Vortex tubes, how do they work?

[Topher925]

Vortex tubes have always puzzled me. They are very popular in industrial applications, their performance can be modeled to an extent, but no one knows for certain how they work.
Wikipedia states "Pressurized gas is injected tangentially" but I have found this is not always the case which leads me to believe that most of the theories that involve angular momentum are incorrect.
http://en.wikipedia.org/wiki/Vortex_tube#Efficiency

They seem to have no limit of operation either, as they can even be used for cryogenic applications. Does anyone have an insight as to how these thing work? I wish I had the time and money to build one that has a translucent structure so I could track the flow of the gas inside.

 

[Andy Resnick]

Jearl Walker's "Flying Circus of Physics" has some information, but it's definitely true the process is complex.

The claim is that a vortex is created in the tube; the air moves faster near the tube axis and slower near the tube wall (this is different than normal axial flow, this appears to be some sort of swirling flow). Viscous interaction between the fast-moving and slow-moving air transfers heat from the center to the periphery. Additional cooling of the core is allowed by passing the central, cooler, air through an aperture and allowing expansion to occur, thus lowering the temperture an additional amount.

The references provided are:

Sibulkin, M. J. Fluid Mech 12, 269 (1962)
Hilsch, R. Rev. Sci. Instrum. 18, 108 (1947)
Fulton, C. D., Refrigerating Eng. 59, 984 (1951)
Hartnett, J. P. and Eckert, E. R. G., Trans. Am. Soc. Mech. Eng. 79, 751 (1957)
Marynovskii, V.S. and Alekseev, V. P., Sov. Phys. Tech. Papers 1, 2233 (1956)
Pengelly, C. D., J. Appl. Phys. 28, 86 (1957)
Van Deemter, J. J., Appl. Sci. Res. A3, 174 (1952)
Cooney, D. O., Ind. Eng. Chem. Fund. 10, 308 (1971)
Scheper, G. W., Refrig. Eng. 59, 985 (1951)

I'm sure someone has taken an image of a vortex tube in operation using smoke or some sort of tracer. If not, that would be an excellent submission to the "Gallery of Fluid Physics".

 

[RonL]

Vortex tubes have always puzzled me. They are very popular in industrial applications, their performance can be modeled to an extent, but no one knows for certain how they work.
Wikipedia states "Pressurized gas is injected tangentially" but I have found this is not always the case which leads me to believe that most of the theories that involve angular momentum are incorrect.
http://en.wikipedia.org/wiki/Vortex_tube#Efficiency

They seem to have no limit of operation either, as they can even be used for cryogenic applications. Does anyone have an insight as to how these thing work? I wish I had the time and money to build one that has a translucent structure so I could track the flow of the gas inside.

Check this out, and see if it helps.

http://www.airtxinternational.com/how_vortex_tubes_work.php

My mind sees a design that can produce a center point, void of anything. Might be a good place for something to happen?

 

[Topher925]

Viscous interaction between the fast-moving and slow-moving air transfers heat from the center to the periphery.

This doesn't make any sense to me? How exactly will friction of the fluid be able to create a temperature difference of 60C let alone REMOVE heat from the center (slower moving) fluid.

Thanks for the link RonL, I think I've seen that site before and I understand the relationships between vortex tube geometry and the aperture. But it doesn't really go into the workings of vortex tubes.

 

[RonL]

This doesn't make any sense to me? How exactly will friction of the fluid be able to create a temperature difference of 60C let alone REMOVE heat from the center (slower moving) fluid.

Thanks for the link RonL, I think I've seen that site before and I understand the relationships between vortex tube geometry and the aperture. But it doesn't really go into the workings of vortex tubes.

Not sure if this will change anything, but think of the perfect center of rotation as zero pressure, and zero temperature, a point of no energy and if a single atom tried to stay in that spot, a second atom would have more energy and tend to bump it out of place, and the two would continually struggle around that center point.
No matter what scale you use, each outward ring of rotation will contain more atoms moving at a faster speed and having more energy based on the average of the mass flow. As you approach the inner wall of the confining tube the frictional value of the flow transfers both mechanical and heat energy to the wall.
Depending on how much restriction is placed against the high speed hot gas discharge, it causes the cold lower pressure central core of air to move out the cold end.
The average of the quantity of bands of rotation that are forced out either end will determine the difference of temperature between the two discharge quantities.

Thats how I see it

Ron

 

[Andy Resnick]

This doesn't make any sense to me? How exactly will friction of the fluid be able to create a temperature difference of 60C let alone REMOVE heat from the center (slower moving) fluid.

Thanks for the link RonL, I think I've seen that site before and I understand the relationships between vortex tube geometry and the aperture. But it doesn't really go into the workings of vortex tubes.

That's a good question. First, the faster-moving air is near the center of the tube, and so it transfers heat to the slower moving air at the periphery, not the other way around. The presumed mechanism of this transfer of heat is friction- it must be some dissipative mechanism. Second, there's a difference between heat and temperature. Small fluxes of heat are able to create large changes in temperature if the specific heat of the material is small.

Have you looked at any of the references?

 

[RonL]

That's a good question. First, the faster-moving air is near the center of the tube, and so it transfers heat to the slower moving air at the periphery, not the other way around. The presumed mechanism of this transfer of heat is friction- it must be some dissipative mechanism. Second, there's a difference between heat and temperature. Small fluxes of heat are able to create large changes in temperature if the specific heat of the material is small.

Have you looked at any of the references?

Well I'm not going to say anything else, but sure hope someone puts an explanation out, air is injected at the outer periphery of the spin chamber, and speed does increase as it is forced into the smaller barrel of the tube, but how can it be spinning faster at the center of rotation? it seems that some portion of the center has to be a dead zone.

 

[rcgldr]

Angular velocity might be higher at the center, but I doubt that speed is, due to viscosity. It would seem that linear speed at the center should be slower.

 

[Phrak]

This doesn't make any sense to me? How exactly will friction of the fluid be able to create a temperature difference of 60C let alone REMOVE heat from the center (slower moving) fluid.

Thanks for the link RonL, I think I've seen that site before and I understand the relationships between vortex tube geometry and the aperture. But it doesn't really go into the workings of vortex tubes.

Perhaps you should start with the basics: PV=nRT.

Other usefull relationships for gases are adiabatic expansion and isothermal expansion for diatomic molecules. (Air consists mainly of diatomic molecules N2 and O2.)

 

[Topher925]

Perhaps you should start with the basics: PV=nRT.

Other usefull relationships for gases are adiabatic expansion and isothermal expansion for diatomic molecules. (Air consists mainly of diatomic molecules N2 and O2.)

I've been down the ideal gas road already and hypothesize that the temperature differences experience in the vortex tube are not great enough to cause such large temperature changes. Does anyone know what happens when you plug up the hot side of a vortex tube?

Andy Resnick, I spent a little bit of time looking for some of those papers online but couldn't find anything. I'm assuming that this is because they are so old. Would you happen to know anyway of getting them easily without me having to go through the trouble of a inter-library loan?

 

[Phrak]

I've been down the ideal gas road already and hypothesize that the temperature differences experience in the vortex tube are not great enough to cause such large temperature changes. Does anyone know what happens when you plug up the hot side of a vortex tube?

Andy Resnick, I spent a little bit of time looking for some of those papers online but couldn't find anything. I'm assuming that this is because they are so old. Would you happen to know anyway of getting them easily without me having to go through the trouble of a inter-library loan?

It's certainly an amazing device in it's simplicity, isn't it?

Im not sure what you mean by a temperature difference in the vortex tube. I think the idea is that the gasses in the section of rotational flow are at nominally the same temperature. There is enough turbulence in the vortex that mixing between layers occurs and heat can be exchanged by convection. There can also be some radiant heat exchange.

There should certainly be a tendency to label the inside vortex 'cold' and the outside 'hot'. But these are to distinguish them from one another. If these are accurate labels, I would be greatly suprised. They should be discounted until proven.

There is a large contact area between layers built into the length of the device. You might ask, why is this length of tube there at all? Without it, one could still extract gasses from the inner vortex and send it one way, and the outer vortex the other way.

Anyway, there are some factoids to consider.

1. The outer layer has greater kinetic energy than the inner layer.
2. It is also under greater pressure than the inner layer.
3. The viscous drag of the walls in the vortex section will ensure that the rotation (whether it be in the laminar or turbulent flow regime) is not a 'ridgid rotation', so that the angular velocitites of the outer layers are somewhat less than the inner layers--but for consideration of point 4).

4. As the gas reaches the hot end of the tube, a portion is directed inward and back toward the cold end on the inside of the vortex, as you know. This gas still has it's original velocity from rotation. So now, at the hot end, it's angular velocity is quit large. Now we know there is a gradient from one end of the tube to the other in the angular velocity profile of the inner vortex.

5. Turbulent flow will result in greater mixing between layers than laminar flow. The velocity profile is also flatter than laminar flow would be. All these are superior attributes of turbulent flow over laminar, I believe.

6. Turbulent flow should have a larger heat exchange with the tubing walls. This is a smaller effect. It's superior to laminar flow too, if you want your cold stream colder and care not so much about the hot side.

 

[Austin0]

I just had a crazy thought: What if the operative principle was simply centrifical force? Starting out with a wide range of molecular velocities, the force acted to sort them out on the basis of, not rest mass ,but momentum
Creating a general movement of the faster molecules to the outside.
It seems like the process happens very fast ,maybe too fast to allow for complicated internal heat transfers between molecules. Although I have no idea what the temp differential would be in an ambient gas as far as the fastest to slowest molecules if you did divide them on that basis.

 

[Andy Resnick]

<snip>

Andy Resnick, I spent a little bit of time looking for some of those papers online but couldn't find anything. I'm assuming that this is because they are so old. Would you happen to know anyway of getting them easily without me having to go through the trouble of a inter-library loan?

Topher,

Unfortunately, I don't. And full disclosure- I haven't looked at them, either :) If I get a chance, I'll head over to our library and see if they have hard copies. I don't have a problem, if I find a good one, making you a copy and snail-mailing it over.

 

[Phrak]

As the vortex spins in the tube, centripital acelleration compresses the gass in the outer vortex. As the outer vortex meets the slug centered in the end of the tube, a portion of the vortex flow is directed inward and back along the tube. As an element of gas moves from the outer the the inner vortex it undergoes expansion cooling. It's angular velocity is still high. Viscous drag with the outer vortex slows the inner vortex as it proceedes back along the tube.

 

[stewartcs]

Vortex tubes have always puzzled me. They are very popular in industrial applications, their performance can be modeled to an extent, but no one knows for certain how they work.
Wikipedia states "Pressurized gas is injected tangentially" but I have found this is not always the case which leads me to believe that most of the theories that involve angular momentum are incorrect.
http://en.wikipedia.org/wiki/Vortex_tube#Efficiency

They seem to have no limit of operation either, as they can even be used for cryogenic applications. Does anyone have an insight as to how these thing work? I wish I had the time and money to build one that has a translucent structure so I could track the flow of the gas inside.

All of the vortex cooling systems I've seen have the air entering tangentially so I've always believed the angular momentum theory. Essentially, in the vortex tube, the speed of the inner vortex remains the same. Angular momentum has been lost from the inner vortex. The energy that is lost shows up as heat in the outer vortex. Thus the outer vortex becomes warm, and the inner vortex is cooled.

CS

 

[CaptainQuasar]

Sorry to inject a tangent but since vortex tubes operate on compressed air, I have a crackpot theory I want to ask about that it reminded me of. Ever since seeing those experimental French automobiles that run on compressed air and learning that one of the difficulties with a hydrogen economy is compressing the hydrogen for transportation and storage, I have wondered if there's any way it would be feasible to compress gasses for these purposes with some sort of device like a diving bell: take a ballasted pair of tanks down to the depth in the ocean where the pressure is at the level you require, let water flow into the larger tank until all of the gas is in the smaller tank, close the valve, haul it all back up to the surface, and voilá, you've got a compressed tank of gas.

It seems like it could be ballasted to a point arbitrarily below buoyancy to reduce the amount of energy necessary to submerge and surface it. But that would seem to make it an unlimited source of energy in the form of compressed air so there must be some physics or engineering detail I haven't got right... what am I missing?

Also, could vortex tubes be used for air conditioning in the compressed-air-powered cars or would it require too much of the energy stored in the tanks?

⚛​

 

[Phrak]

But that would seem to make it an unlimited source of energy... ...what am I missing?

You're way off topic, but as the bell decends the air is compressed and looses boyancy.

Also, could vortex tubes be used for air conditioning in the compressed-air-powered cars or would it require too much of the energy stored in the tanks?

⚛​

[/QUOTE]
I understand the efficiency is about 40%. Throwing away the hot half of the flow would reduce the efficiency even further.

 

[Phrak]

But that would seem to make it an unlimited source of energy... ...what am I missing?

You're way off topic, but as the bell decends the air is compressed and looses boyancy.

Also, could vortex tubes be used for air conditioning in the compressed-air-powered cars or would it require too much of the energy stored in the tanks?

⚛​

I understand the efficiency is about 40%. Throwing away the hot half of the flow would reduce the efficiency even further.

 

[CaptainQuasar]

You're way off topic, but as the bell decends the air is compressed and looses boyancy.

Oh, duh. Buoyancy is dependent on volume. Sorry to be off topic.

⚛​

 

[Phrak]

Oh, duh. Buoyancy is dependent on volume. Sorry to be off topic.

No need to appologize to me. You just might bring down the guys with the big sticks. Better to open another thread. There are dozens of people lying in wait, ready to pounce on any misconception.