Say you're standing on the equator and you have a string below you tied around the equator (40,075 km) that is the length of the equator + 1 meter (40,075.001 km). What is the maximum height you can you lift the string off the ground? Can you create a function of both circumference of the circle (earth) and string to output the distance between the two if pulled tight?


  • For illustration, the result would be pulled from a single point, making a triangle until it met with the earth, in which it would follow the curvature of the earth. Similar to a snow-cone or O>
  • The string does not stretch
  • The earth can be assumed to be a perfect sphere
  • $\begingroup$ How tight is the string? $\endgroup$ – user328032 Apr 18 '16 at 18:50
  • $\begingroup$ Do you know the formula of the circumference of a circle ? $\endgroup$ – callculus Apr 18 '16 at 18:55
  • 2
    $\begingroup$ We get a transcendental equation. One can get an excellent approximation to the solution by using a power series expansion. $\endgroup$ – André Nicolas Apr 18 '16 at 19:05
  • $\begingroup$ I'm far from a math major, and it took me about a day to solve, but you're correct in that the best I could do was a very good approximation. $\endgroup$ – Travis Apr 18 '16 at 19:09
  • 1
    $\begingroup$ See also mathforum.org/mathimages/index.php/Rope_around_the_Earth. $\endgroup$ – lhf Apr 19 '16 at 0:19

We have to solve (with respect to $d$) the trascendental equation: $$ 2\sqrt{(R+d)^2-R^2}+2R\arcsin\frac{R}{R+d}=\pi R+1 $$ or solve (with respect to $x=\frac{d}{R}$) the trascendental equation: $$ 2\sqrt{x^2+2x}+2\arcsin\frac{1}{1+x}=\pi+\frac{1}{R} $$ that is equivalent to: $$ \sqrt{x^2+2x}-\arctan\sqrt{x^2+2x}=\frac{1}{2R}.$$ In order to solve $z-\arctan z=\frac{1}{2R}$ we may apply Newton's method, then $x=-1+\sqrt{z^2+1}$.

Since $z$ is close to zero, $z-\arctan z\approx\frac{z^3}{3}$, hence: $$ d\approx \sqrt[3]{\frac{9R}{32}}\approx \color{red}{121.5\,m}.$$


Draw a picture. Let $P$ be the peak of the stretched string, and let $C$ be the centre of the Earth. Let $G$ be the point on the Earth's surface which is on the line $PC$, and let $T$ and $T'$ be the points of tangency of the string with the Equator after the string has been pushed upwards to full height $PG$.

Let $r$ be the radius of the Earth, and let $\epsilon r$ be the excess amount of string in addition to $2\pi r$. Note that $\epsilon$, in your example, is very small.

Let $\theta=\angle TCG$ Note that $\theta$ is small. The key equation is $$\epsilon r=2r\tan\theta -2r\theta.$$ This holds because $2r\tan\theta$ is the sum of the lengths of the two tangent segments $PT$ and $PT'$, while $2 r\theta$ is the amount of string saved because it no longer covers the minor arc $TGT'$. The difference is equal to the amount $\epsilon r$ of extra string we have available.

Using the fact that $\tan\theta\approx \theta +\frac{\theta^3}{3}$, we get $$\theta\approx \left(\frac{3\epsilon}{2}\right)^{1/3}.$$

Now that we know $\theta$, we can find the height $PG$ of the peak of the string. From the diagram, we can see that $PG=r(\sec\theta-1)$. Since $\theta$ is small, this is well approximated by $\frac{\theta^2}{2}$, and we end up with $$PG\approx \frac{r}{2}\left(\frac{3\epsilon}{2}\right)^{2/3}.$$

  • $\begingroup$ We end up with $$PG\approx \frac{r}{2}\left(\frac{3\delta}{2r}\right)^{2/3}?$$ $\endgroup$ – Narasimham Apr 26 '16 at 21:15
  • $\begingroup$ @Narasimham: Note that I defined $\delta r$ to be the excess amount of string. If you let $\Delta$ be the excess amount, then $\Delta=\delta r$, and we get your expression, with $\frac{3\Delta}{2r}$ instead of your $\frac{3\delta}{2r}$. The reason I defined $\delta$ the way I did was "dimensional." I wanted to end up with an expression that was a constant times $r$. I will check later whether there is a mistake, but dimensionally the formula I gave has the right shape. $\endgroup$ – André Nicolas Apr 26 '16 at 21:22
  • $\begingroup$ Right, my bad. I would use a symbol like $\epsilon$ Shall delete my remark. $\endgroup$ – Narasimham Apr 27 '16 at 3:05
  • $\begingroup$ @Narasimham: I had hesitated between using $\delta$ and using $\epsilon$! Will take that as a suggestion and make the change. $\endgroup$ – André Nicolas Apr 27 '16 at 3:14


Let $\Delta C$ be the extra length of loose string.

$$ r = \frac{C}{2 \pi} ;\ \Delta r = \frac{\Delta C}{2 \pi} = h= \frac{1000\, mm}{2 \pi} = 159. 155 \; {mm \,!} $$

The relation/ratio is the same on Mars or Saturn or any earth based large round cylinder.

In this problem you want $ h $ for given input $ (R, \Delta C) $

We calculate tangent length $T$ shallow horizon range using right triangle property :

$$ T^2 = h ( 2 R -h ) \tag{1} $$

$$ T = R \tan \theta \tag{2}$$

And eliminate $T$ to solve for $ \tan \theta $

$$ \tan \theta= \sqrt{ (h/a) (2-h/a) } \tag{3}$$

From the diagram the slack above involute is indicated by red line,

$$ = IE =\tan \theta - \theta = \Delta C/ (2 h) $$


$$ \sqrt { (h/a) (2-h/a) } -\tan^{-1}{\sqrt{ (h/a) (2-h/a) } } = \Delta C/(2 R) \tag {4} $$

Plugging in numerical values to solve for $h$

$$ R = 40075 * 10^3/( 2 \pi) \, meters ; \, \Delta C =1 m \rightarrow h = 121.508 m \tag{5}$$

$\because h << R$ an approximation using Madhava-Gregory-Leibnitz series

$$ \tan u - \tan ^{-1}u \approx u^3/3 $$

gives us $$ h^3 =\frac {9}{32}\cdot {\Delta C}^2 \cdot R \tag {6}$$

giving only a bit different value

$ h= 121.505 m $.



Half a meter. You take the slack extra meter and lift it half a meter high.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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