Home-made flexible heat pipes
We wandered about this idea when we assembled fanless computers… We had to bend heat pipes to obtain close fit of thermal blocks; moreover, more than once, we had to disassemble the construction and waste our time. In addition there was a theme in the forum, where modders discussed utopian possibility of copper ends joining with a rubber hose :) I mention at once that modding given below is not more than an amusing experiment that we conducted just for the love of the game. We worked out and realized a functioning model of flexible heat pipe, to put it more precisely, it was a big pipe with 16 mm diameter :). But not everything went off smoothly…
So, let’s start step by step. To carry out this experiment, we had to sacrifice a nice pressure relay in good working condition. We cruelly cut out the main component of future flexible heat pipe, a stainless convoluted tube, of the relay. So, cut out the flute with the help of angle grinder.
Let me make a note that the convoluted tube is a real work of art. There are no joints and irregularity, like a cast one… at the same time the thickness of the tube walls is 0.2-02.3 mm! I have no idea how they produce such tubes… Then, sort out a common copper pipe of appropriate diameter.
Then, prepare soldering accessory and carefully solder the convoluted tube to the copper pipes. Scrape the stainless covering with an emery-paper and using special flux.
It may seem that the work is finished.., and that you may plug the hole, fill up the tube with actuating medium and pump air out, may not you? I don’t think so… What will happen with the convoluted tube when we pump air out? Any of the readers of this article can guess that the tube will not be flexible. The convoluted tube will shrink and our experiment will be ignominious failure…
We’ve thought over the situation before breaking a good relay. To prevent the convoluted part of the pipe from shrinkage, we make two steel hinges and fix them parallel to the convoluted tube. As the whole construction is soldered, it’s easier to solder the hinges as well. We did so.
The result is shown in the picture. Of course, flexibility is not perfect, but it’s a good result for the first experiment.
Then, it’s necessary to test the heat pipe when in use. Plug one end of the pipe by soldering a coin to it. Fill up the pipe with rated quantity of actuating medium, in this case, it is two table-spoons of distilled water. We made a fuse of thin wire but decided not to use it, it’s not necessary yet…
Use a coin for the second plug as well. Drill the coin and lead out a thin tubule to the vacuum pump.
Solder securely the both ends of the tubule: one end to the coin, the other to the compressor nipple. The pump is very small, so it will take the pump 20 minutes to finish its work…, alas, I couldn’t find anything better. Drink tea with cakes listening to the chatter of Japanese motor. During this tea-break you have a good possibility to have a discussion about the results of the modding and plan another modding.
After the necessary quantity of air has been pumped out switch off the compressor, pinch the tubule with a clip, nip it off and solder up. If you have decided to replicate the experiment, the sealing technology is described in details by refrigerationists and air-conditioning devices mounters.
As a result we’ve got such strange thing. Test flexibility of the heat pipe. As we hoped, the pipe can be bent but with a little forcing, pressure cannot but give its effect. It even seemed to me that hinge brackets bent a little in spite of the fact that they were made of slab steel of 1.5 mm thickness…
Keep on modding according to the schedule. Dip the heat pipe into a glass with boiling water and ascertain normal operating. Everything was so good that we became bored and decided to sophisticate the construction!
Strictly speaking, we decided to shape this flexible cooler up to make it appropriate for mounting into fanless computer. We decided and made so.
Take two copper disk-shaped workpieces of 5 mm thickness. Saw chases by means of angle saw for better heat dissipation.
Lap the heat dissipating surface of the copper workpieces with a lapping plate. You may lap the underside as well for appearance. It’s desirable to polish it but we didn’t do it.
Sort out two iron hemisphere of appropriate size.., I couldn’t find a good name for them at the time.., in a word, two housings or casings for heat exchangers.
Drill holes in this steel casings; the holes diameter should be equal to the diameter of the future heat pipe. Scrape the metal with a file and solder safely the joint. Of course, one should use acid flux for such operation.
Solder copper heat remover in the same way. If somebody takes risk of replication of the construction, I reveal you a small ruse how to solder such massive component with a handy and small soldering iron of capacity only 65 watt. For that purpose I have a 2000 watt construction fan at hand :) Before soldering the fragment, it’s enough to heat the place for 10 seconds, then you can work with a low-capacity soldering iron. If you solder a 150 gr copper disk, assist yourself with the fan all the time, for such work even 100 watt soldering iron is not enough.
At the other side solder similar “deflector”, such name was given to the device :) To run a few steps forward, I pay your attention to the quality of soldering of the pipe to deflector… Later we took a lot of troubles over this part when pumping air out.
As a matter of fact, there is the result of our work in the picture. It looks horribly… But we didn’t plan to send this device to the beauty contest, so let’s proceed to air exhaust and testing the device.
We had to stop air exhaust and solder joints anew three times… The reason was low quality of metal of which the deflectors were made. The matter was not bad soldering but namely the metal quality. I strongly recommend NOT using iron! Badly soldered fragments appear from nothing, as a result airtightness is broken and air exhaust becomes impossible. It’s very difficult to find minute cracks. To do that, you must generate overpressure in the pipe and dip it into water. The location of the crack can be found by bubbles, but they are very small and difficult to notice at once…
We succeed in air exhaust only from the fourth attempt. Test the pipe again in a glass of boiling water. It works perfectly! The pipe bends almost to 90 degrees. And it springs funnily :)
Because of troubles with airtightness we didn’t try the work of the pipe with a fuse… The first model, which doesn’t have deflectors, will work with fuse perfectly, no doubt. Second variant is not so unambiguous. We planned to make the fuse in the form of a ring and press it to the heat remover. I’m not sure whether this will work. I’ll give a report when we try doing this. The article was waiting to be published for two weeks while we were looking for appropriate copper deflectors without success…
But the primary goal of the experiment has been achieved. We made a working prototype of a flexible heat pipe which is good for processor cooling.
I wish everyone good modding SilentMods.com