Hero’s Engine

Here is a nice simple project that draws good reviews from kids. It is a simple heat engine the design of which is attributed to the Greek engineer Hero of Alexandria, who described its construction about 2000 years ago. It illustrates a number of principles, most centrally the use of steam to convert heat into rotary motion, as we still use to power most generators.

The vessel is a 500 ml Kimax boiling flask that is heat-resistant and extra thick to resist breakage (I’m not a fan of Chinese Bomex). It is about a fourth full of water. The flask is suspended using a heavy tie wrap, safety wire, and a beaded chain used for lights and ceiling fans. The chain has a clasp that can be used to disconnect it and seems to permit less rotational friction. I added a touch of oil to be thorough. Other materials could be used such as a hose clamp, etc., and some use a screwtop metal can for the vessel.

The glass jets are made from lab tubing bent over the alcohol burner you see with its blue flame in the picture at bottom. The burner uses ethanol (thinner) from the hardware store and can be had for about $10 from a science supply such as Cynmar. This one is unusual in that it has a double loop for more heat, so it cost maybe $15. Be careful, the flame can be nearly invisible at times; I had one student nearly touch it. There’s a plumber’s self-igniting MAPP blowtorch in the background that I use gingerly to speed things up in the beginning. Propane would be fine.

To make the nozzles I heated a tube while rotating it, then pulled it apart slowly, collapsing the tube. I cut the tubing by scoring with a file and snapping it, then burnished and “annealed” the nozzles which have roughly 1-2 mm orifices on an arm extending about 6 cm from the center of the flask. Annealing the tubing—heating it close to melting and then gradually cooling it by dabbing it in the flame—makes for much stronger glass by allowing internal stresses to sort themselves out. (Of course I don’t have a real annealing oven, which heats the glass for hours.)

Insert the tubes in opposing directions in a 2-hole lab stopper, start the fire and off it goes once the water and flask are hot. Steam, which is invisible and often superheated water vapor and not the mist of water droplets you actually see (I was wrong on this! the steam plus droplets are called a mixing cloud), is about 1000 times the volume of the liquid it comes from. Be careful not to shatter the glass with too much pressure (I would expect the stopper to come out first, but the glass could have been previously damaged), by heating it with no water on the other side, or by cooling hot glass with cold water. Especially if using a torch, be sure there’s water. The burner here is moved off-center because centrifugal force pushed all the water there.

Caution: Don’t burn yourself. Hot glass doesn’t look hot but causes nasty burns. Scalding hurts and scars! Glass breakage under pressure and shrapnel are serious risks. Wear safety glasses and an oven mitt of some sort (be careful, some of these burn and let water through—I use a good one from science supply, but it too does not prevent scalding) so you can grab the flask or move the flame aside. Be careful, the alcohol burner may flare up if moved suddenly!

The engine will spin pretty fast if you take the time to balance it. A wobbly one could hit the stand and break. I took a bunch of flash pictures to identify which side was spinning out so I could improve the balance. A short neck flask might do better, but this is the one I had. I used an Extech $25 laser tachometer intended for non-contact RPM measurements on cars and clocked around 200 RPM before the spin became unstable.

There’s a science fair project or several in here somewhere, such as the ideal design for the nozzles. There are some interesting questions such as whether it’s better to use a high velocity steam at high pressure, or to use a longer arm on the nozzles, larger orifices, and so on. If you build one of these, please share!

Advertisement