Visual walkthrough — Halo orbits — linearized motion near Lagrange points

3.2.31 · D2 · Physics › Orbital Mechanics & Astrodynamics › Halo orbits — linearized motion near Lagrange points

Prerequisites jinpe hum rely karte hain (agar rusty ho toh skim karo): Simple Harmonic Motion, Circular Restricted Three-Body Problem, Lagrange Points L1–L5.

---

Step 1 — Rotating frame gravity ko ek hill-and-valley mein flatten kar deta hai

KYA HAI. Do bade bodies ko ghumte dekhne ki jagah, hum unke saath saath chalte hain — hum poora camera usi rate $\omega$ par spin karte hain taaki do primaries frozen baith jayein. Normalized units mein $\omega =1$ lo (hum apni ghadi aisi chunte hain ki rotation ka ek radian ek time-unit ho). Is frozen picture mein, ek chhote spacecraft par saare forces ek landscape mein pack ho jaate hain jise effective potential $\Omega$ kehte hain.

KYUN. Ek landscape par ball roll karna sabse simple mental model hai jo hamare paas hai: woh downhill roll karta hai. Agar hum "gravity plus rotation" ko ek single surface mein convert kar sakein, toh "forces kahan balance karte hain?" ka sawaal asaan ho jaata hai "ground kahan flat hai?" — yani, jahan $\nabla \Omega =0$.

PICTURE.

$$\Omega (x,y,z)=\underset{\text{centrifugal bowl (bahar dhakelta hai)}}{\underbrace{\frac{1}{2}(x^2+y^2)}}+\underset{\text{bade body ka pull}}{\underbrace{\frac{1-\mu }{r_1}}}+\underset{\text{chhote body ka pull}}{\underbrace{\frac{\mu }{r_2}}}$$

• $x,y,z$ ::: spacecraft ki position frozen (rotating) frame mein.
• $\mu$ ::: total mass ka woh fraction jo chhote body mein hai (toh $1-\mu$ bade mein hai).
• $r_1,r_2$ ::: spacecraft se bade aur chhote primary tak ki distances.
• $\frac{1}{2}(x^2+y^2)$ ::: bahar jaane par badhta hai — yeh centrifugal tendency hai jo tumhe spin axis se door fling karti hai.

Ek Lagrange point is landscape par ek flat spot hai: $\nabla \Omega =0$. Do bodies ko jodte collinear line par aise teen flat spots hain, $L_1,L_2,L_3$.

---

Step 2 — Zoom in karo: flat spot bowl nahi, saddle hai

KYA HAI. Hum ek collinear point chunte hain (maano $L_2$), origin wahan rakhte hain, aur kisi bhi chhote displacement ko $(\xi ,\eta ,\zeta )$ kehte hain. Hum poochhte hain: us flat spot ke bilkul aas-paas landscape kaisa dikhta hai?

KYUN. Ek flat spot valley bottom ho sakta hai (stable — ball wapas aati hai), hilltop ho sakta hai (unstable — ball bhaag jaati hai), ya ek saddle ho sakta hai (ek mountain pass: ridge ke paar stable, uske saath unstable). Yeh kaun sa hai yeh decide karta hai sab kuch — kya spacecraft rukega ya nahi. Yeh jaanne ke liye hume sirf us point par curvature chahiye — second derivatives.

PICTURE.

Kyunki point $x$-axis par baitha hai, landscape symmetric hai, aur mixed curvatures zero ho jaate hain (${\Omega }_{xy}={\Omega }_{xz}={\Omega }_{yz}=0$). Teen surviving curvatures hain:

$$\boxed{U_{xx}=1+2c(>0),U_{yy}=1-c(<0),U_{zz}=-c(<0)}c\equiv \frac{1-\mu }{r_1^3}+\frac{\mu }{r_2^3}$$

• $c$ ::: ek single positive number jo measure karta hai ki combined gravity yahan kitni sharply fall off karti hai. Collinear points ke liye $c>1$.
• $U_{xx}=1+2c$ ::: bodies ki taraf line ke saath positive curvature → surface upar curve karti hai → potential badhti hai → force tumhe wapas push karta hai... lekin dekho, Step 4 ka Coriolis force ise ridge ke saath instability mein flip kar deta hai.
• $U_{yy}=1-c$ aur $U_{zz}=-c$ ::: dono negative (kyunki $c>1$) → neeche curve karta hai → yeh bare landscape par downhill/repelling directions jaise dikhte hain.

---

Step 3 — Equations of motion, term by term

KYA HAI. Rotating frame mein Newton ka law, flat spot ke aas-paas first order tak expand kiya, $(\xi ,\eta ,\zeta )$ ke liye teen linear equations deta hai.

KYUN. Full CR3BP nonlinear hai aur closed form mein solve nahi hota. Lekin point ke paas displacement tiny hoti hai, toh har force ka leading (linear) term dominate karta hai aur baaki hum throw away kar dete hain. Constant coefficients wale linear equations hum exactly solve kar sakte hain — exponentials aur sines se.

PICTURE.

$$\ddot{\xi }-\underset{\text{Coriolis}}{\underbrace{2\dot{\eta }}}=(1+2c)\xi$$ $$\ddot{\eta }+\underset{\text{Coriolis}}{\underbrace{2\dot{\xi }}}=(1-c)\eta$$ $$\ddot{\zeta }=-c\zeta$$

• $\ddot{\xi },\ddot{\eta },\ddot{\zeta }$ ::: teen axes ke saath accelerations ($\ddot{}$ ka matlab hai "second time-derivative", yani acceleration).
• $-2\dot{\eta }$, $+2\dot{\xi }$ ::: Coriolis force — ek sideways kick jo tumhari velocity ke proportional hai, sirf isliye present hai kyunki frame spin karta hai. Yeh $\xi$ aur $\eta$ ko ek saath couple karta hai.
• Right-hand sides ::: Step 2 se curvatures jo spring/anti-spring constants ki tarah kaam karte hain.

---

Step 4 — Out-of-plane bob: pure simple harmonic motion

KYA HAI. $\ddot{\zeta }=-c\zeta$ ko seedha solve karo.

KYUN. Ise ek mass on a spring se compare karo, $\ddot{x}=-kx$, jiska solution $\sqrt{k}$ par oscillate karta hai. Yahan $k$ ki jagah $c$ hai, toh hum bina kisi nayi machinery ke answer padhh sakte hain — isliye hum SHM ko invoke karte hain general ODE solve karne ki jagah.

PICTURE.

$$\zeta (t)=A_z\sin ({\omega }_{z}t+\psi ),\boxed{{\omega }_z=\sqrt{c}}$$

• $A_z$ ::: spacecraft orbital plane ke upar/neeche kitna bob karta hai.
• ${\omega }_z$ ::: woh kitni tezi se bob karta hai. Kyunki $c>0$, yeh ek real frequency hai → sachchi back-and-forth motion, kabhi runaway nahi.
• $\psi$ ::: ek phase (bob mein kahan se shuru hota hai).

Out-of-plane direction seedha wala hai: yeh disguise mein ek valley hai, aur spacecraft pendulum ki tarah plane ke through jhoolata hai.

---

Step 5 — In-plane pair: $e^{\lambda t}$ guess karo aur ODE ko algebra mein badlo

KYA HAI. $\xi ,\eta$ equations Coriolis se tangled hain. Hum guess karte hain ki dono exponentially grow/oscillate karte hain, $\xi =A{e}^{\lambda t}$, $\eta =B{e}^{\lambda t}$, aur substitute karte hain.

KYUN. Exponential $e^{\lambda t}$ ek maatra function hai jiska derivative khud iska copy hai ($\frac{d}{dt}{e}^{\lambda t}=\lambda e^{\lambda t}$). Woh ek property har derivative ko multiply-by-$\lambda$ mein convert kar deti hai, differential equations ko ordinary algebra mein badal deti hai. Yeh kisi bhi linear constant-coefficient system ke liye standard key hai.

PICTURE.

Substitute karne par ($\dot{\xi }\to \lambda A$, $\ddot{\xi }\to {\lambda }^{2}A$, etc.):

$$\left(\begin{array}{cc}{\lambda }^2-(1+2c)& -2\lambda \\ 2\lambda & {\lambda }^2-(1-c)\end{array}\right)\left(\begin{array}{c}A\\ B\end{array}\right)=\left(\begin{array}{c}0\\ 0\end{array}\right)$$

• Har entry ::: jo bachta hai jab sab kuch ek side par le jaate ho; ${\lambda }^2$ acceleration se, $-2\lambda$ / $+2\lambda$ Coriolis se, constants curvature se.
• $(A,B)$ ::: mode ki shape — ratio $B/A$ batata hai in-plane wobble kitna elongated hai.

Nonzero $(A,B)$ tab hi exist karta hai jab matrix kisi direction ko zero par squash kare, yani jab uska determinant zero ho. $\det =0$ set karne par:

$$({\lambda }^2-(1+2c))({\lambda }^2-(1-c))+4{\lambda }^2=0$$

Expand aur collect karo (maano $\Lambda ={\lambda }^2$):

$$\boxed{{\Lambda }^2+(2-c)\Lambda +(1+2c)(1-c)=0}$$

---

Step 6 — Opposite sign ke do roots: saddle × center structure

KYA HAI. $\Lambda ={\lambda }^2$ mein quadratic solve karo.

KYUN. Har $\Lambda$ ki sign physics decide karti hai: $\Lambda >0$ real $\lambda =\pm \sqrt{\Lambda }$ deta hai (growth/decay), jabki $\Lambda <0$ imaginary $\lambda =\pm i\sqrt{-\Lambda }$ deta hai (oscillation). Hume har collinear point ke liye dono roots check karni chahiye — koi skip nahi.

PICTURE.

$$\Lambda =\frac{-(2-c)\pm \sqrt{(2-c{)}^2-4(1+2c)(1-c)}}{2}$$

Collinear points ke liye $c>1$, toh $(1+2c)(1-c)<0$. Isse do roots ka product negative ho jaata hai — jo ek root positive aur ek negative hone ke liye force karta hai:

$${\Lambda }_1>0\Rightarrow \lambda =\pm \gamma \text{ (real)}{\Lambda }_2<0\Rightarrow \lambda =\pm i{\omega }_{xy}\text{ (imaginary)}$$

• $\gamma =\sqrt{{\Lambda }_1}$ ::: instability rate. Mode $e^{+\gamma t}$ grow karta hai — iske saath displace karo aur tum point se bhaag jaate ho. Yeh saddle ki runaway ridge hai. Exactly isliye $L_1,L_2$ ko station-keeping chahiye, aur yeh invariant manifolds ka seed hai.
• ${\omega }_{xy}=\sqrt{-{\Lambda }_2}$ ::: planar oscillation frequency. Mode $e^{\pm i{\omega }_{xy}t}$ ek sine/cosine hai — ek stable in-plane loop.

---

Step 7 — Eigenvector ellipse fix karta hai: ratio $\kappa$

KYA HAI. Oscillatory root ke liye, $\lambda =i{\omega }_{xy}$ ko wapas matrix mein plug karo aur ratio $B/A=\kappa$ nikalo.

KYUN. Determinant ne bataya ki ek mode exist karta hai; eigenvector uski shape batata hai. Hume $\kappa$ chahiye kyunki in-plane path ek ellipse hai, aur $\kappa$ batata hai ki woh $\xi$ se $\eta$ mein kitna zyada tall hai.

PICTURE.

$$\xi (t)=-A_x\cos ({\omega }_{xy}t+\varphi ),\eta (t)=\kappa A_x\sin ({\omega }_{xy}t+\varphi ),\boxed{\kappa =\frac{{\omega }_{xy}^2+(1+2c)}{2{\omega }_{xy}}}$$

• $A_x$ ::: $x$ ke saath in-plane amplitude.
• $\kappa$ ::: stretch factor. Kyunki typically $\kappa \approx 3$, planar loop ek ellipse hai jo $y$ ke saath elongated hai, circle nahi.
• $\cos$/$\sin$ pairing ::: ek quarter-period out of phase, jo ek ellipse trace karta hai (jaise ek point ke round jaane ke $x,y$).

---

Step 8 — Do frequencies jo match nahi karti: Lissajous, abhi halo nahi

KYA HAI. Ab hamare paas do independent frequencies hain: in-plane ${\omega }_{xy}$ aur out-of-plane ${\omega }_z=\sqrt{c}$. Generally ${\omega }_{xy}\ne {\omega }_z$.

KYUN. Ek closed 3-D loop ke liye wobble aur bob ko sync up karna hoga — same point par same phase par wapas aana hoga. Agar unke periods alag hain, toh path forever drift karta hai aur ek tube fill karta hai: ek Lissajous orbit, halo nahi.

PICTURE.

---

Worked check — Earth–Moon $L_2$ ($c\approx 3.19$)

---

Ek-picture summary

Recall Feynman retelling — plain words mein khud se kaho

Apna camera do bade bodies ke saath spin karo taaki woh rukh jayein. Ab har force ek hill-and-valley landscape hai. Lagrange point ek flat spot hai — lekin valley nahi: yeh ek mountain pass hai, ek saddle. Isse ek taraf roll ho jao (bodies ki taraf line ke saath, sideways Coriolis kick se mix hokar) aur tum loop kar sakte ho — yeh tumhara in-plane wobble hai, ek ellipse jo apni width se karib teen guna tall hai. Plane ke through upar-neeche bob karo aur tumhe ek clean spring oscillation milti hai, kyunki woh direction secretly ek valley hai. Lekin ek nasty runaway direction bhi hai jahan tum pass se slide off karte ho aur kabhi wapas nahi aate — isliye yeh spots ko rukte rehne ke liye constant nudging chahiye. Do nice oscillations ki speeds thodi alag hain, toh unka combined path drift karta hai aur kabhi bilkul close nahi hota — ek tube par Lissajous scribble. Sirf jab tum fine nonlinear corrections add karte ho tab do speeds ek perfect 1:1 match mein lock hoti hain bilkul sahi size par — aur woh point ke aas-paas hovering closed 3-D loop halo hai.

Recall Quick self-test

Ek collinear Lagrange point unstable kyun hota hai? ::: Characteristic quadratic ka ek positive root ${\Lambda }_1$ hota hai, jo real $\lambda =+\gamma$ deta hai; woh mode $e^{\gamma t}$ ki tarah grow karta hai. Out-of-plane equation easy kyun hai? ::: Yeh poori tarah decouple ho jaata hai ($\ddot{\zeta }=-c\zeta$, koi Coriolis nahi), toh yeh pure SHM hai jahan ${\omega }_z=\sqrt{c}$. Linear theory sirf Lissajous kyun deta hai? ::: Generally ${\omega }_{xy}\ne {\omega }_z$, toh loop kabhi close nahi hota; halo ke liye ek nonlinear 1:1 resonance chahiye. $\kappa$ kya batata hai? ::: In-plane ellipse ka $y$-to-$x$ stretch — Earth–Moon $L_2$ ke liye, karib $2.9$.