0

This question cropped up while I was playing with the equation for time dilation. If I set the speed to be $i$ (imaginary unit) the answer from the equation still makes sense, but does that matter if the input data doesn't make sense? Or rather does the input data make sense?

Let $v=i$ and $c=2$ then $T' = T \sqrt{1 -\frac{i^2} {2^2}}$, or $T' = T \sqrt{1 -\frac{-1} {4}}$, or $T' = T \sqrt{\frac{5}{4}}$.

Am I completely out of bounds here or is it something that can be explained?

Brandon Enright
  • 11,937
fimas
  • 111

2 Answers2

2

In what sense does the answer make sense? Does it describe some physical phenomenon?

You've done the algebra right, and gotten a result, but I don't think your inputs (imaginary velocity) mean anything physically. So the answer can't be expected to mean anything physically.

garyp
  • 22,210
  • The physical phenomenon it describes is the time dillation $T'$ between two objects moving relative to eachother. The relative differense in speed between them is $i$. I.e. object $a$ travels away from object $b$ at the speed of $i$ units for 1 unit of time. The speed unit used defines the speed of light in a vacuum as 2. How much does the time differ from $a$ and $b$? In the case above it differs by $\sqrt{\frac{5}{4}}$ units of time. – fimas Jul 25 '14 at 17:25
  • The experimental results for time dilation show that an observer moving relative to a clock sees the hands of a clock moving more slowly than an observer stationary with respect to the clock. Your analysis shows the opposite. Note that in your notation, $T'$ is the time interval between two clock ticks in the stationary frame. Your analysis does not describe the time dilation that is actually observed. It does not describe a physical phenomenon. – garyp Jul 25 '14 at 17:48
  • I guess that answers it then, an imaginary/complex speed value does not make sense since it breaks physics. – fimas Jul 25 '14 at 22:02
  • Another way to see that it can't be of interest physically is to consider the case where $v=c$, which is intermediate between $v<c$ and $v>c$. You get nonsense for $v=c$. –  Oct 02 '14 at 21:49
1

First off Gamma is $\frac{1}{\sqrt{1-v^2/c^2}}$ but that's beside the point.

While by itself it looks okay, if you plug i into a Lorentz transformation everywhere there's a velocity you will also need to either have your time component complex or one of your spatial components as a complex number. This results in that component reversing sign meaning that the proper time or proper separation between events will no longer be equal which was the whole point of a Lorentz transformation in the first place.

Of note is that occasionally time is defined as complex for calculations in general relativity but this is done for both T and T' (I assumed that you were only defining it for one side of the equation); while this does result in what appears to be an imaginary velocity it's just a mathematical trick so the physicist doesn't have to bother keeping the signs straight in the calculations, since at the end it does not change the physics.

Floris
  • 118,905
Reversenorm
  • 21