Classical MechanicsBSc · Notes

1. Why inverse-square attraction is special

For a general central force, the reciprocal-radius equation can remain nonlinear. Inverse-square attraction cancels its explicit u2u^2 denominator, leaving a constant driving term. The orbital geometry then follows from the elementary harmonic differential equation.

Use F=−kr^/r2\mathbf F=-k\hat{\mathbf r}/r^2 with k>0k>0, V=−k/rV=-k/r and V(∞)=0V(\infty)=0. The mass is μ\mu, angular momentum magnitude is ℓ>0\ell>0, and EE is relative mechanical energy. For gravity between two bodies, k=Gm1m2k=Gm_1m_2 and k/μ=G(m1+m2)k/\mu=G(m_1+m_2).

2. Solution of the orbit equation

Binet's equation becomes

u′′+u=μkℓ2.u''+u=\frac{\mu k}{\ell^2}.(20.1)

A constant particular solution is u0=μk/ℓ2u_0=\mu k/\ell^2. The homogeneous equation has solutions cos⁡θ\cos\theta and sin⁡θ\sin\theta, as verified by twice differentiating them. Thus

u=u0+Ccos⁡θ+Dsin⁡θ.u=u_0+C\cos\theta+D\sin\theta.(20.2)

Write C=u0ecos⁡θ0C=u_0e\cos\theta_0 and D=u0esin⁡θ0D=u_0e\sin\theta_0, with e=C2+D2/u0≥0e=\sqrt{C^2+D^2}/u_0\geq0. The addition formula gives u=u0[1+ecos⁡(θ−θ0)]u=u_0[1+e\cos(\theta-\theta_0)]. Choose the polar axis through the closest approach so that θ0=0\theta_0=0 and define the semi-latus rectum

p=1u0=ℓ2μk.p=\frac1{u_0}=\frac{\ell^2}{\mu k}.(20.3)

Here italic pp is an orbital length, while bold p\mathbf p later denotes linear momentum and prp_r denotes radial canonical momentum. The orbit is

r=p1+ecos⁡θ.\boxed{r=\frac{p}{1+e\cos\theta}.}(20.4)

For e>0e>0, θ=0\theta=0 maximizes the denominator and gives periapsis rp=p/(1+e)r_p=p/(1+e). At θ=π/2\theta=\pi/2, r=pr=p, explaining the geometrical name semi-latus rectum. For e=0e=0 the circle has no distinguished periapsis direction.

3. Why the equation describes a conic

With x=rcos⁡θx=r\cos\theta, y=rsin⁡θy=r\sin\theta, multiply the orbit equation by its denominator to get r+ex=pr+ex=p. Thus r=p−exr=p-ex, which must remain nonnegative on the physical branch. Squaring gives

x2+y2=p2−2pex+e2x2,x^2+y^2=p^2-2pex+e^2x^2,(20.5)

or

(1−e2)x2+y2+2pex−p2=0.(1-e^2)x^2+y^2+2pex-p^2=0.(20.6)

For 0≤e<10\leq e<1, complete the square:

(x+pe1−e2)2p2(1−e2)2+y2p21−e2=1.\frac{\left(x+\frac{pe}{1-e^2}\right)^2}{\frac{p^2}{(1-e^2)^2}} +\frac{y^2}{\frac{p^2}{1-e^2}}=1.(20.7)

This is an ellipse with semimajor axis a=p/(1−e2)a=p/(1-e^2), semiminor axis b=p/1−e2=a1−e2b=p/\sqrt{1-e^2}=a\sqrt{1-e^2}, and center at x=−aex=-ae. One focus is the force center at the origin, a distance aeae from the ellipse center.

For e=1e=1, the equation becomes y2=p2−2px=−2p(x−p/2)y^2=p^2-2px=-2p(x-p/2): a parabola with vertex at periapsis x=p/2x=p/2. For e>1e>1, put d=e2−1>0d=e^2-1>0 and complete the square to get

(x−pe/d)2p2/d2−y2p2/d=1.\frac{(x-pe/d)^2}{p^2/d^2}-\frac{y^2}{p^2/d}=1.(20.8)

This is a hyperbola; the physical attractive-scattering orbit uses the branch satisfying the original unsquared condition r=p−ex≥0r=p-ex\geq0. Squaring alone introduces the other geometric branch without making it part of the same physical trajectory.

Ellipse: e = 0.6FPParabola: e = 1FPHyperbola: e = 1.5FPF = force center (focus); P = periapsis. Each panel has the same orbital p.
Figure 20.1 · Conic paths with the force center at a focus. Open branches are cropped by the drawing window; they extend to infinity.

4. Energy and eccentricity

From the orbital energy relation,

E=ℓ22μ(u′2+u2)−ku.E=\frac{\ell^2}{2\mu}(u'^2+u^2)-ku.(20.9)

Since u=(1+ecos⁡θ)/pu=(1+e\cos\theta)/p and u′=−esin⁡θ/pu'=-e\sin\theta/p,

u′2+u2=e2sin⁡2θ+1+2ecos⁡θ+e2cos⁡2θp2=1+2ecos⁡θ+e2p2.u'^2+u^2=\frac{e^2\sin^2\theta+1+2e\cos\theta+e^2\cos^2\theta}{p^2} =\frac{1+2e\cos\theta+e^2}{p^2}.(20.10)

Use ℓ2/(μp)=k\ell^2/(\mu p)=k. Then

E=k2p(1+2ecos⁡θ+e2)−kp(1+ecos⁡θ)=k2p(e2−1).E=\frac{k}{2p}(1+2e\cos\theta+e^2)-\frac{k}{p}(1+e\cos\theta) =\frac{k}{2p}(e^2-1).(20.11)

Therefore

e2=1+2Eℓ2μk2.\boxed{e^2=1+\frac{2E\ell^2}{\mu k^2}.}(20.12)

The angular terms cancel because EE is conserved. At fixed ℓ\ell, the minimum allowed energy is Ec=−μk2/(2ℓ2)E_c=-\mu k^2/(2\ell^2), where e=0e=0. Energies below this have no real orbit for that angular momentum.

Eccentricity Energy Orbit
e=0e=0 E=Ec<0E=E_c<0 Circle
0<e<10<e<1 Ec<E<0E_c<E<0 Ellipse
e=1e=1 E=0E=0 Parabola
e>1e>1 E>0E>0 Hyperbola

This table assumes k>0k>0, ℓ>0\ell>0, and zero potential at infinity. Radial collision trajectories with ℓ=0\ell=0 are degenerate and are not classified by simply setting e=1e=1 in the regular conic formula.

For an ellipse, p=a(1−e2)p=a(1-e^2) gives

E=−k2a,ℓ2=μka(1−e2),ra=a(1+e),rp=a(1−e).\boxed{E=-\frac{k}{2a},\qquad \ell^2=\mu ka(1-e^2),\qquad r_a=a(1+e),\quad r_p=a(1-e).}(20.13)

At fixed aa, more eccentric ellipses have less angular momentum. At fixed ℓ\ell, changing energy changes eccentricity and size together.

5. Speed and escape

From E=μv2/2−k/rE=\mu v^2/2-k/r and E=−k/(2a)E=-k/(2a),

v2=kμ(2r−1a)\boxed{v^2=\frac{k}{\mu}\left(\frac2r-\frac1a\right)}(20.14)

for an ellipse. This is the vis-viva relation. For a circular orbit a=ra=r, it gives vc=k/(μr)v_c=\sqrt{k/(\mu r)}. Setting escape energy to zero gives vesc=2k/(μr)=2vcv_{\rm esc}=\sqrt{2k/(\mu r)}=\sqrt2v_c. This is a speed threshold in the ideal isolated inverse-square model; launch direction determines angular momentum and whether a trajectory encounters a physical surface first.

For a hyperbola, the positive geometric semimajor length aha_h obeys E=k/(2ah)E=k/(2a_h), giving v2=(k/μ)(2/r+1/ah)v^2=(k/\mu)(2/r+1/a_h). Some books use a negative signed aa to retain the same vis-viva formula. We use positive lengths and state the sign explicitly.

For a hyperbolic orbit the allowed angular range is ∣θ∣<θ∞|\theta|<\theta_\infty, where cos⁡θ∞=−1/e\cos\theta_\infty=-1/e. At its limits the denominator vanishes and r→∞r\to\infty. Continuing beyond that range gives an unphysical negative polar radius.

6. Applications

Example 1 · Geometry from the turning radii

An ellipse has rp=2 length unitsr_p=2\,\mathrm{length\ units} and ra=8 length unitsr_a=8\,\mathrm{length\ units}. Adding and subtracting a(1−e)a(1-e) and a(1+e)a(1+e) gives a=5a=5 and e=(8−2)/(8+2)=0.6e=(8-2)/(8+2)=0.6. Then p=a(1−e2)=3.2p=a(1-e^2)=3.2 and b=a1−e2=4b=a\sqrt{1-e^2}=4. The force center is ae=3ae=3 units from the ellipse center.

Example 2 · Launch faster than circular speed

At radius r0r_0, a particle is launched tangentially at v=3/2 vcv=\sqrt{3/2}\,v_c, where vc2=k/(μr0)v_c^2=k/(\mu r_0). Its energy is E=μv2/2−k/r0=−k/(4r0)E=\mu v^2/2-k/r_0=-k/(4r_0) and its angular momentum satisfies ℓ2=μ2r02v2=(3/2)μkr0\ell^2=\mu^2r_0^2v^2=(3/2)\mu kr_0.

Thus a=2r0a=2r_0 and e2=1−3/4=1/4e^2=1-3/4=1/4, so e=1/2e=1/2. Since the radial velocity is zero and the speed exceeds circular speed, the launch point is periapsis: rp=a(1−e)=r0r_p=a(1-e)=r_0. Apoapsis is 3r03r_0. The radius increases initially because r¨=v2/r0−k/(μr02)>0\ddot r=v^2/r_0-k/(\mu r_0^2)>0.

Example 3 · Escape in dimensionless units

Let μ=k=1\mu=k=1, launch radius r0=2r_0=2, and tangential speed v0=1v_0=1. Then E=v02/2−1/r0=0E=v_0^2/2-1/r_0=0, ℓ=r0v0=2\ell=r_0v_0=2, and p=ℓ2=4p=\ell^2=4. The parabola is r=4/(1+cos⁡θ)r=4/(1+\cos\theta), with periapsis rp=2r_p=2. The particle approaches zero speed only as r→∞r\to\infty, not at a finite apoapsis.

Example 4 · A hyperbolic encounter

Take μ=k=1\mu=k=1, asymptotic speed v∞=1v_\infty=1, and impact parameter bimp=2b_{\rm imp}=2. Far away, E=1/2E=1/2 and ℓ=μbimpv∞=2\ell=\mu b_{\rm imp}v_\infty=2. Thus e=1+2(1/2)(4)=5e=\sqrt{1+2(1/2)(4)}=\sqrt5, p=4p=4, and rp=4/(1+5)=5−1r_p=4/(1+\sqrt5)=\sqrt5-1.

The positive energy means the particle returns to infinity after the encounter. The periapsis is smaller than the impact parameter because attraction bends the incoming trajectory inward. The Runge–Lenz chapter will recover the deflection angle from the same constants.

7. Geometry and dynamics together

The conic equation supplies shape and orientation; EE and ℓ\ell determine size and eccentricity; the area law supplies the time dependence. A complete orbital description needs all three. An ellipse drawn with its center at the gravitating body would misrepresent the inverse-square force except in the circular case.

Exercises

Hints and solutions are collected separately.

  1. 20.1 Solve the inverse-square Binet equation and define its two geometric constants.

  2. 20.2 Derive the energy–eccentricity relation.

  3. 20.3 An ellipse has rp=3r_p=3 and ra=9r_a=9 in length units. Find a,e,pa,e,p.

  4. 20.4 For μ=k=1\mu=k=1 and ℓ=2\ell=2, what is the lowest allowed energy?

  5. 20.5 For μ=k=ℓ=1\mu=k=\ell=1, classify the orbit at E=−1/4E=-1/4.

  6. 20.6 Derive the escape speed at radius rr for inverse-square attraction.

  7. 20.7 At fixed semimajor axis, how does angular momentum depend on eccentricity?

  8. 20.8 A particle is launched tangentially at half the circular speed at r0r_0. Find a,ea,e and decide which apsis is the launch point.

  9. 20.9 For a hyperbola with e=2e=2, find its limiting true anomaly.

  10. 20.10 Why does E=0E=0 not mean a parabolic particle is at rest?

  11. 20.11 Use Cartesian coordinates to identify the ellipse center for r=p/(1+ecos⁡θ)r=p/(1+e\cos\theta) with e<1e<1.

  12. 20.12 Why must ℓ=0\ell=0 be treated separately in the energy–eccentricity formula?

Reading progress is saved on this device.

Search the notes

Search all four units and the course references.