Why does light not reach any distance instantly?

[chillpenguin]

After learning about special relativity, and the behavior of light, I do not understand why light is not able to reach any distance instantly. Doesn't time stop at the speed of light? So if no time is passing, yet light is moving, wouldn't that mean that light is traveling infinitely fast? Obviously light doesn't travel that fast, but I can't see what stops it from going that fast if it is able to move without the passing of time.

 

[PeterDonis]

Doesn't time stop at the speed of light?

No. A better way of describing what SR says about lightlike objects (things that move at the speed of light) is that the concept of "time" (more precisely, "proper time") does not apply to them. That is, the question "how much time passes for a light ray?" is not well-defined.

 

[chillpenguin]

Oh I see... The more you learn about physics, the more you realize you don't understand the half of it!

 

[ghwellsjr]

After learning about special relativity, and the behavior of light, I do not understand why light is not able to reach any distance instantly.

How would you know if it did or it didn't? Have you ever thought about how you would determine how long it took for light to get somewhere? You can't watch it like you can watch a baseball being thrown, or even with the aid of high-speed photography, you can't capture the motion of light, can you? For other things, we illuminate them with light so we can tell where they are, but what are you going to do with light? Have you considered that as light travels from the sun to the moon, you can't see it? Even the light that is reflected off the moon coming back to the Earth you can't see while it is in transit. All you can see is when it finally gets to your eyes. And that's the problem. We cannot determine how fast light is traveling but we can measure how long it takes to make a round trip from your laser to a remote target and back to you. In fact, that's how a laser rangefinder works that you can buy at a hardware store; it measures how long it takes for the light to make a round trip between you and a target and then using the defined value for the speed of light, it can calculate the distance. So what we do in Special Relativity, is we say that the light took the same amount of time to get to the target as it did to get back. That's Einstein's second postulate.

Doesn't time stop at the speed of light?

No, this is a common misconception. Einstein said that time is what a clock measures and since a clock has to be made out of some material substances and since material substances cannot travel at the speed of light, we cannot talk about time slowing down for light like we can for material objects.

So if no time is passing, yet light is moving, wouldn't that mean that light is traveling infinitely fast? Obviously light doesn't travel that fast, but I can't see what stops it from going that fast if it is able to move without the passing of time.

Well, since your assumptions are wrong, then there is no quandary for you, is there?

 

[berkeman]

How would you know if it did or it didn't? Have you ever thought about how you would determine how long it took for light to get somewhere? You can't watch it like you can watch a baseball being thrown, or even with the aid of high-speed photography, you can't capture the motion of light, can you?

Of course we can. The velocity of a foot per nanosecond is not hard to measure in the lab...

 

[WannabeNewton]

Of course we can. The velocity of a foot per nanosecond is not hard to measure in the lab...

There's a difference between measuring the one-way speed of light and measuring the two-way speed of light. The latter only requires one clock so there's no issue but the former requires at least two synchronized distant clocks; in other words you need a notion of distant simultaneity. But how do you go about synchronizing the two clocks in the first place? If you use light signals to synchronize the clocks, as is usually done, then you need to know something about the one-way speed of light so we're back to square one. This is why in Einstein's original paper he established by definition that the one-way speed of light is isotropic in inertial frames. Using this we can then synchronize distant clocks. So it's important to distinguish between the two-way speed of light (which can be measured using a single clock) and the one-way speed of light (which requires a synchronization convention such as Einstein's).

 

[ghwellsjr]

How would you know if it did or it didn't? Have you ever thought about how you would determine how long it took for light to get somewhere? You can't watch it like you can watch a baseball being thrown, or even with the aid of high-speed photography, you can't capture the motion of light, can you?

Of course we can. The velocity of a foot per nanosecond is not hard to measure in the lab...

I don't know how, please tell me.

 

[berkeman]

I don't know how, please tell me.

You use techniques similar to optical TDRs: http://en.wikipedia.org/wiki/Optical_time-domain_reflectometer

 

[ghwellsjr]

I don't know how, please tell me.

You use techniques similar to optical TDRs: http://en.wikipedia.org/wiki/Optical_time-domain_reflectometer

I read the article. I saw where the OTDR can measure fiber length, similar to what I said about a laser rangefinder, but I didn't see any claim or mention that it can measure how long it takes for a signal to get from one end of the fiber optic cable to the other.

So I still don't know how. Can you please point me to the place in the article where I misunderstood or overlooked or whatever your understanding is that I don't have.

 

[berkeman]

"How long" is just the horizontal axis on the oscilloscope.

 

[ghwellsjr]

"How long" is just the horizontal axis on the oscilloscope.

The article says the horizontal axis on the oscilloscope is length, not time:

The strength of the return pulses is measured and integrated as a function of time, and is plotted as a function of fiber length.

 

[berkeman]

The article says the horizontal axis on the oscilloscope is length, not time:

That is for the cable tester. Length is time -- about a foot per nanosecond in free space, less in a dielectric like a cable. It is definitely a "time of flight" measurement, which is what you were asking about. It can be done with multiple photodetectors along the path of flight of a short laser pulse in air as well.

 

[ghwellsjr]

The article says the horizontal axis on the oscilloscope is length, not time:

That is for the cable tester. Length is time -- about a foot per nanosecond in free space, less in a dielectric like a cable. It is definitely a "time of flight" measurement, which is what you were asking about.

Are you saying that if the clock in the instrument measures 10 nanoseconds from the time the pulse was emitted to the time when it was detected, then the length is about 10 feet?

It can be done with multiple photodetectors along the path of flight of a short laser pulse in air as well.

Can we deal with this more complicated case later?

 

[AlephZero]

How would you know if it did or it didn't? Have you ever thought about how you would determine how long it took for light to get somewhere?

You can do it with a "simple" mechanical device, to within a few percent:
http://en.wikipedia.org/wiki/Fizeau–Foucault_apparatus

even with the aid of high-speed photography, you can't capture the motion of light, can you?

Yes you can. http://web.media.mit.edu/~raskar/trillionfps/

 

[ghwellsjr]

How would you know if it did or it didn't? Have you ever thought about how you would determine how long it took for light to get somewhere?

You can do it with a "simple" mechanical device, to within a few percent:
http://en.wikipedia.org/wiki/Fizeau–Foucault_apparatus

That's measuring how long it takes for light to get there and back. Do you have a way to measure how long it takes for the light to get there?

even with the aid of high-speed photography, you can't capture the motion of light, can you?

Yes you can. http://web.media.mit.edu/~raskar/trillionfps/

You are again seeing light that has gone from a source to different places on the objects and back to the camera. This does not measure how long it takes for the light to get just from the source to the objects, does it?

 

[berkeman]

Are you saying that if the clock in the instrument measures 10 nanoseconds from the time the pulse was emitted to the time when it was detected, then the length is about 10 feet?

Yes. Are you really not understanding what Aleph and I are saying? The only practical issue for electrical detection is to use a fast laser pulse as the source, and have high bandwidth detectors that can see nanosecond light pulses. The rest is pretty trivial.

 

[ghwellsjr]

Are you saying that if the clock in the instrument measures 10 nanoseconds from the time the pulse was emitted to the time when it was detected, then the length is about 10 feet?

Yes.

Are you sure? I would have thought that if the OTDR measured 10 nsecs, it would have reported 5 feet for the length of the fiber. Please help me understand your answer.

Are you really not understanding what Aleph and I are saying?

I really don't understand how what you are saying is related to measuring how long it takes for light to reach a distance, which is the subject of this thread.

The only practical issue for electrical detection is to use a fast laser pulse as the source, and have high bandwidth detectors that can see nanosecond light pulses. The rest is pretty trivial.

Can we please finish with your first answer regarding the OTDR before going on to other issues?

 

[m4r35n357]

you can't capture the motion of light, can you?

Not to diminish your point in the slightest, but there was an example of just this on UK TV recently (using a state of the art high speed video camera), I think Brian Cox introduced it in one of his pop science programmes but I haven't managed to find it online yet. If anyone can provide a link I'd be interested in seeing it again ;)

EDIT: here's the link
http://www.ted.com/talks/ramesh_raskar_a_camera_that_takes_one_trillion_frames_per_second.html

 

[DrGreg]

To ghwellsjr and berkeman

It seems that neither of you is understanding what the other is saying.

I think ghwellsjr's point is that you can't directly measure the 1-way speed of light; all you can do is measure the 2-way speed. The 1-way speed is set by convention rather than by experiment.

I think berkeman is using the convention that the 1-way and 2-way speeds are equal to each other so there's no need to make any distinction between them. I think all the methods proposed are 2-way measurements.

Have I summarised each of your postions correctly, and does that help you understand each other?

 

[berkeman]

Are you sure? I would have thought that if the OTDR measured 10 nsecs, it would have reported 5 feet for the length of the fiber. Please help me understand your answer.

Yes, 10 feet in free space, and more like 5 feet in fiber.

I really don't understand how what you are saying is related to measuring how long it takes for light to reach a distance, which is the subject of this thread.

If you space out high-speed photodetectors (with beam splitters) and use equal length coax cables to get the detected signals to a high-speed oscilloscope, you can see the optical pulse as it hits each photodetector, with the pulses spaced out in time. That's the way I read the OP's question.

Can we please finish with your first answer regarding the OTDR before going on to other issues?

 

[ghwellsjr]

Not to diminish your point in the slightest, but there was an example of just this on UK TV recently (using a state of the art high speed video camera), I think Brian Cox introduced it in one of his pop science programmes but I haven't managed to find it online yet. If anyone can provide a link I'd be interested in seeing it again ;)

EDIT: here's the link
http://www.ted.com/talks/ramesh_raskar_a_camera_that_takes_one_trillion_frames_per_second.html

This was already asked in post #14 and answered in post #15.

 

[stevmg]

How about this? To the traveler near c, time is going infinitely slowly. To the stationary observer, time proceeds normally. i.e., one can traverse the "entire" universe consuming virtually no time on board the spacecraft while to the observer at one "fixed" point, time goes normally.

P.S., there is NO central or fixed point in the universe.

 

[ghwellsjr]

How about this? To the traveler near c, time is going infinitely slowly.

No, not infinitely slowly. It's finitely slowly according to the gamma function.

To the stationary observer, time proceeds normally.

And to the traveler near c, time proceeds normally. And it's the "stationary" observer whose time is going slowly. More precisely, we should be talking about the rest frame of the "stationary" observer and the rest frame of the traveler because neither one can observe the Time Dilation of the other one.

i.e., one can traverse the "entire" universe consuming virtually no time on board the spacecraft while to the observer at one "fixed" point, time goes normally.

Not "no time" but as small a time as you want. Again, pay attention to the frames.

P.S., there is NO central or fixed point in the universe.

True.

However, this discussion is off topic. This thread is about how long it takes for light to traverse a distance, not a spacecraft .

 

[stevmg]

Correction! One doesn't experience "time dilation" and time doesn't go infinitely slowly.

But, you get what I mean!

 

[stevmg]

All I know about the gamma function is it's relation to factorials.

 

[ghwellsjr]

Are you sure? I would have thought that if the OTDR measured 10 nsecs, it would have reported 5 feet for the length of the fiber. Please help me understand your answer.

Yes, 10 feet in free space, and more like 5 feet in fiber.

Now this is what I don't understand. We're talking about the OTDR that you said could measure how long it takes light to travel a certain distance. The OTDR has only one detector located at the emitter. It can only measure the round trip time it takes for light to go down a fiber optic cable and return. Even if we were talking about free space, the distance would be 5 feet, not 10 feet, correct? For a fiber optic cable it would be something less than 5 feet but in either case, the OTDR does not report anything according to time, it reports information according to length along the fiber optic cable, doesn't it? That's what the article you pointed to says.

I really don't understand how what you are saying is related to measuring how long it takes for light to reach a distance, which is the subject of this thread.

If you space out high-speed photodetectors (with beam splitters) and use equal length coax cables to get the detected signals to a high-speed oscilloscope, you can see the optical pulse as it hits each photodetector, with the pulses spaced out in time. That's the way I read the OP's question.

Can we please finish with your first answer regarding the OTDR before going on to other issues?

I guess we are just about finished with the OTDR.

We'll go on to other schemes. Let's assume that instead of coax cables with dielectrics which add the complication of speeds less than that of light, let's use waveguides that don't have that issue, OK? And let's stipulate that the measured value of the speed of light is one foot per nanosecond, OK?

Since you haven't provided any specifics, I'm going to propose a scenario that I think is in agreement with your proposal. Let me know if it isn't.

We start with a photo source that we can switch on at will. We pass the light through a beam splitter that includes a fast photodetector connected by a five-foot straight waveguide to one input of our high-speed oscilloscope. The other beam continues on along a straight path parallel to the waveguide and on to another high-speed photodetector, ten feet away from the beam splitter. This photodetector is connected to a second straight waveguide, also five feet long going back to the other input of the oscilloscope.

Now we switch on the light and we measure a difference in the arrival times of the two pulses of 10 nanoseconds. But have we measured how long it took for the light to get from one photodetector to the other?

I think not. As the light propagates from the beam splitter and the electrical signal propagates in parallel, they travel in tandem for a distance of 5 feet. Then the waveguide signal goes into the oscilloscope while the light continues parallel to the other waveguide for the remaining 5-foot distance to the second photodetector and then the electrical signal travels back on that second waveguide to the oscilloscope. So all we have measured is the roundtrip time for the light to traverse 5 feet in one direction and the electrical signal to traverse 5 feet back the same distance in the opposite direction and we're calling it the one-way time for the light to traverse the 10 foot distance.

I think it is easy to note that we could have made that first straight waveguide any length we wanted and moved the beam splitter an appropriate distance and we would still get a measurement of 10 nanoseconds. At this stage, I think we can see a similarity to the OTDR example if we just eliminate that first waveguide and put the beam splitter right at the input to the oscilloscope with the first photodetector connected directly into the oscilloscope, don't you concur?

All we have done by this scheme is say that the time it takes for the signal to propagate down the waveguide in one direction is identical to the time for the signal to propagate up in the other direction but we haven't measured it to make sure that the statement is true. In fact there is no way to know the answer to this question. That's why Einstein said that we are free to stipulate that the two times are equal, thus creating the concept of relative time.

Lest anyone think I'm nitpicking, let me remind you that in one reference frame, where we define light to propagate at c in all directions, we do not claim that the signal takes the same amount of time to propagate in both directions in a moving waveguide.

 

[ghwellsjr]

All I know about the gamma function is it's relation to factorials.

The gamma function that I was talking about is:

γ = 1/√(1-β²)

where β is the speed as a fraction of the speed of light.

 

[stevmg]

Start over again -

Emitted light travels from point A to point B (no matter how far) relative to its own frame instantly - zero time.

To observer in another frame, the time consumed is in its own frame relative to the frame that the light is in.

Nowhere in space is a physical frame moving "at the speed of light". Every where else, other than on the light beam, any physical frame is moving at some finite speed below c. Hence, from the aspect of any other frame in the in the universe there will be consumed time from its travels from point A to point B. What is that equation 1/[sqrt(1 - v^2/c^2)]?

Hell, I don't even understand the gamma function, only as it relates to factorials, and even then I don't understand it. Someone try to put the gamma function into something meaningful for me?

Oops! Sorry George. Didn't see your post. Your gamma function is the same as the Lorentz equation I mentioned above.

I am 71 years old, a retired physician (MD) and my expertise with the nuances of relativity and Lorentz is between slim and none, but I find the Lorentz equations fascinating.

 

[ghwellsjr]

Start over again -

Emitted light travels from point A to point B (no matter how far) relative to its own frame instantly - zero time.

That doesn't make any sense. In Special Relativity, the speed of light is defined to be c in any Inertial Reference Frame (IRF). That's Einstein's second postulate. When we use the expression "its own frame", we mean the IRF in which it is at rest. How can there be an IRF in which light can both be at rest and traveling at c? It's a contradiction of well-accepted terms.

To observer in another frame, the time consumed is in its own frame relative to the frame that the light is in.

In all IRF's, light is defined to travel in all directions at c. So if we know the distance, we know the time according to SR. It's real simple. Just definitions.

Nowhere in space is a physical frame moving "at the speed of light".

Let's be clear, frames are not physical, they're man-made constructs. But, yes, no IRF moves at c relative to any other IRF.

Every where else, other than on the light beam, any physical frame is moving at some finite speed below c. Hence, from the aspect of any other frame in the in the universe there will be consumed time from its travels from point A to point B. What is that equation 1/[sqrt(1 - v^2/c^2)]?

Hell, I don't even understand the gamma function, only as it relates to factorials, and even then I don't understand it. Someone try to put the gamma function into something meaningful for me?

Oops! Sorry George. Didn't see your post. Your gamma function is the same as the Lorentz equation I mentioned above.

I am 71 years old, a retired physician (MD) and my expertise with the nuances of relativity and Lorentz is between slim and none, but I find the Lorentz equations fascinating.

What you should do is devise a scenario using the coordinates of an IRF. Then use the Lorentz Transformation process (which includes gamma) to transform to the coordinates of another IRF moving at some speed (less than c) relative to your defining IRF. That's the quickest, easiest, and least confusing way to learn the important aspects of SR.

 

[berkeman]

Now this is what I don't understand. We're talking about the OTDR that you said could measure how long it takes light to travel a certain distance. The OTDR has only one detector located at the emitter. It can only measure the round trip time it takes for light to go down a fiber optic cable and return. Even if we were talking about free space, the distance would be 5 feet, not 10 feet, correct? For a fiber optic cable it would be something less than 5 feet but in either case, the OTDR does not report anything according to time, it reports information according to length along the fiber optic cable, doesn't it? That's what the article you pointed to says.

Sorry if I confused the issue by mentioning the OTDR. I was mainly trying to reference the kind of emitter and detectors that would be used to maesure the time of flight (and hence the distance) of light.