Building A Ranque-Hilsch Vortex Cooling Tube

The Ranque-Hilsch vortex tube is an interesting piece of equipment. It can, without any moving parts or chemicals, separate hot and cold compressed gasses that are passed through it. Interestingly enough, you can cobble one together with very few parts for fairly cheap. [Otto Belden] tossed one together in a weekend back in 2009 just to see if he could do it. His results were fairly good and he shared some video tutorials on its construction.

His latest version, which you can see in the video below, takes compressed air at about 78degrees and spits out about 112degrees on the hot side and  8degrees on the cold side. Not too bad!

44 thoughts on “Building A Ranque-Hilsch Vortex Cooling Tube

  1. I came to use such devices while installing sensitive instruments in extreme hostile environments. Compressed air is an expensive energy. The noise and the heat are problematic waste products.
    No discussion of vortex tubes is complete without looking at where, when and who brought them into the lab and industry.
    Vortex tubes work because of demons.
    Google “Maxwell’s demon”.

    1. The noise of a piston-driven air compressor for a Ranque-Hilsch vortex tube, can be fixed readily by using any suitable rotary pump to compress air rather than using a piston-driven air pump. You know this. The heat generated … You are complaining about heat? You must live in the tropics in an air-conditioned facility. Your objections are inconsequential in the extreme. May I suggest you head back to your chalkboard…? It sounds like you have a lot of valuable hands-on experience that means you could make further improvements. I hope I am not being overly critical here.

  2. Now I really want to know what the physics of this are and why it works.

    I am also curious how compressed the air has to be, and if it’s possible to do this on an enormous scale for something like refrigeration or HVAC.

    1. Also, does the hot end have to be threaded? It seems like in that first link he just uses a cone of wood to partially obstruct the hot end, with a pretty badass adjustment lever setup.

      1. In use, the vortex tube maintains two coaxial vortices: An outer vortex that travels toward the hot end and an inner vortex that returns to the cold air outlet. See the wikipedia page []. The conical nozzle (in this case a machined end of a screw) separates the hot outer fluid from the cool inner fluid. The threads are probably just used for adjustment, by I did not read the article.

        It looks like the exact mechanism for the temperature separation is unknown, though Wikipedia provides some tantalizing hints.

        In any case, between this and the X-ray tube we have a great amateur scientist day at HaD! Kudos.

          Hotter molecules travel faster and “fly” outwards while colder molecules travel in smaller circles,
          when all reach hot end exit, the more energetic molecules exit, whereas less energetic molecules hit the obstacle cone and change directions and create or enter the inner return vortex.

          on the way to the cold exit, the more energetic molecules in the inner vortex get caught up in the inner vortex and get sucked outwards to back towards the hot end.

          Just gotta become a molecule and visualize the journey my friend.

          – Mohinder2020

    2. Hilsch tube cooling is not energy efficient. I doubt that you will ever see Hilsch automotive air conditioning, let alone cooling for an entire house.

      This does not mean, however, that there aren’t practical applications. One application I know of is to cool sewing needles on high-speed industrial sewing machines. The Hilsch tube allows you to generate cold, on the spot, right where you need it, utilizing compressed air, which is common in an industrial setting.

      Another really neat application is the integration of a Hilsch tube and bimetallic themostat into a threaded package that allows for installation in “knockout” holes in control panels. Let’s say you have a motor starter panel that is prone to overheating. An easy way to get cabinet temps under control is to punch out one of the knockouts, thread the hilsch/tstat gizmo into the hole, and secure it with a nut, as you would a conduit fitting. Attach a source of compressed air, and away you go. Here are examples:

      1. I understand their lack of efficiency. Just curious as my A/C went out on my Jeep and I’m thinking of converting my A/C compressor for on-board air. Back in my youth, pumping gas, we used to open up the air hose and inflate our jumpsuits with cold air. It was amazingly refreshing on a hot, humid day. I was wondering, because I don’t use A/C much anyway, if I should do my on-board air conversion and run a vortex tube to cool my cab and exhaust hot air outside. I already have an ExAir Vortex tube sitting around. I really only need cooling while sitting at a stop light or in stop and go traffic. And, I understand that the air would need to be dehydrated–probably the biggest challenge as it would be easy to pre-cool through a coil and store, more cooling, in a tank tucked under the frame.
        Just curious, beyond the efficiency issue, if any of you fellow hackers have tried such a set-up. If you have, good or bad, I’d really appreciate feedback. Thank you all kindly.

    3. I wonder if the dual vortex idea on wiki is a mere theory, or weather it has been scientifically verified.
      Whatever the case, it is not hard to see the thermodynamic principles at work here.
      Basically, the hot air at one end is due to the air beeing squeezed, or compressed.
      The cold air at the other end, is from the expansion of air as it leaves the device.

      Go and do an image search with your favourite search engine. Make sure you ignore some of the vortex tube where the insides are just straight out. The legitimate ones will have the inside air space shaped to funnel the air at one end for expansion, and at the other end for compression.

      After seeing commercial designs, i think there is room for efficiency improvements if the expansion and compression rates were made closer to a hyperbolic, or parabolic shape. After all, do you think a sattelite dish, or a solar energy collector would be anywhere near as efficient if it were to have a linear expansion rate instead of its usual mathematicaly derived progression. This same princple of hyperbolic and the like expansion rates, are also used on speakers (ie “horns”), which raise the efficiency…

  3. why does everyone call it a cooling tube when it’s just as well a HEATING tube.
    wonder if just blowing into it would be enough…heated end for summer, cooling end for winter, just blow…

  4. The late great Owsley Stanley had a great theory about how the same effect could cause global ice ages.

    I’m not sure if I agree with it or not but he tells a great story!

    It’s posted in 5 min sections on Youtube if anyone is interested in hearing it.

  5. My milling machine came with a commercial version of one of these, allowing dry shop air to cool the old computer.

    It is absolutely essential that the air be very dry, otherwise it will condense. These are also used to make snow at ski parks.

    If experimenting, don’t forget that PVC can shatter, especially when cold.

    I was very facinated by it. Neat stuff, but it requires a lot of energy and most of us don’t have sufficiently dry air, especially in summer.

    1. I have used these in machine shops in the place of liquid coolants. Not all machining applications work well with liquid coolants because they are VERY messy and in the case of CNC machining cast iron, the cast iron gets washed around in the machine and becomes more difficult to ensure that you got it all out at the end of the job. If you don’t get it all, it becomes an abrasive and can ruin your machine’s ways. They work very well in that application. I’ve never heard of using them to cool a CPU which would probably work just great, but probably much less energy efficient than a peltier, liquid cooling system, or simply upgrading the fans and keeping any filters clean.

      1. The machine came from GM, so I’d guess that clean, dry shop air was virtually unlimited.

        It also needs air for the coolmist system, and to the bijur lube pump for the ballscrews and ways.

  6. Wikipedia says that “Commercial vortex tubes are designed for industrial applications to produce a temperature drop of about 45 °C (80 °F)”.

    He got 70 °F. Does not sound bad at all.

    You could use this to precool the Peltier cooler to get it to the lower temperatures.

    But I’m too wondering that what happens when you stack the tubes.

    With Peltiers you can do so and I do not the see reason why not here. Of course you get diminishing returns.

    Please correct me when I’m wrong.

    1. i imagine you will hit diminishing returns pretty quickly, the effect of the tube is based on the rapid movement of the air in the tube, hooking one of the outputs to the input of another tube means much less pressure, thus a much diminished effect (could even be so ineffective as to cause back pressure that breaks the vortex and renders the whole thing useless)

  7. Many years ago I saw these in operation in a manufacturing facility. Impressive. Then I asked how much they cost – incredibly over-priced! I applaud this effort to show how affordable the hardware can be.

  8. What I’ve long wondered about these is if it might be possible to make a high volume, low pressure, lower velocity version?

    Scale it all up, have an input pipe one inch or bigger. Target goal of 60~70 degree F cold output from higher than 70 degree input.

    If the thing can be made to work that way, ram air Hilsch air conditioning for moving vehicles? Scoop in air while driving 55+ MPH, dump the hot product outside the vehicle and the cold inside.

    1. That’s an excellent idea. As it is, air-conditioning can have a fairly significant effect on the fuel efficience of a vehicle. This type of system would have a very minimal effect on efficienc, just a small amount of drag.

  9. Bear in mind that compressing a gas in the first place uses a lot of energy just to shift heat (i.e. just like in a fridge compressor), and is wasteful if the heat is not being reclaimed.

    To then use the compressed gas in a device that is perhaps 2% efficient to cool things is even more wasteful of power.

    If you’re going to compress gas in the first place, and energy is scarce, you might as well use a fridge compressor for cooling.

    Extra marks and major kudos however if you create an ammonia absorption refrigerator that works off the waste heat of an air compressor…. Google
    “crossley icy-ball”…

  10. You get about 15psi (102.2kpa) at the front bumper of a car going 100km per hour. To get to 80psi, you’d need a scoop about 50cm^2.

    This would give you about 200kg of drag (someone double-check my math). Pretty steep.

  11. acomputerdog…your mathematical equation is incorrect…..your step number 4 is mathematically incorrect…dive deep on what you did there and you will fund your mistakes…..Math never lies….your skills did…..

  12. The original use of the hilsch tube was used on trains to cool water for drinking, because trains used compressed air. Also we did an experiment of stacking two tubes and it does produce lower temperatures.

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.