4

Is there a trivial solution for the catenary equation with an additional weight attached to a given position on the rope? Like you would hang something from a rope at some given point and the cable weight can not be neglected. For example:

  • Pole distance w: 10m
  • Cable length l: 12m
  • Distributed load Fl = 1N/mm
  • Additional weight Fg = 300N
  • Weights position x: 8.5m

x is measured in the horizontal between the poles. My goal is not to find the exact position of the rope at every position, but to find the exact position of the added weight.

As you can see in the image. The blue line from D to E is the rope, hanging without any attached weight with the length $l$, the gray lines $i$ and $j$ represent a stage, where a weight is added at the point B assuming, that the rope is fully stretched, and the cable weight can be neglected.

  • $l = i + j $

Now in the real world, the rope will not be fully stretched, if the weight is not big enough. So I would like to now how much this will reposition B with some given values.

The blue line is given by the catenary equation

  • $f(x) = \frac{e^{ax} + e^{-ax}}{2a}$

enter image description here

  • Welcome to MSE. Your question is phrased as an isolated problem, without any further information or context. This does not match many users' quality standards, so it may attract downvotes, or be closed. To prevent that, please [edit] the question. This will help you recognize and resolve the issues. Concretely: please provide context, and include your work and thoughts on the problem. These changes can help in formulating more appropriate answers. – José Carlos Santos Nov 10 '22 at 09:44
  • Your problem is interesting, for sure. If you write the equations, even if the solution is far to be obvious, we could try. But show your efforts first – Claude Leibovici Nov 10 '22 at 09:46
  • Use the same technique used to calculate the catenary on the two portions of the rope separated by the weight. – CyclotomicField Nov 10 '22 at 10:56
  • @CyclotomicField I thought about that to, but the othogonal of the lines i and j are not parallel to the gravitational force. The rope parts will not hang symmetricly. – Samuel A. Nov 10 '22 at 11:32

1 Answers1

2

There is a solution, but it is not trivial at all. Well, there are non-linear equations that give you the solution, and an analytical solution maybe for special cases.

Catenary Equations

Basically, you have two catenary curves on both sides of the cable. Where they meet there is a slope discontinuity such as to support the added weight.

fig1

The above is the free body diagram of the weight $W$. On each side of the cable has common constant horizontal tension $H$, and different vertical tension components $V_1$ and $V_2$. If the attached weight was zero then $V_1+V_2=0$. For the general case $V_1+V_2 = W$ is the vertical force balance equation. Each vertical force component of a catenary is related to the horizontal force component via the slope since tension has to act in a tangent fashion to the curve. In equation form this is $$ V(x) = H \frac{\rm d}{{\rm d}x} y(x)$$ and it sets the constraint (2) to describe the force balance below.

The equations for the two curves are

$$ \begin{aligned}y_{1}(x) & =a\left(\cosh\left(\frac{x-x_{1}}{a}\right)-\cosh\left(\frac{x_{1}}{a}\right)\right)\\ y_{2}(x) & =a\left(\cosh\left(\frac{x-x_{2}}{a}\right)-\cosh\left(\frac{S-x_{2}}{a}\right)\right) \end{aligned} $$

where both share the same catenary constant $a=H/w$, with $w$ is the weight per length unit. Also, the two curves have different curve minimum positions $x_1$ and $x_2$. Finaly, $S$ is the horizontal span of the cable.

The above are subject to the constraints $y_1(0)=0$ and $y_2(S)=0$ already. What is missing is $x_1$ and $x_2$, the lowest points on the two curves. Once those are known, the curves are fully defined.

There is a continuity constraint at the weight location $x=x_M$

$$ y_1 (x_M) = y_2 (x_M) $$

or specifically

$$ \cosh\left(\frac{x_M-x_{1}}{a}\right)-\cosh\left(\frac{x_{1}}{a}\right) = \cosh\left(\frac{x_M-x_{2}}{a}\right)-\cosh\left(\frac{S-x_{2}}{a}\right) \tag{1} $$

The additional weight $W$ adds the constraint of slope between the two curves as

$$ W = H \left. \left( \frac{{\rm d}y_2}{{\rm d}x} - \frac{{\rm d}y_1}{{\rm d}x} \right) \right|_{x=x_M} $$

or

$$ W = H \left( \sinh \left( \frac{x_M-x_2}{a} \right) - \sinh \left( \frac{x_M-x_1}{a} \right) \right) \tag{2} $$

These two constraint equations are to be solved for the location of the lowest points of the two curves $x_1$ and $x_2$. The equations are highly non-linear and difficult to solve analytically.

Finally, note that the total length of the cable is determined by the integral $L=\int_{0}^{x_{M}}\sqrt{1+\left(\tfrac{{\rm d}}{{\rm d}x}y_{1}\right)^{2}}\,{\rm d}x+\int _{x_{M}}^{S}\sqrt{1+\left(\tfrac{{\rm d}}{{\rm d}x}y_{2}\right)^{2}}\,{\rm d}x$ which evaluates to

$$ \small L=a\left(\sinh\left(\frac{x_{M}-x_{1}}{a}\right)-\sinh\left(\frac{0-x_{1}}{a}\right)+\sinh\left(\frac{S-x_{2}}{a}\right)-\sinh\left(\frac{x_{M}-x_{2}}{a}\right)\right) \tag{3}$$

Parabolic Approximation

You can find a first-order approximation of the above problem by noting that $\sinh(x) \approx x$ and $\cosh(x) \approx 1 + \tfrac{x^2}{2}$

This will allow us to find the tensions from the length using

$$ H = \sqrt{ \frac{ x_M (S-x_M) (S w+W) W}{2 S^2 (L/S-1) } + \frac{w^2 S^2}{24 (L/S-1)} } $$

in your example $H=4761$

The lowest points on the curves are

$$\begin{aligned} x_1 &= \frac{S}{2} + \frac{W (S-x_M)}{w S} \\ x_2 &= \frac{S}{2} - \frac{W x_M}{w S} \end{aligned}$$

which are $x_1 = 5.05$ and $x_2 = 4.75$ for the example above.

Here is what the approximate solution looks like:

fig2

As you can see there is a small discontinuity at $x_M = 8.5$

John Alexiou
  • 13,816
  • Sorry, but I don't get what your H is. Is this the x proportion of T1 and T2? And those are the forces on the cables? – Samuel A. Nov 14 '22 at 07:58
  • 1
    I added a bit more explanation under the figure. Here $T$ is the tangent tension on each curve, $H$ is the horizontal component of the tension which is constant throughout the span, and $V$ denotes the vertical component of tension that needs to counteract the added weight $W$. – John Alexiou Nov 14 '22 at 13:30
  • 1
    @SamuelA. - I added an approximate solution to this problem by assuming the catenary curves are parabolas (valid for short spans). – John Alexiou Nov 14 '22 at 15:12
  • Thank you very much for your effort. The approximation works well, if the attached weight is not too big relatively to the cable weight per distance, but if it gets bigger, the discontinuity at xM becomes also bigger. So as this problem is highly non-linear, i assume that this is the best approximation. I am working with python, so a numerical approximation would make sense. Do you think I could just adjust H so that y1(xM)-y2(xM) gets smaller than a given threshold? – Samuel A. Nov 15 '22 at 12:13
  • 1
    In working with the real equations you guess $H$ (or really $a=H/w$) and estimate $x_1$ and $x_2$ using (1) and (2). Then refine $a$ using (3) to match the length. In the end, it is still not realistic because it does not account for the elasticity of the cable which stretches more under its own weight and the additional weight. – John Alexiou Nov 15 '22 at 14:09
  • 1
    @SamuelA. - I am very familiar with this subject and have a GitHub repo on a C# application to model multiple spans. You can scan through it to find CatenaryCalculator.cs which holds all equations needed. But I have not included the additional weight functionality and I might try to do this next. – John Alexiou Nov 15 '22 at 14:13
  • That is a good idea, I will try that. Do you think it would be possible to take into account the elasticity of the cable? You could at least add an approximation of the stretch to the cable when the elasticity module and the attached weight is known. – Samuel A. Nov 21 '22 at 08:29
  • 1
    @SamuelA. - the way to include the cable elasticity is to create a graph of tension vs. geometric strain ($\epsilon = L/S-1$) that includes two curves. A) the catenary loading that goes down with strain as the line is more "slack" and B) the elastic tension response that goes up with strain due to the hook's law. Where the two curves meet, equilibrium is reached and the steady state conditions of the cable are reached. That it until creep kicks in and stretches the cable of long time frames. – John Alexiou Nov 21 '22 at 13:22
  • 1
    @SamuelA. - FYI. I found this reference online for the parabolic approximation of the catenary equations. I don't see them including un-even spans. – John Alexiou Nov 21 '22 at 13:28
  • 1
    @SamuelA. - Uneven span catenary equations: https://physics.stackexchange.com/a/279247/392 – John Alexiou Nov 21 '22 at 13:31