The question Benefit of sling shot effect with a space elevator implies that you can get free $\Delta V$ from a space elevator launch. As highlighted in this answer on SciFi quoting Sheffield book "The Web Between the Worlds", and this answer on Space The launch is not really free, it is stealing rotational energy from Earth.

So two questions, given a space elevator launch from 100,000 kilometers, ignoring the energy cost getting mass to and from the launch point, and assuming all mass involved goes up and down the space elevator:

  1. How much mass would need to be launched for the Earth to lose 1 second of rotational speed?

  2. How much mass would need to be lowered to the Earth to gain 1 second of rotational speed?

For prospective, on potential launch mass.

  • $\begingroup$ I did not want to assume that raising and lowering would have the same values. For a change of one second, they may be nearly identical but for a change of a minute or an hour, I would expect to there to be a significant difference. $\endgroup$ Commented Sep 8, 2013 at 11:25
  • 5
    $\begingroup$ There is no such thing as a free lunch. Wait, what.. launch! $\endgroup$ Commented Sep 8, 2013 at 14:07
  • $\begingroup$ The problem is your assumption. TANSTAAFL - You cannot ignore the cost to get to the launch point. $\endgroup$
    – Rory Alsop
    Commented Sep 8, 2013 at 14:32
  • $\begingroup$ @RoryAlsop I used "ignoring the energy cost" as in to "not consider" there are many variables to allow for, that would inhibit calculating an answer, and make the question overly broad. $\endgroup$ Commented Sep 8, 2013 at 14:38
  • 2
    $\begingroup$ The funny thing with getting to the launch point is you pay only to get to GEO. "Ascent" beyond GEO produces energy which you can use to lift other cargo or sell - in essence we have a power plant that uses Earth rotation as power source and slowing down Earth is the only inherent cost. So, yes, ignoring technological, staff, maintenance and such costs, the cost of space elevator launch could be actually negative! $\endgroup$
    – SF.
    Commented Sep 8, 2013 at 16:21

2 Answers 2


This is a simple calculation in the conservation of angular momentum.

The angular momentum of uniform sphere is ${2\over 5}MR^2\omega$. We will assume that Earth is a uniform sphere (close enough for this question), so $M=5.972\times 10^{24}\,\mathrm{kg}$, $R=6371\,\mathrm{km}$ (mean), and $\omega=7.292\times 10^{-5}\,\mathrm{s}$. So the angular momentum of the Earth is $L_e=7.071\times 10^{33}\mathrm{kg\,m^2\over s}$.

The angular momentum of a small object rotating about a point is $mr^2\omega$. $\omega$ does not change when climbing the elevator, just $r$. So the change in angular momentum is $\Delta L=m\omega(r_f^2-r_i^2)$. $\omega$ is that of Earth above.

To change the Earth's rotation rate by $1\,\mathrm{s}$, we need to change its angular momentum by $1\over 86164$ of what it was, so $\Delta L=8.206\times 10^{28}\mathrm{kg\,m^2\over s}$. Assuming a space elevator at the equator, and using the equatorial radius of the Earth, $6378\,\mathrm{km}$, and $100000\,\mathrm{km}$ above that, we get $\Delta L=8.206\times 10^{28}\mathrm{kg\,m^2\over s}=m\omega\left(\left(106378\,\mathrm{km}\right)^2-\left(6378\,\mathrm{km}\right)^2\right)$. To the extent that $m$ reduces (or increases) $M$, there is also a small change in the rotation rate for the given angular momentum when it is released. However that is negligible here.

This can be simplified and approximated, in order to better see the sensitivities, as: $m={2\over 5}M{\Delta T\over T}{R^2\over H^2}$, where $T$ is the rotation period of the Earth, $\Delta T$ is the small amount by which you want to change it, and $H$ is the radius of the release point from the elevator. $m$ is the amount of mass to let go of at the release point to effect a change in rotation period of $\Delta T$.

That gives $m\approx 10^{17}\mathrm{kg}$. About a billion (US-style) of your aircraft carriers.

If each aircraft-carrier size spaceship only carried a handful of people, then we could evacuate the entire human population from the Earth, only slowing Earth down by one second. So I would say, for all practical purposes, yes, a space elevator is free. You just have to pay the energy to climb to that altitude, which is quite small compared to the energy from the velocity that you get.


FWIW, I'm with Arthur C. Clarke on this whole "space elevator" business. There is no "free lunch" (or is it "free launch"?) to be had with any "space elevator" that is supposedly capable of taking a payload with non-negligible mass and with non-negligible air resistance in the lower atmosphere to the stratosphere and beyond. It doesn't really matter then where we put this alleged "space elevator", so I'll take a few shortcuts to answer this as easy as possible. Take it as it is, I have no interests in discussing "space elevators".

For your first question, there is no upper limit, short of removing the total of the Earth's mass. The remaining mass would still have radial velocity equal to the total mass from before, only it's kinetic potential will be lowered as the mass spinning decreases, so it would be more prone to changes in its rotation velocity, if we added some of that mass back later on.

The second question is also relatively easy, if we assume the kinetic potential of the added mass in the direction of the Earth's rotation is equal to zero, i.e. it's lowered directly perpendicular to the Earth's surface and dropped at either of the poles (another answer by @MarkAdler already includes calculations for dropping the mass at the equator):

$$\frac{duration\ of\ Earth\ day\ in\ seconds}{Earth's\ total\ mass} = x \left(\frac{duration\ of\ Earth\ day\ in\ seconds + 1\ }{Earth's\ total\ mass}\right)$$

So this comes out as $1/86400$  of the Earth's mass ($5.972 * 10^{23} kg$), which is $6.9123 * 10^{18} kg$, if my data input was correct. But please do check again. ;)

But assuming my pocket calculator doesn't lie, this would come out as:

  • 72,929,329,875 (nearly 73 billion) of USS Enterprise (CVN-65) aircraft carriers
  • 57,602,623,456,791 (a good 57 and a half trillion) of fully loaded Space Shuttle Orbiters
  • 1
    $\begingroup$ You can't have a space elevator at the pole. It would fall down. The question was about a space elevator. $\endgroup$
    – Mark Adler
    Commented Sep 8, 2013 at 16:05
  • $\begingroup$ @MarkAdler - You can have it, but it won't be stationary to the Earth's surface, ie. it will be in a polar orbit. But as far as feasibility goes, you can't have a "free lunch" (or "free launch" as OP put it) space elevator. Full stop. As soon as some mass on it shifts its center of mass towards the Earth and it doesn't reach orbital velocity (the lower to the Earth, the faster it should go) any more, it will drop down just as well. Not to even speak of atmospheric drag. There's even calculations on the net what path its tether would draw on the Earth's surface. ;) $\endgroup$
    – TildalWave
    Commented Sep 8, 2013 at 16:12
  • $\begingroup$ My answer has nothing to do with "dropping" mass at the equator. My answer only works for a space elevator, where by virtue of being attached to the surface, it launches objects by taking small amounts of Earth's angular momentum and angular kinetic energy. The answer for "dropping" mass at the equator or removing mass from the equator would have an entirely different answer. $\endgroup$
    – Mark Adler
    Commented Sep 8, 2013 at 16:21
  • $\begingroup$ @MarkAdler - You were arguing that you can't have a polar space elevator. I say you can't have a space elevator. It only works on paper as a scribble. When you start attaching mass (that so happens has a physical presence, so also air resistance, among other things) to it at various elevations, nothing of it is "free". And if we scratch that "free" part, you can even add mass to the poles, in theory at least. Something that "space elevator" fails anyway. So I'm answering the other part of the question in as simple terms as possible. $\endgroup$
    – TildalWave
    Commented Sep 8, 2013 at 16:31

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.