Thrust has two parts: the momentum you throw per second, plus a pressure correction for gas not yet
fully expanded. The atmosphere reduces thrust; vacuum thrust is highest.
Effective exhaust velocity $c$ (not raw gas speed) is Chapter 3's $v_e$: it folds the pressure term
into one equivalent velocity, $F = \dot m c$.
Specific impulse is a velocity in seconds: $I_{sp} = c/g_0$; it's impulse per unit propellant
weight, which is why the units are seconds.
Higher $c$ always wins on propellant (it multiplies delta-v and shrinks the mass ratio) — but is
bounded by chemistry (~4.5 km/s) and costs thrust at fixed power.
Two independent tests: enough delta-v (rocket equation) and enough thrust-to-weight (liftoff). A
rocket must pass both; ion engines pass the first and fail the second.
Decision aid — "which relation do I use?"
You know…
You want…
Use
$\dot m$, $v_{\text{ex}}$, $p_e$, $p_a$, $A_e$
thrust
$F = \dot m v_{\text{ex}} + (p_e - p_a)A_e$
$F$, $\dot m$
effective exhaust velocity
$c = F/\dot m$
$I_{sp}$
$c$
$c = I_{sp} g_0$
$F$, $\dot m$
specific impulse
$I_{sp} = F/(\dot m g_0)$
$F$, vehicle mass, local $g$
can it lift off?
$T/W = F/(mg)$; need $> 1$
$c$, $m_p$
total impulse
$I_t = c\, m_p$
$m_p$, $\dot m$ (or $I_t$, $F$)
burn time
$t_b = m_p/\dot m = I_t/F$
The thrust–efficiency split (why the engine zoo has two halves)
Chemical (e.g. Merlin)
Electric (e.g. ion)
Effective exhaust velocity $c$
~2.8–4.5 km/s
~15–40 km/s
Thrust
$10^5$–$10^6$ N (huge)
$10^{-1}$ N (tiny)
Jet power
~GW (from propellant)
~kW (from panels/reactor)
Thrust-to-weight
$> 1$ — can launch
$\ll 1$ — cannot launch
Best for
launch, landing (beat gravity now)
deep-space cruise (save propellant, spend time)
At fixed power, $F = 2P/c$: you get high thrust or high $c$, not both.
Common pitfalls
Pitfall
Reality
"Vacuum → less thrust (no air to push)."
Thrust is higher in vacuum; the atmosphere subtracts $p_a A_e$.
"The pressure term is the rocket pushing on air."
It's the exit gas's own residual pressure; a rocket needs no air at all.
Using vacuum $I_{sp}$ for liftoff, or sea-level for an upper stage.
$I_{sp}$ rises with altitude; use a SL→vac average for stage 1, vacuum value upstairs.
Sizing an upper stage's $T/W$ with the whole vehicle's mass.
It fires after staging — use only stage + payload; and $T/W < 1$ is fine in space.
"More thrust → more delta-v."
Thrust sets burn time ($t_b = I_t/F$), not delta-v; delta-v is $c\ln(m_0/m_f)$.
Sizing thrust from the delta-v (or propellant from $T/W$).
Propellant ← rocket equation; thrust ← thrust-to-weight. Different requirements.
Numbers worth memorizing
$g_0 = 9.80665\ \text{m/s}^2$ (a defined constant in $I_{sp}$, not local gravity).
Merlin 1D: $I_{sp} \approx 282$ s (SL) / $311$ s (vac); $c \approx 2.77$–$3.05$ km/s; $\dot m \approx
306$ kg/s; $F \approx 845$ kN (SL). One Merlin's jet power $\approx 1.2$ GW.
Chemical $c$ ceiling $\approx 4.5$ km/s; ion $c \approx 30$ km/s at milli-newton thrust.
Mission / astrotools additions this chapter
MDR: annotated each maneuver as thrust-critical ($T/W > 1$ needed) or efficiency-critical
(high $I_{sp}$ needed) — the annotation that drives engine selection in Chapters 17–19.
propulsion.py:thrust(mdot, ve, pe, pa, ae), specific_impulse(F, mdot). Grows in Ch. 17–19.