-
Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!
heat pipe
- Thread starter nokiaXP
- Start date
- Joined
- Sep 24, 2002
- Location
- Best 1/4 Mile Time: 13.57 @ 101.13
I don't know about the heat pipe thing.....lokks interesting but I have no details about it.
Thermal paste on the other hand is a REQUIREMENT. Well, at least some kind of TIM (thermal interface material) is needed. Those little pads and junk that come with OEM heatsinks are ok for your internet/email/AOL/solitaire user, but for anyone else....ESPECIALLY for overclockers.....Arctic Silver 3 is pretty much the one and only.
Thermal paste on the other hand is a REQUIREMENT. Well, at least some kind of TIM (thermal interface material) is needed. Those little pads and junk that come with OEM heatsinks are ok for your internet/email/AOL/solitaire user, but for anyone else....ESPECIALLY for overclockers.....Arctic Silver 3 is pretty much the one and only.
A heatpipe has inside it some form of liquid. Let's take the coolermaster hhc001 for example. When placed on cpu, the liquid inside the pipe gets heated and evaporates to the top of the sink where the pipe has cool air flowing over it. The vapors from the liquid condense back into liquid and flow back down into the bottom to heatpipe to repeat the process all over again. Basically, the purpose of a heatpipe is to increase the rate that heat is dissipated away from the cpu and therefore keep it cool.
Thermal compound is neccesary because no matter how smooth you think the bottom of a heatsink is, it still has microscopic valleys and ridges. When placed on the cpu die by itself, the valleys and ridges get sealed in forming air pockets. These air pockets not only greatly reduce the effectiveness of the heatsink, they will raise the temperature by a few degrees. The thermal compound fills the microscopic dents and allows full conduction to occur between the cpu die and heatsink.
Thermal compound is neccesary because no matter how smooth you think the bottom of a heatsink is, it still has microscopic valleys and ridges. When placed on the cpu die by itself, the valleys and ridges get sealed in forming air pockets. These air pockets not only greatly reduce the effectiveness of the heatsink, they will raise the temperature by a few degrees. The thermal compound fills the microscopic dents and allows full conduction to occur between the cpu die and heatsink.
- Joined
- May 10, 2002
- Location
- Los Alamos, NM
A slight clearification on that statement. You may think that gravity is the reason the liquid flows back because the word down was used. In general, gravity isn't a player in the process. What drives the cycle lazerin describes is the thermal difference plus a capillary effect. It is the capillary effect that moves the fluid back to the hot region for evaporation.
One of the original reasons heat pipes were developed was for use in space to remove heat from a source and radiate it to the cold-vacuum of space. Of course no gravity there.
Consider the hhc001, while mounted initial so that gravity might work (if it were the driver) you then place the mobo on its side in your case. This actually turns the heat pipes flat (or nearly so). Thus if gravity were driving the liquid flow, they would probably not work.
I've not taken hhc001 to inspect it but can only assume that the designers weren't stupid and did build them correctly. However I must say that people do prove me wrong from time to time when I make that assumption.
To answer your question, "how would we know?"
1) An indirect method would be to first find out the fluid being used in the heat pipe and if the pipe is pressurized. Determine the fluid's saturation temperature at the pressure (this ignores the pressure drop in the heat pipe) then measure the temperature at the cold end of the heat pipe both loaded and unloaded. It should be at or below the saturation temperature of the fluid. The problem with this technique is that if the fins were over designed (providing more cooling than needed in this design) you would be fooled by the measurement.
2) A direct method would be to determine the normal operational temps of your cpu (loaded and unloaded). Then cut the heat pipe open, i.e. allow the fluid to escape so they won't work. Then measure the cpu performance again. This would tell you the real answer but you've lost a HS in the process.
Several threads over the last week or so on thermal paste, suggest you read them.
Sorry for the long post!
Last edited:
- Joined
- Nov 13, 2001
- Location
- Chicago.
Actually, the shin etsu compound is better than the Arctic Silver 3 compound... although arctic silver IS coming out with a new compound soon, I will hopefully get a test sample within a week or two.
- Joined
- May 10, 2002
- Location
- Los Alamos, NM
No way man! Your description was a good one. I just added a variation/refinement.
