/sci/ - Science & Math

>>15033209
>>15029272

The major flaw remains exactly the same, his derivation of the transverse Doppler effect is not actually transverse. This is because he is considering the difference in between the moving object at two times (marked S and S').

In the article it's claimed this is a derivation of the transverse Doppler effect (page 25). It is not. At point S, the direction of motion (v) is not perpendicular to the line of sight connecting SR. It's only at the very end of the path when the source reaches S' that the velocity is actually perpendicular. But the equation is using this whole motion, when it contains a combination of transverse and longitudinal effects. If the distance between the source and R is changing significantly it's not the transverse Doppler effect.

If one takes a different arrangement (right) where the source starts at a location vt/2 right of the vertical axis, and ends vt/2 to the left. This is a better approximation to the truly transverse effect because symmetry cancels the change in distance. Indeed we see the starting and final distances (SR and S'R) are equal, so by the derivation so are the wavelengths (lambda=lambda'). Just by changing the starting position slightly the "transverse" Doppler effect has disappeared. Because what was derived was not the transverse effect. In relativity there would still be a transverse effect here.

The whole derivation is silly. The author uses these big time steps and then cannot isolate the transverse effect because the angle is changing constantly along the path. This is noted elsewhere in the paper but ignored here. When derived in the normal way one only considers a single instantaneous moment in time. This could be achieved by shrinking t, and hence vt.

I'm quite sure this will fall on deaf ears. This has been pointed out to the author before, he doesn't care. This thread blew up but we all know none of these people begging for the refutation will consider any of this carefully.