Heatpipe retailer

[laxman]

I was wondering if any of you know a good source for heat pipes or heatplanes (the flat versions of heat pipes)? They would need to be short to the length of 10-12cm and with a small diameter. Any thoughts?

[jaydee]

Originally Posted by laxman

I was wondering if any of you know a good source for heat pipes or heatplanes (the flat versions of heat pipes)? They would need to be short to the length of 10-12cm and with a small diameter. Any thoughts?

I don't think you will find anything pre-made. Usually you have to custom order them specific for your use. I would go to www.google.com and start searching.

[bobo5195]

zalman hard drive coolers

[Long Haired Git]

More reading required. Heat pipes are quite specific in their working temperature range and also their wattage capabilities.

[BillA]

also their internal volume, cannot be 're-bent' as volume dictates the charge

[bobo5195]

heatpipes are quite easy to model using 1D FEA analysis however the problem is degration with time is kinda complicated.

[BillA]

FEA is Greek to DIY (someone looking for a source)

[Long Haired Git]

Originally Posted by bobo5195

heatpipes are quite easy to model using 1D FEA analysis however the problem is degration with time is kinda complicated.

I thought one of the benefits of heatpipes was years of maintenace free service due to lack of moving parts (bar the coolant itself, of course).

What causes the degregation, and how marked is the effect? Down to 50% of initial performance over what time frame? How big a drop over, say, 5 years?

[bobo5195]

I can’t find the link anymore. It came up on google when i was searching for other stuff. It was from an engineering product website. Though I doubt the engineering integrity of the site as the marketing guys might have got to it.

It said heatpipes are very good but there are problems with them apparently. The cause of them mucking up seems to be residue due to the phase change. The effect happens of the order of years but after about 5 years or so there could be problems.

FEA = finite element analysis. Which is splitting an object up into blocks and then analysing the forces on each block using elementary procedures. This (well its a version of FEA called finite volume analysis where you measure the flow in and out of a block) can be used in full 3d to model heatpipes but the model is computationally expensive (that high end X2 is not leet enough) and presents problems. This is because the flow is two phase (gas and fluid) and highly temperature and energy dependent. The FEA program mentioned above uses the 1d assumption. Which uses an empirical / analytical method to calculate how a heat pipe would behave over a short section (where what defines short is actually rather complicated). It then uses many of these sections to model the heat transfer of the pipe to a solid section (ie a heatsink model).

[bobo5195]

found it

http://www.crtech.com/heatpipe.html

can't find the bit where it mentions heatpipes being blocked up.

[BillA]

Thermacore has a mktg spiel on heat pipe reliability
before I would conclude that all . . . .
(not gonna catch 'em all with the same net)

[bobo5195]

Had a little search on Inspec and compendex (research paper databases)

This paper:
THE ALUMINUM FLAT HEAT PIPE USING CYCLOPENTANE AS WORKING FLUID
Takahiro Shimura, Hitoshi Sho and Yoshio Nakamura
The Furukawa Electric Co., Ltd.
Ecology & Energy Laboratory
2002 Inter Society Conference
on Thermal Phenomenir


For two heat pipes cyclopentane and hcfc123 obtained dT was:

dt = dT[initial] + a(1- e ^ (-B*t))

where a (kelvin) and b (10^-3 / hours) are constants.
For cyclopentane in their experiment they list
a= 1.73
dt(initial) =1
B = 1.39

for a period up to 1500hrs (realively low amount of time)

How this corresponds to computer heatsinks is unclear but the extrapolation shows that while they degrade over time they only degrade to a lower performance level not to infinity.