Come on robosteve. You are smarter than to call me ignorant and stop replying.
My apologies. I am very tired at the moment, and so rather irritable and lacking in patience. I'll try to explain what's in my mind, though:
Consider the Earth at arbitrary time
t = t0. Call an inertial frame of reference with velocity equal to the Earth at
t0 "FOR
0", which is moving at velocity
v = v0.
Now, consider the Earth at
t = t1 = t0 + dt. Its velocity will, of course, be equal to
v1 = v0 + dv, where
dv = g dt (since
a = g = dv/dt). This corresponds to "FOR
1".
Consider the next infinitesimal increase in
t, which will result in "FOR
2" having velocity
v2 = v1 + g dt. According to Newtonian mechanics, one would expect that FOR
2 is moving at
v0 + 2g dt, but this is not the case.
Using the velocity addition formula, we can calculate
v2 in terms of
v0, as follows:
v2 - v0 = (2 g dt) / (1 + (g dt/c)^2)Now, since
t0 is an arbitrary starting point, we can generalise:
vn = vn-2 + ( (2 g dt) / (1 + (g dt/c)^2) )And, using recursion:
vn = v0 + ( (ng dt) / (1 + (g dt/c)^2) )I'm too tired to check for errors in this working, and to take the limit as n approaches infinity, but it should come out as
vn = v0 + c, such that with constant acceleration, the velocity approaches, but never reaches, the speed of light.