HOW staging helps — derive it. For a single stage, the Tsiolkovsky equation gives
Δv=velnmfm0
where m0 is wet mass, mf is dry (burnout) mass, and ve=Ispg0 is exhaust speed. The mass ratiom0/mf is limited because mf always contains the structural mass. If you carried all the structure to the end, mf is large → Δv small.
For N stages, total Δvadds because each stage burns starting from a smaller vehicle:
WHAT is the tail-off impulse? Model thrust decay after cutoff command at t=0 as an exponential:
F(t)=F0e−t/τ
Why exponential? Chamber pressure pc drains through the nozzle throat at a rate proportional to how much is left (a leaky-tank / first-order process): p˙c∝−pc, and thrust ∝pc. First-order decay ⇒ exponential.
The extra ("tail-off") impulse delivered after the cutoff command:
Derive the separation from momentum. Before separation the two stages move together at v. A separation spring/charge delivers total impulse Jsep between them. Let stage masses be m1 (spent lower) and m2 (upper). By Newton's third law the impulse on each is equal and opposite:
Clearance condition. If separation must produce gap d within time tclear (before upper-stage ignition plume hits the falling stage), and residual differential drag/gravity is negligible over that short window:
vreltclear≥d: upper stage must gain gap d before its plume/ignition.
Is total momentum conserved in separation?
Yes — separation forces are internal, so m1Δv1+m2Δv2=0.
Why does the sum of logs beat one big log?
ln is concave/saturating; staging resets the mass ratio, avoiding the diminishing return of a single huge m0/mf.
Recall Feynman: explain to a 12-year-old
A big rocket is like a stack of water bottles taped together. Each bottle squirts water backward to push the rocket up. When one bottle is empty, it's just dead weight — dragging you down. So the rocket lets go of the empty bottle. But two problems: (1) The squirting doesn't stop the instant you turn it off — a little water keeps dribbling out (that's tail-off). (2) You can't just drop the empty bottle onto the full one below/above, or they'll bump. So little springs give a gentle push apart first, then the next bottle starts squirting only after they've drifted a safe distance. Lighter rocket = faster rocket. That's the whole trick!
Dekho, multistage rocket ka funda simple hai: jab kisi stage ka fuel khatam ho jaata hai, uska empty tank sirf dead weight ban jaata hai — usko carry karte rehna matlab bekaar mein mass ko accelerate karna. Isliye rocket us khaali stage ko phenk deta hai. Har stage ke baad mass ratio "reset" ho jaata hai, aur total Δv=∑ve,iln(m0,i/mf,i) badh jaata hai. Yehi staging ka magic hai.
Par yahan do delicate cheezein hoti hain. Pehli: thrust tail-off. Jaise hi cutoff command doge, engine turant band nahi hota — chamber aur pipe mein bacha propellant thoda aur jalta hai, thrust exponential se girta hai, F(t)=F0e−t/τ. Iska extra impulse Jtail=F0τ hota hai, aur velocity kick Δv=F0τ/m. Guidance ko yeh hidden impulse pehle se count karna padta hai, warna burnout velocity galat ho jaayegi.
Doosri cheez: separation. Springs ya chote retro-rockets se ek internal push Jsep diya jaata hai jisse dono stages alag drift karein. Yaad rakho — yeh internal force hai, isliye total momentum conserve hota hai (m1Δv1+m2Δv2=0), sirf relative velocity banti hai: vrel=Jsep/μ, jahaan μ=m1m2/(m1+m2) reduced mass hai. Common galti: log total mass (m1+m2) se divide kar dete hain — galat! Reduced mass use karo.
Aur ek zaroori baat: upper stage ko turant ignite mat karo. Pehle wait karo taaki gap ban jaaye — clearance condition vreltclear≥d. Warna neeche wale girte stage se plume ya collision mission kharab kar sakta hai. Yaad rakhne ka mantra: "Tail, Split, Clear" — pehle tail-off, phir push apart, phir clear hone ke baad hi next engine on.