Do Heat Pipes Contain Coolant?

[Cuke]

Has anyone opened up a heat pipe to see what's inside? In theory an coolant inside them should help with convection cooling if the system has a vertical orientation, with the cooling fins above the heat source.  Many systems have a horizontal orientation for the heat pipes, or have an unusual winding path (i.e. laptops). One would think this would be detrimental to convection inside the pipe and cooling would suffer. Copper is a good thermal conductor and should provide good cooling by itself. Solid copper should be better than a hollow pipe. Does anyone else feel like heat pipes are more of a marketing thing than an effective cooling solution? Maybe LTT could do some testing?

[GoldenLag]

some heatpipes contain some form of liquid coolant. though it evaporates really easily if you open it. its ever so slightly better in some cases. i believe the Lumia 920 had liquid cooling as well as the razor phone. but its just that little liquid in the copper heatpipe (and there is very little coolant. 

 

the Copper is very expencive and therefore it is cost effective to transport the heat using copper pipes and then using cheaper aluminium fins for the cooling area. 

[dcgreen2k]

Heat pipes do contain a liquid of some sort, and they work by allowing the substance inside the hot end of the pipe to evaporate and condense back into liquid once it reaches the cool end of the pipe, where it transfers the heat energy it has picked up to another cooling device like a heatsink.

[Cuke]

I understand how the heat pipes work, which is why I question their validity as a cooling solution.  A system with heat pipes does not contain a pumping system and relies on gravity to return the condensate to the hot side of the system to keep the convection flow going. If the hot side is above the cold side, gravity cannot return the condensed liquid and it becomes trapped in the cooling fin end of the tube. There is only one orientation which will allow this system to work and that is one in which the hot side is below the cold side. Looking at many GPU, CPU, laptop, mobile device coolers it looks like they simply wouldn't work, and the only cooling they have comes from the thermal conductance of the copper in their coolers. Additionally, the only cooler that looks like convection cooling (in the pipe) should work properly is a cpu cooler in a desktop system where the CPU/MOBO is horizontal, and the cooling fins are above the CPU.

[Mira Yurizaki]

45 minutes ago, Cuke said:

I understand how the heat pipes work, which is why I question their validity as a cooling solution.  A system with heat pipes does not contain a pumping system and relies on gravity to return the condensate to the hot side of the system to keep the convection flow going. If the hot side is above the cold side, gravity cannot return the condensed liquid and it becomes trapped in the cooling fin end of the tube. There is only one orientation which will allow this system to work and that is one in which the hot side is below the cold side. Looking at many GPU, CPU, laptop, mobile device coolers it looks like they simply wouldn't work, and the only cooling they have comes from the thermal conductance of the copper in their coolers. Additionally, the only cooler that looks like convection cooling (in the pipe) should work properly is a cpu cooler in a desktop system where the CPU/MOBO is horizontal, and the cooling fins are above the CPU.

 

Wikipedia has an explanation behind it:

Quote

For the heat pipe to transfer heat, it must contain saturated liquid and its vapor (gas phase). The saturated liquid vaporizes and travels to the condenser, where it is cooled and turned back to a saturated liquid. In a standard heat pipe, the condensed liquid is returned to the evaporator using a wick structure exerting a capillary action on the liquid phase of the working fluid. Wick structures used in heat pipes include sintered metal powder, screen, and grooved wicks, which have a series of grooves parallel to the pipe axis. When the condenser is located above the evaporator in a gravitational field, gravity can return the liquid. In this case, the heat pipe is a thermosiphon. Finally, rotating heat pipes use centrifugal forces to return liquid from the condenser to the evaporator.

 

[Canada EH]

1 hour ago, Cuke said:

One would think this would be detrimental to convection inside the pipe

That is why there is pressure inside the tubing, so there is full liquid contact on the I.D.

Heat transfers from the copper and gets sinked into a mass, then the water evacuates the heat by the transfer of heat from the metal mass to the water, that water then glows to a radiator where the heat is taken away, water reduces in heat and the process starts all over again.

 

You can use different liquids like in colder climates vehicles use anti-freeze along with the water so the water does not freeze. In very cold climates like the arctic other methods are used. There will be some liquids specially engineered to soak up more heat, but its application specific.

Doionized water, inhibited glycol, dielectric fluids.

 

For general use, water is great.

[Cuke]

Ok, capillary action can explain the ability to move the liquid in opposition to gravity, I can accept that.  Capillary action is not a fast process. I just don't believe that dissipating heat from a 100 W CPU by "capillary action" is the predominate cooling method used here. Gravity convection and thermal conductance of copper would win every time. From a production viewpoint, I don't see filling copper tubes with capillary material and a coolant as being less expensive than using solid copper.

[Amazonsucks]

Heat isnt taken away by capillary action, its taken away by the liquid evaporating. The capillary action brings thre liquid back from the condensor section to the processor via the heat pipe interior wall.

 

Vapor chambers work in a similar way, changing the phase of the liquid in a constant closed cycle to remove more heat than a block of metal alone.

 

They dont put gimmicks on satellites or other indusrtrial applications where no one will see them, and heat pipes are used in many places other than PCs to great effect.

[corrado33]

2 hours ago, Cuke said:

Ok, capillary action can explain the ability to move the liquid in opposition to gravity, I can accept that.  Capillary action is not a fast process. I just don't believe that dissipating heat from a 100 W CPU by "capillary action" is the predominate cooling method used here. Gravity convection and thermal conductance of copper would win every time. From a production viewpoint, I don't see filling copper tubes with capillary material and a coolant as being less expensive than using solid copper.

You're wrong. Heat pipes are well tested and shown to have an effective heat transfer coefficient many many times greater than a copper bar. Capillary action may not be super fast, but when properly designed in a controlled environment it's quite fast enough. Plus the fluid they use likely has a very high specific heat so it doesn't require a lot of liquid to move a lot of heat.

[Amazonsucks]

3 hours ago, corrado33 said:

You're wrong. Heat pipes are well tested and shown to have an effective heat transfer coefficient many many times greater than a copper bar. Capillary action may not be super fast, but when properly designed in a controlled environment it's quite fast enough. Plus the fluid they use likely has a very high specific heat so it doesn't require a lot of liquid to move a lot of heat.

And also its the simple fact that the capillary action isnt what removes the heat. Its the opposite. It brings the condensed liquid back to the processor and the liquid evaporating and moving toward the cooler area with the fan and fins is what actually removes the heat. 

 

Capillary action returns the "cold" liquid to the heat source. It does not move hot liquid away.

 

The phase change is much more efficient at moving heat to the cooling fins than a solid piece of copper of the same geometry, and heat pipes are much more difficult to create than a simple copper extrusion or casting as well, so its not about reducing material use really. Its about quickly moving heat from one place to another using the phase change of the liquid absorbing the heat.

[Cuke]

22 hours ago, corrado33 said:

You're wrong. Heat pipes are well tested and shown to have an effective heat transfer coefficient many many times greater than a copper bar.

I'm not saying they don't work. I understand their theory and agree with it. I just don't believe that capillary action can move enough coolant to be effective. After looking at many of the coolers out there, I think the manufactures may believe this also because many of the devices have cooling fins located above and below the heat source to compensate for different orientations of the device. Some specify that the cooler can only be installed with a certain orientation. There will always be a cooling fin above the heat source, and any fins below the heat source would only provide for heat removal by conductance.

 

As a side note. The reason I have an interest in this topic is because I am contemplating building a pure convection/powerless (no pumps or fans) cooling system for my entire system, and I'm considering using custom built heatpipes for everything (anything that generates heat) in a sort of steampunk theme.

[dgsddfgdfhgs]

17 minutes ago, Cuke said:

I just don't believe that capillary action can move enough coolant to be effective. After looking at many of the coolers out there, I think the manufactures may believe this also because many of the devices have cooling fins located above and below the heat source to compensate for different orientations of the device.

you may think this way, if it is not effective meaning that it is cool enough.

and the liquid inside have very low evaporation point. so it doesnt matter how the tubes are orientated, again if it matters, it is then cooled enough

 

 

[corrado33]

54 minutes ago, Cuke said:

I'm not saying they don't work. I understand their theory and agree with it. I just don't believe that capillary action can move enough coolant to be effective. After looking at many of the coolers out there, I think the manufactures may believe this also because many of the devices have cooling fins located above and below the heat source to compensate for different orientations of the device. Some specify that the cooler can only be installed with a certain orientation. There will always be a cooling fin above the heat source, and any fins below the heat source would only provide for heat removal by conductance.

Every laptop in the last 10 years or so uses heatpipes.

 

Not to mention that many OEM computers can be used either horizontal or vertical. 

 

Take, for example, the new mac mini.

 

 

See the heat pipes? They're 100% horizontal with no cooling fins above them. However, mac minis can also be used in a vertical orientation. 

 

 

Heat pipes work, regardless of the orientation. 

[Amazonsucks]

5 hours ago, Cuke said:

I'm not saying they don't work. I understand their theory and agree with it. I just don't believe that capillary action can move enough coolant to be effective. After looking at many of the coolers out there, I think the manufactures may believe this also because many of the devices have cooling fins located above and below the heat source to compensate for different orientations of the device. Some specify that the cooler can only be installed with a certain orientation. There will always be a cooling fin above the heat source, and any fins below the heat source would only provide for heat removal by conductance.

 

As a side note. The reason I have an interest in this topic is because I am contemplating building a pure convection/powerless (no pumps or fans) cooling system for my entire system, and I'm considering using custom built heatpipes for everything (anything that generates heat) in a sort of steampunk theme.

Youre misunderstanding the role of capillary action in the heat pipes entirely.

 

It is not to move hot liquid AWAY from the processor at all. Its actually being used to move the cooled liquid back to the processor, after the VAPOR condenses in the end of the heat pipe with fins and the airflow from fans.

 

What actually removes the heat from the processor is the liquid VAPORISING and carrying the heat away in the vapor, to the cold area(with the fins and fans) where the vapor exchanges the heat its carrying and condenses back into its liquid state. Then the condensed liquid is moved back to the processor in a continuous loop.

 

The orientation of the pipes doesnt matter because the cooled liquid flows back to the processor by capillary action, which is why they have the sintered capillary walls inside the pipes. The capillary action is flowing TOWARD the heat source, not away. Orientation only matters if the heat pipe's working fluid would not be able to flow enough against gravity by capillary action in sufficient volume that it would reduce the effectiveness.

 

Heat pipes, like vapor chambers, are PHASE CHANGE coolers, meaning that the change in the working fluid's state from liquid to vapor is what's actually doing the cooling.

 

Its also important to understand that heat pipes take heat from a tiny surface area(the GPU or CPU die or package in this case) and maximize the surface area for heat exchange by conducting it to the fins, which are attached to the heat pipe. The fans blowing air over the fins is the final transfer of heat from the heat source to the air. 

 

 

[LogicalDrm]

You haven't seen LTTs video series called "Sketchy Heatsinks" then, haven't you? The last video of that series explores meaning of having heatpipes in design vs just block of heatsink aluminum or even some spread of fins.

[Amazonsucks]

6 hours ago, LoGiCalDrm said:

You haven't seen LTTs video series called "Sketchy Heatsinks" then, haven't you? The last video of that series explores meaning of having heatpipes in design vs just block of heatsink aluminum or even some spread of fins.

All i found was something about them making their own heatsink. But im not commenting on how well designed a cooler is anyway. 

 

Im explaining that capillary action is the way cool liquid gets drawn back to the heat source, not how heat is extracted, in heat pipes.

[LogicalDrm]

1 hour ago, Amazonsucks said:

All i found was something about them making their own heatsink. But im not commenting on how well designed a cooler is anyway. 

 

Im explaining that capillary action is the way cool liquid gets drawn back to the heat source, not how heat is extracted, in heat pipes.

I'm replying to OP, not you. They asked for LTT to make video.

[Amazonsucks]

1 hour ago, LoGiCalDrm said:

I'm replying to OP, not you. They asked for LTT to make video.

Oh my bad

[WoodenMarker]

Quote

The main reason for the effectiveness of heat pipes is the evaporation and condensation of the working fluid. The heat of vaporization greatly exceeds the specific heat capacity. Using water as an example, the energy needed to evaporate one gram of water is 540 times the amount of energy needed to raise the temperature of that same one gram of water by 1 °C. Almost all of that energy is rapidly transferred to the "cold" end when the fluid condenses there, making a very effective heat transfer system with no moving parts.

https://en.wikipedia.org/wiki/Heat_pipe

Capillary action doesn't need to be very fast because it doesn't take a lot of liquid to transfer a lot of heat.

[L0G1C]

On 11/21/2018 at 3:43 PM, Cuke said:

Has anyone opened up a heat pipe to see what's inside? In theory an coolant inside them should help with convection cooling if the system has a vertical orientation, with the cooling fins above the heat source.  Many systems have a horizontal orientation for the heat pipes, or have an unusual winding path (i.e. laptops). One would think this would be detrimental to convection inside the pipe and cooling would suffer. Copper is a good thermal conductor and should provide good cooling by itself. Solid copper should be better than a hollow pipe. Does anyone else feel like heat pipes are more of a marketing thing than an effective cooling solution? Maybe LTT could do some testing?

Some fancy ones do. Vapor chamber ones do as the coolant is the vapor

[Bitter]

I think what he's stuck on is looking at the system as if it were a thermosiphon, if you're unfamiliar with phase change cooling and some of the more exotic cooling principles it can be hard to get a handle on especially without a background in science.

 

As for what he wants to build, a thermosiphon water cooling loop...that sounds cool but will be a lot of trial and error I think. And that WILL heavily depend on orientation of components as well as their locations.

[PC Pig]