This is a question bank for the parent topic. Each line below hides an answer — cover the right side, commit to a guess, then reveal. Every answer gives you the reasoning, not just a verdict.
Before we trap you, we build every symbol from scratch and draw every picture the answers lean on. Skip nothing here — the reveals below assume these.
Where the throat and exit sit, and what ϵ means geometrically:
The three states a nozzle can be in — this single figure is the source of half the reveals below. Watch the exhaust plume shape and the wall-pressure trace:
The thrust of a rocket is
F=m˙ve+(pe−pa)Ae.
The first piece is obvious momentum: throw m˙ kg/s out the back at speed ve. But where does the second piece come from? Look at the control volume — the box drawn around the engine — and add up the pressure pushing on every face:
We keep saying "ϵ sets Me." Here is the actual law, from conservation of mass in isentropic flow:
AtAe=Me1[γ+12(1+2γ−1Me2)]2(γ−1)γ+1.
Read it as a machine: feed in geometry ϵ=Ae/At, and it hands back the supersonic Me. The reason it is non-linear — why quadrupling ϵ barely nudges Me — is the picture below: for a fast supersonic flow the area balloons far faster than the Mach number climbs.
Once Me (hence pe) is known, the ideal exhaust speed follows from energy conservation: hot pressurised gas in the chamber trades its thermal/pressure energy for kinetic energy as it expands.
ve=γ−12γRT0[1−(p0pe)(γ−1)/γ]
Chamber energy is set by T0 (heat) and R,γ (which gas); the bracket is the fraction cashed in, governed by how far pressure fell from p0 down to pe. As pe→0 the bracket →1 and vesaturates — the root of every "diminishing returns" answer below.
False. Over-expanded means the gas expanded too far and its exit pressure fell below ambient, so pe<pa and the outside air pushes back on the flow.
A perfectly expanded nozzle at sea level stays perfectly expanded all the way to space.
False. Ambient pa falls from ∼101 kPa to ∼0 during ascent while a fixed nozzle keeps the same pe; matching one altitude guarantees mismatch at all others.
An extendable nozzle lets an upper-stage engine also produce good thrust at sea level.
False. Both stowed and deployed states have very high ϵ (≳100); such a nozzle would be badly over-expanded and would separate at sea level. It only ever fires in near-vacuum.
A larger expansion ratio always increases exhaust velocity by a useful amount.
False.ve depends on p0,pe,T0,γ and saturates as pe→0: the bracket [1−(pe/p0)(γ−1)/γ]→1, so extra ϵ buys smaller and smaller gains.
The stowed (retracted) state of an extendable nozzle exists to give sea-level performance.
False. It exists to save length under the fairing/interstage during launch. The engine still ignites in near-vacuum in both states.
An aerospike lets the exhaust boundary adjust itself to ambient pressure across the whole ascent.
True. The exhaust plume on the open (spike) side is bounded by outside air, so it self-adjusts its shape as pa falls — giving continuous compensation a fixed bell cannot.
In vacuum (pa=0) the pressure-thrust term (pe−pa)Ae is always negative.
False. With pa=0 it equals peAe≥0, so it is non-negative and adds to thrust; it can never be negative in vacuum.
Doubling the expansion ratio doubles the exit-Mach number.
False. The area–Mach relation is strongly non-linear (see the ϵ–Me curve); e.g. ϵ jumping from 21 to 84 (×4) only moves Me from ≈3.8 to ≈4.75.
Specific impulse Isp and exhaust velocity carry the same information (up to a constant).
True.Isp=ve/g0, so Isp is just ve divided by the fixed constant g0; a higher ϵ that raises ve raises Isp identically.
"Make ϵ=100 so the nozzle is perfect in vacuum, and just accept weak sea-level thrust."
The error is ignoring flow separation: at sea level such over-expansion drives shocks inside the bell (right panel of the three-state figure), exhaust detaches from the wall asymmetrically, and side loads can tear the nozzle apart — not merely "weak" thrust.
"Since ve∝ϵ, quadrupling ϵ doubles the exhaust velocity."
There is no clean ϵ law. The correct ve=γ−12γRT0[1−(pe/p0)(γ−1)/γ] saturates; ϵ only enters indirectly by setting pe.
"At sea level an over-expanded nozzle loses thrust only because the (pe−pa)Ae term is negative."
Incomplete — beyond the negative pressure term, the internal shocks cause flow separation and unsteady vibration/structural loads, which are separate loss and failure mechanisms.
"The extension adds structural mass proportional to how much thrust it recovers, so it barely helps."
The extension sits at a large radius where pressure is low, so it can be thin-walled and light; the mass penalty is small relative to the Isp gain, which is why upper stages use it.
"An aerospike is efficient because it has no throat, so no shock losses ever form."
An aerospike absolutely has a throat (the gas must go sonic somewhere). Its advantage is the free boundary on the spike side (see the bell-vs-spike figure), not the absence of a throat.
"Under-expanded means the nozzle is too big for the altitude."
Reversed. Under-expanded (pe>pa) means the nozzle is too small / ϵ too low — it stopped expanding while pressure energy remained.
Why does atmospheric pressure "crush the exhaust inward" in an over-expanded nozzle?
Because pe<pa, the outside air pressure exceeds the flow's, pushing the boundary in and forming oblique shocks that can separate the flow from the wall.
Why can't a single fixed bell nozzle be optimal from launch to orbit?
Because the matching condition pe=pa needs pe to track a falling pa, but a fixed geometry locks pe at one value — it can be right at only one altitude.
Why do we want pe=pa rather than the largest possible pe?
When pe=pa all available pressure energy has been converted into exhaust velocity; any leftover pe>pa is momentum you didn't extract, and pe<pa costs you via back-pressure and shocks.
Why does raising ϵ shrink the pressure-thrust term while growing the momentum term?
More expansion lowers pe (small (pe−pa)Ae) but converts that pressure into higher ve, so momentum thrust m˙ve rises as the pressure term fades.
Why is the aerospike called a "compensating" nozzle while the extendable one is not (across ascent)?
The aerospike's plume boundary responds continuously to pa during the whole climb (bell-vs-spike figure); the extendable nozzle only makes one discrete change and always operates in near-vacuum.
Why does the exit-Mach number rise so slowly as ϵ grows large?
Area grows roughly with the geometric spread of a highly supersonic, nearly parallel flow, so enormous extra area is needed to squeeze out each additional bit of Mach number — hence the flattening ϵ–Me curve.
What happens to the thrust equation term (pe−pa)Ae exactly at perfect expansion?
It becomes zero, so thrust is purely m˙ve — the "cleanest" case where all pressure energy has already gone into velocity.
At pa=0 (deep vacuum), what limits how much thrust a longer nozzle can add?
The saturation of ve as pe→0: the bracket [1−(pe/p0)(γ−1)/γ] approaches 1 (flat right end of the ve curve), so momentum thrust plateaus and only the small peAe term keeps shrinking.
If pe could reach exactly 0, would that be the ideal nozzle?
Only in the limit — it would need infinite exit area (infinite ϵ) and infinite length/mass, so real designs stop where added mass outweighs the vanishing ve gain.
What is the danger boundary when a high-ϵ upper-stage nozzle is (mistakenly) fired at sea level?
Severe over-expansion triggers flow separation and asymmetric side loads inside the bell, risking structural failure — which is precisely why these engines only ignite after the atmosphere has thinned.
What happens to an aerospike's advantage as the vehicle reaches deep vacuum?
The compensation benefit fades because with pa→0 every nozzle is effectively under-expanded the same way; the aerospike's edge is largest in the varying-pressure lower atmosphere.
For an ideally under-expanded nozzle in vacuum, is the pressure term helping or hurting?
Helping — with pa=0 and pe>0 the term peAe is positive, though it represents velocity you could have gained instead had the flow expanded further.
Recall Quick self-test on the built-up symbols
What does ϵ=Ae/At compare? ::: The wide exit area against the narrow throat area — how much the funnel spreads.
Why does ve stop growing as pe→0? ::: The energy-cash-in bracket [1−(pe/p0)(γ−1)/γ] tops out at 1.
Where does the (pe−pa)Ae thrust term physically come from? ::: The only nozzle face where ambient pressure fails to cancel — the exit disc.
