Staging events — separation dynamics, thrust tail-off: Visual walkthrough

Step 1 — Two stages, one shared velocity

WHY start here. Every derivation needs a "before" picture. If we don't fix what things looked like before the push, we can't measure the change the push creates. Change always means "after minus before."

PICTURE. Two blocks glued together, one arrow $v$ pointing forward (the direction of flight). We name the masses now so we never sneak them in later:

$m_{1}$ = mass of the spent lower stage (heavy, empty tanks — kilograms, kg).
$m_{2}$ = mass of the upper stage (the payload-carrier we want to keep going).

Step 2 — The internal shove: a spring between the stages

WHY a spring model. We don't care about the messy chemistry of the bolts — only about how much push they deliver. Physics has one clean number for "total push delivered over the whole shove": the impulse, written $J_{se p}$ (units newton-seconds, N·s). This comes straight from the Impulse-Momentum Theorem.

PICTURE. The spring pushes both blocks. Newton's third law says the two pushes are equal in size, opposite in direction — the spring can't push one side harder than the other.

The amber arrows are the two pushes: same length (equal magnitude $J_{se p}$ ), opposite directions.

Step 3 — What one push does to one stage (impulse → velocity change)

WHY this tool and not another. We want the change in velocity caused by a push. The impulse-momentum theorem is exactly the rule that connects those two:

$J = m Δ v .$

Read it term by term:

$J$ = the impulse (total push) delivered to this body,
$m$ = the body's mass (its resistance to speed change),
$Δ v$ = the resulting change in the body's velocity (its "after" speed minus its "before" speed).

Rearranged, $Δ v = J / m$ : same push, bigger mass ⇒ smaller speed change. That is Step 1's "mass = stubbornness" idea made into algebra.

PICTURE. Apply it to each stage. The upper stage gets the forward push $+ J_{se p}$ ; the lower stage gets the backward push $- J_{se p}$ (the minus sign just means "opposite direction"):

$m_{2} Δ v_{2} = + J_{se p}, m_{1} Δ v_{1} = - J_{se p} .$

$Δ v_{2}$ = how much the upper stage speeds up (it will be positive — forward).
$Δ v_{1}$ = how much the lower stage changes (negative — it falls behind).

Solving each for the velocity change:

$Δ v_{2} = + \frac{J _{se p}}{m _{2}}, Δ v_{1} = - \frac{J _{se p}}{m _{1}} .$

Step 4 — The quantity that actually matters: relative velocity

WHY a difference. A collision happens (or is avoided) based only on how the two bodies move relative to each other. Two cars both doing 100 km/h side by side never crash; it's the difference that decides. So we compute:

$v_{r e l} = Δ v_{2} - Δ v_{1} .$

$Δ v_{2}$ = upper stage's forward gain,
$Δ v_{1}$ = lower stage's change (negative),
subtracting a negative adds — the gap opens faster than either stage moves alone.

PICTURE. In the frame that moves with the original shared velocity $v$ , the two stages fly apart symmetrically. The arrow between them, $v_{r e l}$ , is the sum of both individual arrows' lengths.

Substitute Step 3's results:

$v_{r e l} = \frac{J _{se p}}{m _{2}} - (- \frac{J _{se p}}{m _{1}}) = \frac{J _{se p}}{m _{2}} + \frac{J _{se p}}{m _{1}} .$

Factor out the common $J_{se p}$ :

$v_{r e l} = J_{se p} (\frac{1}{m _{2}} + \frac{1}{m _{1}}) .$

The bracket $(\frac{1}{m _{2}} + \frac{1}{m _{1}})$ = "combined easiness of separating" — both stages' willingness to move, added.

Step 5 — Naming the bracket: the reduced mass $μ$

WHY invent $μ$ . It lets us write the messy two-body separation as if it were one single particle being kicked. Compare:

One particle: $v = J / m$ .
Two stages: $v_{r e l} = J_{se p} / μ$ .

Identical shape — the two-body relative motion is a one-body problem in disguise, with mass $μ$ .

Deriving $μ$ . We want $\frac{1}{μ} = \frac{1}{m _{1}} + \frac{1}{m _{2}}$ . Add the fractions over a common denominator:

$\frac{1}{μ} = \frac{m _{2} + m _{1}}{m _{1} m _{2}} ⟹ μ = \frac{m _{1} m _{2}}{m _{1} + m _{2}} .$

Numerator $m_{1} m_{2}$ = the two masses multiplied,
Denominator $m_{1} + m_{2}$ = their total mass.

PICTURE. A visual of how $μ$ always sits below the smaller of the two masses — it is dominated by the lighter stage.

Putting it together gives the parent note's boxed result:

$v_{r e l} = \frac{J _{se p}}{μ}, μ = \frac{m _{1} m _{2}}{m _{1} + m _{2}}$

Step 6 — Edge cases: pushing the formula to its limits

PICTURE. Three limiting scenarios side by side.

Case A — equal masses ( $m_{1} = m_{2} = m$ ). $μ = \frac{m \cdot m}{m + m} = \frac{m ^{2}}{2 m} = \frac{m}{2} .$ So $v_{r e l} = 2 J_{se p} / m$ : each stage takes half the "effective mass," both move equally, the gap opens fast. Symmetric skaters, equal push-off. ✓

Case B — one stage enormously heavier ( $m_{1} \to \infty$ , wall-like booster). $μ = \frac{m _{1} m _{2}}{m _{1} + m _{2}} = \frac{m _{2}}{m _{2} / m _{1} + 1} m_{1} \to \infty m_{2} .$ The reduced mass collapses to just $m_{2}$ , so $v_{r e l} = J_{se p} / m_{2}$ — only the light upper stage moves, the giant booster stays put. Like pushing off a wall: you fly, the wall doesn't. ✓

Case C — zero push ( $J_{se p} = 0$ ). $v_{r e l} = \frac{0}{μ} = 0.$ No shove, no separation — the stages keep coasting glued together at $v$ . If the springs fail, nothing moves apart. This is the failure mode that makes staging risky, and the maths shows it honestly. ✓

The one-picture summary

This single diagram compresses all six steps: glued stages at $v$ (Step 1) → spring delivers equal-opposite impulse $J_{se p}$ (Steps 2–3) → each stage's $Δ v = \pm J_{se p} / m$ (Step 3) → their difference is the gap-opening speed $v_{r e l}$ (Step 4) → written compactly with reduced mass $μ$ (Step 5).

Recall Feynman retelling — say it in plain words

Imagine two ice-skaters holding a compressed spring between them. Before release they glide together at one speed. Let go: the spring shoves them apart with exactly equal, opposite pushes — it can't cheat one side. Now, a light skater flies off fast and a heavy one barely moves, because the same push changes a small mass more. What a mission planner cares about is not either skater's own speed, but how fast the gap between them grows — that's the difference of their two speed-changes, and subtracting the backward one adds to it. When you add up "how easily each moves" ( $1/ m_{1} + 1/ m_{2}$ ) and flip it over, you get one tidy number, the reduced mass $μ = \frac{m _{1} m _{2}}{m _{1} + m _{2}}$ , which is always smaller than either skater. So the whole two-skater problem becomes a one-body sentence: gap-speed $=$ push $\div$ reduced mass, $v_{r e l} = J_{se p} / μ$ . Push zero, no gap; make one skater a wall, only the other flies. That is separation dynamics.

Recall Quick self-test

Why divide by $μ$ and not $m_{1} + m_{2}$ ? ::: Because the two bodies push on each other (internal force); the relative motion of a two-body system behaves like a single particle of mass $μ$ , so $v_{r e l} = J_{se p} / μ$ . Using $m_{1} + m_{2}$ makes $v_{r e l}$ wrongly small → collision risk. What does $J_{se p} = 0$ predict? ::: $v_{r e l} = 0$ : stages stay together — the failed-separation case. As $m_{1} \to \infty$ , what does $μ$ become? ::: $μ \to m_{2}$ ; only the upper stage moves (push-off-a-wall limit).

Related: Conservation of Momentum · Impulse-Momentum Theorem · Reduced Mass and Two-Body Problem · Tsiolkovsky Rocket Equation · Multistage Rocket Optimization · Gravity Losses and Ascent Trajectory