Is the brightness of light invariant?

[Quantum Quack]

Just a quick diagram....

<img src=http://www.ozziesnaps.com/doppler.gif>

Question :
How can both black ships record a doppler shift and yet still record the same speed of light as compared to if they were stationary to the star in the middle?
Given that time dilation and length contraction are not significant factors.

[ to stop the animation from driving you nuts hit the stop button on the browser ]

 

[dav57]

You guys aren't even trying to learn, are you? Are you really happy in your ignorance? Come on, put some effort in! Don't make me do all the work for you!

I will remain blissfully ignorant until you tell me why the wavelength of a beam of light changes inversely to the frequency as you approach it.

 

[DaleSpam]

Dalespam, thanks for that. But I want to make it very clear that QQ, Prosoothus and myself have a full understanding of the Doppler shift.

I don't think so. Several of you are working under the greatly mistaken assumption that the Doppler shift implies a changing c. In fact, Doppler of any wave (sound or light) requires a constant wave-propagation velocity. In other words, because the product of the wavelength and frequency is constant (constant wave velocity) when one goes up the other must go down. If that product were not constant then there would be no Doppler effect at all. E.g. the wavelength could decrease, the propagation velocity could decrease, and the frequency could remain unchanged. It is clearly an indication that some of you do not understand the Doppler effect when it is submitted as evidence of varying propagation velocity when it is in fact it requires a constant velocity in order to even occur.

It's why the frequency rises when you move faster towards an already emitted wave which is causing us the bother. And also why the wavelength of EM waves changes.

So your question is about an accelerating observer such that the relative velocity between detector and emitter changes between the time of detection and the time of emission?

That is why I included my comments last time about a brief pulse of light. Think about that situation for a while. The emitter emits a brief pulse of light. How can the detector's current velocity possibly important? The pulse of light has not arrived yet so the detector's current velocity cannot possibly be used in the Doppler effect. After some time the detector detects the pulse of light. How can the emitter's current velocity possibly be important? The pulse of light has already left the emitter so the emitter's current velocity cannot possibly be used in the Doppler effect. The two relevant velocities are the emitter at the time of emission and the detector at the time of detection. Any subsequent motion of the emitter and any previous motion of the detector are not relevant.

-Dale

PS I have not included any relativity comments here, strictly classical Doppler.

 

[DaleSpam]

So in doppler shift situation how does one measure the speed of light?

A well-designed "c-measurement" experiment will be insensitive to Doppler effects. In other words, the frequency and wavelength will be irrelevant to the outcome, only the propagation velocity will be important.

This is like asking how you would detect c using a Fourier transform. You can't it is sensitive to frequencies, not speed of propagation. Frequency and c are distinct (though related) features, you can easily examine them independently.

Think of it as length and weight of steel re-bar. Obviously as length increases so does weight, but obviously they can be measured independently. They are distinct (though related) features.

-Dale

 

[dav57]

I don't think so. Several of you are working under the greatly mistaken assumption that the Doppler shift implies a changing c. In fact, Doppler of any wave (sound or light) requires a constant wave-propagation velocity. In other words, because the product of the wavelength and frequency is constant (constant wave velocity) when one goes up the other must go down. If that product were not constant then there would be no Doppler effect at all.

Not true. When approaching a sound wave and increasing your speed, you experience a Doppler shift, yet your measurements show an increased frequency and an unchanged and consistent wavelength.

 

[dav57]

So your question is about an accelerating observer such that the relative velocity between detector and emitter changes between the time of detection and the time of emission?

No, I'm talking about rel vel between observer and beam of light - the vary thing which is claimed to be c and invariant!

 

[Physics Monkey]

Hi dav57,

Solve Maxwell's equations to find the radiation emitted from a source moving relative to you. The wavelength is, after all, simply a description of a certain property of the electromagnetic field. You will then see quite clearly why the wavelength changes. Those same equations further imply that while the frequency and wavelength may change depending on the motion of the source, their product is simply the speed of light. Perhaps a better question for you to ask yourself is why you insist it can't change.

Think of it like this. In common experience what we mean by time and distance is really determined by rigid rods and clocks. Clearly, the relationship between two such coordinate systems is a matter to be settled by experiment. In other words, we should just go out and look to find out how one observer's description of a phenomenon is related to another's description. It happens that for low velocities the Galilean description is the appropriate relationship, but why must this be true always? In fact, we have gone out and looked, and it turns out that different observers descriptions are related by the Lorentz transformation. Why would you insist that your intuition from the world of small velocities be valid? I continue to look forward to the day that we can all take rides on a really fast rocket, see first hand Einstein's vision, and finally lay this whole nonsense to rest.

 

[Physics Monkey]

Some more thoughts ...

Your definition of "velocity relative to a light beam" seems no good because the velocity so defined doesn't do anything that relative velocity ought to do. For example, suppose two objects are next to the same beam of light. They both have the same velocity relative to the light beam by your definition, but this doesn't tell you anything about their velocity relative to each other. Since they have the same velocity "with respect to something" you might think they should be at rest with respect to each other, but that clearly doesn't have to be true. There is a huge difference between saying that light moves at speed c relative to every inertial observer (which is perfectly reasonable and true as far as we know), and saying that everything moves at speed c relative to light (which is nonsense). Light doesn't have a reference frame, and so you can't move relative to it.

Also, why the strange fear of mathematics? If you are thinking logically about a problem, you're already using mathematics. The great visualizer of our age, Richard Feynman, once said that mathematics is about patterns. How is that not exactly what you want?

As for the suggestion that there are deep unanswered questions, I certainly agree. What I don't understand is why you think that abandoning all the experimental and theoretical understanding we've accumulated over the years is somehow the way to make progress.

 

[DaleSpam]

No, I'm talking about rel vel between observer and beam of light - the vary thing which is claimed to be c and invariant!

The "relative velocity between observer and beam of light" is always c. That is why it is irrelevant to the Doppler effect. The only velocity that is important is the velocity of the emitter at the time of emission relative to the velocity of the detector at the time of detection. Nothing else matters for the Doppler effect, neither according to SR nor classical physics.

I don't know why you are having such a difficult time with this concept, but it does not represent a failure of physics nor of the efforts and knowledge of people on this board.

-Dale

 

[DaleSpam]

Not true. When approaching a sound wave and increasing your speed, you experience a Doppler shift, yet your measurements show an increased frequency and an unchanged and consistent wavelength.

You are simply wrong here. The speed of sound is constant (given constant temperature pressure etc.) just like the speed of light. The main difference between sound and light is that sound has an "absolute reference frame" and the analysis must be done in that frame (the frame where the air is at rest).

Here is how the analysis goes for sound:
If an emitter is emitting a sound wave while moving relative to the air then during the time between two wavecrests it has moved so the wavelength will be changed. Since the speed of sound is constant the frequency will change in the opposite direction of the wavelength. For a detector moving relative to the air the process is simply the (time) reverse of the emission.

Again, the Doppler effect requires that the speed of sound be constant!

-Dale

 

[dav57]

Hi dav57,

Solve Maxwell's equations to find the radiation emitted from a source moving relative to you. The wavelength is, after all, simply a description of a certain property of the electromagnetic field. You will then see quite clearly why the wavelength changes. Those same equations further imply that while the frequency and wavelength may change depending on the motion of the source, their product is simply the speed of light. Perhaps a better question for you to ask yourself is why you insist it can't change.

PM, Thanks, but once again you're speaking of the "source" moving and producing varying wavelengths. I have no problem with that.

It seems that relativity always puts the reason for the Doppler shift as being due to the "source" moving and this is where my problem lies.

If you extinguish the source and concentrate on what is left (a beam of light travelling through space) then an observer is now free to move about with a relative speed based on him and the beam (not the source).

I'm asking why the wavelength changes (ignore relativity and think classically because the classical bit outweighs the time dilation and length contraction) when the observer increases his speed towards the beam and also why he observes a frequency change? He can no longer consider himself moving relative to a source because the source ceased to exist possible billions of years ago.

 

[dav57]

You are simply wrong here. The speed of sound is constant (given constant temperature pressure etc.) just like the speed of light. The main difference between sound and light is that sound has an "absolute reference frame" and the analysis must be done in that frame (the frame where the air is at rest).

Here is how the analysis goes for sound:
If an emitter is emitting a sound wave while moving relative to the air then during the time between two wavecrests it has moved so the wavelength will be changed. Since the speed of sound is constant the frequency will change in the opposite direction of the wavelength. For a detector moving relative to the air the process is simply the (time) reverse of the emission.

Again, the Doppler effect requires that the speed of sound be constant!

-Dale

Why are you talking about the EMITTER changing the wavelengh all the time?

If you approach an already emitted sound wave then you as an observer will measure an increased frequency and yet the wavelengh remains unchanged.

Think about the observer moving, NOT the emitter.

Measuring sound in this case is different to light because light "appears" to change its wavelength as you the observer accelerate towards it.

 

[dav57]

Some more thoughts ... Light doesn't have a reference frame, and so you can't move relative to it.

If you can't move relative to light then how has science derived a figure for our velocity relative to it?

I guess you're glad I'm not one of your students...ahhem

 

[Quantum Quack]

I still think my diagram of two ships heading towards a star is a worth while example of a stationary source and doppler effects that are a ship produced pheno and not a star produced pheno.

How fast does a wave of light travel....my guess is it should be 'c'.
And what is this 'c' relevant to? My guess is the ships.
And if the ships can cause by their velocity an increased rate of waves then if they measured the speed of those waves which is used to measure the speed of light what would their measuring ascertain? My guess is that the waves are inceased in their rates or arrival at the ships.
How is the rate of the waves increased with out increacing the relative speed of the waves with regards to the ship?

So if the waves are increasing their rate and yet the rate is still as it was emmitted from the source then one can only conclude that the waves are arriving faster at the ship if the ship is heading with more velocity towards the source. Therefore the speed of those waves would be recorded as being >'c'.

Obviously this must be wrong...but I fail to see how it could be.

 

[Pete]

I'm asking why the wavelength changes (ignore relativity and think classically because the classical bit outweighs the time dilation and length contraction) when the observer increases his speed towards the beam and also why he observes a frequency change? He can no longer consider himself moving relative to a source because the source ceased to exist possible billions of years ago.

Sure he can. The when the light was emitted, the source had some motion. The doppler effect is determined by that motion in the observer's rest frame.

If you can't move relative to light then how has science derived a figure for our velocity relative to it?

No... that figure is the speed of light relative to anything. Not the velocity of anything relative to light.

I still think my diagram of two ships heading towards a star is a worth while example of a stationary source and doppler effects that are a ship pheno and not a star pheno.

The thing is, QQ, that according to the observer, it's always the source that's moving. According to the observer, the observer is at rest, right?

If the observer considers themselves to be moving, then they're obviously using either clocks or rulers or both from a different frame... and that means trouble. If they're not using their own clocks, then they're not getting an accurate doppler measurement. But if they use their own clocks together with rulers that are moving relative to them (eg at rest with respect to the source), then if they measure the speed of the light beam, they don't get c.

 

[dav57]

The thing is, QQ, that according to the observer, it's always the source that's moving. According to the observer, the observer is at rest, right?

Not if the observer is accelerating, Pete. And the source was NOT accelerating when it emitted the light.

 

[Pete]

Even if the observer is accelerating, they are always at rest relative to themselves.

Actually, an accelerating observer brings difficulties. I don't know how to express the doppler shift for an accelerating observer (or an accelerating source for that matter).

Would you be willing to consider an observer that accelerates, then coasts while measuring the beam, accelerates, coasts and measures, etc?

 

[Quantum Quack]

I am confident that a gendanken about doppler effects from a stationary light source could be constructed. It wouldn't be that hard to have observers maintaining a record of frequency in stationary or co-moving frame with that light source.


A single rail with two platforms moving towards a central light source that is also fixed and stationary to that same rail for example.

It could easilly be even experimented on I would imagine.

I'll draw a little diagram to show what I mean. I also fail to see how it is not subject to reciprication either.

 

[Pete]

I am confident that a gendanken about doppler effects from a stationary light source could be constructed. It wouldn't be that hard to have observers maintaining a record of frequency in stationary or co-moving frame with that light source.

I'm not sure what your intention is, QQ... are you asking a moving observer to use clocks and rulers in the source's rest frame to measure the frequency and wavelength of the light beam?

What's the point of that?

 

[Quantum Quack]

Given that time dilation and length contraction are not a significant factor I see no reason not to.
Even if they were considered and produced an error of 80% we still will record the light as >'c'. I feel.