cooling: heat pipes vs. solid copper lines

what do you all think, which one would provide better efficiency on heat transfer?

- consider same thickness - one is solid copper, the other one is heat pipe filled with gas.

I'm sure manufacturers prefer heat pipes because it's most likely the cheaper solution, as well as cuts on laptop weight. But is it actually better?

 

Heat pipe all the way, it might look like solid copper is better, but using an evaporating fluid to carry the heat from one end of the pipe to another is actually better.

The resistance heat transfer in conduction is as follow: L/(kA) where L is the length of the pipe, k the thermal conductivity (~400W/mK for copper) and A the surface area perpendicular to the direction of the heat flow, even taking into consideration convection at the surface of the pipe, a solid copper pipe would have a rather large resistance.

With the evaporating fluid, you basically only get a resistance at each end of the pipe due to condensation/evaporation, fluid flow rate is also a consideration though. Resistance in those cases is 1/(hA) where h is the convection coefficient and A the surface area where the fluid is in contact with the hot/cold surface. The convection coefficient in those cases is usually quite high. I don't have the value for a heat pipe, but for a shell and tube heat exchanger, we're usually talking 10 000 to 20 000 W/m²K in condensation.

tl;dr to transfer heat from point A to point B over a significant distance, a heat pipe filled with a fluid is better.

Edit: when using resistances in heat transfer, the equation for the heatflow can be written as Q = (T2-T1)/R

Edit2: This is an educated guess on my part, i could be wrong, i have a pretty good grasp of industrial heat exchangers, but not of heat pipes in particular. I can say for certain though that in a properly designed heat exchanger, conduction will almost always be the highest resistance to heat transfer.

 

It would seem the heatpipe is better for cooling however it's all about dissipating that heat buildup. So long as there's a nice radiator to have air pass through it and cool the whole set up a solid chunk of copper works much better over heat pipes, I've learned from experience heat pipes are great for transferring heat but have a much lower threshold for keeping maximum temps down. The solid copper takes a lot more time to heat up and generally can keep a chip much cooler since it isn't heating up as fast as the heat pipes which quickly match processor heat. Then again it also depends on the number of heat pipes and as stated earlier what kind of radiator is used to get rid of the heat

 

^Basically what he's saying is that the heat pipes take heat away from the CPU/GPU area over to the actual heatsink fins and fan.

With a solid copper rod, you wouldn't really be shuttling heat away from the critical area.

 

Heatpipes have the BEST thermal conductivity beside water. Plain copper has the best thermal dissipation and capacity but that is related to surface area as it comes in contact with airflow.

 

Heat pipes work better simply because there's something physically carrying the heat away. A solid copper bar is, well... solid, whereas a heat pipe actually has things going on inside it.

I'm sure I could think up some sort of poor anology ralating to everyday items if you want.

 

While i understand what you're saying, the term thermal conductivity is used to describe the amount of heat that can pass through a surface of a specified thickness (per meter/foot/etc.) that is not in movement. Heatpipes are the best way to transport heat from point A to point B, but saying they have better thermal conductivity is wrong, they do have a better heat transfer coefficient. Water has poor thermal conductivity of ~0.6W/mK, the term you're looking for is specific heat (Cp) and/or vaporization enthalpy (ΔHv) if phase change is involved. Given it's high Cp, water makes an excellent fluid to transport heat though, same for it's ΔHv.

Regarding what Ari3sgr3gg0 said, he's right, in the end you still need a good heatsink with a fan to dissipate the heat and that's where copper excels because you're going to need a large surface area to dissipate the heat and you'll have to contend with thermal conductivity of the material you use. Copper isn't cheap, but it's the best you can get in terms of thermal conductivity with a reasonable price. There are other materials like silver and diamond that have better thermal conductivity, but you can guess the price tag that comes with it. The heat pipe is just a means of transporting heat and there's nothing like an evaporating fluid to absorb large amounts of heat.

The best thing would be to stick the heatsink and fan right on top of the component, but not gonna happen in a notebook.

EDIT: Sorry for nitpicking about the terms, comes with being a TA for a heat transfer class .

 

I would like to point out that:

- heat pipe would have faster heat transfer due to the fast moving gases inside;

- given the time needed, the copper line will also heat up to the point of the heat pipe, at least over the heating source (say the CPU).

- the copper line should not have better heat dissipation because that is primary related to the involved surface area (and the material used, thought both metal), and we're talking about the same size/diameter heat pipe and copper line.

- we could probably assume that the heat pipe could be made of brass instead of copper, so that could be small difference.

So could the bigger mass of the copper rod help in transferring "bigger amount" of heat at a time, thus compensating for the slower heat transfer itself ?

 

Solid copper bar wouldn't even work if it is long enough. One end could be melting and the other end could be ice cold. The hot end will cool itself via convection.


--

 

No miro_gt. The mass of the heatpipe versus the copper rod has nothing to do with it. The copper rod could store more heat in it at the same temperature, but that means nothing with regards to how it can transfer heat. Heatpipes are far and away more effective than solid copper at transferring energy between a "hotter" and "cooler" area. If you replaced the heatpipes in almost all laptops with solid copper rods, the laptops would fry. The evaporation and condensation of the refrigerant in a heatpipe can transfer more heat over greater distances. I think you are thinking the way you are because you know that copper is great at conducting heat. That might be relatively true, but there are things that are way better.

Take a look at the evolution of desktop heatsinks from 6 years ago to today. The heatsinks use a lot less copper but are far better. The ability of heatpipes to transfer large amounts of heat over great distances allowed heatsinks to reach ridiculous sizes that would be impractical with using pure copper. With old heatsinks, the base would be small, and fins would have to come out of it immediately because the heat transfer through a thick or large base would be too poor. Now fins are inches away from the base. Heatpipes are just that good.

 

^This

A rod may be able to hold more heat in terms of thermal capacity, so it will take much longer to heat up (but eventually it will if the thermal output of the proc. is more than the dissipation rate of the rod) and your laptop will be cooler initially, but it's not particularly good at moving it from one point to another. Eventually the rod will reach its thermal capacity, and a solid rod of metal is not a particularly good heatsink, meaning putting a fan on it wont do much unless it's an old style heatsink with a copper base and massive fins coming directly from the base as Qing Dao said.

Remember, strictly speaking, heatpipes are a heat transfer device.. not a cooling device unto themselves.

 

Yep, copper is good if you have to transport heat through a solid medium, however the best way to transport heat from point A to point B remains a moving fluid. If you can go through a evaporation/condensation cycle, even better. This is avoided in industries because of the energy and equipment cost associated with compressing a refrigerant. A heat pipe however relies on capillarity and pressure changes to do the evaporation/condensation cycles. They might be a bit complex, but once it's manufactured, it's basically a free condensation/evaporation cycle.