[Intermediate] · Orbits, why getting to space is mostly about sideways speed, and what’s actually flying.
Orbit is not altitude
The most common misconception in the subject: space is not far, it’s fast.
The Kármán line is 100 km up — a drive you’d make without thinking. Getting to that altitude is comparatively easy. Staying there is the hard part.
An orbit is a trajectory where you fall toward Earth and keep missing, because your sideways velocity carries you past the horizon as fast as you drop. Newton’s cannonball: fire hard enough and the ground curves away as fast as the ball falls.
The numbers make the point:
- Reaching 100 km vertically: ~1.4 km/s
- Staying in low Earth orbit: ~7.8 km/s
Roughly 90% of the energy goes into horizontal speed, not height. This is why rockets pitch over almost immediately after launch, and why suborbital flights (Blue Origin’s New Shepard) are a fundamentally easier problem than orbital ones.
Delta-v is the currency
Δv — the total velocity change a mission needs — is the budget everything is planned against. Rocket equation:
The logarithm is brutal. Doubling your Δv doesn’t double the fuel — it squares the mass ratio. This is why rockets are almost entirely propellant, why staging exists (drop the empty tanks so you stop accelerating them), and why every kilogram of payload is fought over.
Rough Δv from Earth’s surface:
| Destination | Δv |
|---|---|
| Low Earth orbit | ~9.4 km/s (incl. losses) |
| Geostationary transfer | +2.4 |
| The Moon (surface) | +~5.9 |
| Mars (surface) | ~+4.5 with aerobraking |
Once you’re in LEO you are, energetically, halfway to anywhere.
Orbital mechanics that feel wrong
To speed up, you slow down. Firing your engine forward (retrograde) drops you to a lower orbit, where orbital velocity is higher. To catch something ahead of you, you go lower and overtake, then raise back up. Intuition from driving is actively misleading here.
Hohmann transfer — the fuel-efficient way between two circular orbits: one burn to enter an elliptical path, one at the far end to circularise. Slow but cheap.
Launch windows exist because the geometry has to line up. Mars windows come every ~26 months, which is why missions cluster and why a slip costs two years.
Gravity assists — steal a little orbital energy from a planet by flying past it. Voyager, Cassini and New Horizons all depended on them. The planet loses an utterly negligible amount of its orbital energy; the spacecraft gains a great deal.
Lagrange points — five positions where the gravity of two bodies and the orbital motion balance. JWST sits at Earth–Sun L2, 1.5 million km out, so Earth and Sun stay in the same direction and one sunshield blocks both.
What’s actually flying
Crewed: the ISS (continuously occupied since 2000, retirement planned ~2030), China’s Tiangong, and commercial vehicles — SpaceX Crew Dragon carries astronauts routinely.
Launch: the dominant change of the last decade is reusability. Falcon 9 boosters land and refly, which cut cost per kilogram substantially and changed what missions are affordable. Starship aims at full reuse at much larger scale.
Robotic exploration is where most science happens: Perseverance and Curiosity on Mars; JWST and Hubble; Parker Solar Probe flying through the Sun’s corona; Europa Clipper en route to Jupiter’s ocean moon; Voyager 1 and 2 in interstellar space after 47+ years.
The honest framing: robots do the science, humans do the difficult, expensive, inspiring part. A rover costs a fraction of a crewed mission and needs no return trip. The case for humans is partly scientific (a geologist on Mars would outpace a rover enormously) and substantially political and cultural — and that’s worth saying plainly rather than pretending otherwise.
Why it’s hard
Vacuum — no convection, so thermal management is entirely radiative. Spacecraft overheat as readily as they freeze.
Radiation — no atmosphere or magnetosphere beyond LEO. Damages electronics (hence radiation-hardened, deliberately old, slow chips) and is a major unsolved obstacle for long crewed missions.
Debris — tens of thousands of tracked objects, millions untracked, at orbital speeds where a paint fleck is dangerous. Kessler syndrome — a cascade of collisions making some orbits unusable — is a genuine concern, not a hypothetical.
Microgravity on humans — bone density loss, muscle atrophy, fluid shifts, vision changes, immune effects.
Distance and latency — Mars is 4–24 light-minutes away. No real-time control, so spacecraft must be substantially autonomous.
The engineering discipline
Spaceflight is where systems engineering was invented, and the reasons are structural: you cannot iterate, you cannot repair, and the cost of failure is total.
Everything in that folder shows up here: requirements traceability, interface control documents (Mars Climate Orbiter is the canonical failure → note 01), FMEA, redundancy and margin, technology readiness levels, and V&V against an environment you cannot fully reproduce.
It’s also where real-time and safety-critical software lives → kinds of software engineering — flight software is written in restricted language subsets with formal review, and the Apollo Guidance Computer’s software is still studied.
Related
- systems engineering — the discipline this field created
- planets — the destinations
- ODEs — how trajectories are computed
- control theory — guidance and attitude control
Source: [reference] — written Aug 2026.
Software engineering
Infrastructure & security
Computer science
Hardware & engineering