Skip to content

Problem Irodov 1.199

Two identical small discs, each of mass mm, rest on a smooth (frictionless) horizontal plane, joined by a light spring of natural length l0l_0 and stiffness ϰ\varkappa that is initially undeformed. At some instant one disc is given a velocity v0v_0, horizontal and perpendicular to the spring’s own line, while the other disc is momentarily still. Let ε=Δl/l0\varepsilon = \Delta l/l_0 denote the spring’s fractional elongation. Given that the maximum value of ε\varepsilon reached during the subsequent motion is small compared to 11, find that maximum fractional elongation εmax⁡\varepsilon_{\max}.

The two discs, each of mass m, joined by the undeformed spring of natural length l_0; one disc is launched with v_0 perpendicular to the spring's line, the other starts at rest.

The two discs, each of mass mm, joined by the undeformed spring of natural length l0l_0; one disc is launched with v0\vect v_0 perpendicular to the spring’s line, the other starts at rest.

Given
  • mm
  • l0l_0
  • ϰ\varkappa
  • v0v_0
Find
εmax⁡=(Δl/l0)max⁡\varepsilon_{\max}=(\Delta l/l_0)_{\max}, given εmax⁡≪1\varepsilon_{\max}\ll 1

Solution

Step 1Reduce to a one-body problem in the center-of-mass frame

Let r=r1−r2\vect r=\vect r_1-\vect r_2 be the separation of the two discs, with r=l0r=l_0 and r˙=v0\dot{\vect r}=\vect v_0 (purely tangential, since v0⊥r\vect v_0\perp\vect r) at the initial instant. The spring pulls each disc toward the other, so the two forces on them are equal and opposite; write F\vect F for the one acting on disc 1.

m1r¨1=Fm2r¨2=−F(1)\begin{aligned} m_1\ddot{\vect r}_1 &= \vect F \\ m_2\ddot{\vect r}_2 &= -\vect F \end{aligned} \tag{1}

Dividing each line of (1) by its own mass isolates the two accelerations, and subtracting the second from the first gives the acceleration of the separation itself, since r¨=r¨1−r¨2\ddot{\vect r}=\ddot{\vect r}_1-\ddot{\vect r}_2.

r¨=Fm1−(−Fm2)=(1m1+1m2)F\begin{align*} \ddot{\vect r} &= \frac{\vect F}{m_1} - \left(-\frac{\vect F}{m_2}\right) \\ &= \left(\frac{1}{m_1}+\frac{1}{m_2}\right)\vect F \tag{2}\end{align*}

The bracket in (2) is the reciprocal of a single mass, and naming it turns a pair of coupled equations into one equation of motion for the separation alone.

1μ=1m1+1m2=m1+m2m1m2(3) \frac{1}{\mu} = \frac{1}{m_1}+\frac{1}{m_2} = \frac{m_1+m_2}{m_1m_2} \tag{3}

With that name, (2) is the equation of motion of a single fictitious particle of mass μ\mu moving under the spring’s central force F(r)\vect F(r).

μ r¨=F(r)(4) \mu\,\ddot{\vect r} = \vect F(r) \tag{4}

Inverting (3) and putting in the two equal masses of this problem,

μ=m1m2m1+m2=m⋅mm+m\begin{align*}\mu &= \frac{m_1m_2}{m_1+m_2} \\ &= \frac{m\cdot m}{m+m}\end{align*}
μ=m2\mu = \dfrac{m}{2}
(5)
Step 2Conserve angular momentum and energy of the relative motion

The force in (4) is central, directed along r\vect r, so it has no moment about the origin of the relative coordinate and the angular momentum of the relative motion is conserved; a spring is also conservative, so the mechanical energy of that motion is conserved too. Work in polar coordinates (r,φ)(r,\varphi), in which the velocity splits into a radial part r˙\dot r and a transverse part rφ˙r\dot\varphi.

The angular momentum is the fictitious particle’s mass times the moment of its transverse velocity about the origin.

L=μ r (rφ˙)=μr2φ˙(6) L = \mu\,r\,(r\dot\varphi) = \mu r^2\dot\varphi \tag{6}

The energy is the kinetic energy built from both parts of that velocity, plus the spring’s potential energy, which depends on the elongation r−l0r-l_0 alone.

E=12μ(r˙2+r2φ˙2)+12ϰ(r−l0)2(7) E = \frac{1}{2}\mu\bigl(\dot r^2+r^2\dot\varphi^2\bigr)+\frac{1}{2}\varkappa(r-l_0)^2 \tag{7}

At t=0t=0 the spring is at its natural length, r=l0r=l_0, and v0\vect v_0 is perpendicular to r\vect r, so the whole relative velocity is transverse: r˙=0\dot r=0 and rφ˙=v0r\dot\varphi=v_0. Evaluating (6) and (7) at that instant,

L=μl0v0,E=12μv02L=\mu l_0 v_0,\qquad E=\dfrac{1}{2}\mu v_0^2
(8)
Step 3Locate the turning point of the radial motion

The spring is at its most stretched when the separation stops changing, r˙=0\dot r=0. At that instant the transverse speed is fixed by angular momentum conservation alone.

rφ˙=Lμrr\dot\varphi = \frac{L}{\mu r}

The energy (7) there is therefore purely transverse kinetic plus spring potential, the radial term having dropped out.

12μ(Lμr)2+12ϰ(r−l0)2=E\frac{1}{2}\mu\left(\frac{L}{\mu r}\right)^2+\frac{1}{2}\varkappa(r-l_0)^2 = E

Substituting (8) for LL and EE and dividing through by μ\mu,

l02v022r2+ϰ(r−l0)22μ=v022.\frac{l_0^2v_0^2}{2r^2}+\frac{\varkappa(r-l_0)^2}{2\mu} = \frac{v_0^2}{2}.
Step 4Expand for a small elongation

Write r=l0(1+ε)r=l_0(1+\varepsilon) with ε≪1\varepsilon\ll1 at the point of maximum stretch, and expand the radial factor to first order in ε\varepsilon.

l02r2=(1+ε)−2≈1−2ε\frac{l_0^2}{r^2}=(1+\varepsilon)^{-2}\approx 1-2\varepsilon

Substituting into the previous line and multiplying through by 2/v022/v_0^2,

(1−2ε)+ϰl02ε2μv02=1−2ε+ϰl02μv02ε2=0ε(ϰl02μv02ε−2)=0.\begin{align*}(1-2\varepsilon) + \frac{\varkappa l_0^2\varepsilon^2}{\mu v_0^2} &= 1 \\ -2\varepsilon+\frac{\varkappa l_0^2}{\mu v_0^2}\varepsilon^2 &= 0 \\ \varepsilon\left(\frac{\varkappa l_0^2}{\mu v_0^2}\varepsilon-2\right) &= 0.\end{align*}

The root ε=0\varepsilon=0 is the starting configuration; the maximum elongation is the other root, where the bracket vanishes.

εmax⁡ϰl02μv02=2\varepsilon_{\max}\frac{\varkappa l_0^2}{\mu v_0^2} = 2
εmax⁡=2μv02ϰl02=mv02ϰl02\varepsilon_{\max} = \dfrac{2\mu v_0^2}{\varkappa l_0^2} = \dfrac{mv_0^2}{\varkappa l_0^2}
(9)

using the reduced mass (5).

Check

εmax⁡\varepsilon_{\max} is dimensionless, as a fractional elongation must be: [m][v02]/([ϰ][l02])=kg⋅m2/s2[m][v_0^2]/([\varkappa][l_0^2]) = \text{kg}\cdot\text{m}^2/\text{s}^2 over (N/m)⋅m2=kg m2/s2(\text{N/m})\cdot\text{m}^2=\text{kg}\,\text{m}^2/\text{s}^2, a ratio of two energies. And εmax⁡→0\varepsilon_{\max}\to0 as ϰ→∞\varkappa\to\infty, a spring too stiff to stretch appreciably, consistent with the small-elongation expansion used to get it.

Remark

εmax⁡\varepsilon_{\max} is exactly twice the ratio of the relative motion’s kinetic energy μv02/2\mu v_0^2/2 to the characteristic spring energy scale ϰl02/2\varkappa l_0^2/2: the small parameter that makes the linearization in Step 4 self-consistent is precisely the assumption the problem hands the solver, εmax⁡≪1\varepsilon_{\max}\ll1, which after the fact says the launch speed v0v_0 must be much smaller than l0ϰ/ml_0\sqrt{\varkappa/m}, a natural speed scale set by the spring’s own stiffness and the disc mass.