Heat Pipe Construction ...
 

How important is it to have the inside of a heat pipe in a vacuum? Can I just poor some suitable liquid in a pipe, close it off and bada-bing you're good to go? Also If you're not competing with gravity, no need for a wick right?

Todd Trowbridge

The purpose of a vacuum is to reduce the boiling temperature. If you are using alcohol, and 40C (minimum temperature of whatever you're cooling) is fine, then go ahead. The purpose of a wick is to return the recondensed liquid to the base, but if your heatpipes cold end is higher (the higher it is, the better it is) than the hot end, then you will not need a wick.

Wow thank you for your very fast response.

Todd Trowbridge

thats right. as i understand it, a heatpipe takes a liquid that boils at a low temp, lets it boil, then when it hits the top of the pipe in gas form, it cools, becomes liquid, and falls again. the phase change both ways first absorbs the heat and then releases it. its best to have a vaccum because as was demonstrated to me, in a low enough pressure, alcohol can boil from the heat of your hand. the demonstation was with a glass container that had alcohol in a very low pressure atmosphere. if you put it in your hand, it boiled crazily. you want to apply this principle as well as condensation to heat removal. ive seen heatpipes that stick out from the top of the case and basically become a radiator at the top, to make cooling quick. if i had the facilites, time, and equipment, id definitly try one out. good luck, and PICS! :)

A good solution would be to make a custom block (one big fitting, like 1", make it so there's no where the liquid can go but out the tube, I think Owen's copper cap block would be almost a perfect choice, mounting is up to you) and get a heater core, and tubing. Have the heater core's barb (once again, only one, you'll have to fill the other two) sticking straight out the top, and invert it with the fitting in the top of your case. Run the thick tube between the two, and, well, the vacuum is up to you.

Its not actuallya a vacuum inside the heatpipe since there is the vapor of whatever coolant your using. It would take some math and trial and error but it should be possible to create a heatpipe. You'd have to do the math to figure out the details howevere the principle would be quite simple. Add a quantity of alcohol(or any coolant) to your heatpipe. Leave a small hole at the top of the pipe to allow air/vapors to escape. Now heat the base of the pipe to much higher than the normal operating temp. This will cause the coolant to boil off and should push most of the air out of the heatpipe. Now quickly seal the pipe and allow it to cool. As the heatpipe and vapors cool the pressure inside the pipe will drop.

Although it would take experimentation to get it just right it should work pretty well.

Wow I never thought of that. That's an excellent (not to mention cheap) way to do it.

Water, as in conventional water cooling, apears to be the best liquid for heat pipes use above freezing. This is due to the latent heat of vaporization. The figures are 2257 kJ/kg for water and 896 kJ/kg for alcohol.

A decent explination here:
"Heat is the transfer of energy due to differences in temperature. Exactly how much the temperature changes due to transfer of heat depends on the material in question. For example, it takes about three and a half times as much energy to change the temperature of iron as it does to change the temperature of the same mass of lead by the same amount. The relationship between heat added (or removed), Q, and change in temperature, ∆T, is given by: Q = m c ∆T where Q is the amount of heat, m is the mass, ∆T is the change in temperature and c is the specific heat of the material. In general the specific heat may change with temperature and will change as the phase (solid, liquid or gas) of the material changes. In addition to the heat required to change temperature, additional energy is required when a material changes phase. The energy per unit mass required to change the nature of the bonding in the substance as it changes from solid to liquid or from liquid to gas is called the Latent Heat, and is given by Q = L m, where, again, Q is heat added or removed, m is mass and L is the latent heat. Latent heat for freezing/melting is called the Latent heat of fusion, LF and latent for boiling/condensing is called the Latent heat of vaporization, LV."

As you can see water can transfer much more heat with the limitation of not being able to go below 0c. Unfortunatly water requires a much lower absolute pressure to evaporate as required, that aside comercial manufacturers report that effective thermal conductivity can range from 10 to 10,000 times the effective thermal conductivity of copper depending on the length of the heat pipe.

A good graphical explination here

the basic text is Heat Pipes, by G. P. Peterson

C'mon that was not that technical.....BTW...is the text you mention available on line? Googled around a little & couldn't find it...for, ahem ...free. OK so I'm a cheap bastard who's interesed. Actually I am currently looking into the feasability of active rather than passive systems. If the pressure is low enough and the pump strong enough to create a partial vacuum, could a fluid loop take advantage of the latent heat of vaporization? Im sure there would be lots o' cavitation involved, and that the pump would have to wear well under those conditions, but could it be done? With most pumps water would not be feasable ...how bout Freon...123 or 111 perhaps? Not much of a vacuum required with those to get 'em to change phase, and a portion the of pump power will actually serve to cool the load. I dunno... not that deep into it yet to know if it is doable.

I think the problem with water may be that it has an insanely high boiling point! I'm pretty sure that the boiling point is the lowest temperature your system can be after stabilizing. Yeah yeah, the vacuum thing, but do you really want to deal with that much vacuum? Why bother with actively doing it? That will be a bit of a hassle, to say the least, and I doubt it would turn up much. The highest I've seen a manufacturer reporting is 200,000W/m^2k, but that was for a flat plate. I think that a well done (heat pipes are in their relative infancy, in my opinion) heat pipe will offer all the performance we'll ever need.

take a look through this

http://www.benchtest.com/heat_pipe1.html

he has made his own heatpipes it might be usefull.

Yes maintaining a vacuum deep enough is the main issue. For water to evaporate @ 10c (50f) it must be maintained at about .18 psia or about 29.56" of vacuum. That is why I am intrested in other fluids, R11 has a boiling point of about 23.8c (74.9f) at 0 psig... in other words in a 72f room it would remain a liquid in an open canister but would immediatly boil if placed in your palm. Unfortunatly it does not have the incredible latent heat of vaporization that water sports.

Here's a good phase diagram of water:

http://www.lsbu.ac.uk/water/phase.html

So from what i gather, a heatpipe with alcohol is very possible and even plausible but it will be about 1/3 as efficient as water in a vacuum. That statement is a bit misleading because it will still be more effiecient then just a copper block. Also while were on the subject of pumps, what would make a good wick for a heat pipe?

Todd Trowbridge

you could use 3 or 4 heatpipes with alcohol instead of one with water... I think this is how the shuttle ICE air collers heatpipes are done?

The current design is looking like 3 x 3/8 inch pipes with a varying amount of acetone ... testing will determine how much.

Todd Trowbridge

as soon as you guys master the basics, you may elicit some useful comments
have fun posting foolishness in the meantime

a factoid is not a device
nor will a collection of factoids yield a effective device
successful designs are based on integrated facts

Does that book talk about about tact too? :-P

BillA is one of those people that comes with procooling and there ain't much you can do about his approach to helping people. To be honest, I rather like his methods of "pointing people in the right direction" rather than spoon feeding them the answers and waiting for them to come back at the next hurdle.

You could do a LOT worse than listen to the advice he is offering.

He may be a little "grumpy" at times, but that is his way.

I'd still wager that he knows more about cooling than most of the members put together.

8-ball

Thats cool, sometimes you have to trade other skills for intelligence .. people skills in this case? lol. Thats cool tho. Either way, is he saying that using alchohol would be a stupid approach? Im a bit lost on that. Also what would be a good wick to use if i had to fight gravity?

Todd Trowbridge

I think what BillA is trying to do is point you in the direction of useful references which should help you to gain the knowledge required to manufacture a heat pipe.

I am of the opinion that this approach is much more beneficial to the community as more people will gain the necessary understanding of cooling principles rather than the same people having to answer the same questions over and over.

That's not to say you shouldn't expect any help, but people are much more likely to get involved in a discussion if they believe that the others involved are as keen to actually go out and read around the subject and possibly bring something back to the table themselves.

With regards to manufacturing heatpipes in general.

Heat pipes are more often than not, a large scale product requiring very expensive and large machinery to construct effectively, hence the need for large quantities to be produced. This machinery can produce heatpipes at low cost, better than you could with quite a bit of money but none of the machinery.

The wick is a perfect example of this.

A common wick technology is to sinter copper powder onto the internal surface of the heatpipe, but this is not something you would be able to do in your home workshop.

If you want to make a homemade heat pipe, I would suggest using gravity thus negating the need for a wick.

8-ball

Does anyone know where I can "legitimately" acquire this work? My local library does not carry it, and I don't want to buy it. I don't really like the idea of an entire book, and it will probably have a lot of technical terms that I just don't understand yet. :(

try on-line used book dealers, very hard to find - just keep looking
$120 otherwise (as I recall)

or get smart, get it through an inter-library loan

note: not an intro, this is an engineering text for practitioners

The best bet really is the intra-library loan. Really, the reason that the book is there is that the necessary information can't be boiled down into a 4-page article on the web. It's a multivariate system that depends upon too many factors to have one simple "best" design and manufacturing technique.

I'm just barely smart enough to realize how ignorant I am on the subject (and I have doubts about that smartness on a daily basis).