Solibelus said:
Hi, I'm trying to derive the displacement of an object with proper acceleration (constant) α.
t is our laboratory time.
The goal:
[tex]x(t) = {1 \over α} [ \sqrt{1+α^2t^2} -1 ][/tex]
How to approach this? :(
Note for anyone who's wondering: no, this is not homework...
The easiest approach is the textbook approach using 4-velocities, or proper velocities.
You'll find the following worked out in MTW's "Gravitation", in tensor notation. I've re-written it to use non-tensor notation. I've also chosen to use subscripts in places where MTW uses superscripts for readability.
Let u be the 4-velocity (proper velocity) of the object, and let ##\tau## be proper time. We wish to solve for ##u(\tau)##, we will take care of the conversion to find x(t) later.
Then by definition we can write the components of the 4-velocity ##u_0## and ##u_1## as:
##u_0 = dt/d\tau \quad u_1 = dx/d\tau##
where ##x(\tau)## is the displacement in terms of proper time.
The proper acceleration a is just the rate of change of the proper velocity with respect to proper time, i.e
##a_0 = \frac{d }{d\tau} u_0## and ##a_1 = \frac{d }{d\tau} u_1##
The 4-magnitude of the proper velocity is a constant. With the sign conventions being used, it's -1. Thus ##{u_1}^2 - {u_0}^2 = -1##.
The 4-magnitude of the proper velocity is a constant. With the sign convetions being used it's just ##\alpha## (no minus sign). Thus
##a_1^2 - a_0^2 = \alpha^2##
Because the magnitude of the proper velocity is always constant, the proper acceleration is always perpendicular to the proper velocity
##u_1\, a_1 - u_0 \,a_0 = 0##
If you wish, you can derive this mathematically and formally from the fact that ##u_1 = \sqrt{u_0^2 - 1}## using the chain rule.
The simple way to proceed is to note that we have
##{u_1}^2 - {u_0}^2 = -1 \quad u_1\,a_1 - u_0\,a_0 = 0 \quad a_1^2- a_0^2= \alpha^2##
where the last equation comes from the fact that the magnitude of the proper acceleration is constant and equal to ##alpha##.
We can write down the solution ##a_0 = \alpha \, u_1 \quad a_1 = \alpha \, u_0## which we can easily verify by inspection.
This is equivalent to the differential equation:
##\frac{d}{d\tau} u_0 = \alpha \, u_1 \quad \frac{d}{d\tau} u_1 = \alpha \, u_0## Combining this with the fact that ##u_1^2 - u_0^2 = -1## we immediately write down the solution
##u_0 = \cosh (\alpha \, \tau) \quad u_1 = \sinh (\alpha \, \tau)##
We integrate to find:
##t = \frac{1}{\alpha} \sinh(\alpha \tau) \quad x = \frac{1}{\alpha} \cosh (\alpha \, \tau) ##
This doesn't have x=0 at t=##\tau##=0, which is just a matter of an integration constant. But we'll address this issue later, we'll just note at this point we can add a constant value to x and we will still have a solution, the proper velocity won't be affected by a choice of origin for x.
To solve your problem, we don't need to write down the proper acceleration, but it's a good check. We see that it's magnitude is constant as desired.
##a_0 = \alpha \, \sinh(\alpha \, \tau) \quad a_1 = \alpha \, \cosh (\alpha \, \tau)##
To find x as a function of t (rather than x and t as a function of tau) we start with
##t = \frac{1}{\alpha} \sinh(\alpha \tau) \quad x = \frac{1}{\alpha} \cosh (\alpha \, \tau) ##
We note that
##x^2 - t^2 = \frac{1}{\alpha^2}##
Solving for x, we have
##x = \sqrt{t^2 + \frac{1}{\alpha^2}}##
The solution won't be affected if we add a constant to x to make it equal to 0 when t=0 as we mentioned previously. Adding this offset, we have
##x = \sqrt{t^2 + \frac{1}{\alpha^2}} - \frac{1}{\alpha}##
which is equal to (though formatted differently) than the expression you wished to find.