# Compressed Air Levitation and the Coanda Effect

What do you want to levitate today? [Latheman666] uses his air compressor to make all kinds of stuff float in mid air. Light bulb, key chain, test tube, ball bearing, tomato… pretty neat trick to try in your shop.

It is interesting to see what physics explain this behavior. The objects do not float just because they are pushed upwards by the airflow, that would be an unstable equilibrium situation. Instead, they obtain lift in a very similar way as the wings of an airplane. Not all objects will levitate using this trick: the object has to be semi-spherical at the top.

[Applied Science] nicely shows this behavior by levitating a screwdriver first, then an identical object but with a flat top. The flat top screwdriver fails to levitate. The curvature provides the path for a smooth airflow, because of the Coanda effect, creating a zone of low pressure at the top, making the situation analogous to that of an airplane wing. Therefore, for this to work, you need an object with some kind of airfoil shaped surface. Another great demonstration is that of [NightHawkInLight], using a high speed camera.

A very impressive experiment that needs nothing more than an air compressor!, we are sure you will try it next time you work with one. For more on this topic of levitation with air streams, check the ping pong ball levitation machine.

## 23 thoughts on “Compressed Air Levitation and the Coanda Effect”

1. Ostracus says:

I imagine the other needed thing is relatively light weight.

1. krinje says:

You can levitate a full sized file/rasp, not “heavy”, but not especially light either. It’s a very common shop trick to do.

2. telek9 says:

Isn’t this kind of an extension of the high school science bit where you levitate ping pong balls using a hair dryer?

3. notarealemail says:

Haha, could someone please backhand math what it would take to levitate a bowling ball? :-D

1. RW ver 0.0.2 says:

A duct cleaning truck.

1. RW ver 0.0.2 says:

Though I wouldn’t bet against a commercial leaf blower and a LOT of spin.

1. notarealemail says:

I need to go try something… hold on a minute… going into the garage… :)

2. Thinkerer says:

You could do it experimentally – drop it out of an airplane, measure its terminal velocity then provide that in an airflow that’s sufficiently fast to bring it to equilibrium.

Though less fun, you can take data from the quora page listed below, which gives about 418 ft/s terminal velocity, and doing this in a one foot diameter tube would require 328ft³/s or 19,698 ft³/min (CFM) of airflow. As you increase the size of the tube, the airflow requirement increases as the square of the diameter.

Note that your computer fans are rated in the tens or hundreds of CFM. Ever been to one of those skydiving wind tunnels? Yeah, that.

https://www.quora.com/What-is-the-terminal-velocity-of-a-bowling-ball

1. Whatnot says:

feet.. seriously.

3. Annie says:

An insulation blower or 6″ air amplifier.

4. Far tastier to use a Malteser, lie on your back and blow. When you’re done, just open your mouth and reward your impressive levitating skills with some chocolate.

1. notarealemail says:

I’m a blue-collar American; Whoppers® for me! :P

5. My senior design project in undergrad was a drone build around the coanda effect. Generally pretty impracticable but still fun to experiment with.

1. JonRB says:

I would LOVE to see a write up of [the interesting bits of] this

6. The link on the Coanda effect in the article has my favorite explanation: surface tension pulling air downward as it passes over the curved surface. In order to take horizontally (or upwards) moving air and have it moving down afterwards, there has to be an equal and opposite force pressing up on the wing.

This makes calculating how much air you need to flow easier. Assume air passing over the head of the lightbulb is diverted 90 degrees. Calculate out how much mass of air you’d need to do that in order to lift the bulb. Done. No spooky “low pressure” zones needed. Just Newton’s Third.

1. Whatnot says:

You don’t lift it, you keep it suspended. There is a slight difference since this is about an equilibrium.

7. An aside, but the V8 F1 cars of a few years ago made use of the coanda effect to direct their exhaust gases towards the rear diffuser.
It was found in earlier seasons that good gains could be made by utilising the high speed exhaust gas to ‘seal’ the edges of the rear diffuser and create more downforce, an idea that was banned by specifying that the exhausts had to exit in a certain position and angle upwards.
Never to be deterred the ingenious engineers made use of a duct and the coanda effect to ‘bend’ the passage of the air back down towards the diffuser again.
A very tricky thing to balance though as the engine operated over a wide band and therefore had a very large variation in exhaust gas temperature, mass flow and velocity; but they made it work!
Meanwhile I’m sat in my garage giggling at a levitating screwdriver.

8. Cknopp says:

Formula one used this to shape and direct the exhaust gases to drive the rear diffusor a few years ago.

Watch “The #TechF1Show – F1’s ‘Coanda’ exhausts explained” on YouTube

1. RW says:

Heh, that’s one of those “I wonder if it was me”, was talking up the idea of exhaust driven coanda surfaces on auto forums about 3 years before they appeared in F1.

9. Law says:

The fascinating aspect is the amount of stability. Just thinking out loud, is this more efficient than a helicopter? Could the every-person (WKA every-man) flying car be developed from this?

1. Whatnot says:

You’ll need to hire a lot of Mexicans to run around with giant compressors underneath those flying cars. Or in Europe the Poles or Ukrainians.