1. 1/22/2001Physics 222 Special Relativity Lecture 3

2. 1/22/2001Physics 222 Transformation of velocities

3. 1/22/2001Physics 222 Doppler Effect Consider the propagation of a light wave. The traveling electromagnetic wave contains electric and magnetic fields that vary sinusoidally in both space and time. First consider the transformation of light that is emitted parallel to the motion of the emitter: ’

4. 1/22/2001Physics 222 Let u = radial velocity of the two frames (+ when separating), t e = time period of emitted wave crests, t r = time period of received wave crests. The source moves a distance x between emitting successive crests: Classical mechanics: the distance between successive crests is Special relativity: the time between crests as seen by the observer is stretched by time dilatation:

5. 1/22/2001Physics 222 We can generalize this result for arbitrary angle between the emitted direction and the direction of motion of the emitter:  Only the radial component u r = u cos  produces the classical transformation, but the entire velocity u produces the time dilation factor  :

6. 1/22/2001Physics 222 Michelson-Morley Experiment We can now understand the significance of the Michelson-Morley experiment:

7. 1/22/2001Physics 222 The wavelength along each path is shifted according to its orientation to the ether velocity Classical mechanics: When is parallel to the light direction, the velocity of propagation is u ’ = c + u. When it is anti-parallel, it is u ’ = c - u. When it is perpendicular:

8. 1/22/2001Physics 222 In classical mechanics, time runs at the same rate in all frames. So the time  t ’ it takes for light to travel a path length L is (  refers to parallel/antiparallel) Referring to the figure above, the red and blue light paths have transit times that depend upon the relative directions of the incident light and the ether drift:

9. 1/22/2001Physics 222 If u is parallel to the incident light (as shown), If u is perpendicular to the incident light ( pointing down in the picture), The two time differences are not the same, so the interference patterns will shift as the apparatus is turned!

10. 1/22/2001Physics 222 Special relativity: The speed of light is the same in all frames! So the transit times do not depend at all on the orientation of the interferometer with respect to any moving reference frame. The interference fringes will not change as the experiment is rotated. The Michelson-Morley experiment found that the fringe pattern was the same in all orientations of the apparatus, and also when observed at all times of day and night. Since the Earth’s surface velocity in its rotation is ~1,000 mi/hr, this rotation velocity should add when it is parallel to any hypothetical ether drift, and subtract when it is antiparallel. One is then forced to the conclusion that there is no ether, and the speed of light is constant in all frames.