Understanding Light Behavior in a Moving Train: Explained

[ewr]

Say a person is on a train moving to the right. He then flashes a light upwards perpendicular to the motion of the train. Directly on top of the person there is a mirror. If the light travels in a straight line why does it hit the mirror and bounce back? Why does the light act in a similar way to, say a tennis ball being thrown upward in a moving train? Why doesn't the light travel to the back of the train hitting just to the left of the mirror?

If the train is moving to the right and the light is flashed upwards, why do the light also move in a horizontal direction along with the motion of the train?

I understand why a tennis ball seems to travel up and down when thrown upwards from the perspective of the person in the moving train. But I don't understand why this is the case for light.

Thank you for the help.

 

[Ibix]

Same reason as the ball, in short. Conservation of momentum.

 

[Drakkith]

I understand why a tennis ball seems to travel up and down when thrown upwards from the perspective of the person in the moving train. But I don't understand why this is the case for light.

Why would it be any different?

 

[PeroK]

The light on the train doesn't know that it is "moving to the right". As far as the person on the train, the torch, the light and the mirror are concerned, they are at rest and it's the person on the platform that is moving to the left.

Also, the train is on the Earth and the Earth is moving round the sun, so souldn't the light to go off in some entirely different direction depending on how fast the train is moving round the sun?

Someone yesterday asked a very similar question:

https://www.physicsforums.com/threads/how-old-is-light.884668/

 

[jartsa]

Let us consider a light bulb that is moving to the right, and a parabolic reflector moving with the light bulb, and a ray of light that is not moving with the reflector.

Now because the ray is not moving with the reflector, the reflector collides with the ray, and after the collision the ray is moving with the reflector.

And that is a reason why moving parabolic mirrors produce light beams that are tilted to the direction of the motion of the mirror.(Imagine a upwards pointing parabolic mirror moving to the right, and a ray that points upwards in the frame where the mirror moves to the right)

 

[PeroK]

Let us consider a light bulb that is moving to the right, and a parabolic reflector moving with the light bulb, and a ray of light that is not moving with the reflector.

Now because the ray is not moving with the reflector, the reflector collides with the ray, and after the collision the ray is moving with the reflector.

And that is a reason why moving parabolic mirrors produce light beams that are tilted to the direction of the motion of the mirror.(Imagine a upwards pointing parabolic mirror moving to the right, and a ray that points upwards in the frame where the mirror moves to the right)

That, if anything, only confuses things.

 

[Dale]

Let us consider a light bulb that is moving to the right, and a parabolic reflector moving with the light bulb, and a ray of light that is not moving with the reflector.

Now because the ray is not moving with the reflector, the reflector collides with the ray, and after the collision the ray is moving with the reflector.

That is a good thing to point out. You can also get a similar analysis by considering the beam produced by a spherically symmetric source and a pinhole.

 

[jartsa]

That, if anything, only confuses things.

Maybe I just happen like mechanical explanations.

Let's say an electron is bouncing up and down between two walls, or between a floor and a ceiling. Said electron radiates EM-waves equally to the left and to the right.

in a frame where the electron is moving to the right, it radiates more to the right than to the left. Why is that?

Let's see ... the direction of the acceleration is the same in both frames, but the shape and the orientation of the electric and magnetic fields of the moving elecron are different in the two frames, so that's the reason, probably.

 

[harrylin]

Say a person is on a train moving to the right. He then flashes a light upwards perpendicular to the motion of the train. Directly on top of the person there is a mirror. If the light travels in a straight line why does it hit the mirror and bounce back? Why does the light act in a similar way to, say a tennis ball being thrown upward in a moving train? Why doesn't the light travel to the back of the train hitting just to the left of the mirror?

If the train is moving to the right and the light is flashed upwards, why do the light also move in a horizontal direction along with the motion of the train?

I understand why a tennis ball seems to travel up and down when thrown upwards from the perspective of the person in the moving train. But I don't understand why this is the case for light.

Thank you for the help.

Maybe it's useful to phrase what others already said in again other words. A tennis ball that is thrown straight upward relative to the train, is thrown at an angle relative to the tracks. And just the same, a light ray that is emitted straight upwards relative to the train, is emitted at an angle relative to the tracks.

 

[jbriggs444]

Maybe it's useful to phrase what others already said in again other words. A tennis ball that is thrown straight upward relative to the train, is thrown at an angle relative to the tracks. And just the same, a light ray that is emitted straight upwards relative to the train, is emitted at an angle relative to the tracks.

The words "light ray" conjure up an image of a line with an unambiguous direction and may not be the best to use. "light pulse" is more helpful in my opinion.

 

[Raymond Potvin]

Same reason as the ball, in short. Conservation of momentum.

If what you say was true, then we would see a star where it actually is, not where it was when it sent its light. Light is independent from the motion of a source, and if the train is moving the same way a star is moving, then it should not follow the motion of the train. Why would motion be different for stars than for trains?

Why would it be any different?

For the same reason.

 

[Dale]

The speed of light is independent of the source, but the direction is not. The rules for both trains and stars are the same, both involve the same relativistic aberration. If you think that they are different then you are misunderstanding something.

 

[jbriggs444]

If what you say was true, then we would see a star where it actually is, not where it was when it sent its light. Light is independent from the motion of a source, and if the train is moving the same way a star is moving, then it should not follow the motion of the train. Why would motion be different for stars than for trains?

Funny that you mention this. There is a second reason that we do not see stars where they actually are. https://en.wikipedia.org/wiki/Aberration_of_light
If you consider reception of light and transmission of light as inverse phenomena then this is exactly the same reason that a light is not always seen to shine in the direction it is pointed.

 

[Raymond Potvin]

The speed of light is independent of the source, but the direction is not. The rules for both trains and stars are the same, both involve the same relativistic aberration. If you think that they are different then you are misunderstanding something.

Motion is about two parameters: speed and direction. Considering that one is independent and the other is not seems contradictory. A ball depends from the motion of the source on both parameters, its speed and its direction, so why would light be independent only for speed? When a photon is sent from a star in the direction of the earth, the star is no more there when the photon strikes the Earth because it has moved away from the direction the photon has towards the earth, so why would it be different for a train?

 

[Ibix]

When a photon is sent from a star in the direction of the earth, the star is no more there when the photon strikes the Earth because it has moved away from the direction the photon has towards the earth, so why would it be different for a train?

It isn't - that's what we keep telling you. The only special thing about light is that the magnitude of its velocity is unaffected by a frame change, while both are affected for things traveling slower than light.

 

[Raymond Potvin]

It isn't - that's what we keep telling you. The only special thing about light is that the magnitude of its velocity is unaffected by a frame change, while both are affected for things traveling slower than light.

Why only velocity would be unaffected? Is there any experiment that shows that the direction of light is affected by the direction of its source?

 

[Ibix]

Why only velocity would be unaffected? Is there any experiment that shows that the direction of light is affected by the direction of its source?

Speed is unaffected, to be precise. Why? It's a consequence of the geometry of spacetime. Or else it can be seen as a fundamental fact from which relativity can be deduced. We're pretty much running into the limits of our understanding here.

Direction of light depending on the relative speed of source and detector: https://en.m.wikipedia.org/wiki/Aberration_of_light

Edit: Also, see the FAQ on the experimental basis for special relativity, a sticky thread at the top of this forum.

 

[PeroK]

Why only velocity would be unaffected? Is there any experiment that shows that the direction of light is affected by the direction of its source?

You are misunderstanding the whole concept. Imagine a light source "at rest" and two observers moving with different velocities relative to the source. They cannot possibly measure the same velocity for the light, unless all three are moving in the same direction.

If the light comes out of the source in, say, the y direction, and one observers is moving along the x-axis, and the other is at rest at the origin, then they will measure different x components of the light's velocity.

This would be true for the motion of any object.

What is different about light is that it's speed is the same to both observers. That would not be true of a ball or a train.

In short, the speed of a light beam can be the same for all observers, but the velocity of the light cannot be the same for all observrs.

 

[Raymond Potvin]

Direction of light depending on the relative speed of source and detector: https://en.m.wikipedia.org/wiki/Aberration_of_light

Here is the animation from that wiki page:

On the right part of the drawing, the Earth is considered at rest while the star is moving, and the direction of the photon is not affected by its motion, but if a ball had been sent instead of a photon, it would not have reached the Earth since its direction would have been affected by the motion of the star.

 

[PeroK]

Here is the animation from that wiki page:

On the right part of the drawing, the Earth is considered at rest while the star is moving, and the direction of the photon is not affected by its motion, but if a ball had been sent instead of a photon, it would not have reached the Earth since its direction would have been affected by the motion of the star.

Those animations would apply exactly if, say, the star was a gun, the Earth the target and the light ray a projectile. It's pure geometry. If the projectile hits the Earth in one frame, it must hit in the other.

For example, in the animation on the left, why would a projectile inevitably miss the Earth?

 

[Ibix]

Here is the animation from that wiki page:

On the right part of the drawing, the Earth is considered at rest while the star is moving, and the direction of the photon is not affected by its motion, but if a ball had been sent instead of a photon, it would not have reached the Earth since its direction would have been affected by the motion of the star.

The photon's direction is affected by the motion of the star, too. It's moving vertically in the left image and diagonally in the right image. The same would be true of a ball. And if the light pulse/ball/whatever hits the Earth in the left picture it must also hit in the right picture. Otherwise you aren't describing a self-consistent scenario.

 

[Raymond Potvin]

Those animations would apply exactly if, say, the star was a gun, the Earth the target and the light ray a projectile. It's pure geometry. If the projectile hits the Earth in one frame, it must hit in the other.

For example, in the animation on the left, why would a projectile inevitably miss the Earth?

It would hit the Earth on the left, but miss it on the right, so it would not be symmetrical as it is the case for light.

 

[Ibix]

It would hit the Earth on the left, but miss it on the right, so it would not be symmetrical as it is the case for light.

No. If it hits in one case it hits in both. Otherwise you are describing a circumstance where a ball hits a target according to one observer but not according to another moving with respect to the first. This is self-contradictory. The ball must either hit or not - it can't do both.

I think you need to step back and examine frame changes in simple Newtonian mechanics, because this seems to be where you have the problem.

 

[Raymond Potvin]

The photon's direction is affected by the motion of the star, too. It's moving vertically in the left image and diagonally in the right image. The same would be true of a ball. And if the light pulse/ball/whatever hits the Earth in the left picture it must also hit in the right picture. Otherwise you aren't describing a self-consistent scenario.

At the right, a ball sent in the direction of the Earth would not hit the earth, it would have to be sent at an angle to the direction of the Earth to account for the motion of the star if it had to hit the earth.

 

[PeroK]

At the right, a ball sent in the direction of the Earth would not hit the earth, it would have to be sent at an angle to the direction of the Earth to account for the motion of the star if it had to hit the earth.

You are misunderstanding those animations fundamentally. They are the same scenario from two different viewpoints. What happens in one must happen in the other. They are not two different scenarios.

You may also be thinking that in the one on the right, the star sees the light move at an angle. But, if you look carefully, you'll see that the star is always directly above the light, which is consistent with the left animation. In the right one, the star has emitted the light vertically downwards in its frame.

And, in the left one, although the light moves vertically from the star's point of view, from the Earths' point of view, the light comes at an angle. And that it what is shown on the right.

 

[Raymond Potvin]

In the right one, the star has emitted the light vertically downwards in its frame.

And, in the left one, although the light moves vertically from the star's point of view, from the Earths' point of view, the light comes at an angle. And that it what is shown on the right.

On the left, I agree that aberration will change the direction of light seen from the Earth since it is moving, but I think that you take for granted that, at the right, the photons are traveling sideways to their direction to account for the motion of the star. It is not what the animation shows, on the contrary, it shows that the photon is aiming straight at the earth. Photons cannot travel sideways like balls if they cannot add the motion of their source to their own motion.

 

[PeroK]

On the left, I agree that aberration will change the direction the light is seen from the Earth since it is moving, but I think that you take for granted that, at the right, the photons are traveling sideways to their direction to account for the motion of the star. It is not what the animation shows. It shows that the photon is aiming straight at the earth. Photons cannot travel sideways like balls if they cannot add the motion of their source to their own motion.

No. On the right, look at the relative motion of the star and the photon. That photon was aimed vertically from the star. In the star's frame, whichever animation you chose, the photon is aimed in the same direction: Directly down, not towards the Earth.

In any case, both animations show the same photon on the same path.

If a photon were aimed at 45 degrees in the star's frame it would miss the Earth in both animations.

Somehow you need to take a step back and see that that these two animations show precisely the same scenario.

When you say photons can't move sideways you're missing the point that there is no absolute sideways. The sideways motion is entirely due to the relative motion of the star and the Earth. The photon moves in the direction it was emitted. The same physical direction in both animations.

The star has no absolute motion, only motion relative to the Earth. The photons move side ways or not depending on whether the observer has sideways motion relative to the star.

 

[DrGreg]

This animation from an old post of mine might help. (Click on the image to enlarge.)

Maybe this diagram helps

Here's something (it could be a pulse of light, it could be a ball) bouncing up and down in a train.

An observer in the train (top) infers the thing is moving vertically up and down.

An observer on the ground (bottom) infers the thing is "sliding sideways" in a zig-zag path.

This is valid in both relativistic and Newtonian mechanics.

 

[Dale]

Considering that one is independent and the other is not seems contradictory.

It is not contradictory. Both facts follow from the Lorentz transform. It may be surprising to you, but that doesn't imply any inherent inconsistency.

When a photon is sent from a star in the direction of the earth, the star is no more there when the photon strikes the Earth because it has moved away from the direction the photon has towards the earth, so why would it be different for a train?

It isn't different, I am not sure why you think it is. In all cases the result is determined by the Lorentz transform.

 

[Dale]

It would hit the Earth on the left, but miss it on the right, so it would not be symmetrical as it is the case for light.

Please show your math that leads you to believe this. This is such an absurdly wrong claim that I cannot believe that you have ever even attempted to work it out.

Photons cannot travel sideways like balls if they cannot add the motion of their source to their own motion.

Show your math. This is just the Lorentz transform.

 

[PeroK]

At the right, a ball sent in the direction of the Earth would not hit the earth, it would have to be sent at an angle to the direction of the Earth to account for the motion of the star if it had to hit the earth.

Another way to look at this is to imagine a pipe leading vertically down from the star. A photon that is fired down the pipe stays in the pipe.

In the left animation, the photon strikes the Earth as the Earth moves to the end of the pipe as the photon emerges.

And, in the right animation the photon also stays in the pipe - it must as its the same scenario. In this animation, the photon hits the Earth because the source and pipe are moving towards the Earth and the end of pipe reaches the Earth just as the photon emerges.

In both cases, the source, pipe and photon all have the same motion in the horizontal direction. On the left, in the source's frame, this is zero motion. On the right, in the Earth's frame this is the relative speed of the star and the Earth.

The same is true whatever the source fires down the pipe. If it stays in the pipe and hits the Earth in one frame, it must stay in the pipe and hit the Earth in the other frame.

 

[PeroK]

An idea to show that light and particles must behave in the same way.

Imagine a source emits a particle (ball) at near light speed and shortly afterwards a photon in the same direction. Imagine this in the rest frame of the source.

The two emissions move in the same direction and after a time, the photon catches the particle and collides with it.

This motion and collision must be observed in all other reference frames. If another reference frame moves perpendicular to the motion of the particle and photon, then it must measure the two moving in the same straight line and colliding.

Any "sideways" motion of the particle caused by the relative motion of the reference frame must be exhibited by the photon as well.

If the photon refused to move sideways in the second reference frame and went in a different direction, it would miss the particle and give an inconsistent observation.

Therefore, photons and particles must behave in the same way in this respect.

 

[Raymond Potvin]

Please show your math that leads you to believe this. This is such an absurdly wrong claim that I cannot believe that you have ever even attempted to work it out.

It is not a mathematical problem, it's a factual one. If we aim a photon from a moving vehicle and at 90 degree to its direction, it should follow a 90 degree path whatever the speed of the vehicle if the direction of light is independent from the motion of the source, but if we throw a ball in the same direction from the same vehicle, it will not follow a 90 degree path because a ball suffers the motion of its source.

 

[A.T.]

if the direction of light is independent from the motion of the source,

But it isn't.

 

[Ibix]

if the direction of light is independent from the motion of the source

You keep noticing that this is contradictory. I have said, @PeroK has said, @harrylin has said, @Dale has said, @jbriggs444 has said, and now @A.T. has said, that the direction of light is not independent of the motion of the source. When are you going to stop worrying about the physical consequences of something that everyone agrees doesn't happen...?