The Engineering of the Drinking Bird

This toy has fascinated me since childhood. To me its motion is almost hypnotic. Here’s how it operates. Wet the bird’s beak thoroughly with room
temperature water — the opaque container makes it looked chilled, but it isn’t … then
stand it up right . . . It’ll take a few seconds for it to start drinking . . . Notice
that all of the action right now takes place in the stem here. As I speed up the action you see liquid rising
and the bird rocking back and forth. If I return to normal speed, you can see the
bird slowly … very, very slowly …. Rock forward … Until it takes a drink, which
it will do again and again. In this video I’ll detail the bird’s clever
engineering design, explain how it uses thermodynamics, and link its action to some of the greatest
and most impactful devices created by engineers. This toy has long history, but its current
incarnation is due to Miles V. Sullivan, a scientist at Bell Labs. He specialized in methods of manufacturing
semiconductors, but as a sideline invented toys. Its reported that this bird delighted U.S.
President Herbert Hoover, an engineer, who failed to figure out how it worked, and it
also defeated the great scientist Albert Einstein, who spent three and half months studying it. Its reported that he refused to take the bird
apart. With the benefit of hindsight, let’s start
by exploring how it works and examining the key engineering design aspects. First, let’s ask is the water ornamental
or essential? At first the bird acts just as if the water
were still there. Now let’s speed up the bird’s motion you
see at 15 minutes it is still drinking. At 30 still drinking. 45 minutes still drinking. 60 minutes still drinking. 75 minutes still drinking. And five or ten minutes later, at eighty or
eight-five minutes it takes its last drink. The liquid still rises a bit, but it never
rises enough to make the bird tip over, which shows that the motion is not perpetual — as
long as there is water, the bird keeps drinking. Let’s look inside the bird to get an idea
of how it works. Underneath the bird’s hat, beak and fabric
covering lies a glass bulb, smaller than the bulb at the base, and also rounder. Now, watch as I put a few drops of isopropyl
alcohol on the top bulb to cool it. The liquid rapidly rises to the head, this
changes the bird’s center of gravity so that it will tilt forward. The head now fills with liquid and then … there … … it … drinks. It becomes upright and the liquid drains from
the head … liquid rises again to the head and … the bird drinks again. This cycle repeats until all of the isopropyl alcohol on the bird’s heat evaporates. Why does the liquid rise? The place to begin is with the bird’s manufacture. The bird is filled through this “tap”
— a small pipe built into the head — with methylene chloride dyed red, which is then
frozen, a vacuum applied to evacuate the air, the tap sealed (and, of course, later hidden
by the bird’s hat) . . . And then the methylene chloride melts: It turns to liquid and then
some of it evaporates (turns into vapor). The key to the bird’s operation is that
the vapor in the head and in the base are separated by the liquid in the base. It’s hard to see, but the tube extends into
the base, nearly reaching the bottom. This separates the vapor in base and the vapor
in the tube and …. of course, the head. So, at rest the pressure in these two spaces are
equal, but when the bird’s beak is wet, the temperature falls and, as I’ll explain
in a moment, the pressure in the head drops below that in the base and the liquid rises. Of course this liquid in the head causes the
bird to . . . tilt forward, to drink … and when it drinks, the vapor in the head and the base are connected, the pressures nearly equalize — a slug of vapor rises to the
top and some liquid drains from the head and and then the cycle repeats. To see the pressure equalize I’ll slow down
the bird as I tilt it forward. Right now the head is half full. When I tilt it you see a slug of vapor go
from bottom to top. I’ve titled it far enough forward that the
liquid in the head is below the top of the tube and the liquid in the base is below the
section of the tube that almost reaches the bottom of the bird. This allows the pressure to equalize, and
as the bird becomes upright the liquid returns to the base before the cycle starts again. In operation it doesn’t tilt quite this far
forward and so the pressures don’t fully equalize. Why, though, does the pressure in the head
drop as the temperature falls? You can see the answer if I shoot cool, compressed
gas across the bird’s head. As the cool gas strikes, you see liquid condensing
inside the head; and , as you see on the left, this causes the liquid in the base to rise. The cool gas withdraws energy as heat from
the head causing some of the methylene chloride vapor inside to condense – to turn into a
liquid — this decreases dramatically the amount of vapor in the head — liquid is
1,000 times more dense than vapor — this in turn lowers the pressure in the head and
causes the liquid to rise. I used compressed gas to cool the head because
I can control the amount of cooling; the bird, though, cools its head by “drinking.” The head is wrapped in fabric that absorbs
water. As I put drops on its beak you can see the
water beads up at first . . . then saturates the fabric and spreads rapidly across the
bird’s face. On the right side you can see it creeping
to back of the head. If I now turn the bird around … you can
see that the water has spread to the back. As I continue adding drops on the beak the
saturated area on the back increases. When this water evaporates into the air it
removes energy from the bulb as heat — you feel this effect every time you step out of the shower, the evaporating water withdraws energy as heat and chills you. This evaporation, this withdrawal of heat,
lowers the temperature and begins the condensation of the vapor, which starts the cycle as I
showed you with the cool, compressed gas. As long as the head is wet and heat is withdrawn
from it, the bird will always “drink,” but if you were to operate the bird in humid
air it would slow down because little water would evaporate, and if the air were at 100%
humidity the bird would stop because no water would evaporate at all. Now, to make this dramatic condensation happen
when the temperature is lowered just slightly — the evaporating water lowers the temperature
by only about three-tenths of a degree — the bird’s designer choose a highly volatile liquid. This means one whose boiling point is near
ambient temperature because for small changes in temperature there is a large change from
vapor to liquid and so the variation of pressure is large. Watch what happens as I “heat” the base
of the bird with my hand. You see the liquid level in the base dropping:
that’s because energy from my hand is converting some of the liquid into vapor, which increases
the pressure in this region . . . and that causes the liquid to rise to the head. Eventually I heat the vapor so much that it shoots up the stem. Now watch as I place
my hand around the head. Heat from my hand converts liquid to vapor,
which increases the pressure and forces the liquid back to the base. To test this explanation of the bird’s operation,
let’s activate the bird in different ways. As I noted it is the temperature difference
between its top and bottom that drives liquid to rise to the head. So, let’s see what happens if I point a
light at the base of the bird, which I’ve painted black so it will absorb the energy from the light better. As I heat the base of the bird, the liquid
rises, as before but … the bird tips backwards. The wet nose tilted the center of gravity
… and so I added some modelling clay to the nose to get the bird to tilt forward. And now when I turn on the light the liquid
rises, the birds drinks just as if there were liquid in front of it until . . . I turn the light off … and the bird drinks for a little bit longer until eventually . . . it comes to rest. Next, let’s see what happens if we use this:
Whiskey. Again, thoroughly wet the bird’s beak with
the liquid . . . stand it upright … and then we see again
the liquid rising in the bird . . . and then … it drinks. We can also now understand why the bird’s
rate of drinking differs among the three methods I used to “activate” the bird: a heat
lamp, whiskey and water. Roughly, the heat bird takes three drinks
for every one of the water bird, the whiskey bird takes two for every drink of the water
bird. The reason the bird drinks whiskey faster
than water is because the rate of evaporation of the alcohol is greater than that of water. This means that heat is withdrawn faster from
the head and so more vapor condenses in a shorter amount of time, which accelerates
the pressure difference. The heat lamp causes the greatest difference
of all, which highlights how an engineer thinks about this bird. To an engineer this bird is a heat engine. A heat engine turns heat differences into
work — mechanical motion. To see that recall that when the bird is just
about to drink that its head is at a lower temperature than its base, which is at ambient
temperature. Then when it “drinks” the pressure in
the head and base start to equalize so liquid returns to the base, but the overall temperature
of the bird is now just a little below ambient temperature. When it return to upright the base draws in
energy as heat . . . the head then rejects some energy as heat and the bird drinks again. These two flows define a heat engine: a device,
operating in a cycle that absorbs heat from a high temperature reservoir, converts part
of it into work, and rejects the remainder into a low temperature reservoir. The fact that this is a heat engine means
it’s related to the great machines that make our globalized world happen: among those
the mighty steam turbine that generates electricity, the giant diesel engine that propels container
ships across the oceans, and the great gas turbine that flies us around the globe. I’m Bill Hammack, the engineer guy.


Add a Comment

Your email address will not be published. Required fields are marked *