Mini tornado experiment

Since experimenting with the cyclone for my small dust collector, I have been thinking more about how cyclones actually work. What promoted this experiment was Jim Ryan's square separator, which achieves a surprising amount of separation even without trying to be a cyclone.

It occurred to me that if centrifugal separation activity can occur in a square enclosure, it could also occur without any enclosure. And that prompted me to try to produce a small sort of "tornado" in my workshop. Maybe "tornado" is overstating it a bit, but I have from time to time seen little eddies or twisters that pick up leaves and twigs from the ground in otherwise relatively calm weather conditions. It's one of those that I set out to create.

I knew I needed some suction from the top. I first tried it with just a box fan blowing air up, but without success. I then made a cardboard shroud to concentrate the suction on a smaller area.

I suspended the box fan from the ceiling, with the bottom of the shroud about 50 cm off the floor. A second, small fan is used to give the air a light rotation, and a blue camping mat helps to deflect that air to make for a better initial swirling motion.

Some wood shavings spread on the floor give the tornado something to pick up, which makes it visible.

A real tornado is visible because there is enough of a pressure drop in the funnel to cause a temperature drop, and that in turn causes water to condense out of the air. But my workshop tornado was not going to be nearly as powerful.

It took some experimentation to reliably get a mini tornado out of my setup. With my initial experiments, I tried too hard to get a strong circulating motion on the ground by producing a sort of wind shear with two fans blowing air past each other. But I think the strong winds from these fans blew away any eddies before they formed. Relatively calm air with only a light twist to it was more conducive to forming a twister.

It's possible that I could have formed a twister without the cardboard shroud on the box fan, but I didn't realize I had too much wind from my other fans until after I put a cardboard shroud around the box fan.

Given some shavings to pull up from the ground, the twister becomes visible. But I wasn't successful at getting a long visible funnel. Any debris I tried sucking up with this twister, including flour, got ejected from the funnel before rising very far. The same is true for a real tornado - otherwise, the debris field from a tornado would be much larger than it is. It's condensation that makes the funnel of a real tornado visible.

What's most interesting is that the air inside the tornado is moving much faster than the circulating breeze or the upward suction that I started with. The high airspeeds get produced by spinning air getting pulled in towards the center. Like a pirouetting figure skater pulling her arms in, the air also accelerates as it is pulled in.

Conservation of angular momentum dictates that, for angular momentum to be conserved, the air must speed up as it gets to the middle. But this is counterintuitive --- what pushes the air to go faster, and where does the kinetic energy come from?

The energy comes from the upwards suction. With centrifugal force from the spinning air pushing outwards, pressure inside the funnel is lower than in the surrounding air. Lower pressure is lower potential energy, and in the process of getting to the middle, potential energy (through centrifugal force) is converted to kinetic energy, resulting in higher wind speeds.

Or you could just use the Bernoulli principle to explain that the air at lower pressure moves at a higher speed.

This speeding up of air towards the middle really helps a cyclone work. Centrifugal force on an object moving in a circular motion is given by this formula:

F = m V²

r
Where:
F = force (Newtons)
m = mass (kg)

V = velocity (m/s)
r = radius

To achieve maximum force on a particle of mass m, we want to maximize speed (V) and minimize radius. And at the middle of the cyclone, r becomes smaller, while air speed (V) increases, achieving maximum force F to pull heavier particles away from the middle where air is sucked out of the cyclone.

And that is what makes a dust separating cyclone work so well.

After I put this article online, Mike Meneghetti sent me a cool picture of his tornado machine, so I just had to include it here.

Mike writes:

I made this one a few years ago. It uses a plenum at the base. Air is sent up through the vertical pipes, each of which has holes drilled 6 inches on center. This provides the sheer wind necessary for a vortex to form. A small fan on the top creates a low pressure area causing the air to rise and causing the sheer wind to begin spinning. It is illuminated by dry ice in warm water in the base. The cool thing about this design is that spectators can reach in and temporarily disrupt the vortex. It almost immediately resumes.

The design is not completely original. I saw the idea on a YouTube instructional video put out by the Tornado Research Institute. I saw some ways of improving on their concept and put them in place.

I have been perusing your web site off and on for a few years now. Keep up the great stuff.
Mike