Few things exhibit as much as disdain for basic physics
as the space levels in Nintendo's "Star Fox,"
except the barrel roll.
PEPPY (VOICEOVER): Do a barrel roll!
That part might be legit, and today I'm
going to demonstrate how it could work.
Space battles in games and movies
are laughably unrealistic.
Ships that bob and weave, laser cannons galore,
and giant explosions all make for good drama but bad physics.
Joe Hanson, over at "It's Okay to be Smart"
has a great episode on how pop culture gets
space battles all wrong.
I'll leave that to him, though, because I
want to focus on one thing that the gaming
world may have gotten right-- the "Star Fox" barrel roll.
In general, "Star Fox" flushes physics down the toilet
during the space levels.
But the infamous barrel roll, the maneuver
that seems most crazy and that Peppy
is shouting at us to execute every five seconds--
PEPPY (VOICEOVER): Do a barrel roll!
--might be the most physically viable part of the game.
Now, let me be clear, we're only talking about the space levels.
When The Arwing is in a planetary atmosphere,
a barrel roll is child's play, it's aerobatics 101.
Just like a regular airplane, you
flip the ailerons, which are these panels on the wings,
in opposite directions-- up on one wing, down on the other.
That creates a differential in the atmospheric force applied
to the top and bottom of each wing.
The net force will push up on one wing, down on the other,
torquing the plane like a doorknob
around its longitudinal axis.
To stop after 360, you quickly reverse the ailerons,
thus flipping the direction of the forces on the wings
and torquing the ship back to a state of zero rotation.
So in an atmosphere, the maneuver
is all about the ailerons.
In fact, MatPat, over at Game Theorists,
correctly pointed out that the technical name of this movie
is an aileron roll.
In a true barrel roll, the whole aircraft
moves along a helical path, corkscrewing rather than
just rotating around its access while moving straight.
But the meme-tactic named, barrel roll, has stuck,
so I'll just keep using it here.
Back to space.
Ailerons don't do anything for you
in space, because there's no air pushing on the wings.
So how do you torque the ship to get it spinning
and then stop it after a full 360?
Because in case it isn't obvious,
once you get the ship spinning, it won't stop on its own.
There's basically no friction in space.
So any rigid spinning object, like a planet,
will continue to spin at exactly the same rate,
along the same axis, until something applies a torque
to change that spin.
That tendency to keep rotating is called an objects' angular
momentum, and physicists represent it
by a vector or an arrow.
Roughly speaking, that narrow points
along the axis of rotation, in the same direction
that the thumb of your right hand
would point if you curl your fingers in the sense
of the object's spin.
So say you want to do a barrel roll that's clockwise,
as viewed from the rear.
To do that, the shipments needs to acquire
a forward-pointing angular momentum vector.
To stop, that angular momentum must be removed.
Now, on a planet, the clockwise and subsequent
counter-clockwise torques required to do this
are provided by the atmosphere.
But in space, there's no air to push on the wings.
So I'll repeat, how do you torque to ship?
One option is to put small thrusters on the top and bottom
of each wing, one quick burst to start to ship rotating and then
another opposing burst to stop it.
That could certainly work, but it
means carrying extra fuel for the thrusters
to get those big torques over and over.
That would weigh down the ship, so it seems wasteful.
Also, I don't see any thursters like that
in the Arwing blueprint.
But what if you could store up the angular momentum in advance
and carry it with you?
And I don't mean having the ship doing a nonstop barrel roll,
I mean happy angular momentum on demand without thrusters.
Could you do it?
Yep, using gyroscopes-- or to be more technically
accurate, flywheels, which I'll sometimes
refer to as gyros, because it's easier and 'cause I'm hungry.
Now, the trick is a lot easier to show you
than to tell you about.
So here we go.
Here's the basic idea.
By rotating the flywheel quickly,
I end up transferring its angular momentum to myself,
so I start spinning.
Pretend the axis of my torso is a longitudinal axis
of the Arwing, with the ship's nose toward my head
and the engine and wings down by my crossed legs.
Imagine the ship has been pre-loaded
with an already-spinning flywheel,
like the spinning bike wheel you see on the screen.
That flywheel is spinning clockwise,
as seen from the rear of the ship,
or counterclockwise as seen from the nose.
So its angular momentum vector points
toward the nose of the ship, which
here is the top of your screen.
You can also think of this vector as the net angular
momentum of the entire ship flywheel system, all of which
currently happens to reside just in the wheel,
with zero angular momentum in the fuselage.
When I flip the flywheel 180 degrees,
its individual angular momentum starts
pointing toward the rear of the ship, i.e.
toward the bottom of the screen.
But the net angular momentum of the entire ship wheel system
must remain unchanged.
That can happen only if the entire fuselage acquires
its own large, forward-pointing angular momentum.
That, combined with the wheel's angular momentum,
nets out to the same total angular momentum
vector we began with.
It's just vector arithmetic.
Flipping the flywheel thus causes the ship to barrel roll.
To stop the roll, you just flip the flywheel back
to its original orientation.
Here's a view from the nose of the ship.
You're flying along and all of a sudden--
PEPPY (FROM VIDEO): Do a barrel roll!
(REMIXED) --barrel roll--
All right, Peppy.
PEPPY (SINGING): Do a barrel-- barrel roll!
Do a barrel-- barrel roll!
Do a barrel roll!
So that, more or less, is a barrel roll in space.
Sorry, an aileron roll, minus the aileron.
Space telescopes, like Hubble and Kepler,
also use angular momentum conservation
to turn without thrusters.
The difference is that their flywheels
point in a fixed direction and gets spun opposite
to the way they want the telescope to turn,
kind of like the teacup ride at Disney World.
But similar to our "Star Fox" setup,
the ISS does rotate pre-spun flywheels
to control its orientation.
They're called control moment gyros.
If you want to feel this physics in action,
just get a bike wheel with some pegs,
clamps to add mass, a spinning chair, and a couple of friends.
You can do what I did.
All of this has just been a proof of concept demo.
I don't have any numbers on the mass and rotational inertia
of a "Star Fox" Arwing.
So I can't tell you how massive or dense the gyros
would need to be, whether you'd flip them
mechanics or with magnets, or how fast they'd
have to spin in the first place to store up enough angular
momentum for the barrel rolls.
So from an engineering perspective,
there are still lots of unanswered questions, including
what a space meter is, which is apparently
the official unit of measurement in the "Star Fox" universe.
But assuming you can spin a massive enough flywheel fast
enough, without it breaking apart, and assuming
the friction in the wheel is really low,
then unlike with thrusters, you can keep storing and reusing
that angular momentum to do as many barrel rolls as you like.
Just knock yourself out.
In fact, automotive engineers are
trying to make flywheels the braking mechanism in cars
to avoid losing energy to friction every time you saw.
Now, before we sign off, there's one little caveat
that I swept under the rug.
Our flywheel flipped fast but not instantaneously.
In the process, both of flipping and of flipping back,
its angular momentum vector sweeps
through a bunch of partially-sideways directions.
To compensate, the ship would have
to do some wonky rotating as the flywheel flips,
in order to keep the total angular momentum
That means during each barrel roll,
the ship's nose will tilt. Some combo of left or right, or up
or down, depending on exactly how we flip the gyro.
But there is a way to stop.
Can you come up with a simple design to keep the ship on axis
during the barrel rolls without thrusters?
Don't put your answers in the comment.
Instead, email them to firstname.lastname@example.org.
Your answers need to be clear.
If I can't figure out what you're saying, no dice.
We'll shout out people who submit correct answers
on the next episode of "Space Time."
Last week, we talked about what might destroy planet Earth,
and you guys have a lot to say.
A lot of you ended up raising many
of the same alternate Earth destruction scenarios.
So this week, rather than shout people by name,
I'm going to address the scenarios you all
Before I do that, let me make a disclaimer about how I chose
scenarios for the episode.
I wanted them to be based on understand physics that
had some non-microscopic probability of occurring.
Or if I allowed in any speculative physics,
like the big rip, there had to be
at least some experimental observational
evidence behind it.
So bear in mind that many of the scenarios you raised,
while technically not impossible,
fail to meet one or both of those criteria.
Among the more speculative scenarios that were raised,
Grey Goo-- the possibility that advanced
nanobots might disassembled the planet atom by atom--
and protons decay-- that maybe protons are unstable,
and in quadrillions of years, individual nuclei will
start falling apart.
I guess technically not impossible,
but they also have no experimental basis,
so now you know why I didn't include them.
Another possibility raise is that maybe there
exists an exotic but more stable state of nuclear matter
involving strange quarks.
And that, if some of that matter appeared on Earth,
it would trigger a chain reaction
wherein every nucleus on Earth would revert into that state.
Physicists have considered this possibility,
and if it were likely, it probably
would have occurred in particle accelerators already.
So we're probably cool.
Lots of you asked whether Earth could suffer a collision when
the Andromeda and Milky Way galaxies merge
in about 4 billion years.
The probability is near zero, way lower
than one in a billion.
There's so much empty space between star systems
and galaxies that the probability
of any two of them colliding would
be similar to the probability of two tennis balls
colliding if I filled all of space with tennis balls
but had a gap of more than three miles between any two of them.
Likewise, people ask whether rogue planets
or rogue black holes, streaking through the Milky Way,
could come in and collide with Earth.
Again, extremely low odds of collision, even if there's
hundreds of millions of those things flying around the Milky
Actually, I will address two comments individually.
Culwin brought up that Ceres is actually the largest asteroid.
This is actually my mistake.
I thought that a dwarf planet and an asteroid designation
were mutually exclusive, they appear not to be.
But even if Ceres is an asteroid,
it's only about three and a half times as massive as 4 Vesta,
still not enough energy to destroy the Earth.
Finally, Paul Anzel asked a very thought-provoking question.
He wanted to know how much energy it would take,
not just to overcome the gravitational attraction
between all the matter on Earth, but to split apart
every interatomic bond as well.
I actually don't know, but this could be worked out.
You just need to know how much energy there
is in interatomic bonds, per atom on Earth,
and multiply that by an estimate of the number of atoms
I encourage you to work it out, and let me know what you get.
Great comments, as always, keep 'em up.
And if you'd like to show, please remember to subscribe.
PEPPY (SINGING, REMIXED) Barrel roll!
Do a barrel-- barrel roll!
Do a barrel roll!