wnvl1 schreef: ↑ma 18 apr 2022, 10:44
Het electrische en magnetische veld gaan dus veranderen.
Nee, die gaan niet veranderen.
Een keuze van coördinaten in elke context kan geen fysisch effect hebben in de klassieke natuurkunde (niet-kwantumfysica).
Ik vind dit wel een goede van Gerard t Hooft:
"There are good ways to define what a Lorentz transformation is, and what its significance is for physics. Consider all *linear* transformations of the four coordinates x,y,z and t, such as x’ = a x + b y + c t , etc. Now look at what happens to the velocity of light in the new coordinates. In general, light signals in different directions would seem to go with different speeds now. Mathematical question: which linear transformations would *not* affect the speed of light? Answer: if the equation
x^2+y^2+z^2=c^2 t^2 (where c=speed of light) stays the same. There are many solutions, but in some of these solutions, other laws of physics do not stay the same, such as the *scale transformations*: x’= a x, y’=a y, ... t’=a t. After such transformations light has the same velocity but atoms would get different sizes and clocks would go with different speeds.
This problem is easy to cure. Just take all linear transformations that keep the value of x^2+y^2+z^2 - c^2 t^2 invariant. That is the group of Lorentz transformations. Add to this the *inhomogeneous* linear transformations, such as x’=x+a, then you get the translations and time translations as well. Combined, we call these the Poincare transformations.
All laws of physics, except gravity, are invariant under these transformations. This is an observation about the laws of nature. No further discussion needed. Is the statement exact? Not quite: gravity does not obey; there, you need non-linear coordinate transformations, which gives you a new theory, even more powerul: general relativity. Again, are all laws of physics exactly invariant under these non-linear coordinate transformations? As far as we know, yes, but one may always speculate on new surprises."
Misschien dat hij nog wil helpen met dit (lange) topic trouwens:
https://www.researchgate.net/profile/Gerard-T-Hooft