Make a Heat Pipe! / General Science and Electronics / Forums

Forums

4hv.org :: Forums :: General Science and Electronics

« Previous topic | Next topic »

Make a Heat Pipe!

1 2 next

Move Thread

LAN_403

Nik

Sat Sept 17 2016, 01:26AM

Registered Member #53 Joined: Thu Feb 09 2006, 04:31AM
Location: Ontario, Canada
Posts: 638

We've all seen them but I've always been curious about how hard it would be to make one and how well it would work. (Turns out it's really easy and they work really well!) After a short Google search revealed that a head pipe is just a tube filled with a substance in both its liquid and gaseous state, the most common one seems to be water with as hard a vacuum as possible pulled. Water at the bottom of the pipe boils into steam which travels up the pipe, condensing on the walls and giving up the heat it absorbed at the bottom. So it was time to hit the lab!

Hypothesis: Heat pipes are pretty simple and should be trivial to build.

Materials: Copper pipe, Copper fittings, Soldering gear, Flat Black paint.

Procedure, Building: After putting a cap and a valve onto a 70cm length of pipe I painted it flat black (for better thermal camera performance). I put a little water into the pipe (less than 2ml) and then headed the pipe with a torch until the water was boiling, the pipe was hot all the way to the valve and there was a lot of steam leaving the valve. Then I quickly closed the valve and turned off the heat.

Procedure, Testing: I hung pipe up with some wire to try to isolate it from any of the heat conducting surfaces, attached a thermocouple to the top and bottom of the pipe and then placed the lower end into a dish of water on a hot plate (the water was to get better thermal contact with the pipe).
I let the pipe heat up for a couple minutes while recording the temperature and IR video, after it was warmed up I chilled the upper end with some canned air, recorded the temperature, removed the pipe from the hot plate and then chilled it again.
Next I emptied the pipe and did the whole heating/cooling cycle again.

Observations: Aside from one interesting hiccup it worked very well. Heating the lower end of the heat pipe caused the temperature of the whole pipe to rise fairly evenly with only a 5 degree gradient from top to bottom for nearly the entire heating and cooling cycle. Running the heating and cooling on the same pipe while empty showed that the copper pipe itself does not conduct heat very well all on its own.

First graph: Heating the heat pipe. There is a delay of almost 2 minutes before the pipe begins to function. This is due to an oversight on my part, with degassed water there was no nucleation points for the boiling to begin. Once the boiling started the temperature at the hot end dropped sharply and the temperature along the rest of the pipe began to rise quickly and evenly along the length.

Second graph: No surprises, cooling the upper end of the pipe very quickly cooled the hot end even though it was still sitting on the hot plate, after removing the hot plate I cooled it further and the whole pipe quickly returned to room temperature.

Third graph: With the empty pipe I followed the same procedure as the full pipe. The heat conduction from one end to the other was terrible, worse than I had expected given how many times I have burned my hand on freshly soldered pipe and wire.

Conclusions: Any one can make a heat pipe! A home made heat pipe might be just the thing you need to cool your high power silicon. By putting a radiator (like the kind used for PC liquid cooling) at the top of your heat pipe it would be possible to get some fantastic cooling compared to an aluminum brick style heat sink.

The Setup:

First graph and thermal video:

Second graph and thermal video:

Third graph and thermal video:

klugesmith

Sat Sept 17 2016, 04:46AM

Registered Member #2099 Joined: Wed Apr 29 2009, 12:22AM
Location: Los Altos, California
Posts: 1716

Excellent work there, Nik. Thanks for sharing. Your imgur host has plenty of entertaining links off to the side.

All heat pipes are sensitive to gravity, which helps condensed liquid return to the heated end when that end is lower. Since you're all instrumented up, could you see what happens if the tube is tilted the other way, or vertical with heat flow upward vs. downward?

The commercial copper-and-water heat pipes I've seen open have a porous coating on the inner surface, to serve as a wick. Maybe regular lamp wicking would work in home-brewed HP's, at least ones the size of a pencil or soda straw. Commercial ones that size can be demonstrated by stirring a cup of hot water & having to let go within a couple of seconds. I haven't tried the complementary experiment, stirring cold water.

Nik

Sat Sept 17 2016, 03:15PM

Registered Member #53 Joined: Thu Feb 09 2006, 04:31AM
Location: Ontario, Canada
Posts: 638

Thanks, and cat curling is a noble sport, we are made to learn it here in Canadian publick schools.

Like you said, commercial pipes all have a a wicking material built in to fight gravity. If you heat the upper end (in my non-wicked pipe) it acts more like the empty pipe with the heat being transferred down by conduction of the copper. I did think about loading it with a wick but couldn't think of something appropriate to use, lamp wick is probably what I needed >_<

If you hold my homemade heatpipe at the top and put a lighter to the bottom you feel the heat within 5 seconds and have to put it down shortly after. Cooling the lower end does cool the whole pipe but not as quickly as cooling the stop. Eventually I want to have a finished, desktoy, version that doesn't have the unsightly valve on it. I will definitely be looking to have a wick in that one.

2Spoons

Sun Sept 18 2016, 10:34PM

Registered Member #2939 Joined: Fri Jun 25 2010, 04:25AM
Location:
Posts: 615

Some years back a colleague at work made a heat pipe similar to yours. It was nearly a metre long and about two centimetres in diameter. It had one curious property though - shaken sharply up and down it would make a clang, like there was a piece of metal inside. This became the subject of much discussion around the engineering team, with some asserting there actually was a piece of metal left inside. Inevitably a tray of beer was wagered on the contents of the pipe, and it was duly cut open to reveal nothing but water and empty space.
So what was causing the clanging sound? I'm pretty sure that because there was nothing but water and vapor in the tube the water was able to move as a solid slug up and down the tube, with vapor bubbles ahead and behind collapsing and expanding. The clang occurred when the slug of water slammed into the end of the tube as the vapor bubble collapsed completely - since there was no air to cushion the impact by acting as a spring.

I'd be curious to know if yours does the same thing.

On another note, carbon fiber is a commonly used wick material, though your liquid must 'wet' the carbon. I think a water /methanol mix is often used.

Patrick

Sun Sept 18 2016, 10:48PM

Registered Member #2431 Joined: Tue Oct 13 2009, 09:47PM
Location: Chico, CA. USA
Posts: 5639

2Spoons wrote ...

On another note, carbon fiber is a commonly used wick material, though your liquid must 'wet' the carbon. I think a water /methanol mix is often used.

I tested this for you just now. it wet-out totally and instantly, but mine has the Volan coating on it. Not sure if that matters.

]volan_datasheet.pdf[/file]

Nik

Tue Sept 20 2016, 03:27AM

Registered Member #53 Joined: Thu Feb 09 2006, 04:31AM
Location: Ontario, Canada
Posts: 638

2Spoons wrote ...

Some years back a colleague at work made a heat pipe similar to yours. It was nearly a metre long and about two centimetres in diameter. It had one curious property though - shaken sharply up and down it would make a clang, like there was a piece of metal inside. This became the subject of much discussion around the engineering team, with some asserting there actually was a piece of metal left inside. Inevitably a tray of beer was wagered on the contents of the pipe, and it was duly cut open to reveal nothing but water and empty space.
So what was causing the clanging sound? I'm pretty sure that because there was nothing but water and vapor in the tube the water was able to move as a solid slug up and down the tube, with vapor bubbles ahead and behind collapsing and expanding. The clang occurred when the slug of water slammed into the end of the tube as the vapor bubble collapsed completely - since there was no air to cushion the impact by acting as a spring.

I'd be curious to know if yours does the same thing.

On another note, carbon fiber is a commonly used wick material, though your liquid must 'wet' the carbon. I think a water /methanol mix is often used.

It makes the clink! It sounds like a ball bearing is inside (but there isn't!) I even shot some high speed video of it happening in a glass bottle:

I'll have to go though my "bin of miscellaneous fibers" (aka, that damned box of string) to see what might wets bets and has thin capillary structures, carbon fiber sounds fun but I'm trying to keep to a 0 budget. Although it would give me a reason to lay with carbon fiber...

klugesmith

Tue Sept 20 2016, 04:19AM

Registered Member #2099 Joined: Wed Apr 29 2009, 12:22AM
Location: Los Altos, California
Posts: 1716

Here are a couple ways you could record the clinking fluid in slow motion.

1. Neutron radiography, where the metal tube is relatively lucent and the water relatively opaque.

2. Make a heat pipe out of a glass tube. Where your metal tube has a valve, you could close off the top (steam-venting) end with a small rubber stopper, or pinch shut a rubber tube. After things have cooled down, heat the glass near the top & let the partial vacuum suck it closed. With practice, the DIY closures might withstand clinkage shocks (water hammer blows).

Maybe it's been done already & is all over youtube. Like cell phone videos that answer the old refrigerator-light-when-door-is-shut question.

Nik

Tue Sept 20 2016, 04:26AM

Registered Member #53 Joined: Thu Feb 09 2006, 04:31AM
Location: Ontario, Canada
Posts: 638

I have a glass heat pipe (boiler sight glass, it is terrible due to the high head capacity and low thermal conductivity of the glass and plugs). When I get back from this work trip I will film it at 1200fps, the glass pipe doesn't make the clinking noise but more of a tap.

Ash Small

Fri Nov 04 2016, 08:53PM

Registered Member #3414 Joined: Sun Nov 14 2010, 05:05PM
Location: UK
Posts: 4245

Any idea why they use a water/methanol mix in commercial heat tubes?

Is it just due to cost, or does it perform better than straight methanol?

This is a really interesting subject, you gould slit open copper tube, and flatten it out, cut it into squares and drill a hole in the middle, and solder them to the pipe. You could even have multiple tubes, or coil the tubes.

EDIT: I guess the water makes the methanol less explosive during the pipe sealing process? Might be a bit dodgy using a blowtorch with just methanol in the pipe

Kolas

Fri Nov 04 2016, 11:47PM

Registered Member #102 Joined: Thu Feb 09 2006, 08:15PM
Location: Raleigh, NC
Posts: 169

I would think that methanol is used because of its considerably lower boiling point. Allowing the heat pipe to regulate temperatures more friendly to silicon devices.

1 2 next

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob