Peculiar Fluid Dynamics Creates Hot And Cold Air

We’re fascinated by things with no moving parts or active components that work simply by virtue of the shape they contain — think waveguides and resonators for microwave radiation. A similarly mystical device from the pneumatics world is the Hilsch Vortex Tube, and [This Old Tony] decided to explore its mysteries by whipping up a DIY version in his shop.

Invented in the 1930s, vortex tubes are really just hollow tubes with an offset swirl chamber. Incoming compressed air accelerates in the swirl chamber and heads up the periphery of the long end of the tube, gaining energy until it hits a conical nozzle. Some of the outer vortex escapes as hot air, while the rest reflects off the nozzle and heads back down the pipe as a second vortex inside the outer one. The inner vortex loses energy and escapes from the short end as a blast of cold air – down to -50°C in some cases. [Tony]’s build doesn’t quite approach that performance, but he does manage to prove the principle while getting a few good-natured jabs into fellow vloggers [AvE] and [Abom79].

We’ve covered vortex tubes before, but as usual [Tony]’s build shines because he machines everything himself, and because he tries to understand what’s making it work. The FLIR images and the great video quality are a bonus, too.

61 thoughts on “Peculiar Fluid Dynamics Creates Hot And Cold Air

  1. There are so many conjunctive applications for all sorts of Chem eng basics – the permutations of which hardly explored, need suitable lab instrumentation to assess or rather provide definitive means to offer full assessments with, as near as possible, hackfree audit trail as vector to prepare reports for managers answerable to the board of directors- wink wink…

    Anecdote comes to mind, geesh what fun !
    [Wrote a report for a non exec director for a public company re assessing various alloys for anti-bacterial properties re length of time in contact with various food & preservative chemicals – most natural, he was looking at my draft with a young women in attendance, ostensibly for minutes recorded in the board room, then leaned over & whispered in my ear “Why did you refer to asses in the introduction, don’t you know we have lots of women in marketing ?” – LOL]

      1. 1. Chem Eng – short title for Chemical Engineer covers a great deal re energy transfer for various processes
        in all the main 3 forms of heat transfer.
        2. Conjunctive applications suggest that combining approaches might offer hitherto unknown benefits.
        3. The permutations & arrangements are vast, I’ve been in Engineering for decades & now my eldest is
        exploring all sorts of oddities in his PhD at Curtin University @ Bentley, Western Australia. We often have
        our heckles raised how new people in the industry fail to grasp combinatorial signigificance – whether they
        are on nootropics or not ;-)
        4. A massive area not addressed are particular niche areas of special types of lab instrumentation, there
        aren’t enough players, they cross each others paths, have no strategems, minimal industry breadth etc
        5. Data logging with appropriate connection to reporting/interface programs such as Knime & Tableau are
        a start along with IT cloud big data collation eg IBM & Oracle but, still have a long way to go, worth acquiring
        a position in that market place by simplest means to start seeking a collaborative well funded partner Eg me.
        6. Ultimately the market in respect of public co growth appreciates this by increases in share prices and by
        plans to offer dividends sooner & a market/sales/development plan to lift that even more as rewards for all..

        Care to offer expansion, speculation on that or in relevance to these sorts of posts Dax ?
        Especially this one as there are a couple of obvious products that could easily stem from this approach
        short term Eg Where continuous gas flow can offer cooling effects without requiring traditional infrastructure
        .. open to all & especially some interesting pertinent comments from Dax/pongbongo are welcome etc

        Cheers :D

          1. :-)
            At the least I am offering considered opinion with the benefit of decades education, training & experience
            in comparison to Dax’s minimal uninspired simplistic one liners that betray all sorts of facile issues which
            lead nowhere & offer negligible advancement, hey give Dax a problem which can’t be easily searched
            via search engines, put yer jibes where you sit :P

            Rather than anti-intellectual prejudices, consider amending & augmenting your diet, challenge yourself
            with some mental/cognitive exercises & reach out to peers that do likewise – ie get out more, get involved,
            the behavioural patterns with advanced profiling software tell heaps – and it aint as simple as poker …

          2. “At the least I am offering considered opinion with the benefit of decades education, training & experience”

            Or so you believe.

            Seriously, go see a doctor. You’re not making much sense.

          3. I’m agreeing with Dax with regards to the not making much sense. There is information there, but it doesn’t appear related to the topic under discussion except in the most vague and/or peripheral manner.

          4. “There is information there”

            I don’t see that either. It’s mostly just namedropping, technobabble or business consultant jargon, like: “Conjunctive applications suggest that combining approaches might offer hitherto unknown benefits.”

            He’s clearly manic and high on something, whether it’s that “nootropic” self-medication, a blood clot in the brain, or schitsophrenia. Either way, it’s gotten way worse since the last time I’ve seen him around.

          5. I think this is one of those cases where you CAN use too much gooder language and be lost in translationisms.

            And seriously, if you’re going to try and be overly academic, inserting ” a ” where appropriate between words will help.

            Also, Chem Eng deals with a heck of a lot more than heat transfer ;)

  2. How efficient are they? I’m guessing there’s a reason they’re not used in refrigerators and home heating systems. Wonder if you could use that air pressurising drop pipe thing to produce the air, this tube to cool or heat it, then pump it into your home, use solar panels to power the whole water moving side and you’ve got a green air conditioner or something.

    1. They are quite inefficient when you consider the energy expended to compress the air. However they are quite useful in some production processes where you need to cool a portion of a part quickly. I have used them in injection molding production lines to cool parts as they come out of the mold, before other pieces are attached. Think of the little metal clips on car trim panels. If the plastic is not cool (hard) enough, attaching the metal clip will deform the plastic.

    2. Let’s just say if you wanted to replace a bar fridge with one, you’d probably have to tow the compressor with a 1 ton truck and leave it parked outside on it’s dual axle trailer.

    3. Not efficient, but compact and portable. One of the best applications I’m aware of is cooling something like a dust suit or exposure suit. These are small enough to not get in the way, but still provide some meaningful cooling.

    4. If you run them off air produced in a trompe, because you just happen to have a supply of gravity fed water, and need both the hot and cool sides for different tasks (or different stages of the same task, such as distillation) then you have a very useful, low maintenance system.

  3. My first encounter with one of these mind-blowing devices (pun intended) was coming onto a new job 7 years ago. My new boss was showing me how to use the department’s vertical milling machine (which differed significantly from a standard Bridgeport), and without saying anything about it he turned on the vortex tube he used to cool the end mill and blosw chips away. Well I had some questions about that, let me tell you! Unfortunately, he didn’t really know the science behind it, so we just got to work. I used it from time to time as needed while working there, but always forgot to look it up by the time I got home from work.

    1. Very good point. People use them for milling. But the usefulness is debatable. While it provides very cold air, it it provides only little of it. If the pumping stream is used to cool the workpiece directly, for most applications cooling is vastly better. It’s really only for cases in which you want to go below room temp.

  4. I’ve always wanted to build one of these as no-moving-parts HVAC for a car, with a funnel mounted up front for capturing the air. Then I can choose to vent the hot or cold streams into the cabin as I see fit. I’m sure it wouldn’t make a huge difference on its own, but it would be a nice auxiliary system.

      1. What exactly is the minimum pressure needed? Or flow rate?

        Suppose you make a large version out of storm drain pipes or concrete tubes and supply it with air from a wind funnel or possibly a windmill, or prehistoric Nubian slaves in the Pharaoh’s court, could you make a meaningful temperature difference to fridgerate a small room? Make ice?

        There’s also a type of air compressor possible where there’s a small waterfall. You basically have a float under the waterfall with a hollow cavity, and the bubbles trapped in the falling water rise up under the float, which is weighted down and therefore collects air pressure inside.

        1. I wish I knew, honestly. Call it a gut instinct, but based on what is known about pilot tubes, you could likely only expect about 2-2.5 inches of water column at highway speeds (70 mph). So maybe that could be overcome with added flow? That would be an interesting question to answer… can you achieve the same effect with very limited pressure, but lots of flow?

          Pulling from gut instinct again, I suspect that the higher the air velocity you have, the better the effect. If that is true, pressure would be a very important factor in getting a vortex tube to function. Regardless, I think you’re right that you’d have to scale up to be able to work with lower pressure, but the effect may still be there.

          1. It’s trivial to turn low speed and high volume into high speed and low volume. It’s just a matter of how large a funnel you want to fix to the car. ´The obvious problem is the extra aerodynamic drag, because the system is very inefficient.

            I was rather interested in making a passively powered system for a house, for cooling a closet or other small space enough to make ice and/or preserve food. Something that you could in principle make out of clay and rocks and bits of wood (animal skin bellows?), that could blow cold air into a cellar of some kind.

          2. I don’t think that’s trivial at all with what you propose. With such a funnel you could turn low velocity into higher velocity, but you’ll see a negligible change in flow rate as very little compression is taking place. Keep in mind, you’re working with a source (ram air) that only offers a couple inches of water column in pressure… that’s very little, as in less than you could blow through a straw. In the end, you will see an increase in velocity, but not a meaningful amount (depending on your goal) and there will be a point where you will see a loss in air velocity as you continue to reduce the size of the exit.

            Ultimately you’d need to calculate what the maximum velocity you could achieve with the pressure you have available. You’ll end up with an idea orifice size to serve as the funnels exit. From there you can work backwards using the flow rate and known amount of compression to determine what the ideal inlet size would be for the funnel; going smaller would reduce your velocity/flow rate and going larger would unnecessarily increase your aerodynamic drag.

            That’d be some very rough math, but could get you in a ball park of what is feasible.

          3. As far as I understand, the pressure isn’t really the important point but the centripetal forces in the vortex, which creates the stratification or air. There’s hardly any pressure in the vortex tube itself – just in the nozzle that injects the air into the tube.

            If you can create a strong swirling vortex without high pressure, such as through a long gradually narrowing cone, it should do more or less the same thing.

          4. But therein lies the problem… you’ll have a great deal of difficulty generating the high velocities you’ll need without the pressure to serve as the driving force. Taper the cone as much as you want, there will still be a limitation to what you’ll be able to achieve. I fear that to make it work at the pressures we’re talking about, the device would have to be comically large and use an immense amount of air.

            Although that spawned an interesting thought… what about mechanically generating the vortex? Say you make a set of vanes that fit inside the tube, shaped in the shape of a cross and use a motor to spin them. Do you think the same effect could be reached?

          5. “Do you think the same effect could be reached?”

            I don’t know. I’ve already been thinking about making a very wide flat channel that spirals onto itself like a nautilus shell. I’ve also thought about spinning the vortex tube itself. Gotta have to simulate it, or build it, and I don’t know how to simulate fluids.

          6. “the device would have to be comically large and use an immense amount of air.”

            For a car perhaps, but for the purpose of cooling a cellar… could be the size of a barn for all we care. If it runs on wind and makes you ice blocks for free, and it’s fairly simple to build out of dirt and sticks, size can be excused.

            Forget about air conditioning a car – think of an African village that could build a communal cellar that’s cold enough to freeze meat.

        2. Tangent to the above discussion –

          Instead of a vortex, what about a once-through system. Think of a flat plate that makes a sharp bend inwards, and air is blown over the bend. If the centripetal force of the airflow turninng causes hot air to bunch up closer to the plate and cool air to collect on top, the two flows can be separated by a vane.

          Once separated, most of the colder air can be directed back to the intake for another go, creating a self-amplifying loop, while the remainder gets used for whatever application you wish.

        3. Dax wrote:
          “There’s also a type of air compressor possible where there’s a small waterfall. You basically have a float under the waterfall with a hollow cavity, and the bubbles trapped in the falling water rise up under the float, which is weighted down and therefore collects air pressure inside”

          Our good buddy Dan#942164212 mentioned “trompe” a few comments above…

          https://en.wikipedia.org/wiki/Trompe

    1. Nice idea & most logical too – commended Brad, foremost in my mind too now & then, always on the lookout
      for new volatiles to accomplish that in conjunction with efficient mechanical systems, some 20,000 new chems
      roughly each year just in that field & getting mre exasperated keeping track, prefer the natural organics of course
      and then can achieve such mutually inspired aim…

      Fwiw Einstein & his student thought of that general issue re no moving parts too some considerable time back :-)
      https://en.wikipedia.org/wiki/Einstein_refrigerator
      and
      visitors here from phys.org (Dax !) might think about contributing specifics for a change instead of slanted one
      line comments dragging tripe from other forums when they have the opportunity to possessively infer, deduce
      or just plain ask pleasantly crafted intelligent questions – aye Dax/pongobongo/Benni et al :P
      http://phys.org/news/2008-09-einstein-green-refrigerator-comeback.html

      Bum, site seems down for me at moment, could be the satellite proxy since I’m bit away from land &
      thuraya has minor issues now & then :shrug:

      ps: Reply meant for Brad but maxillian comes up oddly in header, cant change it – edit function… ?
      pps: Dax, you have a history of launching into insult or crappy tangential put downs, please consider instead
      stimulating that imagination that is so dormant as your other nick shows on phys.org – who are you there ?
      The guy who pedantically complains on mindless minutiae/spelling or offers developed links, useful
      comments offering opportunity, something beyond banal one-liners for a change please – can yah ?

        1. He just needs some imodium for his logorrhea.

          Though seriously, word salads and rambling, and general inability to understand or produce language in the old are usually symptoms of a stroke. People with mild receptive aphasia produce unnecessarily large volumes of text or speech, often confused, off-topic and lacking any clear meaning or point, but the person himself doesn’t notice because the area of the brain that would evaluate the output is damaged.

          The victim thinks they’re talking normally, while everyone else is scratching head trying to figure out what the hell they’re on about.

          1. Although you are way off topic, examine the details of my posts & the connectedness along with
            approachability to advance beyond the simplest interpretation of Physics. Math & descriptive choices,
            rather than inane one liners betraying dull lack of interest, try & entertain some complexity free of fear
            :-)

            Its only a “word salad” to those who have difficulty understanding how these issue relate, just take a
            few minutes to “join the dots” , I advise take it step by step. Instead of making put down claims to
            bolster your lagging self esteem, instead ask smart (technical) questions, that would be far more sensible
            & offer dialectic which the interested can engage in too & explore maturely & which the more skilled
            such as those in the medical profession would appreciate as challenging & provoke wider study…
            Unfortunately many who are not skilled across multiple disciplines conflate intent to educate essentials
            with unusual breadth, with a cognitive aspect, which is doubtless also beyond their understanding.

            I’m sorry Dax & your casual idle supported nicks missed out on higher education, as I stated to you on
            phys.org some months ago – its not your fault unless you stay in that comfort zone. Rather than
            snipe, lift your game be rather more technical at least & learn – it offers resistance to propaganda :-)

            Suggest study up a few papers re methods of enhancing the widest range of cognitive skills, nootropics
            not essential but, at the least start with combinatorial changes to diet, to improve; attention span,
            depth of permutations in language construction, how to assess feedback – along with that – honestly
            at what is the intent of your barbs. I’m fortunate I recently completed (with distinction) food science
            including microbiology at Curtin University after nil biology/biochem from electronics base.
            Do you imagine I could have done that if I had any cognitive deficit – if you doubt, check me out, my
            student no is 7602128 http://www.curtin.edu.au/ though of course they will likely require a release from
            me re qualifications/degrees, from 1976-82 then 2008 then 2010, now informal study into nootropics of
            my own formulation with trials planning underway with serious funding… Ask rather than facile claim ok ?

            Um, Do you have Any underlying interest in subjects here & prepared to delve outside your (contracting)
            comfort zone, hey test me – ask questions – something rather beyond idle inane simplistic reliance upon
            search engine results as many on phys.org seem to have to rely upon – ffs ;-)

            One clear indication of age related cognitive decline – even a little are variant tell take signs of fringe
            contracting comfort zones. Yes I’ve seen a doctor re checkups for cholesterol, blood pressure,
            liver function tests (as on high mineral complexes) etc – Suffice to say most doctors can’t keep up
            & somewhat rattled the few like me have the presence of mind to self medicate & maintain a
            growing skill set despite their age as well as improved physical fitness too…

            Start with languages and a testable predicate “..Intelligence inextricably linked with increased
            vocabulary skills – such that increased exercises in acquiring wider understanding of language
            aids intelligence and increased intelligence bearing exercise increases vocabulary range…”
            The link is an imperative across all walks of life – oh & few exercises in touch typing will help too :P

            Thats the last from me here, try harder next time – don’t be afraid of even a little complexity, cheers…

          2. See, this is what I’m talking about.

            I’m not being facetious here. You’re jumping from idea to idea, rambling and making incoherent remarks and disconnected references. Please, see a doctor.

            “now informal study into nootropics of my own formulation with trials planning underway with serious funding…”

            Are you eating amphetamines?

          3. Wow. The cut and paste that adds all the extra returns is pretty irritating. And the dielectical escretascratum of the jargon has jumped the tracks some time back.

          4. This reminds me of how covert ops will leave messages encoded inside of a seemingly legit piece of text – a newspaper or radio ad for instance. In this case, it must be a dammed novel of a report back to HQ…

        1. Well that is to be expected, because you are incompetent, fortunately most people on HAD are not. I expect your just try to strap a huge aluminium heat-sink onto the roof of your car…

  5. I’m probably on the wrong track but the simple explanation to me of how it functions is that hot air is lighter than cold air which is why hot air balloons fly. So any volume of air is a mixture of hot and cold air and the temperature of any random air volume is the average temperature of the total. When you spin a volume of air the heavier cold molecules are flung to the outside and the warmer lighter volume moves to the center as in any centrifuge. This seems so obvious it is probably wrong.

    1. Close – though the molecules don’t change in mass – the average kinetic energy of the hot
      molecules is higher (density lower in equilibria with centripetal forces) – there are some infra red
      radiative transfer issues as well but, the bulk is squarely in respect of Statistical Mechanics
      fundamentals. There are also resonance issues as well which are interesting & could well be
      aided by appropriate transducers to improve efficiency obviously at the expense of adding
      equipment, tuning & energy input – though it may not be that significant – depending on the
      size of the vortex flow channels (& maybe more than one staggered etc) & the expected
      mass flow unit operation equivalent – probably simplified by a cross section single molecule thick
      function integrated over the path length – close to some idealised rocket science flow parameters
      & who knows it might be applicable if you can inject fuel & ignite at entry to hot end, which of course
      would increase flow – perhaps disastrously… Reminds me of a can of degreaser as accelerant
      for those explosive camp fires in the wilds of the south west of Western Australia ;-)

      But, hey all good fun, we have limited time, so enjoy, while yah can :-)

    2. “This seems so obvious it is probably wrong.”

      It is wrong, but only because of a slight misunderstanding.

      The weight of the air molecules doesn’t change with temperature. Their density does. In hot air the molecules are spaced further apart, therefore it weighs less per volume, therefore it has buoyancy compared to the surrounding cooler air and the balloon rises.

      The ideal gas law:
      PV = nRT
      that is:
      Pressure x Volume = (amount of gas) x (ideal gas constant) x Temperature

      When volume is constant (balloon doesn’t grow or shrink), pressure is constant (balloon is open from the bottom), and the gas constant is constant (duh), then any increase in temperature must be met with a decrease in the amount (moles) of gas in the balloon. In effect, the hot air molecules in the balloon move faster, collide with other air molecules more frequently and push the excess gas out from the balloon, making it lighter.

      1. What’s actually happening with the vortex is somewhat similiar to what’s happening in a cylcone dust separator. The air is made to spin around, and in order to move from the outer diameter to the inner diameter, the air molecule needs to lose momentum – in other words become colder – so the air molecules that end up in the middle are those which are the coldest.

        The cold molecules end up in the middle because random collisions between molecules in the spinning vortex cause some molecules to speed up in the direction of spin and take a wider arc, and others to slow down and lose momentum and take a shorter path around, which causes the molecules to separate according to their kinetic energy. That also why the coldest temperature occurs in the middle of the cold exhaust tube – only the very coldest molecules end up in the center where they barely spin at all.

        The end stop valve is for providing some back pressure, which forces some of the hot air back in the center and pushes the cold air backwards through the other end.

        1. Thanks for the details. My technical background is rather thin. I used to fly a lightplane many years ago and takeoffs in hot weather used to require much longer runways because of the thinner density of summer air and that’s where my thinking began in this.

  6. Excellent video! I enjoyed the phone calls, which reminded me of some of the things people waste my time asking.

    As I read through the comments here, I’m glad I wasn’t the only person who simply said “What the…?!” at the end of some of the posts. There is some absolute nonsense in here, though credit where credit is due, they are beautiful gems of nonsense putting many other pieces of nonsense to shame. :D I can appreciate that.

  7. I made one in shop class in about 1965. The Jr. High school library had the Scientific American book of experiments. It is all about the relation of temperature and velocity for gas molecules. The lowest velocity (or lowest energy) molecules collect in the center and exit the small hole due to a little back-pressure adjusted by the hot end valve. It was pretty neat and made a LOT of noise, so I didn’t get to play with it much at the time. In the 1970’s I as working at a place that made lumber mill equipment and to my surprise, electronic enclosures were cool by a vortex tube with the cold side pipe entering the cabinet. Mills have lots of compressed air available.

    I don’t get the experimentation here. The data is out there from the people who have been making these commercially for 60 years plus plenty of more academic papers on efficiency and various improvements. Seems like that is a good place to start.

    1. The basic complaint seems to be that there seems to be no intermediate level info available – just some random tinker plans to build one, or very hard to read papers that discuss some advanced optimization – no basic info about the construction, theory and sizing of the parts.

  8. A similar vortex technology, dubbed Helikon, was used by South Africa for uranium enrichment as part of its nuclear weapons programme:

    http://www.globalsecurity.org/wmd/intro/u-aerodynamic.htm

    Apparently the working fluid was a hydrogen+UF6 gas mixture, the gas was input to the spiral at about the speed of sound, and enrichment took dozens of stages working in series to achieve a meaningful enrichment level… but it worked on the same general principle as tube in the video.

  9. Hmmm, 100psi air is 6x pressure of ambient. That gets pretty cold expanding. Use the work of expansion to scrub some of the gas against the housing until it’s warm. Sounds easy enough to imagine a functional equivalent without all the mumbo-jumbo.

  10. The Wirbelrohr! I remember reading about this in physicist Mark Silverman’s book “And Yet it Moves.” There’s a whole chapter devoted to it. Apparently a proposed mechanism is the Ranque-Hilsch effect, which involves the presence of a particular sound frequency. As Silverman puts it:

    “Buried within that roar, however, is a pure tone, a ‘vortex whistle’ as it has been called, that emerges from the selective amplification of background noise…. It is this pure tone or whistle, whose frequency increases with the velocity of swirling–and hence with the pressure of the compressed air–that is purportedly responsible…” He goes on to explain acoustic streaming, and that if the pure tone is *acoustically* cancelled out, the temperature difference drops dramatically.

    The chapter is called “The Whirbelrohr’s Roar (… or rather Whistle)”

Leave a Reply to DaxCancel 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.