Special theory of relativity question

In summary: L/c’. However, if the same clock is moving with velocity ‘v’, the direction of the light pulse is shown inclined with respect to the light source in the diagrams. This is because the speed of light is constant in any inertial frame, so the direction of light must change with respect to the source in order for both observers to agree that the light was detected. This is a key principle of relativity, where there is no
  • #36
PAllen said:
Oh, and you think you can distinguish motion of observer relative to source versus motion of source relative to observer?! You will not get far in relativity until you can let go of such notions.

It is not an instrument error but a natural phenomenon. Aberration in special relativity was one whole section of Einstein's 1905 paper. Not only is it not inconsistent with his theory, but he used his theory to give a clearer understanding of this long observed phenomenon.

If it is not an instrumental error then why the diagram for aberration shows a telescope. And also why the formula for aberration depends upon on telescope parameters.
 
Physics news on Phys.org
  • #37
lovetruth said:
How can we use ancient theory to solve problems. I might well pretend that Earth is flat and Earth is the centre of the universe.

Sure, it wouldn't change the fact c is measured the same for all observers.

Just because you can't go through the mental exercise of slowing down the light experiment, getting the feel of relative motion for that, and then moving it up speed where relativistic effects becomes apparent does not mean PAllen's suggested course for thought was a poor analogy. In fact it is so good, it is a natural way to go about it (Einstein did, and I am sure you like him).

I would guess thinking through the idea of time/distance and c would help clarify things.

Once you do have a good grasp of SR's implications, you will be fascinated by the light clock and all that you can derive from it.
 
Last edited:
  • #38
PAllen said:
The analogy had a precise purpose. Galilean relativity is extremely accurate for speeds slow compared to light. For such speeds, all laws are Galilean invariant. Yet even here, no SR effects, angles are not preserved across Galilean transforms. The point was that your belief that angles should be preserved is completely absurd.

The system in consideration includes light. If one apply galilean law than, speed of light should not be same in all frames and the angles are variant. Sadly, this is not true. This system includes light and thus, galilean relativity is inapplicable.
 
  • #39
lovetruth said:
If it is not an instrumental error then why the diagram for aberration shows a telescope. And also why the formula for aberration depends upon on telescope parameters.

Questioning things is almost noble.

Questioning people, you might just be being a boob.
 
Last edited:
  • #40
lovetruth said:
If it is not an instrumental error then why the diagram for aberration shows a telescope. And also why the formula for aberration depends upon on telescope parameters.

A telescope is just the particular measuring device for which the effect was first noted. The telescope is just a typical measuring device for an observer.

Why not do as Mentz suggested: apply the Lorentz transform to the equation of a line, e.g. y=x for some t, to primed coordinates. You will see that the equation changes, giving a different slope.

[Edit: if you want to include light, rather than spatial angle, try transforming: y=ct
In the primed coordinates, you will see the the light is not perpendicular to the x' axis.]
 
Last edited:
  • #41
lovetruth said:
The system in consideration includes light. If one apply galilean law than, speed of light should not be same in all frames and the angles are variant. Sadly, this is not true. This system includes light and thus, galilean relativity is inapplicable.

The point was that bullets change angle, and bullets are *certainly* within the accuracy range of Galilean relativity. I was addressing the notion that 'laws the same in all frames' implies 'angles the same in all frames'.
 
  • #42
nitsuj said:
Questioning things is almost noble.

Questioning people, you might just be being boob.

I think you meant noob. Also I don't believe in arguments. But i like engaging in intellectual conversation. I only want people to justify their statements with strong proofs and not abstract belief or intuition.
 
  • #43
PAllen said:
A telescope is just the particular measuring device for which the effect was first noted. The telescope is just a typical measuring device for an observer.

Why not do as Mentz suggested: apply the Lorentz transform to the equation of a line, e.g. y=x for some t, to primed coordinates. You will see that the equation changes, giving a different slope.

[Edit: if you want to include light, rather than spatial angle, try transforming: y=ct
In the primed coordinates, you will see the the light is no longer perpendicular to the x axis.]

Read my reply #30


Lorentz transform only affects the direction which is along the relative velocity v. Therefore, there is only length contraction and not height or width contraction. Therefore the system in consideration will have contraction in the horizontal direction. Since the light is moving in only vertical direction, it will not move at an angle by applyin lorentz transformation.
 
  • #44
lovetruth said:
Read my reply #30Lorentz transform only affects the direction which is along the relative velocity v. Therefore, there is only length contraction and not height or width contraction. Therefore the system in consideration will have contraction in the horizontal direction. Since the light is moving in only vertical direction, it will not move at an angle by applyin lorentz transformation.

This is simply false, as I already explained. Apply the formulas and you will see it is false. There can be no real communication if you simply insist mathematically false statements are true. See post #32 or simply apply the Lorentz transform and you will see this is as absurd as insisting that 1+1=3.
 
  • #45
PAllen said:
The point was that bullets change angle, and bullets are *certainly* within the accuracy range of Galilean relativity. I was addressing the notion that 'laws the same in all frames' implies 'angles the same in all frames'.

If we assume angle changes in different frames then, it will not be wrong to assume different velocity. The fact is: galilean relativity is only for matter. SR is for both matter and light.
 
  • #46
lovetruth said:
If we assume angle changes in different frames then, it will not be wrong to assume different velocity. The fact is: galilean relativity is only for matter. SR is for both matter and light.

The relevant principle is the *speed* of light is constant for all inertial frames, *not* the angle (thus not the velocity, which is different from speed) of light of light is the same for all inertial frames. It isn't. Not by experiment, not by theory. As I have mentioned, Einsteins 1905 paper had a whole section expounding on aberrations and the *non-constancy* of angle of light between frames, while the *speed* of light was preserved.
 
  • #47
PAllen said:
This is simply false, as I already explained. Apply the formulas and you will see it is false. There can be no real communication if you simply insist mathematically false statements are true. See post #32 or simply apply the Lorentz transform and you will see this is as absurd as insisting that 1+1=3.

The principle of time dilation and length contraction are equivalent to lorentz transformation. Thus, applying time dilation and length contraction will give same result. I have applied the lorentz transformation result and don't see any difference. If you think I have done some mistake, please post your calculation here.
 
  • #48
PAllen said:
The relevant principle is the *speed* of light is constant for all inertial frames, *not* the angle (thus not the velocity, which is different from speed) of light of light is the same for all inertial frames. It isn't. Not by experiment, not by theory. As I have mentioned, Einsteins 1905 paper had a whole section expounding on aberrations and the *non-constancy* of angle of light between frames, while the *speed* of light was preserved.

I haven't read THE Einstein paper but read the theory in books. If he had mentioned non-constancy of angle then, he must had given a formula. Please post the formula here.
Also note that angle should not only depend on light source velocity 'v' but also the direction of the light with respect to v. If the light is emitted horizontally, there must be no angle change.
 
  • #49
lovetruth said:
I haven't read THE Einstein paper but read the theory in books. If he had mentioned non-constancy of angle then, he must had given a formula. Please post the formula here.
Also note that angle should not only depend on light source velocity 'v' but also the direction of the light with respect to v. If the light is emitted horizontally, there must be no angle change.

I have only read books mentioning the theory and that simple idea is not blatantly mentioned because it is intuitive when you understand the main point c is constant for every observer.

Interesting you didn't see it in the last line of post #18.
 
  • #50
nitsuj said:
I have only read books mentioning the theory and that simple idea is not blatantly mentioned because it is intuitive when you understand the main point c is constant for every observer.

Interesting you didn't see it in the last line of post #18.

As I said I demand a formula giving the change in angle as a function of velocity v.
 
  • #51
lovetruth said:
The principle of time dilation and length contraction are equivalent to lorentz transformation. Thus, applying time dilation and length contraction will give same result. I have applied the lorentz transformation result and don't see any difference. If you think I have done some mistake, please post your calculation here.

No, the Lorentz transformation also includes the features of relativity of simultaneity not captured by length contraction and time dilation.

The equations (for a vertical light path):

y=ct, x=0

becomes in the transformed coordinates (I will use x`, y`, t`):

y`/gamma - (v/c) x` = c t` , x` + v t` = 0

describing an angled light path. Despite being angled, the light goes from (x`,y`)=(0,0) at t`=0 to (-v, c/gamma) at t` = 1. Using euclidean distance formula on the x`,y` difference, you get c, so the speed is c/1 = c. Angled light, same speed.

Since even at a glance, it is obvious the Lorenz transform will not transform y=ct to y' = ct`, are you sure you don't want to change your pseudonym?

[EDIT: after a little more rearrangement, you get:

y` = ct`/gamma
x`= -vt`

]
 
Last edited:
  • #52
PAllen said:
No, the Lorentz transformation also includes the features of relativity of simultaneity not captured by length contraction and time dilation.

The equations (for a vertical light path):

y=ct, x=0

becomes in the transformed coordinates (I will use x`, y`, t`):

y`/gamma - (v/c) x` = c t` , x` + v t` = 0

describing an angled light path. Despite being angled, the light goes from (x`,y`)=(0,0) at t`=0 to (-v, c/gamma) at t` = 1. Using euclidean distance formula on the x`,y` difference, you get c, so the speed is c/1 = c. Angled light, same speed.

Since even at a glance, it is obvious the Lorenz transform will not transform y=ct to y' = ct`, are you sure you don't want to change your pseudonym?

You have done it all wrong.
y'=y=ct
x'=gamma*(x-vt)
t'=gamma*(t-vx/c^2)
 
  • #53
lovetruth said:
As I said I demand a formula giving the change in angle as a function of velocity v.

If a light ray has angle A in a given frame, than in a frame moving at speed v in the +x direction, the angle in the new frame (A`) is given by:

cot(A`) = (cot(A) - (v/c) cosec(A)) gamma

This differs from Galilean aberration by factor of gamma. I don't know if this effect has been observed - the difference from the Galilean formula is *extremely* small for the Earth's motion.
 
  • #54
lovetruth said:
You have done it all wrong.
y'=y=ct
x'=gamma*(x-vt)
t'=gamma*(t-vx/c^2)

This borders on insanity. You start with:

y=ct, x=0

to describe a light ray going in +y direction at x=0.
Then you make the following substitutions:

y' for y
(t'+x'v/c^2) gamma for t
(x' + v t') gamma for x

Rearrange, and you get what I wrote.
 
  • #55
PAllen said:
This borders on insanity. You start with:

y=ct, x=0

to describe a light ray going in +y direction at x=0.
Then you make the following substitutions:

y' for y
(t'+x'v/c^2) gamma for t
(x' + v t') gamma for x

Rearrange, and you get what I wrote.

check,
http://en.wikipedia.org/wiki/Lorentz_transformation
 
  • #56
lovetruth said:

You don't know the first thing about using it. You are using it in the wrong direction, and you are using it wrong. To go from an equation in one frame to another, you need to substitute for all instances of x,y,t. Further, to go from an equation in x,y,t to x',y',t' you want to use the expressions for x in terms of (x',t'), t in terms of (t',x') etc.
 
  • #57
LoveTruth said:
All in all, I ask does any theory (including the relativity) predicts the light will move at an angle with respect to the light source when the light source is moving.

This is a good question so I'll have another go.

The answer is to use the wave model of light. When the light beam is turned on, imagine a circular wave front expanding from the emitter. After the moment of emission, the receiver moves a certain distance before the wavefront intersects with it. The light then appears to have traveled at an inclined path in the ground frame. For a collimated beam, a detailed analysis would show that interference only supports the path between the emitter and the detector it was aimed at in the rest frame.

So, the theory(s) required is wave optics, or QED as described by R. Feynman, with the assumption that speed of light is invariant.
 
  • #58
A couple of final points:

1) I gave the equation for a light path in (x`,y`,t`) given y=ct,x=0 in (x,y,t) as:

y` = ct`/gamma
x`= -vt`

Equivalently, if you have a light path in (x`,y`,t`) given by y`=ct`,x`=0 , you would have
in (x,y,t):

y = ct /gamma
x = vt

2) To amplify a little more on Mentz114 explanation:

Imagine observer B (moving at +v in x relative to A) sees his light source emit a single spherical wave front. Restricting to a plane, we have a circular wave front. B sees the wave front arrive simultaneously at both ends of his mirror, which he interprets as an orthogonal light signal, because his mirror is perpendicular to the line from the source to the center of the mirror.

A sees the light pulse with a spherical wave front in *his* frame. He sees B's mirror intersect the wave front at an angle (because it has moved since the signal was emitted). However, A sees the wave front arrive one end of B's mirror before the other end. A sees B calling these two separate arrival times simultaneous. Thus A sees how B interprets the signal as orthogonal, even though to A it is clearly received at an angle by B. Thus the explanation of the different perceived angle is the relativity of simultaneity.
 
  • #59
Mentz114 said:
This is a good question so I'll have another go.

The answer is to use the wave model of light. When the light beam is turned on, imagine a circular wave front expanding from the emitter. After the moment of emission, the receiver moves a certain distance before the wavefront intersects with it. The light then appears to have traveled at an inclined path in the ground frame. For a collimated beam, a detailed analysis would show that interference only supports the path between the emitter and the detector it was aimed at in the rest frame.

So, the theory(s) required is wave optics, or QED as described by R. Feynman, with the assumption that speed of light is invariant.

Wow, I cannot imagine the light clock with the light "beam/particle" as a wave. That's beyond me. (I thought light was like ping pong balls lol j/k)
 
  • #60
lovetruth said:
I think that since the speed of light isn't affected by the light source in any way, so must be the direction of the light.
That is wrong. Funny enough you phrased it correctly in your originally post: Relativity theory tells that the speed of light is constant in any inertial frame.
Think if instead of the laser, i have a point light source. Also, bullet and the gun analogy can't be applied here bcoz bullet is matter and light is light.
Both have momentum, and the law of conservation of momentum (unchanged since Newton!) applies to both
Both behave differently otherwise, we won't need theory of relativity in the first place. Otherwise, Galilean relativity should suffice then.
Thx for your valuable time thou
You're welcome. :smile:
What is different is, again, that SR has a limit speed, which is equal to the speed of light. I hope that it is clear now. :wink:
 
  • #61
PAllen said:
If a light ray has angle A in a given frame, than in a frame moving at speed v in the +x direction, the angle in the new frame (A`) is given by:

cot(A`) = (cot(A) - (v/c) cosec(A)) gamma

This differs from Galilean aberration by factor of gamma. I don't know if this effect has been observed - the difference from the Galilean formula is *extremely* small for the Earth's motion.

While I don't know that the relativistic correction is big enough to see for effects of the Earth's changing velocity relative to astronomic sources, there is another context where the relativistic corrections is believed to be observed. This same equation explains (a portion of) relativistic beaming, where a larger solid angle of emitted light transforms to a smaller solid angle when the source is rapidly approaching; and the converse for a receding source. This is believed to explain why relativistically moving plasma jets from compact sources typically look similar if they are orthogonal to our line of view, but an approaching jet is *much* brighter than a receding jet.
 
  • #62
Mentz114 said:
This is a good question so I'll have another go.

The answer is to use the wave model of light. When the light beam is turned on, imagine a circular wave front expanding from the emitter. After the moment of emission, the receiver moves a certain distance before the wavefront intersects with it. The light then appears to have traveled at an inclined path in the ground frame. For a collimated beam, a detailed analysis would show that interference only supports the path between the emitter and the detector it was aimed at in the rest frame.

So, the theory(s) required is wave optics, or QED as described by R. Feynman, with the assumption that speed of light is invariant.

Thanx Mentz. You have clearly resolved all doubts. Light in nature is produced as spherical wave by oscillating charged particles. For gettin a linear beam of light, the 3D spherical wave is passed thru an aperture in some opaque screen.
If we imagine the linear light beam source is composed of a point light source and a opaque screen with an aperture, we can see the light beam comin out of aperture as a light beam coming out at an angle. If there is some distance between the point light source and the aperture, there will be a delay for the light to move from the point source to the movin aperture. Thus only an angled light beam will come out of aperture as calculated by simple geometry.
Please post ur comment if u think my understandin is wrong or u have a better answer.
 
  • #63
lovetruth said:
Thanx Mentz. You have clearly resolved all doubts. Light in nature is produced as spherical wave by oscillating charged particles. For gettin a linear beam of light, the 3D spherical wave is passed thru an aperture in some opaque screen.
If we imagine the linear light beam source is composed of a point light source and a opaque screen with an aperture, we can see the light beam comin out of aperture as a light beam coming out at an angle. If there is some distance between the point light source and the aperture, there will be a delay for the light to move from the point source to the movin aperture. Thus only an angled light beam will come out of aperture as calculated by simple geometry.
Please post ur comment if u think my understandin is wrong or u have a better answer.

Yes, you are right about the aperture. It will move relative to the point source after the light is emitted and so produce an angled beam. No QED required.:smile:

I made a primitive movie of the situation with a point source. It's here

www.blatword.co.uk/space-time/wavemove.mpeg

size is only 560Kb.
 
Last edited:
  • #64
Mentz114 said:
Yes, you are right about the aperture. It will move relative to the point source after the light is emitted and so produce an angled beam. No QED required.:smile:
Exactly, that's also what I meant with "if it went straight up then it would go through the side of the laser!" :smile:
I made a primitive movie of the situation with a point source. It's here

www.blatword.co.uk/space-time/wavemove.mpeg

size is only 560Kb.

Looks good!

Cheers,
Harald
 
  • #65
Mentz114 said:
Yes, you are right about the aperture. It will move relative to the point source after the light is emitted and so produce an angled beam. No QED required.:smile:

I made a primitive movie of the situation with a point source. It's here

www.blatword.co.uk/space-time/wavemove.mpeg

size is only 560Kb.

Thx for the reply.
 

Similar threads

Replies
14
Views
1K
Replies
111
Views
8K
Replies
53
Views
4K
Replies
45
Views
4K
Replies
1
Views
1K
Back
Top