a liquid and flows down the the bottom of the heatpipe and then starts all over again. Now you are probally wondering why use heatpipes?

The heatpipe enables you to kinda skip over most of the heatsink material thus by=passing the inherent resistance and thus going to the top of the heatsink and redistributing heat their where it can be effectively dissipated.

Quickly think of shuttle computers how they have heatpipes on their heatsinks that travel to another heatsink with a fan, heatpipes are a conduate for heat to flow from one place to the next.

IF and that is a big IF you could make material with 100% efficency thus with no inherent resistance you wouldnt need heatpipes but do to material defects you need them.

Hope that helps, this is a materials engineering point of view, if you want more technical i can supply you with tons of formulas and you can have the fun of calcuting it out..ps it takes a few pages.
Click to expand...

Ok, well I was referring to heatpipe CPU coolers as a whole, including the pipe(s) and the fins (which is the same in function as a heatsink basically.)

And thermal resistance is not due to materials defects. Materials with nearly perfect crystall lattices and virtually pure compositions can be produced, but they still have an inherent resistance to thermal conduction. Pure copper many fewer grain boundaries and dislocations than cold-worked copper, orders of magnitude less. But thermal conductivity doesn't change by nearly that much - just a few percent maybe.

And what equations are you referring to? Steady state heat conduction is fairly simple and can be summed up in just one equation. Heat of phase transformation is also a mere couple of equations at most. The most involved aspect of the problem would be with convective transfer of heat into the air or the phase change fluid, but you didn't really mention anything about that.
 
Ok, well I was referring to heatpipe CPU coolers as a whole, including the pipe(s) and the fins (which is the same in function as a heatsink basically.)

And thermal resistance is not due to materials defects. Materials with nearly perfect crystall lattices and virtually pure compositions can be produced, but they still have an inherent resistance to thermal conduction. Pure copper many fewer grain boundaries and dislocations than cold-worked copper, orders of magnitude less. But thermal conductivity doesn't change by nearly that much - just a few percent maybe.

And what equations are you referring to? Steady state heat conduction is fairly simple and can be summed up in just one equation. Heat of phase transformation is also a mere couple of equations at most. The most involved aspect of the problem would be with convective transfer of heat into the air or the phase change fluid, but you didn't really mention anything about that.


__________________
thermal resistance in part is due to grain boundaries and many other factors. However, they are a factor. You cannot manufacture 100% perfect materials that you speak of. And yes every material has a inherent resistance in it. Yes cold-worked copper would have orders of magnitude greater grain boundaries and disloactions due to it cold working.

The equations you would need to use are 3 dimensional which I know must be done with computers using such approximations as finite element methods. Also you would need convection since that is what the fan is providing, conduction within the material itself, constact resistance from both the solder in the heatsink itself, the thermalpaste and the thermal paste before the heat spreader.

Pretty much too much work to do. I didnt go into too much detail since i wanted to keep my post shorter than longer. Unless you employ a computer you have no chance doing these calcuations especially if it isnt steady state (such as throttling loading etc)

heatpipes can be summed to a conduate for thermal heat transfer to another region of the heatsink which uses convection to cool the metal, while the heatpipe uses phase change (liquid-gas) to move heat instead of conduction which a heatsink would use.
 
I'm amused by those Zalman HD coolers, they have heatpipes that run up and over the drive, but I see no fins on them. I wonder how much they work.

I doubt they do much, and they take up a buncha room in a 5.25 bay.
 
Originally posted by c00z

thermal resistance in part is due to grain boundaries and many other factors. However, they are a factor. You cannot manufacture 100% perfect materials that you speak of. And yes every material has a inherent resistance in it. Yes cold-worked copper would have orders of magnitude greater grain boundaries and disloactions due to it cold working.
Click to expand...

My point was that grain boundaries and dislocations play a small role in conductivity. I know there are no 100% perfect materials, I didn't say 100% perfect. But there are single crystal materials which eliminates grain boundaries and some can be annealed to drastically reduce dislocation density. What I was stating as an example was that a single crystal of copper would have nearly the same thermal conductivity as a highly worked one.

heatpipes can be summed to a conduate for thermal heat transfer to another region of the heatsink which uses convection to cool the metal, while the heatpipe uses phase change (liquid-gas) to move heat instead of conduction which a heatsink would use.
Click to expand...

Yes, that's pretty much what I said before...

Are you a materials scientist/engineer, by the way? You seem to have a pretty good grasp on the subject at hand. I'll be getting my bachelor's degree in a couple months :D and then hopefully another degree 2 years down the road...
 
materials engineer....well working towards that...i am in a 5 year program which earns me essentially two degrees, one a materials engineering degree and the other a general business degree. Thus as I am in third year I am only about a 2 1/2 year engineer
 
You must log in or register to reply here.
Back
Top